identifier
stringlengths
11
32
pdf_url
stringlengths
17
4.62k
lang
stringclasses
120 values
error
stringclasses
1 value
title
stringlengths
2
500
source_name
stringlengths
1
435
publication_year
float64
1.9k
2.02k
license
stringclasses
3 values
word_count
int64
0
1.64M
text
stringlengths
1
9.75M
https://openalex.org/W3012942594
https://pub.epsilon.slu.se/27189/1/lundberg_s_et_al_220223.pdf
English
null
Behavioral Profiles of Adolescent Alcohol-Preferring/Non-preferring (P/NP) and High/Low Alcohol-Drinking (HAD/LAD) Rats Are Dependent on Line but Not Sex
Frontiers in neuroscience
2,022
cc-by
13,956
ORIGINAL RESEARCH published: 13 January 2022 doi: 10.3389/fnins.2021.811401 Behavioral Profiles of Adolescent Alcohol-Preferring/Non-preferring (P/NP) and High/Low Alcohol-Drinking (HAD/LAD) Rats Are Dependent on Line but Not Sex Stina Lundberg1*, Erika Roman1,2 and Richard L. Bell3 1 Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden, 2 Department of Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences, Uppsala, Sweden, 3 Department of Psychiatry, Stark Neuroscience Research Institute, Indiana University School of Medicine, Indianapolis, IN, United States 1 Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden, 2 Department of Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences, Uppsala, Sweden, 3 Department of Psychiatry, Stark Neuroscience Research Institute, Indiana University School of Medicine, Indianapolis, IN, United States Initial contact with alcohol generally occurs during adolescence, and high consumption during this period is associated with increased risk for later alcohol (AUDs) and/or substance use disorders (SUDs). Rodents selectively bred for high or low alcohol consumption are used to identify behavioral characteristics associated with a propensity for high or low voluntary alcohol intake. The multivariate concentric square fieldTM (MCSF) is a behavioral test developed to study rodents in a semi-naturalistic setting. Testing in the MCSF creates a comprehensive behavioral profile in a single trial. The current aim was to examine the behavioral profiles of adolescent, bidirectionally selectively bred male and female high alcohol-consuming (P and HAD1/2) and low alcohol-consuming (NP and LAD1/2) rat lines, and outbred Wistar rats. Alcohol-naïve rats were tested once in the MCSF at an age between postnatal days 30 and 35. No common behavioral profile was found for either high or low alcohol-consuming rat lines, and the effect of sex was small. The P/NP and HAD2/LAD2 lines showed within pair-dependent differences, while the HAD1/LAD1 lines were highly similar. The P rats displayed high activity and risk-associated behaviors, whereas HAD2 rats displayed low activity, high shelter-seeking behavior, and open area avoidance. The results from P rats parallel clinical findings that denser family history and risk-taking behavior are strong predictors of future AUDs, often with early onset. Contrarily, the HAD2 behavioral profile was similar to individuals experiencing negative emotionality, which also is associated with a vulnerability to develop, often with a later onset, AUDs and/or SUDs. Citation: Lundberg S, Roman E and Bell RL (2022) Behavioral Profiles of Adolescent Alcohol-Preferring/Non-preferring (P/NP) and High/Low Alcohol-Drinking (HAD/LAD) Rats Are Dependent on Line but Not Sex. Front. Neurosci. 15:811401. doi: 10.3389/fnins.2021.811401 Keywords: adolescence, alcohol non-preferring rats, alcohol-preferring rats, high alcohol-drinking rats, low alcohol-drinking rats, multivariate concentric square field (MCSF) Edited by: Cristina Núñez, University of Murcia, Spain Reviewed by: Ricardo Marcos Pautassi, Medical Research Institute Mercedes and Martín Ferreyra (INIMEC), Argentina Heidi Lesscher, Utrecht University, Netherlands Heidi Lesscher, Utrecht University, Netherlands *Correspondence: Stina Lundberg Stina.Lundberg@farmbio.uu.se Specialty section: This article was submitted to Neuropharmacology, a section of the journal Frontiers in Neuroscience Specialty section: This article was submitted to Neuropharmacology, a section of the journal Frontiers in Neuroscience Received: 08 November 2021 Accepted: 06 December 2021 Published: 13 January 2022 Received: 08 November 2021 Accepted: 06 December 2021 Published: 13 January 2022 Keywords: adolescence, alcohol non-preferring rats, alcohol-preferring rats, high alcohol-drinking rats, low alcohol-drinking rats, multivariate concentric square field (MCSF) INTRODUCTION Addiction is a complex disorder that is influenced by environmental factors and their interaction with a genetic make-up. Additionally, a vulnerability for addiction is associated with certain personality traits, such as impulsivity and sensation seeking, and share comorbidity with multiple psychiatric disorders, including anxiety and depression (Babor, 1992; Cloninger et al., 1996; January 2022 | Volume 15 | Article 811401 1 Frontiers in Neuroscience | www.frontiersin.org Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. breeding based on alcohol consumption in adolescence. Here, the high alcohol-consuming line showed increased shelter-seeking and decreased risk-taking behaviors when tested in adolescence (Fernández et al., 2020). The same behavioral trends were seen in an earlier replicate of the same breeding procedure using the light-dark box (Fernández et al., 2017). In the present study, we aimed to use the MCSF to further examine the link between a genetic propensity for high or low alcohol consumption and the adolescent behavioral profile before first contact with alcohol by examining naïve male and female P/NP and HAD/LAD (replicates 1 and 2) rats. Additionally, a Wistar cohort was included to provide an outbred comparison to the selectively bred lines. The goal was to evaluate both differences based on selected alcohol consumption (within-pair differences) and differences within the same selection criteria, i.e., among all three high or low alcohol-consuming lines. The hypotheses were to find similar within-pair differences in the behavioral profiles as seen in the study of the adult animals (Roman et al., 2012) and that the HAD/LAD replicates would be similar within each selection criteria while differing from the P/NP lines. Lesch and Walter, 1996; Merikangas et al., 1998; Kranzler and Rosenthal, 2003; Moss et al., 2007). For alcohol use disorders (AUDs), the genetic risk is around 50% (Verhulst et al., 2015; Reilly et al., 2017), reflecting the importance of a family history of AUDs as a risk factor (Reilly et al., 2017). To examine the heritability of a vulnerability to develop AUDs, bidirectional selective breeding strategies in laboratory animals have been used (Bell et al., 2012, 2016, 2017). Breeding programs with bidirectional selection for high vs. MATERIALS AND METHODS Animals, Housing, and Ethics Statement Subjects were high alcohol-consuming P, HAD1, and HAD2 rats; low alcohol-consuming NP, LAD1, and LAD2 rats delivered to the laboratory at the day of weaning (Indiana University School of Medicine, Indianapolis, IN, United States); and outbred Wistar rats (RccHan:WI, Envigo, Indianapolis, IN, United States), for details see Table 1. Animals were pair-housed, by sex and line, in opaque plastic cages (22 × 44 × 20 cm) with sawdust bedding and ad libitum access to food (7001 Teklad 4%, Envigo, Madison, WI, United States) and water in a temperature-controlled (24 ± 1◦C) and humidity-controlled (∼50%) animal room on a reversed 12 h/12 h dark-light cycle, with dark onset at 10:00. The animals were identified by tail markings with a marker pen, which were darkened as needed. Animals were acclimated to their home-cages for 5 days before any procedures were initiated. Experimental procedures were conducted during the dark period of the dark-light cycle. The animal experimental protocol was approved by the Institutional Animal Care and Use Committee of the Indiana University School of Medicine and is consistent with NIH’s Guide for the Care and Use of Laboratory Animals (National Research Council, 2011). The multivariate concentric square fieldTM (MCSF) test is a behavioral test based on an ethoexperimental foundation, and it is designed to generate a behavioral profile that assesses exploration, risk-associated, and shelter-seeking behaviors (Meyerson et al., 2006). This allows for detection of individual and treatment- associated differences during both adolescence and adulthood (e.g., Palm et al., 2014; Karlsson and Roman, 2016; Wille-Bille et al., 2018; Lundberg et al., 2019; Gore-Langton et al., 2021). The MCSF has previously been used to characterize differences between selectively bred line-pairs and within each selection criteria (P/NP, HAD/LAD replicates, sP/sNP, and AA/ANA) in adult males (Roman et al., 2007, 2012; Roman and Colombo, 2009). Within-pair differences varied among the pairs (Roman et al., 2007, 2012; Roman and Colombo, 2009), and substantial differences were found among the high alcohol-consuming lines (Roman et al., 2012). More recently, an adapted version of the MCSF was used to profile differences after short-term selective INTRODUCTION low alcohol consumption have generated seven pairs of high/low alcohol-consuming rat-lines across the world: the University of Chile bibulous/abstainer (UChB/UChA) rats (Quintanilla et al., 2006), the Finnish ALKO alcohol/non-alcohol (AA/ANA) rats (Sommer et al., 2006), the Sardinian alcohol-preferring/non- preferring (sP/sNP) rats (Colombo et al., 2006), the Warsaw high-/low-preferring (WHP/WLP) rats (Dyr and Kostowski, 2008), the Indiana alcohol-preferring/non-preferring (P/NP), and high/low alcohol-drinking (HAD/LAD, replicates 1 and 2) rat lines (McBride et al., 2014). Selective breeding of P/NP and the HAD/LAD replicate rat lines was based on identical selection criteria. P/NP rats were selected from Wistar rats (Murphy et al., 2002) and the HAD/LAD replicate lines from N/NIH rats (Murphy et al., 2002). Selectively bred lines have made it possible to determine behavioral (Roman et al., 2012), neurobiological (Bell et al., 2012), and genetic (McBride et al., 2012; Bell et al., 2017) characteristics associated with selection for high vs. low voluntary alcohol consumption. The high alcohol- consuming lines are also useful for evaluating pharmaceutical candidates to treat AUDs (Bell et al., 2012, 2016, 2017). The existing literature indicates that studies of male and adult animals have predominated in the past, with a modest increase in studies of sex-dependent behavior more recently (Bell et al., 2014). In addition, the consequences of peri-adolescent alcohol drinking have been a more recent focus of study, and both P and HAD rats show increased peri-adolescent alcohol consumption in both continuous and binge-like paradigms (Bell et al., 2014). Nonetheless, little is known about alcohol-naïve adolescent behavior of these lines, especially in females, and such studies would provide further information regarding modeling of family history for AUDs during development (Bell et al., 2013, 2014). Statistical Analysis Statistical analyses were carried out using Statistica 13 (TIBCO Software Inc., Tulsa, OK, United States) unless otherwise specified. Normality was examined with the Shapiro-Wilk’s test. Body weights were normally distributed, and differences were examined using a factorial ANOVA, with line and sex as between- subject factors, followed by post hoc Tukey HSD tests. The behavioral parameters from the MCSF were skewed and analyzed using non-parametric statistics. Continuous parameters were examined with Kruskal-Wallis ANOVA by ranks with post hoc Mann-Whitney U-tests with continuity corrections. Zone and behavior occurrences were analyzed with Maximum-Likelihood Chi-square test. Activity over time in the MCSF was analyzed in R 3.6.1 (R Core Team, 2019) with the nparLD package (Noguchi et al., 2012) with line and sex as between-subject factors and time as the within-subject factor. Time-dependent post hoc tests were performed with Wilcoxon’s matched pairs test and group- dependent tests with Mann-Whitney U-test with continuity corrections. The MCSF parameters were analyzed both with sex separated and with sex collapsed. All non-parametric post hoc comparisons evaluated: (1) differences within each line-pair, (2) differences within each selection criteria, and (3) differences between the Wistar rats and each of the selectively bred lines. Bonferroni corrected multiple comparisons were applied, and only one level of significance is reported (corresponding to p < 0.05). Effect sizes were calculated as previously described (Fritz et al., 2012). FIGURE 1 | Schematic layout of the MCSF. The arena is divided into zones by walls (solid, black lines) or imagined boundaries (dashed lines). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room. Modified with permission from Lundberg et al. (2019). in the CTRCI, 5–10 in the corridors as well as the hurdle, and 500 in the middle of the bridge. Prior to testing, the animals were habituated to the handling procedures for three consecutive days. This consisted of individual handling, weighing, and adaptation to a small transportation cage. Between postnatal day (PND) 30 and 35, the animals were tested once in the MCSF. On the testing days, the animals to be tested were transported in their home cages, in conjunction with lights off, to a holding room. There, the animals were left undisturbed for 2 h before testing was initiated. The transportation cage was then used to transport the animals, one-at-a-time, to the testing room, and back after the test was concluded. The Multivariate Concentric Square FieldTM Test Line n/sex Generation Number of litters Pups/litter and sex Age at weaning or delivery (PND) P 20 S82–S83 11 1–3 22–25 HAD1 19–20 S70 8 2–3 23–25 HAD2 20 S69 9 1–4 22–25 NP 12 S81–S82 6 2 23–27 LAD1 12 S70 6 1–4 24–26 LAD2 12 S69 11 2 22–25 Wistar 20 n.a. n.a. n.a. 22 n.a., not applicable; PND, postnatal day. The trials were recorded by an overhead video camera and manually scored using EthoVision XT 13 (Noldus, Inc., Wageningen, Netherlands). The latency to first visit (L, s), frequency of visits (F, #), and duration spent (D, s) in each zone were scored together with the number of rearings, groomings, and stretched attend postures (SAPs). In addition, the number of urinations and fecal boli in the arena was counted after each trial. The frequency and duration spent in the corridors were summed into total corridor (totcorr) measures, and the latency (L, s) to leave the center after the start of the trial was identified. Further, the duration per visit (D/F, s), percental frequency (%F), and percental duration (%D) were calculated for each zone, including the total corridor measures. The total activity (TOTACT, i.e., sum of all frequencies), number of zones visited, and latency (L, s) to fully explore the arena and occurrences (Occ) for each zone not visited or behaviors not performed by all individuals in a group were also determined. For the corridors, only the total corridor measures were used for analysis, except for occurrences which were analyzed on the individual corridor level. Furthermore, the activity measures total activity and rearing were split into four 5-min periods for analysis of the activity profile over time during the trial. FIGURE 1 | Schematic layout of the MCSF. The arena is divided into zones by walls (solid, black lines) or imagined boundaries (dashed lines). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room. Modified with permission from Lundberg et al. (2019). The Multivariate Concentric Square FieldTM Test The MCSF (Figure 1) is described in detail elsewhere (Meyerson et al., 2006; Roman and Colombo, 2009). Briefly, the arena is 100 × 100 cm and divided into zones: the center (70 × 70 cm) with a central circle (CTRCI, 25 cm in diameter) surrounded on three sides by corridors. Along the fourth side, an elevated and illuminated “bridge” is divided into the slope, bridge entrance (BE), and bridge. One corner has a sheltered dark corner room (DCR) and another a raised hurdle with hole-board flooring. Light conditions in the different zones were as follows (lux): 15 January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org Frontiers in Neuroscience | www.frontiersin.org 2 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. TABLE 1 | Details about the animals in the study; number of animals per sex, to which generation they belong, from how many litters they were derived, and how many animals from each litter entered the study. Line n/sex Generation Number of litters Pups/litter and sex Age at weaning or delivery (PND) P 20 S82–S83 11 1–3 22–25 HAD1 19–20 S70 8 2–3 23–25 HAD2 20 S69 9 1–4 22–25 NP 12 S81–S82 6 2 23–27 LAD1 12 S70 6 1–4 24–26 LAD2 12 S69 11 2 22–25 Wistar 20 n.a. n.a. n.a. 22 n.a., not applicable; PND, postnatal day. FIGURE 1 | Schematic layout of the MCSF. The arena is divided into zones by walls (solid, black lines) or imagined boundaries (dashed lines). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room. Modified with permission from Lundberg et al. (2019). in the CTRCI, 5–10 in the corridors as well as the hurdle, and 500 TABLE 1 | Details about the animals in the study; number of animals per sex, to which generation they belong, from how many litters they were derived, and how many animals from each litter entered the study. wall not leading to a corridor, and each trial lasted for 20 min. Between rats, the arena was sprayed with a 10% ethanol solution, wiped down, and allowed to air dry. After testing, all animals were returned to the animal room where they were weighed. Frontiers in Neuroscience | www.frontiersin.org RESULTS Five animals were excluded from the analyses: four due to technical issues during testing (two HAD1 females, one Wistar female, and one NP male) and one LAD1 female that spent > 70% of the trial climbing on the walls of the arena, which resulted in very little zone-dependent data. Body weights for the animals at the time of testing are found in Supplementary Figure 1. Females weighed less than males in P, NP, and Wistar animals, while no sex-dependent differences were detected in HAD1, LAD1, HAD2, or LAD2 rats. Overall, the Wistar rats were heaviest, followed by P and then NP rats. While lower than P, NP, and Wistar rats, no body weight differences were detected across the HAD1, LAD1, HAD2, and LAD2 lines. When examining activity over time in the MCSF, as indicated by total activity and number of rearings (Supplementary Figures 2A,B, respectively), the pattern over time was similar between P and NP rats. P rats had higher total activity in the second and fourth 5-min periods than NP rats, while the total activity was comparable in the first and third 5-min periods. Both lines had lower initial activity that then sharply increased and peaked in the second 5-min period. For P rats, the total activity then declined to an intermediate level during the second half of the test. For NP rats, the decline was evident only in the fourth 5-min period, where it was comparable to the total activity during the initial 5 min (Supplementary Figure 2A). In number of rearing actions (Supplementary Figure 2B), no line-dependent differences were found. The number of rearings increased in P rats from the first to the second 5-min period; apart from this, no time-dependent differences were found within P or NP rats. In conclusion, P rats showed higher exploratory and risk- associated behaviors in the MCSF, while NP rats showed higher shelter-seeking behavior. No sex-dependent differences were detected among the MCSF parameters in P, NP, HAD1, LAD1, or HAD2 animals. Among LAD2 animals, females urinated fewer times in the arena than males, and among Wistar animals, females had lower percentage frequency of visits to the center than males (Supplementary Table 1). Due to the weak effect of sex, further results are discussed with data collapsed across sex. Statistical Analysis First on each testing day, a non-test rat was allowed to explore the MCSF arena to avoid any first-to-be-tested effect. Males were always tested before females, and the two animals in a cage were never tested directly after each other. A trial started as the rat was released into the center of the arena facing the Principal component analysis (PCA) and partial least squares projections to latent structures discriminant analysis (PLS-DA) were carried out in SIMCA 15 (Sartorius Stedim Data Analytics January 2022 | Volume 15 | Article 811401 3 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. AB, Umeå, Sweden) as a multivariate complement to the standard statistics to further the multivariate interpretation of the MCSF. The PCA examined the relationship among one set of multiple variables, in this case MCSF parameters, and the PLS-DA examined the relationship between two sets of variables, one with multiple variables and one with a single categorical variable, in this case line. PCAs were carried out separately by line-pair to provide an overview of the relationship between each high and low alcohol-consuming line-pair, and PLS-DAs were carried out to examine the differences between all the lines. Autofit was used to generate the models; components were excluded if the eigenvalue was < 2.0 or if Q2 had a large negative value. Latencies, occurrences, percentage duration, and percentage frequency were not included in the models, and other parameters were excluded when advised by the software. scores of NP rats, although some overlap between the lines was seen. Within each line, males and females displayed no separation. The parameter contributions to the analysis are displayed in the loading plot (Supplementary Figure 3A). The separation of P vs. NP scores occurred primarily in the first component (Figure 5A); P rat data were associated with total activity, zone frequencies (except the DCR), and duration at the bridge entrance and on the bridge. NP rat data were associated with duration spent in the corridors and duration per visit to the hurdle, slope, bridge entrance, and bridge (Supplementary Figure 3A). Direct parameter comparisons showed that P rats visited the corridors, hurdle, slope, bridge entrance, and bridge more than NP rats (Figures 2A, 3A), which translated into higher total activity in P vs. NP rats (Figures 2A, 4A). Statistical Analysis P rats had shorter durations in the corridors and DCR, while having longer duration at the bridge entrance and bridge than NP rats (Figures 2A, 3B). The duration per visit was lower for P rats than NP rats in the center, corridors, DCR, and on the bridge (Figures 2A, 3C). Finally, P rats had lower percentage duration in the DCR, and lower percentage duration and frequency in the corridors than NP rats (Figure 2A and Supplementary Table 1). Instead, P rats had higher percentage duration at the bridge entrance, and higher percentage duration and frequency to the bridge than NP rats (Figure 2A and Supplementary Table 1). Within Pair Comparisons Effect sizes for differences within each line-pair are presented in Figure 2A; results for frequencies, total duration, and duration per visit are displayed in Figure 3. The activity measures of total activity (i.e., sum of all zone frequencies) and rearing are shown in Figure 4 (totals for the 20-min trial) and Supplementary Figures 2A–F (over time in 5-min bins); and remaining results are found in Supplementary Table 1. HAD1 and LAD1 Rats The PCA analysis (n = 60, 2 components, R2X = 0.54, Q2 = 0.27) score plot (Figure 5B) shows a large overlap between the lines. Individual scores from both lines clustered together, indicating small variance within the main population, and an outlying group with a large negative contribution of component 1. Results in the loading plot (Supplementary Figure 3B) indicated that this outlying group had higher duration per visit to the DCR, hurdle, and slope. In addition, these outlying individuals did not visit all of the zones in the MCSF arena (Supplementary Data 1). RESULTS However, for transparency, both numerical and statistical results are declared for each sex in Supplementary Table 1 together with the results with sex collapsed. P and NP Rats The PCA analysis (n = 63, 4 components, R2X = 0.63, Q2 = 0.28) score plot (Figure 5A) confirmed the small effect of sex and indicated an influence of line on behavior in the MCSF by P and NP rats. Individual scores of P rats had higher positive loading on component 1 than the Direct parameter comparisons indicated a small difference within the line-pair. HAD1 and LAD1 rats did not differ January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org 4 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. FIGURE 2 | Heatmap of effect sizes for the significant differences comparing (A) the high and low alcohol-consuming rats within each selectively bred pair, (B three high consuming lines, (C) the three low consuming lines, and (D) the selectively bred lines relative to the outbred Wistar rats. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24, Wistar n = 39. BE, bridge entrance; CTRCI, central circle; corr, corridor; D, duration (s); DCR, dark corner ro D/F, duration per visit (s); F (#), frequency; H1/HAD1, high alcohol-drinking line, replicate 1; H2/HAD2, high alcohol-drinking line, replicate 2; L, latency (s); L1/L low alcohol-drinking line, replicate 1; L2/LAD2, low alcohol-drinking line, replicate 2; n.s., non-significant; NP, alcohol non-preferring line; Occ, occurrence (for not visited or behaviors not performed by all individuals); P, alcohol preferring line; SAP, stretched attend posture; TOTACT, total activity (i.e., sum of all frequen totcorr, total corridor; Wi, Wistar. FIGURE 2 | Heatmap of effect sizes for the significant differences comparing (A) the high and low alcohol-consuming rats within each selectively bred pair, (B) the three high consuming lines, (C) the three low consuming lines, and (D) the selectively bred lines relative to the outbred Wistar rats. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24, Wistar n = 39. P and NP Rats BE, bridge entrance; CTRCI, central circle; corr, corridor; D, duration (s); DCR, dark corner room; D/F, duration per visit (s); F (#), frequency; H1/HAD1, high alcohol-drinking line, replicate 1; H2/HAD2, high alcohol-drinking line, replicate 2; L, latency (s); L1/LAD1, low alcohol-drinking line, replicate 1; L2/LAD2, low alcohol-drinking line, replicate 2; n.s., non-significant; NP, alcohol non-preferring line; Occ, occurrence (for zones not visited or behaviors not performed by all individuals); P, alcohol preferring line; SAP, stretched attend posture; TOTACT, total activity (i.e., sum of all frequencies); totcorr, total corridor; Wi, Wistar. January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. IGURE 3 | Results from the MCSF test in the selectively bred rats. Zone (A) frequency (#), (B) duration (s), and (C) duration per visit (s) in P, NP, HAD1, LAD1, AD2, and LAD2 rats; (D) frequency (#), (E) duration (s), and (F) duration per visit (s) in the CTRCI for all lines. Data are presented as median with upper and lower uartiles. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24. *p < 0.0023 comparing high and low alcohol-consuming rats within each electively bred pair; P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats (post hoc Mann-Whitney U-test with ontinuity correction). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room; HAD1, high alcohol drinking line, replicate 1; HAD2, high alcohol-drinking ne, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line; otcorr, total corridor. f d f h ( d d d d ff b d FIGURE 3 | Results from the MCSF test in the selectively bred rats. Zone (A) frequency (#), (B) duration (s), and (C) duration per visit (s) in P, NP, HAD1, LAD1, HAD2, and LAD2 rats; (D) frequency (#), (E) duration (s), and (F) duration per visit (s) in the CTRCI for all lines. Data are presented as median with upper and lower quartiles. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24. P and NP Rats FIGURE 5 | Scatter plots of the individual scores from the PCA analyses of the MCSF parameters in (A) P and NP rats [n = 63, 2 out of 4 components i li d R2X 0 46 Q2 0 29] (B) HAD1 d LAD1 t FIGURE 4 | Activity in the MCSF expressed as (A) total activity (i.e., sum of all frequencies) and (B) rearing in the selectively bred lines and outbred Wistar rats. Data are presented as median with upper and lower quartiles. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24, Wistar n = 39. *p < 0.0023 comparing high and low alcohol-consuming rats within each selectively bred pair; P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats; L1 p < 0.0023 relative to LAD1 rats (post hoc Mann-Whitney U-test with continuity correction). HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line. FIGURE 4 | Activity in the MCSF expressed as (A) total activity (i.e., sum of all frequencies) and (B) rearing in the selectively bred lines and outbred Wistar rats. Data are presented as median with upper and lower quartiles. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24, Wistar n = 39. *p < 0.0023 comparing high and low alcohol-consuming rats within each selectively bred pair; P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats; L1 p < 0.0023 relative to LAD1 rats (post hoc Mann-Whitney U-test with continuity correction). HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line. The HAD1 and LAD1 rats showed only minor changes in their activity (i.e., total activity and rearing) over time in the MCSF (Supplementary Figures 2C,D). LAD1 rats showed no time- dependent changes, while HAD1 rats had higher total activity in the second compared to the fourth 5-min period (Supplementary Figure 2C). P and NP Rats HAD1 rats had a higher number of rearings in the second 5-min period than LAD1 rats (Supplementary Figure 2D). HAD1 rats also had a higher number of rearings in the second compared to the first and fourth 5-min periods (Supplementary Figure 2D). In conclusion, HAD1 and LAD1 rats had a high degree of similarity in their behavioral profiles as assessed with the MCSF. FIGURE 5 | Scatter plots of the individual scores from the PCA analyses of the MCSF parameters in (A) P and NP rats [n = 63, 2 out of 4 components visualized, R2X(1−2) = 0.46, Q2(1−2) = 0.29], (B) HAD1 and LAD1 rats [n = 60, 2 out of 2 components visualized, R2X(1−2) = 0.54, Q2(1−2) = 0.27], and (C) HAD2 and LAD2 rats [n = 64, 2 out of 3 components visualized, R2X(1−2) = 0.46, Q2(1−2) = 0.24]. The plots are colored according to line and the shapes represent males (circles) and females (inverted triangles). The corresponding loading plots can be found in Supplementary Figure 3. HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line. P and NP Rats *p < 0.0023 comparing high and low alcohol-consuming rats within each selectively bred pair; P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats (post hoc Mann-Whitney U-test with continuity correction). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room; HAD1, high alcohol drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line; totcorr, total corridor. activity and rearing did not differ between HAD1 and LAD1 rats (Figures 4A,B). However, a lower proportion of HAD1 rats defecated in the arena, resulting in a lower fecal boli count in HAD1 vs. LAD1 rats (Figure 2A and Supplementary Table 1). in frequency nor duration in any of the zones (Figures 2A, 3A,B,D,E). Nor did the duration per visit differ between HAD1 and LAD1 rats (Figures 2A, 3C,F), except in the DCR where HAD1 rats had longer duration per visit than LAD1 rats (Figures 2A, 3C). The activity measures of total January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org 6 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. FIGURE 4 | Activity in the MCSF expressed as (A) total activity (i.e., sum of all frequencies) and (B) rearing in the selectively bred lines and outbred Wistar rats. Data are presented as median with upper and lower quartiles. P n = 40; NP n = 23; HAD1 n = 37; LAD1 n = 23; HAD2 n = 40, LAD2 n = 24, Wistar n = 39. *p < 0.0023 comparing high and low alcohol-consuming rats within each selectively bred pair; P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats; L1 p < 0.0023 relative to LAD1 rats (post hoc Mann-Whitney U-test with continuity correction). HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line. Frontiers in Neuroscience | www.frontiersin.org HAD2 and LAD2 Rats The PCA analysis (n = 64, 3 components, R2X = 0.56, Q2 = 0.20) score plot (Figure 5C) confirmed a small effect of sex and indicated an influence of line on behavior in the MCSF by HAD2 and LAD2 rats. Individual scores of HAD2 rats loaded widely across all four quadrants, while the scores of LAD2 rats clustered together predominantly in the lower right quadrant (Figure 5C), indicating lower variance in the LAD2 rat data. Within each line, males and females displayed no separation. The parameter contributions to the analysis are displayed in the loading plot (Supplementary Figure 3C). LAD2 rats January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. were characterized by total activity, rearing, duration in the corridors, and frequency of visits to the center, CTRCI, corridors, DCR, and hurdle. (Figures 2B, 3C). P rats had higher duration in the center and CTRCI than either HAD line, whereas the duration was higher in the hurdle and lower in the slope in P vs. HAD2 rats (Figures 2B, 3B,E). A higher proportion of P rats visited the entire arena than either HAD line; a higher proportion of P rats visited the CTRCI, DCR, and two of the corridors than HAD2 rats; and the proportion of P rats visiting the slope was higher than that of HAD1 rats (Figure 2B and Supplementary Table 1). Lastly, P rats had a shorter latency in leaving the center at the start of the trial, shorter latency in fully exploring the entire arena, and shorter latencies in first entering the hurdle and slope than HAD2 rats, whereas no latencies differed between P and HAD1 rats on these measures (Figure 2B and Supplementary Table 1). Direct parameter comparisons showed that HAD2 vs. LAD2 line differences were enriched among parameters relating to the center and CTRCI (Figure 2A and Supplementary Table 1). HAD2 rats had a lower frequency of visits to the center and CTRCI [in absolute numbers (Figures 3A,D) and by percentage frequency (Figure 2A and Supplementary Table 1)], a lower duration in the CTRCI [in absolute numbers (Figure 3E) and by percentage duration (Figure 2A and Supplementary Table 1)], and higher duration per visit to the center (Figures 2A, 3C) than LAD2 rats, and fewer HAD2 than LAD2 rats visited the CTRCI (Figure 2A and Supplementary Table 1). HAD2 and LAD2 Rats Furthermore, among the other parameters, HAD2 rats had higher durations per visit in the corridors and on the bridge entrance than LAD2 rats (Figures 2A, 3C). HAD2 rats also reared less (Figure 4B) and a lower proportion of HAD2 animals visited corridor “b” and all zones compared to LAD2 rats (Figure 2A and Supplementary Table 1). The percentage frequency of visits to the slope, the number of grooming bouts, and the proportion of animals grooming were higher in HAD2 than LAD2 rats (Figure 2A and Supplementary Table 1). The two HAD replicates also showed some differences. The total activity was higher in HAD1 vs. HAD2 animals (Figures 2B, 4A), driven by higher frequency of visits to the corridors and CTRCI in HAD1 than HAD2 rats (Figures 2B, 3A,D). HAD1 rats also reared more than HAD2 rats (Figures 2B, 4B). HAD1 rats had shorter duration on the slope (Figures 2B, 3B) and shorter duration per visit in all zones except the DCR, hurdle, and CTRCI (Figures 2B, 3C,F) than HAD2 rats. The latency in leaving the center at the start of the trial and in first entry of the hurdle, bridge entrance, and bridge were lower in HAD1 than HAD2 rats (Figure 2B and Supplementary Table 1). For HAD1 rats, a higher proportion visited the entire arena and corridor “b,” while a lower proportion visited the slope and bridge entrance than the proportion of HAD2 rats (Figure 2B and Supplementary Table 1). pp y There were no differences between the lines in total activity in any of the 5-min periods (Supplementary Figure 2E). HAD2 rats increased in total activity from the first to the second and third 5- min periods, and the total activity during the second period was also higher than in the fourth 5-min period. LAD2 rats increased in total activity from the first to the second 5-min period, while no other time-dependent differences were seen. In rearing over time (Supplementary Figure 2F), HAD2 rats had lower number recorded during the first, second, and fourth 5-min periods than LAD2 rats. LAD2 rats were stable in their number of rearings over time, while HAD2 rats increased their number of rearings from the first to the second and third 5-min period. In conclusion, HAD2 rats had lower activity and higher avoidance of open areas than LAD2 rats in the MCSF test. HAD2 and LAD2 Rats The total activity per 5-min period (Supplementary Figures 2A,C,E) was higher in P rats than either HAD line throughout the test, while the HAD1 and HAD2 lines were similar except in the initial 5 min where HAD1 rats had higher activity. In the pattern over time, P and HAD2 animals showed similar tendencies, although the actual frequency counts where approximately twice as high in P than HAD2 rats, with a clear increase to a peak total activity in the second 5-min period (Supplementary Figures 2A,E), as opposed to HAD1 rats that had a much flatter curve. The rearing over time (Supplementary Figures 2B,D,F) was also higher for P rats than either HAD line, except in the second period where P and HAD1 rats did not differ. HAD1 rats had higher frequency of rearing than HAD2 throughout the test. For rearing, the pattern over time was more similar among the lines, with increases between the first and second 5-min period in all three lines. Frontiers in Neuroscience | www.frontiersin.org Comparisons Among High Alcohol-Consuming Rat Lines Effect sizes for differences among the high alcohol-consuming rat lines are presented in Figure 2B; results for frequencies, total duration, and duration per visit are displayed in Figure 3. The activity measures of total activity and rearing are displayed in Figure 4 (totals for the 20-min trial) and Supplementary Figures 2A–F (over time in 5-min periods); and remaining results are found in Supplementary Table 1. In the PLS-DA analysis of the selectively bred lines (n = 187, 3 components, R2X = 0.57, R2Y = 0.26, Q2 = 0.22; Figure 6), the three high alcohol-consuming lines’ scores differed mainly in the first component (Figure 6A). The P rats had scores with a negative contribution on component 1, the scores for the HAD1 rats clustered around the origin, and the scores for the HAD2 rats had a largely positive contribution from component 1 as well as loading mostly in the lower right quadrant (Figure 6A). The HAD1 line had an outlying group in the upper right quadrant (Figure 6A), which constituted an outlying cluster in the PCA analysis as well (Figure 5B). In the loading plot (Figure 6B), the relationship between the class variable (line) in the form of aggregate dummy variables and the parameters (MCSF variables) P rats differed substantially from the two HAD lines. P rats had the highest number of visits to all zones (Figures 2B, 3A,D), which resulted in higher total activity than either HAD line (Figures 2B, 4A). P rats also reared more than either HAD line (Figures 2B, 4B). Moreover, P rats had a lower duration per visit to each zone than either HAD line (Figures 2B, 3C), except for the CTRCI, where the duration per visit did not differ among the lines (Figures 2B, 3F), and at the bridge entrance, the P rats had a lower duration per visit than HAD2, but not HAD1, rats January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org 8 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. is visualized. The P rat aggregate variable loaded with a large negative contribution from component 1 and a low contribution from component 2, which was associated with frequency of visits to all zones, total activity, duration in the CTRCI, and number of rearings. Comparisons Among High Alcohol-Consuming Rat Lines The HAD1 rat aggregate variable loaded in the upper right quadrant with a larger contribution from component 2 compared with component 1, which was associated with frequency of grooming, as well as duration and duration per visit to the hurdle and DCR. The HAD2 rat aggregate variable loaded in the lower right quadrant with relatively equal contributions from components 1 and 2, which was associated with duration on the slope, number of urinations and fecal boli, and duration per visit in the center, slope, and bridge (Figure 6B). In conclusion, there were no common behavioral profiles across the P and HAD replicate lines. P rats were highly active in the arena and showed increased risk-associated behavior, while HAD2 rats had low activity and increased risk-avoidance and shelter-seeking behavior. Lastly, the HAD1 rat data clustered between that of the other two high alcohol-consuming lines. breaking down the total activity and rearing frequency in 5-min periods (Supplementary Figure 2), NP rats had higher scores than either LAD line at all time-points with a few exceptions; NP and LAD1 rats did not differ in total activity during the first 5-min period (Supplementary Figures 2A,C), and none of the low alcohol-consuming lines differed in rearing activity during the last 5-min period (Supplementary Figures 2B,F). In activity pattern overtime, the low alcohol-consuming lines were similar in rearing activity where all three lines had flat curves, which is in contrast to their high-consuming counterparts (Supplementary Figures 2B,D,F). Contrary, in total activity over time, there was more correspondence within each line-pair than among the low alcohol-consuming lines (Supplementary Figures 2A,C,D). g ( pp y g , , ) In the PLS-DA analysis (n = 187, 3 components, R2X = 0.57, R2Y = 0.26, Q2 = 0.22; Figure 6), the scores (Figure 6A) for the NP rats had a small to moderate negative contribution from component 1 and a low contribution from component 2. The scores for the LAD1 rats loaded around the origin with a few outlying individuals in the upper right quadrant, whereas the scores for the LAD2 rats loaded mostly in the lower right quadrant. The loading plot (Figure 6B) shows that the aggregate variables for the low alcohol-consuming lines loaded closer to the origin, while retaining a similar vector orientation, compared with their high alcohol-consuming counterparts, indicating that variance among the low alcohol-consuming lines was smaller than that of the high alcohol-consuming lines. Comparisons Among High Alcohol-Consuming Rat Lines The NP rat aggregate variable had a negative contribution from component 1 with minimal contribution from component 2. These results were associated with duration spent in the corridors and, to a lesser extent, the frequency of visits to each zone. The LAD1 rat aggregate variable loaded in the upper right quadrant, close to the origin, which was associated with duration per visit to the DCR, and duration in the DCR and bridge entrance. The LAD2 rat aggregate variable loaded in the lower right quadrant, which was associated with duration spent in the center and on the bridge, and duration per visit to the center and CTRCI. In conclusion, no common behavioral profiles were found across the NP and LAD replicate rat lines, although the differences were fewer than that observed between the high alcohol-consuming lines. The two LAD replicate rat lines showed a large degree of similarity, while NP rats had higher activity and exploration in the MCSF than either LAD rat line. Frontiers in Neuroscience | www.frontiersin.org Alcohol-Consuming Rat Lines Effect sizes for differences among the low alcohol-consuming rat lines are presented in Figure 2C; results for frequencies, total duration, and duration per visit are displayed in Figure 3. The activity measures of total activity and rearing are shown in Figure 4 (totals for the 20-min trial) and Supplementary Figures 2A–F (over time in 5-min periods); and remaining results are found in Supplementary Table 1. pp y The NP rats differed substantially from the two LAD lines. NP rats had a higher frequency of visits to the center, corridors, DCR, hurdle, and CTRCI zones than either LAD line and a higher frequency of visits to the slope than LAD2 rats (Figures 2C, 3A,D); this resulted in higher total activity in NP rats than either LAD line (Figures 2C, 4A). NP rats also reared more than either LAD line (Figures 2C, 4B). NP rats had a lower duration per visit to the center and corridors than either LAD line, as well as a lower duration per visit to the bridge entrance than the LAD1 rats and a lower duration per visit to the slope and on the bridge than LAD2 rats (Figures 2C, 3C). NP rats had a longer duration in the DCR than LAD1 rats, and a lower duration on the bridge and a higher duration in the corridors and CTRCI than LAD2 rats (Figures 2C, 3B,E). The NP rats had a higher latency to first entry of the bridge entrance and bridge than LAD1 rats (Figure 2C and Supplementary Table 1). Lastly, NP rats engaged in less grooming behavior than LAD1 rats and urinated fewer times, produced fewer fecal boli, and had a lower proportion of individuals defecating than LAD2 rats (Figure 2C and Supplementary Table 1). Comparisons Relative to Wistar Rats FIGURE 6 | Scatter plots of (A) individual scores and (B) variable loadings from the PLS-DA [n = 187, 2 out of 3 components visualized, R2X(1−2) = 0.47, R2Y(1−2) = 0.21, Q2(1−2) = 0.18] of the MCSF parameters (variable set 1) and selectively bred line (variable set 2). The plots are colored according to line and the shapes represent males (circles) and females (inverted triangles). BE, bridge entrance; CTRCI, central circle; D, duration; DCR, dark corner room; D/F, duration per visit; F, frequency; HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low FIGURE 6 | Scatter plots of (A) individual scores and (B) variable loadings from the PLS-DA [n = 187, 2 out of 3 components visualized, R2X(1−2) = 0.47, FIGURE 6 | Scatter plots of (A) individual scores and (B) variable loadings from the PLS-DA [n = 187, 2 out of 3 components visualized, R2X(1−2) = 0.47, R2Y(1−2) = 0.21, Q2(1−2) = 0.18] of the MCSF parameters (variable set 1) and selectively bred line (variable set 2). The plots are colored according to line and the shapes represent males (circles) and females (inverted triangles). BE, bridge entrance; CTRCI, central circle; D, duration; DCR, dark corner room; D/F, duration per visit; F, frequency; HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line; TOTACT, total activity (i.e., sum of all zone frequencies); totcorr, total corridor. had higher frequency than HAD1 rats in 7 out of 8 zones, than HAD2 in all zones, than LAD1 in 5 zones, and than LAD2 in 6 zones while only having higher frequency to the bridge entrance and bridge than NP rats and lower frequency to the corridors and hurdle than P rats. The same pattern is present for differences in duration per visit (Figures 2D, 3C,F, 7C,F); Wistar rats had lower duration per visit to most zones compared to the HAD and LAD lines, while the differences compared to P and NP rats were fewer. Differences among the zone durations were few overall (Figures 2D, 3B,E, 7B,E), but most noticeable were a higher duration in the CTRCI in Wistar vs. either HAD or LAD line, higher duration in the corridors in Wistar vs. Comparisons Relative to Wistar Rats Comparisons Relative to Wistar Rats Effect sizes for differences relative to that of the outbred Wistar rats are presented in Figure 2D; results for frequencies, total duration, and duration per visit are displayed in Figure 7. The activity measures of total activity and rearing are shown in Figure 4 (totals for the 20-min trial) and Supplementary Figures 2G,H (over time in 5-min periods); and remaining results are found in Supplementary Table 1. The two LAD replicate lines were largely similar. The LAD1 and LAD2 rats did not differ in frequency or duration in any zone (Figures 2C, 3A,B) nor in total activity or rearing (Figures 2C, 4). However, LAD1 rats groomed more, had shorter latency to the bridge entrance and bridge (Figure 2C and Supplementary Table 1), and had lower duration per visit on the slope and bridge (Figures 2C, 3C) than LAD2 rats. When Wistar rats reared more during the trial than either HAD or LAD line, while they did not differ from the P or NP rats (Figures 2D, 4B). Wistar rats also had higher total activity than either HAD or LAD line, while they did not differ from the P or NP rats (Figures 2D, 4A). This is indicative of the differences seen in the zone frequencies (Figures 2D, 3A,D, 7A,D); Wistar rats January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org 9 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. FIGURE 6 | Scatter plots of (A) individual scores and (B) variable loadings from the PLS-DA [n = 187, 2 out of 3 components visualized, R2X(1−2) = 0.47, R2Y(1−2) = 0.21, Q2(1−2) = 0.18] of the MCSF parameters (variable set 1) and selectively bred line (variable set 2). The plots are colored according to line and the shapes represent males (circles) and females (inverted triangles). BE, bridge entrance; CTRCI, central circle; D, duration; DCR, dark corner room; D/F, duration per visit; F, frequency; HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line; TOTACT, total activity (i.e., sum of all zone frequencies); totcorr, total corridor. DISCUSSION In this study, the alcohol-naïve behavioral profiles of adolescent males and females from the Indiana selectively bred lines for high or low alcohol intake and preference (P/NP, HAD1/LAD1 and HAD2/LAD2) were examined using the MCSF test. The effect of sex on the behavioral profiles was minor across the rat lines, while there were considerable differences both within the line-pairs and between the selected phenotypes. The P/NP and HAD2/LAD2 line-pairs showed within-pair differences, while the HAD1 and LAD1 lines were highly similar. There was no common behavioral profile associated with either high or low alcohol-consuming phenotypes. The high alcohol-consuming lines differed substantially, especially in activity levels, where P rats showed the highest and HAD2 rats displayed the lowest activity. The low alcohol-consuming lines were more similar than their high alcohol-consuming counterparts, although the NP line differed from both LAD replicate lines, which were similar to each other. Comparisons Relative to Wistar Rats In rearing activity over time (Supplementary Figures 2B,D,F,H), Wistar rats had higher levels than either HAD or LAD line throughout the trial, except compared to HAD1 rats in the second 5-min period. Wistar rats did not differ in rearing activity over time compared to P or NP rats. Over time, rearing in Wistar rats increased from the first 5-min period compared to all subsequent periods. This pattern over time corresponded to a higher degree with the high alcohol-consuming lines, although no other line showed such a marked and sustained increase in rearing activity over the trial. In conclusion, the Wistar rats differed substantially from either HAD or LAD line, especially in overall exploration and risk-associated behaviors in the CTRCI. Differences compared to the P and NP lines were fewer but still pronounced, especially in activity at the start of the trial. The result of the PLS-DA analysis that included all of the rat lines (Supplementary Figure 4) supported this overall pattern; Wistar rats differed from all the selectively bred lines, although there were more similarities with the P and NP lines than with the HAD and LAD replicates. In rearing activity over time (Supplementary Figures 2B,D,F,H), Wistar rats had higher levels than either HAD or LAD line throughout the trial, except compared to HAD1 rats in the second 5-min period. Wistar rats did not differ in rearing activity over time compared to P or NP rats. Over time, rearing in Wistar rats increased from the first 5-min period compared to all subsequent periods. This pattern over time corresponded to a higher degree with the high alcohol-consuming lines, although no other line showed such a marked and sustained increase in rearing activity over the trial. In conclusion, the Wistar rats differed substantially from either HAD or LAD line, especially in overall exploration and risk-associated behaviors in the CTRCI. Differences compared to the P and NP lines were fewer but still pronounced, especially in activity at the start of the trial. The result of the PLS-DA analysis that included all of the rat lines (Supplementary Figure 4) supported this overall pattern; Wistar rats differed from all the selectively bred lines, although there were more similarities with the P and NP lines than with the HAD and LAD replicates. Behavioral Profiles of the P and NP Rat Lines The present results revealed differences in adolescent behavior compared to previous studies with adult animals of the P and NP lines. An earlier study using the MCSF to profile adult male rats from five of the bidirectionally selectively bred line-pairs (including P/NP and HAD/LAD replicates) found lower activity in P than the NP rats (Roman et al., 2012). This contrasted with the present study where P rats had the highest levels of activity across all six lines. However, in studies using the open field test to assess locomotor activity, both adult male (Badishtov et al., 1995; Nowak et al., 2000; McKinzie et al., 2002; Roman et al., 2012) and female (McKinzie et al., 2002) P rats displayed higher locomotor activity than that seen in NP rats. Similar studies in adolescent or juvenile P/NP rats reported conflicting results; when tested at PND 33–40 P rats, independent of sex, FIGURE 7 | Results from the MCSF test in the outbred Wistar rats (n = 39). FIGURE 7 | Results from the MCSF test in the outbred Wistar rats (n = 39). Zone (A) frequency (#), (B) duration (s), and (C) duration per visit (s); (D) frequency (#), (E) duration (s), and (F) duration per visit (s) in the CTRCI. Data are presented as median with upper and lower quartile. P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats; L1 p < 0.0023 relative to LAD1 rats; H2 p < 0.0023 relative to HAD2 rats; L2 p < 0.0023 relative to LAD2 rats (post hoc Mann-Whitney U-test with continuity correction). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room; HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line; totcorr, total corridor. FIGURE 7 | Results from the MCSF test in the outbred Wistar rats (n 39). Zone (A) frequency (#), (B) duration (s), and (C) duration per visit (s); (D) frequency (#), (E) duration (s), and (F) duration per visit (s) in the CTRCI. Data are presented as median with upper and lower quartile. Comparisons Relative to Wistar Rats P, HAD2 or LAD2 rats, lower duration in the DCR in Wistar vs. P or NP rats, and lower duration on the slope in Wistar vs. HAD2 or LAD2 rats. Overall, Wistar rats were the fastest to leave the center and start exploring the rest of the arena and had low latencies to many of the zones; however, the latency to fully explore the arena only differed between Wistar and HAD2 and LAD2 rats and Wistar rats differed in zone occurrences only relative HAD2 rats for two separate zones (Figure 2D and Supplementary Table 1). When examining the total activity over time (Supplementary Figures 2A,C,E,G), Wistar rats had higher total activity than all selectively bred lines in the initial 5-min period and remained higher than either HAD or LAD line throughout the trial, except compared to LAD1 in the third 5-min period. From the second 5-min period, Wistar and NP rats had comparable total activity levels, while compared to P rats, Wistar rats had lower total activity level during the last two 5-min period. Over time, Wistar rats decreased their total activity from the first half of the trial to the second; this temporal pattern was unique to the Wistar rats. January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org Frontiers in Neuroscience | www.frontiersin.org 10 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. FIGURE 7 | Results from the MCSF test in the outbred Wistar rats (n = 39) Zone (A) frequency (#), (B) duration (s), and (C) duration per visit (s); (D) frequency (#), (E) duration (s), and (F) duration per visit (s) in the CTRCI. Da are presented as median with upper and lower quartile. P p < 0.0023 relat to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD rats; L1 p < 0.0023 relative to LAD1 rats; H2 p < 0.0023 relative to HAD2 rats; L2 p < 0.0023 relative to LAD2 rats (post hoc Mann-Whitney U-test w continuity correction). BE, bridge entrance; CTRCI, central circle; DCR, dar corner room; HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1 LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; alcohol preferring line; totcorr, total corridor. Behavioral Profiles of the HAD and LAD Replicate Rat Lines The replicate-dependent differences between the HAD/LAD lines also differed from previous studies. Using various behavioral tests, adult male HAD and LAD rats have been reported to be similar both within and between each replicate pair (Viglinskaya et al., 1995; Overstreet et al., 1997; Hwang et al., 2004). Similar observations have been reported for adolescent male and female rats (Rodd et al., 2004). However, differences have been reported between the HAD/LAD replicates using the open field and elevated plus maze tests (Roman et al., 2012). In the present study, the HAD2 rats differed from the other HAD/LAD lines with lower activity and higher risk avoidance, while the HAD1, LAD1, and LAD2 lines followed the main trend of similarity seen in the literature. Notably, HAD2 rats had lower risk-associated behavior in the CTRCI compared with the LAD2 rats. This has previously been observed among adult males of HAD1 vs. LAD1 rats (Roman et al., 2012). This previous MCSF study of the adult lines reported limited differences within each HAD/LAD line- pair, while showing more pronounced differences between the replicates: higher exploratory activity in HAD1 vs. HAD2 rats and lower risk-assessment behavior in LAD1 vs. LAD2 rats (Roman et al., 2012). Further interpretation of similarities to the adult behavior is difficult since previous work has shown that general and exploratory activity do not segregate well in the MCSF during adolescence and that risk-assessment behavior in the MCSF is expressed to a lower degree during adolescence than in adulthood (Lundberg et al., 2019). Support for a greater similarity between HAD1 vs. LAD1 rats than between HAD2 vs. LAD2 or P vs. NP rats was reported in a previous study examining the effects of repeated alcohol-deprivations on alcohol intake by LAD1, LAD2, and NP rats (Bell et al., 2004). This study revealed higher alcohol intake in LAD1 rats after multiple deprivations (Bell et al., 2004). This suggests that the separation in the selection phenotype y y The present results of differences between P and HAD lines may be explained by the fact that these rat lines were derived from different foundation stocks: the P/NP lines from Wistar rats and the HAD/LAD lines from N/NIH rats (Murphy et al., 2002; Bell et al., 2012, 2016). Heterogeneity Within the Selection Criteria That there are differences between the present adolescent results and previously reported adult behavior in the MCSF is not entirely surprising, as a previous study reported effects of age on behavior in the MCSF in outbred male Wistar rats (Lundberg et al., 2019). The present study indicates a similar pattern for P/NP behavior in the MCSF, which unexpectedly was the reverse of previous findings in adult P/NP males (Roman et al., 2012). However, the results from this previous study in male HAD/LAD rats (Roman et al., 2012) were more similar to the present results. Nowak et al. (2000) discussed apparent differences in the literature regarding interpretations of anxiety-like behavior (Stewart et al., 1993; Hwang et al., 2004; Pandey et al., 2005; Zhang et al., 2010) as well as locomotion and novelty-seeking behavior (Badishtov et al., 1995; Nowak et al., 2000; McKinzie et al., 2002; Roman et al., 2012) in P vs. NP rats, and to some extent HAD vs. LAD rats. They proposed that the findings may be connected to increased responsivity to novelty in P and HAD animals and a higher propensity to express “escape” rather than “exploratory” behaviors in a novel, inescapable environment (Nowak et al., 2000). However, this hypothesis is not supported by our findings, as P rats showed increased risk-associated behavior and increased activity, which may be attributed to higher novelty seeking and exploration. Also, the findings of low activity and high shelter- seeking behavior in HAD2 rats contradict this hypothesis (Nowak et al., 2000). Thus, further research is needed to tease apart these phenotypic subtypes. Behavioral Profiles of the P and NP Rat Lines P p < 0.0023 relative to P rats; NP p < 0.0023 relative to NP rats; H1 p < 0.0023 relative to HAD1 rats; L1 p < 0.0023 relative to LAD1 rats; H2 p < 0.0023 relative to HAD2 rats; L2 p < 0.0023 relative to LAD2 rats (post hoc Mann-Whitney U-test with continuity correction). BE, bridge entrance; CTRCI, central circle; DCR, dark corner room; HAD1, high alcohol-drinking line, replicate 1; HAD2, high alcohol-drinking line, replicate 2; LAD1, low alcohol-drinking line, replicate 1; LAD2, low alcohol-drinking line, replicate 2; NP, alcohol non-preferring line; P, alcohol preferring line; totcorr, total corridor. January 2022 | Volume 15 | Article 811401 11 Frontiers in Neuroscience | www.frontiersin.org Frontiers in Neuroscience | www.frontiersin.org Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. between HAD1 vs. LAD1 rats may not be as stark as that observed between HAD2 vs. LAD2 or P vs. NP rats. had higher activity than NP animals after a saline injection (Rodd et al., 2004), whereas testing at PND 20 or 28 revealed no difference between P and NP rats (McKinzie et al., 2002). The present finding of higher risk-associated behavior in adolescent P rats also contrasts with earlier work that described adult P rats displaying behaviors interpreted as higher anxiety-like behavior (Stewart et al., 1993; Hwang et al., 2004; Pandey et al., 2005; Zhang et al., 2010), although other studies reported no such differences compared with adult NP rats (Badishtov et al., 1995; Viglinskaya et al., 1995; Roman et al., 2012). The previous study using the MCSF to profile adult selectively bred lines demonstrated different effects on risk-taking behavior in P vs. NP rats whether the difference in general activity was adjusted for or not. When the data was not adjusted for general activity, male P rats were less risk-taking than NP males. Contrarily, when the data was adjusted for general activity, male P rats had higher percentage frequency of visits to risk areas than NP males (Roman et al., 2012). In the present study, the results were in the same direction whether the analysis included an adjustment for general activity or not. Frontiers in Neuroscience | www.frontiersin.org Behavioral Profiles of the HAD and LAD Replicate Rat Lines This can also be supported by the results herein comparing the selectively bred lines with Wistar rats where the P/NP rats were more similar to the Wistar rats than the HAD/LAD lines. However, there is evidence of considerable variation within Wistar rats which are dependent on supplier origin, e.g., in behavior and voluntary alcohol intake (Palm et al., 2011a,b, 2012; Goepfrich et al., 2013; Momeni et al., 2015; Wood et al., 2016). Considering there has been over 40 years of breeding since the establishment of the P/NP lines (Lumeng et al., 1977), it is not certain that the similarities between P/NP and Wistar rats herein are due to shared lineage. Unfortunately, the N/NIH rats are no longer commercially available and are only maintained in small research colonies, therefore we were unable to include them as additional comparisons in this study. Due to the different foundations stocks, different genes for different behavioral traits are likely to have co-segregated with the genes for alcohol intake and preference in the different line-pairs (e.g., Tabakoffet al., 2009; McBride et al., 2012, 2013; Saba et al., 2015). The present results of differences between HAD1 and HAD2 rats, as opposed to previous research, may be a recent development between the now separate breeding colonies for the line-pairs or represent a January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org Frontiers in Neuroscience | www.frontiersin.org 12 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. previously undetected difference in adolescent behavior that has been present throughout the history of these rat lines. Thus, studies with different selectively bred line-pairs can be expected to aid in the process of tailoring treatments, including pharmacological, to different subtypes of AUDs and thereby increasing treatment success rates. In a previous study characterizing alcohol-naïve, adult male behavior in the MCSF, the Finnish AA and Sardinian sP lines (not included in this study) constituted opposing behavioral extremes, while the P and HAD1/2 lines were intermediate and more similar (Roman et al., 2012). The adolescent behavior of the P and HAD1/2 lines seems to be more dissimilar than their adult counterparts. Interestingly, the lines from the short-term selective breeding based on alcohol consumption in adolescence also demonstrated a clear difference in the adolescent behavioral profile (Fernández et al., 2020), where the high alcohol-consuming line seems to be most similar to the HAD2 profile in the current study. ETHICS STATEMENT The animal study was reviewed and approved by the Institutional Animal Care and Use Committee of the Indiana University School of Medicine, Indianapolis, United States. Behavioral Profiles of the HAD and LAD Replicate Rat Lines This highlights the importance to study the alcohol-naïve adolescent behavior of several high alcohol-consuming lines to investigate the full spectrum of associated phenotypes, i.e., subtypes of AUDs modeled by these selected lines. FUNDING This study was funded by grants from NIH/NIAAA (U24AA013522 and U24AA015512, to RB). SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnins. 2021.811401/full#supplementary-material ACKNOWLEDGMENTS We acknowledge technical assistance from Ms. Erin Jarvis and Mr. Nelson Tobar, and technical support from the Uppsala University Behavioral Facility (UUBF), Disciplinary Domain of Medicine and Pharmacy, Uppsala University. AUTHOR CONTRIBUTIONS Importantly, individual rat line-pairs selectively bred for high or low voluntary alcohol intake and preference do not necessarily represent the full spectrum of AUDs. A corollary is that differences found within any individual line-pair does not necessarily represent all phenotypic differences observed between individuals who are family history positive (FHP) or negative (FHN) for AUDs. Instead, the selectively bred rat lines appear to represent individual and/or subgroup differences, primarily within individuals who are FHP for AUD, which support the clinical diagnosis of AUDs as a spectrum disorder rather than an either-or phenomenon (Crabbe et al., 2010; Froehlich, 2010). Overall, the current findings support the heterogeneity of genotypes and phenotypes associated with a risk for developing AUDs. This stems from our observation that there does not appear to be a common behavioral “pathway” linking these selectively bred rat lines, especially during adolescence. Nevertheless, the behavioral profiles of the high alcohol-consuming lines can, as outlined above, provide important information about different AUD subtypes. For instance, the behavioral profile of the P rat has some parallels with clinical findings observed in individuals with a denser family history of AUDs, including risk-taking behavior and an earlier onset of AUD behavior (Cloninger et al., 1996; Moss et al., 2007). In contrast, the HAD2 behavioral profile may be interpreted as resembling individuals with higher negative emotionality, which has been associated with a vulnerability to develop AUDs and/or SUDs with later onset (Cloninger et al., 1996; Merikangas et al., 1998; Moss et al., 2007). ER, RB, and SL were responsible for the study concept and design, provided critical revision of the manuscript for important intellectual content, and contributed to interpretation of findings. SL contributed to the data collection, performed the data analysis, and drafted the manuscript. RB provided resources and funding. All authors critically reviewed content and approved the final version for publication. DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. Badishtov, B. A., Overstreet, D. H., Kashevskaya, O. P., Viglinskaya, I. V., Kampov-Polevoy, A. B., Seredenin, S. B., et al. (1995). To drink or not to drink: open field behavior in alcohol-preferring and-nonpreferring rat strains. Physiol. Behav. 57, 585–589. doi: 10.1016/0031-9384(94)0 0299-k Babor, T. F. (1992). Types of alcoholics, I: evidence for an empirically derived typology based on indicators of vulnerability and severity. Arch. Gen. Psychiatry 49:599. doi: 10.1001/archpsyc.1992.01820080007002 REFERENCES L., and Ehlers, C. L. (2010). Human and laboratory rodent low response to alcohol: is better consilience possible? Addict. Biol. 15, 125–144. doi: 10.1111/j.1369-1600.2009.00191.x Dyr, W., and Kostowski, W. (2008). Warsaw high-preferring (WHP) and Warsaw low-preferring (WLP) lines of rats selectively bred for high and low voluntary ethanol intake: preliminary phenotypic characterization. Alcohol 42, 161–170. doi: 10.1016/j.alcohol.2008.01.008 Meyerson, B. J., Augustsson, H., Berg, M., and Roman, E. (2006). The Concentric Square Field: a multivariate test arena for analysis of explorative strategies. Behav. Brain Res. 168, 100–113. doi: 10.1016/j.bbr.2005.10.020 Fernández, M. S., Báez, B., Bordón, A., Espinosa, L., Martínez, E., and Pautassi, R. M. (2017). Short-term selection for high and low ethanol intake yields differential sensitivity to ethanol’s motivational effects and anxiety- like responses in adolescent Wistar rats. Prog. Neuropsychopharmacol. Biol. Psychiatry 79, 220–233. doi: 10.1016/j.pnpbp.2017.06.027 Momeni, S., Segerström, L., and Roman, E. (2015). Supplier-dependent differences in intermittent voluntary alcohol intake and response to naltrexone in Wistar rats. Front. Neurosci. 9:424. doi: 10.3389/fnins.2015.00424 Moss, H. B., Chen, C. M., and Yi, H. (2007). Subtypes of alcohol dependence in a nationally representative sample. Drug Alcohol Depend. 91, 149–158. doi: 10.1016/j.drugalcdep.2007.05.016 Fernández, M. S., Bellia, F., Ferreyra, A., Chiner, F., Jiménez García, A. M., D’Addario, C., et al. (2020). Short-term selection for high and low ethanol intake during adolescence exerts lingering effects in stress-induced ethanol drinking and yields an anxiety-prone phenotype. Behav. Brain Res. 380, 112445. doi: 10.1016/j.bbr.2019.112445 Murphy, J. M., Stewart, R. B., Bell, R. L., Badia-Elder, N. E., Carr, L. G., McBride, W. J., et al. (2002). Phenotypic and genotypic characterization of the indiana university rat lines selectively bred for high and low alcohol preference. Behav. Genet. 32, 363–388. doi: 10.1023/a:1020266306135 Fritz, C. O., Morris, P. E., and Richler, J. J. (2012). Effect size estimates: current use, calculations, and interpretation. J. Exp. Psychol. Gen. 141, 2–18. doi: 10.1037/ a0024338 National Research Council (2011). Guide for the Care and Use of Laboratory Animals, 8th Edn. Washington, DC: National Academies Press, doi: 10.17226/ 12910 Froehlich, J. C. (2010). What aspects of human alcohol use disorders can be modeled using selectively bred rat lines? Subst. Use Misuse 45, 1727–1741. doi: 10.3109/10826084.2010.482424 Noguchi, K., Gel, Y. R., Brunner, E., and Konietschke, F. (2012). nparLD: an r software package for the nonparametric analysis of longitudinal data in factorial experiments. J. Stat. Softw. 50, 1–23. doi: 10.18637/jss.v050.i12 experiments. J. Stat. Softw. 50, 1–23. doi: 10.18637/jss.v050.i12 Goepfrich, A. REFERENCES 13:41. doi: 10.3389/fnbeh.2019.00041 Bell, R. L., Rodd, Z. A., Boutwell, C. L., Hsu, C. C., Lumeng, L., Murphy, J. M., et al. (2004). Effects of long-term episodic access to ethanol on the expression of an alcohol deprivation effect in low alcohol–consuming rats. Alcohol. Clin. Exp. Res. 28, 1867–1874. doi: 10.1097/01.alc.0000148101.20547.0a Behav. Neurosci. 13:41. doi: 10.3389/fnbeh.2019.00041 McBride, W. J., Kimpel, M. W., McClintick, J. N., Ding, Z.-M., Hyytia, P., Colombo, G., et al. (2012). Gene expression in the ventral tegmental area of 5 pairs of rat lines selectively bred for high or low ethanol consumption. Pharmacol. Biochem. Behav. 102, 275–285. doi: 10.1016/j.pbb.2012.04.016 Bell, R. L., Rodd, Z. A., Engleman, E. A., Toalston, J. E., and McBride, W. J. (2014). Scheduled access alcohol drinking by alcohol-preferring (P) and high-alcohol- drinking (HAD) rats: modeling adolescent and adult binge-like drinking. Alcohol 48, 225–234. doi: 10.1016/j.alcohol.2013.10.004 McBride, W. J., Kimpel, M. W., McClintick, J. N., Ding, Z.-M., Hyytia, P., Colombo, G., et al. (2013). Gene expression within the extended amygdala of 5 pairs of rat lines selectively bred for high or low ethanol consumption. Alcohol 47, 517–529. doi: 10.1016/j.alcohol.2013.08.004 Bell, R. L., Sable, H. J. K., Colombo, G., Hyytia, P., Rodd, Z. A., and Lumeng, L. (2012). Animal models for medications development targeting alcohol abuse using selectively bred rat lines: neurobiological and pharmacological validity. Pharmacol. Biochem. Behav. 103, 119–155. doi: 10.1016/j.pbb.2012.07.007 McBride, W. J., Rodd, Z. A., Bell, R. L., Lumeng, L., and Li, T.-K. (2014). The alcohol-preferring (P) and high-alcohol-drinking (HAD) rats–animal models of alcoholism. Alcohol 48, 209–215. doi: 10.1016/j.alcohol.2013.09.044 Cloninger, C. R., Sigvardsson, S., and Bohman, M. (1996). Type I and Type II alcoholism: an update. Alcohol Health Res. World 20, 18–23. McKinzie, D. L., McBride, W. J., Murphy, J. M., Lumeng, L., and Li, T.-K. (2002). Effects of amphetamine on locomotor activity in adult and juvenile alcohol- preferring and -nonpreferring rats. Pharmacol. Biochem. Behav. 71, 29–36. doi: 10.1016/s0091-3057(01)00610-4 Colombo, G., Lobina, C., Carai, M. A. M., and Gessa, G. L. (2006). Phenotypic characterization of genetically selected sardinian alcohol-preferring (sP) and - non-preferring (sNP) rats. Addict. Biol. 11, 324–338. doi: 10.1111/j.1369-1600. 2006.00031.x Merikangas, K. R., Mehta, R. L., Molnar, B. E., Walters, E. E., Swendsen, J. D., Aguilar-Gaziola, S., et al. (1998). Comorbidity of substance use disorders with mood and anxiety disorders: results of the international consortium in psychiatric epidemiology. Addict. Behav. 23, 893–907. doi: 10.1016/s0306- 4603(98)00076-8 Crabbe, J. C., Bell, R. REFERENCES Badishtov, B. A., Overstreet, D. H., Kashevskaya, O. P., Viglinskaya, I. V., Kampov-Polevoy, A. B., Seredenin, S. B., et al. (1995). To drink or not to drink: open field behavior in alcohol-preferring and-nonpreferring rat strains. Physiol. Behav. 57, 585–589. doi: 10.1016/0031-9384(94)0 0299-k Babor, T. F. (1992). Types of alcoholics, I: evidence for an empirically derived typology based on indicators of vulnerability and severity. Arch. Gen. Psychiatry 49:599. doi: 10.1001/archpsyc.1992.01820080007002 January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org 13 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. Bell, R. L., Franklin, K. M., Hauser, S. R., and Engleman, E. A. (2013). “Next stop dependence-binge drinking on the road to alcoholism: preclinical findings on its neurobiology from rat animal models,” in Binge Eating And Binge Drinking: Psychological, Social And Medical Implications, ed. S. B. Harris (Hauppauge, NY: Nova Publishers), 1–60. Karlsson, O., and Roman, E. (2016). Dose-dependent effects of alcohol administration on behavioral profiles in the MCSF test. Alcohol 50, 51–56. doi: 10.1016/j.alcohol.2015.10.003 Kranzler, H. R., and Rosenthal, R. N. (2003). Dual diagnosis: alcoholism and co- morbid psychiatric disorders. Am. J. Addict. 12, s26–s40. doi: 10.1111/j.1521- 0391.2003.tb00494.x Bell, R. L., Hauser, S. R., Liang, T., Sari, Y., Maldonado-Devincci, A., and Rodd, Z. A. (2017). Rat animal models for screening medications to treat alcohol use disorders. Neuropharmacology 122, 201–243. doi: 10.1016/j.neuropharm.2017. 02.004 Lesch, O. M., and Walter, H. (1996). Subtypes of alcoholism and their role in therapy. Alcohol Suppl. 31, 63–67. Lumeng, L., Hawkins, T., and Li, T. (1977). “New strains of rats with alcohol preference and non-preference,” in Alcohol and Aldehyde Metabolizing Systems, eds R. Thurman, J. Williamson, H. Drott, and B. Chance (Cambridge, MA: Academic Press), 537–544. Lumeng, L., Hawkins, T., and Li, T. (1977). “New strains of rats with alcohol preference and non-preference,” in Alcohol and Aldehyde Metabolizing Systems, Bell, R. L., Hauser, S., Rodd, Z. A., Liang, T., Sari, Y., McClintick, J., et al. (2016). “A genetic animal model of alcoholism for screening medications to treat addiction,” in International Review of Neurobiology, eds R. L. Bell and S. Rahman (Cambridge, MA: Academic Press), 179–261. doi: 10.1016/bs.irn.2016.02.017 preference and non-preference, in Alcohol and Aldehyde Metabolizing Systems, eds R. Thurman, J. Williamson, H. Drott, and B. Chance (Cambridge, MA: Academic Press), 537–544. Lundberg, S., Högman, C., and Roman, E. (2019). Adolescent exploratory strategies and behavioral types in the multivariate concentric square field(TM) test. Front. Behav. Neurosci. REFERENCES doi: 10.1186/1741-7007-7-70 j Palm, S., Roman, E., and Nylander, I. (2012). Differences in basal and ethanol- induced levels of opioid peptides in Wistar rats from five different suppliers. Peptides 36, 1–8. doi: 10.1016/j.peptides.2012.04.016 Verhulst, B., Neale, M. C., and Kendler, K. S. (2015). The heritability of alcohol use disorders: a meta-analysis of twin and adoption studies. Psychol. Med. 45, 1061–1072. doi: 10.1017/s0033291714002165 Viglinskaya, I. V., Overstreet, D. H., Kashevskaya, O. P., Badishtov, B. A., Kampov- Polevoy, A. B., Seredenin, S. B., et al. (1995). To drink or not to drink: tests of anxiety and immobility in alcohol-preferring and alcohol-nonpreferring rat strains. Physiol. Behav. 57, 937–941. doi: 10.1016/0031-9384(94)00 368-f Pandey, S. C., Zhang, H., Roy, A., and Xu, T. (2005). Deficits in amygdaloid cAMP-responsive element–binding protein signaling play a role in genetic predisposition to anxiety and alcoholism. J. Clin. Invest. 115, 2762–2773. doi: 10.1172/jci24381 Quintanilla, M. E., Israel, Y., Sapag, A., and Tampier, L. (2006). The UChA and UChB rat lines: metabolic and genetic differences influencing ethanol intake. Addict. Biol. 11, 310–323. doi: 10.1111/j.1369-1600.2006.00030.x Wille-Bille, A., Miranda-Morales, R. S., Pucci, M., Bellia, F., D’Addario, C., and Pautassi, R. M. (2018). Prenatal ethanol induces an anxiety phenotype and alters expression of dynorphin & nociceptin/orphanin FQ genes. Prog. Neuropsychopharmacol. Biol. Psychiatry 85, 77–88. doi: 10.1016/j.pnpbp.2018. 04.005 R Core Team (2019). R: A Language And Environment For Statistical Computing. Vienna: R Foundation for Statistical Computing. Reilly, M. T., Noronha, A., Goldman, D., and Koob, G. F. (2017). Genetic studies of alcohol dependence in the context of the addiction cycle. Neuropharmacology 122, 3–21. doi: 10.1016/j.neuropharm.2017.01.017 Wood, C. M., Nicolas, C. S., Choi, S.-L., Roman, E., Nylander, I., Fernandez- Teruel, A., et al. (2016). Prevalence and influence of cys407∗Grm2 mutation in Hannover-derived Wistar rats: mGlu2 receptor loss links to alcohol intake, risk taking and emotional behaviour. Neuropharmacology 115, 128–138. doi: 10.1016/j.neuropharm.2016.03.020 Rodd, Z. A., Bell, R. L., McKinzie, D. L., Webster, A. A., Murphy, J. M., Lumeng, L., et al. (2004). Low-dose stimulatory effects of ethanol during adolescence in rat lines selectively bred for high alcohol intake. Alcohol. Clin. Exp. Res. 28, 535–543. doi: 10.1097/01.alc.0000122107.08 417.d0 Zhang, H., Sakharkar, A. J., Shi, G., Ugale, R., Prakash, A., and Pandey, S. C. (2010). Neuropeptide Y signaling in the central nucleus of amygdala regulates alcohol- drinking and anxiety-like behaviors of alcohol-preferring rats. Alcohol. Clin. Exp. Res. 34, 451–461. doi: 10.1111/j.1530-0277.2009.01109.x Roman, E., and Colombo, G. (2009). REFERENCES A., Gluch, C., Friemel, C. M., and Schneider, M. (2013). Behavioral differences in three Wistar Han rat lines for emotional reactivity, cognitive processing and ethanol intake. Physiol. Behav. 110–111, 102–108. doi: 10.1016/ j.physbeh.2012.12.019 Nowak, K. L., Ingraham, C. M., Mckinzie, D. L., Mcbride, W. J., Lumeng, L., Li, T.-K., et al. (2000). An Assessment of novelty-seeking behavior in alcohol- preferring and nonpreferring rats. Pharmacol. Biochem. Behav. 66, 113–121. doi: 10.1016/s0091-3057(00)00206-9 Gore-Langton, J. K., Werner, D. F., and Spear, L. P. (2021). The effects of age, sex, and handling on behavioral parameters in the multivariate concentric square fieldTM test. Physiol. Behav. 229, 113243. doi: 10.1016/j.physbeh.2020.1 13243 Overstreet, D. H., Halikas, J. A., Seredenin, S. B., Kampov-Polevoy, A. B., Viglinskaya, I. V., Kashevskaya, O., et al. (1997). Behavioral similarities and differences among alcohol-preferring and -nonpreferring rats: confirmation by factor analysis and extension to additional groups. Alcohol. Clin. Exp. Res. 21, 840–848. Hwang, B. H., Stewart, R., Zhang, J.-K., Lumeng, L., and Li, T.-K. (2004). Corticotropin-releasing factor gene expression is down-regulated in the central nucleus of the amygdala of alcohol-preferring rats which exhibit high anxiety: a comparison between rat lines selectively bred for high and low alcohol preference. Brain Res. 1026, 143–150. doi: 10.1016/j.brainres.2004.08.042 Palm, S., Hävermark, Å, Meyerson, B. J., Nylander, I., and Roman, E. (2011a). When is a Wistar a Wistar? Behavioral profiling of outbred Wistar rats from five different suppliers using the MCSF test. Appl. Anim. Behav. Sci. 135, 128–137. doi: 10.1016/j.applanim.2011.08.010 January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org 14 Behavioral Profiles After Alcohol-Consumption Selection Lundberg et al. Palm, S., Momeni, S., Lundberg, S., Nylander, I., and Roman, E. (2014). Risk- assessment and risk-taking behavior predict potassium- and amphetamine- induced dopamine response in the dorsal striatum of rats. Front. Behav. Neurosci. 8:236. doi: 10.3389/fnbeh.2014.00236 Stewart, R. B., Gatto, G. J., Lumeng, L., Li, T.-K., and Murphy, J. M. (1993). Comparison of alcohol-preferring (P) and nonpreferring (NP) rats on tests of anxiety and for the anxiolytic effects of ethanol. Alcohol 10, 1–10. doi: 10.1016/ 0741-8329(93)90046-q Palm, S., Roman, E., and Nylander, I. (2011b). Differences in voluntary ethanol consumption in Wistar rats from five different suppliers. Alcohol 45, 607–614. doi: 10.1016/j.alcohol.2010.11.005 Tabakoff, B., Saba, L., Printz, M., Flodman, P., Hodgkinson, C., Goldman, D., et al. (2009). Genetical genomic determinants of alcohol consumption in rats and humans. BMC Biol. 7:70. Frontiers in Neuroscience | www.frontiersin.org January 2022 | Volume 15 | Article 811401 REFERENCES Lower risk taking and exploratory behavior in alcohol-preferring sP rats than in alcohol non-preferring sNP rats in the multivariate concentric square field (MCSF) test. Behav. Brain Res. 205, 249– 258. doi: 10.1016/j.bbr.2009.08.020 Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Roman, E., Meyerson, B. J., Hyytiä, P., and Nylander, I. (2007). The multivariate concentric square field test reveals different behavioural profiles in male AA and ANA rats with regard to risk taking and environmental reactivity. Behav. Brain Res. 183, 195–205. doi: 10.1016/j.bbr.2007.06.009 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Roman, E., Stewart, R. B., Bertholomey, M. L., Jensen, M. L., Colombo, G., Hyytiä, P., et al. (2012). Behavioral profiling of multiple pairs of rats selectively bred for high and low alcohol intake using the MCSF test. Addict. Biol. 17, 33–46. doi: 10.1111/j.1369-1600.2011.00327.x Saba, L. M., Flink, S. C., Vanderlinden, L. A., Israel, Y., Tampier, L., Colombo, G., et al. (2015). The sequenced rat brain transcriptome – its use in identifying networks predisposing alcohol consumption. FEBS J. 282, 3556–3578. doi: 10. 1111/febs.13358 Copyright © 2022 Lundberg, Roman and Bell. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Sommer, W., Hyytiä, P., and Kiianmaa, K. (2006). The alcohol-preferring AA and alcohol-avoiding ANA rats: neurobiology of the regulation of alcohol drinking. Addict. Biol. 11, 289–309. doi: 10.1111/j.1369-1600.2006.00037.x January 2022 | Volume 15 | Article 811401 Frontiers in Neuroscience | www.frontiersin.org Frontiers in Neuroscience | www.frontiersin.org 15
https://openalex.org/W2605151894
https://dergipark.org.tr/tr/download/article-file/328840
English
null
Uromyces behenis (DC.) UNGER İÇIN YENI BIR KONAKÇI TÜR
Trakya University journal of natural science/Trakya university journal of natural sciences
2,017
cc-by
1,197
Anahtar kelimeler: Uromyces, pas mantarları, yeni konak, Türkiye. Anahtar kelimeler: Uromyces, pas mantarları, yeni konak, Türkiye. Results Rust species recorded so far on Silene as the host belong to the genera Aecidium Pers., Puccinia Pers. and Uromyces (Link) Unger. Uromyces is known to include 12 species worldwide which rely on Silene species and 5 of these species were determined in Turkey by Bahçecioğlu & Kabaktepe (2012). The identification of the fungal samples showed that the fungus sampled on S. echinospermoides was Uromyces behenis (DC.) Unger, Einfluss des Bodenstc.: 216 (1836). (Fig. 1). Macro- and micromorphological features of the fungus were given below. Aecidia usually hypophyllous, seated on spots that vary in size and yellow or purple. It is very conspicuous, solitary or clustered, cup shaped, whitish yellow. Aecidispores densely and minutely verruculose, yellowish, 15-21μm in diam. Uredinia aecidioid. Teleutosori hypophyllous and on the stems, often mixed with uredinia, scattered, gregarious or circinate, covered by epidermis, brownish black or black. Teleutospores subgloboid or obovoid, rounded above 25-37 × 19-27μm, wall smooth, pale brown, up to 11μm at apex, pore apical, pedicels up to 80μm, faintly yellowish, persistent. In the present study, a new host Silene species, S. echinospermoides, was reported for the rust fungus Uromyces behenis, contributing to its host range of. http://dergipark.gov.tr/trkjnat Trakya University Journal of Natural Sciences, 18(1): 39-40, 2017 ISSN 2147-0294, e-ISSN 2528-9691 DOI: 10.23902/trkjnat.268602 http://dergipark.gov.tr/trkjnat Trakya University Journal of Natural Sciences, 18(1): 39-40, 2017 ISSN 2147-0294, e-ISSN 2528-9691 DOI: 10.23902/trkjnat.268602 Research Article/Araştırma Makalesi Introduction Microphotographs of the spores were taken under a light microscope (Noveks B series 1000). The current names of fungi were given according to www.indexfungorum.org. Names of host plants and families are given according to http://www.theplantlist.org. Voucher specimens are deposited in Inönü University Herbarium (INU). Silene L. is one of the largest genera of flowering plants in the world, comprising ca. 750 species of which approximately half occur in the Mediterranean basin. The southern parts of Balkan Peninsula and southwest Asia are the two main centers for the genus (Greuter 1995). Silene is represented in Turkey by 31 sections and 138 species, of which 42% are endemic taxa (Genç et al. 2007). Hasan AKGÜL1, Mustafa SEVİNDİK1, Şanlı KABAKTEPE*2, Ayşen ÖZÇANDIR1, Candan KEMALOĞLU AYKURT1, Ilgaz AKATA3 Hasan AKGÜL1, Mustafa SEVİNDİK1, Şanlı KABAKTEPE*2, Ayşen ÖZÇANDIR1, Candan KEMALOĞLU AYKURT1, Ilgaz AKATA3 1Akdeniz University, Science Faculty, Department of Biology, Antalya. 2Inonu University, Battalgazi Vocat Sch, TR-44210 Battalgazi, Malatya. 3Ankara University, Science Faculty, Department of Biology, Ankara. *Corresponding author: e-mail: sanli.kabaktepe@inonu.edu.tr Received (Alınış): 24 November 2016, Accepted (Kabul): 16 March 2017, Online First (Erken Görünüm): 8 April 2017, Published (Basım): 15 June 201 ceived (Alınış): 24 November 2016, Accepted (Kabul): 16 March 2017, Online First (Erken Görünüm): 8 April 2017, Published (Ba Abstract: In this study, Silene echinospermoides Hub.-Mor. (Caryophyllaceae) is reported as a new host species for the rust fungus Uromyces behenis (DC.) Unger. A short description, distribution and photographs of macro- and micromorphological features of the fungus were given. Key words: Uromyces, rust fungi, new host, Turkey. Uromyces behenis (DC.) Unger İçin Yeni Bir Konakçı Tür Özet: Bu çalışmada, Silene echinospermoides Hub.-Mor. (Caryophyllaceae), Uromyces behenis (DC.) Unger pas mantarı türü için yeni bir konakçı tür olarak rapor edilmiştir. Bu fungusun kısa deskripsiyonu, yayılışı ve makro ve mikromorfolojilerine ait fotoğrafları verilmiştir. Discussion Uromyces behenis, known as the yellow champion rust on Silene, is a common rust fungi widespread in Europe and Asia and was recorded so far on 25 different Materials and Methods Fungi samples and the host plants were collected in 2016 from Antalya province in Turkey. The host specimens were prepared according to the conventional herbarium techniques. Host plants were identified using the Flora of Turkey and the East Aegean Islands (Cullen & Coode 1967). Fungal spores were scraped from dried host specimens and mounted in lactophenol. H. Akgül et al. 40 H. Akgül et al. Fig. 1. Images of the rust fungus Uromyces behenis and the host species Silene echinospermoides. A- a general view of the dried herbarium specimen of the host; B- stereo microscopic view of the infected host plant; C- light microscopic view of Aecidiaspores;D- light microscopic view of Teleutospores. Fig. 1. Images of the rust fungus Uromyces behenis and the host species Silene echinospermoides. A- a general view of the dried herbarium specimen of the host; B- stereo microscopic view of the infected host plant; C- light microscopic view of Aecidiaspores;D- light microscopic view of Teleutospores. Table 1. Host range and distribution of Uromyces behenis in Turkey. Table 1. Host range and distribution of Uromyces behenis in Turkey. Host Distribution in Turkey References Silene spp. Kahramanmaraş (Bahcecioglu et al. 2006, Gobelez 1962) Silene alba Muhl. ex Rohrb. Kars, Ardahan (Bahçecioğlu & Kabaktepe 2012) Silene echinospermoides Hub.-Mor. Antalya The record reported in the present study Silene italica (L.) Pers. Sivas (Bahcecioglu & Yıldız 2005) Silene marschallii C. A. Mey. Erzurum (Henderson 1964, Tamer et al. 1989) Silene vulgaris (Moench) Garcke Sivas, Ordu (Bahcecioglu & Yıldız 2005, Kabaktepe & Bahcecioglu 2006). Silene echinospermoides Hub.-Mor. Silene italica (L.) Pers. Silene marschallii C. A. Mey. Silene vulgaris (Moench) Garcke Silene species (Farr & Rossman 2016). It was also reported on Oberna cserei (Caryophyllaceae) Ikonn. from Ukraine (Dudka et al. 2004). The yellow champion name comes from aecidia on the host plant. U. behensis is known found to infect 5 Silene taxa in Turkey (Table 1). As a result of this study, the number of Silene species serving as a host for U. behenis reached to 26. Specimen examined: Silene echinospermoides Hub.- Mor. (Caryophyllaceae), Turkey, Antalya, Kumluca, Yukarı Alakır, 1200m. 30/05/2016, A. Özcandir 1454. References USDA. Retrieved November 11, from http://nt.ars- grin.gov/fungaldatabases 1. Bahcecioglu, Z. & Yildiz, B. 2005. A study on the microfungi of Sivas Province. Turkish Journal of Botany, 29: 23–44. 7. Genç, G.E., Kandemir, A. & Genç, İ. 2007. A new species of Silene (Caryophyllaceae) from east Anatolia, Turkey. Nordic Journal of Botany, 25: 58–63. 2. Bahcecioglu, Z., Kabaktepe, S. & Yildiz, B. 2006. Microfungi isolated from plants in Kahramanmaras Province, Turkey. Turkish Journal of Botany, 30: 419–434. 8. Gobelez, M. 1962. La mycoflore de Turquie 1. Mycopathologia Applicata. 19(4): 296–314. 3. Bahçecioğlu, Z. & Kabaktepe, Ş. 2012. Checklist of rust fungi in Turkey. Mycotaxon, 119: 494. 9. Greuter, W. 1995. Silene L. (Caryophyllaceae) in Greece: A subgeneric and sectional Classification. Taxon, 44: 543– 581. 4. Cullen, M.J.E. & Coode, J. 1967. Silene L. In (Ed.) Davis, P.H., Flora of Turkey and the East Aegean Islands. Edinburgh University Press, Edinburgh. Volume 2, 179- 242. 10. Henderson, D.M. 1964. Uredinales from S.W. Asia III. Notes from the Royal Botanic Garden, Edinburgh. 25: 197– 277. 5. Dudka, I.O., Heluta, V.P., Tykhonenko, Y.Y., Andrianova, T.V., Hayova, V.P., Prydiuk, M.P., Dzhagan, V.V., & Isikov, V.P. 2004. Fungi of the Crimean Peninsula. M.G. Kholodny Institute of Botany, National Academy of Sciences of Ukraine. 11. Kabaktepe, S. & Bahcecioglu, Z. 2006. Microfungi identified from the Flora of Ordu Province in Turkey. Turkish Journal of Botany, 30: 251–265. 12. Tamer, A.U., Altan, Y. & Gucin, F. 1989. Gulveren koyu (Erzurum Senkaya) florasında belirlenen bazı parazit funguslar. Anadolu Üniversitesi Fen-Edebiyat Fakültesi Dergisi, 1(2): 45–55. 6. Farr, D.F. & Rossman, A.Y. 2016. Fungal Databases, Systematic Mycology and Microbiology Laboratory. ARS,
https://openalex.org/W3194326708
https://huj.uoh.edu.iq/index.php/HUJ/article/download/huj-10355/98
Persian
null
Analysis of the OPEC oil price crisis during the period between the years 1961 - 2019)
Govarî zankoy Hełebce/Govarî zankoy Helebce
2,020
cc-by
16,542
303 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq (٢019-1961( شیكردنەوەی قەیراᣍن نرᣐن نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍن ، ساسان حكمت عىل ٢، حميد عبدەللا صالح ٣1 سلوى توفیق محمد جوگرافیا ، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ کۆلرᢀێ ژى ئاداب ، بەس2,1 زانستەكۆمەᢼڵ يەتيەكان، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ كۆلرێ ژى پەروەردەى بنەڕەᣍت ، بەس3 Corresponding author's e-mail: salwa.muhammed@su.edu.krd 303 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq (٢019-1961( شیكردنەوەی قەیراᣍن نرᣐن نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍن ، ساسان حكمت عىل ٢، حميد عبدەللا صالح ٣1 سلوى توفیق محمد جوگرافیا ، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ کۆلرᢀێ ژى ئاداب ، بەس2,1 زانستەكۆمەᢼڵ يەتيەكان، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ كۆلرێ ژى پەروەردەى بنەڕەᣍت ، بەس3 Corresponding author's e-mail: salwa.muhammed@su.edu.krd 303 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq (٢019-1961( شیكردنەوەی قەیراᣍن نرᣐن نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍن ، ساسان حكمت عىل ٢، حميد عبدەللا صالح ٣1 سلوى توفیق محمد جوگرافیا ، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ کۆلرᢀێ ژى ئاداب ، بەس2,1 زانستەكۆمەᢼڵ يەتيەكان، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ كۆلرێ ژى پەروەردەى بنەڕەᣍت ، بەس3 Corresponding author's e-mail: salwa.muhammed@su.edu.krd ، ساسان حكمت عىل ٢، حميد عبدەللا صالح ٣1 سلوى توفیق محمد جوگرافیا ، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ کۆلرᢀێ ژى ئاداب ، بەس2,1 زانستەكۆمەᢼڵ يەتيەكان، زانکۆى سەالحەددین، هه ولرᢀێ ڕ، هه رێىم كوردستان، عرᢀێ اق ژ كۆلرێ ژى پەروەردەى بنەڕەᣍت ، بەس3 Corresponding author's e-mail: salwa.muhammed@su.edu.krd Summary Analysis of the OPEC oil price crisis during the period between the years 1961 - 2019( Analysis of the OPEC oil price crisis during the period between the years 1961 - 2019( since oil was discovered in a commercial to date Competitor acquiring، did not lose the oil and economic im- portance at the international level، in particular the global level in general، because the oil with close ties to the economy of the consuming and producing countries in addition to the inability of the world to find a source of renewable and non-renewable sources of energy، both to replace oil. Because of its importance in the world economy، which made the oil market is not restricted or legally binding rules of the market، but affected by a range of political factors contradictory gradual disengagement strategy between the big industrialized countries and consuming countries، underdeveloped and productive، there is no doubt that this Working Group was the Republic of the main influential in oil prices. Thus our view of the WORTHIEST to address this issue through research patterns opec((Oil Prices Mixed during an extended period between the years 1961 - 2019( with a view to know the pace of oil prices during the period mentioned in order to achieve this we must talk about the method of determining the price of oil with the shed light on the prices before the founding of the Organization of )opec(، and during the time period divided into several phases of time study with the analysis of the price of oil during each stage، at the end of the research reached a set of conclusions which may be of use to researchers in determining the fluctuation of oil prices during the specified period of time. گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 گۆڤاری زانكۆی هه ڵه بجه :گۆڤارێی زانسیت ئه كادیمیه زانكۆی هه ڵه بجه دە ری دە كات به رگ )2020( ساᣡڵ4 ژمارە5 ڕێككه وته كان2020/12/30: ڕێككه وᣍت بᣀڵ وكردنــه وە|2020/10/ڕێككــه وᣍت په ســه ندكردن:٦| 2020//٩12:ڕێكــه وᣍت وە رگرتــن ئیمه ییل توێژە رsalwa.muhammed@su.edu.krd ماᣚغ چاپ و بᣀڵ و كردنه وە م. سلوى توفیق محمد، م. ساسان حكمت عیل، پ.ی.د. ملخص وعليــه راينــا مــن االجــدر التطــرق ایل هــذا الموضــوع مــن خــالل البحــث الموســوم ب(تذبــذب اســعار نفــط ئوثــك خــالل الرئيــیی ᣚغ ) بهــدف معرفــة وتــرێ ة اســعار النفــط خــالل الفــرت ة المذكــورة والجــل تحقيــق ذلــك البــد مــن التحــدث 201 – ٩1٦٩1 فــرت ة ممتــدة بــᣌێ غ الســنوات عــن اســلوب تحديــد ســعر النفــط مــع تســليط الضــوء عــیل اســعارە قبــل تاســيس منظمــة ئوبيــك ، ومــن خالاللفــرت ة الزمنيــة قســمنا الدراســة ایل عــدة ی تحديد ی ختام توصل البحث ایل مجموعة من االستنتاجات الیت ی قد تفيد الباحثᣌێ غ ᣚغ مراحل زمنية مع تحليل سعر النفط خالل كل مرحلة ، وᣚغ تذبــذب اســعار النفــط خــالل الفــرت ة الزمنيــة المحــددة . پوختەی لێكۆڵینەوە لــه گه ڵدابێــت نــه وت گرنــیگ ئابــوری لــه ســه ر ئاســیت جیهــاᣍغ و ژ لــه ســه ردە یم دۆزینــه وە ی نــەوەت بــه شــێوە یه یک بازرگانیانــه تاکــوو ئه مــڕۆس هه رێــىم دا هه یــه و هه روە هــا په یوە نــدە یک به تیــیغ بــه ئابــوری دە وڵه تــاᣍغ به کاربــه ر و به رهه مهێنــه ر هه یــه و تــا ئێســتاش هیــچ ســه رچاوە یه یک وزە جــا نــوێ بــووە وە بــن یــا خــود نــوێ نه بــووە وە نه یانتوانیــوە کێرب ڕکــیᢀ ێ بکــه ن. لــه گه ڵدابێــت نــه وت گرنــیگ ئابــوری لــه ســه ر ئاســیت جیهــاᣍغ و ژ لــه ســه ردە یم دۆزینــه وە ی نــەوەت بــه شــێوە یه یک بازرگانیانــه تاکــوو ئه مــڕۆس هه رێــىم دا هه یــه و هه روە هــا په یوە نــدە یک به تیــیغ بــه ئابــوری دە وڵه تــاᣍغ به کاربــه ر و به رهه مهێنــه ر هه یــه و تــا ئێســتاش هیــچ ســه رچاوە یه یک وزە جــا نــوێ بــووە وە بــن یــا خــود نــوێ نه بــووە وە نه یانتوانیــوە کێرب ڕکــیᢀ ێ بکــه ن. ئــه و گرنگیــه گه ورە یــه ی کــه وا نــه وت لــه ابــوری جیهــاᣍغ دا هه یــه ᣍت وایکــردوە کــه بــازاڕی نــه وت ئــازاد نه بێــت و ته نهــا وابه ســته ی یاســاکاᣍغ بــازاڕ نه بێــت به ڵکــو کاریگــه رە بــه کۆمه ڵــه فاکته رێــی ســیاس و ســرت اترێ ژی دژ به یــه ک لــه نێــوان دە وڵه تــاᣍغ پیشه ســازی گــه ورە ی به کاربــه ری نــه وت و هه روە هــا دە وڵه تــاᣍغ دواکه وتــووی به رهه مهێنــه ری کــه بێگومــان ئه مــه کاری کردوە تــه ســه ر نــرᣐغ نــه وت. )یه ئامانجســتنه ڕووی نــرᣐغ 201٩-1٦٩1(ئــه م توێژینه وە مــان کــه بــه ناونیشــاᣍغ (هه ڵکشــان و داکشــاᣍغ نــرᣐغ نــه وᣍت ئۆپیــک لــه مـاوە ی نێــوان ســاᢼڵᣍغ نه وتــه لــه و ماوە یــه دا. لــە رێــگای بەکارهێنــاᣍغ هــەردوو میتــودی وەســیغ و شــیکردنەوە داتــا وزانیارییــەکان .لــه ســه رە تا دا بــاس لــه شــێوازی دیاریکــردᣍغ نــرᣐغ نــه وت دە که یــن لــه گــه ڵ تیشــک خســتنه ســه ر نــرᣐغ نــه وت پێــش دامه زرانــدᣍغ ڕێکخــراوی ئۆپیــک دواتــر مــاوە ی نێــوان ســاᢼڵᣍغ لێکۆڵینه وە مــان دابــه ش کــردوە بــۆ چه نــد قۆناغێــک لــه گــه ڵ شــیکردنه وە ی نــرᣐغ نــه وت لــه و قۆناغانــه دا و لــه کۆتاییشــدا چه ندیــن دە رئه نجاممــان خســتوە ته ڕوو. ملخص )201 – ٩1٦٩1 (تذبذب اسعار نفط ئوبك خالل فرت ة ممتدة بᣌێ غ السنوات منــذ اكتشــاف النفــط بشــكل تجــاري وحــیت وقتنــا الحاظــر، لــم يفقــد النفــط اهميتــه االقتصاديــة عــیل المســتوى الــدویلی بشــكل خــاص والمســتوى العالــىمی بشــكل عــام، وذلــك النالنفــط لــه عالقــة وطيــدة باقتصــاد الــدول المســتهلكة والمنتجــة اضافــة ایل عــدم قــدرة العالــم عــیل ايجــاد مصــدر مــن مصــادر الطاقــة ســواء المتجــددة والغــرێ المتجــددة ليحــل محــل النفــط . )201 – ٩1٦٩1 (تذبذب اسعار نفط ئوبك خالل فرت ة ممتدة بᣌێ غ السنوات منــذ اكتشــاف النفــط بشــكل تجــاري وحــیت وقتنــا الحاظــر، لــم يفقــد النفــط اهميتــه االقتصاديــة عــیل المســتوى الــدویلی بشــكل خــاص والمســتوى العالــىمی بشــكل عــام، وذلــك النالنفــط لــه عالقــة وطيــدة باقتصــاد الــدول المســتهلكة والمنتجــة اضافــة ایل عــدم قــدرة العالــم عــیل ايجــاد مصــدر مــن مصــادر الطاقــة ســواء المتجــددة والغــرێ المتجــددة ليحــل محــل النفــط . ی االقتصــاد العالــىمی الــذي جعــل مــن ســوق النفــط غــرێ مقيــد او ملــزم بقواعــد الســوق ولكنهــا متاثــرة بمجموعــة عوامــل ونظــرا الهميتــه الكبــرێ ة ᣚغ سياســية وســرت اتيجية متناقضــة بــᣌێ غ الــدول الصناعيــة الكبــرێ ة والمســتهلكة والــدول المتخلفــة والمنتجــة ، والشــك فيــة ان هــذا العامــل كان المؤثــر ی اســعار النفــط. وعليــه راينــا مــن االجــدر التطــرق ایل هــذا الموضــوع مــن خــالل البحــث الموســوم ب(تذبــذب اســعار نفــط ئوثــك خــالل الرئيــیی ᣚغ ) بهــدف معرفــة وتــرێ ة اســعار النفــط خــالل الفــرت ة المذكــورة والجــل تحقيــق ذلــك البــد مــن التحــدث 201 – ٩1٦٩1 فــرت ة ممتــدة بــᣌێ غ الســنوات عــن اســلوب تحديــد ســعر النفــط مــع تســليط الضــوء عــیل اســعارە قبــل تاســيس منظمــة ئوبيــك ، ومــن خالاللفــرت ة الزمنيــة قســمنا الدراســة ایل عــدة ی تحديد ی ختام توصل البحث ایل مجموعة من االستنتاجات الیت ی قد تفيد الباحثᣌێ غ ᣚغ مراحل زمنية مع تحليل سعر النفط خالل كل مرحلة ، وᣚغ تذبــذب اســعار النفــط خــالل الفــرت ة الزمنيــة المحــددة . ی االقتصــاد العالــىمی الــذي جعــل مــن ســوق النفــط غــرێ مقيــد او ملــزم بقواعــد الســوق ولكنهــا متاثــرة بمجموعــة عوامــل ونظــرا الهميتــه الكبــرێ ة ᣚغ سياســية وســرت اتيجية متناقضــة بــᣌێ غ الــدول الصناعيــة الكبــرێ ة والمســتهلكة والــدول المتخلفــة والمنتجــة ، والشــك فيــة ان هــذا العامــل كان المؤثــر ی اســعار النفــط. https://www.huj.uoh.edu.iq ملخص Summary حميد عبدەللا صالح، گه یشᣌت غ به م توێژینه وە یه 2020 © CCBY-NC_ND 4.0 كراوە یه له ژێر ڕە زامه ندی گۆڤاری زانكۆی هه ڵه بجه :گۆڤارێی زانسیت ئه كادیمیه زانكۆی هه ڵه بجه دە ری دە كات به رگ )2020( ساᣡڵ4 ژمارە5 ڕێككه وته كان2020/12/30: ڕێككه وᣍت بᣀڵ وكردنــه وە|2020/10/ڕێككــه وᣍت په ســه ندكردن:٦| 2020//٩12:ڕێكــه وᣍت وە رگرتــن ئیمه ییل توێژە رsalwa.muhammed@su.edu.krd ماᣚغ چاپ و بᣀڵ و كردنه وە م. سلوى توفیق محمد، م. ساسان حكمت عیل، پ.ی.د. حميد عبدەللا صالح، گه یشᣌت غ به م توێژینه وە یه 2020 © CCBY-NC_ND 4.0 كراوە یه له ژێر ڕە زامه ندی Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 304 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq پوختەی لێكۆڵینەوە ووشەى كليل:(نەوت،بەبازاركردن،ئەوبيك،) پێشەیک : ئــه و گرنگیــه گه ورە یــه ی کــه وا نــه وت لــه ابــوری جیهــاᣍغ دا هه یــه ᣍت وایکــردوە کــه بــازاڕی نــه وت ئــازاد نه بێــت و ته نهــا وابه ســته ی یاســاکاᣍغ بــازاڕ نه بێــت به ڵکــو کاریگــه رە بــه کۆمه ڵــه فاکته رێــی ســیاس و ســرت اترێ ژی دژ به یــه ک لــه نێــوان دە وڵه تــاᣍغ پیشه ســازی گــه ورە ی به کاربــه ری نــه وت و هه روە هــا دە وڵه تــاᣍغ دواکه وتــووی به رهه مهێنــه ری کــه بێگومــان ئه مــه کاری کردوە تــه ســه ر نــرᣐغ نــه وت. )یه ئامانجســتنه ڕووی نــرᣐغ 201٩-1٦٩1(ئــه م توێژینه وە مــان کــه بــه ناونیشــاᣍغ (هه ڵکشــان و داکشــاᣍغ نــرᣐغ نــه وᣍت ئۆپیــک لــه مـاوە ی نێــوان ســاᢼڵᣍغ نه وتــه لــه و ماوە یــه دا. لــە رێــگای بەکارهێنــاᣍغ هــەردوو میتــودی وەســیغ و شــیکردنەوە داتــا وزانیارییــەکان .لــه ســه رە تا دا بــاس لــه شــێوازی دیاریکــردᣍغ نــرᣐغ نــه وت دە که یــن لــه گــه ڵ تیشــک خســتنه ســه ر نــرᣐغ نــه وت پێــش دامه زرانــدᣍغ ڕێکخــراوی ئۆپیــک دواتــر مــاوە ی نێــوان ســاᢼڵᣍغ لێکۆڵینه وە مــان دابــه ش کــردوە بــۆ چه نــد قۆناغێــک لــه گــه ڵ شــیکردنه وە ی نــرᣐغ نــه وت لــه و قۆناغانــه دا و لــه کۆتاییشــدا چه ندیــن دە رئه نجاممــان خســتوە ته ڕوو. یووشەى كليل:(نەوت،بەبازاركردن،ئەوبيك،) ) لە كۆی بەكاربردᣍغ وزەی جیهاᣍغ پێك دەهێیغᢀ ئەمە جگە %3 نزیكەی (٧ ژ نەوت گرنگرت ین سەرچاوەی وزەیە لە جیهاندا و پشكەكەس لــەوەی كــە كەرەســتەی خــاوی چەندیــن پیشەســازییە. داهــاᣍت نەوتیــش ڕێژەیــەیک بــەرز لــە داهــاᣍت دەوڵەتــاᣍغ هەناردەكــەری نــەوت پێــك دەهێنێــت بــە جۆرێــك كــە ئابــوری ئــەو دەوڵەتانــەی بــە شــێوەیەیک ســەرەیک پشــیت ᣍچᢀ دەبەســتێت هــەر گۆڕانكاریــەیک دیــار لــە نــرᣐغ جیهــاᣍغ نــەوت دا ڕووبــدات ڕەنگدانــەوەی لــە ســەر ئابــوری ئــەو دەوڵەتانــە و ئابــوری جیهــاᣍغ دەبێــت چونكــە ئــەم ســەرچاوەیە بــە بزوێنــەری ســەرەیک ئابــوری جیهــاᣍغ دادەنرێــت . گرنــیگ لێكۆڵینــەوە: پشــت بەســᣌت غ بــەو ڕاســتیانە دەكرێــت بووترێــت كــە هــەر ئــەو زانیاریانــەی كــە ئاماژەیــان بــۆ كــرا گرنــیگ بابــەᣍت لێكۆڵینەوەكــە دیــاری ).201٩-1٦٩1 دەكــەن كــە تایبەتــە بــە (شــیكردنەوەی هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــە نێــوان ســاᢼڵᣍغ کێشەی لێكۆڵینەوە :گرفیت لێكۆڵینەوەكە بریتیە لە ڕووداᣍغ هەڵكشان و داكشاᣍغ گەورە لە نرᣐغ نەوᣍت ئۆپیك دا كە هەندێك جار ڕەنگدانەوەی نەرێیغ هەبووە لە سەر ئەنداماᣍغ ئۆپێك و دەوڵەتاᣍغ تری هەناردەكەری نەوت لە دەرەوەی ئۆپیك . Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 305 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq پــالᣍن لێكۆڵینــەوە :بــۆ بەدیهێنــاᣍغ ئامانــیب لێكۆڵینەوەكــە و بــە هــۆی رسوشــیت بابەتەكــەوە بــە باشــمان زاᣍغ لێكۆڵینەوەكــە دابــەش بكەیــن بــۆ ئــەم تەوەرانــە : تەوەرانــە : تەوەری یەكەم : كورتەیەك دەربارەی ڕێكخراوی ئۆپیك و شێوازی دیاریكردᣍغ نرᣐغ نەوت پێش دامەزراندᣍغ ئۆپیك . .201٩-1٦٩1 تەوەری دووەم : دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ .201٩-1٦٩1 تەوەری سێیەم : شیكردنەوەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ هەروەها خستنەڕووی چەندین دەرئەنجام و ڕاسپاردە . تەوەری یەكەم : كورتەیەك دەربارەی ڕێكخراوی ئۆپیك و شێوازی دیاریكردᣍغ نرᣐغ نەوت پێش دامەزراندᣍغ ئۆپیك . .201٩-1٦٩1 تەوەری دووەم : دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ .201٩-1٦٩1 تەوەری سێیەم : شیكردنەوەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ .201٩-1٦٩1 ەوەری سێیەم : شیكردنەوەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ كورتەیــەك دەربــارەی ڕێكخــراوی ئۆپیــك و شــێوازی دیاریكــردᣍغ نــرᣐغ نــەوت پێــش دامەزرانــدᣍغ ئۆپیــك :لــەم تــەوەرەدا بــە كــورᣍت بــاس لــە دوو -1 : بابــەت دەكرێــت كــە بریتــᣌێ غ لــە Organization Of( لــە پیــیت یەكــەیم نــاوی وشــەكاᣍغ ڕێكخراوەكەیــە بــە زمــاᣍغ ئینگلــرێ غیOPEC ڕێكخــراوی ئۆپیــك :زاراوەی ئۆپیــك- 1 -1 ،) دەوڵــەᣍت بەرهەمهێنــەر و هەناردەكــەری نــەوت (عــرᢀێ اق5( ) ئــەم ڕێكخــراوە لــە ئەنجــایم كۆبوونــەوەیPetroleum Exporting Countries ) دامــەزرا . هــۆكاری راســتەوخۆی دامەزرانــدᣍغ ئۆپیــك دەگەڕێتــەوە بــۆ1٦٩0( كویــت، ســعوودیە، ئــرᢀێ ان، فەنزەویــال) لــە بەغــدا لــە ئەیلــویل ســاᣡڵ دا . https://www.huj.uoh.edu.iq : گریمانەی لێكۆڵینەوە :دەكرێت گریمانەكاᣍغ لێكۆڵینەوەكە لەم خاᢼڵ نەدا دیاری بكەین ڕاســتە نــرᣐغ نــەوت وەك هــەر كاᢼڵ یــەیک تــر خواســت و خســتنەڕوو دیــاری دەكات بــەᢼڵ م ئــەو هۆكارانــەی كــە كار دەكەنــە ســەر خواســت و -1 خســتنەڕووی نــەوت جیـاوازە لــەو هۆكارانــەی كــە كار دەكەنــە ســەر خواســت و خســتنەڕووی كاᢼڵ كاᣍغ تــر بــە هــۆی تایبەتمەنــدی و گرنــیگ نــەوت وەك ســەرچاوەیەیک ســەرەیک وزە و كەرەســتەیەیک خــاوی پیشەســازی گرنــگ. .1٦٩0 ڕۆᣡڵ ڕێكخراوی ئۆپیك لە دیاریكردᣍغ نرᣐغ نەوت دا گۆڕانكاری دیاری بەسەر دا هاتووە لە كاᣍت دروستبوونیەوە لە ساᣡڵ-2 .)201٩-1٦٩1( نرᣐغ نەوᣍت ئۆپیك هەڵكشان و داكشاᣍغ گەورەی بە خۆیەوە بینیوە لە نێوان ساᢼڵᣍغ-3 : گریمانەی لێكۆڵینەوە :دەكرێت گریمانەكاᣍغ لێكۆڵینەوەكە لەم خاᢼڵ نەدا دیاری بكەین ڕاســتە نــرᣐغ نــەوت وەك هــەر كاᢼڵ یــەیک تــر خواســت و خســتنەڕوو دیــاری دەكات بــەᢼڵ م ئــەو هۆكارانــەی كــە كار دەكەنــە ســەر خواســت و -1 خســتنەڕووی نــەوت جیـاوازە لــەو هۆكارانــەی كــە كار دەكەنــە ســەر خواســت و خســتنەڕووی كاᢼڵ كاᣍغ تــر بــە هــۆی تایبەتمەنــدی و گرنــیگ نــەوت وەك ســەرچاوەیەیک ســەرەیک وزە و كەرەســتەیەیک خــاوی پیشەســازی گرنــگ. : ئامانᣑب لێكۆڵینەوە :ئەوەی تایبەتە بە ئامانیب لێكۆڵینەوەكە دەكرێت لە دوو خاڵ دا چڕ بكرێتەوە .201٩-1٦٩1 دیاریكردᣍغ دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ-1 دیاریكردᣍغ ئەو هۆكارانەی كە نرᣐغ نەوᣍت ئۆپیك دیاری دەكەن-2 : ئامانᣑب لێكۆڵینەوە :ئەوەی تایبەتە بە ئامانیب لێكۆڵینەوەكە دەكرێت لە دوو خاڵ دا چڕ بكرێتەوە .201٩-1٦٩1 دیاریكردᣍغ دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ-1 : ئامانᣑب لێكۆڵینەوە :ئەوەی تایبەتە بە ئامانیب لێكۆڵینەوەكە دەكرێت لە دوو خاڵ دا چڕ بكرێتەوە .201٩-1٦٩1 دیاریكردᣍغ دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ-1 دیاریكردᣍغ ئەو هۆكارانەی كە نرᣐغ نەوᣍت ئۆپیك دیاری دەكەن-2 میتودی لێكۆڵینەوە : بە مەبەسیت پێکاᣍغ ئامانیب لێكۆڵینەوە هەوڵدراوە هەردوو میتودی وەسیغ و شیکردنەوە داتا وزانیارییەکان بەکاربێهرغ ێت دیاریكردᣍغ ئەو هۆكارانەی كە نرᣐغ نەوᣍت ئۆپیك دیاری دەكەن-2 میتودی لێكۆڵینەوە : بە مەبەسیت پێکاᣍغ ئامانیب لێكۆڵینەوە هەوڵدراوە هەردوو میتودی وەسیغ و شیکردنەوە داتا وزانیارییەکان بەکاربێهرغ ێت پــالᣍن لێكۆڵینــەوە :بــۆ بەدیهێنــاᣍغ ئامانــیب لێكۆڵینەوەكــە و بــە هــۆی رسوشــیت بابەتەكــەوە بــە باشــمان زاᣍغ لێكۆڵینەوەكــە دابــەش بكەیــن بــۆ ئــەم تەوەرانــە : پــالᣍن لێكۆڵینــەوە :بــۆ بەدیهێنــاᣍغ ئامانــیب لێكۆڵینەوەكــە و بــە هــۆی رسوشــیت بابەتەكــەوە بــە باشــمان زاᣍغ لێكۆڵینەوەكــە دابــەش بكەیــن بــۆ ئــەم تەوەرانــە : 1٦٩0-1٩5كەمكردنــەوەی نــرᣐغ نــەوت لــە الیــەن كۆمپانیــا نەوتیــە زەبەالحەكانــەوە لــە ســاڵەكاᣍغ ٩ بەرگریكردن لە بەرژەوەندییەكاᣍغ دەوڵەتاᣍغ بەرهەمهێنەر و هەناردەكەری نەوت .-1 )1٩2) لــە ســاᣡڵ (٧Achnacary Agreement(ســنورێك بــۆ دەســەᢼڵ تداری كارتــیل نــەوᣍت نێودەوڵــەᣍت كــە بــە پــیᢀ ێ ڕێككەوتننامــەی ئەكنــاكاری-2 .دامــەزرا بــوو گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 دلنیای كردنەوەی وڵەتاᣍغ بەرهەم هێنەر لەرووی جێگرێ ی ویەكسانەوە نرخ .-3 .بەردەوامبوون لەهەناردەكردن وپێداᣍغ نەوت بە وᢼڵ تاᣍغ بەكاربەر -4 .پێداᣍغ دڵنیاᣍی بەو وڵەتانەی وەبەرهێنانیان كردووە لە پیشەسازی نەوت-5 جگــە لــە پێنــج ئەنــدایم دامەزرێنــەری ئۆپیــك لــە ســاڵەكاᣍغ دواتــر چەندیــن دەوڵــەᣍت هەناردەكــەری نــەوت پەیوەندیــان بــە ڕێكخراوەكــە كــرد وەك لــە ) دەوڵەتــە .14( )دەردەكەوێــت، ئەمــرۆ دەوڵەتــاᣍغ ئەنــدایم ئۆپیــك ژمارەیــان1() و نەخشــەی ژمارە1( خشــتەی دلنیای كردنەوەی وڵەتاᣍغ بەرهەم هێنەر لەرووی جێگرێ ی ویەكسانەوە نرخ .-3 .بەردەوامبوون لەهەناردەكردن وپێداᣍغ نەوت بە وᢼڵ تاᣍغ بەكاربەر -4 .پێداᣍغ دڵنیاᣍی بەو وڵەتانەی وەبەرهێنانیان كردووە لە پیشەسازی نەوت-5 جگــە لــە پێنــج ئەنــدایم دامەزرێنــەری ئۆپیــك لــە ســاڵەكاᣍغ دواتــر چەندیــن دەوڵــەᣍت هەناردەكــەری نــەوت پەیوەندیــان بــە ڕێكخراوەكــە كــرد وەك لــە ) دەوڵەتــە .14( )دەردەكەوێــت، ئەمــرۆ دەوڵەتــاᣍغ ئەنــدایم ئۆپیــك ژمارەیــان1() و نەخشــەی ژمارە1( خشــتەی .بەردەوامبوون لەهەناردەكردن وپێداᣍ نەوت بە وᢼتاᣍ بەكاربەر -4 .پێداᣍغ دڵنیاᣍی بەو وڵەتانەی وەبەرهێنانیان كردووە لە پیشەسازی نەوت-5 جگــە لــە پێنــج ئەنــدایم دامەزرێنــەری ئۆپیــك لــە ســاڵەكاᣍغ دواتــر چەندیــن دەوڵــەᣍت هەناردەكــەری نــەوت پەیوەندیــان بــە ڕێكخراوەكــە كــرد وەك لــە ) دەوڵەتــە .14( )دەردەكەوێــت، ئەمــرۆ دەوڵەتــاᣍغ ئەنــدایم ئۆپیــك ژمارەیــان1() و نەخشــەی ژمارە1( خشــتەی Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq https://www.huj.uoh.edu.iq https://www.huj.uoh.edu -Volume 5, Issue 4, December 2020 )دەوڵەتاᣍغ ئەندام لە ڕێكخراوی ئۆپیك1( خشتەی ژ ت ڵە وەد ڵی سا بوون ە ئە ندام تێبینی 1 راق ێع 1960 رەنێ زر ە دام یت ڵە وەد 2 ت ەی كو1960 رەنێ زر ە دام یت ڵە وەد 3 سعود یە1960 رەنێ زر ە دام یت ڵە وەد 4 ران ێئ 1960 رەنێ زر ە دام یت ڵە وەد 5 نزو ەف ێلا 1960 رەنێ زر ە دام یت ڵە وەد 6 قەتەر 1961 7 لیبیا 1962 8 نوس ە ند ەئ یا 1962 ڵی سا ەل 2009 ندام ەئ ه ۆی خ یتێ ەڵپە سپ ارد و ل ە كانوون دوو ی ەمی 2016 ی ند ەو پەی كرد ەوە بە ك ۆپی ئ 9 ئی مارات 1967 10 ری زائ ەج1969 11 ای جر ەی ن1971 12 ئی كواد ۆر 1973 ڵی سا ەل 1992 ندام ەئ ه ۆی خ یتێ ەڵپە سپ ارد و ل ڵی سا ە 2007 ی ند ەو پەی كرد ك ۆپی ئ ەب ەوە 13 نۆ اب گ1973 - 1975 ڵی سا ەل 1973 بوو ب ندام ەئ ە ی هاوب ە ش و ل ڵی سا ە 1975 بوو ندام ەئ ە ەس ڕە ی ن و ل ڵی سا ە 1994 ندام ەئ ه ۆی خ یتێ ەڵپە سپ ارد و ل ە تە مووز ی 2016 ۆپی ئ ەب ی ند ەو پەی ك كرد ەوە 14 ال گۆ نەئ2007 ژ ت ڵە وەد ڵی سا بوون ە ندام ەئ تێبینی 1 راق ێع 1960 رەنێ زر ە دام یت ڵە وەد 2 ت ەی كو1960 رەنێ زر ە دام یت ڵە وەد 3 سعود یە1960 رەنێ زر ە دام یت ڵە وەد 4 ران ێئ 1960 رەنێ زر ە دام یت ڵە وەد 5 نزو ەف ێلا 1960 رەنێ زر ە دام یت ڵە وەد 6 قەتەر 1961 7 لیبیا 1962 8 نوس ە ند ەئ یا 1962 ڵی سا ەل2009 ندام ەئ ه ۆی خ یتێ ەڵپە سپ ارد و ل ە كانوون دوو ی ەمی 2016 ی ند ەو پەی كرد ەوە بە ك ۆپی ئ 9 ئی مارات 1967 10 ری زائ ەج1969 11 ای جر ەی ن1971 12 ئی كواد ۆر 1973 ڵی سا ەل1992 ندام ەئ ه ۆی خ یتێ ەڵپە سپ ارد و ل ڵی سا ە2007 ی ند ەو پەی كرد ك ۆپی ئ ەب ەوە 13 نۆ اب گ1973 - 1975 ڵی سا ەل 1973 بوو ب ندام ەئ ە ی هاوب ە ش و ل ڵی سا ە 1975 بوو ندام ەئ ە ەس ڕە ی ن و ل ڵی سا ە 1994 ندام ەئ ه ۆی خ یتێ ەڵپە سپ ارد و ل ە تە مووز ی 2016 ۆپی ئ ەب ی ند ەو پەی ك كرد ەوە ا Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 307 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq .88،ص1سەرچاوە:ئامادەكراوە بە پشت بەسᣌت غ بە: الخویلی ،٧٩٩ ) دا بۆمان دەردەكەوێت كە :2( ) ئەوا لە خشتەی2015( سەبارەت بە بەرهەمهێنان و هەناردەكردن و یەدەیکگ نەوᣍت ئۆپیك و جیهان لە ساᣡڵ ) %43( ڕێكخــراوی ئۆپیـك بــە ڕێژەیــەیک بــەرز بەشــداری دەكات لــە بەرهەمهێنــان و هەناردەكــردᣍغ نــەوت لــە جیهــان یــەك لــە دوای یــەك بــە ڕێــژەی-1 .)%5و (٦.٦ 1 سالم عبد الحسن رسن، اقتصاديات النفط، الجامعة المفتوحة، طرابمس ، ٩٩٩- : پشت بەسᣌت غ بە گ گکی ل گ گکی کی گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات uj p-ISSN: 2412-9607, e-ISSN: 2617-3360 55.ص1 سالم عبد الحسن رسن، اقتصاديات النفط، الجامعة المفتوحة، طرابمس ، ٩٩٩- : سەرچاوە:ئامادەكراوە بە پشت بەسᣌت غ بە ) ی یــەدەیکگ مســۆگەری جیهــان، بۆیــە پێــش بیــیغ دەكریــت ك%81( ڕێكخــراوی ئۆپیـك خـاوەᣍغ ڕێژەیــەیک زۆر بــەرزی یــەدەیکگ مســۆگەرە زیاتــر لــە--2 گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 55.ص1 سالم عبد الحسن رسن، اقتصاديات النفط، الجامعة المفتوحة، طرابمس ، ٩٩٩- : سەرچاوە:ئامادەكراوە بە پشت بەسᣌت غ بە ) ی یــەدەیکگ مســۆگەری جیهــان، بۆیــە پێــش بیــیغ دەكریــت كــە%81( ڕێكخــراوی ئۆپیـك خـاوەᣍغ ڕێژەیــەیک زۆر بــەرزی یــەدەیکگ مســۆگەرە زیاتــر لــە--2 55.ص1 سالم عبد الحسن رسن، اقتصاديات النفط، الجامعة المفتوحة، طرابمس ، ٩٩٩- : سەرچاوە:ئامادەكراوە بە پشت بەسᣌت غ بە ) ی یــەدەیکگ مســۆگەری جیهــان، بۆیــە پێــش بیــیغ دەكریــت كــە%81( ڕێكخــراوی ئۆپیـك خـاوەᣍغ ڕێژەیــەیک زۆر بــەرزی یــەدەیکگ مســۆگەرە زیاتــر لــە--2 55.ص1 سالم عبد الحسن رسن، اقتصاديات النفط، الجامعة المفتوحة، طرابمس ، ٩٩٩- : سەرچاوە:ئامادەكراوە بە پشت بەسᣌت غ بە ) ی یــەدەیکگ مســۆگەری جیهــان، بۆیــە پێــش بیــیغ دەكریــت كــە%81( ڕێكخــراوی ئۆپیـك خـاوەᣍغ ڕێژەیــەیک زۆر بــەرزی یــەدەیکگ مســۆگەرە زیاتــر لــە--2 Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات 308 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq لــە بــەر ئــەوەی والیەتەیەكگرتــووەكاᣍغ ئەمریــكا گەورەتریــن بەرهەمهێنــەر و هەناردەكــەری نــەوت بــووە تــا جەنــیگ جیهــاᣍغ دووەم بۆیــە شــێوازی دیاریكــردᣍغ نــرᣐغ نــەوت لــە جیهانــدا لــەم قۆناغــەدا قۆنــاᣘغ پێــش جەنــیگ جیهــاᣍغ دووەم بــە پلەیــەیک بــەرز كەوتبــووە ژێــر كاریگەریــەᣍت ئامانــج و بریــیت بــوو لــە دیاریكــردᣍغ نــرᣐغ نــەوᣍت خـاو لــە جیهانــدا بــە پــیᢀ ێ سیســتەیم یــەك ژ سیاســەتەكان و بەرژەوەندییــەكاᣍغ ئــەم وᢼڵ تــە، كــە دەرئەنجامەكــەس خــاᣡڵ بنــەڕەᣍت كــە كەنــداوی مەكســیك بــوو .ڕێككەوتننامــەی ئەكنــاكاری كــە پێشــرت ئاماژەمــان ᣍچᢀ دا جەخــیت لــە ســەر پەیرەوكــردᣍغ ئــەم سیســتەمە كــرد و نــرᣐغ نــەوᣍت لــە هــەر شــوێنێی جیهــان بــە گوێــرەی نــرᣐغ نــەوت لــە كەنــداوی مەكســیك دیــاری كــرد ᣍبᢀ گــوێ دان بــە شــوێن و تێچــووᣍغ بەرهەمهێنــان و تێچــووᣍغ گواســتنەوەی، ئەمــەش ئــەوە دەگەیەنێــت كــە نــرᣐغ نــەوت بــۆ نموونــە لــە بەنــدەری (مرســیلیا)ی فەرەنــی یەكســانە بــە نــرᣐغ نــەوت لــە كەنــداوی مەكســیك لــە گــەڵ نــرᣐغ گواســتنەوەی لــە كەنــداوی مەكســیكەوە بــۆ مرســیلیا، واتــە كڕیــاری نــەوت لــە هــەر خاڵێكــەوە نــەوᣍت هەنــاردە : دەكرێ شێوازی دیاریكردᣍغ نرᣐغ نەوت پێش دامەزراندᣍغ ڕێكخراوی ئۆپیك لەم خاᢼڵ نەدا دیار بكرێت ڕێككەوتــیغ كۆمپانیــا نەوتیــە گــەورەكان لــە (ئەكنــاكاری) كــە پێشــرت ئامــاژەی بــۆ كــرا تایبــەت بــوو بــە دیاریكــردᣍغ نــرᣐغ نــەوᣍت خــاو و دابەشــكردᣍغ -أ بــازاڕەكاᣍغ نــەوت بــە شــێوەیەك كــە لــە بەرژەوەنــدی كۆمپانیــاكان و ئــەو دەوڵەتانــە بێــت كــە كۆمپانیــاكان ســەر بەوانــن . ڕێككەوتــیغ (ئەكنــاكاری) بــووە مایــەی سیســتەمێك بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت كــە پــیᢀ ێ دەوتــرا (دیاریكــردᣍغ نــرᣐغ نــەوت بــە پــیᢀ ێ سیســتەیم یــەك خــاᣡڵ یان سیســتەیم كەنداوی مەكســیك)The Single Basing-Point System Of The Posted Prices( بنەڕەᣍت سیســتەیم كەنداوی مەكســیك ) وە بــە دیاریكــردᣍغ نــرᣐغ بەرمیلێــك لــە نــەوᣍت ئەمریــی بــە 1٩3لــە گــەڵ تێچــووᣍغ گواســتنەوە، كــە دەســت بــە جێبــەᣐبᢀ كــردᣍغ كــرا لــە ســاᣡڵ (٦ )دۆالر .بــە پــیᢀ ێ ئــەم سیســتەمە نــرᣐغ نــەوᣍت ئــەو دەوڵەتانــەی كــە دواتــر بــوون بــە ئەنـدایم ڕێكخــراوی ئۆپیـك گــرێ درابــوو بــەو ســەردەمەی كــە 1.0(٩ هەناردەكــەری نــەوت بوونــە (بــۆ نموونــە عــرᢀێ اق و فەنزەوێــال). https://www.huj.uoh.edu.iq HUJ-Volume 5, Issue 4, December 2020 .لــە ئاینــدە دا ڕۆᣡڵ لــە بــۆاری بەرهەمهێنــان و هەناردەكــردᣍغ نــەوت لــە جیهانــدا بەرزبێتــەوە 2015 ) بەرهەمهێنان و هەناردەكردن و یەدەیکگ نەوᣍت ئۆپیك و جیهان لە ساᣡڵ2( خشتەی . سەرچاوە:52.p;5.3 28and.p،p21،3.،٦3.1:Vienna،Austria،Tables ،201 ٦S بوار كان ە ك ۆپی ئ هان یج 1 -نان ێ مه ە ره ەب یۆ مل ن ل/ ی رم ەب ڕۆژ ە ان 32.32 75.08 %هان یج ی 43 100 2 -هە نارد ە كردن یۆ مل ن ل/ ی رم ەب ڕۆژ ە ان 23.57 41.65 %هان یج ی 56.6 100 3 -ەگی د یە مس ر ۆگە ی مل ار /ب لی رم ە1211 1493 %هان یج ی 81.2 100 . 52.p;5.3 28and.p،p21،3.،٦3.1:Vienna،Austria،Tables ،201 ٦Statistical Buletin Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University : شێوازی دیاریكردᣍغ نرᣐغ نەوت پێش دامەزراندᣍغ ئۆپیك-2-1 : دەكرێ شێوازی دیاریكردᣍغ نرᣐغ نەوت پێش دامەزراندᣍغ ڕێكخراوی ئۆپیك لەم خاᢼڵ نەدا دیار بكرێت ڕێككەوتــیغ كۆمپانیــا نەوتیــە گــەورەكان لــە (ئەكنــاكاری) كــە پێشــرت ئامــاژەی بــۆ كــرا تایبــەت بــوو بــە دیاریكــردᣍغ نــرᣐغ نــەوᣍت خــاو و دابەشــكردᣍغ -أ بــازاڕەكاᣍغ نــەوت بــە شــێوەیەك كــە لــە بەرژەوەنــدی كۆمپانیــاكان و ئــەو دەوڵەتانــە بێــت كــە كۆمپانیــاكان ســەر بەوانــن . ڕێككەوتــیغ (ئەكنــاكاری) بــووە مایــەی سیســتەمێك بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت كــە پــیᢀ ێ دەوتــرا (دیاریكــردᣍغ نــرᣐغ نــەوت بــە پــیᢀ ێ سیســتەیم یــەك خــاᣡڵ یان سیســتەیم كەنداوی مەكســیك)The Single Basing-Point System Of The Posted Prices( بنەڕەᣍت سیســتەیم كەنداوی مەكســیك ) وە بــە دیاریكــردᣍغ نــرᣐغ بەرمیلێــك لــە نــەوᣍت ئەمریــی بــە 1٩3لــە گــەڵ تێچــووᣍغ گواســتنەوە، كــە دەســت بــە جێبــەᣐبᢀ كــردᣍغ كــرا لــە ســاᣡڵ (٦ )دۆالر .بــە پــیᢀ ێ ئــەم سیســتەمە نــرᣐغ نــەوᣍت ئــەو دەوڵەتانــەی كــە دواتــر بــوون بــە ئەنـدایم ڕێكخــراوی ئۆپیـك گــرێ درابــوو بــەو ســەردەمەی كــە 1.0(٩ هەناردەكــەری نــەوت بوونــە (بــۆ نموونــە عــرᢀێ اق و فەنزەوێــال). https://www.huj.uoh.edu.iq لــە بــەر ئــەوەی والیەتەیەكگرتــووەكاᣍغ ئەمریــكا گەورەتریــن بەرهەمهێنــەر و هەناردەكــەری نــەوت بــووە تــا جەنــیگ جیهــاᣍغ دووەم بۆیــە شــێوازی دیاریكــردᣍغ نــرᣐغ نــەوت لــە جیهانــدا لــەم قۆناغــەدا قۆنــاᣘغ پێــش جەنــیگ جیهــاᣍغ دووەم بــە پلەیــەیک بــەرز كەوتبــووە ژێــر كاریگەریــەᣍت ئامانــج و بریــیت بــوو لــە دیاریكــردᣍغ نــرᣐغ نــەوᣍت خـاو لــە جیهانــدا بــە پــیᢀ ێ سیســتەیم یــەك ژ سیاســەتەكان و بەرژەوەندییــەكاᣍغ ئــەم وᢼڵ تــە، كــە دەرئەنجامەكــەس خــاᣡڵ بنــەڕەᣍت كــە كەنــداوی مەكســیك بــوو .ڕێككەوتننامــەی ئەكنــاكاری كــە پێشــرت ئاماژەمــان ᣍچᢀ دا جەخــیت لــە ســەر پەیرەوكــردᣍغ ئــەم سیســتەمە كــرد و نــرᣐغ نــەوᣍت لــە هــەر شــوێنێی جیهــان بــە گوێــرەی نــرᣐغ نــەوت لــە كەنــداوی مەكســیك دیــاری كــرد ᣍبᢀ گــوێ دان بــە شــوێن و تێچــووᣍغ بەرهەمهێنــان و تێچــووᣍغ گواســتنەوەی، ئەمــەش ئــەوە دەگەیەنێــت كــە نــرᣐغ نــەوت بــۆ نموونــە لــە بەنــدەری (مرســیلیا)ی فەرەنــی یەكســانە بــە نــرᣐغ نــەوت لــە كەنــداوی مەكســیك لــە گــەڵ نــرᣐغ گواســتنەوەی لــە كەنــداوی مەكســیكەوە بــۆ مرســیلیا، واتــە كڕیــاری نــەوت لــە هــەر خاڵێكــەوە نــەوᣍت هەنــاردە An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 309 HUJ-Volume 5, Issue 4, December 2020 بكردایــە دوور یــان نزیــك بــە جۆرێــك نــرᣐغ نــەوᣍت بــۆ دیــاری دەكــرا وەك ئــەوەی نەوتەكــە لــە كەنــداوی مەكســیكەوە هاتبێــت .پەیرەوكــردᣍغ ئــەم سیســتەمە بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت لــە جیهانــدا نــەك تەنیــا ئامانجــەكاᣍغ كۆمپانیــا نەوتیــە گــەورەكاᣍغ فەراهــەم دەكــرد بگــرە ئامانجــەكاᣍغ والیەتــە جێبەᣐبᢀ دەكرد بۆیە جێگەی ســەر ســووڕمان نیە كە دواڕۆژی پیشەســازی نەوت لەو ســەردەمە دا برییت بوو لە (ئەوەی ژ یەكگرتوەكاᣍغ ئەمریكاس لــە بەرژەوەنــدی ئەمریــكا دایــە لــە بەرژەوەنــدی كۆمپانیاكاندایــە و ئــەوەی لــە بەرژەوەنــدی كۆمپانیاكاندایــە لــە بەرژەوەنــدی ئەمریكادایە).(عفیــیغ ، ).240،ص1٧٧٩ سیســتەیم یــەك خــاᣡڵ بنــەڕەᣍت بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت داڕمــا كاتێــك كەشــتیگەیل بەڕیتــاᣍغ لــە جەنــیگ جیهــاᣍغ دووەم دا بــە ئــاگا هــات كــە نــرᣐغ ئــەو نەوتــەی كــە هـاوردەی دەكات ᣍبᢀ بنەمــا بەرزەبۆیــە فشــاری خســتە ســەر كۆمپانیــا نەوتیــەكان و ناچــار كــران دان بــە ناوچــەی كەنــداوی عەرەبیــش دا بنــᣌᢀێ غ وەك خــاᣡڵ بنــەڕەᣍت دووەم بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت، وە نــرᣐغ نــەوت لــە بەنــدەری عەبــاداᣍغ ئــرᢀێ ان دیاریكــرا لــە هەمــان ئاســیت نــرᣐغ نــەوᣍت كەنــداوی مەكســیك . دواتریــش حكومــەᣍت ئەمریـكا فشــاری خســتە ســەر كۆمپانیــاكان بــۆ ئــەوەی نــرᣐغ نــەوت لــە بەنــدەری( رأس تنــورە )ی ســعودی هەمــان نــرᣐغ نــەوت بێــت لــە كەنــداوی مەكســیك لــە ئەنجامــدا سیســتەمێی نــوێ بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت لــە دوای جەنــیگ جیهــاᣍغ دووەم پەیرەوكــرا كــە نارسابــوو بــە سیســتەیم دوو خــاᣡڵ بنــەڕەᣍت بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت. دۆزینــەوەی نــەوت بــە بڕێــی زۆر لــە دەوڵەتــاᣍغ ڕۆژهــەᢼڵᣍت ناوەڕاســت هــۆكاری ســەرەیک دانــاᣍغ سیســتەمێی نــوێ بــوو بــۆ دیاریكــردᣍغ نــرᣐغ -ب The Duel Basing-Point نــەوت لــە دوای جەنــیگ جیهــاᣍغ دووەم كــە نارسابــوو بــە (سیســتەیم دوو خــاᣡڵ بنــەڕەᣍت بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت ) بــە پــیᢀ ێ ئــەم سیســتەمە نوێیــە نــرᣐغ نــەوᣍت دەوڵەتــاᣍغ رۆژهــەᢼڵᣍت ناوەڕاســت كــە دواتــر بوونــە ئەنــدایم ئۆپیــك System Of The Posted Prices نزیــك بووەلــە نــرᣐغ نــەوت لــە خــاᣡڵ كەنــداوی مەكســیك چونكــە كڕیــاراᣍغ نــەوت دەوڵەتــاᣍغ ڕۆژهــەᢼڵᣍت ناوەڕاســتیان سەرپشــك كردبــوون كــە خــاᣡڵ دیاریكــردᣍغ نــرᣐغ نــەوت هەڵرب ژێــرن (كەنــداوی مەكســیك )یان(كەنــداوی عــەرەᣍب ) كامیــان نزیكــرت ە لێیانــەوە . و زی ر ێی ی رᣍب ) وی )ی ( ی وی رب ێر ( و ی ری رᣍ رᣐ ) لــە Equalization Pointكاتێــك كۆمپانیــا نەوتیــە گــەورەكان ڕازی بــوون بــەم سیســتەمە نوێیــە بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت( خــاᣡڵ یەكســانكردن نێــوان نــرᣐغ نــەوت لــە هــەر دوو كەنــداو (مەكســیك و عــەرەᣍب ) دیاریكــرا كــە بەنــدەری ناپــۆیل لــە باشــوری ئیتالیابــوو. https://www.huj.uoh.edu.iq HUJ-Volume 5, Issue 4, December 2020 309 HUJ-Volume 5, Issue 4, December 2020 هەرچەنــدە لــە پەنجــاكاᣍغ ســەدەی ڕابــردوو ڕێككەوتننامــەی (ســوود بــە نیــوەᣍی ) لــە نێــوان كۆمپانیــا نەوتیــە بیانیــەكان و دەوڵەتــاᣍغ بەرهەمهێنــەری -ج ) شــلینك وەردەگــرێ ا كــە یەكســان 4( نــەوت واژوو كــرا و داهــاᣍت دەوڵەتــاᣍغ هەناردەكــەر بــە پــیᢀ ێ نــرᣐغ ڕاگەیەنــراو دیاریكــرا كــە بــۆ هــەر یــەك تــەن نــەوت ) بــەᢼڵ م كۆمپانیــا گــەورەكاᣍغ نــەوت تاكــە الیــەن بــوون 225 ،ص1٧٩4 ،بــوو بــە بەهــای نیــو تــەن نــەوت بــە گوێــرەی نــرᣐغ ڕاگەیەنــراو .(عبدالوهــاب لــە دیاریكــردᣍغ تێچــووᣍغ نــەوت و گواســتنەوەی و دیاریكــردᣍغ نرخەكــەی و دەوڵەتــاᣍغ بەرهەمهێنــەری ڕۆڵێــی دیاریــان لــەم بــوارە دا نەبــوو بۆیــە دەرئەنجامــەكان وەك پێویســت لــە بەرژەوەنــدی دەوڵەتــاᣍغ بەرهەمهێنــەر نەبــوو . ) لــە Equalization Pointكاتێــك كۆمپانیــا نەوتیــە گــەورەكان ڕازی بــوون بــەم سیســتەمە نوێیــە بــۆ دیاریكــردᣍغ نــرᣐغ نــەوت( خــاᣡڵ یەكســانكردن نێــوان نــرᣐغ نــەوت لــە هــەر دوو كەنــداو (مەكســیك و عــەرەᣍب ) دیاریكــرا كــە بەنــدەری ناپــۆیل لــە باشــوری ئیتالیابــوو. هەرچەنــدە لــە پەنجــاكاᣍغ ســەدەی ڕابــردوو ڕێككەوتننامــەی (ســوود بــە نیــوەᣍی ) لــە نێــوان كۆمپانیــا نەوتیــە بیانیــەكان و دەوڵەتــاᣍغ بەرهەمهێنــەری -ج ) شــلینك وەردەگــرێ ا كــە یەكســان 4( نــەوت واژوو كــرا و داهــاᣍت دەوڵەتــاᣍغ هەناردەكــەر بــە پــیᢀ ێ نــرᣐغ ڕاگەیەنــراو دیاریكــرا كــە بــۆ هــەر یــەك تــەن نــەوت ) بــەᢼڵ م كۆمپانیــا گــەورەكاᣍغ نــەوت تاكــە الیــەن بــوون 225 ،ص1٧٩4 ،بــوو بــە بەهــای نیــو تــەن نــەوت بــە گوێــرەی نــرᣐغ ڕاگەیەنــراو .(عبدالوهــاب لــە دیاریكــردᣍغ تێچــووᣍغ نــەوت و گواســتنەوەی و دیاریكــردᣍغ نرخەكــەی و دەوڵەتــاᣍغ بەرهەمهێنــەری ڕۆڵێــی دیاریــان لــەم بــوارە دا نەبــوو بۆیــە دەرئەنجامــەكان وەك پێویســت لــە بەرژەوەنــدی دەوڵەتــاᣍغ بەرهەمهێنــەر نەبــوو . دەكرێت دەرئەنجایم شێوازەكاᣍغ دیاریكردᣍغ نرᣐغ نەوت پێش دامەزراندᣍغ ئۆپیك لەم خاᢼڵ نەدا چڕبكرێنەوە : لــە بــەر ئــەوەی نــرᣐغ نــەوت لــە الیــەن كۆمپانیــا گــەورەكان و ئــەو دەوڵەتانــەی كــە خـاوەᣍغ كۆمپانیــاكان بــوون دیــاری دەكــرا بۆیــە شــێوازەكاᣍغ -1 . دیاریكــردᣍغ نــرᣐغ نــەوت بــە شــێوەیەك دارێژرابــوو كــە لــە بەرژەوەنــدی ئــەو دەوڵەتانــە و كۆمپانیــا بیانیەكانــدا بــوو هەمیشــە نــرᣐغ نــەوᣍت دەوڵەتــاᣍغ ڕۆژهــەᢼڵᣍت ناوەڕاســت كــە دواتــر بــوون بــە ئەنــدایم ئۆپیــك كەمــرت بــووە لــە نــرᣐغ شایســتەی خــۆی و -2 لــە هــەر یــەك لــەABI بــە پێــوەری34- نــرᣐغ نــەوᣍت خــاوی ئەمریــی بــۆ نموونــە ئامــارەكاᣍغ تایبــەت بــە نــرᣐغ نــەوᣍت خــاوی ســووك كــە چڕییەكــەی .٩٧)ی نــرᣐغ نــەوت بــووە لــە كەنــداوی 1و52.2( كەنــداوی عــەرەᣍب و كەنــداوی مەكســیك بــەᢼڵ م نــرᣐغ نــەوت لــە كەنــداوی عــەرەᣍب لــە نێــوان ).3 مەكسیك(خشــتەی كۆمپانیــا گــەورەكان بــە پــیᢀ ێ بەرژەوەنــدی دەوڵەتەكانیــان گۆڕانكاریــان لــە دیاریكــردᣍغ نــرᣐغ نــەوت دا دەكــرد كــە دواتــر دابەزیــیغ نــرᣐغ نــەوت -3 . ) هۆكارێــی ڕاســتەوخۆ بــوو بــۆ دامەزرانــدᣍغ ئۆپیــك1٦٩0و1٩5لــە ســاڵەكاᣍغ (٩ Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات https://www.huj.uoh.edu.iq https://www.huj.uoh.edu.iq گواســتنەوە، كاریگــەری دیــاری لــە ســەر نــرᣐغ نــەوت هەیــە .لــەم بــوارەدا دەكــرێ بووترێــت كــە تاكــوو شــوێیغ هەناردەكــردᣍغ نــەوت نزیــك بێــت لــە بــازاڕەكاᣍغ نــەوت نــرᣐغ زیاتــر دەبێــت، دیــارە پێچەوانــەی ئــەم بۆچوونــەش ڕاســتە .بــۆ نموونــە دەڵــᣌᢀێ غ نــرᣐغ نــەوᣍت كەركــوك كــە لــە بەنــدەرەكاᣍغ ســەر ) 4دەریــای ناوەڕاســتەوە (بەنــدەركاᣍغ بانیــاس و تەڕابلــوس) هەنــاردە دەكرێــت بــە هــۆی نزیــی لــە ســەرەتای حەفتــاكاᣍغ ســەدەی ڕابــردوو بــە بــڕی (٩ ).5،ص٧1٩80،ســەنیت ئەمریــی زیاتــر بــووە لــە نــرᣐغ هەمــان جۆرلــە نــەوᣍت هەنــاردە كــراو لــە بەنــدەری (فــاو) لــە ســەر كەنــداوی عــەرەᣍب (حمــادي پشــت بەســᣌت غ بــەو ڕاســتیانەی كــە ئاماژەیــان بــۆ كــرا دەربــارەی كاریگەریــەᣍت جــۆر و شـوێیغ هەناردەكــردᣍغ نــەوت لــە ســەر نرخەكــەی و بــە هــۆی بــووᣍغ جیاوازی دیار لە جۆر و شـوێیغ هەناردەكردᣍغ نەوᣍت ئۆپیك بۆیە كاتێك باس لە دیاردەی هەڵكشــان و داكشــاᣍغ نرᣐغ نەوᣍت ئۆپیك دەكەین پشــت بــە تێكــڕای نرخــەكان دەبەســتᣌێ غ و جــۆر و شـوێیغ هەناردەكــردᣍغ نەوتەكــە بــەالوە دەنێــᣌێ غ بــە هــۆی : ئــەم جیاوازیانــە لــە جــۆر و شـوێیغ هەناردەكــردᣍغ نــەوᣍت ئۆپیـك دا نــەك تەنیــا لــە نێــوان ئەندامــاᣍغ ڕێكخراوەكــە دا هەیــە بەڵكــو تەنانــەت لــە جــۆری - أ نــەوᣍت تاكــە ئەندامێــی ئۆپیـك و شـوێیغ هەناردەكــردᣍغ نەوتەكــەی بــەدی دەكرێــت . ئــەم جیاوازیانــە لــە جــۆر و شـوێیغ هەناردەكــردᣍغ نــەوᣍت ئۆپیـك دا نــەك تەنیــا لــە نێــوان ئەندامــاᣍغ ڕێكخراوەكــە دا هەیــە بەڵكــو تەنانــەت لــە جــۆری - أ نــەوᣍت تاكــە ئەندامێــی ئۆپیـك و شـوێیغ هەناردەكــردᣍغ نەوتەكــەی بــەدی دەكرێــت . ئــەوەی دەكرێتــە پێوانــە و داهــاᣍت دەوڵەتــاᣍغ ئۆپیــی ᣍچᢀ دیــاری دەكرێــت نــرᣐغ نــەوᣍت خـاوی نیشــانەكراوی راگەیەنــراوە كــە تایبەتــە بــە نــرᣐغ نــەوᣍت -ب ).A.P.I( پلەیــە بــە پێــوەری34ســووك كــە چڕییەكـەی ) ی تایبــەت بــە تێكــڕای نــرᣐغ ڕاســتەقینەی نــەوᣍت ئۆپیــك و ئامــارەكاᣍغ تێكــڕای نــرᣐغ خــاوی نیشــانەكراوی ڕاگەیەنــراو 4( ئامــارەكاᣍغ خشــتەی ) جێگــرێ بــووە و یەكســان بــووە بــە 1٧٩0-1٦٩1( ئامــاژە دەكــەن كــە تێكــڕای نــرᣐغ خــاوی نیشــانەكراوی ڕاگەیەنــراو لــە شەســتەكاᣍغ ســەدەی ڕابــردوو ) دۆالر/بەرمیــل ئەویــش بــە هــۆی :1.80( )(پ ی ب پێو ر چڕیی وو ) ی تایبــەت بــە تێكــڕای نــرᣐغ ڕاســتەقینەی نــەوᣍت ئۆپیــك و ئامــارەكاᣍغ تێكــڕای نــرᣐغ خــاوی نیشــانەكراوی ڕاگەیەنــراو 4( ئامــارەكاᣍغ خشــتەی ) جێگــرێ بــووە و یەكســان بــووە بــە 1٧٩0-1٦٩1( ئامــاژە دەكــەن كــە تێكــڕای نــرᣐغ خــاوی نیشــانەكراوی ڕاگەیەنــراو لــە شەســتەكاᣍغ ســەدەی ڕابــردوو ) دۆالر/بەرمیــل ئەویــش بــە هــۆی :1.80( )1٦٩1( )دۆالر/بەرمیــل لــە ســاᣡڵ1.5تێكــڕای نــرᣐغ ڕاســتەقینەی ئۆپیـك لــە هەمــان مـاوەدا هەڵكشــان و داكشــاᣍغ كــەیم بەخۆیــەوە بینیــوە و لــە (٧-1 . HUJ-Volume 5, Issue 4, December 2020 HUJ-Volume 5, Issue 4, December 2020 لــە كەنــداوی عــەرەᣍب و كەنــداوی مەكسABI بــە پێوەری-34 ᣍ خــاوی ســووك كــە چڕییەكــەی )1٦٩0-1٩45( ڵ سا نرخ ی نە وت ل ە كە نداو ی عەرە ۆ ب(د الر/ب ل) ی رم ە نرخ ی نە وت ل ە كە نداو ی ی كس ەم ۆ ك(د الر/ب ل) ی رم ە % ڕێژەی 2:1 1 2 1945 1.05 1.36 77.2 1946 1.05 1.70 61.8 1947 1.40 2.68 52.2 1948 2.12 2.68 79.1 1949 1.88 2.68 70.1 1950 1.75 2.76 63.4 1951 1.75 2.76 63.4 1952 1.75 2.76 63.4 1953 1.86 2.76 64.6 1954 1.97 2.88 65.7 1955 1.97 3.00 65.7 1956 1.97 3.00 65.7 1957 2.12 3.25 65.2 1958 2.12 3.25 65.2 1959 1.94 3.14 61.8 1960 1.80 3.14 57.3 لــە كەنــداوی عــەرەᣍب و كەنــداوی مەكســیك لــە نێــوان ســاڵەكاᣍغABI بــە پێوەری-34 ) نــرᣐغ نــەوᣍت خــاوی ســووك كــە چڕییەكــەی3( خشــتەی )1٦٩0-1٩45( Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University ســەرچاوە:ئامادەكراوە بــە پشــت بەســᣌت غ بــە :نــواف نایــف اســماعیل،تحدید اســعار النفــط العــرᣍب الخــام ᣚغ الســوق العالمیــە،دار الرشــید ٦ .0 و جــدول (٦)ص48)ص5( ،جــدول1٩81،للنᣄژ ،برێ وت،لبنان )201٩-1٦٩1 ( قەیراᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساڵەكاᣍغ-2 : نەوت وەك هەموو كاᢼڵ یەیک تر نرخەكەی دەگۆڕێت بە گۆڕییغ جۆر(كوالێــیت ) : ســەبارەت بــە جــۆری نــەوت (كوالێــیت )، باشــرت ین و گرانرت یــن جــۆری نــەوت نــەوᣍت ســووكە كــە چڕییەكــەی بــە پــیᢀ ێ پێــوەری -1 -2 )Marker Crude( پلەیە.ئەم جۆرە نەوتە كە بە خاوی نیشانەكراو34))American petroleum institute)API(پەیمانگای پرت ڕۆᣡڵ ئەمریی ناودەبرێــت، باشــرت ین جــۆری نەوتــە و بەهــای نــەوᣍت خــاو كــەم دەكات بــە بەرزبوونــەوەی چڕییەكــەی ) واتــە ئــازاد لــە ســەر كەشــیت Free On Board –F.O.B( شــوێیغ جوگــراᣚغ هەناردەكــردن : شــوێیغ هەناردەكــردن نــارساوە بــە فــۆب- 2 -2 Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University ســەرچاوە:ئامادەكراوە بــە پشــت بەســᣌت غ بــە :نــواف نایــف اســماعیل،تحدید اســعار النفــط العــرᣍب الخــام ᣚغ الســوق العالمیــە،دار الرشــید ٦ .0 و جــدول (٦)ص48)ص5( ،جــدول1٩81،للنᣄژ ،برێ وت،لبنان )201٩-1٦٩1 ( قەیراᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساڵەكاᣍغ-2 : نەوت وەك هەموو كاᢼڵ یەیک تر نرخەكەی دەگۆڕێت بە گۆڕییغ جۆر(كوالێــیت ) : ســەبارەت بــە جــۆری نــەوت (كوالێــیت )، باشــرت ین و گرانرت یــن جــۆری نــەوت نــەوᣍت ســووكە كــە چڕییەكــەی بــە پــیᢀ ێ پێــوەری -1 -2 )Marker Crude( پلەیە.ئەم جۆرە نەوتە كە بە خاوی نیشانەكراو34))American petroleum institute)API(پەیمانگای پرت ڕۆᣡڵ ئەمریی ناودەبرێــت، باشــرت ین جــۆری نەوتــە و بەهــای نــەوᣍت خــاو كــەم دەكات بــە بەرزبوونــەوەی چڕییەكــەی ) واتــە ئــازاد لــە ســەر كەشــیت Free On Board –F.O.B( شــوێیغ جوگــراᣚغ هەناردەكــردن : شــوێیغ هەناردەكــردن نــارساوە بــە فــۆب- 2 -2 An Academic And Scientntific Journal Issued By Halabja University HUJ-Volume 5, Issue 4, December 2020 )201٩-1٦٩1( ) دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ4( خشتەی )201٩-1٦٩1( ) دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ4( خشتەی ڵ سا نرخ ی نە وت(د ۆ الر/ب ل) ی رم ە 1961 1.32 1979 29.19 1997 18.86 1962 1.33 1980 35.52 1998 12.28 1963 1.36 1981 34.00 1999 17.44 1964 1.42 1982 32.38 2000 27.6 1965 1.45 1983 29.04 2001 23.12 1966 1.50 1984 28.02 2002 24.36 1967 1.52 1985 27.01 2003 28.01 1968 1.57 1986 13.53 2004 36.05 1969 1.63 1987 17.73 2005 50.59 1970 1.21 1988 14.24 2006 61.00 1971 1.70 1989 17.31 2007 69.04 1972 1.82 1990 22.26 2008 94.10 1973 2.07 1991 18.62 2009 60.86 1974 11.00 1992 18.44 2010 77.38 1975 10.43 1993 16.33 2011 107.46 1976 11.60 1994 15.53 2012 109.87 1977 12.50 1995 16.86 2013 105.87 1978 12.79 1996 20.29 2014 96.29 2015 49.49 2016 55 ، 37 2017 65 2018 86 2019 59 ، 46 )201٩-1٦٩1( ) دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ4( خشتەی سەرچاوە :. in us .ddlars.per barrel 2015 to 1٦٩0 The Statistics Portal ;Averageannual OPEC Crude Oil From ) نــرᣐغ نــەوᣍت ئۆپیــك داكشــان و هەڵكشــاᣍغ بەخۆیــەوە دیــوە، دوای ئــەوەی لــە ســەرەتای ئــەم1٩٩0-1٩8لــە نیــوەی دووەیم هەشــتاكان (٦-ب ) دۆالر و بــە ڕێــژەی 13.4) دۆالر واتــە بــە بــڕی (٦13.53( دابــەزی بــۆ1٩8) دۆالرەوە لــە ســاᣡڵ ٦2.٧01( نیوەیــەدا داكشــاᣍغ گــەورە ڕوویــداوە و لــە ) 14.24( دا داكشــا بــۆ1٩88 ) دۆالر بــۆ هــەر بەرمیلێـك بـەᢼڵ م لــە ســاᣡڵ1٧.٧3( جارێــی تــر بەرزبوویــەوە بــۆ1٩8) كــەیم كــردوە و لــە ســاᣡڵ ٧%50( . ) دۆالر بــۆ هــەر بەرمیلێــك22.2دۆالر و لــە دوو ســاᣡڵ كۆتــاᣍی ئــەم نیوەیــە دیســان ڕووی لــە هەڵكشــان كــرد تــا گەیشــتە (٦ سەرچاوە :. in us .ddlars.per barrel 2015 to 1٦٩0 The Statistics Portal ;Averageannual OPEC Crude Oil From ) نــرᣐغ نــەوᣍت ئۆپیــك داكشــان و هەڵكشــاᣍغ بەخۆیــەوە دیــوە، دوای ئــەوەی لــە ســەرەتای ئــەم1٩٩0-1٩8لــە نیــوەی دووەیم هەشــتاكان (٦-ب ) دۆالر و بــە ڕێــژەی 13.4) دۆالر واتــە بــە بــڕی (٦13.53( دابــەزی بــۆ1٩8) دۆالرەوە لــە ســاᣡڵ ٦2.٧01( نیوەیــەدا داكشــاᣍغ گــەورە ڕوویــداوە و لــە ) 14.24( دا داكشــا بــۆ1٩88 ) دۆالر بــۆ هــەر بەرمیلێـك بـەᢼڵ م لــە ســاᣡڵ1٧.٧3( جارێــی تــر بەرزبوویــەوە بــۆ1٩8) كــەیم كــردوە و لــە ســاᣡڵ ٧%50( . https://www.huj.uoh.edu.iq ) لــەو ماوەیــەدا كــەیم كــردووە%23( )ســەنت و بــە ڕێــژەی3)واتــە بــە بــڕی (٦1٧٩0( )دۆالر/بەرمیــل لــە ســاᣡڵ1.21( داكشـاوە بــۆ ᣐبᢀ ی خۆیــەᣍت كــە بووترێــت كــە نــرᣐغ ڕاســتەقینەی نــەوت ئــەو نرخــە بــوو كــە كۆمپانیــا نەوتیــە بیانیــەكان نەوتیــان ᣍچᢀ دەفرۆشــت و داهــاᣍت -2 .)%25-20( ئەندامــاᣍغ ئۆپیــی ᣍچᢀ دیــاری دەكــرا لــە شەســتەكاᣍغ ســەدەی ڕابــردوو كەمــرت بــوو لــە نــرᣐغ خــاوی نیشــانەكراوی ڕاگەیەنــراو بــە ڕێــژەی )بــە پێچەوانــەی شەســتەكان نــرᣐغ نــەوᣍت ئۆپیــك هەڵكشــاᣍغ گــەورەی بەخۆیــەوە دی، لــەو ماوەیــەدا 1٩80-1٧٩1(لــە حەفتــاكاᣍغ ســەدەی ڕابردوو ) كــە زیاتــرە لــە 1٩80( ) دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ35.52( دا بەرزبوویــەوە بــۆ1٧٩1 )دۆالرەوە لــە ســاᣡڵ1.نــرᣐغ بەرمیلێــك نــەوت لــە (٧ .بــە گشــیت نــرᣐغ نــەوᣍت ئۆپیــك لــە حەفتاكانــدا ســاڵ لــە دوای ســاڵ جگــە لــە ســاᣡڵ 1٧٩1 )جــار هێنــدەی نــرᣐغ بەرمیلێــك لــە نــەوت لــە ســاᣡڵ20( )دۆالر بــووە بــۆ هــەر بەرمیلێـك كــە (٩) جــار زیاتــر بــووە لــە تێكــڕای نــرᣐغ نــەوت12.845( زیــادی كــردوە و تێكــڕای نــرᣐغ نــەوت لــەو ماوەیــەدا1٧٩5 لە شەستەكاندا ) داكشــان و هەڵكشــاᣍغ دیــار لــە نــرᣐغ نــەوᣍت ئۆپیــك دا ڕوویــداوە كــە دەكرێــت لــە ڕێــگای ئــەم 1٩٩0-1٩81( لــە هەشــتاكاᣍغ ســەدەی ڕابــردوو-3 : خاᢼڵ نــەوە دیــاری بكرێــت ) نــرᣐغ نــەوᣍت ئۆپیـك بــە بــەردەوایم ڕووی لــە داكشــان كــردوە، لــەو ماوەیــەدا نــرᣐغ نــەوᣍت ئۆپیــك 1٩85-1٩81( لــە نیــوەی یەكــەیم هەشــتایەكاندا-أ ) واتــە بــە نزیكــەی (٧) دۆالر و بــە ڕێــژەی 1٩85( ) دۆالر لــە ســاᣡڵ2.٧01( دا داكشــاوە بــۆ1٩81 )دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ34( لــە )دۆالر بــووە بــۆ هــەر بەرمیلێــك .30.05()كــەیم كــردوە و نــرᣐغ نــەوت لــە نیــوەی یەكــەیم هەشــتاكاندا%20.5(٦ Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 312 https://www.huj.uoh.edu.iq https://www.huj.uoh.edu.iq HUJ-Volume 5, Issue 4, December 2020 HUJ-Volume 5, Issue 4, December 2020 ) دۆالر بــۆ هــەر بەرمیلێــك22.2دۆالر و لــە دوو ســاᣡڵ كۆتــاᣍی ئــەم نیوەیــە دیســان ڕووی لــە هەڵكشــان كــرد تــا گەیشــتە (٦ ژ ئــەو زانیاریانــەی كــە ئاماژەیــان بــۆ كــرا دەریدەخــەن كــە لــە نیــوەی دووەیم هەشــتاكاᣍغ ســەدەی ڕابــردوو نــرᣐغ نــەوᣍت ئۆپیــك چەندیــن جــار تــووس ) دۆالر بــووە بــۆ هــەر بەرمیلێــك كــە 1.٧01( دیــاردەی داكشــان و هەڵكشــان بــووە و بــە گشــیت نــرᣐغ نــەوᣍت ئۆپیــك لــە نیــوەی دووەیم هەشــتاكان ) دۆالر.1كەمــرت ە لــە نــرᣐغ لــە نیــوەی یەكــەیم هەشــتاكاندا بــە زیاتــر لــە (٧ Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University سەرچاوە :. in us .ddlars.per barrel 2015 to 1٦٩0 The Statistics Portal ;Averageannual OPEC Crude Oil From ) نــرᣐغ نــەوᣍت ئۆپیــك داكشــان و هەڵكشــاᣍغ بەخۆیــەوە دیــوە، دوای ئــەوەی لــە ســەرەتای ئــەم1٩٩0-1٩8لــە نیــوەی دووەیم هەشــتاكان (٦-ب ) دۆالر و بــە ڕێــژەی 13.4) دۆالر واتــە بــە بــڕی (٦13.53( دابــەزی بــۆ1٩8) دۆالرەوە لــە ســاᣡڵ ٦2.٧01( نیوەیــەدا داكشــاᣍغ گــەورە ڕوویــداوە و لــە ) 14.24( دا داكشــا بــۆ1٩88 ) دۆالر بــۆ هــەر بەرمیلێـك بـەᢼڵ م لــە ســاᣡڵ1٧.٧3( جارێــی تــر بەرزبوویــەوە بــۆ1٩8) كــەیم كــردوە و لــە ســاᣡڵ ٧%50( . ) دۆالر بــۆ هــەر بەرمیلێــك22.2دۆالر و لــە دوو ســاᣡڵ كۆتــاᣍی ئــەم نیوەیــە دیســان ڕووی لــە هەڵكشــان كــرد تــا گەیشــتە (٦ ژ ئــەو زانیاریانــەی كــە ئاماژەیــان بــۆ كــرا دەریدەخــەن كــە لــە نیــوەی دووەیم هەشــتاكاᣍغ ســەدەی ڕابــردوو نــرᣐغ نــەوᣍت ئۆپیــك چەندیــن جــار تــووس ) دۆالر بــووە بــۆ هــەر بەرمیلێــك كــە 1.٧01( دیــاردەی داكشــان و هەڵكشــان بــووە و بــە گشــیت نــرᣐغ نــەوᣍت ئۆپیــك لــە نیــوەی دووەیم هەشــتاكان ) دۆالر.1كەمــرت ە لــە نــرᣐغ لــە نیــوەی یەكــەیم هەشــتاكاندا بــە زیاتــر لــە (٧ ) دۆالر بــووە بــۆ هــەر بەرمیلێــك كــە زیاتــرە لــە تێكــڕای نــرᣐغ لــە شەســتەكان 23.55( تێكــڕای نــرᣐغ نــەوᣍت ئۆپیــك لــە هەشـتاكاᣍغ ســەدەی ڕابــردوو-ج An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 https://www.huj.uoh.edu.iq https://www.huj.uoh.edu.iq . ) دۆالر یــەك لــە دوای یــەك10.) و (٧22.1( و حەفتــاكاᣍغ ســەدەی ڕابــردوو بــە بــڕی ) داكشــان و هەڵكشــاᣍغ دیــار لــە نــرᣐغ نــەوᣍت ئۆپیــك دا 2000-1٩٩1( هاوشــێوەی هەشــتاكاᣍغ ســەدەی ڕابــردوو لــە دە ســاᣡڵ كۆتــاᣍی ســەدەكە-4 : ڕوویــداوە كــە دەكرێــت لــەم خاᢼڵ نــە دا چڕبكرێتــەوە دیــاردەی داكشــان بــووە، نــرᣐغ نــەوت لــەم چــوار ســاڵە دا لــە ژ لــە چــوار ســاᣡڵ یەكــەیم ئــەم دە ســاڵە دا بــە بــەردەوایم نــرᣐغ نــەوᣍت ئۆپیــك تــووس-أ ) دۆالر لــە 15.53( ) و كــەیم كــردوە بــۆ1٩٩0( ) دۆالر لــە نــرᣐغ ســاᣡڵ3.٦4( كــە كەمــرت ە بــە بــڕی1٩٩1 ) دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ18.٦2( . ) دۆالر لــە مــاوەی ئــەم چــوار ســاڵە دا كــەیم كــردوە3( واتــە بــە زیاتــر لــە1٩٩4 ســاᣡڵ ) دۆالر بۆ هەر بەرمیلێك .20.2) و (٩1.٦8 نرᣐغ نەوᣍت ئۆپیك یەك لە دوای یەك هەڵكشاوە بۆ (٦1 و ٦٩٩1٩٩5 لە ساᢼڵᣍغ-ب ) دۆالر بۆ هەر بەرمیلێك .12.28( ) و18.8 نرᣐغ نەوᣍت ئۆپیك داكشاوە یەك لە دوای یەك بۆ (٦1٩٩8 و1 لە ساᢼڵᣍغ ٧٩٩-ج ) دۆالر بۆ هەر بەرمیلێك .2) و (٦.٧1.٧44( ) نرᣐغ نەوت هەڵكشا یەك لە دوای یەك بۆ2000و1 لە دوو ساᣡڵ كۆتاᣍی ئەم دە ساڵە دا (٩٩٩-د ) دۆالر بــووە كــە كەمــرت ە لــە تێكــڕای نــرᣐغ لــە هەشــتاكاᣍغ ســەدەی 18.22( تێكــڕای نــرᣐغ نــەوᣍت ئۆپیــك لــە دە ســاᣡڵ كۆتــاᣍی ســەدەی بیســتەمدا-ه ) دۆالر یــەك لــە 5.3) و(٦1.٦83( ) دۆالر و زیاتــر لــە تێكــڕای نــرᣐغ لــە شەســتەكان و حەفتــاكاᣍغ ســەدەی ڕابــردوو بــە بــڕی5.33( ڕابــردوو بــە بــڕی دوای یــەك . ) ئامارەكان ئاماژە دەكەن بە:21( ئەوەی تایبەتە بە هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە دە ساᣡڵ یەكەیم سەدەی-5 ) نــرᣐغ نــەوᣍت ئۆپیــك هەڵكشــاᣍغ گــەورەی تۆمــار كــرد دوای ئــەوەی لــە ســەرەتای2008-2001( ) لــە21( لــە هەشــت ســاᣡڵ یەكــەیم ســەدەی- أ دا 2008 ) دۆالر، لــە ســاᣡڵ4.4 بــە بــڕی (٦2000 ) دۆالر بــوو كــە كەمــرت بــوو لــە نــرᣐغ ســاᣡڵ23.12( ) نــرᣐغ بەرمیلێــك نــەوت تەنیــا21( ســەدەی ٩)دۆالر كــە ژمارەیــەیک پێوانــەᣍی بــوو تــا ئــەو كاتــە تۆمــار نەكــرا بــوو .4.1( گەیشــتە ٦) و 0.8 نــرᣐغ نــەوت داكشــاوە یــەك لــە دوای یــەك بــۆ (٦2010 و200) ســاڵەكاᣍغ ٩21( لــە دوو ســاᣡڵ كۆتــاᣍی دەســاᣡڵ یەكــەیم ســەدەی-ب .٧٧) دۆالر .38( . https://www.huj.uoh.edu.iq ) دۆالر بووە كە بەرزترین تێكڕای نرᣐغ نەوᣍت تۆماركراوە بە بەراورد بە دە ساڵەكاᣍغ پێشوو52.45( تێكڕای نرᣐغ نەوت لەم دە ساڵە دا-ج ) نــرᣐغ نــەوᣍت ئۆپیــك هەڵكشــان و داكشــاᣍغ زۆر گــەورەی تێــدا ڕوویــداوە، 2015-2011( ) لــە21( لــە نیــوەی یەكــەیم دە ســاᣡڵ دووەیم ســەدەی-٦ ئامــارەكاᣍغ ئــەو ســاᢼڵ نە ئامــاژە دەكــەن : ) یــەك لــە دوای یــەك 2012و2011( نــرᣐغ نــەوᣍت ئۆپیــك لــە دوو ســاᣡڵ یەكــەیم ئــەم ماوەیــە بــە شــێوەیەیک چاوەڕواننەكــراو بەرزبــووە لــە ســاڵەكاᣍغ-أ ) دۆالر، ئــەو ژمارانــەی كــە ئاماژەیــان بــۆ كــرا دەربــارەی نــرᣐغ نــەوت لــەو دوو ســاڵە دا بەرزتریــن نــرᣐغ نەوتــە كــە تــا 10.٩8 و ٧10.٧4گەیشــتوەتە(٦ ئەمــڕۆ تۆمــار كرابێــت . ) دۆالر كەمــرت ە لــە نــرᣐغ ســاᣡڵ 4( ) دۆالر كــە105.8 دا نــرᣐغ نــەوᣍت ئۆپیــك داكشــانێی كــەیم بەخــۆوە دی و دابــەزی بــۆ (٧2013 لــە ســاᣡڵ-ب .2012 )4.٩4.٦٩ و ٩2) نــرᣐغ نــەوᣍت ئۆپیــك بــە شــێوەیەیک گــەورە داكشــا یــەك لــە دوای یــەك بــۆ (٩2015و2014( لــە دوو ســاᣡڵ كۆتــاᣍی ئــەم ماوەیــە-ج دۆالر . ٩) دۆالر بووە كە بەرزترین تێكڕای نرᣐغ نەوتە تۆمار كرابێت .3.8( تێكڕای نرᣐغ نەوᣍت ئۆپیك لە پێنج ساᣡڵ ئاماژە بۆكراودا-د ئەو هۆكارانەی كە دەبنە مایەی گۆڕانكاری نرᣐغ نەوت : ٩) دۆالر بووە كە بەرزترین تێكڕای نرᣐغ نەوتە تۆمار كرابێت .3.8( تێكڕای نرᣐغ نەوᣍت ئۆپیك لە پێنج ساᣡڵ ئاماژە بۆكراودا-د ئەو هۆكارانەی كە دەبنە مایەی گۆڕانكاری نرᣐغ نەوت : گرنگرت ین ئەو هۆكارانەی كە دەبنە مایەی گۆڕانكاری نرᣐغ نەوᣍت خاوی ئۆپیك بریتᣌێ غ لە : هاوســەنیگ نێــوان خســتنەڕوو و خواســت لــە ســەر نــەوت : بــە یــەیک لــە هــۆكارە بەهــرᢀێ غەكان دادەنرێــت لەبــازاڕی نــەوᣍت جیهــاᣍغ كــەوا كۆمــەᣡڵ -1 فاكتــەری تــر لەخــۆ دەگرێــت وكاریگــەری بەهــرᢀێ غ بــە جێدەهێڵــت لەســەر نــرᣐغ نــەوت بێگومــان كاتێــك خســتنەڕووی نــەوت لــە بــازاڕ دا كــەم دەبێــت و خواســت زیــاد دەبێــت نــرᣐغ نــەوت ڕوو لــە هەڵكشــان دەكات گومــاᣍغ تێــدا نیــە پێچــەواᣍغ ئەمــەش راســتە . ئەمــەش لــە بــەر كۆمەڵــە هۆیــەك كــەوا گرنگرین ئەو هۆكارانەی كە دەبنە مایەی گۆڕانكاری نرᣐغ نەوᣍ خاوی ئۆپیك بریتᣌێ غ لە : هاوســەنیگ نێــوان خســتنەڕوو و خواســت لــە ســەر نــەوت : بــە یــەیک لــە هــۆكارە بەهــرᢀێ غەكان دادەنرێــت لەبــازاڕی نــەوᣍت جیهــاᣍغ كــەوا كۆمــەᣡڵ -1 فاكتــەری تــر لەخــۆ دەگرێــت وكاریگــەری بەهــرᢀێ غ بــە جێدەهێڵــت لەســەر نــرᣐغ نــەوت بێگومــان كاتێــك خســتنەڕووی نــەوت لــە بــازاڕ دا كــەم دەبێــت و خواســت زیــاد دەبێــت نــرᣐغ نــەوت ڕوو لــە هەڵكشــان دەكات گومــاᣍغ تێــدا نیــە پێچــەواᣍغ ئەمــەش راســتە . ئەمــەش لــە بــەر كۆمەڵــە هۆیــەك كــەوا Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 https://www.huj.uoh.edu.iq هەندێكیان دور مەودان ئەواᣍغ تر كاتیᣌێ غ ،گرنگرت ین ئەو هۆكارانە(رووداᣍغ شەڕ و نائاریم سیاس ناوخۆی ونێو دەوڵەᣍت لەنزیك كانەكاᣍغ دەرهێناᣍغ نەوت،فاكتــەرە جیوسیاســیەكان وئابۆریــەكان وئەمنیــەكان وجیولوجیــەكان وئاووهەواییــەكان هەموییــان بەیەكــەوە كاریگەریــان هەیــە لەســەر هاوكێشــەی خســتنەڕو وخۆاســت،بەجۆری بــووᣍغ ئــەو فاكتەرانــە یاخــۆد یــەیک لەوانــە كاردەكاتــە ســەر كەمبــووᣍغ بــڕی خســتنەڕوو زیادبــووᣍغ داواكاری كۆی گشــیت بڕی نەوᣍت خســتنەڕوو 200٩-200پاشــانش زیادبووᣍغ نرᣐغ نەوت وبەپیᢀ ێ زانیارییەكاᣍغ دەســتە و ریكخراوە جیهانەیەكان لەســاالᣍغ ٧ ) 12،ص٧200 ملیــون برمیــل )(حســن،٧88( ملیــون برمیل)بــووە لــە رۆژیكــدا، ئەمەلەكاتیــدا بــڕی داواكاری جیهــاᣍغ نزیكــەی85،5( نزیكــەی لەروژیكــدا ئەمــەش ئــەوە دەگەیــیغ خواســت زیاتــر بــووە لــە خســتنەروو بۆیــە نــرخ بــە رێژەیــی بــاش بەرزبووتەوە.ئــەوەی پێوەنــدی بەالیــەᣍغ خواســت هەیــە بەهەمــان شــێوە كۆمــەᣡڵ فاكتــەر كاریگەریــان هەبــووە لەســەر زیابــووᣍغ خۆاســت وەك گەشــەی ئابــۆری جیهــاᣍغ وزیابــووᣍغ دانیشــتوان وبەرزبوونــەوەی ئاســیت بژیویان،لەكۆتاییــدا ئەمانــە روڵیــان دەبێــت لەســەر زیادبــووᣍغ بەكارهێنــاᣍغ وزە لەناویانشــدا نەوت،بەجــۆری ســاᢼڵ نە بەرێــژەی ،٧٧ملیــون برمیــل رۆژانــە لەســاᣡڵ 2 زیــاد دەكات، ســەرچاوەكان ئامــاژە بــۆ زیادبــووᣍغ خۆاســت جیهــان دەكــەن لــە ســەر نــەوت بەجۆری%1،٦٧ ،ᣍبᢀ گومــان 2015 ،٦٩ملیــون برمیــل لەرۆژیكــدا لەســاᣡڵ8 پاشــان بــۆ نزیكــەی2010 ٩ملیــون برمیــل لەرۆژیكــدا لەســاᣡڵ4،2 بەرزبووتــەوە2000 )4،ص٦٧2012،..(تقریــر اوبیــك2014 ئەمــەی دوایــان دەگەرێتــەوە بــۆ دابەزیــیغ نــرᣐغ نــەوت لــە ســاᣡڵ دابەزییغ سەنی ئۆپیك لە بواری بەرهەمهێنان و هەناردەكردᣍغ نەوت .- -2 ᣍبᢀ ســوود نیــە لــرᢀێ ەدا ئامــاژە بــەو ڕاســتیە بكرێــت كــە ڕێــژەی بەشــداریكردᣍغ ئۆپیــك لــە بەرهەمهێنــان و هەناردەكــردᣍغ نــەوت لــە جیهــان دا لــە ســاᣡڵ ).%5 و ٦.٦%43( ) دا ئــەم ڕێژانــە دابەزیــون یــەك لــە دوای یــەك بــۆ2015( ) بــووە كــە لــە ســاᣡڵ%٩1 و%55.5( یــەك لــە دوای یــەك1٧٩3 )، ئەمــەش الوازبــووᣍغ ســەنیگ ئۆپیــك لــە بــواری بەرهەمهێنــان و هەناردەكــردᣍغ نــەوت لــە جیهانــدا دەســەلمێنێت و5،ص٧٩1(محمــود،٩٧٩ نەوᣍت بێگومان كاریگەری نەرێیغ هەبووە لە سەر توانای ئۆپیك بۆ دیاریكردᣍغ نرᣐغ بێگومــان كاریگــەری نەرێــی هەبــووە لــە ســەر توانــای ئۆپیــك بــۆ دیاریكــردᣍ نــرᣐ نــەوᣍ . یــەك ڕیــزی و یــەك هەڵویســیت دەوڵەتــە ســەرەكیەكاᣍغ بەكاربــەری نــەوت . ئــەم یــەك ڕیزیــە پێــش دامەزرانــدᣍغ ڕێكخــراوی ئۆپیــك لــە -3 بــە پــیᢀ ێ ڕێككەوتننامــەی (ئاكنــاكاری) پێكهێرغ ابــوو، بــەدی دەكــرا . دواتــر1٩2نێــوان كۆمپانیــا نەوتیەكانــدا كارتــەᣡڵ نــەوᣍت نێودەوڵــەᣍت كــە لــە ســاᣡڵ ٧ دوای دروســتبووᣍغ ئۆپیــك و بــاᢼڵ دەســیت لــە دیاریكــردᣍغ نــرᣐغ نەوتــدا لــە نیــوەی یەكــەیم حەفتــاكاᣍغ ســەدەی ڕابــردوو دا دەوڵەتــە پیشەســازییە ی ئەیلــویل 21 ) . https://www.huj.uoh.edu.iq یــان لــەInternational Energy Agency) I.E.Aســەرمایەدارییە بەكاربــەراᣍغ نــەوت لــە جیهانــدا (ئاژانــی وزەی نێودەوڵــەᣍت ) دەوڵــەت و دەوڵەتەكانیــش بریــیت بــوون 20( دا گەیشــتە1٧٩4 ی تᣄژ یــیغ یەكــەیم15 دامەزرانــد و ژمــارەی ئەندامــاᣍغ ئــەم ڕێكخــراوە لــە1٧٩4 ،لــە (والیەتەیەكگرتــووەكاᣍغ ئەمریــكا، بەڕیتانیــا، ئەڵمانیــای ڕۆژئــاوا، هۆڵەنــدا، بەلجیــكا، فنلەنــدا، ئیتالیــا، ژاپــۆن، نەمســا، دانیمــارك، ســوید، نەرویــــــج ) ئامانــیب دروســتكردᣍغ ئــەم 18٩-18،ص٧1٩81،ئیســپانیا، لۆگســمبۆرگ، یۆنــان، كەنــەدا، ســویᣄا، ئوســرت الیا، نیوزلەنــدا، توركیــا ).(اســماعیل ێكخ اوە ب ی ت ب وە لە : یەكخستیغ دەوڵەتە پیشەسازییە سەرمایەدارەكان بەرامبەر بە ڕێكخراوی ئۆپیك و الوازكردᣍغ لە بواری دیاریكردᣍغ نرᣐغ نەوت .- ا كەمكردنــەوەی بەكارهێنــاᣍغ نــەوت لــە الیــەن دەوڵەتــاᣍغ ئەنــدام لــە ڕێكخراوەكــە بــە مەبەســیت كەمكردنــەوەی خواســت لــە ســەر نــەوᣍت - ب ئۆپیــك و دابەزیــیغ نــرᣐغ . ڕێكەوتن لە سەر پالنێی تایبەت بە دابەشكردᣍغ نەوت لە كاᣍت ڕووداوی لەناكاو . - ت ڕێكەوتن لە سەر پالنێی تایبەت بە دابەشكردᣍغ نەوت لە كاᣍت ڕووداوی لەناكاو . - ت پێداویســیت ســاڵێی ئەندامــاᣍغ ڕێكخراوەكــە بــكات لــە ژ ᣐبᢀ بــە ᣐبᢀ كــردᣍغ پرۆگرامێــی تایبــەت بــە كۆگاكــردᣍغ بڕێــی زۆر لــە نــەوت كــە بــەس - ث كاᣍت نەبــووᣍغ توانــای هاوردەكــردᣍغ نــەوت لــە بــەر هــەر هۆكارێــك بێــت . پێداویســیت ســاڵێی ئەندامــاᣍغ ڕێكخراوەكــە بــكات لــە ژ ᣐبᢀ بــە ᣐبᢀ كــردᣍغ پرۆگرامێــی تایبــەت بــە كۆگاكــردᣍغ بڕێــی زۆر لــە نــەوت كــە بــەس - ث كاᣍت نەبــووᣍغ توانــای هاوردەكــردᣍغ نــەوت لــە بــەر هــەر هۆكارێــك بێــت . هەوڵدان بۆ دابینكردᣍغ ئەلتەرناتیڤ بۆ نەوᣍت ئۆپیك . - ج دابەزیــیغ بەهــای دۆالری ئەمریــی بەرامبــەر دراوە نێۆدەوڵەتیــەكان : ئەمــەش یەكیكــە لــە وهۆكارانــەی دادەنرێــت كاردەكاتــە ســەر نــرᣐغ -4 نەوت بەهۆی ئەو پەیوەندیە بەهرᢀێ غەی نێوان دۆالر ونرᣐغ نەوت هەیانە چونكە هەموو مامەڵە بازرگانیەكاᣍغ نەوت بەو دراوە دەكرێت،بەجۆری هــەر گۆڕانكاریــك لــە بەهــای دۆالر رووبــدات دەررئەنجــایم دەبێــت لەســەر نــرᣐغ نــەوات بــە پوزەتیــف ونەگەتیــڤ وپاشــانش لەســەر ئابــۆری An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University https://www.huj.uoh.edu.iq ) دابەزیــن21-1،ص٧200وᢼڵ تــاᣍغ بەرهــەم هێنەر،بەجــۆری كاᣍت بەهــای دۆالر دادەبەزێــت نــرᣐغ نــەوت بــەدۆالر هەڵدەكشێت.(عبداالســدی،٩ وبەرزبوونــەوە نــرᣐغ نــەوت كاریگــەری خــراپ بەجێدەهێڵــت لەســەر ئابــۆری وᢼڵ تــاᣍغ بەرهــەم هێنــەر چونكــە ئابوریــان بەرێژەیــی بــەرز رۆڵ دەبیــیغ لەزیادبــوون وكەمبوونــەوەی خــەرᣐب حكــویم لەهەمــوو بــوارەكان. دابەزیــیغ بــڕی یــەدەیکگ نەوتــیێ : نــەوت ویــەدەیکگ نــەوᣍت لەئەنجــایم هــۆكارە جیولوجیــەكان دورســت بــووە بەدرێــژاᣍی چەندهــا هەزارســاڵ لــە -5 نێــون قۆژبنــەكاᣍغ زەوی،وەهەروەهــا توانــای نوێوبوونــەوەی زیادبــووᣍغ نیــە بەڵكــو لەگــەڵ بەكارهێنــاᣍغ بەشــێوەیی نــا هۆشــمەندانە وهونــەری وەك هــەر كاᢼڵ یــی تــر تــەواو دەبێــت وەك ئــەوەی لــەدۆᣐغ وᢼڵᣍت بەحریــن وئیمــارات كەلــە رابــردوودا دوو وᢼڵᣍت بەرهــەم هێنــەری نــەوت بــوون، بۆنموونــە بەریتانیــا یــەیک لەوᢼڵ تانەیــە كــەوا دەســیت كــردوە بــە ژمــاردᣍغ كۆتــا بەرمیــیل نــەوت لــە و كێڵگانــەی دەكەونــە دەریــای باكــور بــەᣍچᢀ زانیاریــەكاᣍغ بــڕی . 200ملیــون برمیــل لەرۆژیكــدا لــە ســاᣡڵ ٧1،ملیــون برمیــل بــووە لــە رۆژیكــدا كــەᣐب دابزیــوە بــۆ كەمــرت ٧3،1 نزیكــەی1ابەرهەمهێنــاᣍغ لەســاᣡڵ ٩٩٩ زیادبــووᣍغ چــاالىک (زیادكــردᣍغ نــرخ لەســەر یەك)(المضاربە)لەبــازاری ســودا كاری نــەوت: بــە مەبەســیت بەدەســت هینــاᣍغ قازانــیب -٦ زیاتــر لەریــگای ئــەو پیشــبینانەی بــۆ نــرᣐغ نــەوت دەكرێــت كۆمــەᣡڵ وەبەرهێنــاᣍغ تــازە هاتنــە نــاو بــازاڕی نــەوت وەك دامــەزراوەكاᣍغ وەبەرهێنــان وەك دەســتەی خانەنیشــنان وكومپانیــاکاᣍغ دڵنیــاᣍی وەك پارێزگاریــك لەبەرامبــەر هــەر مەترســیی وەك هەڵەوەســان وسســتبووᣍغ نــرᣐغ )،هەرلەم روانگــەوە لــەم چەنــد ســاالᣍغ دوایــدا گریبســتەكاᣍغ نــەوت لەبازارەكانــدا بەرزبوونەوەیــی بــە خۆیــەوە بیــیغ 4،ص٩2011،دۆالر(الســوب ســاڵ لــە دوای ســاڵ بەجــۆری كەوتــە ســەرەوەی پیشــبینیەكان بــڕی بەرهــەم هێنــاᣍغ راســتەقینە وبەكارهینــاᣍغ جیهــاᣍغ لــە نــەوت ئەمــەش بــووە هــۆی زیادبــووᣍغ رێــژەی داواكاری وعەمباركــردᣍغ نــەوت ئەمــەش لــە كۆتاییــدا پاڵەپســتوی لەســەرنرᣐغ نــەوت بــە ᣐبᢀ هێیشــیت . https://www.huj.uoh.edu.iq قــرێ اᣍغ ئابــوری جیهــاᣍغ : رووداᣍغ قرێ انــەكان لەســەر ئاســیت جیهــان كاریگــەری بەهــرᢀێ غی هەبــوو لــە ســەر نــرᣐغ نــەوت وەك قــرێ اᣍغ دارᣍی ئــەم -٧ دواییــەی جیهــان كەبــوو هــۆی پەكخســتیغ بەشــیك زور لــە چاالكییــە ئابۆرییــە جیهانیــەكان بەتایبــەت بــازاری خانووبــەرە لــە وᢼڵ تەیەكگرتــووەكاᣍغ .ئەمــە ســەرەڕای قــرێ اᣍغ دارᣍی ئاســیا ســاᣡڵ 2015و2014 ئەمریــكا وهەروەهــا كزكــردᣍغ بــازارەكاᣍغ دراو ودامــەزراوە بانكیــەكان بەتایبــەت لــە ســاᢼڵᣍغ ᣍبᢀ گومــان هەمــوو ئەمانــە كاریگەریــان بــەᣐبᢀ هێشــت لەســەر هــرᢀێ غی وەبەرهێنــان بــە ᣐبᢀ هێشــت وای كــرد داواكاری وخۆاســت لەســەر نــەوت 1٩٩٩ )13،ص2008،كــەم بێتــەوە نرخیــش داببەزێت(اوبیــك جەنــگ ونائــارایم لــە ناوچــە دەوڵەمەنــد ەكان بەنــەوت:وەك ئەونائارامییانەیلەناوچــەی رۆژهـەᢼڵᣍت ناوەراســت بــەدᣍی كــرا لــە ســاᢼڵᣍغ رابــردوو -8 لــە پینــج وᢼڵᣍت چڕدەبێتــەوە ئەوانش(عــرᢀێ اق وئــرێ ان وســعودیا%٦2،5 كــۆی بــڕی نــەوᣍت جیهــان لــە خــۆ دەگرێــت لــەم بــڕە نزیكــەی%80 كــە نزیكــەی گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 هــۆكارە رسوشــتیەكان: بەتایبــەت ئــەو هۆكارنــەی شــوێن پەنجــەی خۆیــان بــە جــیᢀ ێ دەهێڵــᣌێ غ لەســەر نــرᣐغ نــەوت وەك ئــەو زریانیانــەی -٩ لەئەنجایم 2008-2002 لەناوچەكاᣍغ دەرهینان وگۆاستنەوەی نەوت روودەدن رنگدانەوەی دەبێت لەرسەر نرᣐغ نەوت وەك ئەوەی لەنێوان ساᢼڵᣍغ خویدا بەســەر ناوچەكاᣍغ پاᢼڵ وتیغ نەوت دا لە وᢼڵ تە 2005 هەڵكردᣍغ زریاᣍغ ایفان لە كەندوای مكســیك رویدا وهەروەها زریاᣍغ كاترینا كەلەســاᣡڵ ٧)1،ص200یەكگرتــووەكاᣍغ ئەمریــكا، ئەمانــەش بوونــە هــۆی كەمبــووᣍغ بــڕی بەرهــەیم نــەوت وزیادبــووᣍغ نــرᣐغ نەوت.(محمــود،٧ ناسەقامگرێ ی جیوسیاس:رواداوە سیاسیەكان لەچەند شێوەیەك خۆیان دەبینینەوە وەك: -10 . https://www.huj.uoh.edu.iq قرێ اᣍغ ئەتوویم ئرێ ان لە گەڵ وᢼڵ تاᣍغ گروپ پێنج لەگەڵ یەك-ا .2003 جەنیگ داگرێ كردᣍغ عرᢀێ اق لەالیەن ئەمریكاوە لەساᣡڵ-ب داگرێ كردᣍغ ناوچەیی فرەواᣍغ عرᢀێ اق لەالیەن ریكخراوی ترێ ورسیت دەوڵەᣍت ئیسالیم نارساو بە داعش.-ج روداوە سیاسیەكان وᢼڵᣍت فەنزویال كاری كردتە سەر راوستاᣍغ بەرهەمهێناᣍغ نەوت.-د )23،ص200 نائارامییە خیᣀڵ یەتیەكاᣍغ وᢼڵᣍت نیجرێ یا وەك یەیک لەئەندایم ریكخراوی اوبیك .(النرت ،٧-ه ئەمانــە بەســەر یەكــەوە كاریگــەری زۆریــان هەبــوو لەســەر كەمكردنــەوەی بــڕی نــەوᣍت بەرهــەم هاتوولــە ناووᢼڵ تــاᣍغ ریخــراوی اوبیــك بەرێــژەی ســیᢀ ێ هــۆكارە رسوشــتیەكان: بەتایبــەت ئــەو هۆكارنــەی شــوێن پەنجــەی خۆیــان بــە جــیᢀ ێ دەهێڵــᣌێ غ لەســەر نــرᣐغ نــەوت وەك ئــەو زریانیانــەی -٩ لەئەنجایم 2008-2002 لەناوچەكاᣍغ دەرهینان وگۆاستنەوەی نەوت روودەدن رنگدانەوەی دەبێت لەرسەر نرᣐغ نەوت وەك ئەوەی لەنێوان ساᢼڵᣍغ خویدا بەســەر ناوچەكاᣍغ پاᢼڵ وتیغ نەوت دا لە وᢼڵ تە 2005 هەڵكردᣍغ زریاᣍغ ایفان لە كەندوای مكســیك رویدا وهەروەها زریاᣍغ كاترینا كەلەســاᣡڵ ٧)1،ص200یەكگرتــووەكاᣍغ ئەمریــكا، ئەمانــەش بوونــە هــۆی كەمبــووᣍغ بــڕی بەرهــەیم نــەوت وزیادبــووᣍغ نــرᣐغ نەوت.(محمــود،٧ روداوە سیاسیەكان وᢼڵᣍت فەنزویال كاری كردتە سەر راوستاᣍغ بەرهەمهێناᣍغ نەوت.-د )23،ص200 نائارامییە خیᣀڵ یەتیەكاᣍغ وᢼڵᣍت نیجرێ یا وەك یەیک لەئەندایم ریكخراوی اوبیك .(النرت ،٧-ه ئەمانــە بەســەر یەكــەوە كاریگــەری زۆریــان هەبــوو لەســەر كەمكردنــەوەی بــڕی نــەوᣍت بەرهــەم هاتوولــە ناووᢼڵ تــاᣍغ ریخــراوی اوبیــك بەرێــژەی ســیᢀ ێ Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 HUJ-Volume 5, Issue 4, December 2020 وەك لــە باســكردᣍغ هــۆكارەكان دەردەكەوێــت كــە ناتوانرێــت هەڵكشــان و داكشــاᣍغ نــرᣐغ 2015-1٦٩1 نــرᣐغ نــەوᣍت ئۆپیــك لــە نێــوان ســاڵەكاᣍغ نــەوت بــە تاكــە هۆكارێــك دیــاری بكرێــت ئەمــە جگــە لــەوەی كــە كاریگەریــەᣍت هــۆكارەكان لــە ســەر هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت لــە قۆناغێكــەوە بــۆ قۆناغێــی تــر دەگۆڕێــت وەك لــەم خاᢼڵ نــەدا دەردەكەوێــت : )كــە زیاتربــە جێگرێ بــووᣍغ نــرᣐغ نــەوᣍت خــاوی نیشــانەكراوی راگەیەنــراو و اتــا 1٧٩1-1٦٩1( قۆنــاᣘغ یەكــەم/ ئــەم قۆناعــە دەكەوێتــە نێــوان ســاᢼڵᣍغ ڕووداᣍغ هەڵكشــان و داكشــاᣍغ كــەم لــە نــرᣐغ ڕاســتەقینەی نــەوᣍت ئۆپیـك لــەو ماوەیــەدا روویــداوە ئەمــەش دەگەڕێتــەوە بــۆ ڕۆᣡڵ ڕێكخــراوی ئۆپیــك كــە ) ســاڵ .10( دا و بــۆ مــاوەی زیاتــر لــە1٦٩0 تــواᣍغ نــرᣐغ نــەوᣍت ڕاگەیەنــراو جێگــرێ بــكات لــە ئاســیت نــرᣐغ نــەوت لــە ئــاᣍب ڕێكخــراوی ئۆپیــك لــە شەســتەكاندا ســاڵ لــە دوای ســاڵ ســەنیگ لــە بــواری بەرهەمهێنــان و هەناردەكــردᣍغ نــەوᣍت جیهــان بەرزدەبــووەوە لــە ئەنجــایم زیادبــووᣍغ ژمــارەی ئەندامــاᣍغ و ڕووداᣍغ گۆڕانــكاری لــە سیاســەᣍت نــەوᣍت ئۆپیــك دا . ᣍبᢀ گومــان ئۆپیــك لــە توانایــدا بــوو نــرᣐغ نــەوت لــە شەســتەكاندا زیــاد بــكات بـەᢼڵ م زیادكــردᣍغ بەرهەمهێنــاᣍغ نــەوت لــە الیــەن ئەندامــاᣍغ ئۆپیـك لــەو ماوەیــەدا بــە شــێوەیەیک بــەردەوام (بەرهەمهێنــاᣍغ نــەوت لــە الیــەن ).(العقیل، بدون 1٧٩0 ) ملیۆن بەرمیل/ڕۆژانە لە ســاᣡڵ23. دا زیادی كرد بۆ (٧1٦٩0 )ملیۆن بەرمیل/ڕۆژانە لە ســاᣡڵ8.ئەنداماᣍغ ئۆپیك لە (٧ )1 )دا رێگــر بــوو لــە زیادكــردᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــەو ماوەیــەدا .(شــێوەی22 ســنە الطبــع، ص )كــە زیاتربــە جێگرێ بــووᣍغ نــرᣐغ نــەوᣍت خــاوی نیشــانەكراوی راگەیەنــراو و اتــا 1٧٩1-1٦٩1( قۆنــاᣘغ یەكــەم/ ئــەم قۆناعــە دەكەوێتــە نێــوان ســاᢼڵᣍغ ڕووداᣍغ هەڵكشــان و داكشــاᣍغ كــەم لــە نــرᣐغ ڕاســتەقینەی نــەوᣍت ئۆپیـك لــەو ماوەیــەدا روویــداوە ئەمــەش دەگەڕێتــەوە بــۆ ڕۆᣡڵ ڕێكخــراوی ئۆپیــك كــە ) ســاڵ .10(دا و بــۆ مــاوەی زیاتــر لــە1٦٩0نــەوت لــە ئــاᣍ نــەوᣍت ڕاگەیەنــراو جێگــر بــكات لــە ئاســیت نــرᣐغ تــواᣍغ نــرᣐغ ڵ ) ( و بۆ و ی زی ر ی رᣐ و ᣍب ی رو جێ رێ ب وᣍ رᣐ وᣍ ڕ ڕێكخــراوی ئۆپیــك لــە شەســتەكاندا ســاڵ لــە دوای ســاڵ ســەنیگ لــە بــواری بەرهەمهێنــان و هەناردەكــردᣍغ نــەوᣍت جیهــان بەرزدەبــووەوە لــە ئەنجــایم زیادبــووᣍغ ژمــارەی ئەندامــاᣍغ و ڕووداᣍغ گۆڕانــكاری لــە سیاســەᣍت نــەوᣍت ئۆپیــك دا . ᣍبᢀ گومــان ئۆپیــك لــە توانایــدا بــوو نــرᣐغ نــەوت لــە شەســتەكاندا زیــاد بــكات بـەᢼڵ م زیادكــردᣍغ بەرهەمهێنــاᣍغ نــەوت لــە الیــەن ئەندامــاᣍغ ئۆپیـك لــەو ماوەیــەدا بــە شــێوەیەیک بــەردەوام (بەرهەمهێنــاᣍغ نــەوت لــە الیــەن ).(العقیل، بدون 1٧٩0 ) ملیۆن بەرمیل/ڕۆژانە لە ســاᣡڵ23. https://www.huj.uoh.edu.iq HUJ-Volume 5, Issue 4, December 2020 HUJ-Volume 5, Issue 4, December 2020 ملیــون بەرمیــل لەرۆژیكدا،ئەمــەش كاریگــەری هەبــوو لــەرس دابەزیــیغ بــڕی نــەوᣍت خســتەڕوو لەبــارزاری جیهــان نەوت،چونكــە ئــرێ ان و عــرᢀێ اق بــەدوو بەرهەمهێنــەری ســەرەیک رۆژهــەᢼڵᣍت ناوەراســت دادەنرێــت. ســەرچاوە ئەلتەرناتیڤــەكاᣍغ نــەوت، (گازی رسوشــیت ، خەڵــووز، وزەی خــۆر، وزەی بــا، وزەی نــاوەیک ...تــاد) كاریگەریــی ئەوتویــان نابێــت -11 هــەر زوری ئامــراز ژ پێداویســیت جیهــان پڕدەكاتــەوە لــەرووی وزەوە،ئەمــە لەكاتیكــدا بــەس ژ لەمەودایــی كورتدا،چونكــە نــەوت نزیكــەی ســیᢀ ێ بــەس وئامــرێ ەكان لــەرووی دایزانــەوە بــۆ بەكارهێنــاᣍغ نــەوت وگازی رسوشــیت دورســت كــراوان. ناكــۆیک نێــوان ئەندامــاᣍغ ئۆپیــك لــە الیــەك و نەبــووᣍغ هاریــكاری نێــوان ئۆپیــك و دەوڵەتــاᣍغ هەناردەكــەری نــەوت لــە دەرەوەی ئۆپیــك كــە دەبێتــە -4 . مایــەی زیادبــووᣍغ خســتنەڕووی نــەوت لــە بــازاڕە جیهانییەكانــدا و دابەزیــیغ نــرᣐغ 201٩-1٦٩1 شیكردنەوەی دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ-3 ئــەو ئامــار و زانیاریانــەی كــە لــە تــەوەری دووەم دا ئاماژەیــان بــۆ كــرا دەریدەخــەن كــە هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت دیاردەیــەیک بــاو بــووە لــە .هــۆكارەكاᣍغ هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت تەنیــا هــۆكاری ئابــوری نــᣌێ غ ، بەڵكــو بــە هــۆی تایبەتمەنــدی نەوتــە وەك 2015-1٦٩1 نێــوان ســاᢼڵᣍغ كاᢼڵ یــەیک ســرت اتییب (لــە بــواری وزە و كەرەســتەی خــاو دا) وەك ســەرچاوەیەیک رسوشــیت كــە نــوێ نابێتــەوە . 201٩-1٦٩1 شیكردنەوەی دیاردەی هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە نێوان ساᢼڵᣍغ-3 ئــەو ئامــار و زانیاریانــەی كــە لــە تــەوەری دووەم دا ئاماژەیــان بــۆ كــرا دەریدەخــەن كــە هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت دیاردەیــەیک بــاو بــووە لــە .هــۆكارەكاᣍغ هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت تەنیــا هــۆكاری ئابــوری نــᣌێ غ ، بەڵكــو بــە هــۆی تایبەتمەنــدی نەوتــە وەك 2015-1٦٩1 نێــوان ســاᢼڵᣍغ كاᢼڵ یــەیک ســرت اتییب (لــە بــواری وزە و كەرەســتەی خــاو دا) وەك ســەرچاوەیەیک رسوشــیت كــە نــوێ نابێتــەوە . لــرᢀێ ەدا بــە بــاش دەزانرێــت پێــش شــیكردنەوەی دیــاردەی هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت بــە پــیᢀ ێ ئــەو قۆناغانــەی كــە ئاماژەیــان بــۆ كــرا لــە تــەوەری دووەم دا، بــە كــورᣍت ئــەو هۆكارانــە دەستنیشــان بكەیــن كــە دەبنــە مایــەی هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوᣍت خـاوی ئۆپیــك، هۆكارەكانیــش بریتــᣌێ غ لــە : ئــەو هۆكارانــەی كــە ئاماژەیــان بــۆ كــرا بەرپرســن لــە هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت لــە بــازاڕەكاᣍغ جیهانــدا و بەرپرســن لــە هەڵكشــان و داكشــاᣍغ . HUJ-Volume 5, Issue 4, December 2020 وەك لــە باســكردᣍغ هــۆكارەكان دەردەكەوێــت كــە ناتوانرێــت هەڵكشــان و داكشــاᣍغ نــرᣐغ 2015-1٦٩1 نــرᣐغ نــەوᣍت ئۆپیــك لــە نێــوان ســاڵەكاᣍغ نــەوت بــە تاكــە هۆكارێــك دیــاری بكرێــت ئەمــە جگــە لــەوەی كــە كاریگەریــەᣍت هــۆكارەكان لــە ســەر هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت لــە قۆناغێكــەوە بــۆ قۆناغێــی تــر دەگۆڕێــت وەك لــەم خاᢼڵ نــەدا دەردەكەوێــت : )كــە زیاتربــە جێگرێ بــووᣍغ نــرᣐغ نــەوᣍت خــاوی نیشــانەكراوی راگەیەنــراو و اتــا 1٧٩1-1٦٩1( قۆنــاᣘغ یەكــەم/ ئــەم قۆناعــە دەكەوێتــە نێــوان ســاᢼڵᣍغ ڕووداᣍغ هەڵكشــان و داكشــاᣍغ كــەم لــە نــرᣐغ ڕاســتەقینەی نــەوᣍت ئۆپیـك لــەو ماوەیــەدا روویــداوە ئەمــەش دەگەڕێتــەوە بــۆ ڕۆᣡڵ ڕێكخــراوی ئۆپیــك كــە ) ســاڵ .10( دا و بــۆ مــاوەی زیاتــر لــە1٦٩0 تــواᣍغ نــرᣐغ نــەوᣍت ڕاگەیەنــراو جێگــرێ بــكات لــە ئاســیت نــرᣐغ نــەوت لــە ئــاᣍب ڕێكخــراوی ئۆپیــك لــە شەســتەكاندا ســاڵ لــە دوای ســاڵ ســەنیگ لــە بــواری بەرهەمهێنــان و هەناردەكــردᣍغ نــەوᣍت جیهــان بەرزدەبــووەوە لــە ئەنجــایم زیادبــووᣍغ ژمــارەی ئەندامــاᣍغ و ڕووداᣍغ گۆڕانــكاری لــە سیاســەᣍت نــەوᣍت ئۆپیــك دا . ᣍبᢀ گومــان ئۆپیــك لــە توانایــدا بــوو نــرᣐغ نــەوت لــە شەســتەكاندا زیــاد بــكات بـەᢼڵ م زیادكــردᣍغ بەرهەمهێنــاᣍغ نــەوت لــە الیــەن ئەندامــاᣍغ ئۆپیـك لــەو ماوەیــەدا بــە شــێوەیەیک بــەردەوام (بەرهەمهێنــاᣍغ نــەوت لــە الیــەن ).(العقیل، بدون 1٧٩0 ) ملیۆن بەرمیل/ڕۆژانە لە ســاᣡڵ23. دا زیادی كرد بۆ (٧1٦٩0 )ملیۆن بەرمیل/ڕۆژانە لە ســاᣡڵ8.ئەنداماᣍغ ئۆپیك لە (٧ )1 )دا رێگــر بــوو لــە زیادكــردᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــەو ماوەیــەدا .(شــێوەی22 ســنە الطبــع، ص لــرᢀێ ەدا بــە بــاش دەزانرێــت پێــش شــیكردنەوەی دیــاردەی هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت بــە پــیᢀ ێ ئــەو قۆناغانــەی كــە ئاماژەیــان بــۆ كــرا لــە تــەوەری دووەم دا، بــە كــورᣍت ئــەو هۆكارانــە دەستنیشــان بكەیــن كــە دەبنــە مایــەی هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوᣍت خـاوی ئۆپیــك، هۆكارەكانیــش بریتــᣌێ غ لــە : ئــەو هۆكارانــەی كــە ئاماژەیــان بــۆ كــرا بەرپرســن لــە هەڵكشــان و داكشــاᣍغ نــرᣐغ نــەوت لــە بــازاڕەكاᣍغ جیهانــدا و بەرپرســن لــە هەڵكشــان و داكشــاᣍغ . HUJ-Volume 5, Issue 4, December 2020 دا زیادی كرد بۆ (٧1٦٩0 )ملیۆن بەرمیل/ڕۆژانە لە ســاᣡڵ8.ئەنداماᣍغ ئۆپیك لە (٧ )1 )دا رێگــر بــوو لــە زیادكــردᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــەو ماوەیــەدا .(شــێوەی22 ســنە الطبــع، ص An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 3 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq 1٧٩1-1٦٩1 )هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ1( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دەگەڕێتەوە بۆ كۆمەڵە هۆكارێك كە بریتᣌێ غ لە : 1٩80-1٧٩2قۆناᣘغ دووەم: هەڵكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە حەفتاكاᣍغ سەدەی ڕابردوو بەهرᢀێ غبــووᣍغ ســەنیگ ئۆپیــك لــە بــواری گفتوگۆكــردن لــە گــەڵ كۆمپانیــا نەوتیــەكان لــە ئەنجــایم بــاᢼڵ دەســیت ئۆپیــك لــە بــواری هەناردەكــردᣍغ نــەوت -أ ی نــەوᣍت هەناردەكــراو لــە جیهانــدا لــە نیــوەی یەكــەیم حەفتاكانــدا) و لــە ئەنجــایم گۆڕیــیغ سیاســەᣍت نــەوᣍت هەندێــك لــە %٩0 لــە جیهانــدا (زیاتــر لــە دەوڵەتــاᣍغ ئۆپیــك بــە هــۆی گۆڕیــیغ رژێمــە سیاســیەكانیان . بەرزبوونــەوی هۆشــیاری نــەوᣍت الی ئەندامــاᣍغ ئۆپیــك و هەســتكردنیان بــەو ســتەمەی كــە لێانكــراوە لــە بــواری دیــاری كــردᣍغ نــرᣐغ نــەوت دا كــە -ب نرخێــی دادپەروەرانــە نەبــوو وە پێویســتە بەرزبكرێتــەوە بــە تایبــەᣍت لــە بــەر ئــەوەی كــە نــەوت ســەرچاوەیەیک لــە بــن هاتــووە و نــوێ نابێتــەوە . بەرزبوونــەوەی بــەردەوایم نــرᣐغ كاᢼڵ پیشەســازییەكان و بنەڕەتیــەكان لــە ئەنجــایم دیــاردەی هەᢼڵ وســان لــە جیهانــدا و دابەزیــیغ بەهــای دۆالر -ج بەرامبــەر دراوەكاᣍغ تــر كــە بــووە مایــەی دابەزیــیغ بەهــای راســتەقینەی داهاتــە نەوتیــەكاᣍغ دەوڵەتــاᣍغ ئەنــدام . HUJ-Volume 5, Issue 4, December 2020 بەرزبوونــەوەی بــەردەوایم نــرᣐغ كاᢼڵ پیشەســازییەكان و بنەڕەتیــەكان لــە ئەنجــایم دیــاردەی هەᢼڵ وســان لــە جیهانــدا و دابەزیــیغ بەهــای دۆالر -ج بەرامبــەر دراوەكاᣍغ تــر كــە بــووە مایــەی دابەزیــیغ بەهــای راســتەقینەی داهاتــە نەوتیــەكاᣍغ دەوڵەتــاᣍغ ئەنــدام . بــۆ نموونــە كاتێــك نــرᣐغ فــەریم نــەوᣍت )105،ص1٩81،٧) دۆالر .(اسماعیل-) دۆالر بووە بۆ هەر بەرمیلێك بەهای ڕاستەقینەی دابەزیوە بۆ نێوان (٦12.5( دا1ئۆپیك لە ساᣡڵ ٧٧٩ دا كاتێــك بڕیــاری كەمكردنــەوەی دەســت بــە جــیᢀ ێ 1٧٩3 بەكارهێنــاᣍغ نــەوت وەك چەكێــی ســیاس لــە الیــەن دەوڵەتــە عەرەبیەكانــەوە لــە ســاᣡڵ-د ی كەمكــردەوە) و دواتــر %10 ) دا (ســعوودیە بــە ڕێــژەی%5( لــە كوەیــت بــە ڕێــژەی1٧٩3 ی تᣄژ یــیغ یەكــەیم ســاᣡڵ1بەرهەمهێنــاᣍغ نەوتیــان لــە ٧ ) بەرزكــردەوە و جەختیــان %25( ی تᣄژ یــیغ دووەم لــە كوەیــت بەســرت ا ڕێــژەی كەمكردنــەوەی بەرهەمهێنــاᣍغ نەوتیــان بــۆ5-4 لــەو كۆنگرەیــەی كــە لــە لــە ســەر بڕیــاری پێشــویان كــردەوە كــە تایبــەت بــوو بــە ڕاگرتــیغ هەناردەكــردᣍغ نــەوت بــۆ هەریــەك لــە ویالیەتەیەكگرتــووەكاᣍغ ئەمریــكا و هۆڵەنــدا ئــەم ێ ڕی پ بەرزبوونــەوی هۆشــیاری نــەوᣍت الی ئەندامــاᣍغ ئۆپیــك و هەســتكردنیان بــەو ســتەمەی كــە لێانكــراوە لــە بــواری دیــاری كــردᣍغ نــرᣐغ نــەوت دا كــە -ب نرخێــی دادپەروەرانــە نەبــوو وە پێویســتە بەرزبكرێتــەوە بــە تایبــەᣍت لــە بــەر ئــەوەی كــە نــەوت ســەرچاوەیەیک لــە بــن هاتــووە و نــوێ نابێتــەوە . بەرزبوونــەوەی بــەردەوایم نــرᣐغ كاᢼڵ پیشەســازییەكان و بنەڕەتیــەكان لــە ئەنجــایم دیــاردەی هەᢼڵ وســان لــە جیهانــدا و دابەزیــیغ بەهــای دۆالر -ج بەرامبــەر دراوەكاᣍغ تــر كــە بــووە مایــەی دابەزیــیغ بەهــای راســتەقینەی داهاتــە نەوتیــەكاᣍغ دەوڵەتــاᣍغ ئەنــدام . بــۆ نموونــە كاتێــك نــرᣐغ فــەریم نــەوᣍت )105،ص1٩81،٧) دۆالر .(اسماعیل-) دۆالر بووە بۆ هەر بەرمیلێك بەهای ڕاستەقینەی دابەزیوە بۆ نێوان (٦12.5( دا1ئۆپیك لە ساᣡڵ ٧٧٩ دا كاتێــك بڕیــاری كەمكردنــەوەی دەســت بــە جــیᢀ ێ 1٧٩3 بەكارهێنــاᣍغ نــەوت وەك چەكێــی ســیاس لــە الیــەن دەوڵەتــە عەرەبیەكانــەوە لــە ســاᣡڵ-د ی كەمكــردەوە) و دواتــر %10 ) دا (ســعوودیە بــە ڕێــژەی%5( لــە كوەیــت بــە ڕێــژەی1٧٩3 ی تᣄژ یــیغ یەكــەیم ســاᣡڵ1بەرهەمهێنــاᣍغ نەوتیــان لــە ٧ ) بەرزكــردەوە و جەختیــان %25( ی تᣄژ یــیغ دووەم لــە كوەیــت بەســرت ا ڕێــژەی كەمكردنــەوەی بەرهەمهێنــاᣍغ نەوتیــان بــۆ5-4 لــەو كۆنگرەیــەی كــە لــە لــە ســەر بڕیــاری پێشــویان كــردەوە كــە تایبــەت بــوو بــە ڕاگرتــیغ هەناردەكــردᣍغ نــەوت بــۆ هەریــەك لــە ویالیەتەیەكگرتــووەكاᣍغ ئەمریــكا و هۆڵەنــدا ئــەم ) ملیــۆن بەرمیــل بــوو ڕۆژانــە و لــە دەســتداᣍغ5( دا كــە نزیكــەی1٧٩8 ی كانــووᣍغ یەكــەیم2 ڕاگرتــیغ تــەواوی هەناردەكــردᣍغ نــەوᣍت ئــرᢀێ ان لــە ٦-ه ڕژێمێــی دۆســیت ڕۆژئــاوا (ڕژێــىم شــای ئــرᢀێ ان) بــووە مایــەی كەمبوونــەوەی نــەوᣍت كۆگاكــراو لــە دەوڵەتــاᣍغ ئەنــدام لــە ئاژانــی وزەی نێودەوڵــەᣍت و ) ملیــۆن 3( ) ملیــۆن بەرمیل/ڕۆژانــە چونكــە ئەندامــاᣍغ تــری ئۆپیــك توانیــان تەنیــا قەرەبــۆی2( كەمبوونــەوەی نــەوت لــە بــازاڕەكاᣍغ جیهانــدا بــە بــڕی بەرمیــل لــە نــەوᣍت ئــرᢀێ ان بكەنــەوە . HUJ-Volume 5, Issue 4, December 2020 بــۆ نموونــە كاتێــك نــرᣐغ فــەریم نــەوᣍت )105،ص1٩81،٧) دۆالر .(اسماعیل-) دۆالر بووە بۆ هەر بەرمیلێك بەهای ڕاستەقینەی دابەزیوە بۆ نێوان (٦12.5( دا1ئۆپیك لە ساᣡڵ ٧٧٩ دا كاتێــك بڕیــاری كەمكردنــەوەی دەســت بــە جــیᢀ ێ 1٧٩3 بەكارهێنــاᣍغ نــەوت وەك چەكێــی ســیاس لــە الیــەن دەوڵەتــە عەرەبیەكانــەوە لــە ســاᣡڵ-د ی كەمكــردەوە) و دواتــر %10 ) دا (ســعوودیە بــە ڕێــژەی%5( لــە كوەیــت بــە ڕێــژەی1٧٩3 ی تᣄژ یــیغ یەكــەیم ســاᣡڵ1بەرهەمهێنــاᣍغ نەوتیــان لــە ٧ ) بەرزكــردەوە و جەختیــان %25( ی تᣄژ یــیغ دووەم لــە كوەیــت بەســرت ا ڕێــژەی كەمكردنــەوەی بەرهەمهێنــاᣍغ نەوتیــان بــۆ5-4 لــەو كۆنگرەیــەی كــە لــە لــە ســەر بڕیــاری پێشــویان كــردەوە كــە تایبــەت بــوو بــە ڕاگرتــیغ هەناردەكــردᣍغ نــەوت بــۆ هەریــەك لــە ویالیەتەیەكگرتــووەكاᣍغ ئەمریــكا و هۆڵەنــدا ئــەم بڕیارانــە بوونــە مایــەی : فراوانبووᣍغ گاپ لە نێوان دەرخستە و خواسیت جیهاᣍغ نەوت .-1 ەرمایەدارەكان كەم ونەوی نەوᣍت كۆگاك او لە دەوڵەتە پیشە ازی ە2 1.32 1.33 1.36 1.42 1.45 1.5 1.52 1.57 1.63 1.21 1.7 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 نرخی نەوت ( دۆالر / بەرمیل ) 31 HUJ-Volume 5, Issue 4, December 2020 317 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq https://www.huj.uoh.edu.iq 317 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq 1٧٩1-1٦٩1 )هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ1( شێوەی 1.32 1.33 1.36 1.42 1.45 1.5 1.52 1.57 1.63 1.21 1.7 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 نرخی نەوت ( دۆالر / بەرمیل ) 1.32 1.33 1.36 1.42 1.45 1.5 1.52 1.57 1.63 1.21 1.7 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 نرخی نەوت ( دۆالر / بەرمیل ) )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دەگەڕێتەوە بۆ كۆمەڵە هۆكارێك كە بریتᣌێ غ لە : 1٩80-1٧٩2قۆناᣘغ دووەم: هەڵكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە حەفتاكاᣍغ سەدەی ڕابردوو بەهرᢀێ غبــووᣍغ ســەنیگ ئۆپیــك لــە بــواری گفتوگۆكــردن لــە گــەڵ كۆمپانیــا نەوتیــەكان لــە ئەنجــایم بــاᢼڵ دەســیت ئۆپیــك لــە بــواری هەناردەكــردᣍغ نــەوت -أ ی نــەوᣍت هەناردەكــراو لــە جیهانــدا لــە نیــوەی یەكــەیم حەفتاكانــدا) و لــە ئەنجــایم گۆڕیــیغ سیاســەᣍت نــەوᣍت هەندێــك لــە %٩0 لــە جیهانــدا (زیاتــر لــە دەوڵەتــاᣍغ ئۆپیــك بــە هــۆی گۆڕیــیغ رژێمــە سیاســیەكانیان . بەرزبوونــەوی هۆشــیاری نــەوᣍت الی ئەندامــاᣍغ ئۆپیــك و هەســتكردنیان بــەو ســتەمەی كــە لێانكــراوە لــە بــواری دیــاری كــردᣍغ نــرᣐغ نــەوت دا كــە -ب نرخێــی دادپەروەرانــە نەبــوو وە پێویســتە بەرزبكرێتــەوە بــە تایبــەᣍت لــە بــەر ئــەوەی كــە نــەوت ســەرچاوەیەیک لــە بــن هاتــووە و نــوێ نابێتــەوە . HUJ-Volume 5, Issue 4, December 2020 كە دوو دەوڵەᣍت گرنیگ هەناردەكەری نەوت بوون .1٩80 ئرᢀێ ان لە ساᣡڵ- هەڵگرێ ساᣍغ جەنیگ عرᢀێ اق-و خۆماڵیكردᣍغ نەوت لە الیەن هەندێك لە ئەنداماᣍغ ئۆپیك جەزائرێ و عرᢀێ اق بە نموونە .-ی Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 8 https://www.huj.uoh.edu.iq HUJ-Volume 5, Issue 4, December 2020 ئــەو هۆكارانــەی كــە ئاماژەیــان بــۆ كــرا بەرپرســن لــە هەڵكشــاᣍغ نــرᣐغ نــەوت لــە حەفتــاكاᣍغ ســەدەی ڕابــردوو دا و بــووە مایــەی بەرزبوونــەوەی نــرᣐغ ).2 دا (شــێوەی1٩80 ) دۆالر/بەرمیــل لــە ســاᣡڵ35.52( بــۆ1٧٩2 ) دۆالر/بەرمیــل لــە ســاᣡڵ1.82( نــەوᣍت راســتەقینەی ئۆپیــك لــە 1٩80-1٧٩2 ) هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ2( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی لە ئەنجایم ئەو هۆكارانەی كە ئاماژەیان بۆ كرا ســەنیگ ئۆپیك لە بواری دیاریكردᣍغ نرᣐغ نەوتدا ســاڵ لە دوای ســاڵ بەهرᢀێ غتر دەبوو و دەتوانرێت بووترێــت كــە بــازاڕی نــەوت لــە حەفتاكانــدا لــە بــازاڕی كڕیــارەكان (هاوردەكــەرەكان) گــۆڕا بــۆ بــازاڕی فرۆشــیاران (هەنــاردەكاران) بۆیــە دەكــرا نــرᣐغ هەڵكشــاᣍغ زیاتــر بەخۆیــەوە ببینێــت گــەر ئــەم دوو هــۆكارە نەبوایــە :1٩80-1٧٩2 نــەوت لــەو ماوەیــەدا دا كــە بــە هەمــوو تواناكانیــەوە كاری دەكــرد بــۆ ڕێگرتــن لــە هەڵكشــاᣍغ نــرᣐغ نــەوᣍت 1٧٩4 دامەزرانــدᣍغ ئاژانــی وزەی نێودەوڵــەᣍت لــە كۆتــاᣍی ســاᣡڵ- . ئۆپیــك ناكۆكیەكاᣍغ نێوان ئەنداماᣍغ ئۆپیك كە ڕێگر بوو لە بەدیهێناᣍغ یەك ڕیزی هەڵویســتیان دەربارەی نرᣐغ نەوت . ئەم ناكۆكیانە لە دوو هەڵویســیت - ، ئەبــو زەᣍ كانــووᣍغ یەكــەیم1، كاراكاس كانــووᣍغ یەكــەیم ٧٧٩1ئەندامــاᣍغ ئۆپیــك بــەدی دەكرێــت لــە كۆنگــرەكاᣍغ (الدوحــە لــە كانــووᣍغ یەكــەیم ٦٧٩ 1.82 2.07 11 10.43 11.6 12.5 12.79 29.19 35.52 1972 1973 1974 1975 1976 1977 1978 1979 1980 نرخی نەوت ( دۆالر / بەرمیل ) HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq ئــەو هۆكارانــەی كــە ئاماژەیــان بــۆ كــرا بەرپرســن لــە هەڵكشــاᣍغ نــرᣐغ نــەوت لــە حەفتــاكاᣍغ ســەدەی ڕابــردوو دا و بــووە مایــەی بەرزبوونــەوەی نــرᣐغ ).2 دا (شــێوەی1٩80 ) دۆالر/بەرمیــل لــە ســاᣡڵ35.52( بــۆ1٧٩2 ) دۆالر/بەرمیــل لــە ســاᣡڵ1.82( نــەوᣍت راســتەقینەی ئۆپیــك لــە 1.82 2.07 11 10.43 11.6 12.5 12.79 29.19 35.52 1972 1973 1974 1975 1976 1977 1978 1979 1980 نرخی نەوت ( دۆالر / بەرمیل ) نرخی نەوت ( دۆالر / بەرمیل ) )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی لە ئەنجایم ئەو هۆكارانەی كە ئاماژەیان بۆ كرا ســەنیگ ئۆپیك لە بواری دیاریكردᣍغ نرᣐغ نەوتدا ســاڵ لە دوای ســاڵ بەهرᢀێ غتر دەبوو و دەتوانرێت بووترێــت كــە بــازاڕی نــەوت لــە حەفتاكانــدا لــە بــازاڕی كڕیــارەكان (هاوردەكــەرەكان) گــۆڕا بــۆ بــازاڕی فرۆشــیاران (هەنــاردەكاران) بۆیــە دەكــرا نــرᣐغ هەڵكشــاᣍغ زیاتــر بەخۆیــەوە ببینێــت گــەر ئــەم دوو هــۆكارە نەبوایــە :1٩80-1٧٩2 نــەوت لــەو ماوەیــەدا دا كــە بــە هەمــوو تواناكانیــەوە كاری دەكــرد بــۆ ڕێگرتــن لــە هەڵكشــاᣍغ نــرᣐغ نــەوᣍت 1٧٩4 دامەزرانــدᣍغ ئاژانــی وزەی نێودەوڵــەᣍت لــە كۆتــاᣍی ســاᣡڵ- . ئۆپیــك Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University ناكۆكیەكاᣍغ نێوان ئەنداماᣍغ ئۆپیك كە ڕێگر بوو لە بەدیهێناᣍغ یەك ڕیزی هەڵویســتیان دەربارەی نرᣐغ نەوت . ئەم ناكۆكیانە لە دوو هەڵویســیت - ، ئەبــو زەᣍب كانــووᣍغ یەكــەیم1، كاراكاس كانــووᣍغ یەكــەیم ٧٧٩1ئەندامــاᣍغ ئۆپیــك بــەدی دەكرێــت لــە كۆنگــرەكاᣍغ (الدوحــە لــە كانــووᣍغ یەكــەیم ٦٧٩ )105،ص1٩81) كــە ڕەنگدانــەوەی هەبــوو لــە ســەر بڕیــارەكاᣍغ ئۆپیــی تایبــەت بــە دیاریكــردᣍغ نــرᣐغ نــەوت .(اســماعیل،ا1٧٩8 و هەڵكشــان و داكشــاᣍغ لــە نیــوەی1٩8٦-1٩81 قۆنــاᣘغ ســێیەم: داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــە نیــوەی یەكــەیم هەشــتاكاᣍغ ســەدەی ڕابــردوو دەگەڕێتــەوە بــۆ :1٩٩0-1٩8دووەمیــدا ٧ بــەردەوام بــوو، بـەᢼڵ م كاریگەریــەᣍت نەبــوو لــە ســەر 1٩8٦-1٩81 هەرچەنــدە جەنــیگ دوو دەوڵــەᣍت ســەرەیک ئەنــدایم ئۆپیـك لــە مـاوەی نێــوان ســاᢼڵᣍغ-أ بــڕی خســتنەڕووی نــەوت لــە بــازاڕەكاᣍغ جیهانــدا بەهــۆی : پابەنــد نەبــووᣍغ ئەندامــاᣍغ ئۆپیـك بــە بەشــە نــەوᣍت دیاریكــراو بۆیــان لــە الیــەن ڕێكخراوەكــەوە، بــە تایبــەᣍت ســعوودیە كــە بــڕی بەرهەمهێنــاᣍغ لــە ســاᣡڵ - .) ملیــۆن بەرمیــل /رۆژانــە5( دا بەرزكــردەوە بــۆ1٩8٦ عــرᢀێ اق كــە الیەنێــی جەنــگ بــوو تــواᣍغ زاڵ بێــت بــە ســەر كێشـەی هەناردەكــردᣍغ نــەوت لــە ڕێــگای كەنــداوی عــەرەب و بۆریــە نەوتیــەكان كــە بــە ســوریا - و لوبنــان تێدەپــەڕن لــە ڕێــگای هەناردەكــردᣍغ نــەوت بــە خــاىک ســعودیە بــۆ بەنــدەری (ینبــع) لــە ســەر دەریــای ســور و زیادكــردᣍغ نــەوᣍت هەناردەكــراو بــەو بۆریانــەی كــە بــە توركیــا تێدەپەڕیــن . زیادبووᣍغ بڕی نەوᣍت بەرهەمهێرغ او و هەناردەكراو لە الیەن دەوڵەتاᣍغ دەرەوەی ئۆپیك .- ناكۆكیەكاᣍغ نێوان ئەنداماᣍغ ئۆپیك كە ڕێگر بوو لە بەدیهێناᣍغ یەك ڕیزی هەڵویســتیان دەربارەی نرᣐغ نەوت . https://www.huj.uoh.edu.iq ئەم ناكۆكیانە لە دوو هەڵویســیت - ، ئەبــو زەᣍب كانــووᣍغ یەكــەیم1، كاراكاس كانــووᣍغ یەكــەیم ٧٧٩1ئەندامــاᣍغ ئۆپیــك بــەدی دەكرێــت لــە كۆنگــرەكاᣍغ (الدوحــە لــە كانــووᣍغ یەكــەیم ٦٧٩ )105،ص1٩81) كــە ڕەنگدانــەوەی هەبــوو لــە ســەر بڕیــارەكاᣍغ ئۆپیــی تایبــەت بــە دیاریكــردᣍغ نــرᣐغ نــەوت .(اســماعیل،ا1٧٩8 و هەڵكشــان و داكشــاᣍغ لــە نیــوەی1٩8٦-1٩81 قۆنــاᣘغ ســێیەم: داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــە نیــوەی یەكــەیم هەشــتاكاᣍغ ســەدەی ڕابــردوو دەگەڕێتــەوە بــۆ :1٩٩0-1٩8دووەمیــدا ٧ ناكۆكیەكاᣍغ نێوان ئەنداماᣍغ ئۆپیك كە ڕێگر بوو لە بەدیهێناᣍغ یەك ڕیزی هەڵویســتیان دەربارەی نرᣐغ نەوت . ئەم ناكۆكیانە لە دوو هەڵویســیت - ، ئەبــو زەᣍب كانــووᣍغ یەكــەیم1، كاراكاس كانــووᣍغ یەكــەیم ٧٧٩1ئەندامــاᣍغ ئۆپیــك بــەدی دەكرێــت لــە كۆنگــرەكاᣍغ (الدوحــە لــە كانــووᣍغ یەكــەیم ٦٧٩ )105،ص1٩81) كــە ڕەنگدانــەوەی هەبــوو لــە ســەر بڕیــارەكاᣍغ ئۆپیــی تایبــەت بــە دیاریكــردᣍغ نــرᣐغ نــەوت .(اســماعیل،ا1٧٩8 پابەنــد نەبــووᣍغ ئەندامــاᣍغ ئۆپیـك بــە بەشــە نــەوᣍت دیاریكــراو بۆیــان لــە الیــەن ڕێكخراوەكــەوە، بــە تایبــەᣍت ســعوودیە كــە بــڕی بەرهەمهێنــاᣍغ لــە ســاᣡڵ - .) ملیــۆن بەرمیــل /رۆژانــە5( دا بەرزكــردەوە بــۆ1٩8٦ عــرᢀێ اق كــە الیەنێــی جەنــگ بــوو تــواᣍغ زاڵ بێــت بــە ســەر كێشـەی هەناردەكــردᣍغ نــەوت لــە ڕێــگای كەنــداوی عــەرەب و بۆریــە نەوتیــەكان كــە بــە ســوریا - و لوبنــان تێدەپــەڕن لــە ڕێــگای هەناردەكــردᣍغ نــەوت بــە خــاىک ســعودیە بــۆ بەنــدەری (ینبــع) لــە ســەر دەریــای ســور و زیادكــردᣍغ نــەوᣍت هەناردەكــراو بــەو بۆریانــەی كــە بــە توركیــا تێدەپەڕیــن . زیادبووᣍغ بڕی نەوᣍت بەرهەمهێرغ او و هەناردەكراو لە الیەن دەوڵەتاᣍغ دەرەوەی ئۆپیك .- Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University زیادبووᣍغ بڕی نەوᣍت بەرهەمهێرغ او و هەناردەكراو لە الیەن دەوڵەتاᣍغ دەرەوەی ئۆپیك .- Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University زیادبووᣍ بڕی نەوᣍ بەرهەمهێراو و هەناردەكراو لە الیەن دەوڵەتاᣍ دەرەوەی ئۆپیك .- Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 https://www.huj.uoh.edu.iq دا (شــێوەی 1٩٩8 ) دۆالر لــە ســاᣡڵ12.28( دا بـۆ1) دۆالر بــۆ هــەر بەرمیلێـك لــە ســاᣡڵ ٦٩٩20.2داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیـك لــە زیاتــر لــە (٩-ب ٩)ی نەتــەوە 8)، دەگەڕێتــەوە بــۆ ڕووداᣍغ قەیــراᣍغ ئابــوری لــە ئاسـیا و دووبــارە هەناردەكــردᣍغ نــەوᣍت عــرᢀێ اق بــۆ بــازاڕە جیهانیــەكان بــە ᣍچᢀ ی بڕیــاری (٦4 . یەكگرتــوەكاᣍغ تایبــەت بــە پرۆگــرایم نــەوت بەرامبــەر خــۆراك 2000 ) دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ28( بــۆ نزیكــەی1٩٩8 ) دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ12.28( هەڵكشــاᣍغ نــرᣐغ نــەوت لــە-ج دا دەگەڕێتــەوە بــۆ ڕێككەوتــیغ نێــوان ئۆپیــك و دەوڵەتــاᣍغ هەناردەكــەری نــەوت لــە دەرەوەی ئۆپیــك بــۆ نموونــە (مەكســیك و نەرویــــــج ) لــە ســەر )5،ص200كەمكردنــەوەی بەرهەمهێنانیــان بــە مەبەســیت بەرزكردنــەوەی نــرᣐغ نــەوت. (العــرغ غی،٧ ناكرێت رۆڵ و كاریگەرᣍی سیاســەᣍت نەوᣍت روسـیا بەالوە برغ ێت كە كاریگەری نەرێیغ لە ســەر دیاریكردᣍغ نرᣐغ نەوᣍت ئۆپیك و داكشــاᣍغ بە گشــیت -د لــە نــەوەدەكاᣍغ ســەدەی ڕابــردوو دا هەبــووە .روســیا بــۆ زاڵبــوون بەســەر قەیرانــە ئابــوری و داراییــەكاᣍغ دا بڕێــی زۆری نــەوᣍت هەنــاردەی بــازاڕەكاᣍغ 34 32.38 29.04 28.02 27.01 13.53 17.73 14.24 17.31 22.26 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 نرخی نەوت ( دۆالر / بەرمیل ) 34 32.38 29.04 28.02 27.01 13.53 17.73 14.24 17.31 22.26 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 نرخی نەوت ( دۆالر / بەرمیل ) نرخی نەوت ( دۆالر / بەرمیل ) )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دەگەڕێتەوە بۆ :2000-1٩٩1 قۆناᣘغ چوارەم:داكشان و هەڵكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە دە ساᣡڵ كۆتاᣍی سەدەی ڕابردوو دەگەڕێتــەوە بــۆ الوازی ســەنیگ 1٩٩5-1٩٩1 داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك گــەر بــە ڕێژەیــەیک كەمیــش بووبێــت لــە نیــوەی یەكــەیم ئــەم دە سـاڵە دا-أ ئۆپیــك لــە بــواری دیاریكــردᣍغ نــرᣐغ نــەوت دا بــە هــۆی قووڵبوونــەوەی ناكۆكیــەكاᣍغ نێــوان ئەندامــاᣍغ لــە دورگــەی عــەرەب و ســەر كەنــداوی عــەرەᣍب . وەك سوپاســگوزاری بەرامبــەر هەڵویســیت هاوپەیمانــان لــە جەنــیگ كەنــداوی دووەمــدا (جەنــیگ رزگار كــردᣍغ كوەیــت) هەندێــك لــە دەوڵەتــاᣍغ عــەرەᣍب ئەنــدایم ئۆپیــك (ســعودیە،ئیمارات،كوەیت،قەتەر) پەیــرەوی سیاســەتێی نەوتیــان دەكــرد كــە بەرژەوەندیــەكاᣍغ ئەندامــاᣍغ ئاژانــی وزەی نێودەوڵەتیــان تیــا ڕەچــاو كرابێــت لــە بــواری دیاریكــردᣍغ نــرᣐغ نــەوت دا، كــە ڕێگــر بــوو لــە هەڵكشــاᣍغ نــرᣐغ نــەوت لــەو ماوەیــە دا . دا (شــێوەی 1٩٩8 ) دۆالر لــە ســاᣡڵ12.28( دا بـۆ1) دۆالر بــۆ هــەر بەرمیلێـك لــە ســاᣡڵ ٦٩٩20.2داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیـك لــە زیاتــر لــە (٩-ب ٩)ی نەتــەوە 8)، دەگەڕێتــەوە بــۆ ڕووداᣍغ قەیــراᣍغ ئابــوری لــە ئاسـیا و دووبــارە هەناردەكــردᣍغ نــەوᣍت عــرᢀێ اق بــۆ بــازاڕە جیهانیــەكان بــە ᣍچᢀ ی بڕیــاری (٦4 . https://www.huj.uoh.edu.iq : دەگەڕێتەوە بۆ1٩٩0-1٩8 هەڵكشان و داكشاᣍغ نرᣐغ نەوت لە نێوان ساڵەكاᣍغ ٧-ب ، دەگەڕێتــەوە بــۆ ڕێككەوتــیغ ئەندامــاᣍغ ئۆپیــك لــە ســەر جێگرێ كــردᣍغ 1٩8 دا، بــە بــەراورد لــە گــەڵ ســاᣡڵ ٦1٩8 هەڵكشــاᣍغ نــرᣐغ نــەوت لــە ســاᣡڵ ٧- )3 دا .(شــێوەی1٩8) دۆالر بــۆ هــەر بەرمیلێــك لــە كۆتــاᣍی ســاᣡڵ ٦18( نــرᣐغ نــەوت لــە ئرᢀێ ان - دەگەڕێتەوە بۆ كۆتاᣍی هاتیغ شەڕی عرᢀێ اق1٩8 دا، بە بەراورد لە گەڵ ساᣡڵ ٧1٩88 داكشاᣍغ نرᣐغ نەوت لە ساᣡڵ- دا دەگەڕێتــەوە بــۆ ئاڵۆزییــەكاᣍغ نێــوان دەوڵەتــاᣍغ كەنــداو كــە بــە داگرێ كــردᣍغ كوەیــت1٩٩0 و1٩8 هەڵكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــە ســاڵەكاᣍغ ٩- . لــە الیــەن عرᢀێ اقــەوە كۆتــاᣍی هــات و گەمــاڕۆی نێودەوڵــەᣍت خرایــە ســەر هەناردەكــردᣍغ نــەوᣍت عــرᢀێ اق 1٩٩0-1٩81 )هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ3( شێوەی : دەگەڕێتەوە بۆ1٩٩0-1٩8 هەڵكشان و داكشاᣍ نرᣐ نەوت لە نێوان ساڵەكاᣍ ٧-ب ، دەگەڕێتــەوە بــۆ ڕێككەوتــیغ ئەندامــاᣍغ ئۆپیــك لــە ســەر جێگرێ كــردᣍغ 1٩8 دا، بــە بــەراورد لــە گــەڵ ســاᣡڵ ٦1٩8 هەڵكشــاᣍغ نــرᣐغ نــەوت لــە ســاᣡڵ ٧- )3 دا .(شــێوەی1٩8) دۆالر بــۆ هــەر بەرمیلێــك لــە كۆتــاᣍی ســاᣡڵ ٦18( نــرᣐغ نــەوت لــە ئرᢀێ ان - دەگەڕێتەوە بۆ كۆتاᣍی هاتیغ شەڕی عرᢀێ اق1٩8 دا، بە بەراورد لە گەڵ ساᣡڵ ٧1٩88 داكشاᣍغ نرᣐغ نەوت لە ساᣡڵ- دا دەگەڕێتــەوە بــۆ ئاڵۆزییــەكاᣍغ نێــوان دەوڵەتــاᣍغ كەنــداو كــە بــە داگرێ كــردᣍغ كوەیــت1٩٩0 و1٩8 هەڵكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك لــە ســاڵەكاᣍغ ٩- . لــە الیــەن عرᢀێ اقــەوە كۆتــاᣍی هــات و گەمــاڕۆی نێودەوڵــەᣍت خرایــە ســەر هەناردەكــردᣍغ نــەوᣍت عــرᢀێ اق 1٩٩0-1٩81 )هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ3( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دەگەڕێتەوە بۆ :2000-1٩٩1 قۆناᣘغ چوارەم:داكشان و هەڵكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە دە ساᣡڵ كۆتاᣍی سەدەی ڕابردوو دەگەڕێتــەوە بــۆ الوازی ســەنیگ 1٩٩5-1٩٩1 داكشــاᣍغ نــرᣐغ نــەوᣍت ئۆپیــك گــەر بــە ڕێژەیــەیک كەمیــش بووبێــت لــە نیــوەی یەكــەیم ئــەم دە سـاڵە دا-أ ئۆپیــك لــە بــواری دیاریكــردᣍغ نــرᣐغ نــەوت دا بــە هــۆی قووڵبوونــەوەی ناكۆكیــەكاᣍغ نێــوان ئەندامــاᣍغ لــە دورگــەی عــەرەب و ســەر كەنــداوی عــەرەᣍب . وەك سوپاســگوزاری بەرامبــەر هەڵویســیت هاوپەیمانــان لــە جەنــیگ كەنــداوی دووەمــدا (جەنــیگ رزگار كــردᣍغ كوەیــت) هەندێــك لــە دەوڵەتــاᣍغ عــەرەᣍب ئەنــدایم ئۆپیــك (ســعودیە،ئیمارات،كوەیت،قەتەر) پەیــرەوی سیاســەتێی نەوتیــان دەكــرد كــە بەرژەوەندیــەكاᣍغ ئەندامــاᣍغ ئاژانــی وزەی نێودەوڵەتیــان تیــا ڕەچــاو كرابێــت لــە بــواری دیاریكــردᣍغ نــرᣐغ نــەوت دا، كــە ڕێگــر بــوو لــە هەڵكشــاᣍغ نــرᣐغ نــەوت لــەو ماوەیــە دا . https://www.huj.uoh.edu.iq یەكگرتــوەكاᣍغ تایبــەت بــە پرۆگــرایم نــەوت بەرامبــەر خــۆراك 2000 ) دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ28( بــۆ نزیكــەی1٩٩8 ) دۆالر بــۆ هــەر بەرمیلێــك لــە ســاᣡڵ12.28( هەڵكشــاᣍغ نــرᣐغ نــەوت لــە-ج دا دەگەڕێتــەوە بــۆ ڕێككەوتــیغ نێــوان ئۆپیــك و دەوڵەتــاᣍغ هەناردەكــەری نــەوت لــە دەرەوەی ئۆپیــك بــۆ نموونــە (مەكســیك و نەرویــــــج ) لــە ســەر )5،ص200كەمكردنــەوەی بەرهەمهێنانیــان بــە مەبەســیت بەرزكردنــەوەی نــرᣐغ نــەوت. (العــرغ غی،٧ ناكرێت رۆڵ و كاریگەرᣍی سیاســەᣍت نەوᣍت روسـیا بەالوە برغ ێت كە كاریگەری نەرێیغ لە ســەر دیاریكردᣍغ نرᣐغ نەوᣍت ئۆپیك و داكشــاᣍغ بە گشــیت -د لــە نــەوەدەكاᣍغ ســەدەی ڕابــردوو دا هەبــووە .روســیا بــۆ زاڵبــوون بەســەر قەیرانــە ئابــوری و داراییــەكاᣍغ دا بڕێــی زۆری نــەوᣍت هەنــاردەی بــازاڕەكاᣍغ Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 320 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq . جیهــان دەكــرد بــە مەبەســیت دابینكــردᣍغ داهــاᣍت زیاتــر لــە دراوی گرانبەهــا 2000-1٩٩1 )یهەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ4( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دەگەڕێتەوە بۆ :2013-2001 هەڵكشاᣍغ گەورە لە نرᣐغ نەوᣍت ئۆپیك دا لە نێوان ساڵەكاᣍغ- :قۆناᣘغ پێنجەم نــرᣐغ نــەوت لــە ســەدەی ڕابــردوو دا نرخێــی دادپەروەرانــە نەبــووە و نەگونجـاوە لــە گــەڵ بەهــای نــەوت وەك ســەرچاوەی وزە و كەرەســتەی خــاوی -أ دەیــان كاᢼڵ ی پیشەســازی لــە الیــەك و وەك ســەرچاوەی لــە بــن هاتــوو كــە نــوێ نابێتــەوە بــە بــەراورد بــە بەهــا و نــرᣐغ ســەرچاوەكاᣍغ تــری وزە لــە الیــەیک تــر . بەردەوامبــووᣍغ دیــاردەی هەᢼڵ وســان لــە جیهانــدا، كــە بــووە مایــەی بەرزبوونــەوەی نــرᣐغ كاᢼڵ پیشەســازی و بنەڕەتیــەكان و خزمەتگوزاریــەكان كــە -ب ئەندامــاᣍغ ئۆپیــك هــاوردەی دەكــەن . 18.62 18.44 16.33 15.53 16.86 20.29 18.86 12.28 17.44 27.6 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 نرخی نەوت ( دۆالر / بەرمیل ) . جیهــان دەكــرد بــە مەبەســیت دابینكــردᣍغ داهــاᣍت زیاتــر لــە دراوی گرانبەهــا . جیهــان دەكــرد بــە مەبەســی دابینكــردᣍ داهــاᣍ زیاتــر لــە دراوی گرانبەهــا 2000-1٩٩1 )یهەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ4( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 18.62 18.44 16.33 15.53 16.86 20.29 18.86 12.28 17.44 27.6 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 نرخی نەوت ( دۆالر / بەرمیل ) 2000-1٩٩1 )یهەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ4( شێوەی 2000-1٩٩1 )یهەڵكشان و داكشاᣍ نرᣐغ نەوᣍ ئۆپیك لە ماوەی نێوان ساᢼᣍ4( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 18.62 18.44 16.33 15.53 16.86 20.29 18.86 12.28 17.44 27.6 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 نرخی نەوت ( دۆالر / بەرمیل ) )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دەگەڕێتەوە بۆ :2013-2001 هەڵكشاᣍغ گەورە لە نرᣐغ نەوᣍت ئۆپیك دا لە نێوان ساڵەكاᣍغ- :قۆناᣘغ پێنجەم نــرᣐغ نــەوت لــە ســەدەی ڕابــردوو دا نرخێــی دادپەروەرانــە نەبــووە و نەگونجـاوە لــە گــەڵ بەهــای نــەوت وەك ســەرچاوەی وزە و كەرەســتەی خــاوی -أ دەیــان كاᢼڵ ی پیشەســازی لــە الیــەك و وەك ســەرچاوەی لــە بــن هاتــوو كــە نــوێ نابێتــەوە بــە بــەراورد بــە بەهــا و نــرᣐغ ســەرچاوەكاᣍغ تــری وزە لــە الیــەیک تــر . بەردەوامبــووᣍغ دیــاردەی هەᢼڵ وســان لــە جیهانــدا، كــە بــووە مایــەی بەرزبوونــەوەی نــرᣐغ كاᢼڵ پیشەســازی و بنەڕەتیــەكان و خزمەتگوزاریــەكان كــە -ب ئەندامــاᣍغ ئۆپیــك هــاوردەی دەكــەن . https://www.huj.uoh.edu.iq دابەزییغ بەهای دۆالر بەرامبەر دراوە گرانبەهاكاᣍغ تر بۆ نموونە (یۆرۆ).-ج ڕووداوەكاᣍغ ڕۆژهەᢼڵᣍت ناوەڕاست كە برییت بوون لە :-د ᣡ وی ێی م ج ی .ئاڵۆزی پەیوەندیەكاᣍغ نێوان والیەتە یەكگرتوەكاᣍغ ئەمریكا و دەوڵەتاᣍغ ڕۆژئاوا لە گەڵ ئرᢀێ ان بە هۆی پرۆگرامە ناوەكیەكەی- گۆڕانكارییــە سیاســیەكاᣍغ نێــو دەوڵەتــاᣍغ عــەرەᣍب ، لــە ئەنجــایم ئــەوەی ᣍچᢀ دەوترێــت (ســەرهەڵداᣍغ بەهــاری عــەرەᣍب ) كــە بــووە مایــەی مەتــرس بــۆ- . ئەندامــاᣍغ ئاژانــی وزەی نیۆدەوڵــەᣍت و زیادبــووᣍغ خواســتیان لــە ســەر كۆگاكــردᣍغ نــەوت و 200 زیــاد بــووᣍغ خواســت لــە ســەر نــەوت لــە ئەنجــایم گەشــەی ئابــوری بــە تایبــەᣍت لــە چــᣌێ غ و هینــد .گەشــەی ئابــوری چــᣌێ غ لــە ســاڵەكاᣍغ ٩-ه )1) بــووە .(دبیە،المصــدر الســابق،ص٧%10( ) و%8( یــەك لــە دوای یــەك2010 ٩) دۆالر بوو بۆ هەر بەرمیلێك4( دا زیاتر لە2008 ٦) دۆالر بۆ هەر بەرمیلێك، كە لە ساᣡڵ1( بۆ كەمرت لە200داكشاᣍغ نرᣐغ نەوت لە ساᣡڵ ٩ ) ئەمــەش دەگەڕێتــەوە بــۆ قەیــراᣍغ داراᣍی جیهــاᣍغ لــە ئەنجــایم مایەپــووچ بــووᣍغ چەندیــن بانــی گــەورەی ئەمریــی و هەروەهــا داڕمــاᣍغ 5 (شــێوەی كاریگەریــەᣍت بــەردەوایم لــە ســەر ئابــوری جیهــان و 2010 ســەری هەڵــدا و تــا ســاᣡڵ2008 كۆمپانیــاكاᣍغ ڕەهــیغ خانووبــەرە لــە نیــوەی دووەیم ســاᣡڵ An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 321 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq .نــرᣐغ نــەوت هەبــوو 2012-2001 )هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ5( شێوەی 2012-2001 )هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼᣍغ5( شێوەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی و بــەدواوەوە (شــێوەی ٦) دەگەڕێتــەوە بــۆ زیــاد بــووᣍغ خســتنەڕووی نــەوت لــە بــازاڕەكاᣍغ 2011 داكشــاᣍغ نــرᣐغ نــەوت لــە ســاᣡڵ- : قۆنــاᣘغ شەشــەم جیهانــدا لــە ئەنجــایم : ڕێكنەكەوتــیغ ئەندامــاᣍغ ئۆپیــك لــە ســەر بەرنامەیــەیک تایبــەت بــە بەرهەمهێنــاᣍغ نــەوت و زیــاد بــووᣍغ بەرهەمهێنــاᣍغ هەندێــك لــە ئەندامــاᣍغ بــە -أ ) ملیــۆن 10.2( بەرزكــردەوە بــۆ نزیكــەی2013 شــێوەیەیک بەرچاو.بــۆ نموونــە ســعودیە بەرهەمهێنــاᣍغ لــە (٦.٩)ملیــۆن بەرمیــل لــە ڕۆژێكــدا لــە ســاᣡڵ ) ملیــۆن بەرمیــل .3.5( ) ملیــۆن بەرمیــل بەرزكــردەوە بــۆ2. . عرᢀێ اقیــش لــە هەمــان مــاوەدا بەرهەمهێنــاᣍغ نــەوᣍت لــە (٩2015 بەرمیــل لــە ســاᣡڵ زیــاد بــووᣍغ بەرهەمهێنــاᣍغ نــەوت لــە الیــەن دەوڵەتــاᣍغ بەرهەمهێنــەری نــەوت لــە دەرەوەی ئۆپیــك .بــۆ نموونــە بەرهەمهێنــاᣍغ نــەوت لــە والیەتــە -ب .٩) ملیــۆن بەرمیــل . https://www.huj.uoh.edu.iq بەرازیلیــش لــە هەمــان مــاوەدا بــڕی 4( زیــادی كــرد بــۆ2013 .٧) ملیــۆن بەرمیــل ڕۆژانــەوە لــە ســاᣡڵ4( یەكگرتــوەكاᣍغ ئەمریــكا لــە ) ملیــۆن بەرمیــل .2.4( ) ملیــۆن بەرمیلــەوە ڕۆژانــە زیــاد كــرد بــۆ2( بەرهەمهێنــاᣍغ نــەوᣍت لــە 2015-2011 شێوەی (٦) هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ 23.12 24.36 28.01 36.05 50.59 61 69.04 94.1 60.86 77.38 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 نرخی نەوت ( دوالر / برمیل ) 23.12 24.36 28.01 36.05 50.59 61 69.04 94.1 60.86 77.38 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 نرخی نەوت ( دوالر / برمیل ) نرخی نەوت ( دوالر / برمیل ) گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی و بــەدواوەوە (شــێوەی ٦) دەگەڕێتــەوە بــۆ زیــاد بــووᣍغ خســتنەڕووی نــەوت لــە بــازاڕەكاᣍغ 2011 داكشــاᣍغ نــرᣐغ نــەوت لــە ســاᣡڵ- : قۆنــاᣘغ شەشــەم جیهانــدا لــە ئەنجــایم : ڕێكنەكەوتــیغ ئەندامــاᣍغ ئۆپیــك لــە ســەر بەرنامەیــەیک تایبــەت بــە بەرهەمهێنــاᣍغ نــەوت و زیــاد بــووᣍغ بەرهەمهێنــاᣍغ هەندێــك لــە ئەندامــاᣍغ بــە -أ ) ملیــۆن 10.2( بەرزكــردەوە بــۆ نزیكــەی2013 شــێوەیەیک بەرچاو.بــۆ نموونــە ســعودیە بەرهەمهێنــاᣍغ لــە (٦.٩)ملیــۆن بەرمیــل لــە ڕۆژێكــدا لــە ســاᣡڵ ) ملیــۆن بەرمیــل .3.5( ) ملیــۆن بەرمیــل بەرزكــردەوە بــۆ2. . عرᢀێ اقیــش لــە هەمــان مــاوەدا بەرهەمهێنــاᣍغ نــەوᣍت لــە (٩2015 بەرمیــل لــە ســاᣡڵ زیــاد بــووᣍغ بەرهەمهێنــاᣍغ نــەوت لــە الیــەن دەوڵەتــاᣍغ بەرهەمهێنــەری نــەوت لــە دەرەوەی ئۆپیــك .بــۆ نموونــە بەرهەمهێنــاᣍغ نــەوت لــە والیەتــە -ب .٩) ملیــۆن بەرمیــل . بەرازیلیــش لــە هەمــان مــاوەدا بــڕی 4( زیــادی كــرد بــۆ2013 .٧) ملیــۆن بەرمیــل ڕۆژانــەوە لــە ســاᣡڵ4( یەكگرتــوەكاᣍغ ئەمریــكا لــە ) ملیــۆن بەرمیــل .2.4( ) ملیــۆن بەرمیلــەوە ڕۆژانــە زیــاد كــرد بــۆ2( بەرهەمهێنــاᣍغ نــەوᣍت لــە 2015-2011 شێوەی (٦) هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 23.12 24.36 28.01 36.05 50.59 61 69.04 60.86 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 107.46 109.87 105.87 96.29 49.49 2011 2012 2013 2014 2015 نرخی نەوت ( دوالر / برمیل ) )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌ بە خشتەی و بــەدواوەوە (شــێوەی ٦) دەگەڕێتــەوە بــۆ زیــاد بــووᣍغ خســتنەڕووی نــەوت لــە بــازاڕەكاᣍغ 2011 داكشــاᣍغ نــرᣐغ نــەوت لــە ســاᣡڵ- : قۆنــاᣘغ شەشــەم جیهانــدا لــە ئەنجــایم : ڕێكنەكەوتــیغ ئەندامــاᣍغ ئۆپیــك لــە ســەر بەرنامەیــەیک تایبــەت بــە بەرهەمهێنــاᣍغ نــەوت و زیــاد بــووᣍغ بەرهەمهێنــاᣍغ هەندێــك لــە ئەندامــاᣍغ بــە -أ ) ملیــۆن 10.2( بەرزكــردەوە بــۆ نزیكــەی2013 شــێوەیەیک بەرچاو.بــۆ نموونــە ســعودیە بەرهەمهێنــاᣍغ لــە (٦.٩)ملیــۆن بەرمیــل لــە ڕۆژێكــدا لــە ســاᣡڵ ) ملیــۆن بەرمیــل .3.5( ) ملیــۆن بەرمیــل بەرزكــردەوە بــۆ2. . HUJ-Volume 5, Issue 4, December 2020 : و بەدواوەوە201هەڵكشان وداكشاᣍغ نرᣐغ نەوت لە ساᣡڵ ٦- :قۆناᣘغ حەوتم ) دوالر خــۆی 2٧-31( لــەم قۆناغــەدا جاریكیــرت نــرᣐغ نــەوت دابەزینــی مــرᢀێ ژ ووی وبەرچـاوی بــە خۆیــەوە بیــیغ بەجــۆری هەڵبــەز ودابەزیــیغ كەلەنێــوان گرتــەوە بروانــه شــيوە ى زمــارە (٧)،ئــه مــه بەكەمرت يــن بەهــا دادەنرێــت نــەوت لەمــاوەی دوانــزە ســاᣡڵ رابــردوو نــەوت بەخۆیــەوە بینبیت ،هەرئەمــەش پــاᣡڵ بەوᢼڵ تــاᣍغ بەرهــەم هێنەرنــا كــەوا داوای كۆبوونــەوەی بەپەلــە بكــەن تاكــو نــەرᣐغ نــەوت جێگــرێ بكــەن بەپــیᢀ ێ ڕای تایبەتمەنــدەكاᣍغ ئــەم بــۆارە دەگوڕێتــەوە بۆ"بــۆ بــووᣍغ بڕیــی زۆری نــەوت وەك خســتنەروو لــە بــازاڕەكاᣍغ جیهــان زیاتــر لــە وەی داواكاری 201دابەزیــیغ نــرᣐغ نــەوت لــە ســاᣡڵ ٦ ملیــون برمیــل رۆژانــە.5 لەســەر بــوو لەالیــەن بــە كاربــەراᣍغ نــەوت بــە رێــژەی /٦)12/201/٦https://www.aljazeera.net/ebusiness( پــاش چەندیــن كوبوونەوەیــەك ئەندامــاᣍغ ئوپیــك بەســەرۆكایەᣍت ســعودیا وᢼڵ تــاᣍغ بەرهــەم201وســەرەتاᣍی ســاᣡڵ ٧201لەكۆتاییــەكاᣍغ ســاᣡڵ ٦- ملیــون برمیــل رۆژانــە بــر1،2 ملیــون برمیــل رۆژانــە بەرێــژەی1،8 هێنــەری ســەربەخۆ گەیشــᣌت غ بــە رێكەوتــن بەدابەزانــدᣍغ بــڕی بەرهــەم بەرێــژەی ٦برمیــل رۆژانــە بــۆ وᢼڵ تــاᣍغ ســەربەخۆ،لە وریــگای ئــەم كــردارە نــرᣐغ نــەوت بەرزبوونەوەیــی بەخۆیــەوە بیــیغ بەجــۆری بەهــای 00000وᢼڵ تــاᣍغ ئوپیــك و جاریكــرت نــرᣐغ نــەوت هەڵكشــانیی تــری تۆمــار 201دوالر، ئــەم هەڵكشــانە بــەردەوام بــوو تاكــو كۆتــاᣍی ســاᣡڵ ٧55 برمیلیــك نــەوت گەیشــتە نزیكــەیtا ))t yۆا ۆ ٧دۆالر )بــۆ هەربرملیــك نــەوت تــا كاتیــی 0-دا نــرᣐغ نــەوت هەڵبــەز ودابەزینــی زۆری كــرد بەجــۆری نــرᣐغ نــەوت لەنێــوان( ٦٦2018 لەســاᣡڵ- بەهــۆی كەشــانەوەی والیەتــە یەكگرتــووەكاᣍغ ئەمریــكا لــە رێكەوتنامــەی ئەتــوویم لەگــەڵ2018 زۆر جێگرێ بوو.بــەᢼڵ م لــە نێــوەی یەكــەیم ســاᣡڵ دا نــرᣐغ هەڵكشــا بــۆ ســەرەوەی ژ ئــرێ ان،كار كــردن بــۆ زیاتــر كــردᣍغ رسغ اكان لەســەر ئــرێ ان و توندكــرردᣍغ گەمــاروی ئابۆری،نــرᣐغ نــەوت تەكانیــی بــاس ملیــون برمیــل نــەوت لــە 2،1 ٧دۆالر بــۆ هــەر برمیلیــك چونكــە ئــرێ ان بەســێیەم وᢼڵ ت دێــت لــەرووی بەرهــەم هێنانــەوە وبەشــدار دەبێــت بەرێــژەی5 ، سەرەرای فشاری ئەمریكا لەسەر وᢼڵ تاᣍغ بەرهەم هێنەر بۆ2018 رۆژیكدا.،نرᣐغ نەوت بەم شێوەیە بە بەرزی مایەوە سەرەرای بەدراێژای ساᣡڵ كــەم كردنــەوەی نــرخ لەریــگای زیادكــردᣍغ بەرهەم،بـەᢼڵ م ئەندامــاᣍغ ئوبیــك بەســەرۆكایەᣍت جەزائــر نەچوونــە ژێــر بــاᣡڵ ئــەو داوایــە ورازی نەبــوون بــە كــەم كردنــەوەی رێــژەی بەرهــەم هێنــان ودابەزانــدᣍغ نــرخ، ئەمــە بــووᣍغ دەنــیگ بــاس زیادكــرᣍغ رسغ ای زیاتربەســەر وᢼڵᣍت ئــرێ ان بــۆ هــۆی ئــەوەی بەهــای نــەوت /٩)1/201/٩https://www.aljazeera.net/news/ebusiness(دوالر بــۆ هەربرمیلیــك8هەڵبكشــێت بــۆ ســەرەوە بــگات بــە نزیكــەی ٦ )بەهاكــەی جاریكــرت داكشــانیی زۆر گــەوەرەی بــە خۆیــەوە بیــیغ بەجــۆری بەهــای برمیــیل نــەوᣍت خــاو 201 ســەبارەت بەنــرᣐغ نــەوت لەســاᣡڵ (٩- دوالر) بــۆ برمیلیــك بەشــێوەیی لەنــاكاو وچاوڕواننەكــراو دور لــە هەمــوو پێشــبییغ و خۆێندینــەوەكاᣍغ جیهــاᣍغ بــۆ دۆᣐغ5،٩4دابــەزی بــۆ نزیكــەی( ٦ ٧)-3 ،ص2018 ،نەوت.لەگــەڵ ئەمەشــدا كۆمــەᣡڵ هــۆكار هەبــوون لــە پشــت ئــەم داكشــانە بــەم شــێوەی خــۆارەوە: (الــرت یک )ملیــون بەرمیــیل نــەوᣍت ئاســاᣍی ،بەم شــێوەیە 11،ملیــون برمیــل رۆژانــە ئەمەســەرەرای (٧2 هەڵكشــاᣍغ قەبــارەی نــەوᣍت بــەردی ئەمریــی بەرێــژەی-ا ئەمریــكا لــەرووی قەبــارەی بەرهــەم هێنانــەوە پێــش روســیا كــەوت ،ئەمــەش كاریگــەری لەســەردابەزییغ رێــژەی داواكاری نــەوت هەبــوو لەجیهانــدا. https://www.huj.uoh.edu.iq عرᢀێ اقیــش لــە هەمــان مــاوەدا بەرهەمهێنــاᣍغ نــەوᣍت لــە (٩2015 بەرمیــل لــە ســاᣡڵ زیــاد بــووᣍغ بەرهەمهێنــاᣍغ نــەوت لــە الیــەن دەوڵەتــاᣍغ بەرهەمهێنــەری نــەوت لــە دەرەوەی ئۆپیــك .بــۆ نموونــە بەرهەمهێنــاᣍغ نــەوت لــە والیەتــە -ب .٩) ملیــۆن بەرمیــل . بەرازیلیــش لــە هەمــان مــاوەدا بــڕی 4( زیــادی كــرد بــۆ2013 .٧) ملیــۆن بەرمیــل ڕۆژانــەوە لــە ســاᣡڵ4( یەكگرتــوەكاᣍغ ئەمریــكا لــە ) ملیــۆن بەرمیــل .2.4( ) ملیــۆن بەرمیلــەوە ڕۆژانــە زیــاد كــرد بــۆ2( بەرهەمهێنــاᣍغ نــەوᣍت لــە گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 107.46 109.87 105.87 96.29 49.49 2011 2012 2013 2014 2015 نرخی نەوت ( دوالر / برمیل ) گۆڤارێكی زانستی ئهكادیمیه زانكۆی ههڵهبجه دهری دهكات uj p-ISSN: 2412-9607, e-ISSN: 2617-3360 )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 107.46 109.87 105.87 96.29 49.49 2011 2012 2013 2014 2015 نرخی نەوت ( دوالر / برمیل ) Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات 322 HUJ-Volume 5, Issue 4, December 2020 HUJ-Volume 5, Issue 4, December 2020 دە وڵه تــاᣍغ عــه رە ᣍب به رهه مهێنــه ری نــه وت بڕیاریــان دا کــه نــه وت وە ک چه کێــی ســیاس کاریگــه ر به کاربــهێــᣌغ غ1٧٩3 ئۆکتۆبــه ری ســاᣡڵ ) هــه وᣡڵ زۆری دا کــه نــرᣐغ نــه وت بــه جێگــرێ ی بمێنێتــه وە ئه ویــش بــه په یرە وکــردᣍغ سیاســه ᣍت 1٩85_1٩82( _ئۆپیــک لــه مــاوە ی نێــوان ســاᢼڵᣍغ3 که مکردنــه وە ی پشــی به رهه مهێنــاᣍغ بــه ᢼڵ م هه نــدێ لــه ئه ندامــاᣍغ رێکخراوە کــه پابه نــدی ئــه و پشــکه نه بــوون کــه بۆیــان دیــاری کــرا بــوو هــه ر بۆیــه ئــه م 0 65 86 0 2016 2017 2018 2019 نرخی نەوت ( دۆالر / بەرمیل ) 32 HUJ-Volume 5, Issue 4, December 2020 32 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq 201٩-201شێوەی ژمارە ٧هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ ٦ 201٩-201شێوەی ژمارە ٧هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ ٦ )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 0 65 86 0 2016 2017 2018 2019 نرخی نەوت ( دۆالر / بەرمیل ) 201٩-201شێوەی ژمارە ٧هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼᣍغ ٦ )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی 0 65 86 0 2016 2017 2018 2019 نرخی نەوت ( دۆالر / بەرمیل ) نرخی نەوت ( دۆالر / بەرمیل ) )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی لــه الیــه ن پێنــج دە وڵــه ᣍت به رهه مهێنــه ری نــه وت کــه 1٦٩0 ی ســیپته مبه ری ســاᣡڵ14 _ رێکخــراوی دە وڵه تــاᣍغ هــاوردە ی نــه وت (ئۆپیــک) لــه1 و ئه ندە نوســیا و1٦٩1 بریــیت بــوون لــه عــرᢀێ اق و فه نزە وێــال و ئــرᢀێ ان و کوە یــت و ســعودیه لــه به غــدا دامــه زرا، دواتریــش هه ریه کــه لــه قه تــه ر لــه ســاᣡڵ دا بوونــه ئه نــدایم 1٧٩3 و ئیکــوادۆر و گابــۆن لــه ســاᣡڵ1٧٩1 و نه یجرێ یــا لــه ســاᣡڵ1 و ئیمــارات و جه زائرێ یــش لــه ســاᣡڵ ٧٦٩1٦٩3لیبیــاش لــه ســاᣡڵ لــه ئه ندامێــیت رێکخراوە کــه هاتنــه دە رە وە و لــه ئێســتا دا رێکخراوە کــه لــه 1 و ٦٩٩1٩٩2 رێکخراوە کــه بــه ᢼڵ م ئــه م دوو دە وڵه تــه ی کۆتــاᣍی لــه ســاᢼڵᣍغ یــازدە دە وڵــه ت پێکدێــت . HUJ-Volume 5, Issue 4, December 2020 جەنــیگ بــازرگاᣍغ نێــوان ئەمریــكا وچیــᣌێ غ كەدەكرێــت ببێتــە هــۆی نائــارایم ئابــۆری وپاشــانش سســت بــووᣍغ گەشــەی ئابــۆری بەتایبـەت ئەگــەر ئیمــە -ب ئــەو راســتەیە بزانــᣌێ غ وᢼڵᣍت چیــᣌێ غ دووەم وᢼڵᣍت جیهانــە لــەرووی بەكارهێنــان وداواكاری نــەوت. زیادبــووᣍغ قەبــارەی نــەوᣍت خســتنە بــازار زیاتــر لــە بــڕی نــەوᣍت خۆاســرت او،بەهۆی كزبــووᣍغ گەشــەی ئابــۆری لــە ئەگــەری جەنــیگ ئەمریــكا وچیــᣌێ غ - ج وهەروەهــا فشــاری وᢼڵ تــاᣍغ زلهــرᢀێ غ لەســەر وᢼڵ تــاᣍغ بەرهــەم هێنــەر بــۆ كــەم كردنــەوەی بــری نــەوت . و بر و م ر م ێ ر بۆ ᣍ ب ر ر و ᣍ زهرێ ر و رو و فشــاری ئەمریــكا لەســەر وᢼڵ تــاᣍغ ئوپیــك بەتایبــەت ســعودیا لەرێــگای بــەكار هێنــاᣍغ دۆســیەكاᣍغ رۆژنامەنــووس جمــال خاشــقیب كێشــەی یەمــن بــۆ - د كەمكردنــەوەی نــرᣐغ نــەوت وزیــاد كــردᣍغ بــڕی بەرهــەم هێنــان. Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 3 HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq 201٩-201شێوەی ژمارە ٧هەڵكشان و داكشاᣍغ نرᣐغ نەوᣍت ئۆپیك لە ماوەی نێوان ساᢼڵᣍغ ٦ )4( سەرچاوە : ئامادەكراوە بە پشت بەسᣌت غ بە خشتەی دە رئه نجام- ٤ لــه الیــه ن پێنــج دە وڵــه ᣍت به رهه مهێنــه ری نــه وت کــه 1٦٩0 ی ســیپته مبه ری ســاᣡڵ14 _ رێکخــراوی دە وڵه تــاᣍغ هــاوردە ی نــه وت (ئۆپیــک) لــه1 و ئه ندە نوســیا و1٦٩1 بریــیت بــوون لــه عــرᢀێ اق و فه نزە وێــال و ئــرᢀێ ان و کوە یــت و ســعودیه لــه به غــدا دامــه زرا، دواتریــش هه ریه کــه لــه قه تــه ر لــه ســاᣡڵ دا بوونــه ئه نــدایم 1٧٩3 و ئیکــوادۆر و گابــۆن لــه ســاᣡڵ1٧٩1 و نه یجرێ یــا لــه ســاᣡڵ1 و ئیمــارات و جه زائرێ یــش لــه ســاᣡڵ ٧٦٩1٦٩3لیبیــاش لــه ســاᣡڵ لــه ئه ندامێــیت رێکخراوە کــه هاتنــه دە رە وە و لــه ئێســتا دا رێکخراوە کــه لــه 1 و ٦٩٩1٩٩2 رێکخراوە کــه بــه ᢼڵ م ئــه م دوو دە وڵه تــه ی کۆتــاᣍی لــه ســاᢼڵᣍغ یــازدە دە وڵــه ت پێکدێــت . خــاᣡڵ گــۆڕان بــووە لــه مــرᢀێ ژ ووی دە وڵه تــاᣍغ ئۆپیــک چونکــه توانیــان رۆڵیــان لــه دیاریکــردᣍغ نــرᣐغ نــه وت دا هه بێــت چونکــه لــه جه نــیگ 1٧٩3 _ ســاᣡڵ2 . HUJ-Volume 5, Issue 4, December 2020 خــاᣡڵ گــۆڕان بــووە لــه مــرᢀێ ژ ووی دە وڵه تــاᣍغ ئۆپیــک چونکــه توانیــان رۆڵیــان لــه دیاریکــردᣍغ نــرᣐغ نــه وت دا هه بێــت چونکــه لــه جه نــیگ 1٧٩3 _ ســاᣡڵ2 . دە وڵه تــاᣍغ عــه رە ᣍب به رهه مهێنــه ری نــه وت بڕیاریــان دا کــه نــه وت وە ک چه کێــی ســیاس کاریگــه ر به کاربــهێــᣌغ غ1٧٩3 ئۆکتۆبــه ری ســاᣡڵ ) هــه وᣡڵ زۆری دا کــه نــرᣐغ نــه وت بــه جێگــرێ ی بمێنێتــه وە ئه ویــش بــه په یرە وکــردᣍغ سیاســه ᣍت 1٩85_1٩82( _ئۆپیــک لــه مــاوە ی نێــوان ســاᢼڵᣍغ3 که مکردنــه وە ی پشــی به رهه مهێنــاᣍغ بــه ᢼڵ م هه نــدێ لــه ئه ندامــاᣍغ رێکخراوە کــه پابه نــدی ئــه و پشــکه نه بــوون کــه بۆیــان دیــاری کــرا بــوو هــه ر بۆیــه ئــه م هه وᢼڵ نــه ســه رکه وتوو نه بــوون. https://www.huj.uoh.edu.iq )2015( داکشــان کــرد هــه روە ک لــه ســاᣡڵ ژ هه نــاردە ی نــه وت و به رزبوونــه وە ی به هــای دۆالر جارێــی تــر نــرᣐغ نــه وᣍت لــه بــازاڕە کاᣍغ جیهــان دا تــووس ) دۆالر بــووە . 50( دا نــرᣐغ به رمیلێــک نــه وت که مــرت لــه نــەوت بەهەمــان شــێوەی قۆناغــەكاᣍغ تــر هەڵبــەز ودابەزیــی گــەوەرەی بەخۆیــەوە بیــیغ بەجــۆری پــاش ئــەوەی 201بــۆ٩201لەســاالᣍغ نیــوان ٦-٧ دوالر بــەᢼڵ م بەدوبــارە بوونــەوەی هــۆكارەكاᣍغ پیشــوو كێشــەی ئەتــوویم ئــرێ ان 84 نــرᣐغ نــەوت گیشــتە نزیكــەی201 وســەرەتاكاᣍغ ٧201لــە كۆتــاᣍی ٦ وكەشــانەوەی ئەمریــكا لــە رێكەوتننامەكــە وزیادبــووᣍغ بــڕی بەرهــەیم خســتنەرۆی نــەوت زیاتــر لــە قەبــارەی داواكراو،پەرەســەندᣍغ شــەڕی بــازرگاᣍغ دۆالر بــۆ هــەر 5،٩4نێــوان ئەمریــكا وچــᣌێ غ وسســت بــووᣍغ گەشــەی ئابــوری جبهــاᣍغ هەموویــان رۆڵیــان هەبــوو لــە دابەزیــیغ نــرᣐغ نــەوت بــۆ ژێــر ٦ برمیلیــك دابــەزی. راسپاردەکان پێویســتە ریکخــراوی ئوبیــک کاربــکات لەســەر بەهرᢀێ غکــردᣍغ پێگــەی ئیســتای خــؤی لــە چوارچێــوەی بــازاری نــەوت لەرێــگای راکێشــاᣍغ وᢼڵ تــاᣍغ تــری -1 .نــەوᣍت دەرەوەی بــۆ هاتنــە نــاو ئــەم ریکخــراوە ئەنجامــداᣍغ گڤتوگــۆ وراویــژ لەگــەڵ وᢼڵ تــە گــەورەکاᣍغ بەرهــەم هێنــەری نــەوت ئەوانــەی لــەدەرەوەی ئــەم ریکخــراوەن بەئامانــیب پارێــزگاری کــردن -2 .لــە نــرᣐغ نــەوت پێویستە ڕᣍᢀی وشوێیغ قورسرت بگرᢀێ ێتە بەر بەرامبەر بەهەموو وᢼڵ تاᣍغ ئەندایم ئوبیک لەرووی پابەندکردنیان بە رێژەی بەرهەمهێنان-3 کارکــردᣍغ بەخــرᢀێ اᣍی لەالیــەن ریکخــراوی ئوبیــک لــە بــۆاری وەبەرهێنــاᣍغ نــەوت ،چونکــە ئەمــە وادەکات پێگــەی ئــەم رێکخــراوە لــە بــازاری نــەوت لــە-4 .داهاتــوودا بــە پارێزێــت )2015( داکشــان کــرد هــه روە ک لــه ســاᣡڵ ژ هه نــاردە ی نــه وت و به رزبوونــه وە ی به هــای دۆالر جارێــی تــر نــرᣐغ نــه وᣍت لــه بــازاڕە کاᣍغ جیهــان دا تــووس ) دۆالر بــووە . 50( دا نــرᣐغ به رمیلێــک نــه وت که مــرت لــه نــەوت بەهەمــان شــێوەی قۆناغــەكاᣍغ تــر هەڵبــەز ودابەزیــی گــەوەرەی بەخۆیــەوە بیــیغ بەجــۆری پــاش ئــەوەی 201بــۆ٩201لەســاالᣍغ نیــوان ٦-٧ دوالر بــەᢼڵ م بەدوبــارە بوونــەوەی هــۆكارەكاᣍغ پیشــوو كێشــەی ئەتــوویم ئــرێ ان 84 نــرᣐغ نــەوت گیشــتە نزیكــەی201 وســەرەتاكاᣍغ ٧201لــە كۆتــاᣍی ٦ وكەشــانەوەی ئەمریــكا لــە رێكەوتننامەكــە وزیادبــووᣍغ بــڕی بەرهــەیم خســتنەرۆی نــەوت زیاتــر لــە قەبــارەی داواكراو،پەرەســەندᣍغ شــەڕی بــازرگاᣍغ دۆالر بــۆ هــەر 5،٩4نێــوان ئەمریــكا وچــᣌێ غ وسســت بــووᣍغ گەشــەی ئابــوری جبهــاᣍغ هەموویــان رۆڵیــان هەبــوو لــە دابەزیــیغ نــرᣐغ نــەوت بــۆ ژێــر ٦ برمیلیــك دابــەزی. په رتوک : ی عالم البرت ول قضايا برت ولية دولية , بدون مكان طبع. العقيل, خالد بن منصور، بدون سنة، رحلة ᣚغ-1 .، الرياض2018 ،اسواق العالمية للنفط الربــع الرابع2018،الرت یکی ،فهد--2 ،غ ا،دار المنهل،برێ وت ی العالم العرᣍب ی ماضياوحارص ،التطوارات النفطية ᣚغ200الترغ ،سمرێ ،٧--3 .،اقتصاديات النفط، الجامعة المفتوحة، طرابلس1 سالم عبد الحسن رسن، ٩٩٩-4 .،تحدید اسعار النفط العرᣍب الخام ᣚغ السوق العالمیە،دار الرشید للنᣄژ ،برێ وت،لبنان1٩81،نواف نایف اسماعیل-5 . ،دار زهران للنᣄژ والتوزيــــع ، جدە السعودية5 . ط3 ،أقتصاد النفط ،الجزء1 الخویلی ، سيد فتیی أحمد ،٧٩٩-٦ ،مكتبة العرᣍب ی الحديث،قاهرة.3،تسويق البرت ول،ط2003 ،عفيیغ ی ،محمد صديق-٧ ،برێ وت.1،النفط بᣌێ غ السياسةواالقتصاد،ط1٧٩4 ،عبدالوهاب،عبدالمنعم-8 .ی الشوون النفط،دارالطليعة،برێ وت ،مذكرات واراء ᣚغ1٩80،حمادي،سعدون-٩ ،رياض.1ی منازعات عقود النفط،ط ،اشكاليات التحكم ᣚغ1٩81،اسماعيل،خالد منصور-10 ،رياض.1ی منازعات عقود النفط،ط ،اشكاليات التحكم ᣚغ1٩81،اسماعيل،خالد منصور-10 HUJ-Volume 5, Issue 4, December 2020 ) بــه هــۆی ڕووداوە کاᣍغ ڕۆژهــه ᢼڵᣍت ناوە راســت بــه تایبــه ᣍت شــه ڕی عــرᢀێ اق و ئاڵۆزییــه کاᣍغ نه یجرێ یــا لــه الیــه ک 2008_2003( _ لــه مــاوە ی ســاᢼڵᣍغ4 و زیادبــووᣍغ خواســت لــه ســه ر نــه وت لــه ســه ر ئاســیت جیهــاᣍغ و گه شــه ی ئابــوری چــᣌێ غ و هینــد یارمه تیــدە ر بــوو لــه هه ڵکشــاᣍغ نرخه کــه ی لــه مانــیگ )دۆالر بــۆ هــه ر به رمیلێــک کــه ئه مــه ش گه ورە تریــن هه ڵکشــان بــووە لــه مــرᢀێ ژ ووی نــرᣐغ نــه وت .14) دا بــۆ (٧2008( ته مــووزی ) دا و لــه دوای ســه رهه ڵداᣍغ به هــاری عــه رە ᣍب و بــه هــۆی ئــه و گۆڕانکارییــه سیاســییه ی وᢼڵ تــاᣍغ به رهه مهێنــه ری نــه وᣍت گرتــه وە 2011( _لــه ســاᣡڵ5 .ڕێــژە ی هه ناردە کــردᣍغ نــه وت ڕووی لــه دابه زیــن کــردوە ئه مــه ش بووە تــه هــۆی هه ڵکشــاᣍغ گــه ورە لــه نــرᣐغ نــه وت دا ) بــه هــۆی زۆری ڕێــژەی هه ناردە کــردᣍغ نــه وت لــه الیــه ن ئۆپیــک و دە وڵه تــاᣍغ تــری 2015( ) و هه روە هــا ســاᣡڵ2014( ٦_ لــه ناوە راســیت ســاᣡڵ هه نــاردە ی نــه وت و به رهه مهێنــاᣍغ نــه وᣍت به ردیــن لــه الیــه ن والیه ته یه کگرتــووە کاᣍغ ئه مریــکاوە هه روە هــا ئاڵۆزییــه کاᣍغ ڕۆژهــه ᢼڵᣍت ناوە راســت و شــه ڕی داعــش و فرۆشــتیغ نــه وᣍت ئــرᢀێ ان و ڕێکنه که وتــیغ دە وڵه تــاᣍغ ئۆپیــک و دە وڵه تــاᣍغ تــری به رهه مهێنــه ری نــه وت لــه ســه ر که مکردنــه وە ی ڕێــژە ی ) بــه هــۆی ڕووداوە کاᣍغ ڕۆژهــه ᢼڵᣍت ناوە راســت بــه تایبــه ᣍت شــه ڕی عــرᢀێ اق و ئاڵۆزییــه کاᣍغ نه یجرێ یــا لــه الیــه ک 2008_2003( _ لــه مــاوە ی ســاᢼڵᣍغ4 و زیادبــووᣍغ خواســت لــه ســه ر نــه وت لــه ســه ر ئاســیت جیهــاᣍغ و گه شــه ی ئابــوری چــᣌێ غ و هینــد یارمه تیــدە ر بــوو لــه هه ڵکشــاᣍغ نرخه کــه ی لــه مانــیگ )دۆالر بــۆ هــه ر به رمیلێــک کــه ئه مــه ش گه ورە تریــن هه ڵکشــان بــووە لــه مــرᢀێ ژ ووی نــرᣐغ نــه وت .14) دا بــۆ (٧2008( ته مــووزی ) دا و لــه دوای ســه رهه ڵداᣍغ به هــاری عــه رە ᣍب و بــه هــۆی ئــه و گۆڕانکارییــه سیاســییه ی وᢼڵ تــاᣍغ به رهه مهێنــه ری نــه وᣍت گرتــه وە 2011( _لــه ســاᣡڵ5 ) بــه هــۆی ڕووداوە کاᣍغ ڕۆژهــه ᢼڵᣍت ناوە راســت بــه تایبــه ᣍت شــه ڕی عــرᢀێ اق و ئاڵۆزییــه کاᣍغ نه یجرێ یــا لــه الیــه ک 2008_2003( _ لــه مــاوە ی ســاᢼڵᣍغ4 و زیادبــووᣍغ خواســت لــه ســه ر نــه وت لــه ســه ر ئاســیت جیهــاᣍغ و گه شــه ی ئابــوری چــᣌێ غ و هینــد یارمه تیــدە ر بــوو لــه هه ڵکشــاᣍغ نرخه کــه ی لــه مانــیگ )دۆالر بــۆ هــه ر به رمیلێــک کــه ئه مــه ش گه ورە تریــن هه ڵکشــان بــووە لــه مــرᢀێ ژ ووی نــرᣐغ نــه وت .14) دا بــۆ (٧2008( ته مــووزی .ڕێــژە ی هه ناردە کــردᣍغ نــه وت ڕووی لــه دابه زیــن کــردوە ئه مــه ش بووە تــه هــۆی هه ڵکشــاᣍغ گــه ورە لــه نــرᣐغ نــه وت دا ) بــه هــۆی زۆری ڕێــژەی هه ناردە کــردᣍغ نــه وت لــه الیــه ن ئۆپیــک و دە وڵه تــاᣍغ تــری 2015( ) و هه روە هــا ســاᣡڵ2014( ٦_ لــه ناوە راســیت ســاᣡڵ هه نــاردە ی نــه وت و به رهه مهێنــاᣍغ نــه وᣍت به ردیــن لــه الیــه ن والیه ته یه کگرتــووە کاᣍغ ئه مریــکاوە هه روە هــا ئاڵۆزییــه کاᣍغ ڕۆژهــه ᢼڵᣍت ناوە راســت و شــه ڕی داعــش و فرۆشــتیغ نــه وᣍت ئــرᢀێ ان و ڕێکنه که وتــیغ دە وڵه تــاᣍغ ئۆپیــک و دە وڵه تــاᣍغ تــری به رهه مهێنــه ری نــه وت لــه ســه ر که مکردنــه وە ی ڕێــژە ی Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 گۆڤارێكی‌زانستی‌ئه‌كادیمیه‌‌زانكۆی‌هه‌ڵه‌بجه‌‌ده‌ری‌ده‌كات HUJ-Volume 5, Issue 4, December 2020 https://www.huj.uoh.edu.iq راسپاردەکان پێویســتە ریکخــراوی ئوبیــک کاربــکات لەســەر بەهرᢀێ غکــردᣍغ پێگــەی ئیســتای خــؤی لــە چوارچێــوەی بــازاری نــەوت لەرێــگای راکێشــاᣍغ وᢼڵ تــاᣍغ تــری -1 .نــەوᣍت دەرەوەی بــۆ هاتنــە نــاو ئــەم ریکخــراوە ئەنجامــداᣍغ گڤتوگــۆ وراویــژ لەگــەڵ وᢼڵ تــە گــەورەکاᣍغ بەرهــەم هێنــەری نــەوت ئەوانــەی لــەدەرەوەی ئــەم ریکخــراوەن بەئامانــیب پارێــزگاری کــردن -2 .لــە نــرᣐغ نــەوت پێویستە ڕᣍᢀی وشوێیغ قورسرت بگرᢀێ ێتە بەر بەرامبەر بەهەموو وᢼڵ تاᣍغ ئەندایم ئوبیک لەرووی پابەندکردنیان بە رێژەی بەرهەمهێنان-3 کارکــردᣍغ بەخــرᢀێ اᣍی لەالیــەن ریکخــراوی ئوبیــک لــە بــۆاری وەبەرهێنــاᣍغ نــەوت ،چونکــە ئەمــە وادەکات پێگــەی ئــەم رێکخــراوە لــە بــازاری نــەوت لــە-4 .داهاتــوودا بــە پارێزێــت ) دۆالر بــووە . 50( دا نــرᣐغ به رمیلێــک نــه وت که مــرت لــه نــەوت بەهەمــان شــێوەی قۆناغــەكاᣍغ تــر هەڵبــەز ودابەزیــی گــەوەرەی بەخۆیــەوە بیــیغ بەجــۆری پــاش ئــەوەی 201بــۆ٩201لەســاالᣍغ نیــوان ٦-٧ دوالر بــەᢼڵ م بەدوبــارە بوونــەوەی هــۆكارەكاᣍغ پیشــوو كێشــەی ئەتــوویم ئــرێ ان 84 نــرᣐغ نــەوت گیشــتە نزیكــەی201 وســەرەتاكاᣍغ ٧201لــە كۆتــاᣍی ٦ وكەشــانەوەی ئەمریــكا لــە رێكەوتننامەكــە وزیادبــووᣍغ بــڕی بەرهــەیم خســتنەرۆی نــەوت زیاتــر لــە قەبــارەی داواكراو،پەرەســەندᣍغ شــەڕی بــازرگاᣍغ دۆالر بــۆ هــەر 5،٩4نێــوان ئەمریــكا وچــᣌێ غ وسســت بــووᣍغ گەشــەی ئابــوری جبهــاᣍغ هەموویــان رۆڵیــان هەبــوو لــە دابەزیــیغ نــرᣐغ نــەوت بــۆ ژێــر ٦ برمیلیــك دابــەزی. ـژ لەگــەڵ وᢼڵ تــە گــەورەکاᣍغ بەرهــەم هێنــەری نــەوت ئەوانــەی لــەدەرەوەی ئــەم ریکخــراوەن بەئامانــیب پارێــزگاری کــرد پێویستە ڕᣍᢀی وشوێیغ قورسرت بگرᢀێ ێتە بەر بەرامبەر بەهەموو وᢼڵ تاᣍغ ئەندایم ئوبیک لەرووی پابەندکردنیان بە رێژەی بەرهەمهێنان-3 کارکــردᣍغ بەخــرᢀێ اᣍی لەالیــەن ریکخــراوی ئوبیــک لــە بــۆاری وەبەرهێنــاᣍغ نــەوت ،چونکــە ئەمــە وادەکات پێگــەی ئــەم رێکخــراوە لــە بــازاری نــەوت لــە-4 .داهاتــوودا بــە پارێزێــت 325 HUJ-Volume 5, Issue 4, December 2020 :گۆڤار https://www.huj.uoh.edu.iq ی دائرة القوة,)عدد خاص,ازار ونيسان,. ،(منظمات ᣚغ2010 ,مجلة البرت ول-1 .، دمشق2 (انعكاسات الزيادات االسعار النفط عیل االقتصاد الوطیغ ی )،مجلة تᣄژ ين،العدد200 محمود ،حبيب ، ٧-2 ،مركزالدراسات13 ی ظل تطوارات سوق النفط العالمية)،مجلة االقتصاد الخليیب ی ،العدد (،نفط الخليج ᣚغ200حسن ،يحیێ حمود،واخرون،٧-3 .،الخليج العرᣍب ی ،البرة ی اســتقرار ســوق النفــط الدولية)،مجلــة العلــوم ،(دور سياســات النفــط الســعودية ᣚغ200عبــد االسدي،يوســف عیلی ،واخــرون٩-4 .،،البــرة24االقتصادية،العدد : تᣂᢀێ ن ە کان )،رســاله ماجســترێ200 ٩- 1٩8ی الدواللعربیةخــالل الفــرت ة ٦ ،(تقلبــات أســعارالنفط وانعكاســاتھاعیل تمویــل التنمیــةᣚغ200نعیمــة، حمــادي، ٩-1 .،جامعةحسیبةبنبوعیلی ،جامعةالشــلف )،رســاله ماجســترێ200 ٩- 1٩8ی الدواللعربیةخــالل الفــرت ة ٦ ،(تقلبــات أســعارالنفط وانعكاســاتھاعیل تمویــل التنمیــةᣚغ200نعیمــة، حمــادي، ٩-1 .،جامعةحسیبةبنبوعیلی ،جامعةالشــلف )،رساله ماجسترێ ،کلیه العلوم االقتصادیه 2011-2001( ،(تاثرێ الᣄژ کات النفطیه العالمیه عیل اسعار النفط2013،دبیه ،عیل-2 .، و رقله-و التجاریه و علوم التسیرێ ،جامعه قاصدی مرباح )،رساله ماجسترێ ،کلیه العلوم االقتصادیه 2011-2001( ،(تاثرێ الᣄژ کات النفطیه العالمیه عیل اسعار النفط2013،دبیه ،عیل-2 .، و رقله-و التجاریه و علوم التسیرێ ،جامعه قاصدی مرباح Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University An Academic And Scientntific Journal Issued By Halabja University Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360 : رۆژنامه Huj p-ISSN: 2412-9607, e-ISSN: 2617-3360
https://openalex.org/W2971200364
https://bmcinfectdis.biomedcentral.com/track/pdf/10.1186/s12879-019-4373-9
English
null
Assessing rates and contextual predictors of 5-year mortality among HIV-infected and HIV-uninfected individuals following HIV testing in Durban, South Africa
BMC infectious diseases
2,019
cc-by
8,603
© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Bassett et al. BMC Infectious Diseases (2019) 19:751 https://doi.org/10.1186/s12879-019-4373-9 Bassett et al. BMC Infectious Diseases (2019) 19:751 https://doi.org/10.1186/s12879-019-4373-9 Open Access Abstract Background: Little is known about contextual factors that predict long-term mortality following HIV testing in resource-limited settings. We evaluated the impact of contextual factors on 5-year mortality among HIV-infected and HIV-uninfected individuals in Durban, South Africa. Methods: We used data from the Sizanani trial (NCT01188941) in which adults (≥18y) were enrolled prior to HIV testing at 4 outpatient sites. We ascertained vital status via the South African National Population Register. We used random survival forests to identify the most influential predictors of time to death and incorporated these into a Cox model that included age, gender, HIV status, CD4 count, healthcare usage, health facility type, mental health, and self-identified barriers to care (i.e., service delivery, financial, logistical, structural and perceived health). Results: Among 4816 participants, 39% were HIV-infected. Median age was 31y and 49% were female. 380 of 2508 with survival information (15%) died during median follow-up of 5.8y. For both HIV-infected and HIV-uninfected participants, each additional barrier domain increased the HR of dying by 11% (HR 1.11, 95% CI 1.05–1.18). Every 10- point increase in mental health score decreased the HR by 7% (HR 0.93, 95% CI 0.89–0.97). The hazard ratio (HR) for death of HIV-infected versus HIV-uninfected varied by age: HR of 6.59 (95% CI: 4.79–9.06) at age 20 dropping to a HR of 1.13 (95% CI: 0.86–1.48) at age 60. Conclusions: Independent of serostatus, more self-identified barrier domains and poorer mental health increased mortality risk. Additionally, the impact of HIV on mortality was most pronounced in younger persons. These factors may be used to identify high-risk individuals requiring intensive follow up, regardless of serostatus. gistration: Clinical Trials.gov Identifier NCT01188941. Registered 26 August 2010. Trial registration: Clinical Trials.gov Identifier NCT01188941. Registered 26 August 2010. Keywords: HIV infection, Mortality, Predictors of mortality, Barriers to care Keywords: HIV infection, Mortality, Predictors of mortality, Barriers to care ART in sub-Saharan Africa can achieve life expectancies similar to those who are HIV-uninfected [2, 3]. However, mortality remains high both before [4, 5] and after ART initiation because of inconsistent care [6–10]. Despite 86% of PLWH in South Africa knowing their HIV status, only 56% were on ART, and only 45% were virally suppressed in 2016 [1]. Thus, accurately ascertaining factors contrib- uting to long-term mortality risk following HIV-diagnosis is paramount. Assessing rates and contextual predictors of 5-year mortality among HIV-infected and HIV-uninfected individuals following HIV testing in Durban, South Africa Ingrid V. Bassett1,2,3,4* , Ai Xu5, Janet Giddy6, Laura M. Bogart7, Andrew Boulle8,9,10, Lucia Millham2, Elena Losina2,3,11,12 and Robert A. Parker2,3,4,5 Ingrid V. Bassett1,2,3,4* , Ai Xu5, Janet Giddy6, Laura M. Bogart7, Andrew Boulle8,9,10, Lucia Millham2, Elena Losina2,3,11,12 and Robert A. Parker2,3,4,5 Background South Africa has the largest number of HIV-infected indi- viduals of any country, with over 7 million people diag- nosed with HIV and 270,000 new infections in 2016 [1]. People living with HIV (PLWH) who consistently take * Correspondence: ibassett@mgh.harvard.edu 1Division of Infectious Diseases, Massachusetts General Hospital, 100 Cambridge Street, 16th Floor, Boston, MA 02114, USA 2Medical Practice Evaluation Center, Massachusetts General Hospital, Boston, MA, USA Full list of author information is available at the end of the article Studies assessing the long-term risk of mortality in persons living with HIV in sub-Saharan Africa have , ll list of author information is available at the end of the article Health behavior k d We asked participants whether they had tested for HIV prior to enrollment. We assessed self-reported compet- ing needs at enrollment, by asking if, in the past 6 months, they had ever gone without healthcare because they needed money for basic needs, or if they had gone without basic needs because they needed money for healthcare [39, 40]. Page 2 of 11 Page 2 of 11 Bassett et al. BMC Infectious Diseases (2019) 19:751 Bassett et al. BMC Infectious Diseases (2019) 19:751 focused on age, gender, and CD4 counts to evaluate pre- dictors of risk [11–15]. We and others have found, how- ever, that contextual factors (e.g. barriers to care), emotional health, social support, and competing needs at the time of diagnosis, also likely to have an important effect on survival [16, 17]. In addition, most mortality studies, including those that have examined social and contextual factors, have limited analyses to HIV-infected individuals [7, 18–23]. Little is known about contextual factors that predict long-term mortality in resource-lim- ited settings, for both HIV-infected and HIV-uninfected individuals. By comparing to a concurrently enrolled HIV-uninfected group of individuals, who share similar socioeconomic status, we can further our understanding of what interventions in the outpatient setting may im- prove outcomes in South Africa regardless of HIV status. Healthcare utilization We assessed self-reported healthcare utilization in the year prior to enrollment, including visits to a commu- nity health worker, local clinic, hospital, or private doctor. The total utilization was grouped into three categories: > 5 times, 3–5 times, 1–2 times, did not use healthcare in prior year. We also asked about number of visits to a traditional healer. Demographics and CD4 count data We asked participants to provide demographic informa- tion including: age, gender, relationship status, and hours worked outside the home. We collected baseline CD4 count data from medical records for those who were HIV- infected; missing data is discussed in the Additional file 1. Our objective was to assess contextual predictors of 5- year mortality following HIV testing. Research suggests a strong correlation between mental health, especially de- pression, and mortality among people with HIV [24, 25] and other health conditions [26–28]. Research among PLWH indicates that such associations may be due to a relationship between depression and immune suppression, leading to accelerated disease progression [25, 29, 30]. Moreover, in the US, depression interventions have de- creased mortality risk, including for those with chronic diseases, suggesting a causal effect [31–34]. Research also shows strong associations between social support and mortality in general [35, 36]. Thus, we hypothesized that poor emotional health and social support at HIV diagnosis would be associated with higher 5-year mor- tality rates and could therefore serve as targets for future interventions. Healthcare access We assessed healthcare access using four questions that determined how difficult it might be for a patient to reach the site. We collected data on mode of transporta- tion (public, private, or other) and distance to clinic. Transportation variables were grouped into two categor- ies – public transport (bus, taxi) or other (including pri- vate and other). Study setting/design y g g This analysis includes data from the Sizanani Trial (NCT01188941), a randomized controlled trial that exam- ined the efficacy of health system navigation and short messaging service (SMS) reminders on linkage to and re- tention in HIV/TB care. We enrolled adults prior to HIV testing at 4 outpatient sites, 2 hospital outpatient depart- ments (one urban and one semi-rural) and 2 primary health clinics (semi-rural) in Durban, South Africa from August 2010–January 2013. This trial is described in detail elsewhere [16, 37, 38]. Because we did not find efficacy of the intervention with respect to linkage to HIV care, TB treatment completion, or 1-year mortality between study arms, we pooled data from the intervention and control groups into a single cohort in the current study and compared them to HIV-uninfected individuals enrolled concurrently. Participants Adults ≥18 years with unknown HIV status presenting for HIV testing were eligible for enrollment. Study enroll- ment, consisting of informed consent and a baseline ques- tionnaire, occurred prior to HIV testing. This allowed for assessment of contextual factors, emotional health, and social support prior to knowledge of HIV status. The study was approved by the McCord Hospital Medical Research Ethics Committee, St. Mary’s Hospital Research Ethics Committee, University of KwaZulu- Natal Biomedical Research Ethics Committee, and Partners Institutional Review Board (Protocol 2011-P-001195, Boston, MA). Self-perceived barriers to care We assessed self-perceived barriers to healthcare in the 6 months prior to enrollment using a 12-question instru- ment modified from the ARTAS-II trial [41]. We grouped barriers into 5 domains: 1) concerns about service delivery (waiting time to see a provider, treatment by clinic staff), Page 3 of 11 Bassett et al. BMC Infectious Diseases (2019) 19:751 Page 3 of 11 importance based on permutation importance and vari- able depth relative to root node (see Additional file 1). 2) financial concerns (ability to afford medication or trans- portation), 3) perception of personal health (not being sick enough or being too sick), 4) logistical concerns (unable to get out of work, responsibilities to care for others), 5) structural (impaired clinic access due to clinic hours or transportation difficulties, lack of knowledge about where to find care). We created a total number of barriers vari- able by adding up all barriers in all 5 categories for each participant. We created a similar total number of domains variable by adding up the total number of domains under which a participant indicated they had a barrier. We used a sequential procedure to select the most important variables for the survival model. We first identified the most important of the six domains above to include, and then selected the most important vari- able(s) in that domain to include in subsequent model construction. Variable selection was based on combining the results from two different statistical approaches. Demographic characteristics were considered the most important category to include, and age was identified as the most important covariate in this category. Gender was selected a priori to be included [49]. Random sur- vival forests were then fitted separately to each of the other five categories of covariates, with age, gender, and HIV status included in all models. The second most im- portant category and important covariates within that cat- egory were selected as described above, and the procedure repeated until at least one variable was selected from each category if the category was important. We also evaluated the most important barriers and domains within the self-perceived barriers to care category as detailed in the Additional file 1. Because of the influence of CD4 counts on mortality, we included this in the final model as well. Emotional health and social support We adapted the 5-item Mental Health Inventory screen- ing test and calculated a mental health composite (MHC) score [42]. In addition, we condensed 13 questions about social support into 4 social support scales (emotional/in- formational, tangible, positive interaction, and affection- ate) and calculated the Social Support Index (SSI) from the Medical Outcomes Study [43]. Separately, we averaged each scale and converted to a scale from 0 to 100. A higher number on the scale indicates better emotional health or social support. An MHC ≤52 qualified as a posi- tive depression screen; an SSI below the sample median (75) qualified as a lack of social support [44]. Self-perceived barriers to care For the 92 HIV-infected participants (8%) miss- ing CD4, we used multiple imputation based on gender, age, health facility type, healthcare use in the past year, total number of domains, and mental health score CD4 counts. HIV-uninfected participants were assigned a CD4 count of 775 based on the median CD4 count of the general population in rural KwaZulu-Natal [50]. We also assessed how the association of health care utilization with mortality varied over the five-year period by fitting the model for the first third, second third, and last third of deaths separately. Outcome ascertainment We elicited mortality from the National Population Regis- ter, which is estimated to incorporate at least 90% of deaths nationwide [13, 45]. We used South African ID numbers (SAIDs) obtained at enrollment to match partici- pants to the National Population Register in November 2017; median follow-up was 5.8 years (IQR 5.2–6.5 years). Statistical analysis We fitted a propensity score-weighted Cox propor- tional hazards model to the final set of selected covari- ates. The likelihood ratio test was used to test for interaction effects between HIV status and other covari- ates included in the Cox model. We used the integrated area under the curve (AUC) as the measure of accuracy for the Cox models [51]. We provide a summary of our methods here; further de- tails are in the Additional file 1. Because a substantial fraction of participants were missing SAIDs, we used propensity score (PS) weighting [46] to make the population with SAIDs be representa- tive of the total group. We estimated the probability of having a SAID from a logistic regression model separ- ately for HIV-infected and HIV-uninfected including all available baseline data, and then used inverse probability weighting to make the population with SAIDs more rep- resentative of the total population. To avoid potential confounding of contextual factors by HIV-status, we then used an additional propensity score adjustment so that the HIV-infected and HIV-uninfected were similar. We used random survival forests [47] on all covariates listed in Table 1 to inform the development of our survival model. When the number of variables is comparatively large, a random forest is useful for variable selection be- cause it avoids overfitting [48]. We determined variable We describe the association of each variable with death using hazard ratios (HR), 95% confidence inter- vals, and P-values in the fully adjusted model (incorp- orating propensity score weighting and CD4 value imputation), and in simpler models without CD4 value imputation, without propensity score adjustment, and without both to assess the robustness of our conclu- sions. We used two-tailed P-values < 0.05 as a cut-off for statistical significance. Statistical analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC) and “randomForestSRC” in R version 3.4.2 (www. r-project.org) [52]. Bassett et al. Overall cohort characteristics There were 4816 enrollees, of whom 1897 (39%) were HIV-infected (Table 1). The median age of the entire cohort was 31 years (IQR: 24–41); 33 years for HIV-in- fected individuals and 28 years for those HIV-unin- fected (P < 0.001). Overall, 2339 (49%) were female. The median CD4 count for those with HIV was 196 (IQR: 73–352). Most participants, 3738 (78%), were never married; 2283 (48%) used public transport to travel to the healthcare site and 3582 (74%) underwent HIV test- ing in an hospital outpatient department versus a pri- mary health clinic. The most commonly identified barrier domain was pa- tient perception of personal health, with 1247 (26%) of participants experiencing a barrier in this category. Across barrier domains, service delivery showed the largest differ- ence between HIV-infected and HIV-uninfected partici- pants: 566 (30%) HIV-infected participants identified a service delivery barrier, compared to 586 (20%) HIV-unin- fected participants (P < 0.001). HIV-infected individuals ex- perienced significantly higher burdens of self-identified barriers across all domains. Statistical analysis BMC Infectious Diseases (2019) 19:751 Page 5 of 11 Table 1 Differences between HIV-infected and HIV-uninfected participants at baseline (Continued) Overall, n = 4816 HIV-infected, n = 1897 HIV-uninfected, n = 2919 P* Number of barriers for participants reporting barriers Median (IQR) 3 (2–6) 4 (2–6) 3 (1–5) < 0.001 Number of barrier domains for participants reporting barriers Median (IQR) 3 (1–4) 3 (2–4) 2 (1–4) < 0.001 Gone without healthcare for basic needs, n (%) Yes 920 (19) 414 (22) 506 (17) < 0.001 No 3896 (81) 1483 (78) 2413 (83) Gone without basic needs for healthcare, n (%) Yes 724 (15) 323 (17) 401 (14) 0.002 No 4092 (85) 1574 (83) 2518 (86) Table 1 Differences between HIV-infected and HIV-uninfected participants at baseline (Continued) uninfected individuals was for waiting too long to see a nurse or doctor: 506 (27%) of HIV-infected participants reported this barrier, while only 510 (18%) of HIV-unin- fected individuals did (P < 0.001). Statistical analysis BMC Infectious Diseases (2019) 19:751 Page 4 of 11 Table 1 Differences between HIV-infected and HIV-uninfected participants at baseline Overall, n = 4816 HIV-infected, n = 1897 HIV-uninfected, n = 2919 P* Age, yrs Median (IQR) 31.0 (24–41) 33.0 (27–41) 28.0 (22–42) < 0.001 Sex, n (%) Male 2477 [51] 964 (51) 1513 [52] 0.491 Female 2339 [49] 933 (49) 1406 [48] Marital status, n (%) Never married 3738 (78) 1535 (81) 2203 (76) < 0.001 Currently married 810 (17) 265 (14) 545 (19) Divorce/separated/widowed 239 (5) 85 (5) 154 (5) Education, n (%) Some high school or greater 4148 (87) 1614 (86) 2534 (87) 0.236 Primary school or less 638 (13) 270 (14) 368 (13) Mode of transport, n (%) Public transport (bus, taxi) 2283 [48] 877 (47) 1406 [48] < 0.001 Private transport 1117 [23] 524 (28) 593 (20) Other 1387 [29] 484 (26) 903 (31) Distance from clinic, n (%) Less than 5 km 1177 [25] 352 (19) 825 (28) < 0.001 At least 5 km 3610 (75) 1533 (81) 2077 (72) Health facility type, n (%) Primary health clinics 1234 [26] 404 (21) 830 (28) < 0.001 Hospital outpatient departments 3582 (74) 1493 (79) 2089 (72) Work hours outside home, n (%) None 2697 [56] 944 (50) 1753 [60] < 0.001 Less than 40 h 603 (13) 318 (17) 285 (10) 40 h or more 1516 [32] 635 (34) 881 (30) Prior HIV testing, n (%) Yes 1870 [39] 464 (25) 1406 [48] < 0.001 No 2917 (61) 1421 (75) 1496 [52] Health care use in prior year, n (%) None 715 (15) 256 (14) 459 (16) 0.006 1–2 times 1499 [31] 570 (30) 929 (32) 3–5 times 1732 [36] 684 (36) 1048 [36] > 5 times 841 (18) 375 (20) 466 (16) Visit to traditional healer in prior year, n (%) Yes 1567 [33] 708 (38) 859 (30) < 0.001 No 3220 (67) 1177 (62) 2043 (70) Social support score Median (IQR) 75 (54–87) 67 (50–83) 75 (60–90) < 0.001 Mental health score Median (IQR) 64 (56–80) 64 (56–76) 68 (56–84) < 0.001 Reported barriers to healthcare, n (%) Yes 1809 [38] 830 (44) 979 (34) < 0.001 Table 1 Differences between HIV-infected and HIV-uninfected participants at Overall, n = 4816 HIV-infected, Age, yrs Median (IQR) 31.0 (24–41) 33.0 (27–41) Sex, n (%) Male 2477 [51] 964 (51) Female 2339 [49] 933 (49) Marital status, n (%) Never married 3738 (78) 1535 (81) Currently married 810 (17) 265 (14) Divorce/separated/widowed 239 (5) 85 (5) Education, n (%) Some high school or greater 4148 (87) 1614 (86) Primary school or less 638 (13) 270 (14) Mode of transport, n (%) Public transport (bus, taxi) 2283 [48] 877 (47) Private transport 1117 [23] 524 (28) Other 1387 [29] 484 (26) Distance from clinic, n (%) Less than 5 km 1177 [25] 352 (19) At least 5 km 3610 (75) 1533 (81) Health facility type, n (%) Primary health clinics 1234 [26] 404 (21) Hospital outpatient departments 3582 (74) 1493 (79) Work hours outside home, n (%) None 2697 [56] 944 (50) Less than 40 h 603 (13) 318 (17) 40 h or more 1516 [32] 635 (34) Prior HIV testing, n (%) Yes 1870 [39] 464 (25) No 2917 (61) 1421 (75) Health care use in prior year, n (%) None 715 (15) 256 (14) 1–2 times 1499 [31] 570 (30) 3–5 times 1732 [36] 684 (36) > 5 times 841 (18) 375 (20) Visit to traditional healer in prior year, n (%) Yes 1567 [33] 708 (38) No 3220 (67) 1177 (62) Social support score Median (IQR) 75 (54–87) 67 (50–83) Mental health score Median (IQR) 64 (56–80) 64 (56–76) Reported barriers to healthcare, n (%) Yes 1809 [38] 830 (44) No 3007 (62) 1067 [56] Bassett et al. Balance after propensity score adjustment for estimating impact of HIV-infection on mortality A higher proportion of HIV-infected participants re- ported one or more barriers to healthcare compared to those HIV-uninfected (44% vs. 34%; P < 0.001). Among those who reported any barriers, HIV-infected partici- pants also reported more barriers than HIV-uninfected participants: 4 (IQR: 2–6) vs. 3 (IQR: 1–5; P < 0.001). Likewise, for those who reported any barriers, the total number of barriers spanned more domains for HIV-in- fected participants than for HIV-uninfected individuals: 3 (IQR: 2–4) vs. 2 (IQR: 1–4; P < 0.001). 414 (22%) HIV- infected participants had gone without healthcare for money to spend on basic needs (i.e. food, clothing, hous- ing), whereas only 506 (17%) HIV-uninfected partici- pants had done so (P < 0.001). Similarly, more HIV- infected individuals 323 (17%) had gone without basic needs for money to spend on healthcare compared to 401 (14%) HIV-uninfected participants (P = 0.002). Of 4816 enrollees, only 1154 of HIV-infected (61%) and 1354 of HIV-uninfected (46%) provided valid SAIDs. As shown in Additional file 1: Table S1, there were differ- ences in characteristics between those providing valid SAIDs and those who did not. As shown in Additional file 1: Table S2, initial propensity score weighting of individuals with a valid SAID reduced many of the im- balances between those with a valid SAID and those without valid SAID in both the HIV-infected and HIV- uninfected cohorts. A standard propensity score applied to these weighted population reduced imbalances be- tween HIV-infected and HIV-uninfected participants with valid SAID (Additional file 1: Table S3). Predictors of mortality Results for HIV-in- fected and HIV-uninfected individuals separately are re- ported in Additional file 1: Table S4. risk of dying, with risk increasing as healthcare use in- creased (> 5 times, HR: 2.34, 95% CI: 1.75–3.12; 3–5 times, HR: 1.86, 95% CI: 1.42–2.44, 1–2 times, HR: 1.53, 95% CI: 1.15–2.02 compared to no use in the past year). The effect of health care utilization on mortality was re- duced over time, showing the most substantial effect in the first third (125 deaths, first 3 months, P < 0.001) a smaller but still significant effect in the second third (130 deaths, months 4–21, P < 0.001) of deaths and mar- ginal impact in the last third of deaths (125 deaths, month 22 and later, P = 0.07). risk of dying, with risk increasing as healthcare use in- creased (> 5 times, HR: 2.34, 95% CI: 1.75–3.12; 3–5 times, HR: 1.86, 95% CI: 1.42–2.44, 1–2 times, HR: 1.53, 95% CI: 1.15–2.02 compared to no use in the past year). The effect of health care utilization on mortality was re- duced over time, showing the most substantial effect in the first third (125 deaths, first 3 months, P < 0.001) a smaller but still significant effect in the second third (130 deaths, months 4–21, P < 0.001) of deaths and mar- ginal impact in the last third of deaths (125 deaths, month 22 and later, P = 0.07). The integrated AUC for the overall Cox regression model was 0.755. A similar Cox model was fitted separ- ately to the HIV-uninfected and HIV-infected partici- pants with HIV status removed (and CD4 value also removed from the HIV-uninfected population). The model fit was better for HIV-uninfected compared to H IV-infected participants (AUC 0.837 vs. 0.687). There was a significant interaction of HIV-status with age (HR: 0.63, 95% CI: 0.56–0.70, P < 0.001 for each 10- year increase in age). Each additional 10 years of life in- creased the risk of death by 94% for HIV-uninfected par- ticipants (HR: 1.94, 95% CI: 1.78–2.11), but only by 22% (HR: 1.22, 95% CI: 1.13–1.31; P < 0.001) for HIV-infected participants. As shown in Fig. 1, the HR for HIV-in- fected individuals compared to HIV-uninfected individ- uals varied from 6.59 (95% CI: 4.79–9.06) at age 20 down to 1.13 (95% CI: 0.86–1.48) at age 60. Predictors of mortality Regardless of HIV status, patients using primary health clinics, as opposed to those using hospital outpatient de- partments, had reduced mortality risk (HR: 0.51, 95% CI: 0.38–0.68). A 10-point increase in mental health score decreased the risk of death by 7% (HR: 0.93; 95% CI: 0.89–0.97). Participants who used healthcare services in the year before enrollment in the study were at higher Patient perception of not being sick enough to seek care (1059; 22%; Table 2) was the most common individ- ual barrier reported. Uniformly, a higher proportion of participants who tested positive for HIV reported experi- encing each type of barrier to care than those who tested negative; all differences were statistically significant. The largest discrepancy between HIV-infected and HIV- Bassett et al. BMC Infectious Diseases (2019) 19:751 Page 6 of 11 Table 2 Differences in barriers between HIV-infected and HIV-uninfected participants at baseline Overall, n = 4816 HIV-infected, n = 1897 HIV-uninfected, n = 2919 P* Barrier domain reported, n (%) Service delivery 1152 [24] 566 (30) 586 (20) < 0.001 Too long to see nurse/MD 1016 [21] 506 (27) 510 (18) < 0.001 Not treated with respect 179 (4) 93 (5) 86 (3) 0.002 Financial 831 (17) 423 (22) 408 (14) < 0.001 Cost of transport 731 (15) 377 (20) 354 (12) < 0.001 Cost of medication 760 (16) 392 (21) 368 (13) < 0.001 Personal Health 1247 [26] 596 (31) 651 (22) < 0.001 Not sick enough 1059 [22] 503 (27) 556 (19) < 0.001 Too sick 433 (9) 244 (13) 189 (7) < 0.001 Logistical 656 (14) 334 (18) 322 (11) < 0.001 Could not get off work 382 (8) 213 (11) 169 (6) < 0.001 Taking care of someone else 371 (8) 182 (10) 189 (7) < 0.001 Structural 1110 [23] 540 (29) 570 (20) < 0.001 Did not know where to find care 398 (8) 181 (10) 217 (8) 0.033 Difficult hours 778 (16) 402 (21) 376 (13) < 0.001 Language barrier 231 (5) 112 (6) 119 (4) 0.014 Transport 688 (14) 353 (19) 335 (12) < 0.001 HIV-infected individuals 7% (HR: 1.07, 95% CI: 1.02–1.13; P = 0.002) There was also some evidence that the increased hazard for males was lower for HIV-infected men (HR: 1.38, 95% CI: 1.16–1.65) then for HIV-uninfected men (HR: 2.18, 95% CI: 1.64–2.89; P = 0.007). Predictors of mortality Sensitivity analyses showed that results were robust to the modeling assumptions with HR for HIV-infection ranging from 6.59 to 7.24 at age 20 and 1.13 to 1.29 at age 60; model details in Table 3. Discussion Instituting mental health screening not only during HIV testing but also during routine healthcare appointments could allow providers to identify patients with poor mental health and connect them to additional resources. We and others have found that depressive symptoms are common among HIV-in- fected patients in sub-Saharan Africa and are correlated with decreased likelihoods of obtaining a CD4 count or taking ART [57, 58]. In this study, we used the 5-item Mental Health Inventory screening test; this short survey may be feasible to include in routine healthcare visits. On the other hand, social support was not shown to affect 5-year mortality risk. This may be because we did not measure social support in the form of social integra- tion, which has been shown to be most predictive of mortality [35, 36]. This study highlights that barriers to care negatively affect survival regardless of HIV status. HIV-infected individuals reported experiencing significantly more barriers than HIV- uninfected individuals. This could be related to HIV-in- fected participants having fewer resources as significantly more HIV-infected participants also reported more com- peting needs than their HIV-uninfected counterparts. Among both HIV-infected and HIV-uninfected individuals, perception of personal health, service delivery, and struc- tural barriers were the most frequently reported barrier domains. Within those domains, both HIV-infected and HIV-uninfected participants most commonly reported waiting too long to see a nurse or doctor or not feeling sick enough to seek care as barriers. Recent studies suggest that improvements in clinic operations, including standardizing staff workloads and patient flow, introducing triage, and in- creasing staff size might help shorten wait times in low- and middle-income settings [53, 54]. We and others have found that in sub-Saharan African settings, participants often feel as though they are not sick enough to seek care [16, 55] or avoid seeking care when they do not feel ill for fear that treatment will make them feel worse [56]. Efforts to improve clinic operations and to promote seeking Despite continued efforts to diagnose and link individuals to HIV care in South Africa, HIV-infected individuals re- main at substantially increased risk for long-term mortality when compared to their HIV-uninfected counterparts at the same study sites and with similar socioeconomic status. A recent study reported similar findings in Botswana com- paring a population-based sample of HIV-infected and -un- infected individuals [59]. Discussion Among 2503 participants with valid SAIDs and complete data enrolled at 4 outpatient sites in Durban, South Africa between 2010 and 2013, more self-identified barrier domains and poorer mental health increase 5- year mortality risk, regardless of HIV status. For every 10-point decrease in mental health score, indicating poorer mental health, 5-year mortality increased by 7%. For each additional self-identified barrier domain 5-year mortality increased by 12%. There was some evidence for an interaction between HIV status and risk of 5-year mortality based on number of reported domains. A higher proportion of HIV-infected partici- pants reported self-identified barriers across every do- main when compared to HIV-uninfected participants. After adjusting for the HIV x age interaction, there was some evidence for an interaction of HIV-status with number of domains, with the mortality risk for HIV-un- infected individuals increasing by 23% (HR: 1.23, 95% CI: 1.14–1.32) for each additional domain, but more slowly for Bassett et al. BMC Infectious Diseases (2019) 19:751 Page 7 of 11 Fig. 1 5-year mortality risk among HIV-infected participants varies by age. Hazard ratio is calculated from the primary model, which includes both two-stage propensity score adjustment and multiple imputation of CD4 values in the HIV-infected population Fig. 1 5-year mortality risk among HIV-infected participants varies by age. Hazard ratio is calculated from the primary model, which includes both two-stage propensity score adjustment and multiple imputation of CD4 values in the HIV-infected population Additionally, increased healthcare use in the prior year contributed to increased hazard of death for both HIV-infected and HIV-uninfected individuals. Moreover, those who were HIV-infected had a 4-fold increase in hazard of death during follow-up compared to HIV-un- infected participants at age 31. These results were ro- bust when analyzed across multiple model variations. Additionally, increased healthcare use in the prior year contributed to increased hazard of death for both HIV-infected and HIV-uninfected individuals. Moreover, those who were HIV-infected had a 4-fold increase in hazard of death during follow-up compared to HIV-un- infected participants at age 31. These results were ro- bust when analyzed across multiple model variations. routine care may improve long-term mortality in both HIV-infected and HIV-uninfected individuals. We found that poor mental health also decreased sur- vival regardless of HIV status. Discussion Other studies have evaluated long-term mortality risk in HIV-infected individuals in sub- Saharan Africa [11–15], and a few have examined the im- pact of individual level contextual factors [16, 17]. Unlike Page 8 of 11 Bassett et al. BMC Infectious Diseases (2019) 19:751 Table 3 Predictors of Mortality Among HIV-infected and HIV-uninfected Patients in Durban, South Africa Measure Primary analysis Alternate analyses PS & CD4 MI HR (95% CI) P ǂ CD4 MI HR (95% CI) PS HR (95% CI) Base HR (95% CI) a. Adjusted hazard ratio of factors in the primary and alternative multivariable models, and in univariate analyses HIV Statusx 3.86 (3.08–4.85) < 0.001 3.99 (2.94–5.41) 4.44 (3.51–5.62) 4.54 (3.32–6.21) Male sex 1.57 (1.35–1.82) < 0.001 1.67 (1.35–2.06) 1.67 (1.44–1.93) 1.75 (1.41–2.16) Test received at primary health clinic 0.51 (0.38–0.68) < 0.001 0.60 (0.40–0.91) 0.47 (0.35–0.62) 0.55 (0.37–0.83) Healthcare use in past year: 1–2 times 1.53 (1.15–2.02) 0.003 1.66 (1.09–2.54) 1.65 (1.25–2.18) 1.79 (1.17–2.73) 3–5 times 1.86 (1.41–2.44) < 0.001 1.99 (1.32–3.00) 2.01 (1.54–2.63) 2.16 (1.44–3.25) > 5 times 2.34 (1.75–3.12) < 0.001 2.39 (1.55–3.67) 2.63 (1.98–3.49) 2.68 (1.75–4.10) Total number of healthcare barrier domains reported out of 5 possible 1.12 (1.07–1.16) < 0.001 1.11 (1.05–1.18) 1.14 (1.09–1.18) 1.13 (1.06–1.20) 10-point increase in mental health score corresponding to better mental health 0.93 (0.89–0.97) 0.002 0.94 (0.89–1.01) 0.92 (0.88–0.96) 0.94 (0.88–1.00) CD4 count (per 100) 0.80 (0.76–0.84) < 0.001 0.82 (0.76–0.88) – – b. Adjusted hazard ratios related to the interaction between age and HIV status 10-year age increase for HIV-infected* 1.22 (1.13–1.31) < 0.001 1.20 (1.07–1.35) 1.24 (1.15–1.34) 1.21 (1.08–1.36) 10-year age increase for HIV-uninfected* 1.94 (1.78–2.11) < 0.001 1.89 (1.69–2.11) 1.92 (1.77–2.09) 1.87 (1.67–2.09) HR for HIV-infected at age 20 6.59 (4.79–9.06) < 0.001 6.98 (4.43–11.00) 7.11 (5.07–9.96) 7.24 (4.61–11.38) HR for HIV-infected at age 30 4.24 (3.35–5.35) < 0.001 4.45 (3.18–6.22) 4.59 (3.58–5.89) 4.71 (3.38–6.56) HR for HIV-infected at age 40 2.72 (2.26–3.28) < 0.001 2.83 (2.18–3.67) 2.96 (2.46–3.57) 3.06 (2.38–3.94) HR for HIV-infected at age 50 1.75 (1.43–2.15) < 0.001 1.80 (1.38–2.36) 1.91 (1.60–2.29) 1.99 (1.54–2.57) HR for HIV-infected at age 60 1.13 (0.86–1.48) < 0.001 1.15 (0.81–1.63) 1.24 (0.98–1.56) 1.29 (0.92–1.82) PS: with propensity score adjustment; CD4 MI: with CD4 multiple imputation; HR: adjusted hazard ratio; Base: baseline analysis without propensity score adjustment or CD4 multiple imputation. Discussion x The HIV status variable is calculated as a contrast based on the parameter estimates for the HIV factor,and age factor, incorporating the average difference in age between the HIV+ and HIV- groups. For models including CD4, the HIV status variable also adjusts for the parameter estimate for CD4 and the and the average difference in CD4 between the HIV+ and HIV- groups ǂ p-values were generated from Cox proportional hazards models Bassett et al. BMC Infectious Diseases (2019) 19:751 Page 9 of 11 Page 9 of 11 including efforts to modify beliefs about the risks of HIV treatment, the benefits of seeking care when feeling healthy, and routine mental health monitoring. Such tar- geted interventions could improve health outcomes for high-risk individuals. Many structural and logistical barriers can be recognized early, i.e. at the first clinic visit, and could identify patients that may require more intensive follow-up. Both alleviation of barriers that pose increased mortality risks and development of tools to identify high-risk patients in clinical settings could significantly improve outcomes for HIV-infected and HIV-uninfected individuals. previous studies, however, the present study evaluated HIV-infected and HIV-uninfected individuals and found that some factors previously found to be associated with long-term mortality risk in HIV-infected cohorts, are also risk factors for HIV-uninfected individuals. This work should be considered in the context of sev- eral limitations. We did not adjust for data on ART use in this model; however, CD4 count was an eligibility cri- terion for starting ART during the study period and was included in the model. We did not collect data on other health-related comorbidities for either HIV-infected or HIV-uninfected participants; it is possible that those who reported more barriers to care may have also had higher rates of comorbid conditions. We also may underestimate the effects of mental health on mortality because we did not consider psychotic symptoms, which have been associated with an even higher relative risk of mortality compared to anxiety and depression [60]. Healthcare use in the prior year may be on the causal pathway to mortality, however, we felt this variable was an important predictor of time to death. We found that the effect of healthcare use in the prior year on mortality was reduced over time and may be a stronger predictor of early mortality than later mortality. The model results did not change qualitatively when this variable was removed (data not shown). Discussion In addition, only 52% of the participants provided a valid SAID number for death registry cross-matching and there were significant differ- ences between those with and without a valid SAID. Though we used propensity scores, we were unable to fully adjust for these differences. Lastly, the methods used to determine predictors of mortality in this study could not be readily used in a clinical setting. In addition to directly addressing those characteristics identified as pre- dictive of mortality, it is necessary to develop predictive instruments, for both HIV-infected and HIV-uninfected individuals, that can be implemented in patient care settings to identify at risk patients. Ethics approval and consent to participate pp p p The study was approved by the McCord Hospital Medical Research Ethics Committee, St. Mary’s Hospital Research Ethics Committee, University of KwaZulu-Natal Biomedical Research Ethics Committee and Partners Institutional Review Board (Protocol 2011-P-001195, Boston, MA). Written informed consent was obtained from all participants. Additional file Additional file 1: Statistical analysis and supplementary results tables. Variable selection procedure for random forests. Propensity score adjustment procedure. Imputation of CD4 values. Calculated effect of HIV. Table S1A and 1B – HIV-infected (1A) and HIV-uninfected (1B) comparing those with and without valid SA ID numbers. Table S2A and 2B show the standardized difference between the overall group (HIV-infected or HIV- uninfected) and the group with valid SA ID numbers. (DOCX 74 kb) Funding This study was funded by the US National Institute of Mental Health R01 MH090326 (IVB) and R01 MH108427 (IVB). This work was also supported by the Weissman Family MGH Research Scholar Award (IVB). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the US National Institutes of Health. The funder had no role in the design of the study, data collection, analysis, interpretation of data or in writing the manuscript. Authors’ contributions All authors have contributed significantly to this work and have read and approved this manuscript. IVB, Principal Investigator of this project, led the design and execution of this study as well as all stages of manuscript writing and preparation. JG and LB also contributed significantly to the design and execution of the study. AB facilitated crossmatch and interpretation of National Population Register data. RAP led all data analysis efforts, while EL also contributed substantially to statistical analysis. AX contributed significantly to data analysis and interpretation. LRIM, the Research Assistant, contributed substantially to manuscript drafting, editing, and revision. Abbreviations AUC: Area under the curve; HR: Hazard ratio; IQR: Interquartile range; MHC: Mental health composite; PLWH: People living with HIV; SAID: South African Identification Number; SMS: Short messaging service; SSI: Social support index Conclusions HIV infection remains a significant predictor of 5-year mortality in Durban, South Africa. However, additional screening for all patients can be used to help identify at risk individuals who may require additional healthcare interventions. While HIV-infected patients carry a higher burden of self-identified barriers to care than their HIV- uninfected counterparts, the effects of those barriers on mortality risk is not significantly different between HIV- infected and HIV-uninfected individuals. Similarly, while HIV-infected participants reported worse mental health than HIV-uninfected participants, poor mental health increased mortality risk for both groups. Interventions are needed that address both clinic-level barriers to care, such as long wait times, as well as patient-level barriers, Availability of data and materials The data supporting these research findings is available upon reasonable request to the corresponding author. References Mortality among adults transferred and lost to follow-up from antiretroviral therapy programmes in South Africa: a multicenter cohort study. J Acquir Immune Defic Syndr. 2014;67(2):e67–e75. 7. Cornell M, Lessells R, Fox MP, et al. Mortality among adults transferred and lost to follow-up from antiretroviral therapy programmes in South Africa: a multicenter cohort study. J Acquir Immune Defic Syndr. 2014;67(2):e67–e75. 31. Gallo JJ, Bogner HR, Morales KH, Post EP, Lin JY, Bruce ML. The effect of a primary care practice-based depression intervention on mortality in older adults: a randomized trial. Ann Intern Med. 2007;146(10):689–698. 8. Boulle A, Schomaker M, May MT, et al. Mortality in patients with HIV-1 infection starting antiretroviral therapy in South Africa, Europe, or North America: a collaborative analysis of prospective studies. PLoS Med. 2014;11(9):e1001718. 32. Gallo JJ, Morales KH, Bogner HR, et al. Long term effect of depression care management on mortality in older adults: follow-up of cluster randomized clinical trial in primary care. BMJ. 2013;346:f2570. 33. Bogner HR, Morales KH, Post EP, Bruce ML. Diabetes, depression, and death: a randomized controlled trial of a depression treatment program for older adults based in primary care (PROSPECT). Diabetes Care. 2007;30(12):3005–10. 9. Yiannoutsos CT, Johnson LF, Boulle A, et al. Estimated mortality of adult HIV-infected patients starting treatment with combination antiretroviral therapy. Sex Transm Infect. 2012;88(Suppl 2):i33–43. 34. Bogner HR, Joo JH, Hwang S, et al. Does a depression management program decrease mortality in older adults with specific medical conditions in primary care? An exploratory analysis. J Am Geriatr Soc. 2016;64(1):126–31. 10. Braitstein P, Brinkhof MW, Dabis F, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high-income countries. Lancet. 2006;367(9513):817–824. 35. Holt-Lunstad J, Robles TF, Sbarra DA. Advancing social connection as a public health priority in the United States. Am Psychol. 2017;72(6):517–30. 11. Cornell M, Johnson LF, Wood R, et al. Twelve-year mortality in adults initiating antiretroviral therapy in South Africa. J Int AIDS Soc. 2017;20(1): 21902. 36. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Med. 2010;7(7):e1000316. 12. Haas AD, Zaniewski E, Anderegg N, et al. Retention and mortality on antiretroviral therapy in sub-Saharan Africa: collaborative analyses of HIV treatment programmes. J Int AIDS Soc. 2018;21(2). 37. Bassett IV, Coleman SM, Giddy J, et al. Competing interests h h d l h Competing interests The authors declare that they have no competing interests. The authors declare that they have no competing interests. Page 10 of 11 Page 10 of 11 Page 10 of 11 Bassett et al. BMC Infectious Diseases (2019) 19:751 Bassett et al. BMC Infectious Diseases (2019) 19:751 Author details 1 18. Bassett IV, Regan S, Chetty S, et al. Who starts antiretroviral therapy in Durban, South Africa?... not everyone who should. AIDS. 2010;24(Suppl 1):S37–44. 1Division of Infectious Diseases, Massachusetts General Hospital, 100 Cambridge Street, 16th Floor, Boston, MA 02114, USA. 2Medical Practice Evaluation Center, Massachusetts General Hospital, Boston, MA, USA. 3Harvard Medical School, Boston, MA, USA. 4Harvard University Center for AIDS Research, Harvard University, Boston, MA, USA. 5Biostatistics Center, Massachusetts General Hospital, Boston, MA, USA. 6McCord Hospital, Durban, South Africa. 7RAND Corporation, Santa Monica, CA, USA. 8Centre for Infectious Diseases, Epidemiology and Research, School of Public Health and Family Medicine, University of Cape Town, Cape Town, South Africa. 9Department of Health, Provincial Government of the Western Cape, Cape Town, South Africa. 10Wellcome Centre for Infectious Diseases Research in Africa, Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Cape Town, South Africa. 11Department of Orthopedic Surgery, Brigham and Women’s Hospital, Boston, MA, USA. 12Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA. 19. Walker AS, Prendergast AJ, Mugyenyi P, et al. Mortality in the year following antiretroviral therapy initiation in HIV-infected adults and children in Uganda and Zimbabwe. Clin Infect Dis. 2012;55(12):1707–1718. 20. Walker AS, Prendergast AJ, Mugyenyi P, et al. Mortality in the year following antiretroviral therapy initiation in HIV-infected adults and children in Uganda and Zimbabwe. Clin Infect Dis. 2012;55(12):1707–18. 21. Brinkhof MW, Pujades-Rodriguez M, Egger M. Mortality of patients lost to follow-up in antiretroviral treatment programmes in resource-limited settings: systematic review and meta-analysis. PLoS One. 2009;4(6):e5790. 22. Coetzee D, Hildebrand K, Boulle A, et al. Outcomes after two years of providing antiretroviral treatment in Khayelitsha, South Africa. AIDS. 2004; 18(6):887–895. 23. Flynn AG, Anguzu G, Mubiru F, et al. Socioeconomic position and ten-year survival and virologic outcomes in a Ugandan HIV cohort receiving antiretroviral therapy. PLoS One. 2017;12(12):e0189055. Received: 17 May 2019 Accepted: 12 August 2019 Received: 17 May 2019 Accepted: 12 August 2019 Received: 17 May 2019 Accepted: 12 August 2019 24. Antelman G, Kaaya S, Wei R, et al. Depressive symptoms increase risk of HIV disease progression and mortality among women in Tanzania. J Acquir Immune Defic Syndr. 2007;44(4):470–477. References 25. Ickovics JR, Hamburger ME, Vlahov D, et al. Mortality, cd4 cell count decline, and depressive symptoms among hiv-seropositive women: longitudinal analysis from the hiv epidemiology research study. JAMA. 2001;285(11): 1466–74. 1. Joint United Nations Programme on HIV/AIDS. UNAIDS Data 2017. Geneva, Switzerland: UNAIDS; 2017. 1. Joint United Nations Programme on HIV/AIDS. UNAIDS Data 2017. Geneva, Switzerland: UNAIDS; 2017. 2. Johnson LF, Mossong J, Dorrington RE, et al. Life expectancies of south African adults starting antiretroviral treatment: collaborative analysis of cohort studies. PLoS Med. 2013;10(4):e1001418. 2. Johnson LF, Mossong J, Dorrington RE, et al. Life expectancies of south African adults starting antiretroviral treatment: collaborative analysis of cohort studies. PLoS Med. 2013;10(4):e1001418. 26. Satin JR, Linden W, Phillips MJ. Depression as a predictor of disease progression and mortality in cancer patients. Cancer. 2009;115(22):5349–61. 3. Mills EJ, Bakanda C, Birungi J, et al. Life expectancy of persons receiving combination antiretroviral therapy in low-income countries: a cohort analysis from Uganda. Ann Intern Med. 2011;155(4):209–216. 27. van Dooren FEP, Nefs G, Schram MT, Verhey FRJ, Denollet J, Pouwer F. Depression and risk of mortality in people with diabetes mellitus: a systematic review and meta-analysis. PLoS One. 2013;8(3):e57058. 4. Ingle SM, May M, Uebel K, et al. Outcomes in patients waiting for antiretroviral treatment in the Free State Province, South Africa: prospective linkage study. AIDS. 2010;24(17):2717–2725. 28. Rutledge T, Reis VA, Linke SE, Greenberg BH, Mills PJ. Depression in heart failure a meta-analytic review of prevalence, intervention effects, and associations with clinical outcomes. J Am Coll Cardiol. 2006;48(8):1527–1537. 5. Amuron B, Namara G, Birungi J, et al. Mortality and loss-to-follow-up during the pre-treatment period in an antiretroviral therapy programme under normal health service conditions in Uganda. BMC Public Health. 2009;9:290. 29. Burack JH, Barrett DC, Stall RD, Chesney MA, Ekstrand ML, Coates TJ. Depressive symptoms and CD4 lymphocyte decline among HIV-infected men. JAMA. 1993;270(21):2568–2573. 6. Bassett IV, Chetty S, Wang B, et al. Loss to follow-up and mortality among HIV-infected people co-infected with TB at ART initiation in Durban, South Africa. J Acquir Immune Defic Syndr. 2012;59(1):25–30. 30. Evans DL, Ten Have TR, Douglas SD, et al. Association of depression with viral load, CD8 T lymphocytes, and natural killer cells in women with HIV infection. Am J Psychiatry. 2002;159(10):1752–9. y 7. Cornell M, Lessells R, Fox MP, et al. References Sizanani: A randomized trial of health system navigators to improve linkage to HIV and TB care in South Africa. J Acquir Immune Defic Syndr. 2016;73(2):154–160. 13. Cornell M, Lessells R, Fox MP, et al. Mortality among adults transferred and lost to follow-up from antiretroviral therapy programmes in South Africa: a multicentre cohort study. J Acquir Immune Defic Syndr. 2014;67(2):e67–e75. 38. Bassett IV, Giddy J, Chaisson CE, et al. A randomized trial to optimize HIV/TB care in South Africa: design of the Sizanani trial. BMC Infect Dis. 2013;13:390. 38. Bassett IV, Giddy J, Chaisson CE, et al. A randomized trial to optimize HIV/TB care in South Africa: design of the Sizanani trial. BMC Infect Dis. 2013;13:390. 39. Rand Corporation. HIV Cost and Services Utilization Study. Available at: http://www.rand.org/health/projects/hcsus/Base/. Accessed 31 October 2014. 39. Rand Corporation. HIV Cost and Services Utilization Study. Available at: http://www.rand.org/health/projects/hcsus/Base/. Accessed 31 October 2014. 14. Hoffmann CJ, Schomaker M, Fox MP, et al. CD4 count slope and mortality in HIV-infected patients on antiretroviral therapy: multicohort analysis from South Africa. J Acquir Immune Defic Syndr. 2013;63(1):34–41. 40. Cunningham WE, Andersen RM, Katz MH, et al. The impact of competing subsistence needs and barriers on access to medical care for persons with human immunodeficiency virus receiving care in the United States. Med Care. 1999;37(12):1270–81. 15. Fox MP, Shearer K, Maskew M, et al. Treatment outcomes after 7 years of public-sector HIV treatment. AIDS. 2012;26(14):1823–1828. 16. Bassett IV, Coleman SM, Giddy J, et al. Barriers to Care and 1-Year Mortality Among Newly Diagnosed HIV-Infected People in Durban, South Africa. J Acquir Immune Defic Syndr. 2017;74(4):432–438. 41. Craw JA, Gardner LI, Marks G, et al. Brief strengths-based case management promotes entry into HIV medical care: results of the antiretroviral treatment access study-II. J Acquir Immune Defic Syndr. 2008;47(5):597–606. 17. Mukoswa GM, Charalambous S, Nelson G. The association between social capital and HIV treatment outcomes in South Africa. PLoS One. 2017;12(11):e0184140. 42. Hays RD, Sherbourne CD, Mazel RM. The RAND 36-item health survey 1.0. Health Econ. 1993;2(3):217–27. 42. Hays RD, Sherbourne CD, Mazel RM. The RAND 36-item health survey 1.0. Health Econ. 1993;2(3):217–27. Page 11 of 11 Bassett et al. BMC Infectious Diseases (2019) 19:751 Bassett et al. BMC Infectious Diseases (2019) 19:751 43. Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med. 1991;32(6):705–14. 44. Holmes WC. References Depressive symptoms and their impact on health-seeking behaviors in newly-diagnosed HIV-infected patients in Durban, South Africa AIDS Behav 2012;16(8):2226–2235. 59. Gaolathe T, Wirth K, Holmes MP, et al. Higher mortality in HIV-infected vs -uninfected adults despite ART, Botswana. CROI; 2018; Boston, Massachusetts. 59. Gaolathe T, Wirth K, Holmes MP, et al. Higher mortality in HIV-infected vs -uninfected adults despite ART, Botswana. CROI; 2018; Boston, Massachusetts. 60. Walker E, McGee R, Druss B. Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatry. 2015;72(4):334–41. 60. Walker E, McGee R, Druss B. Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatry. 2015;72(4):334–41. References A short, psychiatric, case-finding measure for HIV seropositive outpatients: performance characteristics of the 5-item mental health subscale of the SF-20 in a male, seropositive sample. Med Care. 1998;36(2):237–43. 45. Johnson LF, Dorrington RE, Laubscher R, et al. A comparison of death recording by health centres and civil registration in South Africans receiving antiretroviral treatment. J Int AIDS Soc. 2015;18(1). 46. Rosenbaum P, Rubin D. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41–55. observational studies for causal effects. Biometrika. 1983;70:41–55 47. Ishwaran H, Kogalur U, Blackstone E, Lauer M. Random survival forests. Ann Appl Stat. 2008;2(3):841–60. 48. Breiman L. Random forests. Mach Learn. 2001;45(1):5–32. 49. Cornell M, Schomaker M, Garone DB, et al. Gender differences in survival among adult patients starting antiretroviral therapy in South Africa: a multicentre cohort study. PLoS Med. 2012;9(9):e1001304. 50. Malaza A, Mossong J, Bärnighausen T, Viljoen J, Newell M-L. Population- based CD4 counts in a rural area in South Africa with high HIV prevalence and high antiretroviral treatment coverage. PLoS One. 2013;8(7):e70126. g g 51. Heagerty PJ, Zheng Y. Survival model predictive accuracy and ROC curves. Biometrics. Mar 2005;61(1):92–105. 52. Ishwaran H, Kogalur U. Random Forests for Survival, Regression, and Classification (RF-SRC), R package version 2.6.1. 2018; https://cran.r-project. org/web/packages/randomForestSRC/index.html. 53. Alamo ST, Wagner GJ, Ouma J, et al. Strategies for Optimizing Clinic Efficiency in a Community-Based Antiretroviral Treatment Programme in Uganda. AIDS Behav. 2013;17(1). 54. Wagenaar BH, Gimbel S, Hoek R, et al. Wait and consult times for primary healthcare services in Central Mozambique: a time-motion study. Glob Health Action. 2016;9. 55. Katirayi L, Chadambuka A, Muchedzi A, et al. Echoes of old HIV paradigms: reassessing the problem of engaging men in HIV testing and treatment through women’s perspectives. Reprod Health. 2017;14(1):124. 56. Katz IT, Dietrich J, Tshabalala G, et al. Understanding treatment refusal among adults presenting for HIV-testing in Soweto, South Africa: a qualitative study. AIDS Behav. 2015;19(4):704–14. 57. Kinyanda E, Levin J, Nakasujja N, et al. Major Depressive Disorder: Longitudinal Analysis of Impact on Clinical and Behavioral Outcomes in Uganda. J Acquir Immune Defic Syndr. 1 2018;78(2):136–143. 57. Kinyanda E, Levin J, Nakasujja N, et al. Major Depressive Disorder: Longitudinal Analysis of Impact on Clinical and Behavioral Outcomes in Uganda. J Acquir Immune Defic Syndr. 1 2018;78(2):136–143. 58. Ramirez-Avila L, Regan S, Giddy J, et al. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2110924724
https://boundaryvalueproblems.springeropen.com/counter/pdf/10.1186/s13661-015-0299-3
Latin
null
The existence of solutions for impulsive p-Laplacian boundary value problems at resonance on the half-line
Boundary value problems
2,015
cc-by
7,200
Abstract By using the continuous theorem of Ge and Ren and constructing suitable Banach spaces and operators, we investigate the existence of solutions for an impulsive p-Laplacian boundary value problem with integral boundary condition at resonance on the half-line. An example is given to illustrate our main results. MSC: 34B40 Keywords: impulsive; p-Laplacian operator; boundary value problem; integral boundary condition; resonance Keywords: impulsive; p-Laplacian operator; boundary value problem; integral boundary condition; resonance Weihua Jiang* Weihua Jiang* *Correspondence: weihuajiang@hebust.edu.cn College of Sciences, Hebei University of Science and Technology, Shijiazhuang, Hebei 050018, P.R. China 1 Introduction Boundary value problems on the half-line arise in various applications such as in the study of the unsteady flow of a gas through semi-infinite porous medium, in analyzing the heat transfer in radial flow between circular disks, in the study of plasma physics, in an analysis of the mass transfer on a rotating disk in a non-Newtonian fluid, etc. [] Many dynamical systems have an impulsive dynamical behavior due to abrupt changes at certain instants during the evolution process. The mathematical description of these phenomena leads to impulsive differential equations. For some general and recent works on the theory of impulsive differential equations we refer the reader to [–]. Impulsive differential equations occur in biology, medicine, mechanics, engineering, chaos theory, etc. [–]. Impulsive boundary value problems have been studied by many papers; see [–]. For example, in [], the authors studied the existence of solutions for the problem ⎧ ⎪⎪⎪⎨ ⎪⎪⎪⎩ (p(t)u′(t))′ = f (t,u(t)), t ∈(,∞)\{t,t,...,tn}, △u′(tk) = Ik(u(tk)), k = ,,...,n, αu() – β limt→+ p(t)u′(t) = , γ limt→∞u(t) + δ limt→∞p(t)u′(t) = . ⎧ ⎪⎪⎪⎨ ⎪⎪⎪⎩ (p(t)u′(t))′ = f (t,u(t)), t ∈(,∞)\{t,t,...,tn}, △u′(tk) = Ik(u(tk)), k = ,,...,n, αu() – β limt→+ p(t)u′(t) = , γ limt→∞u(t) + δ limt→∞p(t)u′(t) = . In [], the impulsive boundary value problem on the half-line ⎧ ⎪⎪⎪⎨ ⎪⎪⎪⎩  p(t)(p(t)x′(t))′ = f (t,xt), t ∈(,∞)\{t,t,...,tn}, △x′(tk) = Ik(xtk), k = ,,...,m, λx() – β limt→+ p(t)x′(t) = a, γ x(∞) + δ limt→∞p(t)x′(t) = b Jiang Boundary Value Problems ( 2015) 2015:39 DOI 10.1186/s13661-015-0299-3 R ES EA RCH Open Access (H) For any constant r > , there exist a function hr ∈L[,+∞) and a constant Mr > , such that |f (t,(+ t)u,v)| ≤hr(t), t ∈[,+∞), |u| < r, |v| < r, |Ii(u,v)| ≤Mr, i = ,,...,k, |u| ≤r(+ tk), |v| ≤r. © 2015 Jiang; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons Attribu- tion License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. Page 2 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 Page 2 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 A boundary value problem is said to be a resonance one if the corresponding homoge- neous boundary value problem has a non-trivial solution. The boundary value problems at resonance have been studied by many papers; see [–]. In [], the author gave the existence of solutions for the p-Laplacian boundary value problem at resonance on the half-line  (ϕp(u′))′(t) = ψ(t)f (t,u(t),u′(t)), t ∈[,+∞), u′(+∞) = , u() =  +∞  h(t)u(t)dt,  (ϕp(u′))′(t) = ψ(t)f (t,u(t),u′(t)), t ∈[,+∞), u′(+∞) = , u() =  +∞  h(t)u(t)dt, where ϕp(s) = |s|p–s, p > . where ϕp(s) = |s|p–s, p > . As far as we know, the impulsive p-Laplacian boundary value problems at resonance on the half-line have not been investigated. In this paper, we will discuss the existence of solutions for the problem ⎧ ⎪⎨ ⎪ (ϕp(u′(t)))′ + f (t,u(t),u′(t)) = , t ∈[,∞)\{t,t,...,tk}, △ϕp(u′(ti)) = Ii(u(ti),u′(ti)), i = ,,...,k,  +∞ (.) ⎧ ⎪⎨ ⎪⎩ (ϕp(u′(t)))′ + f (t,u(t),u′(t)) = , t ∈[,∞)\{t,t,...,tk}, △ϕp(u′(ti)) = Ii(u(ti),u′(ti)), i = ,,...,k, u() = , ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt, (.) (.) u() = , ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt, ⎨ ⎪⎩u() = , ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt, where < t< t< ··· < tk < +∞, △ϕp(u′(ti)) = ϕp(u′(ti + )) – ϕp(u′(ti – )). In this paper, we will always suppose that the following conditions hold. where < t< t< ··· < tk < +∞, △ϕp(u′(ti)) = ϕp(u′(ti + )) – ϕp(u′(ti – )). In this paper, we will always suppose that the following conditions hold. (H) h(t) ≥, t ∈[,+∞),  +∞  h(t)dt = , f : [,+∞) × R→R, and Ii : R→R, i = ,,...,k are continuous. (H) h(t) ≥, t ∈[,+∞),  +∞  h(t)dt = , f : [,+∞) × R→R, and Ii : R→R, i = ,,...,k are continuous. (a) (I – Q)Nλ( ) ⊂ImM ⊂(I – Q)Y, 3 Main results In the following, we will always suppose that q satisfies /p + /q = . Let R+ = [,+∞), J′ = R+\{t,t,...,tk}, Y = L(R+) with norm ∥y∥=  +∞  |y(t)|dt, PC R+ = u : u ∈C J′ ,u′(ti – ),u′(ti + ) exist and u′(ti – ) = u′(ti),i = ,,...,k , X = u : u() = ,u ∈C  R+ ∩PC R+ , sup t∈R+ |u(t)| + t < +∞, lim t→+∞u′(t) exists PC R+ = u : u ∈C J′ ,u′(ti – ),u′(ti + ) exist and u′(ti – ) = u′(ti),i = ,,...,k , with norm ∥u∥= max{∥u +t∥∞,∥u′∥∞}, where ∥u∥∞= supt∈R+ |u(t)|. with norm ∥u∥= max{∥u +t∥∞,∥u′∥∞}, where ∥u∥∞= supt∈R+ |u(t)|. Let Z = Y ×Rk, with norm ∥(y,c,c,...,ck)∥= max{∥y∥,|c|,|c|,...,|ck|}. Then (X,∥·∥) and (Z,∥· ∥) are Banach spaces. Define the operators M : X ∩domM →Z, Nλ : X →Z as follows: Mu = ⎡ ⎢⎢⎢⎣ (ϕp(u′))′(t) △ϕp(u′(t)) ··· △ϕp(u′(tk)) ⎤ ⎥⎥⎥⎦, Nλu = ⎡ ⎢⎢⎢⎣ –λf (t,u(t),u′(t)) λI(u(t),u′(t)) ··· λIk(u(tk),u′(tk)) ⎤ ⎥⎥⎥⎦, where domM = {u ∈X : (ϕp(u′))′ ∈Y,ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt}. where domM = {u ∈X : (ϕp(u′))′ ∈Y,ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt}. where domM = {u ∈X : (ϕp(u′))′ ∈Y,ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt}. It is clear that u ∈domM is a solution of the problem (.) if it satisfies Mu = Nu, where N = N. For convenience, let (a,b)T :=  a b  , denote J= [,t], Ji = (ti,ti+], i = ,,...,k – , Jk = (tk,+∞). Lemma .M is a quasi-linear operator. M : X ∩domM →Y M : X ∩domM →Y M : X ∩domM →Y is a quasi-linear operator and Nλ : →Y, λ ∈[,] is M-compact. In addition, if the following conditions hold: is a quasi-linear operator and Nλ : →Y, λ ∈[,] is M-compact. In addition, if the following conditions hold: (C) Mx ̸= Nλx, ∀x ∈∂ ∩domM, λ ∈(,), (C) deg{JQN, ∩KerM,} ̸= , then the abstract equation Mx = Nx has at least one solution in domM ∩ , where N = N, J : ImQ →KerM is a homeomorphism with J(θ) = θ. then the abstract equation Mx = Nx has at least one solution in domM ∩ , where N = N, J : ImQ →KerM is a homeomorphism with J(θ) = θ. 2 Preliminaries For convenience, we introduce some notations and a theorem. For more details see []. For convenience, we introduce some notations and a theorem. For more details see []. Definition .[] Let X and Y be two Banach spaces with norms ∥· ∥X, ∥· ∥Y , respec- tively. A continuous operator M : X ∩domM →Y is said to be quasi-linear if Definition .[] Let X and Y be two Banach spaces with norms ∥· ∥X, ∥· ∥Y , respec- tively. A continuous operator M : X ∩domM →Y is said to be quasi-linear if (i) ImM := M(X ∩domM) is a closed subset of Y, (i) ImM := M(X ∩domM) is a closed subset of Y, (ii) KerM := {x ∈X ∩domM : Mx = } is linearly homeomorphic to Rn, n < ∞, where domM denote the domain of the operator M. Let X= KerM and Xbe the complement space of Xin X, then X = X⊕X. On the other hand, suppose Yis a subspace of Y and that Yis the complement of Yin Y, i.e. Y = Y⊕Y. Let P : X →Xand Q : Y →Ybe two projectors and ⊂X an open and bounded set with the origin θ ∈ . Let X= KerM and Xbe the complement space of Xin X, then X = X⊕X. On the other hand, suppose Yis a subspace of Y and that Yis the complement of Yin Y, i.e. Y = Y⊕Y. Let P : X →Xand Q : Y →Ybe two projectors and ⊂X an open and bounded set with the origin θ ∈ . Definition .[] Suppose that Nλ : →Y, λ ∈[,] is a continuous operator. Denote Nby N. Let λ = {x ∈ : Mx = Nλx}. Nλ is said to be M-compact in if there exist a vector subspace Yof Y satisfying dimY= dimXand an operator R : × [,] →X being continuous and compact such that for λ ∈[,], Definition .[] Suppose that Nλ : →Y, λ ∈[,] is a continuous operator. Denote Nby N. Let λ = {x ∈ : Mx = Nλx}. 2 Preliminaries Nλ is said to be M-compact in if there exist a vector subspace Yof Y satisfying dimY= dimXand an operator R : × [,] →X being continuous and compact such that for λ ∈[,], (a) (I – Q)Nλ( ) ⊂ImM ⊂(I – Q)Y, (b) QNλx = θ,λ ∈(,) ⇔QNx = θ, Page 3 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 (c) R(·,) is the zero operator and R(·,λ)| λ = (I – P)| λ, (d) M[P + R(·,λ)] = (I – Q)Nλ. (c) R(·,) is the zero operator and R(·,λ)| λ = (I – P)| λ, (d) M[P + R(·,λ)] = (I – Q)Nλ. Theorem .[] Let X and Y be two Banach spaces with the norms ∥· ∥X, ∥· ∥Y , respec- tively, and ⊂X an open and bounded nonempty set. Suppose that Lemma .M is a quasi-linear operator. Proof It is easy to get KerM = {at | a ∈R} := X. Proof It is easy to get KerM = {at | a ∈R} := X. For u ∈X ∩domM, if Mu = (y,c,c,...,ck)T, then Proof It is easy to get KerM = {at | a ∈R} := X. For u ∈X ∩domM, if Mu = (y,c,c,...,ck)T, then For u ∈X ∩domM, if Mu = (y,c,c,...,ck)T, then  ϕp  u′ ′(t) = y(t), △ϕp  u′(ti) = ci, i = ,,...,k. Page 4 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 For t ∈J, we get ϕp  u′(t) =  t  y(s)ds + a. For t ∈J, we get ϕp  u′(t) =  t  y(s)ds + a. For t ∈J, considering △ϕp(u′(t)) = c, we get For t ∈J, considering △ϕp(u′(t)) = c, we get ϕp  u′(t) =  t  y(s)ds + a + c. ϕp  u′(t) =  t  y(s)ds + a + c. For t ∈Ji, i = ,,...,k, considering △ϕp(u′(ti)) = ci, we get For t ∈Ji, i = ,,...,k, considering △ϕp(u′(ti)) = ci, we get ϕp  u′(t) =  t  y(s)ds + a +  ti<t ci. ϕp  u′(t) =  t  y(s)ds + a +  ti<t ci. By ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt and  +∞  h(t)dt = , we find that (y,c,c,...,ck)T sat- isfies By ϕp(u′(+∞)) =  +∞  h(t)ϕp(u′(t))dt and  +∞  h(t)dt = , we find that (y,c,c,...,ck)T sat- isfies  +∞  h(t)  +∞ t y(s)ds +  tk   ti≥t cih(t)dt = . (.) (.) On the other hand, if (y,c,c,...,ck)T satisfies (.), take On the other hand, if (y,c,c,...,ck)T satisfies (.), take u(t) =  t  ϕq  s  y(r)dr +  ti<s ci  ds. mple calculation, we get u ∈X ∩domM and Mu = (y,c,c,...,ck)T. Thus By a simple calculation, we get u ∈X ∩domM and Mu = (y,c,c,...,ck)T. Thus ImM = (y,c,c,...,ck)T | y ∈Y,c,c,...,ck satisfies (.) . □ Obviously, ImM ⊂Z is closed. So, M is quasi-linear. The proof is completed. Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. Proof By (H), (H), the continuity of ϕq and Lebesgue’s dominated convergence theorem, we find that R is continuous and {R(u,λ) | u ∈ ,λ ∈[,]} is bounded. We will prove that R( × [,]) is compact. Since ⊂X is bounded, there exists a constant r > such that ∥u∥≤r, u ∈ . It fol- lows from (H) that there exist a function hr ∈L(R+) and a constant Mr > such that |f (t,u(t),u′(t))| ≤hr(t), |Ii(u(ti),u′(ti))| ≤Mr, i = ,,...,k, t ∈R+, u ∈ . Lemma .M is a quasi-linear operator. □ Take projectors P : X →Xand Q : Z →Zas follows: Take projectors P : X →Xand Q : Z →Zas follows: (Pu)(t) = u′(+∞)t, Q(y,c,c,...,ck)T =  +∞  h(t)  +∞ t y(s)dsdt +  tk   ti≥t cih(t)dt  +∞  h(t)e–t dt e–t,,..., T , where Z= {(ce–t,,...,)T | c ∈R}. Obviously, QZ = Zand dimZ= dimX. Define an operator R : X × [,] →Xas R(u,λ)(t) =  t  ϕq  +∞ s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr dr + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds – u′(+∞)t, t ∈Ji,i = ,,...,k, where X⊕X= X. where X⊕X= X. Page 5 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 By [, ], we get the following lemma. By [, ], we get the following lemma. Lemma .Assume that V ⊂X is bounded. V is compact if { u(t) +t : u ∈V} and {u′(t) : u ∈ V} are both equicontinuous on Ji, i = ,,...,k – , and JT = (tk,T], for any given T > tk, respectively, and equiconvergent at infinity. Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. We have R(u,λ)′(x) – R(u,λ)′(x)  = ϕq  +∞ x λ  f  s,u(s),u′(s) = ϕq  +∞ x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds + ϕp  u′(+∞) – λ  tj≥x Ij  u(tj),u′(tj)  – ϕq  +∞ x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s + ϕp  u′(+∞) – λ  tj≥x Ij  u(tj),u′(tj)  – ϕq  +∞ x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  d + ϕp  u′(+∞) – λ  tj≥x Ij  u(tj),u′(tj) . For u ∈ , λ ∈[,], define F(u,λ)(t) =  +∞ t λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds + ϕp  u′(+∞) – λ  tj≥t Ij  u(tj),u′(tj) . –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  d  +∞  h(t)e–t dt Obviously, Obviously, F(u,λ)(t)  ≤∥hr∥+ ∥hr∥+ kMr  +∞  h(t)e–t dt + ϕp(r) + kMr := K, u ∈ ,λ ∈[,],t ∈R+, F(u,λ)(x) – F(u,λ)(x)  =   x x λ  f  s,u(s),u′(s) F(u,λ)(t)  ≤∥hr∥+ ∥hr∥+ kMr  +∞  h(t)e–t dt + ϕp(r) + kMr := K, u ∈ ,λ ∈[,],t ∈R+, F(u,λ)(x) – F(u,λ)(x) =   x x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds  ≤  x x hr(t)dt + ∥hr∥+ kMr  +∞  h(t)e–t dt  e–x– e–x , u ∈ ,λ ∈[,]. –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds ≤  x x hr(t)dt + ∥hr∥+ kMr  +∞  h(t)e–t dt  e–x– e–x , u ∈ ,λ ∈[,]. Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. For any given T > tk, x,x∈Ji, i = ,,...,k – ,T, x< x, we have R(u,λ)(x) + x – R(u,λ)(x) + x  ≤   + x  x  ϕq  +∞ s λ  f  x,u(x),u′(x) R(u,λ)(x) + x – R(u,λ)(x) + x  ≤   + x  x  ϕq  +∞ s λ  f  x,u(x),u′(x) ≤   + x  x  ϕq  +∞ s λ  f  x,u(x),u′(x) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–x –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–x  d + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds –  + x  x  ϕq  +∞ s λ  f  x,u(x),u′(x) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–x  dx + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds  +  x + x – x + x  u′(+∞) ≤   + x –  + x Tϕq  +∞   hr(x) +  +∞  hr(t)dt + k i=|Ii(u(ti),u′(ti))|  +∞  h(t)e–t dt e–x  dx + ϕp(r) + kMr  + x– x + x ϕq  +∞   hr(x) +  +∞  hr(t)dt + k i=|Ii(u(ti),u′(ti))|  +∞  h(t)e–t dt e–x  dx + ϕp(r) + kMr  +  x + x – x + x r ≤   + x –  + x T + (x– x)  ϕq  ∥hr∥+ ∥hr∥+ kMr  +∞  h(t)e–t dt + ϕp(r) + kMr  +  x + x – x + x r. Page 6 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 Since t,  +t , and t +t are equicontinuous on Ji, i = ,,...,k – ,T, we find that { R(u,λ)(t) +t ,u ∈ ,λ ∈[,]} are equicontinuous on Ji, i = ,,...,k – ,T. Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. It follows from the absolute continuity of integral and the equicontinuity of e–t that {F(u,λ)(t),u ∈ ,λ ∈[,]} are equicontinuous on Ji, i = ,,...,k – ,T. By the uniform continuity of ϕq(t) in [–K,K], we find that {R(u,λ)′(t),u ∈ ,λ ∈[,]} are equicontinuous on Ji, i = ,,...,k – ,T. Page 7 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 For any u ∈ , λ ∈[,], since For any u ∈ , λ ∈[,], since  +∞ t λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds  ≤  +∞ t hr(s) + ∥hr∥+ kMr  +∞  h(t)e–t dt e–s ds → (t →∞) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds ≤  +∞ t hr(s) + ∥hr∥+ kMr  +∞  h(t)e–t dt e–s ds → (t →∞) and ϕq(u) is uniform continuous on [–K – ϕp(r),K + ϕp(r)], for any ε > , there exists a constant T> tk such that and ϕq(u) is uniform continuous on [–K – ϕp(r),K + ϕp(r)], for any ε > , there exists a constant T> tk such that and ϕq(u) is uniform continuous on [–K – ϕp(r),K + ϕp(r)], for any ε > , there exists a constant T> tk such that ϕq  +∞ s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  d + ϕp  u′(+∞)  – u′(+∞)  ≤ε , s > T,u ∈ ,λ ∈[,]. Obviously, there exists a constant T > Tsuch that, for any t > T, Obviously, there exists a constant T > Tsuch that, for any t > T,  + t  ϕq(K) + r T< ε . Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. Thus, for any x,x> T, we have Thus, for any x,x> T, we have R(u,λ)(x) + x – R(u,λ)(x) + x  =   + x  x  ϕq  +∞ s λ  f  r,u(r),u′(r) =   + x   ϕq  s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds – u′(+∞)x –  + x  x  ϕq  +∞ s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds – u′(+∞)x  ≤   + x  T  ϕq  +∞ s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr   s  –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds – u′(+∞)x –  + x  x  ϕq  +∞ s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr + ϕp  u′(+∞) – λ  tj≥s Ij  u(tj),u′(tj)  ds – u′(+∞)x  ≤   + x  T  ϕq  +∞ s λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e Page 8 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr + ϕp  u′(+∞)  ds – u′(+∞)(x– T)  ≤  + x  ϕq(K) + r T+ x– T + x ε +  + x  ϕq(K) + r T+ x– T + x ε < ε, ≤  + x  ϕq(K) + r T+ x– T + x ε +  + x  ϕq(K) + r T+ x– T + x ε < ε, R(u,λ)′(x) – R(u,λ)′(x) ≤ ϕq  +∞ x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds + ϕp  u′(+∞)  – u′(+∞)  + ϕq  +∞ x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds + ϕp  u′(+∞)  – u′(+∞)  < ε + ε < ε. Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set. For u ∈ , λ ∈[,], t ∈Ji, i = ,,,...,k, we have For u ∈ , λ ∈[,], t ∈Ji, i = ,,,...,k, we have  ϕp  Pu + R(u,λ) ′ ′(t) = –λf  t,u(t),u′(t) –  +∞  h(t)  +∞ t –λf (s,u(s),u′(s))dsdt + λ  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–t  ϕp  Pu + R(u,λ) ′ ′(t) = –λf  t,u(t),u′(t) –  +∞  h(t)  +∞ t –λf (s,u(s),u′(s))dsdt + λ  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–t and and ϕp  Pu + R(u,λ) ′(t) =  +∞ t λ  f  r,u(r),u′(r) ϕp  Pu + R(u,λ) (t) =  +∞ t λ  f  r,u(r),u′(r) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  dr + ϕp  u′(+∞) – λ  tj≥t Ij  u(tj),u′(tj) . –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–r  d + ϕp  u′(+∞) – λ  tj≥t Ij  u(tj),u′(tj) . By a simple calculation, we can get By a simple calculation, we can get By a simple calculation, we can get M  Pu + R(u,λ)  = (I – Q)Nλu. M  Pu + R(u,λ)  = (I – Q)Nλu. So, Definition .(d) holds. These, together with Lemma ., mean that Nλ is M-compact in . The proof is completed. □ Lemma .R : × [,] →Xis continuous and compact, where ⊂X is an open bounded set.   h(t)e–t dt + ϕp  u′(+∞)  – u′(+∞)  + ϕq  +∞ x λ  f  s,u(s),u′(s) –  +∞  h(t)  +∞ t f (s,u(s),u′(s))dsdt –  tk   ti≥t Ii(u(ti),u′(ti))h(t)dt  +∞  h(t)e–t dt e–s  ds + ϕp  u′(+∞)  – u′(+∞)  < ε + ε < ε. By Lemma ., we find that {R(u,λ) | u ∈ ,λ ∈[,]} is compact. The proof is com- pleted. □ Lemma .Assume that ⊂X is an open bounded set. Then Nλ is M-compact in . Proof By (H), we get Nλ : →Y, λ ∈[,] is continuous. It is clear that ImP = KerM, QNλx = θ,λ ∈(,) ⇔QNx = θ, i.e. Definition .(b) holds. Page 9 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 Page 9 of 13 For u ∈ , it follows from Q(I –Q)Nλu = θ that (I –Q)Nλu satisfies (.). So, (I –Q)Nλu ∈ ImM, i.e. (I –Q)Nλ( ) ⊂ImM. Furthermore, by ImM = KerQ and z = Qz+(I –Q)z we find that z ∈ImM implies z = (I – Q)z ∈(I – Q)Z, i.e. ImM ⊂(I – Q)Z. Thus, (I – Q)Nλ( ) ⊂ ImM ⊂(I – Q)Z, i.e. Definition .(a) holds. Obviously, R(·,) = . For u ∈ λ = {u ∈ ∩domM : Mu = Nλu}, we get QNλu = θ and ϕp  u′(t) =  +∞ t λf  s,u(s),u′(s) ds + ϕp  u′(+∞) – λ  ti≥t Ii  u(ti),u′(ti) . So, we have So, we have R(u,λ) =  t  ϕq  ϕp  u′(s) ds – u′(+∞)t = (I – P)u, i.e. Definition .(c) holds. i.e. Definition .(c) holds. i.e. Definition .(c) holds. Theorem .Assume that (H), (H), and the following conditions hold: Theorem .Assume that (H), (H), and the following conditions hold: (H) There exist nonnegative functions a(t), b(t), c(t), and nonnegative constants di, gi, ei, i = ,,...,k with (+t)p–a(t),b(t),c(t) ∈Y, and ∥a(t)(+t)p–∥+∥b∥+k i=[di(+ ti)p–+ gi] < such that f (t,x,y)  ≤a(t) ϕp(x)  + b(t) ϕp(y)  + c(t), a.e. t ∈[,+∞),x,y ∈R, Ii(x,y)  ≤di ϕp(x)  + gi ϕp(y)  + ei, i = ,,...,k,x,y ∈R. f (t,x,y)  ≤a(t) ϕp(x)  + b(t) ϕp(y)  + c(t), a.e. t ∈[,+∞),x,y ∈R, Ii(x,y)  ≤di ϕp(x)  + gi ϕp(y)  + ei, i = ,,...,k,x,y ∈R. Page 10 of 13 Page 10 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 Page 10 of 13 (H) There exists a constant e> such that if inft∈R+ |u′(t)| > e, then one of the following inequalities holds: (H) There exists a constant e> such that if inft∈R+ |u′(t)| > e, then one of the following inequalities holds: () u′(t)  +∞  h(t)  +∞ t f  s,u(s),u′(s) ds –  ti≥t Ii  u(ti),u′(ti)  dt > ; () u′(t)  +∞  h(t)  +∞ t f  s,u(s),u′(s) ds –  ti≥t Ii  u(ti),u′(ti)  dt < , where t ∈[,+∞). Then boundary value problem (.) has at least one solution. where t ∈[,+∞). Then boundary value problem (.) has at least one solution. where t ∈[,+∞). Then boundary value problem (.) has at least one solution. where t ∈[,+∞). Then boundary value problem (.) has at least one solution. (.) (.) By (.), (H), and (.), we obtain By (.), (H), and (.), we obtain By (.), (H), and (.), we obtain ϕp  u′(t)  ≤  +∞   a(t) ϕp  u(t)  + b(t) ϕp  u′(t)  + c(t)  dt + ϕp(e) + k  i=  di ϕp  u(ti)  + gi ϕp  u′(ti)  + ei ≤  a(t)(+ t)p– + k  i= di(+ ti)p– ϕp  u + t  ∞  Page 11 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 +  ∥b∥+ k  i= gi ϕp u′ ∞ + ∥c∥+ ϕp(e) + k  i= ei ≤  a(t)(+ t)p– + ∥b∥+ k  i=  di(+ ti)p–+ gi  ϕp u′ ∞ + ∥c∥+ ϕp(e) + k  i= ei. Thus u′ ∞≤ϕq  ∥c∥+ ϕp(e) + k i=ei – (∥a(t)(+ t)p–∥+ ∥b∥+ k i=[di(+ ti)p–+ gi])  . +  ∥b∥+ k  i= gi ϕp u′ ∞ + ∥c∥+ ϕp(e) + k  i= ei ≤  a(t)(+ t)p– + ∥b∥+ k  i=  di(+ ti)p–+ gi  ϕp u′ ∞ + ∥c∥+ ϕp(e) + k  i= ei. Thus u′ ∞≤ϕq  ∥c∥+ ϕp(e) + k i=ei – (∥a(t)(+ t)p–∥+ ∥b∥+ k i=[di(+ ti)p–+ gi])  . Thus u′ ∞≤ϕq  ∥c∥+ ϕp(e) + k i=ei – (∥a(t)(+ t)p–∥+ ∥b∥+ k i=[di(+ ti)p–+ gi])  . □ This, together with (.), means that is bounded in X. □ This, together with (.), means that is bounded in X. Lemma .Assume that (H), (H), and (H) hold. Then = {u ∈KerM | QNu = } is bounded in X, where N = N. Proof For u ∈ , we have u = at, a ∈R, and Q(Nu) = , i.e.  +∞  h(t)  +∞ t f (s,as,a)dsdt –  tk   ti≥t Ii(ati,a)h(t)dt =  +∞  h(t)  +∞ t f (s,as,a)ds –  ti≥t Ii(ati,a)  dt = . By (H), we get ∥u∥= |a| = |u′(t)| ≤e. So, is bounded. The proof is completed. □ □ By (H), we get ∥u∥= |a| = |u′(t)| ≤e. So, is bounded. The proof is completed. □ Proof of Theorem Let = {u ∈X | ∥u∥< r} where r > eis large enough such that By (H), we get ∥u∥= |a| = |u′(t)| ≤e. So, is bounded. The proof is completed. where t ∈[,+∞). Then boundary value problem (.) has at least one solution. In order to prove Theorem ., we show two lemmas. In order to prove Theorem ., we show two lemmas. In order to prove Theorem ., we show two lemmas. Lemma .Suppose that (H)-(H) hold. Then the set Lemma .Suppose that (H)-(H) hold. Then the set Lemma .Suppose that (H)-(H) hold. Then the set = u ∈domM | Mu = Nλu,λ ∈(,) = u ∈domM | Mu = Nλu,λ ∈(,) is bounded in X. Proof For u ∈ , we have QNλu = , i.e. Proof For u ∈ , we have QNλu = , i.e.  +∞  h(t)  +∞ t f  s,u(s),u′(s) dsdt –  tk   ti≥t Ii  u(ti),u′(ti) h(t)dt =  +∞  h(t)  +∞ t f  s,u(s),u′(s) ds –  ti≥t Ii  u(ti),u′(ti)  dt = . By (H), there exists a constant t∈R+ such that |u′(t)| ≤e. Assume t∈Jm, m = ,,...,k. It follows from Mu = Nλu that By (H), there exists a constant t∈R+ such that |u′(t)| ≤e. Assume t∈Jm, m = ,,...,k. It follows from Mu = Nλu that ⎧ ⎪⎪⎪⎪⎪⎪⎨  t t λf (s,u(s),u′(s))ds + ϕp(u′(t)) – λm j=i+Ij(u(tj),u′(tj)), t ∈Ji,i = ,,...,m – , ϕp  u′(t) = ⎧ ⎪⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎪⎩  t t λf (s,u(s),u′(s))ds + ϕp(u′(t)) – λm j=i+Ij(u(tj),u′(tj)), t ∈Ji,i = ,,...,m – ,  t t λf (s,u(s),u′(s))ds + ϕp(u′(t)), t ∈Jm,  t t λf (s,u(s),u′(s))ds + ϕp(u′(t)) + λi j=m+Ij(u(tj),u′(tj)), t ∈Ji,i = m + ,m + ,...,k. (.) ϕp  u′(t) = ⎪⎨ ⎪⎪⎪⎪⎪⎪⎩  t t λf (s,u(s),u′(s))ds + ϕp(u′(t)), t ∈Jm,  t t λf (s,u(s),u′(s))ds + ϕp(u′(t)) + λi j=m+Ij(u(tj),u′(tj)), t ∈Ji,i = m + ,m + ,...,k. (.) (.) Since u(t) =  t u′(s)ds, Since u(t) =  t u′(s)ds, |u(t)| + t ≤ u′ ∞, t ∈[,+∞). where t ∈[,+∞). Then boundary value problem (.) has at least one solution. □ y ( ) g ∥∥ | | | ( )| p p Proof of Theorem .Let = {u ∈X | ∥u∥< r}, where r > eis large enough such that ⊃ ∪ . Proof of Theorem .Let = {u ∈X | ∥u∥< r}, where r > eis large enough such that ⊃ ∪ . Proof of Theorem .Let = {u ∈X | ∥u∥< r}, where r > eis large enough such that ⊃ ∪ . By Lemmas .and ., we have Mu ̸= Nλu, u ∈domM ∩∂ , and QNu ̸= , u ∈KerM ∩ ∂ . Let H(u,δ) = ρδu + (– δ)JQNu, δ ∈[,], u ∈KerM ∩ , where J : ImQ →KerM is a homeomorphism with J(ae–t,,...,)T = at, ρ = !–, if (H)() holds, , if (H)() holds. For u ∈KerM ∩∂ , we have u = at ̸= . Thus ! , if (H)() holds. For u ∈KerM ∩∂ , we have u = at ̸= . Thus H(u,δ) = ρδat – (– δ)  +∞  h(t)  +∞ t f (s,as,a)dsdt –  tk   ti≥t Ii(ati,a)h(t)dt  +∞  h(t)e–t dt t. If δ = , H(u,) = ρat ̸= . If δ = , by QNu ̸= , we get H(u,) = JQN(at) ̸= . For < δ < , we now prove that H(u,δ) ̸= . Otherwise, if H(u,δ) = , then  +∞  h(t)  +∞ t f (s,as,a)dsdt –  tk   ti≥t Ii(ati,a)h(t)dt = ρδa – δ  +∞  h(t)e–t dt. Page 12 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 Thus a  +∞  h(t)  +∞ t f (s,as,a)ds –  ti≥t Ii(ati,a)  dt = ρδa – δ  +∞  h(t)e–t dt. Since |u′(t)| = |a| = ∥u∥= r > e, this is a contradiction with (H) and the definition of ρ. So, H(u,δ) ̸= , u ∈KerM ∩∂ , δ ∈[,]. Since |u′(t)| = |a| = ∥u∥= r > e, this is a contradiction with (H) and the definition of ρ. So, H(u,δ) ̸= , u ∈KerM ∩∂ , δ ∈[,]. Since |u′(t)| = |a| = ∥u∥= r > e, this is a contradiction with (H) and the definition of ρ. So, H(u,δ) ̸= , u ∈KerM ∩∂ , δ ∈[,]. Competing interests Competing interests The author declares that she has no competing interests. where t ∈[,+∞). Then boundary value problem (.) has at least one solution. By the homotopy of degree, we get By the homotopy of degree, we get deg " JQN, ∩KerM, # = deg " H(·,), ∩KerM, # = deg " H(·,), ∩KerM, # = deg " ρI, ∩KerM, # ̸= . By Theorem ., we can find that Mu = Nu has at least one solution in . The proof is completed. □ □ 4 Example Let us consider the following impulsive p-Laplacian boundary value problems at reso- nance on the half-line ⎧ ⎪⎨ ⎪⎩ (ϕp(u′(t)))′ + f (t,u(t),u′(t)) = , t ∈[,∞)\{t,t,...,tk}, △ϕp(u′(ti)) = ci, i = ,,...,k, u() = , ϕp(u′(+∞)) =  +∞  e–tϕp(u′(t))dt, (.) (.) ⎨ △ϕp(u′(ti)) = ci, i = ,,...,k, ⎩u() = , ϕp(u′(+∞)) =  +∞  e–tϕp(u′(t))dt, where < t< t< ··· < tk < +∞, p =  , f (t,x,y) = e–t (+t)   √ sinx + e–t √y + e–t. Corresponding to the problem (.), we have h(t) = e–t, Ii(u,v) = ci, i = ,,...,k. Take hr(t) = ((+ t)–  + r  + )e–t, a(t) = e–t (+t)  , b(t) = c(t) = e–t, di = gi = , ei = ci, i = ,,...,k, e= e(+tk)(+ k i=|ci|), Mr = max≤i≤k{|ci|}. where < t< t< ··· < tk < +∞, p =  , f (t,x,y) = e–t (+t)   √ sinx + e–t √y + e–t. (+t)  Corresponding to the problem (.), we have h(t) = e–t, Ii(u,v) = ci, i = ,,...,k. Take hr(t) = ((+ t)–  + r  + )e–t, a(t) = e–t (+t)  , b(t) = c(t) = e–t, di = gi = , ei = ci, i = ,,...,k, e e(+tk)(+ k |c |)M max {|c |} (+t) e= e(+tk)(+ k i=|ci|), Mr = max≤i≤k{|ci|}. (+t) e= e(+tk)(+ k i=|ci|), Mr = max≤i≤k{|ci|}. By a simple calculation, we find that (H)-(H) and (H)() hold. By Theorem ., we find that the problem (.) has at least one solution. Acknowledgements h k d Acknowledgements This work is supported by the Natural Science Foundation of China (11171088) and the Natural Science Foundation of Hebei Province (A2013208108). The author is grateful to anonymous referees for their constructive comments and suggestions, which led to improvement of the original manuscript. Received: 1 October 2014 Accepted: 28 January 2015 References 1. Agarwal, RP, O’Regan, D: Infinite Interval Problems for Differential, Difference and Integral Equations. Kluwer Academic, Dordrecht (2001) 2. Benchohra, M, Henderson, J, Ntouyas, S: Impulsive Equations and Inclusions. Contemporary Mathematics and Its Applications, vol. 2. Hindawi Publishing Corporation, New York (2006) 3. Lakshmikantham, V, Bainov, DD, Simeonov, PS: Theory of Impulsive Differential Equations. Series in Modern Applied Mathematics, vol. 6. World Scientific, Hackensack (1994) 4. Samoilenko, AM, Perestyuk, NA: Impulsive Differential Equations. World Scientific, Singapore (1995) 5. Zhang, X, Shuai, Z, Wang, K: Optimal impulsive harvesting policy for single population. Nonlinear Anal., Real World Appl. 4, 639-651 (2003) References Page 13 of 13 Jiang Boundary Value Problems ( 2015) 2015:39 6. Carter, TE: Necessary and sufficient conditions for optimal impulsive rendezvous with linear equations of motion. Dyn. Control 10, 219-227 (2000) 7. Chang, CC, Liu, CQ, Huston, RL: Dynamics of multibody systems subjected to impulsive constraints. Multibody Syst. Dyn. 8, 161-184 (2002) y , ( ) 8. Li, C, Liao, X, Yang, X: Impulsive stabilization and synchronization of a class of chaotic delay systems. Chaos 15, 043103 (2005) y 8. Li, C, Liao, X, Yang, X: Impulsive stabilization and synchronization of a class of chaotic delay systems. Chaos 15, 043103 (2005) 9. Shen, JH, Li, JL: Existence and global attractivity of positive periodic solutions for impulsive predator-prey model with dispersion and time delays. Nonlinear Anal., Real World Appl. 10, 227-243 (2009) 10. Agarwal, RP, Franco, D, O’Regan, D: Singular boundary value problems for first and second order impulsive differentia equations. Aequ. Math. 69, 83-96 (2005) q q 11. Tian, Y, Bai, Z: Existence results for the three-point impulsive boundary value problem invol equations. Comput. Math. Appl. 59, 2601-2609 (2010) 11. Tian, Y, Bai, Z: Existence results for the three-point imp equations. Comput. Math. Appl. 59, 2601-2609 (2010) 11. Tian, Y, Bai, Z: Existence results for the three-point imp tence results for the three-point impulsive boundary value problem 11. Tian, Y, Bai, Z: Existence results for the three-point impulsive boundary value problem involving fractional differential equations. Comput. Math. Appl. 59, 2601-2609 (2010) equations. Comput. Math. Appl. 59, 2601-2609 (2010) 12. Jankowski, T: Positive solutions for second order impulsive differential equations involving Stieltjes integral 12. Jankowski, T: Positive solutions for second order impulsive differential equations involving Stieltjes integral conditions Nonlinear Anal 74 3775-3785 (2011) 12. Jankowski, T: Positive solutions for second order im conditions. Nonlinear Anal. 74, 3775-3785 (2011) 13. Nieto, JJ, O’Regan, D: Variational approach to impulsive differential equations. Nonlinear Anal., Real World Appl. 10, 680-690 (2009) 14. Kaufmann, ER, Kosmatov, N, Raffoul, YN: A second-order boundary value problem with impulsive effects on an unbounded domain. Nonlinear Anal. 69, 2924-2929 (2008) 15. Yan, B: Boundary value problems on the half-line with impulsive and infinite delay. J. Math. Anal. Appl. 259, 94-114 (2001) 16. Kosmatov, N: Multi-point boundary value problems on an unbounded domain at resonance. Nonlinear Anal. 68, 2158-2171 (2008) 17. Zhang, X, Feng, M, Ge, W: Existence result of second-order differential equations with integral boundary conditions at resonance. J. References Math. Anal. Appl. 353, 311-319 (2009) 18. Liu, Y, Ge, W: Solvability of nonlocal boundary value problems for ordinary differential equations of higher order. Nonlinear Anal. 57, 435-458 (2004) 19. Du, Z, Lin, X, Ge, W: Some higher-order multi-point boundary value problem at resonance. J. Comput. Appl. Math. 177, 55-65 (2005) 20. Liu, B: Solvability of multi-point boundary value problem at resonance (II). Appl. Math. Comput. 136, 353-377 (2003) l b l f b d l bl h l d l bl 20. Liu, B: Solvability of multi-point boundary value problem at resonance (II). Appl. Math. Comput. 136, 353-377 (2003) 21. Jiang, W: Solvability of boundary value problem with p-Laplacian at resonance. Bound. Value Probl. 2014, 36 (2014). doi:10 1186/1687 2770 2014 36 Liu, B: Solvability of multi-point boundary value problem at resona 0. u, B: So ab ty o u t po t bou da y a ue p ob e at eso a ce ( ). pp . at . Co put. 36, 353 3 ( 003) 21. Jiang, W: Solvability of boundary value problem with p-Laplacian at resonance. Bound. Value Probl. 2014, 36 (2014). doi:10.1186/1687-2770-2014-36 22. Jiang, W: Solvability for p-Laplacian boundary value problem at resonance on the half-line. Bound. Value Probl. 2013, 207 (2013). doi:10.1186/1687-2770-2013-207 23. Ge, W, Ren, J: An extension of Mawhin’s continuation the 23. Ge, W, Ren, J: An extension of Mawhin’s continuation theorem and its application to boundary value problems with a p-Laplacian. Nonlinear Anal. TMA 58, 477-488 (2004) p-Laplacian. Nonlinear Anal. TMA 58, 477-488 (2004) p p 24. Agarwal, RP, Franco, D, O’Regan, D: Singular boundary value problems for first and second order impulsive differential equations. Aequ. Math. 69, 83-96 (2005)
https://openalex.org/W3172151863
https://papers.phmsociety.org/index.php/phme/article/download/1615/579
English
null
Towards a model-based condition assessment of complex marine machinery systems using systems engineering
Proceedings of ... European Conference of the Prognostics and Health Management Society.
2,016
cc-by
9,611
1. BACKGROUND In the present paper, a systems engineering methodology is presented for the analysis and condition assessment of complex marine machinery systems. Two important characteristics of these systems are (i) that they comprise of a multitude of subcomponents which influence the overall system condition/performance and (ii) the continuously varying operating and environmental conditions. The methodology presented herein is capable to evaluate the system level effects of component degradation and faulty states under realistic system operation. By virtue of this, it is employed along with sensor signal data for the identification of degraded states and the allocation of the problem to specific system components. The modelling platform used in this work is the DNVGL COSSMOS (Complex Ship System Modelling & Simulation). Within the maritime industry there is a strong trend of machinery components and systems becoming more integrated and thereby also more complex. Advanced control and monitoring systems are enablers for safer, greener and smarter maritime transport of goods and people but also add to the level of complexity with regards to design, commissioning and operation. The key drivers behind this trend are stricter legislation related to emissions and the tough competitive market driving performance/cost optimization. When different components and sub-systems from a multitude of suppliers are combined to fulfil one or more functions on board a ship, the need for increased multi- discipline knowledge and system understanding is obvious. It can be hard to determine and verify all possible failure modes and their effect, and thereby also the overall system reliability and safety. During operation, the increased level of integration and dependence of embedded control functions significantly escalates the level of information and possible failure scenarios that the operator needs to understand and handle. The fact that ship machinery systems are operated in a host of different working modes under continuously varying operating and environmental conditions further complicates the situation, and makes it hard to deal with process anomalies in an effective manner. Within this context, a systems engineering approach is required in order to achieve and document a safety level equivalent to the traditional and well proven designs. At the methodological level, an automated model-based sensitivity analysis is conducted with respect to a set of component degradation/failure modes. The latter is used along with a clustering algorithm for the precise allocation of the failure to specific components and system particulars. Alexandros S. Zymaris et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Towards a model-based condition assessment of complex marine machinery systems using systems engineering Alexandros S. Zymaris 1, Øystein Åsheim Alnes 2, Knut Erik Knutsen3 and Nikolaos M.P. Kakalis4 1,4DNV▪GL Maritime R&D, Piraeus, 18545, Greece Alexandros.Zymaris@dnvgl.com Nikolaos.Kakalis@dnvgl.com 1,4DNV▪GL Maritime R&D, Piraeus, 18545, Greece Alexandros.Zymaris@dnvgl.com Nikolaos.Kakalis@dnvgl.com 2,3DNV▪GL Strategic Research & Innovation, Høvik, 1363, Norway Oystein.Alnes@dnvgl.com Knut.Erik.Knutsen@dnvgl.com 2,3DNV▪GL Strategic Research & Innovation, Høvik, 1363, Norway Oystein.Alnes@dnvgl.com Knut.Erik.Knutsen@dnvgl.com EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Diagnostics and prognostics algorithms for implementing a condition monitoring system are based on one of two main approaches (G Manno, Knutsen, & Vartdal, 2014) 1999; Pedersen & Pedersen, 2012; Vrijdag, Stapersma, & Terwisga, 2009) process modelling approaches have been used for the analysis of marine propulsion systems. In (Dimopoulos, 2009; Dimopoulos, Kougioufas, & Frangopoulos, 2008; Tostevin & Nealy, 2003) the analysis and optimisation of generic marine energy production systems mostly focusing on steady-state synthesis, design and operation is presented. Moreover, in (Rüde, 2006) the cooling network of a conventional propulsion vessel with a four stroke Diesel engine has been modelled, the purpose of the analysis is the enhancement of safety and reliability studies. Diagnostics and prognostics algorithms for implementing a condition monitoring system are based on one of two main approaches (G Manno, Knutsen, & Vartdal, 2014) 1999; Pedersen & Pedersen, 2012; Vrijdag, Stapersma, & Terwisga, 2009) process modelling approaches have been used for the analysis of marine propulsion systems. In (Dimopoulos, 2009; Dimopoulos, Kougioufas, & Frangopoulos, 2008; Tostevin & Nealy, 2003) the analysis and optimisation of generic marine energy production systems mostly focusing on steady-state synthesis, design and operation is presented. Moreover, in (Rüde, 2006) the cooling network of a conventional propulsion vessel with a four stroke Diesel engine has been modelled, the purpose of the analysis is the enhancement of safety and reliability studies.  A model-based (first principles physical model) approach (Isermann, 2011)  A model-based (first principles physical model) approach (Isermann, 2011)  A model-based (first principles physical model) approach (Isermann, 2011)  A data-driven (statistical and data mining) approach (Baraldi, Maio, Genini, & Zio, 2015; Ge, Song, & Gao, 2013; Wang, 1999; Yin, Ding, Xie, & Luo, 2014) One major challenge for the maritime industry is the lack of reliable statistical failure data for ship equipment similar to the OREDA database for the offshore industry (SINTEF, 2002). Additionally, machinery systems are run very differently from ship type to ship type and also under continuously varying operating and environmental conditions. These facts strongly support the use of model– based approaches within shipping since it requires knowledge about the physics and function of the system and not large amounts of previous failure data. In (Grimmelius et al., 1999) a comparison between data-driven and first principle model-based CM approaches is presented for two marine machinery case studies. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 The systems studied are: (i) a Diesel engine, using only the torsional vibration of the crank shaft, and (ii) a compression refrigeration plant, using many different sensors. Both CM approaches (data-driven and model-based) show promising results. COSSMOS (Dimopoulos, Georgopoulou, Stefanatos, Zymaris, & Kakalis, 2014; Dimopoulos & Kakalis, 2010; Kakalis & Dimopoulos, 2012) is DNVGL’s model-based systems engineering simulation platform. What differentiates the aforementioned approaches from DNVGL COSSMOS is that each one of them is dedicated to a specific type of analysis for a specific system/component. On the other hand, COSSMOS is a generic marine machinery systems platform that is capable of performing steady-state and transient analysis as well as more complex tasks such as optimisation and parameter estimation for all machinery components that can be found on-board a vessel. Another megatrend within the maritime industry is the remarkable growth in broadband satellite installations on board large ships over the past few years. Components and systems can be integrated with sensors that measure, log and send ashore operational system data. Access to comprehensive operational data from on board system sensors that is transferred to shore on a live basis will lead to tremendous opportunities for all stakeholders within the maritime industry to make shipping smarter, safer and more cost-efficient. This revolution in ship connectivity will enable the implementation of a host of new applications such as performance and environmental monitoring, energy efficiency optimisation, remote control/autonomy and condition monitoring (CM). In (Monnin, Voisin, Leger, & Iung, 2011) and (Voisin, Medina-Oliva, Monnin, Leger, & Iung, 2013) a knowledge structuring scheme for achieving fleet-wide PHM utilizing both modelling and monitoring is proposed. Utilizing contextual information about the system and components by means of semantic modelling is identified as a key issue in order to allow for consideration of fleet component similarities and heterogeneities. Monitoring data from a given component are considered within their context and thereby enhancing the identification of the corresponding health condition. The fleet dimension can provide knowledge and data to improve both diagnostic and prognostic models. In the maritime sector a fleet could both be a ship owner fleet of typically 10 to a few hundred vessels or the entire fleet of a classification society such as DNVGL with a fleet of more than 13,000 vessels. Moreover, in (Grimmelius et al., 1999) the aspect of sensor redundancy using virtual sensing is investigated. 1. BACKGROUND The selected case-study is the Diesel-electric marine propulsion system of a 2300 tonnes DWT (deadweight tonnage) anchor handling vessel embedded with its cooling network. Based on the results, the approach is capable to successfully identify faults at various subcomponents of the cooling network system including pumps, regulating valves, heat exchangers and piping. Due to the fact that the system is treated in an integrated manner, a fault can be identified in a component using sensor signals placed in other system locations. One of the first studies on marine energy conversion systems dealing with the computer-aided design of the overall system is (Ito & Akagi, 1986). Later, in (Campora & Figari, 2003; Hansen, Adnanes, & Fossen, 2001; Jefferson, Zhou, & Hindmarch, 2003 ; Kyrtatos & Lambropoulos, 2000; Kyrtatos, Theodossopoulos, Theotokatos, & Xiros, EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 The approach is divided into two steps: (i) the binominal classification between failed and non-failed states and (ii) the specific failure identification on a system, subcomponent/location. The binominal classification procedure aims to compare system measurements with model results and class the system state between two different possibilities: failed or non-failed. If failed, the next step is to allocate the exact position of the failure. This is achieved by performing a holistic sensitivity analysis using the simulation model. The product of the sensitivity analysis (sensitivity maps) provides a quantitative behavioural analysis of the sensor variables with respect to an extensive envelope of system faults. Based on that, the failure pattern from the sensor readings can be matched to the pattern(s) of specific faults. In this work, an unsupervised type of neural network, the self-organising maps (SOM) (Kohonen, 2001) has been used for that purpose. The SOM task is to cluster the unknown failure pattern within clusters of failures that each one of them corresponds to a set of system failure modes (given by the sensitivity analysis). Although in literature a variety of neural networks have been used for condition monitoring and recognition of faults (Baraldi, Compare, Sauco, & Zio, 2013; Grimmelius et al., 1999; Schenk, Natale, Germond, Boss, & Lam, 2002), what differentiates this approach is that it utilises sensor signals along with the first principle sensitivity analysis results. Steiner, 2013). Although Diesel engine CM is a well investigated subject, there are less studies found for the auxiliary cooling network system or the Diesel engines and the other essential propulsion system equipment (such as frequency converters and electrical transformers in Diesel electric systems). One of the few works is (Twiddle & Jones, 2002) where a model-based CM approach for the cooling network of an on-shore Diesel generator has been developed. Steiner, 2013). Although Diesel engine CM is a well investigated subject, there are less studies found for the auxiliary cooling network system or the Diesel engines and the other essential propulsion system equipment (such as frequency converters and electrical transformers in Diesel electric systems). One of the few works is (Twiddle & Jones, 2002) where a model-based CM approach for the cooling network of an on-shore Diesel generator has been developed. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Model- based technologies are a key enabler for virtual sensing / analytical redundancy (Chen & Patton, 1999). In (Blanke, 2001) a model-based methodology for the diagnosis of faults in ship propulsion systems and fault-tolerant control is presented. In this work mathematical models for ship speed, propeller and prime mover (i.e. Diesel engine) have been utilised. Diesel engines are among the most expensive, critical and maintenance intensive components on board the ship, and as a consequence many first principle condition monitoring approaches have been proposed in literature (Kouremenos & Hountalas, 1997; Watzenig, Sommer, & CM is the process of monitoring a set of parameters of condition in a system in order to identify a significant change that is indicative of a failure or a developing fault that is affecting it’s function. Within the maritime industry the use of condition monitoring techniques to perform, for instance, condition based maintenance (CBM) on machinery components and systems is still very much in an infant stage. There are some early adopters within segments with high value vessels such as cruise and offshore but for most ships the on board maintenance tasks are carried out based on basis of either calendar or running hours (Coull, 2015). The need to stay competitive in a tough market is however pushing more and more ship owners and managers towards improving maintenance procedures, boosting uptime and cutting costs. The uptake of CBM is therefore expected to accelerate over the coming years. 2 3. SYSTEM UNDER STUDY The chosen case-study for this work is a 2300 DWT anchor handling vessel built in 2009 and operated in the North Sea. The vessel has a twin screw conventional propulsion plant consisting of: Figure 1. DNVGL Digital twin, MODAM project (DNVGL, 2015; G. Manno, et, & Al., 2015).  Two 4.5 MW 4 stroke propulsion Diesel engines EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 During the recent years, DNVGL has initiated a number of projects in this area: COMPASS (Condition and operation monitoring for performance and ship safety), MODAM (Model based, data driven asset management) (DNVGL, 2015). “Nauticus Twinity” (Ludvigsen, Jamt, Husteli, & Smogeli, 2016) is a first DNVGL digital twin prototype (Glaessgen & Stargel, 2012; Tuegel, Ingraffea, Eason, & Spottswood, 2011). A digital twin, Figure 1, is a model of a physical asset, that encapsulates (i) an information/data model, (ii) a simulation model (e.g. COSSMOS) and model- based & data driven analytics, and (iii) possibly a dependability model and (iv) possibly a visualisation model. What makes a digital twin different from generic models is that they are specific to their physical counterparts. The digital twin model is specifically instantiated for the specific asset. Furthermore, it will follow its corresponding real life twin through its life cycle, through collecting sensor updates and history data. Any change on the physical asset must also be reflected in the model. As such, the digital twin is envisaged as a holistic simulation and analytics platform for performance, energy efficiency optimisation, classification services and condition monitoring. In the rest of the paper, the description of the under study system is presented in section ‎3. The modelling methodology is presented in section ‎4, and the detailed discussion of the binominal classification and the failure identification is given in sections ‎5 and ‎6. Two different case studies of failures at the cooling network are presented in section ‎7: (i) a fault (infraction) at the 3-way valve TCV65LT and (ii) a fault at the LT pump at the main engine (ME) cooling circuit. Figure 1. DNVGL Digital twin, MODAM project (DNVGL, 2015; G. Manno, et, & Al., 2015). 2. OUTLINE OF THE PROPOSED APPROACH  Two reduction gears each with a 3 MW Power Take-In (PTI) electric booster motor The ultimate goal of the work presented in this paper is to have a continuous health state monitoring of the subject cooling system that can support classification and predictive maintenance activities and enhance ship safety and reliability. A condition monitoring approach using first principle model-based analytics for the cooling network system is proposed. The digital twin of the propulsion system and its cooling network is developed in COSSMOS. Apart from the simulation model, a set of available sensor measurements have to be identified.  Two controllable pitch propellers with rudders Vessel maneuverability and station-keeping capabilities are further enhanced by 4 separate thruster units:  Two stern tunnel thrusters  One bow tunnel thruster  One bow retractable azimuth thruster 3 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Illustration of Main Engine LT and HT sub- systems. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 1st stage charge air cooling and cylinder block cooling. The system has a 3-way temperature controlled valve recirculating cooling water inside the HT system. The PID controller regulates the cooling water temperature coming out of the cylinder block to 90°C. Power generation is achieved by two 5 MW shaft generators installed on front-end of the main engines and 4 equally sized 2.1 MW auxiliary Diesel generator sets. The propulsion and power system is designed to fulfill Class redundancy requirements but is also configured to provide the operator with large operational flexibilities. The cooling system is equipped with a limited number of sensors monitoring system temperatures, pressures, cooling water levels etc. Critical functions are protected by alarms. The complete FW system no.2 contains: The cooling of the vessel’s main machinery equipment is divided into two separate freshwater (FW) cooling systems, FW cooling system No. 1 & No.2. Each FW cooling system is arranged for one main engine with shaft generator, reduction gear and PTI, two auxiliary generator sets and two thruster units. To the authors’ knowledge this is the first time the entire cooling network of a Diesel electric system is modelled and assessed for condition monitoring purposes. The FW cooling system includes one main electrical circulation pump, engine driven pumps (for high and low temp systems), engine preheaters, valves and coolers. One 3-way regulating valve is arranged for each FW system with a set-point of 37°C. For each Diesel engine two additional 3-way regulating valves are installed. Crossover lines and one standby pump are arranged between FW Cooling Systems No.1 & No.2 with manual normally closed valves.  25 different plate coolers,  70-80 valves that can adjust cooling water flow (manually and automatically),  70-80 valves that can adjust cooling water flow (manually and automatically),  Approximately 200 meters of cooling water pipe of varying diameter.  Approximately 200 meters of cooling water pipe of varying diameter. Figure 3. Illustration of Main Engine LT and HT sub- systems. The system under study is FW Cooling System No.2 (starboard side) as illustrated in Figure 2. Figure 2. Illustration of FW Cooling System no.2. Propulsion, power and maneuvering system deposits heat to the FW system, which again deposits to a sea water (SW) system via 2 main coolers. The 3-way temperature controlled mixing valve (TCV65LT) maintains a given FW temperature of 37°C by mixing cold and warm water. Figure 3. 4. MODELLING APPROACH – DNVGL COSSMOS Since 2008 DNVGL has introduced MBSE (Model Based Systems Engineering) for modelling, simulation and optimisation of integrated marine energy systems (Dimopoulos et al., 2014; Dimopoulos & Kakalis, 2010; Kakalis & Dimopoulos, 2012). This work has resulted in the modelling framework named as DNVGL COSSMOS. DNVGL COSSMOS (Figure 5) aims at being one tool providing model-based decision support on: Since 2008 DNVGL has introduced MBSE (Model Based Systems Engineering) for modelling, simulation and optimisation of integrated marine energy systems (Dimopoulos et al., 2014; Dimopoulos & Kakalis, 2010; Kakalis & Dimopoulos, 2012). This work has resulted in the modelling framework named as DNVGL COSSMOS. DNVGL COSSMOS (Figure 5) aims at being one tool providing model-based decision support on: Figure 2. Illustration of FW Cooling System no.2. Propulsion, power and maneuvering system deposits heat to the FW system, which again deposits to a sea water (SW) system via 2 main coolers. The 3-way temperature controlled mixing valve (TCV65LT) maintains a given FW temperature of 37°C by mixing cold and warm water.  Optimal design of on-board machinery with respect to energy efficiency, safety and cost effectiveness;  Performance evaluation, diagnostics and operation optimisation under real-service conditions for the entire mission envelope of the machinery system; and The Main Engine cooling system is divided into two sub- systems: a low temperature (LT) system and a high temperature (HT) system. Both sub-systems are equipped with a separate circulation pump. The LT system is used to provide fuel and lubrication oil cooling/heating and 2nd stage charge air cooling. The amount of cooling water flowing through the 2nd stage charge air cooler is regulated by a 3-way flow control valve. The valve PID controller is used to give a charge air temperature of 55°C. The HT system utilizes output cooling water from the LT system for  Assessment of the potential, operational capabilities, and safety of new technologies. As such, the COSSMOS modelling framework provides the user with the capability to define and analyse a wide range of system configurations, model their behaviour in terms of their physics processes 4 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Figure 4. Propulsion system model in DNVGL COSSMOS. (Abbr.: ME=Main engine, SG=Shaft Generator, DG=Diesel Generator, VFD= Variable Frequency drive). (mechanical, electric, thermodynamic, heat transfer, fluid flow, etc.), and perform any model-based application of interest (simulation, optimisation, control, parameter estimation) under both steady-state (design/off-design conditions) and dynamic (time varying, transient operation) conditions. The implementation of COSSMOS is done in gPROMS (PSE, 2016), an equation-oriented modelling language specially designed for process system modelling and simulation that can efficiently handle the numerical solution of highly complex non-linear PDAE (Partial Differential Algebraic Equation) systems in a variety of processes. The COSSMOS component models can be a phenomenological model based on first principles, detailed physical models or even correlations from experimental data. A general description of the component model mathematical equations reads: Figure 4. Propulsion system model in DNVGL COSSMOS. (Abbr.: ME=Main engine, SG=Shaft Generator, DG=Diesel Generator, VFD= Variable Frequency drive). 𝑑𝑌⃗ 𝑑𝑡= 𝐹 (𝑌⃗ (𝑡), 𝑑𝑌⃗ 𝑑𝑥𝑖 (𝑡), 𝑢⃗ (𝑡), 𝑏⃗ (𝑡), 𝑡) (1) 𝐻⃗⃗ (𝑌⃗ (𝑡), 𝑑𝑌⃗ 𝑑𝑥𝑖 (𝑡), 𝑢⃗ (𝑡), 𝑏⃗ (𝑡), 𝑡) = 0 (2) (1) As in any model-based systems engineering approach, the overall system model is synthesized from coupling the individual component models as shown in Figure 5. In this work the vessel’s propulsion system has been modelled; the model view from the COSSMOS GUI is presented in Figure 4. A schematic representation of the propulsion system with the cooling network is given in Figure 2. Moreover, the overall COSSMOS model: propulsion system and cooling network (starboard side) can be seen in Figure 6. The model comprises of approx. 14300 PDAE equations and 295 system components. 4.1. Model Calibration The simulation model has been calibrated based on manufacturers’ data. The comparison of results between the system operation at design conditions (black bars) and the model predictions (grey bars) can be found in Figure 7. These are indicative results and if the model is to be used in a real CM application a more exhaustive calibration procedure (through sensor data) has to be carried out. However, the simulation model is adequate for demonstrating the CM approaches presented in the current work. Let’s assume the following fault detection problem: given the set of signal data in Table 1, the engineer/analyst is asked to identify if this set of signals is corresponding to a degraded system condition (i.e. operation under the presence of fault or failure in the system). Based on the schematic of Figure 8: by using (i) a set of inputs (system operational data & sensor measurements that we have higher confidence in) and (ii) information about the condition of specific system parts (if a priori knowledge is available, e.g. amount of fouling at heat exchangers (HEXs), COSSMOS simulations can be carried out and the results can be used for benchmarking against the values of Figure 7. Comparison between system data (black bars) and model predictions (grey bars). The system positions corresponding to the quantities depicted in the x-axis can be found in Appendix I. Table 1 (signal measurements, column headers with grey background). In this example the inputs are the independent parameter values ( Table 1, white background), the benchmarking results are depicted in Figure 9 and Figure 10. From the differences between these one could assume that there is some kind of fault/degraded state in the LT circuit that affects (among other system variables) the TCV65LT stem position and the pump power. Figure 7. Comparison between system data (black bars) and model predictions (grey bars). The system positions corresponding to the quantities depicted in the x-axis can be found in Appendix I. Figure 7. Comparison between system data (black bars) and model predictions (grey bars). The system positions corresponding to the quantities depicted in the x-axis can be found in Appendix I. In the discussion above the simulation model has been used for analysing the current system state by comparing with the “ideal” one (including HEX fouling) under the specific operating conditions. (2) (2) where 𝑡 is the time, 𝑌⃗ = (𝑌1, … , 𝑌𝑁𝑌), 𝑢⃗ = (𝑢1, … , 𝑢𝑁𝑢) and 𝑏⃗ = (𝑏1, … , 𝑏𝑁𝑏) are the vectors of differential variables, algebraic variables and parameters, respectively. The vectors 𝑌⃗ and 𝑢⃗ constitute the process variables of the system. 𝐹 and 𝐻⃗⃗ are vector functions. The partial derivative base vector 𝑥 , is an appropriate distribution domain, usually expressing geometry dimensions (e.g. length, width, radius, etc.). The PDAE system is completed by the necessary initial and boundary conditions. Figure 5. The DNVGL COSSMOS modelling framework. Figure 5. The DNVGL COSSMOS modelling framework. 5 Figure 6. Overall COSSMOS model: propulsion system & cooling network (vessel’s starboard side). Figure 6. Overall COSSMOS model: propulsion system & cooling network (vessel’s starboard side). EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 historical data exist). Moreover, it is also useful for systems with many independent parameters that influence the operation. In these cases there is a challenge in obtaining real-life data for all possible states (combinations of the independent parameters) that may be needed in order to create a good system representation. Critical and rare events can only be included in the database by using model-based simulations as it is done in this project. Consequently the use of simulations models is particularly important for filling these “gaps” of information in the real-life data. Figure 9. Comparison of TCV65LT stem position. Dark grey bars: values corresponding to a system without any fault. Light grey bars: set of data to be identified if correspond to a condition or not. Figure 8. Methodology schematic. From left to right: Inputs to the COSSMOS model, COSSMOS simulation model, result comparison and benchmarking with sensor signals. Figure 10. Comparison of LT pump loads (as in Figure 9). Figure 8. Methodology schematic. From left to right: Inputs to the COSSMOS model, COSSMOS simulation model, result comparison and benchmarking with sensor signals. Figure 10. Comparison of LT pump loads (as in Figure 9). Table 1. Propulsion system and cooling network TCV65LT valve and LT and pump signal data; Anchor handling (AH) mode. Input variables on left side (white background) and simulated variables on right side (grey background). Table 1. Propulsion system and cooling network TCV65LT valve and LT and pump signal data; Anchor handling (AH) mode. Input variables on left side (white background) and simulated variables on right side (grey background). Input variables Output variables Load ME 2 Load Azimuth Thruster [kW] Load Tunnel Thruster3 Aft2 [kW] Load PTI [kW] El. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 demand STBD side (other) [A] Central Cooler fouling SW temperature TCV65LT stem [0-1] Pump LT Power [kW] FW Temp ME inlet [C] #1 20% 360 960 600 1000 No 15 0.13 37.58 88.1 #2 30% 360 960 600 1486 No 15 0.14 37.60 87.2 #3 20% 360 480 300 1486 Yes 32 0.72 37.44 88.1 #4 30% 360 480 300 1000 No 32 0.82 37.07 87.2 #5 20% 180 960 600 1000 Yes 32 0.69 37.50 88.1 #6 30% 180 960 300 1486 Yes 32 0.78 37.31 87.2 #7 20% 180 480 600 1486 No 15 0.12 37.57 88.1 #8 30% 180 480 600 1000 Yes 15 0.10 37.55 87.2 Table 1. Propulsion system and cooling network TCV65LT valve and LT and pump signal data; Anchor handling (AH) mode. Input variables on left side (white background) and simulated variables on right side (grey background). m and cooling network TCV65LT valve and LT and pump signal data; Anchor handling (AH) left side (white background) and simulated variables on right side (grey background).  Step 1: Perform Binominal classification analysis (as in paragraph 6). If it is identified that there is a hidden fault somewhere in the system, then:  Step 1: Perform Binominal classification analysis (as in paragraph 6). If it is identified that there is a hidden fault somewhere in the system, then: 4.1. Model Calibration This is particularly useful when the system under study is at its infant life stage (where limited 1 If clustering has been applied before, then correlate to an agglomeration of matrix rows (that can be seen as very similar failures). 6.1. Sensitivity analysis map and sensitivity derivatives Sensitivity derivative signs: 𝒔𝒊𝒈𝒏[ 𝜹(𝑴𝑽) 𝜹𝒙𝒋| 𝜹(𝑴𝑽) 𝜹𝒚𝒍| 𝜹(𝑴𝑽) 𝜹𝒖𝒎] 𝑻 , white: -1, black:+1. 1 2 3 4 5 6 7 8 9 10 11 50 100 150 200 250 300 350 MV 𝑥𝑗, 𝑦𝑙, 𝑢𝑚 1 2 3 4 5 6 7 8 9 10 11 50 100 150 200 250 300 350 MV 𝑥𝑗, 𝑦𝑙, 𝑢𝑚 MV In Figure 11, the sensitivity map for the cooling network is presented. The values are in log10 scale. The chosen MV variables are found in Table 2 (others could have been chosen as well). Figure 12. Sensitivity map: cooling network, AH mode. 𝑻 Sensitivity derivative signs: 𝒔𝒊𝒈𝒏[ 𝜹(𝑴𝑽) 𝜹𝒙𝒋| 𝜹(𝑴𝑽) 𝜹𝒚𝒍| 𝜹(𝑴𝑽) 𝜹𝒖𝒎] 𝑻 , white: -1, black:+1. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 𝑥𝑗, 𝑦𝑙, 𝑢𝑚 Table 2. MV for the cooling network case. 1 Central cooler FW temperature OUT 2 DG3 cooler FW temperature before cooler 3 DG4 cooler FW temperature before cooler 4 TCV65HT stem position 5 DG3 cooler FW temperature after cooler 6 DG4 cooler FW temperature after cooler 7 ME pump_30HTpower 8 ME pump_30LTpower 9 LT circulation pump power 10 DG3 FW pump pressure 11 DG4 FW pump pressure Figure 11. Sensitivity map: cooling network, AH mode. Values are in log10 scale: 𝒍𝒐𝒈𝟏𝟎(𝒂𝒃𝒔[ 𝜹(𝑴𝑽) 𝜹𝒙𝒋| 𝜹(𝑴𝑽) 𝜹𝒚𝒍| 𝜹(𝑴𝑽) 𝜹𝒖𝒎] 𝑻 ) . Figure 12. Sensitivity map: cooling network, AH mode. Sensitivity derivative signs: 𝒔𝒊𝒈𝒏[ 𝜹(𝑴𝑽) 𝜹𝒙𝒋| 𝜹(𝑴𝑽) 𝜹𝒚𝒍| 𝜹(𝑴𝑽) 𝜹𝒖𝒎] 𝑻 , white: -1, black:+1. 1 2 3 4 5 6 7 8 9 10 11 50 100 150 200 250 300 350 -10 -5 0 5 1 2 3 4 5 6 7 8 9 10 11 50 100 150 200 250 300 350 MV MV 𝑥𝑗, 𝑦𝑙, 𝑢𝑚 Table 2. MV for the cooling network case. 1 Central cooler FW temperature OUT 2 DG3 cooler FW temperature before cooler 3 DG4 cooler FW temperature before cooler 4 TCV65HT stem position 5 DG3 cooler FW temperature after cooler 6 DG4 cooler FW temperature after cooler 7 ME pump_30HTpower 8 ME pump_30LTpower 9 LT circulation pump power 10 DG3 FW pump pressure 11 DG4 FW pump pressure Table 2. MV for the cooling network case. 1 Central cooler FW temperature OUT 2 DG3 cooler FW temperature before cooler 3 DG4 cooler FW temperature before cooler 4 TCV65HT stem position 5 DG3 cooler FW temperature after cooler 6 DG4 cooler FW temperature after cooler 7 ME pump_30HTpower 8 ME pump_30LTpower 9 LT circulation pump power 10 DG3 FW pump pressure 11 DG4 FW pump pressure 6. FAILURE IDENTIFICATION 6. FAILURE IDENTIFICATION Herein, the scope is to address the problem of locating the internal failure at the corresponding part of the system, extending the methodology described in section ‎5. The system’s sensitivity map can be utilised for this purpose. DNVGL COSSMOS has the ability to perform automated sensitivity analysis using symbolic differentiation (Naumann, 2012). The step below are followed: Herein, the scope is to address the problem of locating the internal failure at the corresponding part of the system, extending the methodology described in section ‎5. The system’s sensitivity map can be utilised for this purpose. DNVGL COSSMOS has the ability to perform automated sensitivity analysis using symbolic differentiation (Naumann, 2012). The step below are followed: y  Step 2: Perform sensitivity analysis using the systems engineering model.  Step 3: Cluster sensitivity analysis results into groups of faults.  Step 4: Identify the group of faults that the specific fault pattern belongs to. 7 7 6.1. Sensitivity analysis map and sensitivity derivatives In the sensitivity map, the sensitivity derivatives of selected system variables with respect to the model input parameters and independent variables are depicted. For condition monitoring, the selected system variables make sense to correspond to sensor measured variables (MV). In more detail, a non-linear differential and algebraic system of equations (like the COSSMOS model) that describes the dynamic behaviour of an engineering system can be written in the form: 𝑅𝑖(𝑥𝑗, 𝑑𝑥𝑗 𝑑𝑡, 𝑦𝑙, 𝑢𝑚) = 0 (3) (3) 𝑥𝑗, 𝑦𝑙, 𝑢𝑚 1 2 3 4 5 6 7 8 9 10 11 50 100 150 200 250 300 350 -10 -5 0 5 MV where 𝑥 (𝑡) and 𝑦 (𝑡) are the vectors of differential and algebraic variables respectively (both are unknowns to be determined by the simulation) and 𝑢⃗ (𝑡) is the vector of the mathematical system input parameters that is a given function of time. The sensitivity derivative of a variable MV is a vector given by: [𝛿(𝑀𝑉) 𝛿𝑥𝑗 | 𝛿(𝑀𝑉) 𝛿𝑦𝑙 | 𝛿(𝑀𝑉) 𝛿𝑢𝑚 ] 𝑇 (4) (4) The overall matrix comprises of 𝑀𝑆𝑉 columns, where 𝑀𝑆𝑉 is the number of the monitored variables. Each column vector element, shows how much the MV is affected by a perturbation in the corresponding system variable (x,y) or parameter (u). The system variables (x,y) and parameters (u) are related to a certain sub-component function and characteristics, and thus any possible failure mode of the sub-component will alter the their values by 𝛿𝑥 leading to a change in the MV. The sensitivity maps provide a holistic view of the internal system relations between MV and all the system variables and can be used to correlate the fault pattern (“fingerprint”) to a specific matrix row1 (i.e. variable (x,y) or parameter (u)); and consequently to a specific system component and failure mode. The advantage of using the sensitivity analysis approach is that it requires only one simulation of the system irrespective of the number of the system variables (x,y) or parameters (u) (and consequently the number of failures). This reduces computational requirements significantly. MV MV Figure 11. Sensitivity map: cooling network, AH mode. Values are in log10 scale: 𝒍𝒐𝒈𝟏𝟎(𝒂𝒃𝒔[ 𝜹(𝑴𝑽) 𝜹𝒙𝒋| 𝜹(𝑴𝑽) 𝜹𝒚𝒍| 𝜹(𝑴𝑽) 𝜹𝒖𝒎] 𝑻 Values are in log10 scale: 𝒍𝒐𝒈𝟏𝟎(𝒂𝒃𝒔[ 𝜹(𝑴𝑽) 𝜹𝒙𝒋| 𝜹(𝑴𝑽) 𝜹𝒚𝒍| 𝜹(𝑴𝑽) 𝜹𝒖𝒎] 𝑻 ) . Figure 12. Sensitivity map: cooling network, AH mode. 7.1. 3-way regulation valve fault case study. Table 1) readings (simulation with fault) the Binominal classification analysis (Step 1, section ‎5) has been carried out. The differences between the signals that correspond to the fault condition and the ones that correspond to the “non- fault” condition are depicted in Figure 14. These data are introduced to the clustering algorithm in order to be allocated in the best matching unit of the map in Figure 13. For this case, the best matching unit is the node (5,7) shown by the arrow in Figure 13 (left). Based on sensitivity matrix data, the components related to the faults/parameters that have been previously matched to node (5,7) are depicted in the Figure 15. The TCV65LT valve is one of them and consequently the approach has correctly identified the failed component between all the different (but influencing the sensor signal set) areas of the system (Figure 16). The Self-Organising Maps (SOM) (Kohonen, 2001), an unsupervised machine learning clustering method, has been utilised aiming to aggregate together system parameters/faults (i.e. the rows of sensitivity matrix), that have the same impact to the sensor readings. The SOMPY Python Library for Self Organizing Maps has been used for that purpose (Moosavi, 2016). The data-set used to build the SOM was the sensitivity matrix (depicted in Figure 11). By using a 2-dimensional representation plane with a square topology (21x21 in size), the agglomerates of the parameters/faults can be seen in Figure 13 (left). Each cycle corresponds to the node of the SOM that has been best matched to a set/agglomerate of parameters/faults – the larger the cycle radius the more the system parameters/faults are associated with the corresponding node. On the right hand side of Figure 13 the U-Matrix 2 , that gives insight into the local distance structures of the data set, is presented. The red areas represent divisions of the data set, while the dots are the nodes presented also in the left graph. In order to assess the SOM performance, as set of #40 different simulation cases has been used to create the true/false positive and true/false negative matrix of Table 3. The data set comprises of #20 different TCV65LT fault realisations (various infraction percentages, different SW and engine load conditions) and #20 system realisations without any fault at the TCV65LT valve. 7.1. 3-way regulation valve fault case study. The latter includes the case where every component is working as new and other cases with faults/failures at components other than the TCV65LT valve. From the #20 system realisations that the TCV65LT valve has failed or degraded, #16 have been correctly identified, i.e. the sensor set signal was correctly matched to SOM node (5,7). For the rest #4 cases, the model prediction has allocated as best matching units the nodes (6,6) (three times) and (7,6) (one time). Although counted for false negatives, it is worth mentioning that the latter two nodes are at the vicinity of the correct node choice that is again node (5,7). In Figure 17, the orange-coloured highlighted area includes nodes that corresponding to the TCV65LT infraction fault, namely node (5,7) (16 true positive results of Table 3) and nodes (6,6) & (7,6) (four false negative cases of Table 3). In the same figure (Figure 17), the blue-coloured square areas correspond to the 20 true negative cases of Table 3. In Figure 21, the heat-maps of the sensor variables are depicted over the SOM. These heat-maps present the distribution of each sensor variable across the SOM node structure (21x21). By comparing them, interdependencies between sensors and sensor redundancies with respect to failures can be identified. For example, sensor variable 1 (LT FW temperature out of the central cooler) can differentiate mainly between parameters/failures that are located to the nodes of the upper right (blue, Figure 21) and lower-middle (red, Figure 21) part of the map. Figure 13. 3-way regulation valve fault case study: 2- dimensional representation plane (21x21) with the nodes (left). Corresponding the U-Matrix (right). Table 3. 3-way regulation valve case study: matching matrix that presents true/false positives and true/false negatives. Predicted Failed Non-Failed Actual Class Failed 16 4 (*) Non-Failed 0 20 Table 3. 3-way regulation valve case study: matching matrix that presents true/false positives and true/false negatives. Table 3. 3-way regulation valve case study: matching matrix that presents true/false positives and true/false negatives. Figure 13. 3-way regulation valve fault case study: 2- dimensional representation plane (21x21) with the nodes (left). Corresponding the U-Matrix (right). From the maintenance log of the vessel’s cooling water system, it is known that there has been a fault at the 3-way valve TCV65LT, Figure 2. Unfortunately no real data were available for the time period of the fault. 7. CASE STUDIES Two different cases of failures at the cooling network are studied. Case study 1 concerns a fault (infraction) at the 3- way valve TCV65LT, Figure 2 has been studied. In the second case study, a fault at the LT pump at the ME cooling circuit has been investigated (Figure 3). 8 2 https://en.wikipedia.org/wiki/U-matrix 7.1. 3-way regulation valve fault case study. Orange-coloured nodes correspond to the TCV65LT infraction fault, namely node (5,7) (16 true positive results of Table 3). Blue-coloured square areas correspond to the 20 true negative cases of Table 3. 7.1. 3-way regulation valve fault case study. The simulation model has been used in order to simulate an infraction at the TCV65LT valve. Based on the sensor ( 9 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Sensor variables & locations:  T LT-FW @Main Cooler outlet  T LT-FW @Cooler Gen-Set 3 outlet  T LT-FW @Cooler Gen-Set 4 outlet  T 3WV @ME, Gen-sets (x3)  Pump loads (x 5) Figure 14. 3-way regulation valve fault case study; top: Sensor variables & locations. Bottom: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. Sensor variables & locations:  T LT-FW @Main Cooler outlet  T LT-FW @Cooler Gen-Set 3 outlet  T LT-FW @Cooler Gen-Set 4 outlet  T 3WV @ME, Gen-sets (x3)  Pump loads (x 5) Figure 14. 3-way regulation valve fault case study; top: Sensor variables & locations. Bottom: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. between all the different areas of the system which influence the sensors signal set values. Figure 17. 3-way regulation valve case study: 2-dimensional representation plane with the SOM nodes. Orange-coloured nodes correspond to the TCV65LT infraction fault, namely node (5,7) (16 true positive results of Table 3) and nodes (6,6) & (7,6) (four false negative cases of Table 3). Blue-coloured square areas correspond to the 20 true negative cases of Table 3. Figure 14. 3-way regulation valve fault case study; top: Sensor variables & locations. Bottom: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. Figure 14. 3-way regulation valve fault case study; top: Sensor variables & locations. Bottom: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. Figure 14. 3-way regulation valve fault case study; top: Sensor variables & locations. Bottom: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. Figure 15. 3-way regulation valve fault case study: components related to the faults/parameters that have been previously matched to the node (5,7). Figure 17. 3-way regulation valve case study: 2-dimensional representation plane with the SOM nodes. 7.2. LT pump fault case study. Figure 15. 3-way regulation valve fault case study: components related to the faults/parameters that have been previously matched to the node (5,7). Like in the 3-way regulation valve fault case study, a SOM was built based on the sensitivity matrix depicted in Figure 11. By using the sensor set variables of Figure 18 (simulation with fault), the clustering model has been used to identify a fault at the LT pump at the ME (Figure 19). The fault has been successfully identified since it was best fitted to the node (5,16). In this node the failure of the LT ME pump has been allocated during the initial training of the model. In this case a smaller sensor set of six variables, Figure 18. Figure 16. 3-way regulation valve fault case study: the approach has identified correctly the failed component There are some interesting aspects in this case: (i) there is not any signal measurement placed on the pump where the fault has been identified and (ii) there is not any measurement directly at the LT ME circuit (apart from the temperature value at the exit of the central cooler). By using signal values at other positions the algorithm has managed to correlate them with this specific fault. Although a simple example, it indicates the potential of using simulation models with real signals so as to reason about faults in parts of the system where no real signals are available. Figure 16. 3-way regulation valve fault case study: the approach has identified correctly the failed component 10 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Sensor specific faults/failures could also be identified using the proposed methodology. Similarly to other system component faults/failures, when sensors faults/failures are under the CM scope, one could again utilise the fact that the pattern of a specific sensor fault may differ from other system or sensor fault patterns. To identify these patterns the sensitivity analysis and SOM approach described in this paper could be used. Figure 19. LT pump fault case study: detail of the ME circuit (ref. Figure 6). A fault at the LT ME pump is indicated. In Figure 22, the heat-maps of the sensor variables are presented. These heat-maps show the distribution of each sensor variable across the SOM node structure (25x25). By comparing them, interdependencies between sensors, sensor redundancies with respect to failures and lack of sufficient sensors to monitor the entire set of possible system failure modes can be identified in a systematic way. Figure 19. LT pump fault case study: detail of the ME circuit (ref. Figure 6). A fault at the LT ME pump is indicated. 11 modes can be identified in a systematic way. Sensor variables & locations:  T LT-FW @Main Cooler outlet  T LT-FW @Cooler Gen-Set 3 outlet  T LT-FW @Cooler Gen-Set 4 outlet  T 3WV @ME, Gen-sets (x3) Figure 18. LT pump fault case study; left: Sensor variables & locations. Right: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. circuit (ref. Figure 6). A fault at the LT ME pump is indicated. Figure 20. LT pump fault case study: 2-dimensional representation plane (25x25) with the SOM nodes (left). Corresponding the U-Matrix (right). Figure 21. 3-way regulation valve fault case study: heatmaps of the sensor variables (Figure 14). Sensor variables & locations:  T LT-FW @Main Cooler outlet  T LT-FW @Cooler Gen-Set 3 outlet  T LT-FW @Cooler Gen-Set 4 outlet  T 3WV @ME, Gen-sets (x3) Figure 18. LT pump fault case study; left: Sensor variables & locations. Right: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. Sensor variables & locations:  T LT-FW @Main Cooler outlet  T LT-FW @Cooler Gen-Set 3 outlet  T LT-FW @Cooler Gen-Set 4 outlet  T 3WV @ME, Gen-sets (x3) Figure 18. LT pump fault case study; left: Sensor variables & locations. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Right: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. indicated. Figure 20. LT pump fault case study: 2-dimensional representation plane (25x25) with the SOM nodes (left). Corresponding the U-Matrix (right). Sensor variables & locations:  T LT-FW @Main Cooler outlet  T LT-FW @Cooler Gen-Set 3 outlet  T LT-FW @Cooler Gen-Set 4 outlet  T 3WV @ME, Gen-sets (x3) Figure 20. LT pump fault case study: 2-dimensional representation plane (25x25) with the SOM nodes (left). Corresponding the U-Matrix (right). Figure 20. LT pump fault case study: 2-dimensional representation plane (25x25) with the SOM nodes (left). Corresponding the U-Matrix (right). Figure 18. LT pump fault case study; left: Sensor variables & locations. Right: Differences (%) between the signals that correspond to the fault condition and the ones that correspond to the “non-fault” condition. Figure 21. 3-way regulation valve fault case study: heatmaps of the sensor variables (Figure 14). Figure 21. 3-way regulation valve fault case study: heatmaps of the sensor variables (Figure 14). Figure 21. 3-way regulation valve fault case study: heatmaps of the sensor variables (Figure 14). 11 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Figure 22. LT pump fault case study: heatmaps of the sensor variables (Figure 18). Figure 22. LT pump fault case study: heatmaps of the sensor variables (Figure 18). Figure 22. LT pump fault case study: heatmaps of the sensor variables (Figure 18). This work has been funded by the Norwegian Research Council (NFR) through its innovation programme for maritime activities and offshore operations (MAROFF) and in close collaboration with our project partners Rolls-Royce Marine and Farstad Shipping. 8. CONCLUSION By encapsulating model-based and data-driven analytics the DNVGL digital twin platform will enable new applications such as performance and environmental monitoring, energy efficiency optimisation, future classification services and potentially condition monitoring. The present paper describes a first principle digital twin of an anchor handling vessel’s propulsion system along with its cooling network developed in DNVGL COSSMOS for condition monitoring purposes. After the model was built and tested, an extensive and automated sensitivity analysis with respect to possible system failures has taken place. The underlying benefit in this approach is that with a single simulation run the analyst obtains all failure patterns related to the system operation in that specific state. Based on this information, a SOM was used to cluster the sensor readings pattern (difference between nominal and actual sensor readings) with the model-based sensitivity results. By assigning the unknown failure pattern (from the sensor readings) to one of the clusters, the algorithm directly identifies the unknown fault to a single or a set of failure modes (that correspond to the specific cluster to which the unknown fault has been matched). The present case-studies have demonstrated the potential of the methodology. A powerful aspect of the approach is its ability to identify faults and failures in areas that are not directly subjected to measurements; by simply using signal values at other positions and employing first principle models, the algorithm is capable of correlating sensor readings with specific failures at other parts of the system. REFERENCES Baraldi, P., Compare, M., Sauco, S., & Zio, E. (2013). Ensemble neural network-based particle filtering for prognostics. Mechanical Systems and Signal Processing, 41, 288-300. Hansen, J. F., Adnanes, A. K., & Fossen, T. I. (2001). Mathematical Modelling of Diesel-Electric Propulsion Systems for Marine Vessels. Mathematical and Computer Modelling of Dynamical Systems, 7(3), 323-355. doi: 10.1076/mcmd.7.3.323.3641 Baraldi, P., Maio, F. D., Genini, D., & Zio, E. (2015). Comparison of Data-Driven Reconstruction Methods for Fault Detection. IEEE Transactions on Reliability, 64(3), 852-860. Isermann, R. (2011). Fault-Diagnosis Applications: Model- Based Condition Monitoring: Actuators, Drives, Machinery, Plants, Sensors, and Fault-tolerant Systems: Springer Blanke, M. (2001). Enhanced maritime safety through diagnosis and fault tolerant control. Paper presented at the IFAC conference CAMS' 2001. Ito, K., & Akagi, S. (1986). An optimal planning method for a marine heat and power generation plant by considering its operational problem. Energy Research, 10 (1), 75-85. doi: 10.1002/er.4440100109 Campora, U., & Figari, M. (2003). Numerical simulation of ship propulsion transients and full-scale validation. Proceedings of the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment, 217, 41-52. Jefferson, M., Zhou, P. L., & Hindmarch, G. (2003 ). Analysis by computer simulation of a combined gas turbine and steam turbine (COGAS) system for marine propulsion. Paper presented at the Proceeding of IMarEST - Part A - Journal of Marine Engineering and Technology. Chen, J., & Patton, R. J. (1999). Robust Model-Based Fault Diagnosis for Dynamic Systems: Springer. Coull, G. (2015). Condition monitoring in the marine sector. Naval Architect, 20-22. Kakalis, N. M. P., & Dimopoulos, G. (2012). Managing the complexity of marine energy systems Position Paper: Det Norske Veritas, Research & Innovation. Dimopoulos, G. G. (2009). Synthesis, Design and Operation Optimization of Marine Energy Systems. PhD Phd Thesis, National Technical University of Athens, Athens, Greece. Kohonen, T. (2001). Self-Organizing Maps: Springer Series in Information Sciences. Dimopoulos, G. G., Georgopoulou, C. A., Stefanatos, I. C., Zymaris, A. S., & Kakalis, N. M. P. (2014). A general- purpose process modelling framework for marine energy systems. Energy Conversion and Management 86 (2014) 325–339. Kouremenos, D. A., & Hountalas, D. T. (1997). Diagnosis and condition monitoring of medium-speed marine diesel engines, . Tribotest Journal, 4(1), 63–91. Kyrtatos, N. P., & Lambropoulos, V. (2000). Modular Simulation of Marine Propulsion Systems Using an Engineering Building Block Approach. Paper presented at the ISME 6th International Symposium on Marine Engineering, Tokyo, Japan. Dimopoulos, G. EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 NOMENCLATURE & ABBREVIATIONS AH Anchor handling (vessel operating mode) b model parameters vector CBM Condition based maintenance CM Condition monitoring COSSMOS Complex Ship System Modelling & Simulation COMPASS Condition and operation monitoring for performance and ship safety DG Diesel generator DWT Deadweight tonnage F vector function describing the transient in time PDAE model equations FW Fresh water GUI Graphical user interface H vector function describing steady in time PDAE model equations HEX Heat exchanger HT High temperature (cooling network) LT Low temperature (cooling network) MBSE Model based system engineering ME Main engine MODAM Model based, data driven asset management MV Measured process variables PDAE Partial differential algebraic equations PID Proportional-integral-derivative control PTI Power take in SOM Self organising maps SG Shaft generator STBD Starboard side (vessel) SW Sea water t time u algebraic process variables vector VFD Variable frequency drive x differentiation domain Y differential process variables vector Next steps will include the utilisation of the methodology along with real-life signal data. Aspects like signal filtering, sensor and model epistemic uncertainties will be also investigated. To the authors’ view the current work is a first step towards a model-based prognostic tool for complex marine energy systems that will incorporate physics of failure with the exact current system condition to reason about possible system state under an envelope of possible future operational scenarios. ACKNOWLEDGEMENT 12 REFERENCES G., & Kakalis, N. M. P. (2010). An integrated modelling framework for the design, operation and control of marine energy systems. Paper presented at the 26th CIMAC World Congress, Bergen, Norway. Kyrtatos, N. P., Theodossopoulos, P., Theotokatos, G., & Xiros, N. (1999). Simulation of the Overall Ship Propulsion Plant for Performance Prediction and Control. Paper presented at the MarPower Conference, Newcastle-upon- Tyne, UK. Dimopoulos, G. G., Kougioufas, A. V., & Frangopoulos, C. A. (2008). Synthesis, design and operation optimization of a marine energy system. Energy, 33(2), 180-188. DNVGL. (2015). Research Review 2015 Next generation winning concepts. In D. S. R. Innovation: (Ed.). DNVGL. (2015). Research Review 2015 Ludvigsen, K. B., Jamt, L. K., Husteli, N., & Smogeli, Ø. (2016). Digital twins for design, testing and verification throughout a vessel’s life cycle. Paper presented at the COMPIT 2016 (to be presented), Lecce, Italy. Next generation winning concepts. In D. S. R. Innovation: (Ed.). Ge, Z., Song, Z., & Gao, F. (2013). Review of Recent Research on Data-Based Process Monitoring. Industrial & Engineering Chemistry Research(52), 3543−3562. Manno, G., et, & Al. (2015). MODAM: Model based Data Driven Asset Management, DNV GL, ABB Marine and BW Gas - Project Plan. Glaessgen, E. H., & Stargel, D. S. (2012). The digital twin paradigm for future NASA and US Air Force vehicles. Paper presented at the 53rd Structures, Structural Dynamics and Manno, G., Knutsen, K. E., & Vartdal, B. J. (2014). An Importance Measure approach to system level Condition Monitoring of Ship Machinery Systems. Paper presented at the CM 2014 and MFPT 2014, Manchester Grimmelius, H. T., Meiler, P. P., Maas, H. L. M. M., Bonnier, B., Grevink, J. S., & Kuilenburg, R. F. v. (1999). 13 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 Monnin, M., Voisin, A., Leger, J.-B., & Iung, B. (2011). Fleet-wide health management architecture. Paper presented at the Annual Conference of the Prognostics and Health Management Society, 2011, Montreal, Quebec, Canada Wang, X. Z. (1999). Data Mining and Knowledge Discovery for Process Monitoring and Control. London: Springer. Monnin, M., Voisin, A., Leger, J.-B., & Iung, B. (2011). Fleet-wide health management architecture. Paper presented at the Annual Conference of the Prognostics and Health Management Society, 2011, Montreal, Quebec, Canada Watzenig, D., Sommer, M. S., & Steiner, G. (2013). Model- Based Condition and State Monitoring of Large Marine Diesel Engines. In S. Bari (Ed.), Diesel Engine - Combustion, Emissions and Condition Monitoring. SOMPY, from Moosavi, V. (2016). SOMPY, https://github.com/sevamoo/SOMPY, from https://github.com/sevamoo/SOMPY Yin, S., Ding, S. X., Xie, X., & Luo, H. (2014). A Review on Basic Data-Driven Approaches for Industrial Process Monitoring. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 61(11). Naumann, U. (2012). The Art of Differentiating Computer Programs: An Introduction to Algorithmic Differentiation. . SIAM series on Software, Environments and Tools. Pedersen, T.-A., & Pedersen, E.-H. (2012). Bond graph modelling of marine power systems. Mathematical and Computer Modelling of Dynamical Systems: Methods, Tools and Applications in Engineering and Related Sciences 18(2), 153-173. doi: DOI:10.1080/13873954.2011.603735 PSE. (2016). gPROMS, from http://www.psenterprise.com/gproms.html gPROMS, Rüde, E. (2006). Entwicklung eines Simulationsmodells für ein Kühlwassersystem zur Identifikation der Zustandsüberwachungsparameter. Diploma Thesis, Technische Universität Hamburg-Hamburg, Hamburg. Schenk, A., Natale, S., Germond, A., Boss, P., & Lam, A. (2002). Tranformer state assessment: Self-orginizing maps as an intelligent on-line data processor. Paper presented at the CIGRE 2002 Paris. SINTEF. (2002). OREDA: Offshore Reliability Data: OREDA Participants. Tostevin, G. M., & Nealy, J. C. (2003). Computer synthesis for the design of marine power plants. Journal of marine design and operations, B (2003), 25-32. Tuegel, E. J., Ingraffea, A. R., Eason, T. G., & Spottswood, S. M. (2011). Reengineering aircraft structural life prediction using a digital twin. Int. J. Aerospace Eng. Twiddle, J. A., & Jones, N. B. (2002). Fuzzy model-based condition monitoring and fault diagnosis of a diesel engine cooling system. Proc Instn Mech Engrs Part I: J Systems and Control Engineering, 216 Voisin, A., Medina-Oliva, G., Monnin, M., Leger, J.-B., & Iung, B. (2013). Fleet-wide Diagnostic and Prognostic Assessment. Paper presented at the Annual Conference of the Prognostic and Health Management Society 2013, New Orleans, USA. Vrijdag, A., Stapersma, D., & Terwisga, T. v. (2009). Systematic modelling, verification, calibration and validation of a ship propulsion simulation model. Proceedings of IMarEST - Part A - Journal of Marine Engineering and Technology A15:2009 3-20. 14 EUROPEAN CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2016 APPENDIX I In the current appendix, namely in Figure 23, the system locations that correspond to the comparison of results presented in Figure 7 are depicted. Figure 23. COSSMOS model: system locations for the comparison of results presented in Figure 7. Figure 23. COSSMOS model: system locations for the comparison of results presented in Figure 7 15
W4318752643.txt
https://zenodo.org/records/7585685/files/Fesenfeld_etal_SDSN_Guide_F.pdf
fr
L'AVENIR DE L'ALIMENTATION EN SUISSE - Guide des principaux leviers et axes politiques pour établir un système alimentaire durable
Zenodo (CERN European Organization for Nuclear Research)
2,023
cc-by
54,831
Sustainable Development Solutions Network Switzerland L’AVENIR DE L’ALIMENTATION EN SUISSE Guide des principaux leviers et axes politiques pour établir un système alimentaire durable Une publication du comité scientifique Avenir Alimentaire Suisse, qui regroupe plus de quarante chercheuses et chercheurs d’institutions suisses de premier plan 1 2 PRÉFACE D'un point de vue scientifique, il est clair que notre système alimentaire n'est pas durable. Pour préserver nos bases de vie et d’économie, une réorientation de l'ensemble de la chaîne de valeur est nécessaire. Celle-ci est en même temps une clé pour atteindre l'Agenda 2030 pour le développement durable. SDSN Suisse a initié le comité scientifique Avenir Alimentation Suisse afin de créer un guide : Il doit permettre à la Suisse de saisir les opportunités à temps et d'éviter des conséquences financières incontrôlables. Le comité scientifique a réalisé un travail de pionnier au niveau international. Dans le cadre d'un processus scientifique interdisciplinaire, une voie d'action globale a été élaborée pour la première fois pour un pays afin de réorienter le système alimentaire en accord avec les objectifs de développement durable. Les chercheurs impliqués créent ainsi une base importante pour la poursuite du débat politique en Suisse et au niveau international. Ce travail important, qui prend la forme d'un guide scientifique largement soutenu, n'a pu voir le jour que grâce à l'engagement infatigable de Lukas Fesenfeld, responsable scientifique du groupe, et de plus de quarante autres chercheurs d'institutions suisses de premier plan. C'est à lui et à eux tous que nous adressons nos plus vifs remerciements. Janet Hering & Heidi Strebel Co-présidence du Sustainable Development Solutions Network SDSN Suisse Carole Küng & Jorge Tamayo Co-direction du Sustainable Development Solutions Network SDSN Suisse 3 4 TABLE DES MATIÈRES PRÉFACE 3 TABLE DES MATIÈRES 5 RÉSUMÉ 6 1 10 INTRODUCTION 1.1 Les Objectifs de développement durable de l’ONU et leur importance pour le système alimentaire suisse 1.2 Durabilité et système alimentaire : La situation actuelle en Suisse 1.3 Points de basculement positifs dans la transformation du système alimentaire 2 MODÈLE ET OBJECTIFS CENTRAUX POUR LA TRANSFORMATION DU SYSTÈME ALIMENTAIRE SUISSE 2.1 Conflits d’objectifs et synergies potentielles  2.2 Un modèle pour le développement du système alimentaire suisse en accord avec les ODD 2.3 Indicateurs cibles mesurables pour la réalisation des ODD Objectif 1 : Adaptation de l’alimentation conformément aux recommandations nationales en matière de nutrition et au régime alimentaire Planetary Health Diet Objectif 2 : Réduction des pertes d’azote et de phosphore dans le système alimentaire Objectif 3 : Réduction des effets écotoxiques des pesticides Objectif 4 : Utilisation durable des terres agricoles en Suisse Objectif 5 : Réduction de la perte de biodiversité Objectif 6 : Réduction de l’empreinte hydrique liée à l’alimentation Objectif 7 : Réduction des émissions de gaz à effet de serre Objectif 8 : Réduction des déchets et des pertes alimentaires Objectif 9 : Revenus de subsistance tout au long de la chaîne d’approvisionnement Objectif 10 : Réduction des risques psychosociaux et physiques pour la santé dans l’agriculture suisse Objectif 11 : Respect des normes de l’OIT sur le travail des enfants tout au long de la chaîne d’approvisionnement 2.4 Comparaison des indicateurs cibles sur le statu quo du système alimentaire suisse 3 VOIES POLITIQUES POUR LA TRANSFORMATION DU SYSTÈME ALIMENTAIRE SUISSE 3.1 Aperçu des recommandations d’actions stratégiques Une voie d’action pour la transformation du système alimentaire suisse 4 10 12 16 19 19 20 21 22 24 25 27 28 29 30 31 31 32 32 33 38 40 42 Paquet de mesures I : Introduction d’un fonds de transformation Paquet de mesures II : Mesures réglementaires I et taxes d’incitation Paquet de mesures III : Mesures de politique agricole et soutien aux zones rurales Paquet de mesures IV : Mesures réglementaires II 46 49 55 56 LA GOUVERNANCE DU SYSTÈME ALIMENTAIRE EN SUISSE 60 4.1 Le rôle de la gouvernance dans la transformation du système alimentaire suisse 4.2 Gouvernance du système alimentaire suisse 2030 : Une proposition 60 CHAPITRE FINAL 66 ÉQUIPE DE PROJET DU COMITÉ SCIENTIFIQUE AVENIR ALIMENTAIRE SUISSE 69 BIBLIOGRAPHIE 70 INDEX DES ILLUSTRATIONS 74 MENTIONS LÉGALES 75 62 5 RÉSUMÉ Saisir les opportunités, assurer la sécurité alimentaire : Des voies vers un avenir alimentaire durable La sécurité alimentaire est menacée. Les guerres, les pandémies, le changement climatique et l’appauvrissement de la biodiversité mettent également en péril l’approvisionnement de la Suisse. En tant que pays interconnecté au niveau international, la Suisse est tributaire d’écosystèmes fonctionnant à l’échelle mondiale et de chaînes d’approvisionnement stables, tant pour la sécurité de son approvisionnement que pour son économie. Parallèlement, le système alimentaire actuel est le moteur de ces crises. Il dépasse les limites planétaires, est notamment à l’origine d’environ 30% des émissions de gaz à effet de serre dans le monde et constitue un facteur déterminant dans l’appauvrissement de la biodiversité. Les risques accrus de pandémie et les troubles sociaux croissants ont également une grande influence sur la manière dont nous produisons, traitons, distribuons et consommons actuellement les denrées alimentaires. Pour garantir la sécurité alimentaire, il est donc indispensable, d’un point de vue scientifique, d’orienter globalement le système alimentaire vers un développement durable tout au long de la chaîne de valeur, et pas seulement l’agriculture au sens strict du terme. Les 17 Objectifs de développement durable (ODD) des Nations Unies, que la Suisse a signés avec tous les autres membres de l’ONU dans le cadre de l’Agenda 2030, en constituent le cadre. Outre les aspects écologiques, il faut également tenir compte des aspects économiques et sociaux, par exemple des revenus souvent faibles dans l’agriculture et l’industrie alimentaire. Ceci est particulièrement important pour assurer une mutation socialement équitable du système alimentaire. La performance et les besoins des différent·e·s actrices et acteurs tout au long de la chaîne de valeur du système alimentaire devraient être appréciés à leur juste valeur et les coûts éventuels devraient être répartis et compensés de manière équitable. Les chances offertes par cette mutation, qui l’emportent nettement sur les coûts, devraient être au centre du débat. La Suisse, pionnière du changement : Faire avancer la transformation ensemble Malgré sa taille modeste, la Suisse peut contribuer de manière substantielle au changement global. Premièrement, la Suisse a une empreinte environnementale par habitant·e supérieure à la moyenne, avec de fortes répercussions à l’étranger. Deuxièmement, la politique, l’économie, la place financière et la science suisses sont fortement interconnectées au niveau mondial et peuvent donner des impulsions décisives pour un 6 système alimentaire plus durable. Troisièmement, la Suisse a la chance, en tant que pionnière du changement, non seulement d’en profiter fortement elle-même, mais aussi de servir de laboratoire et de modèle pour des processus similaires dans d’autres pays du monde. Des efforts orientés vers un système alimentaire durable ont été et sont déjà entrepris à différents niveaux de la chaîne de valeur. Par exemple, dans l’agriculture suisse, les surfaces de biodiversité ont augmenté et les émissions de gaz à effet de serre ont été réduites depuis les années 1990. Un certain nombre de grandes entreprises suisses se sont fixé des objectifs contraignants de réduction des émissions de CO2. Dans le commerce de détail également, plusieurs entreprises ont pris des mesures en faveur de la durabilité. En outre, les consommatrices et les consommateurs sont de plus en plus conscient·e·s de la durabilité. Mais ce qui a été fait jusqu’à présent n’est pas suffisant et se fait trop lentement. En Suisse, par exemple, le taux de nitrates dans les eaux souterraines se maintient depuis 2002 à un niveau particulièrement élevé par rapport au reste du monde. De plus, notre alimentation provoque chaque année le défrichage de grandes surfaces de forêts, notamment dans des zones particulièrement riches en espèces. En outre, avec une part de près de 70%, la plupart des émissions de gaz à effet de serre de notre alimentation se produisent à l’étranger. Il y a urgence, notamment en raison des points de basculement dangereux des écosystèmes, à partir desquels, par exemple, le changement climatique et la disparition des espèces s’accélèrent de manière drastique. Sans une action rapide, la sécurité alimentaire mondiale, y compris en Suisse, ne pourra pas être garantie à long terme et les objectifs de développement durable seront loin d’être atteints. Afin d’aborder ces défis de manière scientifique, SDSN Suisse, la branche Suisse d’une initiative des Nations Unies, a mis sur pied le comité scientifique interdisciplinaire Avenir Alimentaire Suisse. Ce panel de 42 scientifiques a élaboré le présent guide des principaux leviers et axes politiques pour établir un système alimentaire durable. Dans ce document, le comité explique que des objectifs nettement plus ambitieux et des mesures concrètes tout au long de la chaîne de valeur sont nécessaires pour une transformation durable. La transformation du système alimentaire est une tâche qui incombe à l’ensemble de la société. Outre les acteurs tels que l’industrie et l’agriculture ainsi que les consommatrices et les consommateurs, l’État doit tout particulièrement jouer un rôle clé dans la réorientation stratégique de la politique du système alimentaire suisse. D’un point de vue scientifique, l’objectif doit être un approvisionnement en denrées alimentaires selon des principes agroécologiques, qui combine les intérêts économiques avec les intérêts écologiques et sociaux sur l’ensemble du système alimentaire. Plus cette réorientation sera rapide, plus les chances de profiter du changement, d’éviter les crises et de minimiser les coûts seront grandes. Modèle et indicateurs pour un système alimentaire durable Tout d’abord, dans le chapitre 2 du présent rapport, les scientifiques ont élaboré un modèle pour un système alimentaire durable et l’ont assorti de onze indicateurs cibles prioritaires pour atteindre les objectifs de développement durable. Il s’agit notamment d’adapter le régime alimentaire suisse en accord avec le Planetary Health Diet, de réduire les pertes d’azote et de phosphore ainsi que les effets écotoxiques des pesticides, d’utiliser de manière durable les sols exploités par l’agriculture en Suisse et de réduire la déforestation, les pertes de biodiversité, les émissions de gaz à effet de serre, les déchets alimentaires et l’empreinte hydrique liée à l’alimentation. Ces indicateurs d’objectifs écologiques sont complétés par des objectifs sociaux mesurables, tels que la réduction des risques psychosociaux et physiques pour la santé dans l’agriculture, ainsi que la garantie de revenus suffisants pour vivre et le respect des normes de l’OIT sur le travail des enfants tout au long de la chaîne d’approvisionnement. La comparaison de ces onze indicateurs avec le statu quo montre que des efforts évidents sont nécessaires pour transformer le système alimentaire suisse d’ici 2030, conformément aux ODD. Une politique stratégique peut déclencher des points de basculement positifs Sur la base de ce constat, le comité scientifique esquisse au chapitre 3 une voie d’action politique concrète pour que l’État suisse puisse accélérer ce changement en collaboration avec les autres acteurs du système. Pour que la politique du système alimentaire soit efficace, rentable et réalisable, il ne suffit pas de choisir les mesures ; il faut également veiller à ce qu’elles s’enchaînent dans le temps de manière stratégique- ment cohérente. Il est possible de cibler des points de basculement positifs qui déclenchent des processus d’accélération dans la bonne direction. Pour cette raison, le comité scientifique ne se contente pas de proposer des mesures isolées : il les classe par ordre de priorité en fonction de leur impact, de leur urgence et de leur faisabilité, les regroupe en quatre paquets de mesures et les place dans un ordre chronologique jusqu’en 2030. Ici, le principe de base est d’abord l’encouragement ciblé, puis le renforcement progressif des exigences. La piste d’action politique stratégique proposée met particulièrement l’accent sur les opportunités de la mutation. Série de mesures I : Création d’un fonds de transformation Dans une première phase, le comité scientifique recommande la création d’un vaste fonds de transformation d’ici 2025. Ce fonds devrait permettre de financer en temps utile des mesures d’information et de formation ainsi que des instruments d’incitation positifs. Concrètement, il pourrait s’agir par exemple de programmes de formation initiale et continue pour les professions pertinentes de l’ensemble de la chaîne de valeur, mais aussi de programmes d’aide à la conversion des exploitations pour les jeunes agricultrices et agriculteurs. Une promotion ciblée des innovations technologiques, mais aussi des primes de conversion pour les exploitations agricoles, ainsi que la promotion d’offres durables dans les cantines et les restaurants sont des éléments centraux du fonds. Un fonds de transformation à orientation stratégique en phase 1 contribue à générer de nouvelles possibilités de création de valeur, à changer les normes sociales et à augmenter l’acceptation de mesures plus poussées dans les phases ultérieures. Au début, le fonds pourrait être alimenté par des moyens budgétaires supplémentaires de la Confédération combinés à des moyens privés, puis être élargi dans les phases ultérieures par des taxes d’incitation et la réaffectation de moyens existants. L’organisation précise du fonds peut s’inspirer d’exemples de réussite dans d’autres pays, par exemple au Danemark. 7 Série de mesures II : Mesures réglementaires et taxes d’incitation Dans une deuxième phase, il faudrait, à partir de 2025, recourir de plus en plus à des mesures réglementaires et à des taxes d’incitation pour atteindre les objectifs de développement durable. D’un point de vue scientifique, il serait judicieux d’instaurer des taxes d’incitation favorisant la réalité des coûts, d’adapter les droits de douane et de mettre en place des directives nationales basées sur des critères de durabilité pour les achats publics et les cantines des établissements publics. Par ailleurs, il faudrait à ce stade mettre en œuvre de manière contraignante des accords sectoriels visant à introduire des objectifs de réduction basés sur le savoir ainsi que des devoirs de diligence renforcés pour les groupes de sociétés selon des directives internationales. Série de mesures III : Mesures de politique agricole et soutien aux zones rurales La troisième phase (à partir de 2026 environ) se concentre sur des mesures de politique agricole et le soutien aux zones rurales. On peut par exemple envisager une adaptation des paiements directs, des mesures de soutien du marché et des aides à l’investissement, ainsi qu’un impôt négatif sur le revenu dans l’agriculture. Les mesures commerciales devraient également être systématiquement axées sur les systèmes alimentaires durables au plus tard à ce stade. Série de mesures IV : Mesures réglementaires plus profondes Pour la quatrième phase (à partir de 2030 environ), il est possible d’introduire des mesures qui sont encore moins bien acceptées actuellement. Grâce à des mesures d’encouragement et de compensation dans le cadre du fonds de transformation ainsi qu’à de nouvelles possibilités de création de valeur, des mesures réglementaires plus profondes et des taxes d’incitation plus élevées deviendront progressivement plus réalistes d’ici 2030. Accélérer le processus de négociation pour une politique du système alimentaire Une réorientation aussi complète de la politique du système alimentaire suisse dans le sens de l’Agenda 2030 nécessite un processus de négociation orienté vers le compromis pour l’ensemble du système alimentaire. Pour que ce processus soit aussi fructueux que possible et afin de minimiser la polarisation, la Confédération, les cantons et les communes doivent jouer un rôle de leader. Une politique efficace en matière de système alimentaire suppose donc une gouvernance intégrée du système alimentaire. Le chapitre 4 de ce guide montre à quoi celle-ci pourrait ressembler. La recommandation centrale du comité scientifique sur la gouvernance est la création d’une commission sur l’avenir du système alimentaire. Celle-ci a pour but d’accélérer le processus de négociation confidentiel et multilatéral entre les acteurs centraux tout au long de la chaîne de valeur. Il convient de garantir une représentation équilibrée des principaux groupes 8 d’intérêt, une médiation neutre et un suivi scientifique. Les changements rapides nécessaires impliquent en outre des procédures régulières de participation des citoyen·ne·s, non pas en tant que parlements de substitution, mais en tant qu’organes consultatifs. Le comité scientifique propose également de repenser à moyen terme la base légale dans le sens d’une loi générale sur le système alimentaire et de permettre une meilleure intégration des offices concernés pour cette tâche transversale. En résumé, le comité scientifique Avenir Alimentaire Suisse convoqué par SDSN Suisse constate qu’il est urgent de s’engager sur de nouvelles voies vers un système alimentaire durable afin de garantir les objectifs de développement durable et la sécurité alimentaire. Le processus de changement nécessaire offrant toutefois de grandes opportunités pour la Suisse, il devrait être mené rapidement de manière stratégique et collective. 9 1 INTRODUCTION 1.1 Les Objectifs de développement durable de l’ONU et leur importance pour le système alimentaire suisse Auteur·es chapitre 1 : Lukas Fesenfeld, Sol Kislig, Julie Zähringer « Sans une transformation profonde du système alimentaire, le monde ne pourra pas atteindre à la fois les Objectifs de développement durable des Nations Unies et l’accord de Paris sur le climat. » Prof. Johan Rockström et al. dans Nature Food, 20201 Il est scientifiquement indéniable que notre système alimentaire mondial actuel n’est pas durable. La façon dont nous produisons, transformons et consommons les aliments contribue de manière déterminante aux multiples crises écologiques, sociales, politiques, sanitaires et économiques de notre époque, qui touchent à la fois les agriculteurs et agricultrices, les entreprises et les consommateurs et consommatrices. Le système alimentaire actuel, en d’autres termes l’ensemble de la chaîne de valeur des aliments, contribue considérablement au dépassement des limites planétaires. Il est par exemple à l’origine d’environ 30% des émissions mondiales de gaz à effet de serre et joue un rôle déterminant dans la déforestation et la perte de la biodiversité2. Cela est d’autant plus problématique qu’il existe des points de basculement dangereux dans nos écosystèmes. La déforestation de l’Amazonie, par exemple, peut modifier de manière brutale et difficilement réversible autant le climat local que le climat mondial, ce qui favorise ensuite le dégel du pergélisol et le franchissement d’autres points de basculement.3 Les crises aiguës telles que le changement climatique, l’extinction des espèces et la dégradation des terres s’en trouvent clairement et rapidement amplifiées, ce qui menace ensuite les récoltes par des sécheresses persistantes et des phénomènes météorologiques extrêmes. La sécurité alimentaire mondiale et suisse est ainsi de plus en plus menacée. En tant que place économique et pays mondialement interconnecté, la Suisse est particulièrement dépendante de chaînes d’approvisionnement stables, d’écosystèmes fonctionnant à l’échelle mondiale et de l’importation de nombreuses ressources naturelles. Cela est d’autant plus évident lors de situations de crise aiguë : l’offensive russe contre l’Ukraine et la pandémie de coronavirus tendent à montrer que le système alimentaire mondial n’est pas résilient. La guerre a par exemple provoqué une forte hausse des prix des aliments, des engrais et des carburants, entraînant une détérioration de la sécurité alimentaire, en particulier dans les pays à faible et moyen revenu. De plus, ces crises aiguës aggravent les 10 inégalités. Alors que les coûts de la santé augmentent fortement dans le monde entier, y compris en Suisse, en raison notamment de l’obésité et de la malnutrition, 10% de la population mondiale souffrent dans le même temps de dénutrition chronique.4-6 Ce chiffre atteint même 20% en Afrique, alors qu’en Europe, le nombre de personnes concernées n’est que de 2,5%.7 Les inégalités sociales dans le système alimentaire constituent un défi considérable, y compris au sein des différents pays et régions. Ainsi, même dans les pays prospères comme la Suisse, les personnes à faible revenu s’offrent généralement moins souvent des aliments sains que les personnes à revenu élevé.8 C’est l’une des raisons pour lesquelles les classes à faible revenu sont particulièrement touchées par les conséquences sanitaires et les coûts de la malnutrition. Nombre d’agriculteurs, d’agricultrices, de producteurs et de productrices de denrées alimentaires, tout au long de la chaîne de valeur, souffrent également de la forte pression du travail et des faibles revenus. Leur mission est cependant très précieuse, car ce sont eux et elles qui fournissent à la population des aliments frais, de qualité et locaux. Il est donc essentiel d’impliquer toutes les parties prenantes de la chaîne de valeur dans le processus de transformation du système alimentaire et de les rémunérer de manière équitable. Même en Suisse, environ un cinquième des ménages agricoles vit aujourd’hui en dessous du minimum vital (mesuré par rapport à l’aide sociale). Les situations de revenus précaires existent également à d’autres niveaux de la chaîne de valeur liée à l’alimentation, par exemple dans le domaine de la transformation et de la distribution des aliments.9 Elles sont particulièrement problématiques dans de nombreux pays d’où la Suisse im- DÉFINITION DU SYSTÈME ALIMENTAIRE Le système alimentaire comprend toutes les composantes et activités allant de la production des aliments jusqu’à leur consommation, en passant par leur transformation et leur distribution, ainsi que toutes les substances que le système absorbe (inputs) et rejette (outputs).11 Un système alimentaire durable est un système alimentaire qui garantit aujourd’hui l’alimentation et la sécurité alimentaire sans compromettre les bases sociales, économiques et environnementales nécessaires pour garantir la sécurité alimentaire des générations futures.11 SYSTÈME D’ALIMENTATION Économie & technique Production Agriculture Industrie Transformation Emballage Stockage Transport Ressources naturelles Valorisation Élimination Recyclage Déchets Commerce Commercialisation Cuisine Restauration Achats Société & culture porte des denrées alimentaires et des matières premières. Des revenus équitables et suffisants ainsi que la création de valeur sont donc une condition essentielle pour réussir ce processus de transformation. Il s’agira donc à l’avenir de prendre davantage au sérieux les besoins des différents acteurs de l’ensemble de la chaîne de valeur et d’éviter que les mesures de transformation ne portent atteinte à leurs moyens de subsistance. Les éventuelles restrictions de rendement ou charges financières doivent être examinées sous l’angle de leur compatibilité sociale et être compensées de manière ciblée. Les inégalités sociales et économiques constituent un défi majeur pour la transformation du système alimentaire. En termes positifs : la transformation du système alimentaire en vue d’une plus grande durabilité bénéficie d’une acceptation sociale et politique élevée, surtout lorsqu’une attention particulière est accordée à la justice sociale et à la répartition équitable des coûts de la transformation.10 Dans ce contexte, il ne faut pas adopter une perspective qui serait limitée à l’agriculture, à la transformation, au commerce ou à la consommation, mais une perspective globale du système alimentaire (voir encadré p. 10).11,12 La transformation du système alimentaire est donc une tâche politique et sociétale. Convaincu que ces défis doivent être abordés de manière interdisciplinaire et intersectorielle à tous les niveaux, le Sustainable Development Solutions Network (SDSN Suisse) a créé le comité scientifique Avenir Alimentaire Suisse. Les 42 scientifiques ont régulièrement échangé leurs points de vue dans le cadre d’une méthode Delphi itérative (voir les détails dans l’appendice en ligne) et ont rassemblé des connaissances interdisciplinaires sur la transformation du système alimentaire. Le présent guide est né de ces discussions et de ces interrogations systémiques. Fig. 1 : Fesenfeld et al. (2023)237 / Telek En 2015, la Suisse a signé, avec tous les États membres des Nations Unies, l’Agenda 2030 pour le développement durable et ses Objectifs de développement durable (ODD)13. Les 17 objectifs et leurs 169 sous-objectifs contribuent au développement durable dans toutes ses dimensions. On distingue en principe les dimensions sociale, économique et écologique de la durabilité, qui constituent les piliers de l’Agenda 2030. Les objectifs doivent permettre une transformation socialement juste et équitable vers plus de durabilité, sans laisser personne de côté. « Leave no one behind » est donc la promesse centrale de l’Agenda 2030. Les objectifs doivent être mis en œuvre à l’échelle mondiale et par tous les États membres des Nations Unies d’ici à 2030.14 En 2015, la communauté internationale a également adopté l’accord de Paris sur le climat, qui a pour objectif de limiter le réchauffement mondial à bien moins de 2 °C par rapport à l’ère préindustrielle et de viser une augmentation maximale de la température de 1,5 °C.15 La Suisse a ratifié cet accord en 2017. La transformation du système alimentaire mondial constitue l’un des leviers les plus importants pour atteindre les 17 ODD et l’objectif climatique de Paris.16,17 Comme expliqué plus haut, le système alimentaire n’influence pas seulement l’objectif de développement durable de réduction de la faim (ODD 2) ; il est aussi étroitement lié à presque tous les ODD.18 Ainsi, une transformation durable du système alimentaire et des habitudes alimentaires peut également garantir une vie plus saine (ODD 3)19 et contribuer de manière significative à des conditions de travail décentes et à une croissance économique durable (ODD 8) ainsi qu’à des modèles de consommation et de production durables (ODD 12).20 En outre, notre système alimentaire revêt une grande importance pour la disponibilité et la gestion durable de l’eau (ODD 6)21, la réduc11 tion des émissions de gaz à effet de serre (ODD 13)16 et l’utilisation plus durable des écosystèmes terrestres et marins (ODD 14 et 15).22 En raison du rôle clé du système alimentaire mondial dans la réalisation des ODD, sa transformation durable est inévitable. Les Nations Unies ont illustré cette nécessité en organisant pour la première fois un sommet mondial sur le système alimentaire le 23 septembre 2021 (United Nations Food System Summit [UNFSS]). À l’occasion de ce sommet, la Suisse s’est également engagée à rendre son système alimentaire plus durable. 1.2 Durabilité et système alimentaire : La situation actuelle en Suisse En Suisse, transformer le système alimentaire en accord avec les Objectifs de développement durable de l’ONU est à la fois un défi urgent et une grande opportunité. Au sein du pays, le système alimentaire actuel contribue de manière significative à l’augmentation des coûts de la santé ainsi qu’à la perte des ressources naturelles et de la biosphère, indispensables à la survie. En Suisse, l’alimentation est responsable d’environ 25% des impacts négatifs sur l’environnement du pays, causant ainsi autant de pollution que le domaine de l’habitat (qui inclut la consommation d’énergie liée à l’habitat, la construction de logements, les meubles et les appareils ménagers) et nettement plus que la mobilité privée (14%).23 L’impact de la consommation alimentaire de la Suisse se fait également nettement sentir au-delà des frontières du pays : plus de 80% des terres nécessaires à la production des aliments consommés en Suisse se trouvent en dehors du pays et plus de 95% des effets négatifs sur la biodiversité ne se produisent pas en Suisse.24 La production de denrées alimentaires et l’utilisation des terres dans les pays du Sud sont de plus en plus influencées par la consommation et la demande des pays prospères comme la Suisse. Dans le même temps, le changement d’affectation des terres dans le Sud a des répercussions mondiales telles que le changement climatique et la perte de la biodiversité, qui affectent à leur tour directement l’économie et la population suisses.25,26 Bien que l’impact mondial de la consommation alimentaire nationale soit très complexe, cette perspective globale doit être prise en compte afin de garantir la sécurité alimentaire suisse et de transformer durablement le système alimentaire suisse dans le cadre de l’Agenda 2030. Outre le fait qu’elle possède un secteur agroalimentaire innovant, la Suisse joue un rôle important dans le système alimentaire mondial en tant que siège d’entreprises de grands groupes internationaux, d’assurances et de banques. Elle a ici l’opportunité de se positionner sur le marché mondial en tant que place économique attractive et pionnière d’une industrie agroalimentaire axée sur la durabilité. C’est donc non seulement la prévention des coûts privés et publics élevés liés à l’augmentation des risques pour la santé et l’environnement, mais aussi les grandes opportunités économiques d’un marché de plus en plus orienté vers la durabilité qui font de la transformation globale du système alimentaire un investissement exceptionnellement rentable. La Suisse, en tant que place économique et scientifique mondialement interconnectée, avec son degré de liberté politique comparativement élevé, sa prospérité et sa structure territoriale, dispose donc des meilleures conditions pour mettre en place ce processus de transformation. L’influence de différentes entreprises suisses s’étend sur toute la chaîne de valeur. 12 La Suisse a donc le potentiel d’avoir un impact global pour transformer durablement le système alimentaire. Elle est l’une des plus importantes places de négoce de matières premières agricoles au monde et abrite plus de 550 sociétés de négoce de matières premières, y compris des sociétés de négoce agricole dominant le marché comme ADM, Cargill, ECOM, Louis Dreyfus, Volcafe, et Sucafina. Les entreprises basées en Suisse sont responsables de 60% du commerce des céréales, de 55% du commerce du café, de 45% du commerce du sucre et de 35% du commerce du cacao dans le monde. Givaudan, le plus grand producteur mondial d’arômes et de parfums, est un fournisseur important qui a également son siège en Suisse. Les groupes alimentaires suisses comme Nestlé, Lindt et Sprüngli ou Barry Callebaut fabriquent des produits consommés dans le monde entier. En tant que place financière, la Suisse a également une importance mondiale dans la mesure où elle finance et assure des entreprises agricoles et alimentaires dans le monde entier. Les entreprises suisses des secteurs agroalimentaires mettent déjà en place des initiatives pionnières en matière de durabilité. Ainsi, plus d’une douzaine d’entreprises suisses importantes du secteur alimentaire ont adhéré à la Science Based Targets initiative (SBTi), sept d’entre elles s’étant engagées à atteindre l’objectif zéro émission nette. Depuis des années, certains acteurs du secteur agroalimentaire suisse jouent un rôle de pionnier important dans de nombreux domaines. Un exemple parmi d’autres est la mise en place précoce d’un label bio rigoureux (en comparaison internationale) avec le soutien d’un leader du commerce de détail. Le secteur agroalimentaire suisse se caractérise par une forte capacité d’innovation, ce qui peut favoriser son adaptation face aux défis du développement durable. La roue colorée symbolise les 17 ODD. Fig. 2 : United Nations (2015)13 L’impact global de la consommation alimentaire nationale, la position de la Suisse en tant que place économique et scientifique mondialement interconnectée ainsi que la densité d’entreprises influentes dans le système alimentaire mondial ayant des liens avec la Suisse offrent donc l’opportunité à l’économie, à la science et à la société civile suisses d’exercer une influence sur la transformation vers un avenir alimentaire durable à l’intérieur et à l’extérieur des frontières nationales. Outre les initiatives actives de l’économie, de la science et de la société civile, l’État suisse – de l’échelon communal à l’échelon fédéral – est un acteur central du système alimentaire. Grâce à des conditions-cadres étatiques et à un soutien et un pilotage ciblés, l’État suisse peut accélérer de manière décisive le processus de transformation pour atteindre les objectifs de durabilité. En raison du rôle important joué par l’État et ses différentes institutions au niveau communal, cantonal et fédéral, ce guide scientifique développe des propositions sur la manière dont les responsables politiques peuvent contribuer stratégiquement à la mise en œuvre de l’Agenda 2030. La Suisse a déjà fait des progrès dans certains domaines en ce qui concerne la mise en œuvre de l’Agenda 2030 et des Objectifs de développement durable. Cependant, ils ne sont pas suffisamment rapides pour que l’on espère atteindre toutes les valeurs cibles comme prévu d’ici 2030. Selon le « Sustainable Development Report », la Suisse n’est en passe d’atteindre l’objectif d’ici 2030 que pour l’ODD 7 (énergie abordable et propre).27 Pour tous les autres indicateurs, il existe de grandes lacunes et un besoin de rattrapage. La consommation suisse de ressources naturelles dans le monde et l’empreinte écologique dépassent de loin les limites planétaires.28 Cela augmente aussi considérablement les risques que des points de basculement dangereux soient bientôt dépassés dans nos systèmes naturels (p. ex. la déforestation de l’Amazonie).3 Il est donc scientifiquement indéniable que la transformation du système alimentaire est urgente. Cela est particulièrement important pour la sécurité alimentaire et la prospérité économique de la Suisse, pays mondialement interconnecté et dépendant de chaînes d’approvisionnement stables et de l’importation de nombreuses ressources naturelles. À l’aide de trois exemples d’ODD (3, 6 et 13), qui sont étroitement liés au système alimentaire de la Suisse, nous allons maintenant présenter brièvement les domaines dans lesquels la Suisse a encore du retard à rattraper en ce qui concerne la réalisation des objectifs de durabilité sociaux, économiques et écologiques. Un aperçu complet des 17 ODD, des indicateurs correspondants et de leur état actuel en Suisse peut être consulté dans l’annexe statistique du rapport national 2022 de la Suisse.29 Au chapitre 2, nous présenterons en outre une analyse plus détaillée de l’état actuel, afin de rendre mesurables les besoins d’action et les progrès réalisés. 1.2.1 ODD 3 : Permettre à tous de vivre en bonne santé et promouvoir le bien-être de tous à tout âge L’ODD 3 traite de l’amélioration de la santé et du bien-être des personnes. La malnutrition est la première cause de mortalité dans le monde. Les maladies non transmissibles (MNT) représentent 80% de nos coûts de santé et sont en grande partie liées à l’alimentation.30 Dans les pays industrialisés, c’est surtout l’indicateur « surpoids » qui représente une mesure importante pour un système alimentaire favorable à la santé. En Suisse, environ 42% de la population est concernée par l’obésité. Ce pourcentage a augmenté de 4,7 points depuis la fin des années 1990 et ne devrait pas diminuer.28 L’obésité augmente le risque de maladies cardio-vasculaires et métaboliques. L’obésité est une forme particulièrement grave de surpoids. Le pourcentage de la population souffrant d’obésité a augmenté de 3,6 points depuis la fin des années 1990. La prévention du surpoids et de l’obésité permet d’éviter des décès prématurés et d’économiser des frais de santé élevés, ce qui s’avère durable non seulement sur le plan social, mais aussi sur le plan économique. Surpoids et obésité Population de 15 ans et plus vivant en ménage privé Hommes Femmes 60% 60% 50% 50% 40% 6.1 6.7 7.9 8.6 11.2 12.3 40% 30% 30% 20% 33.2 35.6 37.7 38.1 39.3 38.7 10% 0% 1992 1997 Surpoids 2002 2007 2012 2017 7.4 7.8 21.2 21.9 20.9 1997 2002 2007 7.0 20% 4.7 10% 17.2 0% 1992 9.4 10.2 22.6 22.8 2012 2017 Obésité Graphique 1 : OFS (2020)31 1.2.2 ODD 6 : Garantir l’accès de tous à l’eau et à l’assainissement et assurer une gestion durable des ressources en eau Outre l’hygiène et l’assainissement, l’ODD 6 comprend également l’accès à l’eau potable ainsi que la protection et la restauration des écosystèmes liés à l’eau. Notre système alimentaire, en particulier la production agricole, a une grande influence sur ces écosystèmes : d’une part, il sollicite des ressources en eau en Suisse et dans le reste du monde (par le biais des importations) ; d’autre part, il les impacte par ses outputs. Par exemple, les nitrates se retrouvent dans l’eau en grande partie à cause de l’utilisation d’engrais azotés et d’un nombre élevé d’animaux par surface dans l’agriculture. Le taux de nitrate est un bon indicateur de la qualité de l’eau des nappes phréatiques suisses. Des niveaux élevés de nitrates indiquent une faible qualité de l’eau (potable) et peuvent présenter un risque pour la santé humaine. En Suisse, des concentrations supérieures à 25 mg/l ont été détectées dans environ 15% des stations de mesure. Ces concentrations dépassent la valeur limite de l’ordonnance sur la protection des eaux (OEaux). Dans les régions où les cultures sont nombreuses, cette valeur a même été dépassée dans près de 40% des stations de mesure.32 En Suisse, la proportion de nitrates dans les eaux souterraines reste depuis 2002 à un niveau particulièrement élevé par rapport au reste du monde. Ce niveau dépasse nettement les limites planétaires.28,33 Les phosphates aussi se retrouvent dans les eaux souterraines à cause de la production agricole. Des 13 taux de phosphate trop élevés peuvent entraîner une eutrophisation. Cela signifie que l’enrichissement en nutriments modifie ou réduit le fonctionnement de l’écosystème. Globalement, la part des stations de mesure de l’observatoire national des eaux souterraines NAQUA qui ne répondent pas aux exigences de la protection des eaux est en hausse depuis 2017. Une meilleure qualité de l’eau contribue non seulement au maintien d’écosystèmes locaux et mondiaux sains, mais garantit également la santé humaine et permet de réaliser des économies considérables sur le traitement de l’eau potable ainsi que sur « l’aération artificielle » des eaux en Suisse. serre réside dans l’adaptation des modèles de consommation et de production.36 Ces adaptations permettraient non seulement de réduire les risques existentiels liés au changement climatique, mais aussi de diminuer les coûts élevés d’adaptation du secteur agroalimentaire suisse au changement climatique. Empreinte gaz à effet de serre de l’alimentation en millions de tonnes d’équivalent CO2 25000 20000 Nitrates dans les eaux souterraines 15000 20% 10000 15% 5000 10% 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Inland Suisse 5% Limite du système : consommation de denrées alimentaires par les résident·e·s suisses dans le pays et à l’étranger, à domicile et hors domicile 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 Inland 2008 0% Part des stations de mesure NAQUA pour lesquelles l’exigence chiffrée définie par l’ordonnance sur la protection des eaux est dépassée Graphique 2 : OFEV (2022)34 1.2.3 ODD 13 : Prendre d’urgence des mesures pour lutter contre les changements climatiques et leurs répercussions L’ODD 13 se concentre sur le renforcement des mesures de protection du climat ainsi que sur la promotion de la résilience face aux catastrophes naturelles liées au climat. La Suisse s’est engagée à lutter contre le changement climatique en signant non seulement l’Agenda 2030, mais aussi l’accord de Paris sur le climat. La crise climatique qui s’aggrave actuellement rapidement à l’échelle mondiale menace l’existence des humains et de la nature. Le système alimentaire contribue à l’accélération de la crise climatique surtout par l’émission de méthane, notamment à travers l’élevage et la production de riz. Gaz à effet de serre particulièrement néfaste pour le climat à court terme, le méthane augmente le risque de franchir des points de basculement climatiques dangereux.35 Deux tiers de l’empreinte carbone des aliments consommés en Suisse sont produits à l’étranger (voir graphique 3). En matière de politique climatique, la Suisse a donc une responsabilité non seulement au niveau national, mais aussi au niveau international. En même temps, le potentiel de réduction des émissions de gaz à effet de serre par la transformation du système alimentaire est très important : selon la stratégie climatique pour l’agriculture et l’alimentation, les émissions dans l’agriculture peuvent être réduites jusqu’à deux tiers d’ici 2050.36 Selon des études de l’OFEV, le plus grand potentiel de réduction des gaz à effet de 14 Ausland Étranger 25000 Ausland Total Graphique 3 : OFEV / EBP (2022) 37 20000 1.2.4 La réalisation des ODD en Suisse est un défi majeur 15000 10000 Comme le montrent les exemples d’évolutions relatives à l’ODD 3, à l’ODD 6 et à l’ODD 13, il est prévisible que la Suisse ne parviendra pas à atteindre la plupart des ODD d’ici à 2030. 5000 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 non 2011 2012 2013 2014 2015 en 2016 raison 2017 2018 La pression pour agir augmente donc seulement des conséquences existentielles du changement climatique et de la perte de la biodiversité, mais aussi en raison des grands défis de la durabilité sociale, sanitaire et économique. La transformation du système alimentaire place les différents acteurs et décisionnaires politiques du système alimentaire suisse face à des tâches importantes. Par ailleurs, ces défis sociétaux et politiques impliquent souvent des conflits d’objectifs, d’intérêts et de valeurs. La pression exercée sur le secteur agroalimentaire pour qu’il améliore sa durabilité écologique tout en garantissant des revenus suffisants, des prix avantageux et la sécurité alimentaire ne cesse de croître.38 Certes, il existe à moyen terme d’importants potentiels de synergie dans la réalisation commune de différents objectifs de durabilité (p. ex. entre une alimentation équilibrée, des objectifs écologiques et la sécurité de l’approvisionnement, mais aussi entre la protection du climat et la sécurité économique, notamment de l’agriculture).39 Néanmoins, les conflits d’objectifs, d’intérêts et de valeurs doivent être pris au sérieux. Une pondération compréhensible des différents objectifs, intérêts et valeurs est donc, du point de vue du comité scientifique, indispensable à la réussite du processus de transformation (voir chapitre 2). 15 1.3 Points de basculement positifs dans la transformation du système alimentaire % de la transformation réalisée du système alimentaire suisse en accord avec les objectifs de développement durable (ODD) Afin d’aborder et de minimiser les conflits d’objectifs, d’intérêts et de valeurs de manière constructive, le comité scientifique estime qu’il est nécessaire de définir une trajectoire de transformation politique basée sur une perspective globale du système alimentaire (voir chapitre 3). Cette trajectoire doit prendre en compte les interactions entre les acteurs et leurs actions tout au long de la chaîne de valeur et à tous les niveaux (communal, cantonal, national et international). Une telle trajectoire de transformation politique peut déclencher de manière ciblée des dynamiques de points de basculement positives (voir encadré p. 16) et accélérer ainsi la transformation du système alimentaire en accord avec les ODD. Du point de vue du comité scientifique, les stratégies et mesures actuelles concernant le système alimentaire suisse forment un puzzle dont les pièces ne s’emboîtent pas assez bien. Ce point de vue est également partagé par le rapport « L’orientation future de la politique agricole »12 du Conseil fédéral. Ce dernier y souligne que ce puzzle résulte de l’introduction constante de nouvelles mesures, tandis que les instruments inefficaces ne sont pas supprimés, ce qui donne un système toujours plus complexe. Actuellement, on recense par exemple différents documents stratégiques de l’Office fédéral de la santé publique (OFSP) sur l’alimentation qui se chevauchent dans le temps, la stratégie alimentaire de l’Office Défi: Réduire les barrières pour le déclenchement de points de basculement positifs Points de basculement Durée Mobiliser Accélérer Stabiliser Fig. 3 : Fesenfeld et al. (2023) 237 / Telek POINTS DE BASCULEMENT POSITIFS DANS LA TRANSFORMATION DU SYSTÈME ALIMENTAIRE Le système alimentaire actuel est confronté à des défis majeurs, tant au niveau mondial qu’au niveau national. Pour réussir la transformation vers un système alimentaire plus durable, des ajustements sont nécessaires tout au long de la chaîne de valeur, de la production à la consommation en passant par la transformation et la distribution.1,2,11 En raison des liens étroits et des interdépendances entre tous les acteurs du système alimentaire, les modifications d’une activité ont des répercussions sur les autres activités et acteurs du système. La transformation du système alimentaire doit donc suivre une approche systémique. Les « points de basculement positifs » jouent un rôle important à cet égard. Contrairement aux points de basculement négatifs de l’écosystème (tels que la perte de l’Amazonie)3, les points de basculement positifs peuvent être définis comme des points critiques dans un système socio-technique et socio-naturel, 16 où des interventions ciblées peuvent avoir un impact important, positif et à long terme sur le développement durable du système et modifier profondément son fonctionnement.40, 142 Les interventions économiques, politiques ou techniques peuvent favoriser l’émergence de boucles de rétroaction positives.40 Par exemple, les innovations technologiques dans le domaine des substituts animaux et de l’agrophotovoltaïque peuvent rendre attractifs de nouveaux modèles commerciaux et de nouvelles sources de revenus pour ainsi accélérer le changement des normes et des comportements.40 Cela peut ensuite faciliter, avec le temps, l’adoption politique de mesures de transformation plus ambitieuses, car un plus grand nombre d’acteurs du système reconnaissent les avantages de la transformation et la société oppose moins de résistance à cet égard. En accélérant les processus de transformation et en contribuant à réduire les conflits d’objectifs et d’intérêts dans le système existant (voir plus de détails au chapitre 3), de telles dynamiques de points de basculement positives peuvent donc jouer un rôle très important pour la transformation du système alimentaire. fédéral de la sécurité alimentaire et des affaires vétérinaires (OSAV) avec le plan d’action correspondant41 en collaboration avec l’OFEV, des stratégies sectorielles à plus long terme de l’OFEV42 et la stratégie des sols élaborée par l’OFEV43. Tous traitent les mêmes indicateurs cibles sous des angles différents, au lieu de les intégrer dans une stratégie de système alimentaire globale. Le rapport « L’orientation future de la politique agricole 2022 », publié en juin 2022, est un premier pas vers une perspective plus globale du système alimentaire, impliquant tous les acteurs de la chaîne de valeur12. Une telle perspective regroupée du système alimentaire permettra de faire avancer la transformation durable du système alimentaire suisse de manière plus efficace et efficiente. En outre, une réorientation globale de la politique du système alimentaire permettra d’exploiter au mieux les synergies positives des mesures prises tout au long de la chaîne de valeur. Enfin, du point de vue du comité scientifique, la transformation du système alimentaire suisse présuppose un changement des structures institutionnelles du pays. Des ajustements sont donc nécessaires dans ce que l’on appelle la gouvernance du système alimentaire, c’est-à-dire la manière dont les différents acteurs du système alimentaire suisse collaborent et négocient. Ce n’est qu’en mettant l’accent sur une politique globale du système alimentaire, et non plus uniquement sur la politique agricole, que de nouvelles coalitions politiques pourront être formées et que des compromis pourront être trouvés. Une réorientation stratégique de la politique du système alimentaire implique donc des changements dans la coopération entre les principaux acteurs du système pour lutter contre l’immobilisme politique et la polarisation (voir chapitre 4). Dans le chapitre 2 du présent guide, le comité scientifique définit un modèle pour un système alimentaire suisse durable et les indicateurs cibles mesurables correspondants. Ces derniers sont ensuite mis en relation avec l’état actuel afin de démontrer de manière mesurable la nécessité d’agir. Le chapitre 3 esquisse ensuite une possible trajectoire de transformation politique qui permettrait d’atteindre l’objectif ambitieux du chapitre 2 et de transformer le système alimentaire de manière efficace et réalisable du point de vue du comité scientifique. Enfin, le chapitre 4 aborde le rôle central de la gouvernance du système alimentaire dans le processus de transformation. 17 18 2 MODÈLE ET OBJECTIFS CENTRAUX POUR LA TRANSFORMATION DU SYSTÈME ALIMENTAIRE SUISSE 2.1 Conflits d’objectifs et synergies potentielles Auteur·e·s chapitre 2 : Lukas Fesenfeld, Matthias Meier, Thomas Nemecek, Claudio Beretta, Christine Brombach, Philippe Jeanneret, Stefan Mann, Stephan Pfister, Sonja Schönberg La transformation du système alimentaire pose des défis majeurs aux décideurs et décideuses politiques et aux acteurs et actrices du système alimentaire. La pression exercée sur le secteur agroalimentaire pour qu’il améliore sa durabilité écologique tout en garantissant des revenus suffisants, des prix avantageux et la sécurité alimentaire ne cesse de croître.38 Les Objectifs mondiaux de développement durable (ODD) représentent pour la Suisse le cadre d’objectifs international pour l’orientation du système alimentaire. Au niveau national, les articles 54 et 104 de la Constitution fédérale, par exemple, explicitent à nouveau ces différents objectifs. D’une part, la Confédération doit contribuer à soulager les populations dans le besoin et à lutter contre la pauvreté dans le monde, à faire respecter les droits de l’homme et à promouvoir la démocratie, à favoriser la coexistence pacifique des peuples et à préserver les ressources naturelles (article 54). D’autre part, la Confédération doit veiller à ce que l’agriculture, par une production durable et orientée vers le marché, contribue substantiellement à la sécurité de l’approvisionnement de la population, à la conservation des ressources naturelles et à l’entretien du paysage rural ainsi qu’à l’occupation décentralisée du territoire (article 104). À court terme, il existe des conflits d’objectifs entre les dimensions écologique, sociale et économique de la durabilité. Par exemple, dans son rapport sur la « Future orientation de la politique agricole », le Conseil fédéral souligne que des exigences écologiques plus élevées signifient à court terme des coûts supplémentaires pour la production agricole et que, si les modèles alimentaires restent les mêmes, les objectifs écologiques et un taux d’autosuffisance élevé sont difficilement conciliables.12 Cependant, à moyen et long terme, il existe un grand potentiel de synergie dans la réalisation commune de différents objectifs de durabilité. Il existe par exemple des interactions positives entre une alimentation équilibrée, des objectifs écologiques et la sécurité de l’approvisionnement. Ainsi, une réduction de la consommation de viande peut réduire considérablement l’empreinte écologique, avoir un effet positif sur la santé et libérer pour la production d’aliments végétaux des surfaces jusqu’ici utilisées pour la production d’aliments pour animaux. Cela peut augmenter la production nationale de calories et la sécurité de l’approvisionnement.12 Malgré d’importants potentiels de synergie à moyen et long terme, les conflits d’objectifs doivent être pris au sérieux. Le comité scientifique estime donc qu’une pondération compréhensible des différents objectifs est indispensable à la réussite du processus de transformation. Sur la base du « SDG Wedding Cake Model »44 (voir figure 4), une biosphère saine (ODD 6, 13, 14, 15) constitue la base existentielle pour la réalisation des objectifs de durabilité sociale (ODD 1, 2, 3, 4, 5, 7, 11, 16) et économique (8, 9, 10, 12). En d’autres termes, la réalisation d’objectifs de durabilité sociale et économique présuppose donc durablement la réalisation d’objectifs écologiques. Ce principe devrait être respecté lors de l’élaboration des politiques et de la mise en œuvre des objectifs et des trajectoires de transformation. En outre, lors de la pondération relative des différents objectifs, il convient de prendre en compte l’ampleur du dépassement actuel des objectifs. Ce principe est également illustré par le concept de limites planétaires45 ainsi que celui des unités de charge écologique36 et pourrait être appliqué à d’autres indicateurs cibles. Par exemple, pour chaque indicateur cible, le rapport entre la charge actuelle et la valeur cible pourrait être élevé au carré, ce qui permettrait de calculer une pondération relative des indicateurs cibles. Il en résulterait que les indicateurs cibles qui s’écartent particulièrement de la valeur cible seraient davantage pondérés et bénéficieraient d’une priorité plus élevée dans le processus de transformation que ceux qui s’écartent moins de la valeur cible. Dans ce qui suit, le comité scientifique présente un projet de modèle possible pour la transformation du système alimentaire suisse et propose des indicateurs cibles mesurables pour atteindre les différents objectifs de durabilité dans le système alimentaire. Une pondération relative de ces indicateurs cibles n’a volontairement pas été effectuée ici. Bien qu’il existe des dépendances objectives entre les objectifs (voir Wedding Cake Model) et des différences mesurables dans la réalisation des objectifs, la pondération relative des indicateurs d’objectifs est fondamentalement une tâche politique. Sur la base du cadre DPSIR de l’OCDE (abréviation de Driving forces, Pressures, States, Impacts and Responses), nous recommandons de mettre un accent particulier sur l’adressage des indicateurs cibles du côté de la consommation (voir ci-dessous, entre autres, l’indicateur cible 1), car ceux-ci conditionnent la réalisation de la plupart des autres indicateurs cibles. Du point de vue de la consommation, on prend en compte tous les produits et émissions qui sont produits et consommés au sein de la chaîne d’approvisionnement suisse. Du point de vue du comité scientifique, les conflits d’objectifs des ODD ainsi que l’empreinte globale du système alimentaire suisse (c’est-à-dire incluant toutes les externalités importées) ne peuvent être minimisés de manière globale que si les effets de la consommation intérieure sur l’ensemble de la chaîne de valeur sont suffisamment pris en compte. ÉCONOMIE SOCIÉTÉ BIOSPHÈRE Fig. 4 : Rockström & Sukhdev (2016)44 19 2.2 Un modèle pour le développement du système alimentaire suisse en accord avec les ODD Nous esquissons ci-après un modèle pour le système alimentaire que nous souhaitons mettre en place à l’avenir et qui doit permettre d’atteindre au mieux les objectifs mondiaux de développement durable (ODD) dans le domaine de l’alimentation. Il s’agit d’assurer à la population un approvisionnement aussi optimal que possible en denrées alimentaires et en nutriments, de garantir un mode d’alimentation sain à des prix abordables et de générer des revenus adéquats dans le secteur agroalimentaire, tout en garantissant des conditions de travail équitables. Les effets négatifs sur l’environnement et le développement durable en Suisse et à l’étranger doivent être réduits de manière à ne pas dépasser la capacité de charge des écosystèmes et à préserver leurs fonctions. Les ressources renouvelables ne doivent être utilisées que dans le cadre de leur taux de renouvellement et l’utilisation de ressources non renouvelables doit être évitée dans la mesure du possible. La sécurité d’approvisionnement et la résilience doivent rester garanties face à des risques croissants tels que le changement climatique, les ruptures de la chaîne d’approvisionnement ou les pandémies. De bonnes conditions de travail et de bons revenus tout au long de la chaîne de valeur contribuent également à la sécurité de l’approvisionnement. La transition vers le système visé est urgente si l’on veut atteindre les objectifs de durabilité et préserver les écosystèmes mondiaux en tant que moyens de subsistance. Cela est particulièrement important pour la sécurité alimentaire dans un pays mondialement interconnecté comme la Suisse, qui dépend de l’importation de ressources naturelles importantes. Le processus de transformation exigera des ressources et doit être organisé de manière socialement acceptable. Toutefois, du point de vue du comité scientifique, les avantages et les opportunités de la transition vers le système visé l’emportent clairement sur les coûts et les risques de cette transition. La figure 5 permet de visualiser le modèle du système alimentaire durable. Les caractéristiques du système d’objectifs sont présentées ci-dessous. 2.2.1 Comportement de consommation et modèles alimentaires Les adaptations de l’alimentation et de la consommation sont une condition impérative pour la réalisation des ODD, car ces objectifs ne peuvent pas être atteints uniquement par des améliorations de la production. Pour réduire l’impact environnemental global à un niveau écologiquement supportable, un changement marqué de la consommation alimentaire est nécessaire. Cela signifie une alimentation principalement à base de plantes (flexitarienne), qui s’inspire des recommandations alimentaires nationales et du régime alimentaire Planetary Health Diet. Cette alimentation s’accompagne d’une forte réduction de la consommation de viande par rapport aux modes de consommation actuels en Suisse, d’une réduction significative de la consommation de produits laitiers et d’œufs, ainsi que d’une diminution de la consommation excessive de boissons alcoolisées, de sucre et de café. Par contre, la consommation de légumes, de légumineuses, de noix, de céréales et de fruits doit être augmentée. Le comportement de consommation et l’alimentation doivent être adaptés aux besoins nu20 tritionnels individuels, afin d’éviter le gaspillage alimentaire et une consommation supérieure aux besoins d’une alimentation saine. Les légumes et les fruits consommés devraient, dans la mesure du possible, provenir d’une production régionale, saisonnière et adaptée aux conditions locales, sans transport aérien ni serres chauffées. Les produits moins transformés et contenant moins d’additifs sont par ailleurs à privilégier. 2.2.2 Agriculture et secteur alimentaire La production de denrées alimentaires est prioritaire sur les terres arables. La culture d’aliments pour animaux sur les terres agricoles est minimisée et n’a lieu que dans la mesure où elle est optimale du point de vue de la rotation des cultures ou qu’elle n’entre pas en concurrence avec la production de denrées alimentaires (p. ex. fourrage intermédiaire entre deux cultures de denrées alimentaires). Les prairies permanentes qui, pour des raisons agronomiques ou écologiques, ne se prêtent pas aux grandes cultures, sont utilisées en priorité pour la production laitière. Pour compléter le fourrage des prairies et des pâturages, on utilise principalement des sous-produits de la production de denrées alimentaires qui ne peuvent pas être utilisés pour l’alimentation humaine directe. La viande bovine est un coproduit de la production laitière. L’élevage de volailles et de porcs est réduit et l’alimentation se fait en priorité avec des sous-produits de la transformation de denrées alimentaires. Les effectifs d’animaux sont globalement réduits, les animaux sont nourris autant que possible avec des aliments d’origine nationale et la production est organisée de manière à respecter le bien-être des animaux. La réduction du nombre d’animaux permet de libérer des surfaces de prairies qui sont exploitées de manière extensive ou qui ne le sont plus du tout, afin de promouvoir la biodiversité. L’agriculture produit de manière adaptée aux conditions locales. Cela signifie que l’on privilégie à chaque fois l’utilisation optimale des ressources naturelles d’un point de vue agronomique et écologique (compte tenu de la capacité de charge des écosystèmes) et du point de vue de l’approvisionnement alimentaire. La production de denrées alimentaires est efficace et économe en ressources, elle génère donc relativement plus avec moins d’intrants et entraîne moins d’impacts sur l’environnement. La production de denrées alimentaires repose sur des principes agroécologiques qui aident à garantir les rendements tout en préservant les fonctions des écosystèmes. Les technologies et les systèmes de production qui contribuent aux ODD et qui sont sûrs sont utilisés. De plus, la production doit devenir indépendante des sources d’énergie fossiles. L’utilisation des sols organiques est adaptée de manière à stopper la dégradation de l’humus et à préserver ou à développer l’humus sur les sols minéraux. L’économie circulaire est renforcée, en particulier dans le domaine des nutriments, et les moyens auxiliaires agricoles tels que les engrais minéraux, les pesticides et l’irrigation sont utilisés de manière ciblée en fonction des besoins. La ressource biomasse est valorisée de manière optimale par une utilisation en cascade selon les priorités suivantes : 1. Alimentation humaine, 2. Alimentation animale, 3. Utilisation matérielle (p. ex. fibres, engrais), 4. Utilisation énergétique. Les flux secondaires doivent également être utilisés conformément à ces priorités (p. ex. au moyen d’approches telles que « nose to tail » ou « leaf to root »). Les pertes de denrées alimentaires sont minimisées tout au long de la chaîne de valeur. Les revenus et les conditions de travail sur l’ensemble des chaînes de valeur, en Suisse et à l’étranger, sont équitables et favorables à la santé et garantissent le respect des droits de l’homme. La transparence en matière de durabilité est assurée tout au long de la chaîne de valeur. Des aliments de qualité et savoureux sont produits à des prix raisonnables. Les coûts réels se reflètent dans les prix des denrées alimentaires et les revenus de la population suisse permettent de payer ces prix raisonnables. Contribution – production – transformation • agroécologie • adapté au site et économe en ressources • ouvert à la technologie Valorisation – élimination Réduction du gaspillage alimentaire tout au long de la chaîne de valeur Commerce Modèle pour un système alimentaire durable • transparence • revenu équitable • prix socialement acceptables tout au long de la chaîne de valeur Consommation – transformation • alimentation flexitarienne, à base de plantes (Planetary Health Diet) • produits saisonniers Fig. 5 : Fesenfeld et al. (2023)237 / Telek 2.3 Indicateurs cibles mesurables pour la réalisation des ODD Dans ce qui suit, nous déduisons du cadre d’objectifs global des ODD et du modèle pour le développement futur du système alimentaire suisse onze indicateurs cibles mesurables et prioritaires. Les définitions et les justifications de ces indicateurs cibles ont été élaborées par les membres du comité scientifique dans le cadre d’un processus Delphi itératif (voir l’appendice en ligne). Les indicateurs cibles ont été formulés de manière suffisamment ambitieuse, conformément aux ODD, et ont été choisis, dans la mesure du possible, sur la base des connaissances scientifiques actuelles. Certains indicateurs cibles reflètent des valeurs cibles déjà officiellement en vigueur en Suisse, mais d’autres, définis par le comité scientifique, vont au-delà des indicateurs cibles existants. Cela se justifie par le fait que, du point de vue du comité scientifique et compte tenu de l’état actuel de la recherche, les valeurs cibles convenues au niveau national sont en partie insuffisantes pour transformer le système alimentaire suisse conformément aux ODD. Ce choix d’indicateurs et d’objectifs est bien entendu entaché d’incertitudes et ne peut pas être déterminé de manière définitive dans le cadre du présent guide. Le processus largement soutenu du comité scientifique permet toutefois de garantir que les indicateurs couvrent les dimensions centrales du système pour la réalisation des ODD et qu’il existe une forte probabilité que la réalisation des objectifs en lien avec ces indicateurs génère également des effets d’aubaine dans d’autres domaines non mesurés. Dans ce qui suit, nous définissons et justifions onze objectifs mesurables qui, d’un point de vue scientifique, devraient être atteints en priorité pour réaliser les ODD. Ensuite, nous mettons brièvement en relation ces onze objectifs avec l’état actuel de leur réalisation. Cette analyse illustrative du statu quo permet de se faire une idée des besoins d’action actuels. Elle offre également aux décideurs politiques une base leur 21 permettant de pondérer les objectifs centraux, d’identifier les synergies potentielles et de fixer des priorités fondées sur des données probantes lors de l’élaboration des mesures (voir section 2.1. Conflits d’objectifs et synergies potentielles). Objectif 1 : Adaptation de l’alimentation conformément aux recommandations nationales en matière de nutrition et au régime alimentaire Planetary Health Diet Les modes d’alimentation de la population résidant en permanence en Suisse s’orientent vers des recommandations nutritionnelles qui favorisent la santé humaine, garantissent la capacité de régénération des ressources naturelles au niveau local et global, sont compatibles sur le plan socio-économique et tiennent compte du bien-être des animaux. Dans le cadre de l’objectif 1, l’accent est mis sur l’adoption d’un mode d’alimentation plus sain et plus respectueux de l’environnement. Cela suggère une orientation vers les recommandations nutritionnelles nationales46, les conclusions d’Ow et al.47 ainsi que celles d’Ernstoff et al.48 et le Planetary Health Diet de Willett et al.49. Les analyses de données actuelles sur l’alimentation respectueuse de l’environnement et des ressources en Suisse, réalisées par Zimmermann et al.50 ainsi que par Jungbluth et al.51, permettent en outre de résumer les adaptations nécessaires par rapport aux modèles alimentaires actuels de la population adulte en Suisse, tels qu’identifiés dans l’enquête nationale sur la nutrition menuCH52 et les bulletins nutritionnels suisses 2019 et 202153 : – Une augmentation significative de la consommation de légumes et un maintien de la consommation de fruits issus de la production locale et saisonnière. – Une augmentation et une diversification de la consommation de céréales complètes, de pommes de terre, de légumineuses et de noix. – Une réduction de la consommation de produits laitiers transformés et une préférence pour le lait non transformé, ainsi que le remplacement des produits laitiers par des alternatives végétales avec ajout de calcium. – Une réduction de la consommation d’huiles alimentaires et une utilisation ciblée d’huiles appropriées pour les préparations à haute température telles que le rôtissage/grillage. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Une alimentation saine et durable telle que définie par la FAO et l’OMS favorise la santé et le bien-être (ODD 3 – Bonne santé et bien-être), réduit l’impact environnemental (ODD 13 – Climat, ODD 14 – Vie aquatique, ODD 15 – Vie terrestre) et est abordable, sûre, produite de manière équitable et culturellement acceptable (ODD 1 – Pas de pauvreté, ODD 2 – Faim « zéro », ODD 8 – Travail décent, ODD 12 – Consommation et production durables).55 Différents modèles et différentes stratégies56–59 tentent de répondre à ces exigences. C’est pourquoi l’accent a été récemment déplacé de la couverture des besoins nutritionnels vers les différentes dimensions d’une alimentation durable. 60 Dans son rapport pour 2019, la commission EAT-Lancet a publié un régime de référence concret et adapté à l’avenir : le régime Planetary Health Diet61 souligne la grande importance d’une alimentation principalement basée sur les plantes pour respecter les limites planétaires tout en nourrissant la population mondiale de manière saine et adaptée à ses besoins. Le régime Planetary Health Diet présente des similitudes avec les recommandations nutritionnelles de la Société Suisse de Nutrition.46 De plus, les recommandations nutritionnelles actuelles de la Société Suisse de Nutrition sont en cours de révision, en accord avec les recommandations du Planetary Health Diet. Les valeurs cibles ci-dessus s’orientent donc en particulier vers les recommandations nutritionnelles, le Planetary Health Diet61 et une alimentation de la population suisse optimisant les ressources,47 afin de permettre la réalisation des ODD d’un point de vue scientifique. L’alimentation actuelle des adultes en Suisse s’écarte fortement de ces recommandations scientifiques en matière de nutrition. L’enquête nationale sur la nutrition montre que les habitudes alimentaires actuelles dépassent, par exemple, les recommandations en matière de consommation d’aliments d’origine animale et sont nettement inférieures à celles concernant les aliments d’origine végétale.46, 52 La seule combinaison de ces deux éléments réduit en même temps de manière significative la compatibilité des régimes alimentaires actuels avec la santé et l’environnement en Suisse. Justification des valeurs cibles spécifiques par habitant pour la consommation Les valeurs cibles spécifiques par habitant pour la consommation se justifient comme suit : – Une forte réduction de la consommation de viande. – Une forte réduction de la consommation d’alcool et de boissons sucrées. La quantification des recommandations nutritionnelles à l’horizon 2030 et 2040 en vue de la réalisation des ODD et de l’adoption de modes d’alimentation sains pour les adultes en bonne santé en Suisse s’appuie sur les modèles alimentaires actuels, tels qu’identifiés dans l’enquête nationale sur la nutrition menuCH et les données du bilan alimentaire Agristat.54 Le comité scientifique propose une approche en deux étapes, qui tente de réduire le champ de tension entre les recommandations alimentaires et les habitudes alimentaires et qui devrait évoluer au fur et à mesure de l’acquisition de connaissances. 22 a. Céréales complètes et produits dérivés : 230 g/jour Pour des raisons de santé, la part de produits à base de céréales complètes devrait être nettement augmentée.48 Cela permettrait en outre de réduire les pertes alimentaires et de contribuer ainsi à l’ODD 12. Les pâtes et le riz devraient être réduits au profit de plus de produits à base d’orge ou de maïs. b. Légumes féculents (p. ex. pommes de terre) : 100 g/jour La Société Suisse de Nutrition (SSN) recommande de consommer trois portions de féculents par jour, ce qui peut se faire notamment avec des pommes de terre non transformées. Une portion de pommes de terre correspond à 180 à 300 g. Si une portion de pommes de terre non transformées est consommée trois fois par semaine, cela correspond à une quantité quotidienne d’environ 100 g, dont la consommation peut être recommandée. Objectifs pour une alimentation adaptée en accord avec les ODD Groupe alimentaire Consommation recommandée par jour et par personne Consommation recommandée par jour et par personne 2030 2040 Glucides en grammes Céréales complètes et produits dérivés 230 230 Légumes féculents (p. ex. pommes de terre) 100 100 Légumes 250 360 Fruits 240 240 Viande < 50 < 40 Œufs 15 15 Poisson < 20 < 20 Légumineuses 35 75 Noix 30 50 Produits laitiers (lait entier ou produits fabriqués à partir de cette quantité) < 350 < 350 Sucre ajouté < 30 < 30 Source de protéines en grammes Tableau 1 : Fesenfeld et al. (2023)237 / Telek c. Fruits 240 g/jour (2030) et 240 g/jour (2040) et légumes 250 g/jour (2030) et 360 g/jour (2040) Les données des bulletins alimentaires suisses de 2019 et 2021 ainsi que les données du bilan Agristat de 2020 indiquent que la consommation de légumes, en particulier, est nettement trop faible d’un point de vue nutritionnel et que les ODD 2 et 3 sont donc menacés. Il faudrait planifier de manière ciblée une augmentation de la consommation de légumes au sein de la population, tout en minimisant les risques liés à certains impacts environnementaux (p. ex. utilisation de l’eau, pollution des cours d’eau)22 afin de répondre aux ODD 6 et 14. d. Viande 50 g/jour (2030) ou 40 g/jour (2040) maximum Avec environ 800 g/semaine41, 62, les habitudes alimentaires suisses dépassent presque trois fois les recommandations actuelles de la Société Suisse de Nutrition en matière de consommation de viande.46 La viande transformée représente environ 40% de la consommation de viande et contribue au développement de maladies non transmissibles. Ainsi, le niveau actuel de la consommation de viande en Suisse n’est pas en accord avec l’ODD 3 (santé). La viande est en outre l’une des principales sources de pollution du système alimentaire. Les émissions de méthane, particulièrement nocives pour le climat et dues à la production de denrées alimentaires d’origine animale, augmentent notamment les risques de franchir des points de basculement climatiques dangereux et d’accélérer ainsi le changement climatique (ODD 13).35, 63 De plus, le niveau actuel de consommation de viande contribue de manière significative à la déforestation mondiale et à la perte de biodiversité (ODD 15) ainsi qu’à la surexploitation et à la pollution de l’eau (ODD 14 et ODD 6).2 Certes, certains avantages liés à la localisation d’un élevage basé sur les herbages en Suisse peuvent avoir des effets positifs sur l’environnement (p. ex. sur la biodiversité locale), mais d’un point de vue scientifique, il est indéniable que la consommation nationale de produits animaux en Suisse doit baisser de manière significative pour atteindre les ODD.47, 64 e. Œufs 15 g/jour Cette quantité correspond à un peu moins de deux œufs par semaine. Le régime Planetary Health Diet prévoit une quantité similaire d’œufs et met l’accent sur le remplacement des protéines d’origine animale par des protéines d’origine végétale. f. Poisson maximum 20 g/jour La consommation de poisson ne devrait pas augmenter de manière significative par rapport à aujourd’hui, afin de réduire la surexploitation des stocks sauvages ainsi que l’impact environnemental de la pisciculture, cf. ODD 4 et 12. D’un point de vue nutritionnel, les acides gras polyinsaturés EPA et DHA contenus dans le poisson de mer devraient être remplacés par des suppléments. g. Légumes secs (poids sec) 35 g/jour (2030) ou 75 g/jour (2040) En Suisse, la consommation moyenne de légumineuses est de 5 g/jour. Actuellement, les recommandations alimentaires suisses ne mentionnent pas explicitement les légumineuses comme une source importante de protéines végétales, mais comme des féculents. Cependant, l’importance des légumineuses en tant que substitut végétal aux sources de protéines animales et en tant que source importante de fibres alimentaires devrait être davantage soulignée pour travailler à la réalisation des ODD 2, 3 et 12. h. Noix et graines 30 g/jour (2030) ou 50 g/jour (2040) En Suisse, la consommation moyenne de noix, de graines, d’olives et d’avocats est de 10 g/jour au total. La consommation de noix et de graines devrait à elle seule être augmentée à 30-50 g pour que l’on puisse exploiter leur potentiel en tant que source végétale de protéines, de fibres alimentaires et d’acides gras polyinsaturés, et ainsi viser les ODD 2 et 3. i. Lait ou produits laitiers 350 g/jour maximum (2030) En Suisse, les personnes âgées de 18 à 75 ans consomment en moyenne, par personne et par jour, 619 g de lait ou de produits fabriqués à partir de cette quantité de lait, comme le fromage et le yaourt. Comme plus de la moitié de la surface agricole utile en Suisse est constituée de prairies, la production laitière représente une forme d’exploitation importante pour couvrir les besoins en protéines de la population. En même temps, la consommation de ce groupe d’aliments dans le pays doit être réduite afin d’atteindre les objectifs écologiques (notamment l’ODD 13 – Climat ainsi que l’ODD 15 – Vie terrestre).22 Les caractéristiques topographiques et agronomiques de l’agriculture suisse conduisent à une recommandation pour le lait et les produits laitiers qui est supérieure à la quantité du Planetary Health Diet et inférieure à celle des recommandations alimentaires actuelles pour la Suisse. Une réduction de la production laitière doit être accompagnée d’une planification ciblée en ce qui concerne les exportations de produits laitiers ainsi que d’une gestion plus extensive des surfaces de prairies afin d’anticiper, entre autres, les répercussions sur la biodiversité et le paysage.47 23 j. Sucre ajouté 30 g/jour au maximum Une consommation excessive de sucre augmente les risques d’obésité et de maladies associées, ainsi que de diabète. Il convient donc de réduire au maximum la consommation de sucre. La valeur limite s’oriente vers le Planetary Health Diet.61 Objectif 2 : Réduction des pertes d’azote et de phosphore dans le système alimentaire En plus de l’objectif de réduire l’utilisation d’azote et de phosphore liée à l’alimentation, des réductions mesurables des pertes d’azote et de phosphore nuisibles à l’environnement dans la production sont nécessaires pour ramener les effets négatifs sur l’environnement à un niveau compatible avec l’écosystème. C’est pourquoi les pertes nocives d’azote (sans les pertes de N2) et de phosphore de l’agriculture suisse dans l’environnement, qui représentent aujourd’hui respectivement environ 60% des apports en azote et environ 40% des apports en phosphore, seront réduites de 10% d’ici 2025, de 10% supplémentaires d’ici 2030 et encore de 10% d’ici 2045 par rapport à la moyenne des années 2014 à 2016 (97 344 t d’azote, 6087 t de phosphore). En raison de son importance dans la perte de biodiversité et de l’urgence qui en découle (voir ci-dessous), des objectifs de réduction supplémentaires sont nécessaires pour l’ammoniac : d’ici 2030, les émissions d’ammoniac (NH3) provenant notamment de l’élevage seront réduites au niveau régional de manière à ce que les charges critiques d’azote pour les forêts et les écosystèmes (semi-)naturels ne soient pas dépassées en Suisse. Pour cela, il faut les réduire d’au moins 40% pour atteindre un maximum de 25 000 t de NH3-N par an. Pour boucler les cycles des éléments nutritifs, le phosphore sera entièrement récupéré à partir de 2030 dans les eaux usées, les boues d’épuration et les farines animales. L’azote sera récupéré à 70% dans les eaux usées à partir de 2035 et à 80% à partir de 2045. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? L’azote (N) et le phosphore (P) sont essentiels à la croissance des plantes et garantissent des rendements élevés dans l’agriculture. Des rendements élevés à l’hectare réduisent à leur tour la pression exercée pour étendre la surface agricole utile afin de nourrir une population mondiale croissante. Selon les estimations, sans l’utilisation d’engrais azotés, la surface agricole utile aurait dû augmenter dans la même proportion que la population mondiale pour garantir la sécurité alimentaire.65 L’utilisation d’engrais azotés et phosphatés contribue donc de manière significative à la réalisation de l’ODD 2 (Faim « zéro »). 24 Par contre, l’utilisation excessive d’engrais azotés et phosphatés et les fortes densités de bétail dans l’agriculture entraînent une surfertilisation des eaux douces (principalement par les phosphates)66 et des océans (principalement par les nitrates),67 ce qui va à l’encontre de la réalisation de l’ODD 14 – Vie aquatique. Par le biais des dépôts atmosphériques d’azote (ammoniac), les écosystèmes terrestres sont également touchés par la surfertilisation due à une utilisation excessive d’azote,68–71 ce qui rend plus difficile la réalisation de l’ODD 15 – Vie terrestre. La surfertilisation se traduit finalement par une perte de biodiversité et donc par une perte de fonctionnalité des écosystèmes.72 D’autres problèmes liés aux composés azotés réactifs, notamment ceux issus de l’agriculture, sont l’acidification des sols et des eaux par l’ammoniac,73, 74 le réchauffement du climat par le protoxyde d’azote75 (lien avec l’ODD 13 – Mesures relatives à la lutte contre les changements climatiques), la pollution par les nitrates dans les eaux souterraines32 ainsi que le dégagement de poussières fines (par l’ammoniac comme précurseur).73 Les deux derniers points représentent notamment un danger pour la santé humaine (références à l’ODD 3 – Bonne santé et bien-être et à l’ODD 6 – Eau propre et assainissement). Les pertes importantes de nutriments dues à l’utilisation d’engrais azotés et phosphorés et à l’élevage sont également un problème dans l’agriculture suisse. Des efforts isolés dans l’agriculture ont certes permis de réduire les pertes de nutriments. Par exemple, des améliorations dans la gestion des engrais de ferme ont permis de réduire les émissions d’ammoniac provenant de l’agriculture d’environ 4% entre 2005 et 2018. Malgré tout, les apports excessifs d’azote et de phosphore dans l’environnement en Suisse sont les principaux moteurs de la perte de biodiversité.72 Cela est notamment lié au fait que les charges critiques d’azote sont dépassées en Suisse dans environ 90% des sites forestiers, dans presque tous les sites de hauts-marais, dans trois quarts des bas-marais et dans un tiers des prairies et pâturages secs riches en espèces.68 Un dépassement persistant menace à long terme les fonctions de la forêt (fonctions d’exploitation, de protection et de détente) et la pérennité des écosystèmes sensibles. Justification des valeurs cibles spécifiques Un changement de régime alimentaire en faveur d’une alimentation essentiellement végétale permet de réduire l’impact environnemental négatif de l’utilisation d’azote d’environ 30% et celui de l’utilisation de phosphore d’environ 20%.22 Le changement de modèle alimentaire est donc une stratégie importante pour atteindre un niveau inférieur aux limites de charge globales pour l’azote et le phosphore. Toutefois, les limites de charge globales pour l’azote et le phosphore ne sont pas déduites sur la base des pertes de composés réactifs d’azote et de phosphore, mais sont définies – notamment en raison de la disponibilité des données globales – sur la base de l’utilisation d’azote et de phosphore dans les engrais minéraux dans l’agriculture.22, 65 D’une part, les pertes de composés d’azote et de phosphore sont en corrélation avec la quantité d’engrais utilisée. D’autre part, des analyses de sensibilité sur les limites de charge ont montré que, d’un point de vue global, l’utilisation de l’azote et du phosphore sont des indicateurs suffisants pour évaluer les pertes de nutriments.22, 49 Les limites de charge globales actuellement calculées pour l’azote et le phosphore se situent dans une fourchette de 65 à 90 millions de tonnes d’azote par an et de 6 à 12 millions de tonnes de phosphore par an.22, 49, 65 Sur la base de la population mondiale actuelle d’environ huit milliards d’habitants, cela correspond à une utilisation de 8 à 11 kg d’azote par habitant et par an et de 0,75 à 1,5 kg de phosphore par habitant et par an. Aujourd’hui déjà, la limite de charge globale pour l’utilisation agricole d’azote et de phosphore sous forme d’engrais artificiels est dépassée.22, 49 Étant donné que, selon les prévisions actuelles, la population mondiale atteindra environ dix milliards d’habitants d’ici 2050, l’utilisation d’azote et de phosphore dans l’agriculture ne diminuera que d’environ 4% pour l’azote et 7% pour le phosphore, même en cas de passage à une alimentation essentiellement végétale, et dépassera donc toujours la limite de charge globale.49 Outre l’adaptation du mode d’alimentation, d’autres mesures sont donc nécessaires du côté de la production ainsi que dans le domaine de la prévention des pertes et du gaspillage alimentaires pour réduire les pertes dans l’environnement. Cela permettrait d’utiliser plus d’azote et de phosphore dans l’agriculture que la limite de charge actuelle.49 En partant de scénarios dans lesquels, d’ici 2050, les déficits de rendement dans les cultures sont comblés à 90% au niveau mondial grâce à l’amélioration des pratiques agricoles, l’efficacité d’utilisation des éléments nutritifs est augmentée de 30% et le taux de recyclage du phosphore est porté à au moins 50%, et les pertes et le gaspillage alimentaires sont réduits de moitié, selon Willett et al (2019)49. Les limites de charge se situent entre 90 et 130 millions de tonnes par an pour l’azote et entre 8 et 16 millions de tonnes par an pour le phosphore. Pour une population mondiale de dix milliards d’habitants, cela représente 8,13 à 11,25 kg d’azote et 0,8 à 1,6 kg de phosphore par habitant et par an. Certes, la Suisse ne peut influencer directement les mesures de production au niveau mondial et les pertes alimentaires globales que de manière limitée, mais la Confédération peut exercer une certaine influence globale par le biais du commerce et des accords internationaux. En outre, elle peut contribuer directement à la réalisation de l’objectif en mettant en œuvre des mesures appropriées pour la production nationale et en prenant des mesures pour réduire les pertes et le gaspillage alimentaires au sein de la Suisse (voir objectif 8). Des objectifs de réduction de 20% des apports d’éléments nutritifs issus de l’agriculture d’ici 2030 ont déjà été définis au niveau fédéral.76 Afin de suivre également les objectifs globaux concernant l’utilisation d’azote et de phosphore dans la production agricole, le comité scientifique estime qu’une réduction supplémentaire de 10% des apports est nécessaire d’ici 2045.22, 49 Parce que les émissions d’ammoniac représentent environ 60% des pertes d’azote de l’agriculture suisse77 et qu’elles sont principalement responsables du dépassement à grande échelle des charges critiques d’azote en Suisse,68 leur réduction revêt une importance particulière. Une réduction rapide de ces pertes est le levier le plus efficace pour réduire les effets environnementaux négatifs des pertes d’éléments nutritifs en Suisse et pour augmenter l’efficacité de l’utilisation de l’azote dans l’agriculture suisse. Un autre levier important pour réduire les pertes de nutriments est l’augmentation du taux de récupération d’azote et de phosphore dans les eaux usées, les boues d’épuration et les farines animales. Selon l’ordonnance fédérale sur le traitement des déchets (OTD, RS 814.600), la récupération du phosphore dans les eaux usées communales, les boues d’épuration, les cendres de boues d’épuration et les farines animales et osseuses sera obligatoire en Suisse dès 2026. Par souci de cohérence avec les autres objectifs en ce qui concerne l’horizon temporel, l’année 2030 a été choisie comme date limite de mise en œuvre pour la récupération du phosphore dans la définition des objectifs. La récupération de l’azote, notamment dans les eaux usées, n’est pas prévue à ce jour au niveau législatif en Suisse. Ce n’est qu’en récupérant le phosphore et l’azote que ces cycles de matières peuvent être bouclés. En outre, la récupération contribue notamment à réduire la surfertilisation des océans (ODD 14) et à compenser la répartition inégale des éléments nutritifs à l’échelle mondiale.49 Objectif 3 : Réduction des effets écotoxiques des pesticides Les effets écotoxiques des pesticides doivent être réduits de 55% d’ici 2030 et de 75% d’ici 2040 (par rapport à la moyenne 2012–2015), au-delà de l’objectif fixé par la Confédération (50% d’ici 2027). Il s’agit d’examiner les effets causés par la consommation de denrées alimentaires par la population suisse en Suisse et à l’étranger.. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Les pesticides sont utilisés pour garantir les rendements et la qualité des produits, mais ils ont également des effets indésirables sur les organismes non ciblés. Sánchez-Bayo et al. (2019) citent par exemple la perte d’habitats, la pollution par les pesticides et les engrais, les agents pathogènes et les espèces envahissantes ainsi que le changement climatique comme principaux moteurs de la disparition d’espèces d’insectes (ODD 13 – Climat, ODD 14 – Vie aquatique, ODD 15 – Vie terrestre).78 Dans le concept de la protection intégrée des cultures, l’utilisation de produits phytosanitaires chimiques est la dernière mesure à appliquer lorsque les autres mesures n’ont pas été suffisamment efficaces.79 La priorité est donnée aux mesures préventives telles que les variétés résistantes, la rotation des cultures, les cultures mixtes ou la promotion des insectes utiles, les aides à la décision comme les systèmes de prévision ainsi que la lutte non chimique par des méthodes biologiques ou physiques. Mais il faut aussi repenser les exigences du commerce et des consommateurs en matière de produits ; les aliments présentant de petits défauts visuels peuvent tout à fait être encore consommés. La réduction de l’utilisation des 25 pesticides ne devrait toutefois pas entraîner une diminution importante des rendements ou de la qualité. Sinon, il y a un risque que les surfaces cultivées doivent être étendues au détriment d’habitats naturels (parfois à l’étranger) ou que les pertes alimentaires augmentent. Il est toutefois important de souligner qu’une utilisation intensive de pesticides est associée à une diminution des populations d’insectes et à une baisse de la biodiversité, ce qui peut également avoir un impact négatif sur les rendements.78, 80, 81 En outre, les changements de comportement des consommateurs (voir objectif 1) permettent de gagner des surfaces pour la production d’aliments végétaux. Cela peut minimiser les éventuels conflits d’objectifs entre la réduction de l’utilisation des pesticides et la garantie des rendements. La réalisation de l’objectif 3 implique des efforts accrus en matière de recherche et de sélection pour développer des méthodes alternatives de protection des plantes, afin d’éviter autant que possible les pertes de rendement. Cet objectif contribue donc de manière importante à la réalisation de l’ODD 6 – Eau propre et assainissement, de l’ODD 14 – Vie aquatique et de l’ODD 15 – Vie terrestre. Cependant, une diminution des rendements, de la qualité ou de la capacité de stockage peut également mettre en péril la sécurité alimentaire et donc aller à l’encontre de l’objectif 2 – Faim « zéro ». La réduction de l’utilisation des pesticides contribue également à améliorer la santé des agriculteurs et des agricultrices82, 83 et de l’ensemble de la population83, 84. On peut donc également s’attendre à des synergies positives avec l’ODD 3 – Bonne santé et bien-être, notamment en réduisant les risques pour les utilisateurs de produits phytosanitaires. Justification des valeurs cibles spécifiques Le plan d’action phytosanitaire79 prévoit une réduction de 50% du risque potentiel pour les organismes aquatiques d’ici 2027 par rapport à la moyenne 2012 – 2015. Cet objectif est donc repris pour un premier jalon. La biodiversité étant soumise à une forte pression, le comité scientifique estime qu’il faut réduire encore davantage la pollution de l’environnement par les pesticides. La question de savoir à quel niveau une charge écologique supportable à long terme doit être fixée n’est pas définitivement tranchée. C’est pourquoi les objectifs s’orientent également vers ce qui est techniquement réalisable. Möhring et al. proposent des solutions pour réduire l’utilisation de produits phytosanitaires.85 Dans les scénarios d’optimisation calculés par Ow et al., il en résulte des réductions de 40% par rapport à la référence.47 Ces scénarios prennent en compte des adaptations au niveau de la consommation, mais peu d’améliorations au niveau de la production. Le passage à une alimentation basée sur les végétaux s’accompagne généralement d’une augmentation de la consommation de fruits et de légumes, et ces cultures sont parmi celles qui nécessitent le plus de produits phytosanitaires. Les changements côté consommation ne suffisent donc pas à atteindre les objectifs. Mathis et al. ont montré qu’une réduction d’un facteur 4 de l’écotoxicité de l’eau douce pouvait être obtenue dans la culture des pommes et que les risques pour les organismes aquatiques pouvaient être réduits de 98 à 99%.86 On constate toujours que certaines substances actives dominent ;87 c’est pourquoi la substitution de ces substances actives est également une solution prometteuse. Waldvogel et al. ont montré que pour cinq grandes cultures, il est possible 26 de réduire les impacts environnementaux d’un tiers par rapport à une intensité de traitement moyenne, et d’un facteur 8 par rapport à une intensité élevée.88 C’est pourquoi les valeurs cibles fixées sont certes très ambitieuses, mais réalisables du point de vue actuel. La condition préalable est que toutes les options soient épuisées. Toutefois, les rendements et la qualité ne doivent pas être affectés de manière significative et les charges financières et de personnel doivent rester supportables. Bystricky et al. ont par exemple montré que des réductions importantes des traitements phytosanitaires dans le pays entraîneraient une délocalisation à l’étranger (en raison d’importations plus importantes), ce qui provoquerait d’importants conflits d’objectifs.89 Furrer et al. ont calculé qu’une sélection optimale des produits importés permettrait de réduire l’écotoxicité de l’eau douce d’environ 60% pour le même panier de produits.90 Si nous combinons ces potentiels avec de nouvelles possibilités techniques, les objectifs fixés devraient être réalisables. Les méthodes d’évaluation de l’impact environnemental global des pesticides sont disponibles avec USEtox et le modèle de consensus PestLCI 91, 92 ou avec des modèles d’évaluation des risques87 à condition que l’utilisation de produits phytosanitaires dans la production puisse être déterminée. Objectif 4 : Utilisation durable des terres agricoles en Suisse La perte irréversible des fonctions écologiques des sols sur les terres agricoles en Suisse est réduite de 55% d’ici 2030 et de 75% d’ici 2040 (par rapport à la situation actuelle). Cet objectif est atteint par des réductions correspondantes des pertes de matière organique (perte d’humus pour les sols minéraux, disparition de la tourbe pour les sols organiques), une réduction des pertes de biodiversité du sol, une réduction de l’érosion et du tassement du sol ainsi que des apports de polluants persistants. Le potentiel de séquestration cumulé des terres agricoles en Suisse (terres arables et prairies permanentes), estimé aujourd’hui à 10,2 mégatonnes de carbone (ce qui correspond à l’extraction d’environ 37 mégatonnes de CO2 de l’atmosphère), et qui peut être atteint grâce à des pratiques agricoles adaptées sur l’ensemble du territoire, sera atteint à 40% d’ici 2030 et à 90% d’ici 2040. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Le sol est une ressource non renouvelable à court terme et constitue la base même de la production de denrées alimentaires. En outre, le sol joue un rôle important pour la biodiversité et fournit des services écosystémiques essentiels tels que l’approvisionnement en eau potable, le stockage du carbone et la protection contre les risques naturels.43 Ainsi, une utilisation durable des sols agricoles contribue directement à la réalisation de l’ODD 2 – Faim « zéro », de l’ODD 6 – Eau propre et assainissement, de l’ODD 13 – Mesures relatives à la lutte contre les changements climatiques et de l’ODD 15 – Vie terrestre. De ce fait, la capacité d’adaptation des sols au changement climatique diminue également. On part du principe que sans adaptation de la gestion des sols, les pertes de carbone des sols augmenteront en raison de la hausse des températures et de la sécheresse accrue.43 Justification des valeurs cibles spécifiques Les points abordés dans l’objectif 4 se recoupent largement avec les objectifs formulés dans la Stratégie Sol Suisse pour préserver à long terme la ressource sol et ses fonctions écologiques,43 et portent sur les défis jugés pertinents par les parties prenantes pour une gestion durable des sols agricoles.97 Il manque cependant des données couvrant l’ensemble du territoire sur l’évolution dans le temps des principaux indicateurs d’état des sols agricoles en Suisse. C’est pourquoi la Stratégie Sol de la Confédération prévoit d’abord d’obtenir les données nécessaires dans les années à venir. Pour cette raison, il n’est pas encore possible de formuler pour l’objectif 4 d’autres valeurs cibles quantitatives plus concrètes à atteindre d’ici 2030 ou 2040 (p. ex. la part maximale terres agricoles tassées en 2030 et 2040). Une autre alternative consiste à définir le pourcentage de réduction des évolutions négatives d’ici 2030 et 2040, afin de faire face au risque de perte de fonction des sols agricoles et de garantir ainsi l’utilisation à long terme de la ressource sol. Les pourcentages formulés de 50% (d’ici 2030) et 75% (d’ici 2040) coïncident avec les objectifs de réduction des effets écotoxiques des produits phytosanitaires (objectif 3). Les produits phytosanitaires sont entre autres coresponsables des apports de polluants dans les sols et de la dégradation des fonctions du sol qui en découle.98 La valeur cible concernant le potentiel de séquestration cumulé dans les sols agricoles (terres arables et prairies permanentes) est dérivée d’une estimation réalisée pour la Suisse Aujourd’hui, en Suisse, l’utilisation durable des sols n’est pas assurée en de nombreux endroits en raison d’une exploitation souvent inadaptée aux conditions du sol.43 L’utilisation de machines lourdes sur des sols humides est la principale cause de tassement des sols agricoles. Une couverture du sol et un enracinement insuffisants pendant les périodes critiques, un travail du sol trop intensif et un pâturage excessif sur les terrains en pente sont les principales causes de l’érosion du sol. Les modélisations actuelles partent du principe que 20% des terres arables en Suisse sont menacées d’érosion93 et qu’il existe également un risque d’érosion considérable pour les prairies.94 En outre, les fonctions du sol peuvent être considérablement affectées par des polluants. Les voies d’apport de polluants dans les sols agricoles sont l’épandage d’engrais minéraux, d’engrais de ferme et d’engrais de recyclage contaminés par les substances correspondantes, ainsi que l’utilisation de produits phytosanitaires. En outre, les terres agricoles remplissent une fonction importante en ce qui concerne le stockage du carbone.95 Après les océans, le sol est le deuxième plus grand réservoir de carbone.96 L’augmentation de la teneur en carbone du sol contribue à la séquestration du carbone et donc à la réduction de la concentration de CO2 dans l’atmosphère. Mais actuellement, ce sont surtout les sols cultivés en Suisse qui sont potentiellement concernés par la perte de matière organique du sol,43 ce qui a des conséquences négatives sur de nombreuses fonctions du sol et donc sur la fertilité du sol dans son ensemble. 27 des quantités de carbone pouvant être séquestrées de manière cumulative sur un horizon de 20 ans, en supposant des taux de séquestration annuels correspondants.95 Dans l’hypothèse d’un taux de séquestration annuel de 0,63 t C/ha99 il est possible de stocker au total environ 5 mégatonnes de carbone sur un horizon de 20 ans sur les quelque 400 000 ha de terres arables selon la statistique de la superficie, jusqu’à ce que la saturation des terres arables soit atteinte. Par analogie, sur les 920 000 ha de prairies en Suisse, en supposant un taux de séquestration annuel de 0,28 t C/ha, il est possible de stocker100 environ 5,2 mégatonnes de carbone, jusqu’à ce que le niveau de saturation soit atteint. Si l’on parvient à atteindre et à maintenir à l’avenir les taux de séquestration supposés en modifiant les pratiques agricoles sur les terres arables et dans les prairies permanentes, il sera possible d’extraire de l’atmosphère un total d’environ 37 mégatonnes de CO2 au cours des 20 prochaines années et de les stocker durablement dans le sol. 1900 et 40% des oiseaux nicheurs sont en danger.104 Ce sont surtout les agrobiocénoses (écosystèmes modifiés par l’agriculture) qui ont perdu de nombreuses petites structures paysagères telles que les haies en raison d’une exploitation intensive et non durable. Cette perte est encore accentuée par l’utilisation de grandes quantités d’engrais et de pesticides, un ensemencement pauvre en espèces et une utilisation mécanique uniforme. Depuis les années 90, des incitations à la promotion d’habitats proches de la nature et de petites structures ont toutefois été mises en place au niveau des exploitations et adaptées en permanence.105 Les objectifs pour l’agriculture ont été définis comme suit : « L’agriculture protège et favorise les espèces et les habitats indigènes qui, dans leur aire de répartition naturelle, se trouvent principalement sur des terres agricoles ou sont tributaires de l’exploitation agricole. Les populations des espèces cibles sont maintenues et favorisées. Les populations d’espèces indicatrices sont favorisées par la mise à disposition d’habitats appropriés d’une superficie suffisante et présentant la qualité et la répartition spatiale requises. »106 Objectif 5 : Réduction de la perte de biodiversité En résumé, l’objectif 5 joue un rôle central dans la réalisation de divers ODD. Ainsi, la biodiversité fournit des biens naturels et des services écosystémiques qui sont essentiels au maintien des systèmes de production de denrées alimentaires et, plus généralement, au développement de la société et de l’économie : fourniture d’eau potable de qualité suffisante grâce aux forêts et à leurs sols (ODD 6) ; préservation de la fertilité des sols (ODD 15) ; régénération des matières premières et des denrées alimentaires (ODD 2 et ODD 15) ; ressources génétiques pour de nouveaux médicaments ou de nouvelles cultures (ODD 3) ; amélioration de la qualité de vie grâce à des zones de loisirs naturelles (ODD 3) ; protection contre les chutes de pierres et les avalanches grâce aux forêts de montagne (ODD 15) ; absorption des pics de pluie par les marais et les zones humides (ODD 13 et 15) ; régulation et traitement des pathogènes, des adventices et des ravageurs (ODD 15). De plus, à l’échelle mondiale, plus de 75% des cultures, y compris les fruits et les légumes, ainsi que certaines des principales cultures, comme le café, le cacao et les amandes, dépendent de la pollinisation animale. Les écosystèmes marins et terrestres sont actuellement les seuls puits d’émissions de carbone anthropiques, avec un stockage brut de 5,6 gigatonnes de carbone par an (ce qui représente environ 60% des émissions anthropiques mondiales) (ODD 13).101 Dans le monde, de moins en moins de variétés et de races de plantes et d’animaux sont cultivées, élevées, échangées et conservées, malgré de nombreux efforts locaux, notamment de la part des peuples autochtones et des communautés locales. Les produits destinés à la consommation alimentaire nationale (produits dans le pays ou importés) ne sont pas impliqués dans la déforestation d’ici 2030, n’entraînent pas de perte supplémentaire de biodiversité en Suisse et à l’étranger et proviennent de systèmes de production locaux adaptés aux conditions locales afin de maximiser les services écosystémiques. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? En moyenne, environ 25% des espèces d’animaux et de plantes évaluées sont menacées à l’échelle mondiale.101 Cela signifie qu’environ un million d’espèces sont déjà menacées d’extinction. Parmi elles, beaucoup le seront dans quelques décennies si aucune mesure n’est prise pour réduire l’intensité des causes de la perte de biodiversité. Sans mesures appropriées, l’extinction mondiale des espèces, qui est déjà au moins dix à cent fois plus élevée que la moyenne des dix derniers millions d’années, continuera à s’accélérer.101 Cela va à l’encontre de la réalisation de divers ODD, par exemple l’ODD 14 – Vie aquatique et l’ODD 15 – Vie terrestre. La perte de biodiversité et l’interaction croissante entre l’homme et la faune sauvage augmentent en outre le risque de pandémie et compromettent donc également la réalisation de l’ODD 3 – Bonne santé et bien-être.102 La consommation de la Suisse et l’augmentation des importations de biens et de services augmentent la pression sur les ressources naturelles et la biodiversité à l’échelle mondiale. L’empreinte sur la biodiversité de la Suisse est quatre fois supérieure aux valeurs limites mondiales.103 Cela inclut la consommation de denrées alimentaires et l’importation d’aliments pour animaux en provenance de pays dont les « points chauds de biodiversité » sont menacés, comme le Brésil (soja, café, riz), l’Afrique de l’Ouest (cacao), la Colombie (café) et l’Indonésie (huile de palme, cacao), ce qui entraîne la déforestation et la perte de biodiversité dans les écosystèmes tropicaux humides. En Suisse aussi, la diversité biologique a dramatiquement diminué depuis 1900 et la situation actuelle est alarmante. Au total, 60% des insectes sont menacés, 90% des prairies sèches et pâturages secs ont disparu depuis 28 Justification des valeurs cibles spécifiques Bien que le taux de déforestation mondial récent soit inférieur à celui de la décennie précédente, il ne l’est que d’environ un tiers, et la déforestation pourrait à nouveau s’accélérer dans certaines régions. La consommation alimentaire suisse entraîne une déforestation d’environ 102 km2 par an en Suisse et à l’étranger.47 Il est important de souligner que la base de données est ici entachée de grandes incertitudes et qu’il est urgent d’augmenter à l’avenir la transparence et la traçabilité concernant les effets sur la biodiversité et la déforestation des denrées alimentaires et des matières premières importées en Suisse. Une consommation alimentaire sans déforestation d’ici 2030 contribuera à lutter contre la perte, la dégradation et la fragmentation des habitats dans les forêts et autres biomes, en particulier dans les écosystèmes présentant la plus grande biodiversité dans les régions tropicales.107 Toute réduction de la diversité des plantes cultivées, des espèces sauvages apparentées et des races domestiquées signifie que les agroécosystèmes seront moins résistants aux changements climatiques, aux ravageurs et aux pathogènes à venir. En effet, une réduction de la biodiversité entraîne généralement une diminution du fonctionnement des écosystèmes, et ce sur un large éventail d’écosystèmes et de fonctions. Une grande diversité biologique est une sorte de réassurance biologique. Lorsqu’une espèce disparaît, d’autres espèces peuvent prendre le relais dans l’écosystème, qui reste ainsi stable. Inversement, cela signifie aussi que plus les espèces disparaissent, plus les écosystèmes sont instables et plus le risque est grand que – par un effet de cascade – d’autres espèces disparaissent également. L’écosystème concerné finira par s’effondrer et ne pourra plus assurer ses fonctions actuelles. Les services écosystémiques fournis par la biodiversité permettent donc à l’homme d’exister et d’exercer des activités économiques.108, 109 La biodiversité est donc indispensable au fonctionnement des écosystèmes, qui incluent les ressources essentielles à la production de denrées alimentaires.108, 110, 111 Objectif 6 : Réduction de l’empreinte hydrique liée à l’alimentation Les produits destinés à la consommation alimentaire nationale (produits dans le pays ou importés) ne contribuent pas à une pénurie d’eau excessive d’ici 2030 : selon la méthode AWARE, l’empreinte hydrique de la consommation alimentaire suisse par habitant diminue de 30% d’ici 2030 par rapport à 2010. Pour les produits issus de régions où l’eau est rare, une attention particulière est accordée aux bonnes pratiques dans le contexte local. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? En moyenne sur plusieurs années, la pénurie d’eau ne joue pour l’instant un rôle dans la consommation suisse que pour les biens et services importés, mais également transformés dans la production nationale. Il est donc possible de lutter contre la pénurie d’eau en adaptant les importations directes ainsi que les chaînes d’approvisionnement. Avec sa consommation par habitant, la Suisse se situe environ un cinquième au-dessus de la moyenne mondiale de l’empreinte hydrique.112 Une réduction de la consommation et la prévention des déchets peuvent ici aussi contribuer à réduire l’empreinte hydrique. bonne santé et le bien-être (ODD 3). En ce qui concerne le bien-être, l’eau est importante pour la qualité de vie (p. ex. pour se rafraîchir et irriguer les espaces verts dans les villes) et les zones de loisirs aquatiques. L’eau étant souvent surexploitée par les groupes à haut revenu, l’objectif relatif à la pénurie d’eau est également important pour réduire les inégalités (ODD 10). De nombreux conflits internationaux et régionaux sont en partie basés sur des litiges concernant les ressources en eau, et cet objectif est donc également central pour la paix, la justice et des institutions efficaces (ODD 16). L’eau est également une ressource essentielle pour la vie aquatique (ODD 14) et terrestre (ODD 15), car les zones riveraines et les zones humides constituent des habitats précieux. De plus, l’eau est nécessaire dans les écosystèmes pour fixer le carbone de l’air et contribue donc également à l’atténuation du changement climatique (ODD 13). Justification des valeurs cibles spécifiques Depuis la première étude sur les limites planétaires113, la communauté scientifique a débattu du fait que, malgré une disponibilité suffisante de l’eau à l’échelle globale, la pénurie d’eau est critique à l’échelle mondiale. Sur la base d’analyses de la pénurie d’eau, il a été proposé de réduire la consommation d’eau de 40 à 50% en général,114 afin d’atteindre en moyenne un niveau de pénurie d’eau supportable. Les différences régionales devraient être prises en compte et les économies devraient être recherchées en priorité dans les zones où l’eau est rare. Environ la moitié de la population mondiale vit dans des conditions critiques en ce qui concerne la pénurie d’eau.115 Des études récentes examinent les limites au niveau des bassins versants afin de déterminer plus précisément où l’eau est consommée au-delà d’une mesure sûre.45, 116, 117 Le PNUE a recommandé la méthode AWARE pour l’évaluation des empreintes hydriques (en mettant l’accent sur la pénurie)118, et la FAO l’a également recommandée pour les empreintes hydriques119. Cette méthode combine la pénurie d’eau en valeur absolue (disponibilité de l’eau par surface) avec des indicateurs de consommation par rapport à la disponibilité (analogues à l’ODD 6 – Indicateur de pénurie d’eau) et est largement utilisée pour évaluer la pénurie d’eau dans les chaînes d’approvisionnement.112, 120 Selon un rapport sur les valeurs d’orientation commandé par l’OFEV 121, 122, l’eau consommée en Suisse provient à 29% de bassins versants dont les limites sont dépassées (dans l’UE, cette part est de 26%). Étant donné que la Suisse contribue au problème exclusivement par le biais d’importations (denrées alimentaires, fourrage et coton), l’objectif porte en premier lieu Cet objectif joue un rôle central dans la réalisation de plusieurs ODD, notamment l’ODD 6 – Fournir de l’eau potable de qualité suffisante. L’eau est également essentielle pour l’agriculture et joue donc un rôle important pour l’ODD 2 – Faim « zéro », l’ODD 1 – Pas de pauvreté et donc indirectement pour une 29 sur les chaînes d’approvisionnement. Les consommateurs et les producteurs ont donc besoin d’informations sur l’origine des marchandises importées et sur la manière dont elles sont produites. En ce qui concerne la pénurie d’eau, les principales régions affectées par la consommation suisse se trouvent en Inde, au Pakistan, en Chine et en Égypte, mais aussi aux États-Unis et en Espagne. L’objectif de réduction de 30% de l’empreinte hydrique selon la méthode AWARE tient ainsi compte des limites des différents bassins versants et intègre le fait que l’empreinte hydrique suisse par habitant est supérieure de 20% à la moyenne mondiale, ce qui pose déjà des problèmes. Un objectif plus ambitieux peut être poursuivi à long terme. Objectif 7 : Réduction des émissions de gaz à effet de serre La consommation alimentaire nationale doit atteindre la neutralité climatique d’ici 2045 au plus tard. Cela comprend à la fois la production nationale et les importations. Les objectifs de réduction visés par rapport à l’année 2010 sont les suivants : 2045 Compensation par des émissions négatives 2030 2035 Méthane (CH4) –40% –60% Non Protoxyde d’azote (N2O) –40% –60% Oui CO2 –40% –90% Oui Tableau 1 : Fesenfeld et al. (2023)237 / Telek Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Les nations industrielles prospères comme la Suisse contribuent fortement aux émissions mondiales de gaz à effet de serre par habitant. La Suisse s’est donc officiellement fixé pour objectif d’atteindre la neutralité climatique d’ici 2050.42, 123 Toutefois, les ODD (surtout l’ODD 13 – Climat) et les objectifs de Paris ne pourront probablement pas être atteints au niveau mondial si les nations industrielles prospères comme la Suisse n’atteignent la neutralité climatique qu’en 2050.124-126 Pour atteindre la neutralité climatique et l’objectif climatique de Paris, le système alimentaire doit apporter une contribution décisive.127 Les émissions de gaz à effet de serre provenant de la production de produits animaux – en particulier une réduction marquée des émissions de méthane à courte durée de vie – et les changements d’affectation des sols jouent un rôle central dans ce contexte pour réduire le risque de dépassement à court terme des points de basculement climatiques.35, 127, 128 Pour la Suisse, cela signifie que les émissions de gaz à effet de serre doivent être réduites rapidement et de manière significative tout au long de la chaîne de valeur des denrées alimentaires consommées en Suisse (voir également l’objectif 1). L’empreinte de gaz à effet de serre de la consommation alimentaire suisse en 2018 s’élevait à environ 21,5 millions de tonnes d’équivalents CO2, dont 7,2 millions de tonnes en 30 Suisse. Environ 77% des émissions ont eu lieu à l’étranger et ont été causées par l’importation de denrées alimentaires et d’aliments pour animaux.75 Justification des valeurs cibles spécifiques L’année de référence 2010 a été choisie par analogie avec l’étude de Ow et al.47 parce qu’elle a servi d’année de référence pour les scénarios spécifiques à la Suisse dans l’étude et parce que les objectifs de réduction des émissions mentionnés ci-dessus s’appuient principalement sur cette étude.47 a) Méthane (CH4) Une réduction de 40% d’ici 2030 et de 60% d’ici 2035 peut être atteinte en optimisant l’alimentation et la production de denrées alimentaires et en évitant les déchets alimentaires.47, 50 Il s’agit d’un objectif très ambitieux, mais réalisable. Des scénarios plus avancés,47 dans lesquels 70% des prairies pourraient être abandonnées, conduisent certes à une réduction encore plus importante des émissions de gaz à effet de serre, mais aussi à des conflits d’objectifs avec d’autres ODD et à une alimentation unilatérale, et ne sont donc pas pris en compte ici. Les émissions de méthane devraient être réduites rapidement, car cela a un fort effet de refroidissement à court terme. Cela permet de gagner du temps pour développer des technologies à émissions faibles et négatives. Contrairement au protoxyde d’azote et au dioxyde de carbone, les émissions de méthane restantes n’ont pas à être compensées par des technologies à émissions négatives. Dès qu’un nouvel équilibre sera atteint, il n’y aura plus d’effet de réchauffement supplémentaire. Des études récentes indiquent toutefois que le taux de décomposition du méthane dans l’atmosphère est nettement réduit (jusqu’à quatre fois plus lent) en raison du réchauffement climatique et de l’augmentation des incendies de forêt.129 Si ces résultats se confirment, les objectifs de réduction des émissions de méthane devront être examinés et, le cas échéant, formulés de manière encore plus rigoureuse. Cela pourrait toutefois impliquer certains conflits d’objectifs avec d’autres ODD (p. ex. ODD 3 – Santé). b) Protoxyde d’azote (N2O) L’objectif de réduction est fixé à 40% d’ici 2030 et à 60% d’ici 2035, ce qui est ambitieux mais réalisable.47 Les émissions restantes doivent être compensées par des technologies à émissions négatives afin d’atteindre l’objectif global de neutralité climatique. c) Émissions de CO2 Les émissions de CO2 dans le système alimentaire sont principalement dues à l’utilisation d’énergies fossiles (p. ex. pour la production d’engrais, de combustibles et de carburants) ainsi qu’à l’utilisation des terres et à ses changements. L’utilisation d’énergies fossiles doit être évitée autant que possible à l’avenir et remplacée par d’autres sources d’énergie, technologies et systèmes de production (p. ex., une culture accrue de légumineuses). Le modèle d’optimisation de Ow et al.47 ne tient pas compte de toutes les nouvelles technologies et innovations à faibles émissions (p. ex. les énergies renouvelables, la remise en eau des sols organiques) qui devraient être développées dans les décennies à venir ou qui existent déjà (voir la baisse rapide des prix des technologies d’énergie renouvelable).47 Par conséquent, l’objectif de réduction de 40% d’ici 2030 et de 90% d’ici 2045 est très ambitieux, mais réaliste. Les émissions de CO2 restantes doivent être compen- sées par des technologies à émission négative afin d’atteindre l’objectif global de neutralité climatique. Objectif 8 : Réduction des déchets et des pertes alimentaires Les déchets et pertes alimentaires évitables dans les chaînes d’approvisionnement nationales et internationales des produits consommés en Suisse seront réduits de 25% d’ici 2025 et de 50% d’ici 2030 (par rapport à 2017). Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Le sous-objectif 12.3 de l’ODD indique que le gaspillage alimentaire par habitant doit être réduit de moitié d’ici 2030 au niveau du commerce de détail et des consommateurs, et être limité au niveau de l’agriculture et de la transformation, y compris en ce qui concerne les pertes post-récolte. En juin 2022, le Conseil fédéral a publié un rapport sur l’avenir de la politique agricole suisse.130 Il y a été réaffirmé que près de 25% de l’impact environnemental de l’alimentation suisse est dû à des pertes alimentaires évitables. Les pertes et les déchets alimentaires désignent toutes les denrées alimentaires destinées à la consommation humaine qui ne sont pas consommées par l’homme. Les pertes alimentaires évitables sont également appelées gaspillage alimentaire (ou food waste). Elles concernent les parties comestibles des aliments qui sont produites pour la consommation humaine, mais qui ne sont pas effectivement consommées par l’homme.50, 131 Le Conseil fédéral (2022) constate dans son rapport qu’un tiers des denrées alimentaires sont actuellement jetées.130 Cela représente 2,8 millions de tonnes de déchets générés chaque année par la consommation suisse de denrées alimentaires en Suisse et à l’étranger.132 Par personne et par an, cela correspond à environ 330 kg de déchets alimentaires évitables.130 Justification des valeurs cibles spécifiques En fixant spécifiquement les valeurs cibles, nous nous alignons sur le plan d’action du Conseil fédéral130 ainsi que sur l’ODD 12. Selon le plan d’action du Conseil fédéral en deux phases (phase 1 : 2022–2025, phase 2 : 2026–2030), la quantité de pertes alimentaires évitables sera réduite de moitié par rapport à 2017. La valeur indicative qui devrait se dessiner d’ici 2025 pour atteindre l’objectif d’ici 2030 a été fixée à 25% de réduction. D’ici 2030, les déchets et pertes alimentaires devraient être réduits de moitié par rapport à 2017. Ces ambitions sont cohérentes avec le sous-objectif 12.3 formulé dans le cadre de l’objectif de durabilité 12.133 Les valeurs cibles formulées sont en outre cohérentes avec les autres objectifs environnementaux, notamment l’objectif 2 et l’objectif 747. Objectif 9 : Revenus de subsistance tout au long de la chaîne d’approvisionnement Les employés travaillant dans les chaînes d’approvisionnement des produits alimentaires consommés en Suisse doivent bénéficier d’un revenu et d’un salaire suffisants pour vivre d’ici 2030. Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Une transformation durable du système alimentaire implique de manière centrale l’instauration de conditions équitables dans la production, la distribution et la consommation de denrées alimentaires – et ce à tous les niveaux de la chaîne de valeur alimentaire, y compris au-delà des frontières nationales. Cela touche directement plusieurs ODD, notamment l’ODD 1 – Pas de pauvreté, l’ODD 2 – Faim « zéro », l’ODD 3 – Bonne santé et bien-être, l’ODD 8 – Travail décent et croissance économique et l’ODD 10 – Inégalités réduites. Un système alimentaire socialement responsable et équitable exige que, tout au long de la chaîne de valeur alimentaire, les travailleurs puissent vivre de leurs revenus. Les revenus des personnes employées dans l’agriculture de production, en particulier, sont depuis longtemps une source régulière d’inquiétude.134 En Suisse, environ un cinquième des ménages agricoles vivent aujourd’hui en dessous du minimum vital (mesuré par l’aide sociale). Des situations de revenus précaires existent également à d’autres niveaux de la chaîne de valeur alimentaire, par exemple dans le domaine de la transformation et de la distribution des denrées alimentaires.9 En outre, la situation des revenus est particulièrement précaire pour de nombreux producteurs et transformateurs de denrées alimentaires à l’étranger dont les produits sont importés en Suisse.135, 136 Justification des valeurs cibles spécifiques Selon une étude de Leinert et al. (2016), une personne active sur huit en Suisse travaille dans le secteur alimentaire.137 Plus de la moitié d’entre elles travaillent dans des petites ou moyennes entreprises. En moyenne sur trois ans, entre 2018 et 2020, la main-d’œuvre familiale dans les régions de plaine, de collines et de montagne a gagné respectivement 85%, 64% et 56% du salaire de référence dans les secteurs secondaire et tertiaire.138 L’objectif de développement durable 8 de l’Agenda 2030 porte notamment sur la promotion d’un travail décent. En particulier, le sous-objectif 8.5 formule l’exigence que toutes les personnes, indépendamment de leur âge et de leur sexe, doivent recevoir une rémunération égale pour un travail de valeur égale. Du point de vue du comité scientifique, un système alimentaire socialement durable se caractérise entre autres par le fait que toutes les personnes qui y sont employées peuvent obtenir un salaire horaire suffisant pour vivre. Dans les pays d’où la Suisse importe des denrées alimentaires et des matières premières, un salaire horaire vital n’est actuellement souvent pas garanti pour les employés de la production de denrées alimentaires.135, 136 C’est pourquoi le comité scientifique propose de fixer comme objectif contraignant que 31 tous les employés, sans exception, dans le pays et à l’étranger, tout au long de la chaîne de création de valeur des denrées alimentaires consommées en Suisse, reçoivent un revenu et un salaire suffisants pour vivre. Objectif 10 : Réduction des risques psychosociaux et physiques pour la santé dans l’agriculture suisse Les risques psychosociaux et physiques liés à l’activité professionnelle des personnes travaillant dans l’agriculture suisse seront réduits de moitié d’ici 2030 (par rapport à 2016). Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? Un système alimentaire durable devrait contribuer à la santé non seulement des consommateurs, mais aussi des travailleurs impliqués dans la production et la fourniture de denrées alimentaires. Or, de nombreux travailleurs du secteur agricole suisse sont actuellement exposés à des exigences psychiques et physiques considérables.138 Ainsi, selon une étude d’Agroscope, la proportion de personnes menacées de burnout dans le secteur agricole suisse est largement supérieure à la moyenne nationale.139 Dans leur étude de cohorte, Steck et al. ont en outre constaté un taux de suicide plus élevé chez les hommes travaillant dans l’agriculture que dans d’autres secteurs professionnels. 140 En outre, ils ont constaté que cet écart se creusait depuis 2006 : alors que le taux de suicide a baissé dans d’autres professions, il est resté constant dans le secteur agricole. La charge physique due aux travaux lourds et les cadences de travail sont également élevées.138 Ces exemples illustrent l’énorme besoin d’action, surtout dans le secteur agricole, pour réduire les risques pour la santé liés au travail. Justification des valeurs cibles spécifiques Depuis 2015, la santé mentale est inscrite dans les ODD de l’ONU comme l’un des 13 sous-objectifs de l’ODD 3. C’est pourquoi la mention explicite et la prise en compte égalitaire des maladies physiques et psychiques liées au travail ainsi que des accidents dans l’agriculture constituent un indicateur cible central pour la réalisation des ODD. Réduire de moitié les risques psychosociaux et physiques pour la santé dans l’agriculture suisse d’ici 2030 (par rapport à 2020) est ambitieux, mais réalisable pour un pays prospère comme la Suisse. En outre, la réalisation du sous-objectif 8.8. de l’Agenda 2030 pour le développement durable requiert des environnements 32 de travail sûrs pour tous les travailleurs.14 Toutefois, il convient de noter que le travail physique avec des machines souvent de grande taille comporte inévitablement des risques physiques qui ne peuvent être totalement exclus. Mais une réduction de la pression au travail ainsi que des risques psychosociaux pour la santé peut également prévenir les accidents liés au travail physique. Il est possible de minimiser encore les risques psychosociaux pour la santé en surveillant délibérément la santé mentale des personnes travaillant dans l’agriculture, en les impliquant activement et en les faisant participer à l’amélioration de leurs conditions de travail. Objectif 11 : Respect des normes de l’OIT sur le travail des enfants tout au long de la chaîne d’approvisionnement Pourquoi s’agit-il d’un objectif prioritaire pour atteindre les ODD ? La durabilité implique de n’abandonner personne. Un système alimentaire socialement durable n’est pas compatible avec le travail des enfants, quel qu’il soit. Au-delà des considérations morales fondamentales, les dommages directs sur la santé et les conséquences indirectes du travail des enfants sur leurs perspectives de développement social et économique sont considérables et incompatibles avec les droits de l’enfant. Dans le contexte international, le travail des enfants dans l’agriculture est un énorme problème. Au niveau mondial, on estime à 112 millions le nombre d’enfants de moins de 18 ans qui contribuent au revenu familial, le plus souvent par un travail extrêmement mal payé dans l’agriculture. Justification des valeurs cibles spécifiques La Convention sur les pires formes de travail des enfants de l’Organisation internationale du travail (OIT), ratifiée par la Suisse en 2000, va au-delà d’une interdiction et demande à ses parties de concevoir et de mettre en œuvre des plans d’action pour endiguer le travail des enfants.141 En outre, le Conseil fédéral s’est clairement prononcé, avec le sousobjectif 8.7 de l’ODD 8, en faveur de la suppression toute forme de travail des enfants d’ici 2025. Du point de vue de la consommation alimentaire, cela signifie que le travail des enfants doit également être stoppé à l’étranger tout au long de la chaîne de valeur des aliments consommés en Suisse. 2.4 Comparaison des indicateurs cibles sur le statu quo du système alimentaire suisse Le comité scientifique, en collaboration avec EBP Suisse, a réalisé une comparaison illustrative entre le statu quo et les indicateurs cibles prioritaires et actuellement quantifiables pour la transformation du système alimentaire suisse. Cette comparaison, illustrée dans le graphique 4, permet de voir approximativement l’état actuel et l’état souhaité. Il est important de noter que la représentation de la situation actuelle et de la situation souhaitée est une approximation et qu’elle ne peut pas donner une image complète des divers effets sur la durabilité du système alimentaire suisse. Les indicateurs spécifiques pour chaque objectif ont été sélectionnés sur la base des connaissances actuelles et en tenant compte des données existantes. Dans certains cas, il pourrait toutefois être utile d’adapter les indicateurs spécifiques dans le cadre d’une révision régulière et plus précise des objectifs. En outre, le comité scientifique est limité par des lacunes parfois importantes dans les données. Il a parfois fallu renoncer à comparer les valeurs cibles avec le statu quo en raison du manque de données ou utiliser les valeurs disponibles des années précédentes pour se rapprocher du statu quo, et la comparabilité de certaines valeurs réelles avec les valeurs cibles est limitée par des différences méthodologiques. Un contrôle de la réalisation des objectifs nécessite une meilleure collecte régulière d’indicateurs de durabilité pertinents tout au long de la chaîne de valeur. Par exemple, il n’est actuellement pas possible de faire des déclarations valides sur les effets écotoxiques de l’utilisation des pesticides, notamment ceux liés aux importations (objectif 3), ni sur l’empreinte biodiversité de la consommation alimentaire suisse (objectif 5) tout au long de la chaîne de valeur. C’est pourquoi le panel recommande également, dans le chapitre 3, des obligations plus étendues en matière de rapports ainsi que la mise en place d’une infrastructure de mesure et de données portant sur différents indicateurs relatifs aux objectifs climatiques et environnementaux tout au long de la chaîne de valeur. Comme nous l’avons expliqué plus en détail au début du chapitre 2, le comité scientifique recommande en outre de mettre l’accent sur l’adressage des indicateurs cibles côté consommation (voir entre autres l’indicateur cible 1). La réalisation de la plupart des autres indicateurs cibles dépend de la réalisation des objectifs de consommation. Du point de vue de la consommation, on prend en compte les émissions et les groupes de produits les plus importants qui sont produits et consommés au sein de la chaîne d’approvisionnement suisse. La méthodologie de la comparaison illustrative entre le statu quo et les indicateurs cibles prioritaires pour la transformation du système alimentaire suisse est détaillée dans l’appendice en ligne. 33 Objectifs pour une alimentation adaptée d’ici 2030 Valeur actuelle 607 Valeur cible 2030 350 600 500 400 300 230 250 240 161 215 200 101 30 e aj o ut é ue 30 et pr od ui ts L d Su ériv ait és cr co q s à Fr ui ts Lé gu m Po i ss uf Œ Vi an de Fr ui t es um te rre Lé g 35 20 in eu se 15 50 0 95 27 on 98 100 134 100 et Lé prodCér ui éa gu ts l dé es p. me ex s riv .p f és om éc m ul es en de ts Consommation recommandée par jour et par personne en grammes 700 Le graphique visualise la consommation alimentaire actuelle en Suisse par jour et par personne, en grammes, selon la catégorie d’aliments. Il visualise également la consommation par jour et par personne recommandée par le comité scientifique, selon la catégorie d’aliments. Les calculs se basent sur les données de consommation indiquées dans le bilan alimentaire actuel (Agristat, 2022), moins les quantités de déchets alimentaires générées par les ménages et la restauration selon Beretta & Hellweg (2019).54.132 237 Graphique 4 : Fesenfeld Valeur de et al. (2023) / Telek départ Unité 81'500 tN 5'400 tP 40'400 t NH3-N 10.2 Mt C 102 km2 19.1 mia m3 22.5Mt éq. CO2 2.8 mio t 57 %h 12 % 61 %h pa e P ra Pe n rte sé sd qu es e pa N tra ra H tio n 3 n in ex pl oi té Dé fo re st at Em pa ion Em pr ra ei pr n nt ei Dé e nt hy ch e dr ga et i s/ z pa qu pe à ra e eff rte n et sa de lim se en Ri sq pa rre ta ra ue ire n sé de v re ita b pa ven pa le s l us ra s e n n ’ m a R in ss da isq im ur da ns ue um an sp ns Ri l ’ a vit t s gr ou la qu al icu r ch e aî de ltu la s ne t re an d’ rav su té ap ai iss pr l d e ov es isi en on fa ne nt m s en t te sd Po t en t ie l de Pe r Pe r te s d pa e N ra n 20% 20% Objectif de réduction4 montre tout d’abord que la consommation inutilisé Le graphique 30% ne sera alors plus que d’environ 6,1 mégatonnes de 40% 40% actuelle de denrées alimentaires s’écarte40% parfois fortement des carbone.95, 99, 100 50% 50% Valeur cible 65’000 4’300 valeurs cibles (objectif 1).54, 132 En particulier, la consommation 13.4 moyenne de viande devrait passer d’environ 100 grammes par La déforestation liée à l’alimentation est un moteur important jour et par personne actuellement à environ 50 grammes par de la perte de13.5 biodiversité dans le monde et a été estimée 24’000 6.1 jour et par personne d’ici 2030. La consommation de produits à environ 102 km2 par an pour la consommation alimentaire 100% 100% 100% 1.4 6% laitiers et de sucres ajoutés devrait également être nettement suisse. D’ici 2030, l’alimentation suisse ne doit plus provoquer réduite, tandis que la consommation de légumes, de légumide déforestation (objectif 5).47 neuses et de noix devrait augmenter. La consommation de produits céréaliers devrait être augmentée et comprendre prin- En 2018, l’empreinte hydrique agrégée de la Suisse liée à cipalement des céréales complètes. l’alimentation était d’environ 19,1 milliards de m3. D’ici 2030, cette empreinte hydrique devrait être réduite à au moins 0.0 0% 0% 13,1 milliards de m3 (objectif 6).37, 120–122 Les autres objectifs nécessitent également de revoir à la hausse les ambitions, et ce parfois de manière importante. Ainsi, en 2015, les pertes totales d’azote et de phosphore de Selon des calculs récents, l’empreinte de gaz à effet de serre l’agriculture suisse s’élevaient à environ 81 500 t d’azote et liée à l’alimentation en Suisse était d’environ 22,5 Mt CO2eq 5400 t de phosphore. Il faudrait réduire ces pertes annuelles par an en 2018. Pour atteindre les objectifs de durabilité, il d’au moins 20% d’ici 2030 (voir objectif 2). De plus, la faudrait réduire l’empreinte de gaz à effet de serre liée à l’aliperte totale d’azote-ammoniac dans l’agriculture suisse mentation de 40% d’ici 2030 et plafonner à 13,5 Mt CO2eq s’élevait à 40 400 t de NH3-N en 2015. Pour être en accord par an (objectif 7).37, 47 avec les ODD, il faudrait réduire cette valeur de 40% d’ici 2030 (voir objectif 2).77, 142, 143 Les déchets et pertes alimentaires évitables par an ont été estimés à environ 2,8 millions de tonnes en 2017 en Suisse. D’ici 2030, il faudrait réduire cette valeur de 50% pour atteindre un Le potentiel cumulé de séquestration des sols agricoles en maximum de 1,4 million de tonnes par an (objectif 8).130, 132 Suisse (terres arables et prairies permanentes) est estimé aujourd’hui à environ 10,2 mégatonnes de carbone (ce qui correspond à l’extraction d’environ 37 mégatonnes de CO2 de Environ 57% des heures de travail générées par l’alimentation l’atmosphère). Ce potentiel devrait être exploité à 40% d’ici suisse dans le monde ont eu lieu dans des pays et des sec2030 (objectif 4), de sorte que le potentiel de séquestration teurs présentant un risque moyen à très élevé de salaires in34 et L p. ex . e Lé gu Objectifs de réduction d’ici 2030 Valeur de départ Unité 81'500 tN 5'400 tP Objectif de réduction 20% 20% Valeur cible 65’000 4’300 40'400 t NH3-N 10.2 Mt C 40% 40% 102 km2 19.1 mia m3 22.5Mt éq. CO2 30% 40% 2.8 mio t 57 %h 12 % 50% 61 %h 50% 13.4 24’000 6.1 13.5 100% 1.4 ire et ta eff en zà ue Ri sq sa rte pe s/ et ch de lim ga e nt ei pr 0% sé vit re ab pa ven pa le s l us ra s e n n ’ m a R in ss da isq im ur da ns ue um an sp ns Ri l ’ a vit t s gr ou la qu al icu r ch e aî de ltu la s ne t re an d’ rav su té ap ai iss pr l d e ov es isi en on fa ne nt m s en t se de dr i hy e nt ei pr Em Em 100% pa rre ra n pa qu ra e n t pa ion ra n or es ta lo ité Dé f ex p in 6% 0% Dé Po t en tie ld e sé qu es tra tio n Pe rte sd e pa N ra H n 3 ar P an sd p e Pe rte Pe rte sd p e ar N an 0.0 100% Les pertes d’azote annuelles s’élevaient à environ 81 500 t en 2015 (OFAG 2021, Reutimann et al. 2022, Spiess / Liebisch 2020). 77, 142, 143 Elles doivent être réduites de 20 % d’ici 2030. Les pertes de phosphore annuelles s’élevaient à 5400 t en 2015 (OFAG 2021, Reutimann et al. 2022, Spiess / Liebisch 2020).77, 142, 143 D’ici 2030, elles doivent être réduites de 20 %. Les pertes d’ammoniac annuelles s’élevaient à environ 40 400 t en 2015 (OFAG 2021, Reutimann et al. 2022, Spiess / Liebisch 2020). 77, 142, 143 Elles doivent être réduites de 40 % d’ici 2030. Grâce à une gestion durable des sols à long terme, les terres arables et les prairies permanentes pourraient séquestrer un total de 10,2 millions de tonnes de carbone (Beuttler 2019).95 D’ici 2030, 40 % de ce potentiel devrait être réalisé. Déforestation : pour la consommation suisse de denrées alimentaires, environ 102 km2 de forêt sont défrichés chaque année (von Ow et al. 2020). D’ici 2030, ce chiffre devrait être réduit à zéro. L’empreinte hydrique de l’alimentation suisse s’élevait à 19,1 milliards de m3 en 2010 (Nathani et al. 2022). 37 D’ici 2030, elle devrait être réduite de 30 %. En 2010, l’empreinte de gaz à effet de serre de l’alimentation suisse était d’environ 22,5 millions de tonnes d’équivalent CO2 (Nathani et al. 2023).37 D’ici 2030, elle devrait être réduite de 40 %. En 2017, les déchets alimentaires évitables s’élevaient à environ 2,8 millions de tonnes (Beretta, Hellweg 2019). Ils devraient être réduits de moitié d’ici 2030. Revenu de subsistance : environ 57 % des heures de travail générées par l’alimentation suisse dans le monde ont lieu dans des pays et des secteurs qui présentaient un risque moyen à très élevé de revenus n’assurant pas le minimum vital (Social Hotspots Database 2022).144 D’ici 2030, toutes les personnes employées dans la chaîne d’approvisionnement de l’alimentation suisse devraient bénéficier de revenus assurant le minimum vital. Risques pour la santé : en 2016, 12 % des agricultrices et agriculteurs suisses étaient exposé·e·s au risque de burn-out (Reissig 2017).139 D’ici 2030, ce chiffre doit être réduit de moitié, tout comme d’autres risques pour la santé liés au travail. Travail des enfants : environ 61 % des heures de travail générées par l’alimentation suisse dans le monde ont lieu dans des pays et des branches qui présentent un risque moyen à très élevé de travail des enfants (Social Hotspots Database 2022).144 D’ici 2030, les conventions de l’OIT sur le travail des enfants doivent être appliquées dans toute la chaîne d’approvisionnement de l’alimentation suisse. Graphique 5 : Fesenfeld et al. (2023)237 / Telek suffisants pour vivre. Il faudrait réduire ce risque à 0% d’ici 2030 pour transformer le système alimentaire suisse conformément aux Objectifs de développement durable (objectif 9).144 Selon les données disponibles, 12% des agriculteurs et agricultrices suisses étaient exposés au risque de burn-out en 2016 : d’ici 2030, ce chiffre doit être réduit de moitié (objectif 10), tout comme d’autres risques pour la santé liés au travail.139, 144 Environ 61% des heures de travail générées par l’alimentation suisse dans le monde ont eu lieu dans des pays et des secteurs présentant un risque moyen à très élevé de travail des enfants. D’ici 2030, les conventions de l’OIT sur le travail des enfants doivent être appliquées dans toute la chaîne d’approvisionnement de l’alimentation suisse (objectif 11).144 35 36 37 3 VOIES POLITIQUES POUR LA TRANSFORMATION DU SYSTÈME ALIMENTAIRE SUISSE Auteur·e·s chapitre 3 : Lukas Fesenfeld, Stefan Mann, Bettina Scharrer, Claudio Beretta, Elisabeth Bürgi Bonanomi, Janina Grabs Le chapitre 2 a présenté le système d’objectifs visé et les indicateurs cibles mesurables d’un système alimentaire suisse durable. Mais comment les atteindre concrètement sur le plan politique ? Outre les initiatives des acteurs privés, qui apportent une contribution très importante à la transformation du système alimentaire, par exemple sur une base volontaire ou dans le cadre d’accords interprofessionnels, la politique a un rôle central à jouer dans la maîtrise de cette tâche qui concerne l’ensemble de la société. Le présent chapitre montre comment des points de basculement positifs dans le processus de transformation peuvent être déclenchés par une politique globale du système alimentaire et des voies politiques stratégiquement coordonnées. Grâce à cette conception stratégique de la politique tout au long de la chaîne de valeur, le comité scientifique estime que les objectifs formulés au chapitre 2 peuvent être atteints d’ici 2030. Lors de l’élaboration de cette proposition, le comité scientifique s’est notamment penché sur la manière de concevoir une transformation socialement équitable et de supprimer les obstacles actuels à la transformation pour les actrices et les acteurs tout au long de la chaîne de création de valeur. Nous nous appuyons sur le concept des points de basculement positifs (voir l’introduction et le paragraphe suivant) comme leitmotiv pour l’élaboration d’une éventuelle voie de transformation politique et nous nous concentrons sur les chances que représente le processus de transformation pour les multiples actrices et acteurs impliqué·e·s. Tout d’abord, nous souhaitons expliquer brièvement plus en détail les notions de « points de basculement » et de « voies d’action politique » dans le contexte du système alimentaire (cf. également le chapitre d’introduction). Nous décrivons ensuite cinq conditions qui devraient être remplies pour déclencher des points de basculement positifs dans le système alimentaire.145 Nous présentons ensuite, comme pièce maîtresse de ce chapitre, une voie d’action politique détaillée dont la mise en œuvre stratégique peut remplir les cinq conditions de points de basculement et contribuer ainsi à la réalisation du système cible. On appelle points de basculement les « points critiques d’un système où des interventions ciblées entraînent des conséquences importantes et à long terme sur l’évolution d’un système socio-technique ou naturel et modifient profondément (souvent de manière irréversible) son fonctionnement ».145 Une mesure politique peut constituer une telle intervention, qui conduit à l’émergence d’une boucle de rétroaction positive. En cela, le point de basculement déclenché par l’intervention active d’autres points de basculement plus petits, qui peuvent à leur tour déclencher des points de basculement systémiques plus larges. De manière imagée, on peut donc se représenter un point de basculement comme une sorte de chaise qui bascule et qui change d’état en peu de temps (par exemple de debout à allongé). La recherche actuelle montre qu’il existe également des points de basculement positifs dans des systèmes complexes, tels que le système alimentaire ou énergétique. Par exemple, le développement rapide des énergies renouvelables indique un point de basculement positif qui a été atteint grâce à un mélange d’innovations technologiques et de 38 changements de comportement, de normes et de politiques.40 Par exemple, des programmes de soutien ciblés et des innovations technologiques ont fait baisser le prix des énergies renouvelables, ont rendu leur développement plus attractif pour les investisseuses et les investisseurs et les particuliers et ont finalement accéléré l’introduction de mesures politiques visant à réduire l’utilisation des énergies fossiles. 40, 147, 148 En outre, les interfaces entre les systèmes, par exemple le système énergétique et le système alimentaire, peuvent également donner lieu à un changement rapide et à des effets de rétroaction positifs. Par exemple, le développement de centrales agrophotovoltaïques, qui renforcent à la fois le développement des énergies renouvelables et offrent une nouvelle source de revenus aux agriculteurs, peut accélérer les processus de transformation du système alimentaire et énergétique.146, 149 Les points de basculement positifs peuvent donc contribuer à une nette accélération du changement de système en direction des objectifs de durabilité. Toutefois, le déclenchement de ces points de basculement positifs nécessite une élaboration active et stratégique des politiques. Dans leur état naturel, les systèmes ne sont généralement pas « proches » d’un point de basculement qui serait alors déclenché automatiquement ou de manière aléatoire. Souvent, il faut un pilotage politique actif qui provoque délibérément le déclenchement du point de basculement.150 Dans des systèmes complexes comme le système alimentaire, un tel déclenchement peut être le fait de mouvements de la société civile, d’entreprises pionnières, d’innovations technologiques et de mesures politiques.40, 147, 151 L’État joue ici un rôle particulièrement central. Le choix stratégique des mesures politiques ainsi que la conception du paquet de mesures politiques et le moment précis de la mise en œuvre de ces mesures sont décisifs pour déclencher ou non des points de basculement positifs. Le rapport de la Food and Land Use Coalition145 définit cinq conditions qui devraient être remplies pour favoriser la réalisation de points de basculement positifs dans le système alimentaire : La solution ou la mesure politique proposée devrait notamment ... ... rendre les produits et procédés alternatifs durables économiquement compétitifs (condition de prix). ... satisfaire au moins au niveau de performance ou de qualité des produits et procédés alternatifs durables par rapport aux produits et procédés existants et, dans l’idéal, le dépasser (condition de performance). ... rendre les produits et procédés alternatifs durables facilement accessibles aux parties prenantes (condition d’accès). ... rendre les produits et procédés alternatifs durables souhaitables pour les parties prenantes et la société et promouvoir les normes sociales correspondantes (condition de norme). ... augmenter les informations et les capacités des parties prenantes afin de rendre les produits et les processus alternatifs durables réalisables pour elles (condition de capacité). Le développement de telles mesures pour déclencher des points de basculement positifs est extrêmement difficile. Remplir toutes les conditions avec une seule mesure s’avère souvent impossible. Pour cette raison, le comité scientifique propose comme solution le regroupement stratégique et la succession ciblée de mesures dans ce que l’on appelle des paquets politiques et des voies d’action.152 Par exemple, les instruments d’incitation négative basés sur la réglementation et le marché peuvent être mieux acceptés s’ils sont introduits en même temps que des instruments d’incitation basés sur l’information et la formation. Grâce à ce regroupement politique, les barrières politiques peuvent être surmontées et les parties prenantes de différents camps politiques peuvent être amenées à un compromis commun dans de nouvelles coalitions.152 L’acceptation publique des mesures politiques transformatrices peut également être accrue de cette manière (ibid.). Le regroupement stratégique, mais aussi l’agencement stratégique de tels paquets de mesures dans le temps améliorent encore le potentiel de franchissement des obstacles politiques, économiques, technologiques et sociaux.40 Enfin, il convient de tenir compte de la perspective institutionnelle et multi-niveaux (c’est-à-dire commune, canton, Confédération, international) du système alimentaire et, par conséquent, du processus de transformation. Le chapitre 4 est consacré plus précisément au cadre institutionnel de décision pour l’élaboration de la politique du système alimentaire (ce que l’on appelle la gouvernance du système alimentaire). Les différents niveaux impliqués influencent toutefois aussi la conception et la mise en œuvre de voies d’action politique efficaces et réalisables. La conception stratégique des voies d’action politiques est donc le résultat d’une évaluation et d’une pondération laborieuses de différentes mesures politiques et de leur agencement ciblé, les unes avec les autres et les unes après les autres, à travers les différents niveaux impliqués. quatre catégories bien établies : 1. Mesures axées sur l’information et la formation ; 2. Instruments d’incitation positive (c’est-à-dire aides et subventions publiques) ; 3. Instruments d’incitation négative (c’est-à-dire taxes d’incitation publiques et autres instruments de marché visant à réduire les externalités négatives) ; 4. Instruments réglementaires. Dans un premier temps, le comité scientifique recommande un examen complet de toutes les mesures publiques existantes afin d’évaluer quelles mesures existantes servent les objectifs de durabilité et les indicateurs cibles définis pour la transformation du système alimentaire suisse (chapitre 2) ; quelles mesures ne contribuent que partiellement à la réalisation de ces objectifs et devraient donc être réaménagées ; et quelles mesures ne contribuent pas aux objectifs et devraient donc être supprimées. En particulier lors de la transformation et de la suppression de mesures existantes, il convient de prendre en compte si et comment les actrices et les acteurs qui supportent des coûts de transformation particulièrement élevés peuvent être compensé·e·s. 3 / Dans une troisième étape, les expert·e·s ont ensuite pla- Outre cet inventaire complet et l’examen des mesures existantes, le comité scientifique estime qu’il faudrait en outre introduire de nouveaux paquets de mesures stratégiquement harmonisés. Dans ce qui suit, nous présentons maintenant une voie d’action politique prometteuse d’un point de vue scientifique, qui peut d’une part contribuer efficacement à la réalisation des onze indicateurs d’objectifs (voir chapitre 2) et qui est réalisable tant sur le plan politique, technique, juridique, économique, biophysique que social. Concrètement, le comité scientifique a suivi les quatre étapes suivantes : 1 / Sur la base de l’état actuel de la recherche, le comité scientifique a d’abord rassemblé des mesures individuelles importantes qui présentent un fort potentiel de transformation. Ces mesures ont ensuite été réparties par les scientifiques en 2 / Dans un deuxième temps, les membres du comité scientifique ont ensuite classé les mesures par ordre de priorité en fonction de leur impact, de leur urgence et de leur faisabilité pour atteindre les indicateurs cibles (voir chapitre 2) et ont identifié les effets de synergie potentiels avec d’autres mesures. En outre, les scientifiques ont mis en évidence lesquelles des cinq conditions de point de basculement pouvaient être adressées par chaque mesure individuelle. Il est particulièrement important de souligner que le comité scientifique a utilisé des études d’évaluation récentes, si possible spécifiques au contexte, comme base pour l’élaboration des propositions de mesures. Cependant, il existe encore en partie des lacunes évidentes dans la recherche en ce qui concerne l’évaluation de l’efficacité et de la faisabilité des mesures, notamment en matière d’interaction entre différentes mesures. Le comité scientifique souligne donc la nécessité de poursuivre la recherche sur l’évaluation ex ante et ex post dans ce domaine. Sur la base de ces études d’évaluation, certaines recommandations de mesures pourraient donc être adaptées et complétées. Outre les preuves empiriques actuelles, les propositions du comité présentées ici sont basées sur un processus Delphi itératif, dans lequel les membres du comité scientifique ont pu apporter leur expertise interdisciplinaire individuellement ainsi que dans le cadre de discussions de groupe (voir l’appendice en ligne pour les détails méthodologiques du processus Delphi itératif pour le développement de la voie d’action). cé les différentes mesures dans un ordre chronologique stratégique et les ont combinées entre elles afin de maximiser leur potentiel de transformation synergique ainsi que leur faisabilité. 4 / Dans une dernière étape, une enquête a été menée au- près des membres du comité afin d’évaluer quantitativement l’efficacité, l’urgence et le soutien des paquets de mesures. Dans l’enquête, les trois dimensions ont été évaluées pour chaque mesure sur une échelle de Likert à 5 points. La valeur 5 signifie à chaque fois le niveau le plus élevé, c’est-à-dire un soutien total, une efficacité très élevée ou la plus grande urgence perçue pour la mise en œuvre de cette mesure (d’ici 2025). La valeur 1 indique en revanche que la mesure n’est pas du tout soutenue par le comité scientifique, que ses membres voient dans la mesure un potentiel d’efficacité très faible et qu’ils ne mettraient jamais en œuvre la mesure (urgence très faible). Ci-dessous, pour chaque proposition de mesure, cette évaluation quantitative des expert·e·s est rendue visible de manière transparente sous forme de valeur moyenne. Le taux de réponse à l’enquête a été de 65 %. Le comité scientifique recommande de réaliser d’autres études d’évaluation ex ante et ex post afin d’évaluer encore plus précisément les interactions complexes des différentes mesures. La proposition ne doit donc pas être considérée comme définitive, mais comme une suggestion pour le débat scientifique, public et politique. 39 3.1 Aperçu des recommandations d’actions stratégiques En se basant sur la littérature en sciences sociales sur les solutions politiques par paquets et les voies de transformation, le comité scientifique esquisse dans figure 6 une séquence de quatre paquets de mesures stratégiquement coordonnés.40, 153-158 Chaque paquet de mesures comprend un certain nombre de mesures individuelles qui s’imbriquent de manière synergique tout au long de la chaîne de création de valeur et qui, par conséquent, mettent délibérément l’accent non seulement sur les mesures de politique agricole, mais aussi sur l’ensemble du système alimentaire. En outre, chaque paquet de mesures s’adresse aux différentes conditions du point de basculement, augmentant ainsi les chances de créer, au fil du temps, les conditions nécessaires pour permettre un changement systémique. Il est important d’appliquer le principe « encourager et exiger ». La prémisse de base pour la séquence stratégique des mesures est donc de créer d’abord les conditions économiques, sociales, techniques et politiques par des mesures d’encouragement tout au long de la chaîne de création de valeur, afin de permettre à moyen terme des mesures plus exigeantes et donc actuellement en partie controversées. Le comité scientifique s’attache en outre à élaborer des approches socialement équitables. D’une part, le comité scientifique recommande des mesures de compensation ciblées pour les groupes à faible revenu (cf. par exemple la redistribution des recettes d’une taxe éq. CO2). D’autre part, il faut aussi, en principe, des revenus suffisamment élevés pour permettre un mode d’alimentation sain et durable, même si les coûts externes de la production alimentaire se reflètent de manière réaliste dans le prix. Les paquets de mesures visent 40 également à générer des sources de revenus alternatives pour les actrices et les acteurs du système qui tirent actuellement des revenus de processus de production, de transformation, de commerce et de consommation dont certains ne sont pas conformes aux Objectifs mondiaux de développement durable et au modèle correspondant pour le système alimentaire suisse (voir chapitre 2).Concrètement, le comité scientifique recommande l’introduction coordonnée des quatre paquets de mesures suivants d’ici 2030 (cf. également figure 6, p. 42, et le tableau 3, p. 43 – 45) : Dans une première phase, le comité scientifique recommande la création d’un fonds de transformation global qui commencerait le plus rapidement possible à réduire les obstacles existants dans le processus de transformation par des mesures d’information et de formation ainsi que des instruments d’encouragement financiers ciblés, et à aborder les cinq conditions de basculement. Ce fonds global de transformation permet en particulier de construire de nouvelles perspectives économiques et des normes sociales autour des opportunités du processus de transformation. Les ressources du fonds permettent ainsi de créer de nouvelles sources de revenus pour des entreprises et des collaboratrices et collaborateurs ciblé·e·s tout au long de la chaîne de création de valeur. En outre, grâce aux instruments d’encouragement, il est possible d’acquérir avec ces actrices et ces acteurs les compétences nécessaires pour relever les défis de la transformation et de développer des technologies durables et innovantes jusqu’à leur mise sur le marché. Tout cela peut, à terme, renforcer l’adhésion à des mesures de transformation plus profondes, par exemple l’introduction de taxes d’incitation ciblées et d’interventions réglementaires nécessaires. Les nouvelles compétences, les nouvelles normes et l’accès à des technologies et à des sources de revenus innovantes peuvent donc, à terme, réduire les obstacles existants au changement de système et accroître l’adhésion politique et sociale à la transformation. Le comité scientifique propose donc de mettre en œuvre rapidement les mesures 1a* à 2i* (* indique des mesures dans le fonds de transformation) décrites en détail plus loin dans le cadre d’un vaste fonds de transformation. La capacité d’action politique rapide et la promotion ciblée dans le cadre de la politique en lien avec le coronavirus montrent par exemple que des mesures d’encouragement aussi importantes sont également possibles à court terme en Suisse. Comme nous l’avons vu aux chapitres 1 et 2, la pression pour agir est très forte. Pour cette raison, le comité scientifique recommande aux responsables politiques de mettre en place le fonds de transformation dans les meilleurs délais. Le fonds de transformation devrait être doté d’un financement complet et pourrait être financé par des ressources budgétaires supplémentaires de l’État fédéral et, idéalement, accroître son impact grâce à une coordination stratégique avec les investisseuses et les investisseurs privé·e·s. À moyen terme, le fonds pourrait en outre être élargi par l’introduction de nouvelles taxes d’incitation ainsi que par la réaffectation de fonds existants. L’organisation précise du fonds peut s’inspirer d’exemples de réussite dans d’autres pays, par exemple au Danemark.159 Dans la deuxième phase, des taxes d’incitation et des mesures réglementaires efficaces devraient être introduites progressivement sur la base du fonds de transformation (voir description détaillée des mesures 3a à 3g et 4a à 4k ci-dessous). Au début, le niveau d’intervention ne devrait volontairement pas être trop bas, afin de rendre le processus de transformation équitable et réalisable pour l’ensemble des actrices et des acteurs concerné·e·s. D’ici 2030, ces mesures plus coûteuses, mais indispensables d’un point de vue scientifique, devraient être progressivement mises en place de manière plus stricte. L’augmentation au coup par coup et transparente du niveau d’intervention crée une fiabilité et une prévisibilité pour l’ensemble des actrices et des acteurs. Les coûts éventuels pour certain·e·s actrices et acteurs de ce processus (p. ex. les exploitations produisant des animaux et des aliments pour animaux) peuvent et doivent en outre être amortis socialement et compensés de manière ciblée (voir p. ex. les mesures prises dans le cadre du fonds de transformation, comme p. ex. 2g, les primes de compensation pour les reconversions* ainsi que 2h, les programmes de soutien aux jeunes agricultrices et agriculteurs pour la reconversion des exploitations*). Les bénéfices et les coûts générés par le processus de transformation doivent être répartis le plus équitablement possible entre les actrices et les acteurs de la chaîne de création de valeur, d’où la nécessité de certaines adaptations fiscales et réglementaires. Il est en outre essentiel de contrer de manière ciblée les éventuels effets régressifs (c’est-à-dire des coûts relativement plus élevés pour les groupes à faibles revenus). Le comité scientifique a élaboré différentes propositions de mesures à cet effet (voir par exemple les propositions de redistribution des recettes d’une taxe éq. CO2 sur les denrées alimentaires). Dans la troisième phase, d’autres mesures de politique agricole ainsi qu’un soutien supplémentaire aux zones rurales devraient être introduits (2j à 2m). Ces instruments d’incitation positifs et ces mesures de soutien aux agricultrices et agriculteurs pourraient également être introduits en même temps que le deuxième paquet de mesures, mais l’expérience montre qu’ils nécessitent un peu plus de temps pour être négociés, compte tenu du débat politique actuel, et pourraient donc idéalement être mis en œuvre pour la première fois entre 2025 et 2030. Au cours de la quatrième phase, le comité scientifique recommande alors l’introduction de mesures réglementaires supplémentaires, parfois plus profondes (4l à 4p). Ces mesures sont actuellement encore très controversées et nécessitent d’abord le développement du processus de transformation pour être efficaces et politiquement réalisables. En particulier, le soutien apporté par le fonds de transformation à la transformation des modèles d’entreprise actuels et des processus de production et de transformation peut, à terme, réduire les coûts perçus et objectifs des mesures proposées dans le paquet IV pour différent·e·s actrices et acteurs tout au long de la chaîne de valeur. Ainsi, à ce stade, le débat peut se concentrer davantage sur les opportunités que sur les coûts de ces mesures. La voie d’action présentée ici peut être considérée comme une proposition de discussion du comité scientifique. Le message central que les expert·e·s aimeraient faire passer dans le débat avec leur proposition est qu’il faut une réorientation globale et stratégique de la politique alimentaire et agricole. Pour cela, les scientifiques estiment qu’il faudrait davantage considérer les synergies entre les mesures individuelles ainsi que leur séquence stratégique dans le temps au sein de paquets de mesures complets. Grâce à une telle conception politique globale, intégrée et stratégique, le comité scientifique estime que les opportunités (d’un point de vue économique, social, sanitaire et environnemental) l’emportent clairement sur les coûts de la transformation. La conception commune d’une voie d’action politique (voir chapitre 4 – Gouvernance du système alimentaire) permet en outre de lever les barrières existantes et de rendre visibles les avantages de la transformation pour les actrices et acteurs clés du système. La figure 6 sur la page suivante présente la voie d’action politique proposée et le tableau 3, p. 43 – 45, donne un aperçu des mesures dans les quatre paquets. Les mesures individuelles par paquet sont ensuite décrites et justifiées de manière plus détaillée. Les mesures individuelles sont identifiées par des couleurs correspondant à leur type de mesure : Mesures axées sur l’information et la formation en bleu Instruments d’incitation positive (c’est-à-dire aides et subventions publiques) en vert III Instruments d’incitation négative (c’est-à-dire taxes d’incitation publiques et autres instruments de marché visant à réduire les externalités négatives) en orange IV Instruments de régulation en rose I II 41 Une voie d’action pour la transformation du système alimentaire suisse Paquet I 1 2 Paquet II 3 4 5 1 2 Paquet III 3 4 Introduction du fonds de transformation Taxes d’incitation 1a – 2i Mesures réglementaires I 3a – 3g 1 3 Paquet IV 5 Mesures de politique agricole et soutien aux zones rurales 2j – 2m 1 3 4 5 Mesures réglementaires II 4l – 4p % de la transformation réalisée du système alimentaire suisse en accord avec les objectifs de développement durable (ODD) 4a – 4k 50 25 Durée 2022 Conditions Type de mesure 1 Prix Mesures axées sur l’information et la formation 2 Performance Instruments d’incitation positive 3 Accès Instruments d’incitation négative 4 Norme Instruments réglementaires 5 Capacité Fig. 6 : Fesenfeld et al. (2023)237 / Telek 42 2025 2030 Paquet de mesures I : Introduction d’un fonds de transformation Mesure 1a Potentiel de synergie Conditions du point de basculement Soutien Efficacité Urgence Campagnes d’information publiques qui communiquent à grande échelle les avantages et l’importance de la réduction des déchets alimentaires et d’une alimentation à base de plantes* 1b, 1c, 2b, 2f, 3a, 3b, 3d, 3e, 3g, 4a, 4e, 4f, 4o, 4h, 4p Norme, capacité 4,2 3,4 4,7 1b Offres de formation et de conseil pour les consom- 1a, 1c, 2b, 2f, 3a, 3b, 3d, 3e, 3g, 4a, 4e, 4f, 4o, 4h, 4p Norme, capacité 4,1 3,7 4,6 1c 2a, 2b, 2c, 2d, 2e, 2f, 2h, 3b, 3c, 3e, 4a, 4b, 4c, 4e, 4l, 4m, 4o, 4h, 4j, 4k Performance, norme, capacité 4,8 4,5 4,6 2a Promotion de la recherche et du développement 1a, 1b, 1c, 2b, 2f, 2g, 2h, 3a, 3b, 3d, 3e, 3g, 4a, 4b, 4e, 4f, 4o, 4h, 4p Prix, performance, accès, capacité 3,8 3,8 3,7 2b Promotion d’une alimentation durable et saine dans 1a, 1b, 1c, 2b, 2f, 2g, 2h, 3a, 3b, 3d, 3e, 3g, 4a, 4b, 4e, 4f, 4o, 4h, 4p Performance, accès, norme, capacité 4,4 4,3 4,5 2c Promotion du développement et de l’application de 1c, 2c, 2d, 2e, 2g, 2h, 2j, 2k, 3a, 3b, 3c, 3e, 3f, 3g, 4b, 4i, 4k Prix, performance, accès, capacité 3,8 3,7 3,8 2d Promotion du développement technique et de 1c, 2c, 2e, 2g, 2h, 2j, 2k, 3a, 3b, 3c, 3e, 4b, 4d Prix, performance, accès, capacité 3,8 3,7 4 2e Promotion de la sélection et de l’utilisation de va- 1c, 2c, 2d, 2g, 2h, 2j, 2k, 3a, 3b, 3e, 4j, 4k Prix, performance, capacité 4,3 4,1 3,8 2f 1a, 1b, 1c, 2b, 3a, 3d, 4a, 4c, 4e–g, 4o, 4h, 4p Accès, norme, capacité 4,1 3,8 4,1 2g Primes de compensation pour les reconversions 1a, 1b, 1c, 2b, 2c, 2d, 2e, 2h, 2j, 2k, 3a, 3b, 3c, 3d, 3e, 3f, 3g, 4a, 4b, 4f, 4g, 4h, 4i, 4j, 4l, 4p Prix, capacité 4,4 4,3 3,9 2h Programmes de soutien aux jeunes agricultrices et 1a, 1b, 1c, 2b, 2c, 2d, 2e, 2h, 2j, 2k, 3a, 3b, 3c, 3d, 3e, 3f, 3g, 4a, 4b, 4f, 4g, 4h, 4i, 4j, 4l, 4p Prix, accès, capacité 4,6 4,6 4,2 2i 1a, 1b, 1c, 3a, 4c, 4e Prix, accès, norme 3,4 3,4 4,2 matrices et les consommateurs dans différents contextes de vie et de quotidien (p. ex. dans la restauration hors domicile, dans les supermarchés, dans les établissements d’enseignement, dans les exploitations, etc.)* Programmes d’éducation et de formation pour les professions liées au système alimentaire (toute la chaîne de valeur) (notamment en cas de conversion, méthodes de production durables)* indépendants de substituts à la viande animale répondant à des normes sanitaires et environnementales minimales renforcées* la restauration hors domicile* la numérisation et de l’agriculture de précision* l’utilisation du biochar* riétés végétales et de races d’animaux de rente adaptées aux conditions locales, au lieu de se concentrer sur des variétés et des races purement performantes* Soutien financier pour les services spécialisés dans l’alimentation durable (pourcentages de postes et moyens pour les projets) dans les communes ainsi que soutien pour la mise en place de réseaux alimentaires locaux (p. ex. conseils de l’alimentation)* agricoles* agriculteurs pour la reconversion des exploitations vers une production plus durable et davantage axée sur la production végétale lors de la reprise d’une exploitation* Incitations fiscales pour les dons de denrées alimentaires provenant de la restauration, du commerce et de la transformation Tableau 3 : Fesenfeld et al. (2023)237 * indique des mesures dans le fonds de transformation moyenne de l’échelle 1–5 5 = soutien élevé / efficacité élevée / urgence élevée 43 Paquet de mesures II : Mesures réglementaires I et taxes d’incitation Mesure Potentiel de synergie Soutien Efficacité Urgence 3a Introduction d’une taxe éq. CO2 sur les denrées 1a, 1b, 1c, 2a, 2b, 2c, 2d, 2f, 2g, 2h, 2j, 2k, 3b, 3c, 3d, 3e, 3f, 3g, 4a, 4b, 4d, 4f, 4g, 4h, 4i, 4j, 4l, 4n, 4o, 4p Prix, performance, norme 4,2 4,4 3,7 3b Introduction d’une taxe sur les excédents d’azote, 1c, 2a, 2b, 2c, 2d, 2f, 2g, 2h, 2j, 2k, 3a, 3c, 3d, 3e, 3f, 3g, 4a, 4b, 4d, 4f, 4g, 4h, 4i, 4j, 4l, 4n, 4o, 4p Prix, performance, norme 4,3 4 4,2 3c Imposer plutôt que subventionner l’exploitation des 1c, 2a, 2b, 2c, 2d, 2f, 2g, 2h, 2j, 2k, 3a, 3c, 3d, 3e, 3f, 3g, 4a, 4b, 4d, 4f, 4g, 4h, 4i, 4j, 4l, 4n, 4o, 4p Prix, norme 3,9 3,9 3,6 3d Suppression du soutien public à la publicité et au 1a, 1b, 1c, 2b, 2f, 2g, 2h, 3a, 3b, 3e, 3g, 4a, 4b, 4e, 4f, 4o, 4h, 4p Prix, norme 4,2 3,6 4,3 3e Suppression du taux de TVA réduit pour les produits 1a–2h, 3g, 4a, 4n, 4p Prix, norme 4,4 3,9 4,2 3f Prix, performance, accès, norme 4,2 3,9 3,8 alimentaires combinée à une redistribution des recettes (montant limité dans un premier temps, mais trajectoire d’augmentation définie), sans compromis sur le bien-être animal d’ammoniac et de phosphore sols organiques (> 30 % d’humus) marketing pour les produits carnés polluants dans le secteur agricole 1c, 2c, 2d, 2g, 2h, 3a Supprimer progressivement le remboursement de l’impôt sur les huiles minérales tout en accordant un avantage fiscal et en encourageant de manière ciblée l’utilisation de biocarburants de troisième génération ainsi que l’utilisation de technologies neutres en CO2 et les énergies renouvelables (p. e. l’agrophotovoltaïque) 3g Augmentation des droits de douane sur les produits Prix, accès, 1a, 1b, 1c, 2b, 2f, 2g, 2h, 3a, 3b, 3d, 3e, 4a, 4b, 4e, norme 4f, 4o, 4h, 4p 4,4 4,3 4,2 4a Directives nationales pour les marchés publics et Prix, accès, 1a, 1b, 1c, 2b, 2f, 2g, 2h, 3a, 3b, 3d, 3e, 3g, 4b, 4e, norme 4f, 4o, 4h, 4p 4,3 4,2 4,3 4b Mise en place d’obligations de rapport plus com- 2c, 2d, 3a, 3b, 3c, 3e, 4c, 4d, 4f, 4g, 4o Capacité 3,8 3,6 3,8 4c Obligation de surveillance et de reporting pour les 1a, 1b, 1c, 2i, 4e Prix, norme, capacité 3,9 3,8 3,9 Norme, capacité 4 3,8 3,9 Prix, accès, norme, capacité 4,7 4,3 4,7 carnés, les aliments pour animaux importés et les engrais minéraux normes minimales de l’offre dans les cantines des entreprises publiques pour la promotion d’une alimentation saine/durable ainsi que des accords sectoriels correspondants avec la restauration collective d’entreprise plètes et d’une infrastructure de mesure et de données concernant différents indicateurs relatifs aux objectifs climatiques et environnementaux tout au long de la chaîne de valeur pertes de denrées alimentaires, basée sur la base légale existante de la loi sur la protection de l’environnement (LPE, art. 46) 4d Mention explicite des objectifs de réduction des émis- 1a, 1b, 2b, 2d, 3a, 3b, 3c, 3e, 3f, 4b sions provenant du système alimentaire dans les contributions nationales à la protection du climat (NDC) ainsi que des accords sectoriels pour l’introduction de Science-based Targets (SBT) obligatoires pour les entreprises suisses du secteur alimentaire et agricole 4e Réglementation plus souple de la durée de conser- vation des denrées alimentaires, transmission juridique simplifiée aux banques alimentaires, adaptation des normes de qualité et industrielles pour éviter les déchets alimentaires et application obligatoire du principe de prévention des déchets (LPE, art. 30) aux denrées alimentaires Tableau 3 : Fesenfeld et al. (2023)237 44 Conditions du point de basculement 1a, 1b, 1c, 2i, 4c moyenne de l’échelle 1–5 5 = soutien élevé / efficacité élevée / urgence élevée Mesure 4f Potentiel de synergie Conditions du point de basculement Soutien Efficacité Urgence 1a, 1b, 1c, 2b, 2f, 2g, 2h, 3a, 3b, 3d, 3e, 3g, 4a, 4b, 4e, 4o, 4h, 4p Norme, capacité 4 3,7 3,8 4g Meilleure utilisation du droit sur les cartels afin de 2g, 2h, 2l, 2m, 4k Prix, capacité 3,9 3,8 3,6 4h Adapter les directives de formation 1a, 1b, 1c, 2a–h, 4a, 4b, 4c, 4e, 4k Norme, capacité 4,3 4,2 4 4,3 4,1 4,2 2c, 2e, 2g, 2h, 2j, 2k, 3a, Prix, accès, 3b, 3d, 3g, 4i, 4k capacité 4,2 4,2 4,1 4g, 4m, 4n 4,2 4,1 3,9 Étiquetage public des denrées alimentaires concernant les effets sur l’environnement, la santé et le bien-être animal et normes minimales légales pour l’étiquetage privé des denrées alimentaires réduire l’asymétrie d’information et les écarts de pouvoir sur le marché et de faciliter les fusions entre productrices et producteurs 4i Abaissement du nombre d’unités de gros bétail-fumure 2c, 2e, 2g, 2h, 2j, 2k, 3a, Prix (UGBF) de 3 à 2,5 dans la loi sur la protection des eaux 3b, 3d, 3g, 4j, 4k 4j Adaptation de la loi sur l’aménagement du territoire, si possible dans les meilleurs délais, pas de nouvelles exploitations d’élevage hors-sol dans les zones d’agriculture intensive 4k Législation complémentaire qui met en œuvre le devoir de diligence des grands groupes (y compris les banques) conformément aux directives internationales (p. ex. les Principes directeurs des Nations Unies relatifs aux entreprises et aux droits de l’homme ou les Lignes directrices de l’OCDE) Accès, norme, capacité Paquet de mesures III : Mesures de politique agricole et soutien aux zones rurales 2j Adaptation des paiements directs, des mesures de 1c, 2b, 2c, 2d, 2e, 2g, 2h, 2k, 2l, 2m, 3a, 3b, 3c, 3e, 3f, 3g, 4b, 4h, 4i, 4j, 4k Prix, accès 4,5 4,4 4 2k Orienter les mesures commerciales vers des sys- 3g, 4m, 4n Prix, norme, capacité, accès 4,4 4,4 3,8 2l Introduction d’un impôt négatif sur le revenu pour la 1c, 2g, 2h, 2j, 2m Prix, accès 3,7 3,6 3,2 2m Renforcement des services psychosociaux dans les 1c, 2g, 2h, 2j, 2m Accès, norme, capacité 3,7 3,6 3,6 soutien du marché et des aides à l’investissement tèmes alimentaires durables catégorie Working Poor dans l’agriculture régions rurales Paquet de mesures IV : Mesures réglementaires II 4l Passage du Suisse-Bilanz à la méthode OSPAR par 2c, 2e, 2g, 2h, 2j, 2k, 3a, 3b, 3d, 3g, 4j Prix, capacité 4,1 3,9 3,9 4m Mesures de contrôle garantissant l’absence de tra- 4g, 4n Norme, capacité 4,3 3,7 3,9 4n Adaptation des accords commerciaux internationaux 2j, 2k, 4l, 4m Prix, norme, capacité, accès 4,4 4,1 3,6 4o Traçabilité accrue et informations plus précises des 1a, 1b, 1c, 2b, 4a, 4b Norme, capacité 4,1 3,8 3,8 4p Interdiction des promotions à bas prix sur les produits d’origine animale par le commerce de détail et les discounters 1a, 1b, 1c, 2b, 2f, 2g, 2h, 2j, 2k, 3a, 3b, 3d, 3e, 3g, 4a, 4b, 4e, 4f, 4h Prix, norme, capacité 4,3 4,2 4,1 étapes d’ici 2030 vail des enfants pour toutes les denrées alimentaires consommées en Suisse restaurateurs sur l’origine des produits, en particulier pour la viande. En outre, introduction d’un label d’État et d’une distinction pour la restauration répondant à des normes de durabilité élevées Tableau 3 : Fesenfeld et al. (2023)237 45 Paquet de mesures I : Introduction d’un fonds de transformation 1a Campagnes d’information publiques qui communiquent à grande échelle les avantages et l’importance de la réduction des déchets alimentaires et d’une alimentation à base de plantes* Brève description de la mesure : Des campagnes d’information publiques, qui atteignent les citoyen·ne·s par le biais des canaux médiatiques classiques (p. ex. affiches, télévision, journaux) et des médias sociaux, peuvent contribuer à modifier la conscience publique, les connaissances ainsi que les normes sociales en matière de comportement de consommation.146,160 Il est particulièrement important que les citoyen·ne·s soient informé·e·s de manière transparente et claire sur l’importance cruciale d’une alimentation plus végétale, en accord avec le Planetary Health Diet (cf. indicateur d’objectif 1) et la réduction des déchets alimentaires (cf. indicateur d’objectif 8). Des formats de campagne financés par l’État et aussi créatifs que possible devraient être développés à cet effet. Dans l’idéal, ces derniers devraient être visibles à bas niveau dans le quotidien des consommatrices et des consommateurs, en collaboration avec des partenaires (p. ex. des organismes de formation, des entreprises, des supermarchés, des restaurants, des cantines, etc.) (cf. par exemple la campagne Food Ninjas). Justification du choix de la mesure : La mesure agit surtout sur l’indicateur d’objectif 1 relatif à l’adaptation du mode d’alimentation, ainsi que sur l’indicateur d’objectif 8 relatif à la réduction des déchets alimentaires. Cette mesure peut également augmenter l’acceptation de mesures politiques profondes (p. ex. taxes environnementales sur les denrées alimentaires) au fil du temps et ainsi augmenter la faisabilité politique d’une transformation globale.146, 160 1b Offres de formation et de conseil pour les consom- matrices et les consommateurs dans différents contextes de vie et du quotidien (p. ex. dans la restauration hors domicile, dans les supermarchés, dans les établissements d’enseignement, dans les exploitations, etc.)* Brève description de la mesure : Les ressources du fonds de transformation devraient financer un programme d’encouragement pour le développement et la mise en œuvre de projets de formation et de conseil nutritionnels, visant en particulier les consommatrices et les consommateurs dans des contextes de vie réels, p. ex. la restauration hors domicile (en particulier la restauration, les cantines, etc.), dans les supermarchés et dans les écoles (p. ex. les jardins et les cuisines scolaires). Grâce à des approches éducatives innovantes dans un contexte d’achat et de consommation quotidiens, il est possible de renforcer les compétences culinaires pour planifier, composer et préparer une alimentation à base de plantes, ainsi que les compétences en vue de réduire les déchets alimentaires. Pour la mise en œuvre de cette mesure, il faudrait travailler en étroite collaboration avec les entreprises de restauration hors domicile, les supermarchés, les écoles, ainsi que les caisses d’assurance maladie (p. ex. développement des conseils nutritionnels). Le programme d’encouragement aux offres de formation et de conseil devrait être conçu de manière globale et définir clairement les lignes directrices et les critères de soutien. Le développement et la mise en œuvre des offres spécifiques de formation et de conseil sont ensuite réalisés en fonction des besoins et du contexte par des diététicien·ne·s solidement formé·e·s conformément à la LPSan et des organismes de formation, en coopération avec * indique des mesures dans le fonds de transformation 46 des entreprises de restauration hors domicile, des supermarchés et, le cas échéant, d’autres partenaires (p. ex. entreprises et caisses d’assurance maladie). Justification du choix de la mesure : La mesure agit surtout sur l’indicateur d’objectif 1 relatif à l’adaptation du mode d’alimentation ainsi que sur l’indicateur d’objectif 8 et indirectement sur l’indicateur d’objectif 7 relatif à la réduction des émissions de gaz à effet de serre. La recherche comportementale montre que l’enseignement de compétences culinaires (p. ex. les capacités de transformation et de cuisson) est un levier important pour rendre possible des changements de comportement alimentaire.161 La mesure peut donc avant tout modifier des compétences proches du quotidien ainsi que des normes sociales et permettre ainsi des changements de comportement des consommatrices et des consommateurs et du personnel des exploitations. 1c Programmes d’éducation et de formation pour les professions liées au système alimentaire (toute la chaîne de valeur) (notamment en cas de conversion, méthodes de production durables)* Brève description de la mesure : Le fonds de transformation doit permettre d’apporter un soutien supplémentaire au développement et à la mise en œuvre de programmes de formation initiale et de formation continue pour toutes les professions liées au système alimentaire (notamment dans l’agriculture, la transformation, la restauration, l’économie domestique, la technologie alimentaire, etc.). Pour qu’une transformation de la production agricole et du système alimentaire puisse réussir, les programmes d’enseignement de ces professions doivent être cohérents avec les objectifs d’un système alimentaire durable (cf. mesures 4k et 4h). Dans l’agriculture en particulier, les programmes de formation initiale et de formation continue doivent, dans le cadre de la réforme de la formation CFC actuellement en cours (jusqu’en 2024), mais aussi à des niveaux de formation plus élevés, mettre beaucoup plus l’accent sur les pratiques et les méthodes de production nécessaires pour ne plus dépasser les limites de charge écologique des ressources naturelles. Des débuts dans cette direction ont déjà été faits en partie, comme avec le projet CAAST, Community for Applied Agricultural Sustainability Teaching. Il s’agit notamment de mettre davantage l’accent sur une utilisation améliorée et plus efficace des nutriments, sur la réduction de l’utilisation des engrais minéraux, sur la réduction de l’utilisation des produits phytosanitaires de synthèse et sur l’utilisation de la lutte biologique contre les parasites ou de la protection non chimique des cultures en général. En outre, il faudrait proposer davantage d’offres de formation sur la culture de légumineuses, la gestion des pâturages et le pâturage intégral, le vêlage saisonnier, les méthodes de culture agroécologiques, entre autres l’agroforesterie, l’agriculture régénérative, la production intégrée, la numérisation, etc. Des modules de reconversion spécifiques et gratuits devraient également être développés et proposés aux agricultrices et aux agriculteurs qui cessent leur production animale, p. ex. pour se tourner vers la production végétale. Dans les métiers axés sur la transformation des denrées alimentaires, la préparation et le conseil, les ressources du fonds de transformation seront utilisées pour développer et promouvoir des techniques et des compétences axées sur la transformation et la préparation d’aliments d’origine végétale, et une utilisation réduite de produits d’origine animale. Un accent particulier devrait également être mis sur la formation initiale et continue des cuisinières et des cuisiniers et du personnel de la restauration individuelle et collective (cf. 2b). Justification du choix de la mesure : Les programmes de formation initiale et continue pour les professions liées au système alimentaire permettent de définir la voie à suivre pour l’orientation future de la production. Un grand nombre des mesures proposées ci-après exigent impérativement des adaptations des programmes d’enseignement, du matériel pédagogique et des cours de formation continue pour les enseignant·e·s. Les mesures proposées auront un impact important sur l’orientation de la production et sur les méthodes de production et de transformation, ce qui nécessite beaucoup de nouveau savoir-faire dans la mise en œuvre pratique afin que la transformation puisse réussir. La mesure agit surtout sur les indicateurs cibles 2-8. Il existe des synergies potentielles avec d’autres mesures visant directement ou indirectement à promouvoir les connaissances sur la culture et l’utilisation de denrées alimentaires végétales, et à diffuser des méthodes de production agricole durables. 2a Promotion de la recherche et du développement indépendants de substituts à la viande animale répondant à des normes sanitaires et environnementales minimales renforcées* Brève description de la mesure : Les moyens du fonds de transformation ainsi que les moyens existants pour la recherche alimentaire et agricole ainsi que pour la promotion de l’innovation (notamment le Fonds national suisse, Innosuisse, le domaine des EPF et Agroscope) devraient servir à promouvoir de manière ciblée la recherche et le développement de substituts à la viande animale répondant à des normes minimales plus élevées en matière de santé et d’environnement. Certes, de plus en plus de capitaux privés sont déjà investis dans le développement de substituts à la viande animale, mais le développement de ces produits n’est généralement pas lié à des normes de durabilité mesurables. Comme le montrent les exemples réussis de promotion publique ciblée de substituts durables dans d’autres pays,162 un programme d’encouragement public permet de créer efficacement un écosystème pour les start-ups et les innovations de produits. Avec sa forte densité d’excellents instituts de recherche, de bailleurs de fonds privés, de start-ups et d’entreprises alimentaires actives à l’échelle internationale, la Suisse peut jouer un rôle de leader mondial sur le marché en pleine croissance des substituts à la viande animale, tant du point de vue économique que du point de vue de la durabilité. Grâce à un soutien public ciblé dans le cadre du fonds de transformation et lié à des normes minimales mesurables en matière de santé et d’environnement, l’offre de substituts innovants et durables augmente sur le marché. Sans le soutien de l’État et des normes minimales couplées en conséquence dans le domaine de la santé et de l’environnement, le risque existe que les capitaux privés se dirigent davantage vers des innovations de produits qui sont avant tout lucratives sur le plan économique. S’appuyant sur des projets tels que le projet phare de l’ETH World Food System Center « Novel Proteins for Food and Feed »163 ainsi que des plateformes d’innovation existantes, telles que la Swiss Food & Nutrition Valley, il convient donc de mettre en place un programme d’encouragement public complet avec des normes minimales mesurables correspondantes. Justification du choix de la mesure : La mesure agit surtout sur l’indicateur d’objectif 1 relatif à l’adaptation du mode d’alimentation et indirectement sur les indicateurs d’objectif 2 à 7. En transmettant des compétences proches de la vie quotidienne, la mesure peut permettre des changements de comportement des consommatrices et des consommateurs et du personnel des exploitations. Il existe notamment un potentiel de synergie avec d’autres mesures qui s’adressent directement ou indirectement au comportement des consommatrices et des consommateurs, qui aident les exploitations à se tourner vers une production accrue de denrées alimentaires végétales et qui renchérissent la vente et la production de produits d’origine animale. 2b Promotion d’une alimentation durable et saine dans la restauration hors domicile* Brève description de la mesure : Outre la restauration hors domicile dans les restaurants, la restauration collective joue un rôle particulier dans le processus de transformation, avec près d’un million de convives par jour.164 Des mesures publiques peuvent s’adresser en particulier à l’offre de cantines dans les écoles, les hôpitaux et autres établissements publics. Les ressources du fonds de transformation devraient donc être utilisées pour introduire de manière généralisée une offre alimentaire durable et favorable à la santé dans la restauration hors domicile, en particulier dans la restau- ration collective des crèches, des écoles et d’autres établissements publics, en accord avec le Planetary Health Diet49 et l’indicateur d’objectif 1. Les ressources du fonds devraient être utilisées de manière ciblée afin d’adapter l’infrastructure d’achat (p. ex. modification de la présentation et de l’ordre de l’offre) et les connaissances issues de la recherche comportementale165, 166 dans le secteur de la restauration collective. Des offres d’encouragement correspondantes devraient également être mises à la disposition des prestataires privés dans le domaine de la restauration hors domicile. En outre, le fonds pourrait financer de manière ciblée des initiatives de formation continue en matière de développement durable pour le personnel chargé des achats et des cuisines dans le secteur de la restauration hors domicile (voir mesure 1c). Le conseil et l’encouragement ciblés des exploitations devraient être complétés à moyen terme par l’inscription dans la loi de normes minimales durables et favorables à la santé dans les cantines des établissements publics (cf. mesure 4a). Cette annonce légale pourrait déjà accélérer à court terme une adaptation de l’offre dans le secteur de la restauration hors domicile.165 Justification du choix de la mesure : La mesure agit surtout sur l’indicateur d’objectif 1 relatif à l’adaptation du mode d’alimentation et indirectement sur les indicateurs d’objectif 2 à 7. En adaptant l’offre et en transmettant des compétences proches de la vie quotidienne, la mesure peut permettre des changements de comportement des consommatrices et des consommateurs et du personnel des exploitations. Il existe notamment un potentiel de synergie avec d’autres mesures qui s’adressent directement ou indirectement au comportement des consommatrices et des consommateurs, qui aident les exploitations à se tourner vers une production accrue de denrées alimentaires végétales, qui réduisent les déchets alimentaires et qui renchérissent la vente et la production de produits d’origine animale. 2c Promotion du développement et de l’application de la numérisation et de l’agriculture de précision* Brève description de la mesure : Les innovations technologiques dans le domaine de la numérisation et de l’agriculture de précision ont le potentiel de réduire considérablement l’empreinte environnementale de notre agriculture et de notre alimentation, p. ex. l’utilisation de pesticides et d’engrais.167, 168 En outre, ces innovations ont également le potentiel de renforcer la place économique suisse. Cependant, en Suisse, le potentiel de développement et d’application des technologies de l’agriculture intelligente générant une valeur ajoutée écologique mesurable n’est pas encore pleinement exploité à l’heure actuelle et des interventions publiques ciblées sont donc nécessaires.169 Pour cette raison, les ressources du fonds de transformation devraient permettre de réduire les coûts des technologies correspondantes réduisant de manière mesurable l’empreinte environnementale de la production (p. ex. par des programmes d’encouragement de l’innovation ainsi que par le subventionnement de l’utilisation).169 En outre, le taux d’utilisation peut être atteint en augmentant la fiabilité de la technologie, en soutenant les agricultrices et les agriculteurs en cas de problèmes techniques et en les indemnisant pour les avantages environnementaux qu’ils apportent. Au fil du temps, la réduction des coûts et l’augmentation de la qualité de la technologie peuvent donc également être améliorées. L’augmentation de la fiabilité ainsi que l’assistance en cas de problèmes techniques nécessitent tout d’abord un programme d’impulsion de l’État qui soutiendra de manière ciblée le développement de la technologie ainsi que l’assistance technique lors de l’utilisation au cours des cinq prochaines années. En outre, les agricultrices et les agriculteurs peuvent être incité·e·s à utiliser les nouvelles technologies par des paiements directs pour des pratiques respectueuses de l’environnement ainsi que par des taxes d’incitation correspondantes. En ce qui concerne l’augmentation de la fiabilité (perçue) de la technologie, la formation et la promotion des connaissances technologiques constituent un levier important pour augmenter la probabilité d’utilisation de la technologie. Des analyses récentes en Suisse indiquent que les programmes d’information sur la technologie elle-même, mais aussi de sensibilisation à ses effets positifs sur l’environnement, augmentent les taux d’acceptation et d’utilisation chez les agricultrices et les agriculteurs.170 47 Justification du choix de la mesure : La mesure agit surtout sur les indicateurs cibles 2-7. Cette mesure est essentielle pour garantir une utilisation efficace des ressources naturelles limitées et pour réduire la pollution de l’environnement par l’utilisation de produits phytosanitaires et d’engrais. Il existe notamment des synergies potentielles avec d’autres mesures qui s’adressent directement ou indirectement au développement et à l’application de l’agriculture intelligente et d’autres technologies innovantes.40, 145 2d Promotion du développement technique et de l’utilisation du biochar* Brève description de la mesure : Afin d’améliorer encore les procédés techniques de stockage des émissions de gaz à effet de serre dans du biochar (installations de pyrolyse) et de les rendre moins chers, des moyens de recherche et de développement ciblés devraient être mis à disposition à partir des ressources du fonds de transformation. En collaboration avec les plateformes d’innovation existantes, telles que la Swiss Food & Nutrition Valley (2022), il faudrait en outre mettre en place un vaste programme public d’encouragement aux entreprises innovantes.171 En outre, les agricultrices et les agriculteurs devraient être motivé·e·s de manière ciblée par une formation continue, un soutien technique ainsi que par des paiements directs et des taxes d’incitation pour une utilisation optimale du biochar. Le soutien ciblé de l’État aux agriculteurs pour la vente de certificats d’éq. CO2 pourrait également augmenter l’utilisation du biochar. Justification du choix de la mesure : Selon une méta-étude d’Agroscope172, en 2050, jusqu’à 80 % des émissions de gaz à effet de serre qui subsisteraient alors en Suisse pourraient être compensés par l’utilisation de biochar. Outre le potentiel élevé de protection du climat, l’utilisation accrue de biochar entraîne aussi des effets positifs sur la croissance ainsi que sur la résilience des plantes et des animaux face aux parasites et aux risques climatiques, et peut avoir des effets positifs sur la santé. D’un point de vue économique, l’utilisation de biochar pourrait également constituer une source de revenus supplémentaire pour les agricultrices et les agriculteurs, qui pourraient ainsi compenser les émissions de gaz à effet de serre de manière transparente et mesurable, et être actifs sur le marché en pleine croissance de la compensation carbone. Cela pourrait également rendre la transition vers une production accrue d’aliments végétaux destinés à la consommation humaine plus intéressante économiquement pour les agricultrices et les agriculteurs. La mesure contribue notamment à la réalisation de l’indicateur d’objectif 7. De plus, elle contribue indirectement à l’indicateur d’objectif 9. 2e Promotion de la sélection et de l’utilisation de variétés végétales et de races d’animaux de rente adaptées aux conditions locales* Brève description de la mesure : La Suisse a encore beaucoup de retard à rattraper en ce qui concerne la promotion de la sélection et de l’utilisation de variétés végétales et de races d’animaux de rente adaptées aux conditions locales, entre autres les races à double usage, mais pas exclusivement. La sélection et l’utilisation ciblées de variétés végétales et de races d’animaux de rente adaptées aux conditions environnementales locales, présentant une grande efficacité d’utilisation des éléments nutritifs et un bon profil de résistance, ainsi que des rendements relativement bons, constituent la base de l’intégration des services écosystémiques pour l’intensification et la transformation écologiques de la production.173 En outre, la diversification de la production qui en résulte permet une plus grande résilience face aux événements météorologiques extrêmes et aux variations climatiques (cf. ODD 2, objectif 2.4.). 48 Justification du choix de la mesure : La culture de variétés à haut rendement dans la production végétale et l’élevage de races à haut rendement dans la production animale sont liés à une utilisation importante d’intrants externes, notamment des aliments pour animaux, des produits phytosanitaires, des engrais chimiques, qui est rendue possible uniquement grâce à la disponibilité (bon marché) et à l’utilisation importante de ressources fossiles.174 Cette utilisation d’intrants externes garantit certes des rendements élevés, mais elle entraîne souvent une surexploitation massive des ressources naturelles au niveau régional, avec des conséquences graves pour l’environnement à long terme.175 De plus, l’élevage d’animaux favorise des systèmes d’élevage en partie douteux sur le plan éthologique et éthique. La mesure est liée aux indicateurs d’objectifs 2–7. 2f Soutien financier pour les services spécialisés dans l’alimentation durable (pourcentages de postes et moyens pour les projets) dans les communes ainsi que soutien pour la mise en place de réseaux alimentaires locaux (p. ex. conseils de l’alimentation)* Brève description de la mesure : En Suisse, environ 85 % de la population totale vit dans les villes et les agglomérations. En conséquence, une grande partie de la consommation a lieu dans les régions urbaines. En tant qu’acteurs, les communes peuvent rendre leurs systèmes alimentaires urbains plus durables, notamment par le biais de la restauration collective publique, de campagnes de sensibilisation telles que les semaines du goût ou en soutenant des initiatives citoyennes initiées par la société civile et, le cas échéant, en créant des réseaux alimentaires locaux et des conseils de l’alimentation. Afin de mieux exploiter la thématique des « systèmes alimentaires urbains et régionaux » ainsi que les potentiels dont disposent les communes en tant qu’acteurs de la conception, il convient de soutenir financièrement les services spécialisés dans l’alimentation durable existants ou à créer dans les communes. Les fonds peuvent également être utilisés pour financer directement des initiatives citoyennes qui apportent une contribution essentielle à l’amélioration du système alimentaire ou pour participer à la mise en place de réseaux alimentaires locaux et urbains (cf. p. ex. les forums alimentaires de Berne, Bâle et Zurich). Ces réseaux alimentaires locaux constituent souvent un espace d’expérimentation important et sont un laboratoire réel pour des innovations efficaces, afin d’accélérer la transformation du système alimentaire et d’impliquer des acteurs clés tout au long de la chaîne de valeur dans le processus.176, 177 Justification du choix de la mesure : La capacité des pouvoirs publics à promouvoir activement et efficacement la transition vers des systèmes alimentaires urbains durables dans les communes dépend en grande partie de leurs possibilités financières. S’il n’existe pas de mandat de prestations légal avec un cadre budgétaire correspondant, c’est-à-dire si les moyens sont plutôt limités, le sujet est traité de manière insuffisante malgré l’urgence. Alors que les grandes villes, comme Lausanne, Zurich ou Berne, disposent de leur propre service spécialisé dans l’alimentation durable, d’autres villes ne disposent pas des ressources financières nécessaires pour pourvoir un tel service. Les projets ascendants de la société civile ont également besoin d’un soutien financier en fonction de leur complexité, notamment dans la phase de mise en place, par exemple des conseils de l’alimentation. La mesure est directement liée à l’indicateur d’objectif 1 et indirectement à la plupart des autres indicateurs d’objectif. 2g Primes de compensation pour les reconversions agricoles* Brève description de la mesure : Pour réduire par exemple les charges en bétail en Suisse à un niveau durable, des infrastructures qui fonctionnent encore doivent être mises hors service, ce qui entraîne des pertes inacceptables pour les exploitations. Il en va de même pour l’abandon de la gestion des sols organiques. C’est pourquoi des primes de compensation devraient être versées aux agricultrices et aux agriculteurs pour des reconversions agricoles correspondantes qui augmentent de manière significative et mesurable la durabilité de l’exploitation. Par exemple, conformément aux précédents paquets de mesures178, 179, en cas de fermeture de bâtiments d’élevage de moins de 30 ans, les agricultrices et les agriculteurs peuvent bénéficier d’une certaine compensation si la charge en bétail est supérieure à 0,5 UGB/ha. Dans ce cas, les primes compensatoires devraient être limitées aux exploitations qui dépassent au maximum 2,5 UGBD et qui doivent recourir à HODUFLU (programme Internet de gestion harmonisée des transferts d’engrais de ferme et de recyclage dans l’agriculture) (cf. mesure 4i). En outre, les exploitations agricoles devraient recevoir des primes conditionnelles si elles se tournent davantage vers la culture végétale. La promotion ciblée de nouvelles possibilités de création de valeur durable, p. ex. la combinaison de la production végétale et de l’agrophotovoltaïque ainsi que l’utilisation et la production de biochar, devrait faire l’objet d’une attention particulière. Justification du choix de la mesure : La mesure contribue à l’objectif de rendre la structure agricole suisse plus durable, sans que les agricultrices et agriculteurs aient à supporter des dépenses supplémentaires non compensées du fait de la non-utilisation des investissements. Les primes peuvent également augmenter l’acceptation du processus de transformation dans le secteur. La mesure contribue donc à la réalisation des objectifs 2, 3, 5, 6 et 7 et facilite indirectement la réalisation de l’objectif 1. 2h Programmes de soutien aux jeunes agricultrices et agriculteurs pour la reconversion des exploitations vers une production plus durable et davantage axée sur la production végétale lors de la reprise d’une exploitation* Brève description de la mesure : Les transformations nécessaires dans le secteur agricole impliquent d’emprunter de nouvelles voies, et certaines le sont déjà, notamment par la jeune génération de gestionnaires d’exploitation. Les adaptations nécessaires au niveau de l’exploitation entraîneront des besoins d’investissement, par exemple dans la technologie de production de plantes protéagineuses, dans l’utilisation et la production de biochar, ainsi que dans l’utilisation de technologies de l’agriculture intelligente et de l’agrophotovoltaïque. L’aide fédérale actuelle aux investissements se concentre sur la construction de nouvelles étables. Nous proposons ici une réaffectation de ces fonds en faveur des technologies durables et de la mise en œuvre de modes de production agroécologiques, qui impliquent une plus grande concentration sur la culture végétale. Outre les bonifications d’intérêts, des contributions à fonds perdus doivent absolument être accordées. Si les fonds disponibles sont suffisants, la limite d’âge doit être appliquée avec souplesse. Outre ses effets positifs sur l’environnement, cette mesure doit permettre de lutter efficacement contre le changement structurel et la disparition des fermes dans l’agriculture. Les jeunes agricultrices et agriculteurs doivent être motivé·e·s et soutenu·e·s pour accompagner le processus de transformation de manière active et autonome. Justification du choix de la mesure : Cette mesure peut contribuer de manière importante à renforcer encore la capacité d’innovation de l’agriculture et à soutenir de manière ciblée les agricultrices et agriculteurs dans le processus de transformation. Les jeunes agricultrices et agriculteurs, en particulier, peuvent ainsi être motivé·e·s à miser sur une production plus durable et davantage axée sur la culture végétale lors de la reprise d’une exploitation. Cela peut favoriser les dépendances au sentier positives, la capacité et le changement de norme. En outre, la qualité des technologies respectueuses de l’environnement peut être améliorée et leur prix réduit si ces technologies sont utilisées plus largement et donc constamment améliorées dans des conditions d’application réelles. La mesure se rapporte en particulier aux objectifs 2 à 7 ainsi qu’à l’objectif 9.. 2i Incitations fiscales pour les dons de denrées alimentaires provenant de la restauration, du commerce et de la transformation Brève description de la mesure : Les entreprises qui font don de denrées alimentaires à titre gracieux à des institutions caritatives ou à des organisations de distribution peuvent faire valoir des déductions fiscales et compenser ainsi, du moins en partie, les dépenses supplémentaires en personnel et en logistique. Justification du choix de la mesure : Le don gratuit de denrées alimentaires à des institutions d’aide alimentaire ou à des associations telles que Foodsharing représente pour les entreprises un surcroît de travail dans la gestion des denrées alimentaires en comparaison avec leur élimination dans les ordures ménagères ou dans les installations de biogaz. De nombreuses denrées alimentaires sont donc jetées alors qu’elles pourraient être utilisées de manière irréprochable. La Suisse donne nettement moins par habitant·e que les pays voisins (1–2 kg/hab./an, en Italie : selon l’interview avec le Banco Alimentare, 5-6 kg/hab./an). En France, 60 % de la valeur comptable des denrées alimentaires donnés peuvent être directement déduits des impôts, en Espagne 35 % et au Portugal jusqu’à 140 %. Un rapport de la Commission européenne (2017) confirme que les incitations fiscales pour les entreprises, telles qu’elles existent dans certains États membres (p. ex., la France, l’Espagne et le Portugal), ont un effet positif démontré sur le don de denrées alimentaires excédentaires par l’industrie.180. Ce thème a été repris en Suisse par la motion Hegglin. Le Conseil fédéral approuve le contenu de la demande, mais renvoie au plan d’action contre le gaspillage alimentaire qui était alors en cours d’élaboration. Le comité scientifique recommande la mise en œuvre rapide de la mesure, qui a un effet positif direct sur la réalisation de l’objectif 7. Paquet de mesures II : Mesures réglementaires I et taxes d’incitation 3a Introduction d’une taxe éq. CO2 sur les denrées alimentaires combinée à une redistribution des recettes (montant limité dans un premier temps, mais trajectoire d’augmentation définie), sans concession sur le bien-être animal Brève description de la mesure : L’introduction d’une taxe éq. CO2 sur les denrées alimentaires avec un prix de départ de 120 francs par tonne d’équivalent CO2 en 2025 et une augmentation de prix fixée chaque année jusqu’en 2030 (en accord avec les objectifs de réduction des émissions, voir chapitre 2) a un effet incitatif direct sur le comportement de consommation et peut réduire sensiblement la consommation de denrées alimentaires nocives pour le climat et la santé.181-183 La taxe éq. CO2 a des répercussions sur la consommation et concerne donc également les produits importés, ce qui ne réduit donc pas l’avantage concurrentiel du secteur agroalimentaire suisse. Au contraire, le secteur agroalimentaire suisse peut se positionner comme le pionnier international d’une production alimentaire respectueuse du climat grâce à son avantage géographique dans le domaine de la production animale basée sur les herbages, ainsi qu’à un savoir-faire technologique de haut niveau et à un soutien gouvernemental et scientifique adéquat. Cette situation pourrait se traduire dès aujourd’hui mais également à l’avenir par des avantages concurrentiels sur le plan international. Cette mesure ne doit toutefois pas entraîner une diminution du bien-être animal. Il faut donc exclure le recours, par exemple, à des procédés techniques de réduction du CO2 dans le cadre d’un élevage exclusivement en stabulation, qui empêcheraient les animaux d’évoluer à l’air libre. Il est possible de favoriser un élevage adapté au site 49 et respectueux du bien-être animal, conforme à la taxe éq. CO2, en combinant de manière stratégique les autres mesures recommandées dans ce guide, notamment les mesures de soutien ciblées pour les agricultrices et les agriculteurs (cf. p. ex. 2g, 2h et 2j), les réglementations ciblées (cf. p. ex. 4i, 4j, 4k), les autres taxes d’incitation (p. ex. 3b), ainsi que les mesures douanières et commerciales adaptées (p. ex. 2j et 3g). Afin de minimiser les effets régressifs de la mesure sur les groupes à faibles revenus, il faudrait qu’au moins 60 pour cent des recettes de la taxe soit reversées aux consommatrices et consommateurs disposant de revenus faibles et moyens.184 Le remboursement sous la forme d’un coupon climat-alimentation pour les revenus faibles et moyens pourrait par exemple constituer une option. Cela permettrait d’alléger la charge qui pèse sur les groupes à faibles revenus tout en augmentant l’effet incitatif de la mesure. Les 40 pour cent restants des recettes pourraient être utilisés pour refinancer le fonds de transformation recommandé. À terme, la mesure pourrait également être étendue à une taxe environnementale plus complète (basée notamment sur l’impact climatique, l’écotoxicité, la toxicité humaine, l’eutrophisation, les besoins en ressources, l’acidification). Une telle taxe environnementale est nettement plus complexe sur le plan technique et administratif qu’une simple taxe éq. CO2.185 L’Union européenne (UE) travaille actuellement à la mise en place d’un système d’évaluation correspondant et la Suisse pourrait, le cas échéant, collaborer avec l’UE dans ce domaine dans le cadre des accords bilatéraux. Selon une étude d’Agroscope, les processus en cours au sein de l’UE pour la mesure de l’empreinte environnementale des produits (PEF/EEP) ainsi que le protocole « Environmental Assessment of Food and Drink » contribuent considérablement à l’harmonisation de la déclaration environnementale sur les produits agricoles.187 Cependant, il faudra poursuivre le développement de normes uniformes en matière de mesure, de données et de suivi pour qu’une taxe environnementale globale puisse être mise en œuvre.187 Justification du choix de la mesure : Cette mesure a une incidence directe sur les indicateurs cible 1 et 7 et un impact indirect sur l’indicateur cible 8. La succession et la combinaison stratégiques de mesures dans le cadre du fonds de transformation ainsi que des mesures adaptées en matière de commerce extérieur peuvent augmenter l’impact et la faisabilité de la taxe éq. CO2145, 152 Si la visibilité de la taxe éq. CO2 est élevée, il pourrait en résulter, outre l’effet d’incitation économique, un signal positif en termes d’information et de norme sociale. Il est possible d’accroître l’acceptation sociale si les éventuels effets régressifs de la taxe sont compensés par des indemnisations financières sociales visibles (p. ex un coupon climat pour les faibles revenus) et si, au préalable, les produits alternatifs (notamment les substituts aux produits d’origine animale) et les options de création de valeur sont renforcés dans le cadre du fonds de transformation.156 Une communication stratégique des principaux décideurs politiques peut également permettre une meilleure acceptation sur le plan politique.160 3b Introduction d’une taxe sur les excédents d’azote, d’ammoniac et de phosphore Brève description de la mesure : Les économistes s’accordent à dire que les externalités négatives doivent être internalisées afin d’optimiser le bien-être de la société dans son ensemble. Les émissions d’excédents d’azote, d’ammoniac et de phosphore constituent des externalités négatives classiques. Il convient donc de les taxer rapidement. Pour que la taxe d’incitation ait un impact, nous proposons une taxe d’un franc par kilogramme d’azote ou de phosphore excédentaire. La méthode OSPAR permet de calculer l’excédent de manière concrète. La taxe est prélevée en déduction des paiements directs.188 Justification du choix de la mesure : Les simulations189 nous ont montré qu’il ne faut pas surestimer l’effet d’une taxation sur les engrais. Dans le cadre d’une approche théorique du bien-être, une situation avec taxe est néanmoins préférable à une situation sans taxe. Étant donné qu’elle ne s’applique pas aux denrées alimentaires importées, il convient de rechercher des compensations en matière de politique commerciale. La mesure contribue à la réalisation des objectifs 1 à 8. 50 3c Taxation sur la gestion des sols organiques (> 30 % d’humus) en lieu et place des subvention Brève description de la mesure : Alors que le stock de carbone d’une terre arable classique est d’environ 50 tonnes de carbone par hectare, cette valeur est au moins dix fois plus élevée pour les sols marécageux.190 Le drainage et l’exploitation agricole permettent de transformer ce type de sol organique en une terre arable ordinaire, moyennant toutefois des rejets considérables du carbone stocké. Cette pratique est aujourd’hui soutenue par des contributions à la sécurité de l’approvisionnement et des contributions au paysage cultivé. De même, les tourbières laissées à l’état naturel contribuant positivement au bilan carbone, il est fortement recommandé de ne plus encourager la gestion des sols organiques avec des fonds fédéraux. En lieu et place, les agricultrices et les agriculteurs devraient être tenu·e·s de payer le prix du marché pour les émissions de dioxyde de carbone. Il faudrait toutefois que les exploitations bénéficient d’un soutien ciblé pour compenser l’augmentation des coûts liés à la conversion des exploitations (cf. primes de conversion dans le cadre du fonds de transformation). Justification du choix de la mesure : Alors que les paiements directs sont habituellement justifiés par les externalités positives de l’agriculture, des processus de transformation aux externalités massivement négatives sont aujourd’hui également subventionnés. Il est nécessaire de mettre fin à cette pratique. Le processus de conversion représente toutefois un enjeu pour l’ensemble de la société et devrait donc être accompagné de manière ciblée par des mesures de compensation et des primes de conversion. 3d Suppression du financement public destiné à la publicité et au marketing des produits carnés Brève description de la mesure : Proviande reçoit environ six millions de francs par an de la part de la Confédération afin de financer la publicité pour la viande. Alimenter la publicité pour des produits ayant des externalités négatives par des moyens fiscaux va à l’encontre des objectifs de transformation. Il convient donc de mettre fin à ces aides financières dans les plus brefs délais. Justification du choix de la mesure : Il s’agit d’un montant relativement faible, dont l’économie n’a par conséquent qu’un effet limité. Renoncer à promouvoir la consommation de viande suisse peut constituer un geste symbolique de la part de l’État qui pourrait toutefois avoir des conséquences majeures sur le développement de normes sociales et de modèles de consommation, raison pour laquelle cette mesure est proposée. Le positionnement de la viande suisse par rapport à la viande importée ne suffit pas à justifier le maintien de cette subvention, car la publicité en question ne fait qu’encourager la consommation de produits carnés de manière générale. La subvention actuelle est donc en contradiction avec l’indicateur cible 1 et doit faire l’objet d’une modification. 3e Suppression du taux de TVA réduit pour les produits polluants dans le secteur agricole Brève description de la mesure : Les denrées alimentaires et les consommations intermédiaires de l’agriculture sont aujourd’hui toutes soumises à un taux de TVA réduit de 2,5 au lieu de 7,7 pour cent. Cette mesure est tout à fait justifiée pour les intrants et les aliments respectueux de l’environnement. Cependant, les raisons pour lesquelles les intrants polluants, tels que les produits phytosanitaires et les engrais minéraux, profitent également de cette réduction sont difficiles à comprendre. En moyenne, les aliments d’origine animale entraînent eux aussi une augmentation de la pollution, ce qui explique pourquoi il faut éviter de fixer un faible taux de taxation. Il est donc recommandé d’appliquer le taux de taxation normal de 7,7 pour cent à ces produits, ainsi qu’au commerce d’animaux. Justification du choix de la mesure : Il existe aujourd’hui encore des distorsions du marché dues à des réglementations politiques favorisant des produits nocifs pour l’environnement. En conséquence, les scientifiques recommandent de recourir à la TVA pour internaliser les externalités dans le secteur alimentaire.183, 191 En raison du taux de TVA relativement faible de 7,7 pour cent, l’effet d’incitation attendu de cette mesure sur l’achat de produits polluants reste modéré. Cette mesure permet également de mettre d’autres moyens de financement publics à la disposition du fonds de transformation. Cette mesure se rapporte aux objectifs 1 et 2 et présente des synergies positives avec toutes les mesures proposées dans le cadre du fonds de transformation. 3f) Supprimer progressivement le remboursement de l’impôt sur les huiles minérales tout en accordant un avantage fiscal et en encourageant de manière ciblée l’utilisation de biocarburants de troisième génération ainsi que l’utilisation de technologies neutres en CO2 et les énergies renouvelables (p. e. l’agrophotovoltaïque) Brève description de la mesure : La mesure prévoit la suppression progressive du remboursement de l’impôt sur les huiles minérales jusqu’en 2027. Afin que les agricultrices et les agriculteurs ne subissent pas d’augmentation de leurs coûts énergétiques et qu’un approvisionnement énergétique sûr et neutre pour le climat soit garanti, il convient dans le même temps d’encourager de manière ciblée, par le biais de la fiscalité et dans le cadre du fonds de transformation, le recours aux biocarburants de troisième génération (p. ex. à base d’algues) pour une utilisation exclusive dans l’agriculture ainsi que l’utilisation de technologies neutres en CO2 (p. ex. le recours à l’agrophotovoltaïque).192 Justification du choix de la mesure : Actuellement, l’agriculture demeure fortement dépendante des sources d’énergie fossiles. Pour atteindre les objectifs climatiques (objectif 7), il est donc nécessaire de réduire rapidement les sources d’énergies fossiles. Le remboursement de l’impôt sur les huiles minérales va à l’encontre de cet objectif. Afin d’aider les agricultrices et les agriculteurs à se tourner progressivement vers l’utilisation d’énergies renouvelables, il est toutefois nécessaire de promouvoir de manière ciblée ces alternatives et ces technologies respectueuses de l’environnement. Dans le meilleur des cas, le recours à ces technologies permet aux exploitations agricoles de bénéficier d’une nouvelle source de revenus. Les agricultrices et les agriculteurs ne se contentent plus de consommer, mais produisent aussi également des énergies renouvelables. Leur formidable esprit d’innovation leur permet également de faire évoluer positivement le progrès technologique. 3g Augmentation des droits de douane sur les produits d’origine animale, les aliments pour animaux importés et les engrais minéraux Brève description de la mesure : Les droits de douane sur les importations d’aliments pour animaux, en particulier les aliments concentrés riches en protéines, devraient être augmentés afin d’accroître les coûts de production des exploitations d’élevage ne disposant pas de leur propre base fourragère, ce qui entraînerait une baisse de la rentabilité de ces exploitations. Pour l’importation d’aliments pour animaux, on applique le système de prix-seuils et il n’y a pas de restriction des quantités importées. Actuellement, des droits de douane faibles ou inexistants sur les importations d’aliments pour animaux (à nouveau réduits depuis mars 2022) continuent de favoriser l’augmentation des cheptels sans base fourragère propre. Les aliments pour animaux riches en protéines, comme le soja et les huiles végétales pour animaux, peuvent être importés en franchise de droits de douane depuis 2011, même si un droit de douane de 39 francs par 100 kilos est conforme aux règles de l’OMC.106 La hausse du coût de la production due à des droits de douane plus élevés ne doit pas être compensée par des importations de viande bon marché. En conséquence, il convient également d’augmenter les droits de douane sur les importations de viande afin de préserver la production de viande suisse adaptée aux conditions locales. Les engrais minéraux étant également à l’origine des excédents d’azote, il est également nécessaire d’augmenter les droits de douane sur les importations d’engrais minéraux. Justification du choix de la mesure : La hausse des importations d’aliments pour animaux entraîne un recul de la production animale partiellement indépendante du sol ainsi que des effectifs de porcs et de poulets d’élevage, et favorise, pour la production bovine et laitière, une production basée sur les herbages et sur la base fourragère propre. Si les droits de douane sur la viande importée sont plus élevés, la production nationale adaptée aux conditions locales reste préservée. Des campagnes d’information et des contrôles stricts aux frontières doivent permettre de lutter contre le risque de tourisme d’achat (cf. mesure 1a). Afin de pouvoir accélérer la réduction des excédents d’azote élevés en Suisse, il est nécessaire de réduire les importations d’aliments pour animaux (35 pour cent des excédents d’azote) et d’augmenter les prix des importations d’engrais minéraux (31 pour cent des excédents d’azote)142. Ces mesures contribuent à la réalisation des objectifs 1 à 8. 4a Directives nationales pour les marchés publics et normes minimales de l’offre dans les cantines d’entreprises publiques pour la promotion d’une alimentation saine/durable ainsi que des accords sectoriels correspondants avec la restauration collective d’entreprise Brève description de la mesure : Les règles régissant les marchés publics dans le domaine alimentaire devraient être complétées de manière à ce que : 1) les pouvoirs publics puissent, mais aussi soit tenus, de s’approvisionner en produits durables ; 2) les critères d’achat durable soient définis de manière claire et non discriminatoire ; 3) des procédures ouvertes de reconnaissance par l’État contribuent à garantir la qualité des certificats privés. Il faut notamment tenir compte de la nécessité de passer à un mode d’alimentation plus végétal et de consommer en Suisse davantage d’aliments produits de manière durable issus des pays Sud. Il faudrait notamment définir des directives au niveau national pour les marchés publics ainsi que des normes minimales pour l’offre dans les cantines des entreprises publiques afin de promouvoir une alimentation saine et durable. Le comité scientifique recommande notamment la mise en place de directives nationales obligeant les cantines publiques à acheter et à proposer au moins 50 pour cent d’aliments d’origine végétale d’ici 2025 et à atteindre un minimum de 75 pour cent d’aliments d’origine végétale d’ici 2030. Il faudrait également élaborer des directives supplémentaires pour l’achat de denrées 51 alimentaires saines et durables en collaboration avec les exploitants de cantines et les autorités cantonales et communales. Si cette démarche implique des coûts supplémentaires en matière de formation et d’approvisionnement, il convient d’examiner dans quelle mesure les ressources du fonds national de transformation peuvent être utilisées (voir 1c). Afin que l’efficacité de ces mesures ne se limite pas aux marchés publics, il faudrait que la Confédération développe également des accords de branche avec la restauration collective d’entreprise. Justification du choix de la mesure : Près d’un million de personnes mangent chaque jour dans la restauration collective.164 Les cantines des entreprises publiques font notamment figure d’exemple et peuvent influencer de manière déterminante aussi bien l’offre culinaire que les normes sociales permettant d’atteindre le Planetary Health Diet et l’objectif 1 (voir chapitre sur les objectifs). Une étude publiée dans la prestigieuse revue scientifique PNAS165 révèle notamment qu’une augmentation de 50 pour cent des plats d’origine végétale dans les cantines publiques peut accroître la consommation d’aliments d’origine végétale jusqu’à 80 pour cent (ibid.). L’effet est particulièrement prononcé chez les personnes qui consomment beaucoup de viande. En s’appuyant sur les expériences menées dans des cantons et des villes comme Zurich, Berne et Bienne, qui ont adopté des directives et des réglementations pour une alimentation durable et saine au sein des cantines publiques, il est possible d’édicter des directives nationales contraignantes au niveau fédéral et de poursuivre leur concrétisation et leur mise en œuvre au niveau des cantons et des communes. Il faudrait surtout que la proportion d’aliments d’origine végétale soit déjà fixée de manière contraignante au niveau national. Cette mesure peut avoir des effets de rétroaction politiques positifs145 et accroître la prise de conscience et la volonté de prendre des mesures politiques de grande envergure en matière de consommation.156 Cette mesure a un impact direct sur l’objectif 1 et indirect sur les objectifs 2 à 7. 4b Développement d’obligations de déclaration plus complètes ainsi que d’une infrastructure de mesures et de données concernant différents indicateurs relatifs aux objectifs climatiques et environnementaux tout au long de la chaîne de valeur Brève description de la mesure : Comme indiqué au chapitre 2, une révision continue et précise des objectifs n’est possible que sur la base de données fiables et actualisées. De même, il est aujourd’hui difficile de faire la distinction entre les méthodes de production respectueuses de l’environnement et celles qui ne le sont pas, notamment en raison de l’absence de nombreuses bases de données. Ce constat se vérifie notamment au niveau du lien entre l’utilisation d’intrants chimiques et la qualité de l’eau et de l’air, ainsi que pour le lien entre le traitement et la qualité des sols. Parallèlement, la numérisation permet aux entreprises de collecter des données à grande échelle tout au long de la chaîne de création de valeur, sans que cela n’entraîne un surcroît de travail disproportionné pour les exploitations concernées. La collecte et le traitement systématiques et obligatoires de ces données par les entreprises tout au long de la chaîne de valeur en ce qui concerne les principaux indicateurs environnementaux et climatiques devrait bénéficier d’un soutien technique, administratif et financier ciblé de la part de la Confédération. Justification du choix de la mesure : Une base de données solide est la condition préalable à la mise en œuvre et au contrôle fiables des nombreux objectifs et mesures formulés ici. Si, par exemple, les teneurs en nitrates dans l’eau potable sont mises en perspective avec les apports d’azote des exploitations agricoles locales, il est possible de mettre en place des mécanismes qui réagissent aux problèmes de qualité de l’eau par des restrictions de fertilisation à court terme. La mesure concerne donc en particulier les indicateurs cibles 2 à 7 et présente des synergies avec la plupart des mesures recommandées. 52 4c Obligation de surveillance et de reporting pour les pertes de denrées alimentaires sur la base légale existante de la loi sur la protection de l’environnement (LPE, art. 46) Brève description de la mesure : Les entreprises des secteurs de l’agriculture, du commerce de gros et de détail, de la transformation et de la restauration seront tenues de rendre compte des pertes de denrées alimentaires subies dans leurs entreprises ainsi que des mesures mises en œuvre et prévues pour éviter les pertes de denrées alimentaires (par analogie à l’obligation de reporting pour les emballages de boissons). Les exigences imposées aux acteurs peuvent être adaptées à la branche et à la taille de l’entreprise et doivent être proportionnées afin de servir l’objectif de réduction de moitié des pertes alimentaires évitables d’ici 2030. Justification du choix de la mesure : De nombreux acteurs du secteur agroalimentaire qui produisent des déchets alimentaires ne sont pas conscients de l’ampleur de leurs pertes et du potentiel d’économies économiques et écologiques. Une obligation de surveillance et de reporting sur les pertes des denrées alimentaires génère une prise de conscience du problème et une volonté d’agir en prenant des mesures. Elle permet également de coordonner et de hiérarchiser les mesures au niveau national en améliorant l’accessibilité et la qualité des données. Et elle est équitable, car elle fournit une base d’action commune à toutes les entreprises. Cette mesure a un effet positif immédiat sur l’indicateur cible 8. 4d Mention explicite des objectifs de réduction des émissions du système alimentaire dans les contributions nationales à la protection du climat (CDN) ainsi que des accords sectoriels pour l’introduction de Sciencebased Targets (SBT) obligatoires pour les entreprises suisses du secteur alimentaire et agricole Brève description de la mesure : Actuellement, les contributions déterminées au niveau national (CDN) prévues dans le cadre de l’accord de Paris n’indiquent pas explicitement d’objectifs de réduction par la Suisse pour les émissions de gaz à effet de serre provenant du système alimentaire (c’est-à-dire y compris les émissions importées). Cela devrait intervenir lors de la prochaine révision des CDN. Par ailleurs, la Confédération devrait, dans le cadre d’accords sectoriels, obliger les entreprises suisses du secteur alimentaire et agricole à adopter des Science Based Targets. Ces objectifs pourraient être établis dans un premier temps pour l’empreinte carbone des entreprises, avant d’être étendus à des indicateurs supplémentaires, afin d’empêcher le dépassement des différentes limites de la planète.1 Justification du choix de la mesure : Une mention explicite des objectifs de réduction pourrait avoir valeur d’exemple au niveau international et encourager d’autres pays à indiquer des objectifs de réduction pour leurs émissions (souvent) importées dans le secteur alimentaire. Cela permettrait également une meilleure équité mondiale dans la répartition de la charge de la réduction des émissions, étant donné que des pays historiquement riches comme la Suisse sont responsables de bien plus d’émissions que les pays en développement et importent encore aujourd’hui près des deux tiers de leurs émissions liées au secteur alimentaire.193 Les accords sectoriels visant à rendre obligatoire l’introduction de Science Based Targets permettent également aux entreprises d’exploiter leur capacité d’innovation pour réduire leur empreinte tout au long de la chaîne de valeur. Des entreprises pionnières peuvent ainsi également accélérer les processus de transformation technologique, économique et sociale. Cette mesure contribue donc, entre autres, à établir des normes ainsi qu’à atteindre les objectifs de l’indicateur cible 7. Cette mesure présente notamment un potentiel de synergie avec d’autres mesures en faveur de la protection du climat et de l’environnement.. 4e Réglementation plus souple de la durée de conservation des denrées alimentaires, transmission juridique simplifiée aux banques alimentaires, adaptation des normes de qualité et des normes industrielles pour éviter les déchets alimentaires et application obligatoire du principe de prévention des déchets (LPE, art. 30) aux denrées alimentaires. Brève description de la mesure : Un grand nombre de denrées alimentaires sont jetées et gaspillées inutilement en raison de leur date de péremption. Or, pour toutes les denrées alimentaires sans date limite de consommation, nos propres sens constituent un indicateur fiable pour savoir si elles peuvent encore être consommées. Par conséquent, il faudrait reconsidérer la datation sur les aliments qui ne présentent aucun risque pour la santé, voire remplacer la date de péremption par une date de fabrication. Par ailleurs, de nombreuses denrées alimentaires sont gaspillées alors qu’elles sont encore bonnes et qu’elles pourraient être données à des organismes d’aide alimentaire (p. ex. banques alimentaires, Foodsharing, etc.) ou à d’autres organismes chargés de leur distribution. Les raisons de responsabilité, le manque d’incitations, les obstacles à la déclaration, les directives internes et la peur de faire la une des journaux sont souvent évoqués. La simplification des exigences en matière de déclaration lors du don de denrées alimentaires194, ainsi que l’assouplissement des conditions en matière de responsabilité en cas de défaut de qualité (à l’exclusion des lacunes en matière de sécurité alimentaire) peuvent encourager le don de denrées alimentaires (voir également incitations fiscales, mesure 2i). Le principe de prévention des déchets inscrit dans la loi sur la protection de l’environnement pourrait s’opposer à ce que les entreprises se débarrassent des denrées alimentaires ou les utilisent pour nourrir des animaux, tant qu’elles respectent les exigences en matière de sécurité alimentaire. Le principe obligatoire de la prévention des déchets (LPE, art. 30) peut notamment être appliqué de manière efficace aux denrées alimentaires pour lesquelles la sécurité alimentaire est garantie conformément à la loi sur les denrées alimentaires (LDAl) par les mesures suivantes : la vente et le don de denrées alimentaires après la date limite de péremption et par la congélation en temps utile avant la date limite de consommation (voir le guide de remise de l’OSAV), ainsi que par la distribution en temps utile des excédents de la restauration aux clients, aux collaboratrices et collaborateurs, aux organismes caritatifs, au Foodsharing, etc. Justification du choix de la mesure : Selon Deloitte (2021), un aliment sur cinq est gaspillé dans les ménages en raison d’une date de péremption dépassée ou mal interprétée (entre 100 000 et 200 000 tonnes par an).195 Dans le commerce, la majeure partie des quelques 100 000 tonnes de denrées alimentaires est gaspillée en raison de la date, sans oublier une part considérable dans le secteur de la transformation.132 Par conséquent, les mesures prises en matière de datation des aliments et de gestion des aliments datés présentent un potentiel considérable. Le potentiel des dons de denrées alimentaires est nettement supérieur aux quelques 10 000 tonnes qui font l’objet de dons aujourd’hui.132 Tant sur le plan de l’offre que sur celui de la demande, une multiplication des dons de denrées alimentaires existants semble réaliste. La logistique et la propension de certaines entreprises à faire des dons sont des facteurs restrictifs. Dans d’autres pays, les dons par habitant·e sont largement supérieurs. D’un point de vue financier, les dépenses liées au don de denrées alimentaires ne représentent qu’une fraction de la valeur de ces denrées, et d’un point de vue écologique, les transports nécessaires à la redistribution ne représentent qu’une fraction de l’impact environnemental investi dans la production des denrées alimentaires.132 Cette mesure a un effet positif direct sur l’indicateur cible 8. 4f Étiquetage public des denrées alimentaires concernant les effets sur l’environnement, la santé et le bien-être animal, et normes minimales légales pour l’étiquetage privé des denrées alimentaires Brève description de la mesure : Il faudrait introduire un étiquetage alimentaire public sur les effets des produits alimentaires sur l’environnement, la santé et les animaux, sur la base de preuves scientifiques, afin de motiver et d’aider les consommatrices et consommateurs à faire des choix de consommation durables. Il faudrait également instaurer des normes minimales légales pour les labels privé des denrées alimentaires dans le domaine de l’environnement, du bien-être animal et de la santé, sans quoi les consommatrices et consommateurs risquent de ne pas disposer d’éléments de décision factuels et fiables. L’introduction d’étiquetages alimentaires publics ainsi que de normes légales minimales pour les labels privés peuvent également réduire le nombre de labels alimentaires et donc la complexité pour les consommatrices et consommateurs. Justification du choix de la mesure : Plusieurs études révèlent qu’un étiquetage alimentaire simple à comprendre et fiable favorise des choix de consommation durables et bénéfiques pour la santé.196-199 Les étiquetages alimentaires reconnus et soutenus par l’État bénéficient d’une plus grande confiance et généralement d’une plus grande transparence en matière d’évaluation. Ils peuvent donc avoir un impact plus important sur les comportements de consommation qu’un label alimentaire privé. En 2019, l’Office fédéral de la sécurité alimentaire et des affaires vétérinaires (OSAV) a notamment recommandé le Nutri-score, développé en France, sur la base d’une évaluation scientifique concluante.200 Le Nutri-score est désormais visible sur de nombreux produits en Suisse. Les labels de durabilité privés, comme le M-Check de Migros pour le climat et le bien-être animal, ont une portée et une visibilité moindres que les labels alimentaires reconnus par l’État. Une récente étude 201 a effectivement permis de constater l’effet positif de ces labels alimentaires privés en Suisse sur la simplicité perçue pour des décisions d’achat durables. Cependant, leur effet positif direct sur les décisions d’achat demeure relativement faible. La visibilité et la crédibilité d’un étiquetage jouent donc un rôle important dans l’optimisation de son impact. Les étiquetages alimentaires fondés sur des preuves et reconnus par l’État augmentent la visibilité et la crédibilité. Toutefois, les labels alimentaires privé et les initiatives de l’industrie agroalimentaire peuvent tout à fait renforcer le soutien à la création d’un étiquetage alimentaire public.146, 201 La mise en place d’un étiquetage alimentaire complet portant sur différents indicateurs environnementaux (basés notamment sur l’impact climatique, l’écotoxicité, la toxicité humaine, l’eutrophisation, les besoins en ressources, l’acidification) n’est pas si simple. Bien qu’une telle évaluation soit tout à fait réalisable sur le plan technique (cf. mesure 3a),146, 202 sa mise en œuvre nécessite un certain temps. Par conséquent, la mise en place d’un étiquetage public des denrées alimentaires en matière d’environnement pourrait se faire progressivement pour différents indicateurs. Par exemple, dans un premier temps et parallèlement à l’introduction de la taxe éq. CO2 (mesure 3a), l’empreinte carbone des aliments pourrait être indiquée sur les produits, en plus des effets dans les domaines du bien-être animal et de la santé. À terme, l’empreinte environnementale pourrait alors être calculée et indiquée selon d’autres indicateurs (p. ex. écotoxicité, toxicité humaine, eutrophisation, besoins en ressources, acidification). La présentation de l’étiquetage des denrées alimentaires doit toutefois être aussi compréhensible que possible, afin que les consommatrices et les consommateurs puissent prendre une décision d’achat durable et bénéfique pour leur santé, de manière aussi intuitive et rapide que possible dans le cadre de leurs processus de décision quotidiens. La mesure agit directement sur l’indicateur cible 1 et indirectement sur les indicateurs cible 2 à 8. 53 4g Une meilleure utilisation du droit des cartels pour réduire l’asymétrie de l’information et les écarts de pouvoir de marché et faciliter les regroupements de producteur Brève description de la mesure : Le rapport entre le droit de la concurrence et le droit agricole doit être défini et adapté de manière à permettre aux productrices et producteurs agricoles de se regrouper en vue d’une commercialisation commune, y compris en ce qui concerne les accords sur les prix et les quantités, notamment sur le marché du lait. À cet effet, l’art. 37 LAgr doit faire l’objet d’une modification. Des adaptations de la loi sur l’agriculture et du droit des cartels doivent permettre de contrer les processus de forte concentration aux échelons de la transformation et du commerce et de permettre aux productrices et aux producteurs d’instaurer un contre-pouvoir sur le marché (comme dans l’UE avec l’art. 42 du TFUE). La Commission de la concurrence COMCO devrait également examiner attentivement la situation du marché alimentaire et agricole sous l’angle du pouvoir de marché relatif selon l’art. 4 de la LCart, qui a été introduit en 2022, puisqu’il existe en Suisse une forte différence de pouvoir de marché due à l’existence d’un duopole entre Coop et Migros (y compris Denner qui appartient au groupe Migros), qui totalisent à eux deux 80 pour cent de parts de marché. Il est également important de lutter efficacement contre la concurrence déloyale. Justification du choix de la mesure : La condition essentielle pour une production durable est que les prix à la production couvrent au moins les coûts de production. Afin que les agricultrices et les agriculteurs puissent participer de manière équitable à la valeur ajoutée de la production de denrées alimentaires, ce qui contribuerait à assurer la survie de leurs exploitations, le législateur doit s’acquitter de son obligation de prendre des mesures plus efficaces pour compenser les rapports de force inégaux sur le marché. La mesure concerne avant tout l’objectif 9. 4h Adaptation des directives en matière de formation Brève description de la mesure : Les directives en matière de formation et les programmes d’enseignement définissent la manière dont l’enseignement doit être dispensé au degré secondaire I (école primaire), au degré secondaire II (écoles secondaires et formation professionnelle) et au degré tertiaire. En ce qui concerne les mesures 1b et 1c, qui sont par ailleurs cofinancées par le fonds de transformation, les directives doivent également être adaptées en conséquence à tous ces degrés et pour les différents domaines professionnels. Au niveau secondaire supérieur, cela concerne les directives de formation pour les métiers de l’agriculture (p. ex. agriculteur/trice, aviculteur/trice, maraîcher/ère, arboriculteur/ trice, viticulteur/trice), de la transformation des aliments (p. ex. cuisinier/ère, boulanger/ère, boucher/ère, technologue en industrie laitière, meunier/ère) et de l’alimentation en général et de la santé (p. ex. diététicien/ne, cuisinier/ère en diététique). Au degré tertiaire également, les programmes d’enseignement doivent être adaptés pour tous les domaines professionnels et toutes les filières du secteur agricole et alimentaire. Lors de l’adaptation des directives de formation, il convient de prendre en compte l’ensemble du système éducatif agricole et alimentaire, notamment les hautes écoles en agriculture et les instituts de recherche qui exercent une grande influence sur l’orientation du système alimentaire. Justification du choix de la mesure : L’adaptation des directives de formation et des programmes d’études doit être en rapport avec les mesures proposées, notamment 1a–1c, et répondre à la nécessité d’un nouveau savoir-faire en matière de transformation. Ces connaissances sont nécessaires à tous les niveaux de formation et dans tous les domaines professionnels concernés pour transformer globalement le système alimentaire et mettre en œuvre avec succès d’autres méthodes de production et de transformation. Cette mesure contribue à la réalisation des objectifs 1 à 8.. 54 4i Abaissement du nombre d’unités de gros bétail-fumure (UGBF) de 3 à 2,5 dans la loi sur la protection des eaux Brève description de la mesure : La réduction des unités de gros bétail-fumure (UGBF) de 3 UGBF à 2,5 UGBF dans la loi sur la protection des eaux, réduction toujours prévue dans la PA22+, devrait dans tous les cas être intégrée dans la consultation relative à la prochaine étape de la politique agricole. Le durcissement du Suisse-Bilanz qui interviendra à partir de 2023 (suppression des marges d’erreur de +10 %) ne constitue pas une mesure plus efficace suffisante dans le domaine des pertes d’éléments nutritifs et ne rend pas obsolète une baisse des UGBF. Une limitation plus stricte des UGBF est cohérente avec le durcissement du Suisse-Bilanz et concerne également les exploitations qui ne bénéficient pas de paiements directs et qui ne sont pas recensées via le Suisse-Bilanz. La diminution des UGBF prévue par la loi n’empêche certes pas directement les cheptels indépendants du sol, mais elle augmente les coûts d’évacuation des éléments nutritifs excédentaires, avec pour conséquence un effet incitatif certain en faveur de la réduction des cheptels. La réduction de la limite maximale autorisée pour les UGBF augmente la pression sur les exploitations d’élevage partiellement ou totalement indépendantes du sol pour qu’elles réduisent leurs effectifs. Cette mesure contribuera à réduire les émissions, notamment en ce qui concerne l’ammoniac. Par ailleurs, en cas de non-respect des objectifs de réduction pour l’azote (N) et le phosphore (P), le Conseil fédéral pourrait faire usage de sa compétence pour abaisser encore la limite des UGBF.203 Justification du choix de la mesure : La forte densité animale qui existe actuellement, notamment dans les régions les plus denses comme Lucerne et la Suisse orientale, ne permet pas d’atteindre les objectifs environnementaux concernant l’azote (N), le phosphore (P) et l’ammoniac (NH3), ni les objectifs climatiques. La réglementation UGBF augmente également la pression financière en particulier sur les élevages non adaptés aux conditions locales, dépourvus d’une base fourragère suffisante et de surfaces suffisantes pour les nutriments produits. Cette mesure concerne les objectifs 2, 4, 6 et 7. Il existe également des synergies potentielles avec d’autres mesures qui agissent directement ou indirectement sur la réduction des excédents d’éléments nutritifs et sur la réduction des cheptels. 4j Adaptation de la loi sur l’aménagement du ter- ritoire, si possible dans les meilleurs délais, pas de nouvelles exploitations d’élevage indépendantes du sol dans les zones d’agriculture intensive Brève description de la mesure : Les réglementations de l’actuelle loi sur l’aménagement du territoire permettent aux exploitations de « développement interne » (partiellement indépendantes du sol) et aux exploitations fortement ou totalement indépendantes du sol (dans les zones d’agriculture intensive) d’accéder, par le biais de l’art. 16a, al. 3, à des conditions-cadres aussi favorables que les exploitations qui pratiquent une agriculture et une production animale adaptées au site et liées au sol et qui s’orientent en fonction de la surface utile disponible. Parmi ces conditions favorables figurent la promotion des ventes et de la production, des mesures de protection à la frontière, des mesures d’allègement du marché, etc. La LAT doit être adaptée afin que les exploitations sans lien avec le sol déjà existantes et situées dans les zones d’agriculture dite intensive soient déclassées en tant qu’« entreprises industrielles non agricoles » et soient exclues des prestations de soutien prévues par la LAgr. Par ailleurs, aucune nouvelle construction ou installation dépassant le développement interne ne sera plus autorisée et l’affectation de la « zone d’agriculture intensive » sera supprimée ou à nouveau retirée de la LAT. Justification du choix de la mesure : Cette mesure permet de renforcer et d’assurer la cohérence avec toutes les autres mesures qui visent à réduire le nombre de cheptels sans lien avec le sol. De plus, elle empêche toute nouvelle augmen- tation future des cheptels dans le sens d’un fort développement interne, ce qui renforce la mesure des primes de reconversion (cf. 2g) et veille à ce que cette mesure ne soit pas contournée. Cette mesure concerne les objectifs 2, 4, 5 et 7. Il existe également des synergies potentielles avec d’autres mesures qui agissent directement ou indirectement sur la réduction des excédents d’éléments nutritifs et sur la réduction des cheptels. Paquet de mesures III : Mesures de politique agricole et soutien aux zones rurales 4k Législation complémentaire qui met en œuvre le devoir de diligence des grands groupes (y compris les banques) conformément aux directives internationales (p. ex. les Principes directeurs des Nations Unies relatifs aux entreprises et aux droits de l’homme ou les Lignes directrices de l’OCDE), afin que les entreprises soient conscientes de leur responsabilité, analysent les risques, prennent les mesures correspondantes et que les personnes lésées aient accès à un mécanisme de réclamation 2j Adaptation des paiements directs, des mesures de soutien du marché et des aides à l’investissement Brève description de la mesure : Les groupes agroalimentaires, les sociétés commerciales et les banques dont le siège social est en Suisse exercent une influence qui dépasse les frontières nationales grâce à leurs chaînes d’approvisionnement et à leurs investissements. Après l’échec de l’initiative pour des multinationales responsables, le contre-projet est entré en vigueur. Cependant, en comparaison avec les pays voisins et la législation européenne proposée, celui-ci204 ne définit que de manière très limitée les obligations de diligence des entreprises (uniquement en ce qui concerne les minerais de conflit et le travail des enfants) et comporte peu de mécanismes d’application administratifs. Il faudrait adapter ce point en s’appuyant sur des directives internationales, comme les Principes directeurs des Nations Unies relatifs aux entreprises et aux droits de l’homme205, et sur la législation existante et à venir, comme la loi allemande sur le devoir de diligence206 ou la directive européenne sur la diligence raisonnable en matière de développement durable des entreprises207 . Une loi suisse sur le devoir de diligence des entreprises devrait suivre les procédures recommandées et couvrir un large éventail de devoirs de diligence en matière de protection de l’environnement et de droits de l’homme ; inclure les propres activités de l’entreprise ainsi que les fournisseurs ; exiger la réalisation d’une analyse de risques étendue, des mesures de prévention, la détection des non-conformités et l’adoption immédiate de mesures correctives en cas de violations constatées ; et permettre aux personnes lésées d’accéder à un mécanisme de réclamation ainsi qu’à un recours juridique auprès du système judiciaire suisse. Justification du choix de la mesure : La Suisse a une très forte densité d’entreprises commerciales qui négocient des matières premières agricoles essentielles en provenance du monde entier. Une loi complète sur le devoir de diligence inciterait ces grands groupes à améliorer la durabilité environnementale et sociale de ces produits et à prévenir les violations en matière de droits de l’homme, sous peine de se mettre en danger sur le plan légal. Ainsi, la Suisse peut contribuer à améliorer la durabilité du système alimentaire mondial et, surtout, ne pas offrir d’échappatoire aux entreprises qui souhaiteraient se soustraire à leurs responsabilités. En raison de l’attractivité de la Suisse en termes d’environnement économique et de main-d’œuvre qualifiée, cette mesure ne devrait pas entraîner un exode massif des entreprises. Au contraire, la législation plus poussée pourrait conduire à l’implantation de nouvelles entreprises en Suisse axées sur la durabilité, faisant ainsi de la Suisse un pays pionnier sur ce marché mondial en pleine croissance. Cette mesure répond à tous les objectifs. Brève description de la mesure : Les adaptations des paiements directs via la mise en œuvre partielle de la PI 19.475 « Réduire les risques de l’utilisation des pesticides » et via les nouveaux trains d’ordonnance méritent d’être signalées, mais demeurent insuffisantes. Il est nécessaire de faire évoluer la politique agricole sur le long terme afin de simplifier le système et de l’orienter davantage vers les objectifs. Par conséquent, le comité scientifique recommande ci-après, à titre d’exemples (liste non exhaustive), quelques mesures essentielles qui pourraient être intégrées dans la nouvelle PA en cours d’élaboration : a) Pour les contributions à la sécurité de l’approvisionnement et les contributions au paysage cultivé, les contributions doivent être réduites de manière échelonnée dès que les charges critiques d’azote sont dépassées au niveau régional. La charge minimale en bétail doit également être modifiée là (régions de plaine et de collines) où elle a contribué non pas à une extensification, mais à une intensification.208 b) Pour les contributions aux cultures particulières, les contributions pour la culture de légumineuses propres à la consommation humaine directe doivent être plus élevées que celles pour les légumineuses destinées à l’alimentation animale. Les contributions pour la culture conventionnelle de betteraves sucrières doivent être réduites. c) En ce qui concerne les contributions au système de production, le budget pour la renonciation volontaire aux produits phytosanitaires doit être augmenté de toute urgence et de manière significative. Les 92 millions de francs prévus à partir de 2023 ne suffiront pas à créer des incitations encore plus fortes. d) La contribution actuelle pour la PLVH (production de lait et de viande basée sur les herbages) n’est pas assez ambitieuse. Après 2025, elle devra se transformer en une contribution pour une utilisation d’aliments concentrés réduits en protéines. e) Une preuve obligatoire de la couverture sociale des conjointes et conjoints et des partenaires enregistré·e·s doit être intégrée comme exigence de base dans les PER. f) Les suppléments de soutien au prix du lait doivent être échelonnés en fonction de la proportion de protéines en kilogrammes dans les aliments concentrés utilisés chaque année, c’est-à-dire des suppléments complets en cas d’utilisation minimale d’aliments concentrés (base fourragère propre en premier lieu), un couplage négatif en cas d’utilisation croissante d’aliments concentrés. En cas d’années difficiles dues aux conditions météorologiques et climatiques, avec une pénurie nationale de fourrages grossiers, il devrait être possible de déroger à cette réglementation afin d’éviter la malnutrition ou les carences alimentaires chez les bovins. g) De nouvelles aides en matière d’investissement et des facilités juridiques pour l’installation de systèmes agrophotovoltaïques devraient être mises en œuvre Justification du choix de la mesure : Les conditions actuelles pour l’obtention des paiements directs doivent être plus ambitieuses et mieux adaptées aux objectifs de transformation.179 Le cadre réglementaire de l’aide à l’agriculture, déjà complexe, ne doit pas être trop surchargé dans un souci de simplification et d’allègement de la charge pesant sur les agricultrices et les agriculteurs ; c’est pourquoi les propositions formulées ici visent en premier lieu, sur le plan environnemental, à réduire les émissions, 55 notamment les excédents d’azote, et sont complétées par une importante mesure sociale, déjà proposée dans la PA22+, concernant la couverture par les assurances sociales des conjointes et des conjoints et des partenaires enregistré·e·s. Les mesures de politique agricole recommandées ici présentent également des synergies positives avec de nombreuses mesures prises dans le cadre du fonds de transformation ainsi que du paquet de mesures II. Ces mesures contribuent à la réalisation des objectifs 2 à 7 ainsi qu’à celle de l’objectif 9. 2k Orientation des mesures commerciales vers des systèmes alimentaires durables Brève description de la mesure : La nature des relations commerciales est un levier important pour promouvoir des systèmes alimentaires diversifiés et durables. La Suisse peut exploiter la marge de manœuvre que lui laisse le droit de l’OMC et orienter (plus) systématiquement les mesures commerciales vers des critères de durabilité. Elle peut par exemple réglementer l’utilisation des certificats, lier des préférences tarifaires et des contingents tarifaires à des critères de durabilité, réduire les mesures commerciales non tarifaires (telles que les taxes ou les procédures de reconnaissance fastidieuses) pour des produits particulièrement durables et apporter un soutien au marché. Parallèlement, elle peut augmenter les droits de douane sur les produits nocifs dans les limites de sa marge de manœuvre et interdire l’importation de produits particulièrement nocifs. Justification du choix de la mesure : Les droits de douane, les contingents tarifaires et les procédures d’accès au marché sont des mesures importantes de régulation du marché. Dans le cadre de divers débats – comme le débat sur l’alimentation équitable ou le débat sur l’accord AELE-Indonésie et l’initiative sur l’élevage industriel – , la Suisse a réfléchi à la manière dont elle pourrait associer davantage les mesures commerciales à des critères de durabilité. L’objectif de telles mesures serait, d’une part, de créer un « level playing field » pour les productrices et producteurs nationaux et, d’autre part, de soutenir les développements durables dans le pays partenaire et d’empêcher les développements non durables. La Suisse est en mesure de prévoir de telles différenciations de produits dans le cadre d’accords de partenariat (accords commerciaux) ou de les réglementer de manière unilatérale. Ce faisant, elle doit respecter les limites imposées par la Constitution et le droit de l’OMC. Ainsi, la mesure doit être axée sur des objectifs sociaux et environnementaux ; elle doit être proportionnée ou n’aller que jusqu’à ce qui est nécessaire pour atteindre l’objectif ; et elle ne doit pas créer de discrimination entre les fournisseurs qui produisent selon des normes équivalentes. Elle doit agir de manière cohérente, c’est-à-dire appliquer les mêmes conditions à l’intérieur du pays. Ces conditions-cadres permettent à la Suisse de stimuler la durabilité à la fois vers l’intérieur et vers l’extérieur et de considérer le système alimentaire comme un système global. La Suisse est loin d’avoir épuisé sa marge de manœuvre dans ce domaine. L’expérience en matière de procédures est encore limitée. La loi fédérale sur le commerce agricole durable, élaborée dans le cadre du projet du FNS « Des systèmes alimentaires diversifiés grâce à des relations commerciales durables », montre comment de tels processus peuvent être fondés sur la confiance. 2l Instauration d’un impôt négatif sur le revenu pour la catégorie Working Poor dans l’agriculture Brève description de la mesure : Dans le secteur agricole, l’écart entre les revenus est important. Dans près de 20 pour cent des exploitations, le revenu du ménage est inférieur au revenu qui serait généré par l’aide sociale.209 Les femmes travaillant dans l’agriculture, en particulier, ont des revenus très faibles. Aujourd’hui, la demande d’aide sociale n’est quasiment jamais une option pour les familles travaillant dans l’agriculture, car l’aide sociale est un instrument politique fortement axé sur les villes. Par conséquent, nous sommes aujourd’hui confrontés à une forte prévalence de la pauvreté dans le secteur agricole. Même si l’on 56 manque encore d’expériences à grande échelle en matière d’impôts négatifs sur le revenu, il est probable que cet instrument permettrait de réduire le problème de la pauvreté dans l’agriculture, comme le souligne également Tondani.210 En conséquence, il est proposé de prévoir pour le secteur agricole, en cas de revenu inférieur au taux de l’aide sociale, un transfert de retour via la déclaration d’impôt, qui garantirait à chaque ménage agricole un revenu équivalent à celui de l’aide sociale. Justification du choix de la mesure : Dans le secteur agricole, il existe depuis des décennies, certes à tire officieux, des mesures de politique de revenus dont le point faible est qu’elles permettent également aux ménages aisés de bénéficier d’une aide de l’État. Nous proposons un instrument afin de remédier à cette anomalie. Cette mesure répond à l’objectif 9. 2m Renforcement de l’offre psychosociale dans les régions rurales Brève description de la mesure : Ces dernières années, la prise de conscience de l’importance des soins psychosociaux a augmenté et les besoins se sont multipliés en conséquence. Dans les pays développés, le nombre de psychologues ne cesse d’augmenter. Cependant, les infrastructures correspondantes restent, aujourd’hui encore, fortement concentrées dans les zones urbaines. En Suisse, le fait que les agricultrices et les agriculteurs soient environ deux fois plus exposé·e·s au risque d’épuisement que la moyenne de la population doit faire réfléchir. En conséquence, il est proposé d’établir des offres de thérapie et de conseil à faible coût dans les zones rurales et de les communiquer en conséquence dans l’agriculture. Justification du choix de la mesure : Une agriculture durable repose sur la bonne santé mentale de ses actrices et acteurs, notamment lorsqu’il s’agit de mener à bien une transformation ensemble. Or, en Suisse également, la santé mentale des agricultrices et des agriculteurs n’est pas au beau fixe. 211 Cette mesure répond directement à l’objectif 10. Paquet de mesures IV : Mesures réglementaires II 4l Passage du Suisse-Bilanz au Hoftorbilanz par étapes jusqu’en 2030 Brève description de la mesure : Il semble paradoxal que, d’une part, toutes les exploitations agricoles bénéficiant de paiements directs soient tenues de présenter un bilan azoté équilibré avec le Suisse-Bilanz, mais que, d’autre part, de nombreuses régions caractérisées par les grandes cultures présentent une teneur élevée en azote dans les eaux souterraines et de surface. Cela s’explique par les faiblesses du Suisse-Bilanz, qui présente justement des lacunes méthodologiques en ce qui concerne les enrichissements et les pertes en azote. Des études scientifiques ont montré qu’un Hoftorbilanz (OSPAR), associé à des obligations de documentation étendues, entraînerait une diminution des excédents.212 En outre, la méthode OSPAR permettrait également d’enregistrer directement la fixation biologique de l’azote et les dépôts atmosphériques/l’ammoniac. Un tel changement de système doit donc être préparé et mis en œuvre avec rigueur. Justification du choix de la mesure : Aujourd’hui, la qualité de l’eau potable est souvent si mauvaise dans les régions de plaine que des mesures réglementaires supplémentaires sont indispensables. C’est précisément le cas pour la teneur en nitrates qui dépasse régulièrement les valeurs limites réglementaires. Dans ce cas, il est indispensable de passer à un système qui exclut les excédents. Cette mesure devrait apporter une contribution décisive à l’objectif 2. 4m Mesures de contrôle garantissant l’absence de travail des enfants pour tous les produits alimentaires consommés en Suisse Brève description de la mesure : Aujourd’hui, plus de la moitié du travail des enfants est effectué dans le secteur agricole. L’Organisation internationale du travail estime qu’environ 97 millions d’enfants de moins de 16 ans participent à la culture et la récolte de denrées alimentaires au point d’empêcher une scolarisation régulière et de mettre leur santé en danger. Il n’existe aujourd’hui aucun mécanisme permettant d’éviter que les produits concernés ne se retrouvent dans la chaîne alimentaire suisse. En tant que pays dont les normes sociales sont relativement élevées et dont la part d’importations est très importante, la Suisse est particulièrement bien placée pour trouver des solutions qui permettraient d’exclure du commerce agricole les denrées alimentaires issues du travail des enfants, ce qui constituerait un signal fort sur le plan international. Il existe déjà aujourd’hui des initiatives qui montrent qu’il est possible de définir des normes de qualité et de les contrôler tout au long de la chaîne. À partir de ces expériences, il convient de mettre en place un système de contrôle qui vérifie également le processus de production dans le pays d’origine quant à la présence d’une maind’œuvre majeure suffisante. Justification du choix de la mesure : Il est prouvé que le travail des enfants a des conséquences catastrophiques sur les opportunités de formation et donc de participation des générations futures.213 La tolérance du travail des enfants constitue donc l’une des zones d’ombre les plus importantes du commerce agricole. Cette mesure se réfère explicitement à l’objectif 11. 4n Adaptation des accords commerciaux internationaux Brève description de la mesure : La Suisse devrait s’engager au niveau multilatéral en faveur de nouvelles règles commerciales agricoles axées sur les Objectifs de développement durable de l’ONU. Elle peut, au sein de l’OMC, appuyer le processus de réforme qui vient d’être lancé en proposant des idées de réglementation innovantes. Au sein du Comité de la sécurité alimentaire mondiale (CSA), elle peut proposer de travailler ensemble à l’élaboration de principes de commerce et de systèmes alimentaires durables. Elle pourrait ainsi contribuer à réduire le fossé entre Rome (FAO/CSA) et Genève (OMC) et créer un nouvel élan politique en faveur du changement. Justification du choix de la mesure : Le monde a besoin de toute urgence d’un système commercial qui établisse un « level playing field » pour les produits issus d’une production durable, qui permette aux États de consolider les systèmes alimentaires locaux, tout en encourageant et en rendant plus fiable le commerce de denrées alimentaires durables. L’accord actuel de l’OMC (tout comme les règles en matière de propriété intellectuelle et de normes de produits) ne permet pas d’atteindre cet objectif. Alors que les tentatives de réforme n’ont pas progressé pendant des années, on observe actuellement un nouvel élan politique (voir notamment la conclusion de l’accord de l’OMC sur les subventions à la pêche ; le brainstorming de l’OMC sur un nouvel accord agricole ; les principes du CSA en matière d’agroécologie ; la déclaration des Nations Unies sur les droits des petits agriculteurs UNDROP ; ainsi que diverses conférences mondiales sur le commerce et les ODD). Alors qu’il est généralement admis que les règles existantes présentent de nombreuses lacunes, il n’existe guère d’idées concrètes sur la manière dont de nouvelles règles plus équilibrées pourraient être formulées afin de conjuguer intelligemment sécurité, protection et intégration des marchés ; de mieux prendre en compte les besoins des pays les plus pauvres ; de permettre des solutions adaptées au 57 contexte ; et de promouvoir des systèmes alimentaires durables à l’échelle mondiale. En sa qualité d’actrice indépendante, la Suisse est en mesure d’apporter un soutien important aux processus qui génèrent de nouvelles idées concrètes et accélèrent le processus de réforme. Cette mesure répond à tous les objectifs. 4o Traçabilité accrue et informations plus précises auprès des restaurateurs sur l’origine de leurs produits, notamment pour la viande. Par ailleurs, l’instauration d’un label d’État et d’une distinction pour la restauration répondant à des normes de durabilité élevées. Brève description de la mesure : Il existe aujourd’hui un décalage entre l’offre souvent importante de produits répondant à des normes sociales et environnementales élevées dans le commerce de détail et le portefeuille dans la restauration. La restauration a clairement besoin d’une traçabilité accrue et d’indications plus précises sur l’origine des produits, en particulier pour la viande.214 L’absence de normes et de déclarations en matière de durabilité empêche la clientèle des restaurants de prendre des décisions en connaissance de cause. Des obligations supplémentaires en matière de documentation, d’une part, et des incitations sous la forme d’un label de durabilité pour les restaurants et les cantines répondant à des normes élevées correspondantes, d’autre part, devraient contribuer à augmenter le niveau moyen de durabilité de la restauration hors domicile. Justification du choix de la mesure : Jusqu’à présent, la restauration a souvent été exclue des stratégies de durabilité. Ce constat ne tient pas compte de l’importance considérable et grandissante de la restauration hors domicile. Cette mesure se réfère à l’objectif 1 et concerne indirectement les objectifs 2 à 8. 58 4p Interdiction des promotions à bas prix sur les produits d’origine animale par les détaillants et les discounters Brève description de la mesure : Interdire les promotions à bas prix sur les produits d’origine animale diminue l’incitation à l’achat de tels produits. Le commerce de détail classique (composé de Migros, Coop, Volg, Spar et Manor) a enregistré une part de promotion de près de 49 pour cent en termes de volume sur l’ensemble du secteur de la viande. En termes de chiffre d’affaires, la part était de 43 pour cent. L’importance des promotions est légèrement inférieure chez les discounters (Aldi, Lidl et Denner). Les ventes avec promotions représentent ici 44 pour cent et la part du chiffre d’affaires réalisée grâce aux promotions est de 40 pour cent.215 La mesure cible notamment les promotions à bas prix pour les produits d’origine animale, qui risquent de ne pas couvrir les coûts de production, de répercuter la pression sur les prix aux productrices et aux producteurs sans tenir compte des coûts externes. Justification du choix de la mesure : Même si les promotions sont en partie justifiées par les fluctuations saisonnières de l’offre (notamment pour la viande fraîche) et qu’elles permettent de réduire le risque de gaspillage alimentaire, il faudrait dans l’ensemble diminuer considérablement le nombre de réductions accordées sur la vente de produits d’origine animale, notamment sur la viande. D’une part, une consommation élevée de produits d’origine animale telle que la viande a des externalités négatives (notamment sur le climat, la pollution azotée, la santé) et devrait donc être réduite (cf. indicateur cible 1). D’autre part, les promotions – surtout sur le segment des prix bas – augmentent la pression sur les prix pour les productrices et les producteurs. De ce fait, ces derniers ne sont souvent plus en mesure d’assumer les coûts de production réels et de garantir un bien-être animal accru ainsi qu’une meilleure protection de l’environnement. Interdire en conséquence les promotions à bas prix permet donc de lutter contre les défaillances du marché et de minimiser les fausses incitations à la consommation. Cette mesure contribue à la réalisation des objectifs 1 à 7. Il existe un potentiel de synergie avec d’autres mesures qui agissent directement ou indirectement sur la consommation de viande et la réduction des cheptels. 59 4 LA GOUVERNANCE DU SYSTÈME ALIMENTAIRE EN SUISSE 4.1 Le rôle de la gouvernance dans la transformation du système alimentaire suisse Auteur·e·s chapitre 4 : Lukas Fesenfeld, Karin Ingold, Eva Lieberherr, Sol Kislig, Christian Schader Comme nous l’avons vu dans les chapitres précédents, il y a encore beaucoup à faire pour rendre le système alimentaire suisse conforme aux objectifs de développement durable des Nations Unies. Cette forte pression, aussi bien environnementale, sociale et sanitaire qu’économique, qui s’exerce sur les acteurs et actrices du système alimentaire est à la hauteur des opportunités qui s’offrent à eux et elles. Une transformation totale du système alimentaire requiert un objectif politique ambitieux (voir chapitre 2) ainsi que des pistes d’action coordonnées sur le plan stratégique (voir chapitre 3) tout au long de la chaîne de valeur. La décision politique d’entreprendre des projets de transformation aussi ambitieux et de les mettre en œuvre suppose toutefois que les divers objectifs, intérêts et valeurs de principaux acteurs et principales actrices du système alimentaire soient pris en compte. Le processus de transformation exige fondamentalement des processus décisionnels et des institutions qui, au lieu de faire le jeu de la polarisation politique, œuvrent pour une réorientation stratégique de la politique alimentaire qui soit axée sur le long terme et sur une idée de compromis constructif. Dans ce chapitre, nous nous attacherons à démontrer que la question de la gouvernance du système alimentaire revêt une importance centrale dans ce contexte. Nous présenterons des propositions concrètes quant à la forme que pourra prendre cette gouvernance en Suisse d’ici 2030 afin de mettre en œuvre sur les plans politique et pratique les objectifs ambitieux (chapitre 2) et les mesures (chapitre 3) précédemment détaillés. La gouvernance des systèmes alimentaires désigne le processus et les institutions par lesquels les sociétés négocient, mettent en œuvre et évaluent les objectifs, les priorités, les conditions-cadres et les mesures concrètes qui s’appliquent à la conception des systèmes alimentaires.11 La perspective systémique est ici essentielle. Il s’agit en effet d’élaborer une politique globale du système alimentaire, qui ne se concentre pas seulement sur l’agriculture (c’est-à-dire sur la politique agricole au sens strict), mais intègre également les autres acteurs et actrices de la chaîne de valeur (par ex. la production, le commerce et la consommation) ainsi que les effets de l’alimentation et de l’agriculture sur l’environnement et la société. Les différents groupes d’intérêt du système alimentaire suisse se trouvent actuellement dans une situation difficile, caractérisée par des conflits d’objectifs, d’intérêts et de valeurs (voir chapitre 2). Or ceux-ci mènent souvent à un débat fortement polarisé qui empêche tout dialogue constructif et mène à une situation d’impasse politique. C’est ce qui s’est produit au printemps 2021, lorsque la réforme de la politique agricole PA22+76, qui était prévue pour 2022, a été suspendue. Depuis, le Conseil fédéral doit réviser la PA22+ et en présenter une nouvelle version au Parlement. Dans son rapport sur l’orientation future de la politique agricole publié en juin 2022, il engage ce processus de réforme en soulignant notamment l’importance d’une vision globale du système alimentaire et d’une politique intégrée de celui-ci.12 Parallèlement, le Conseil fédéral 60 prévoit également d’agir à court terme à l’échelon des ordonnances dans certains domaines partiels. En avril 2022, il a par exemple adopté, après une procédure de consultation d’une année, un train d’ordonnances en réponse à l’initiative parlementaire 19.475 « Réduire le risque de l’utilisation de pesticides »143. L’entrée en vigueur des ordonnances concernées commence dès 2023. Ces démarches montrent que le Conseil fédéral souhaite et doit agir de plus en plus rapidement en raison de l’urgence de la situation. Mais elles prouvent aussi qu’aucune réorientation globale et consensuelle de la politique en matière de système alimentaire n’est possible sans opérer de changement en matière de gouvernance. Les développements à l’échelle internationale viennent corroborer l’importance de la gouvernance des systèmes alimentaires. Lors du Sommet des Nations Unies sur les systèmes alimentaires, qui s’est tenu à New York en 2021, le rôle central que la gouvernance de la transformation des systèmes alimentaires mondiaux doit jouer dans la réalisation des objectifs de développement durable (ODD) est apparu clairement.216 Dans le cadre de la préparation du Sommet, l’Office fédéral de l’agriculture (OFAG) a détaillé dans une feuille de route la manière dont la Suisse entendait transformer son système alimentaire dans le respect des objectifs de développement durable de l’ONU13 et de la Stratégie pour le développement durable 2030 du Conseil fédéral133. Le plan d’action correspondant, qui présente les mesures concrètes à mettre en œuvre entre 2021 et 2023, souligne également l’importance de poursuivre le dialogue avec toutes les parties prenantes concernées.216 Cet échange doit permettre de négocier des recommandations concrètes à l’attention des instances législatives. On observe donc une dynamique dans laquelle la simple consultation cède le pas à une négociation active et multilatérale incluant l’ensemble des groupes d’intérêt. Il existe de nombreux exemples de pratiques allant dans ce sens tant à l’échelle internationale que nationale. Leur succès dépend toutefois en grande partie de l’objectif visé et de la manière dont le processus est conçu. En Inde, par exemple, le mécanisme multipartite institutionnalisé « Eat Right India », mis en place en 2018 par l’autorité indienne de sécurité alimentaire, a un pouvoir consultatif dans le processus législatif puis un rôle actif d’accompagnement de la mise en œuvre.219 Il se compose de 15 à 20 représentant·es sélectionné·es à partir de précédents processus multipartites dans le domaine de l’agriculture et de l’alimentation. En France, le Conseil National de l’Alimentation (CNA) est consulté depuis 1985 sur la définition de la politique de l’alimentation. Il compte 63 membres issu·es du monde scientifique, de l’agriculture, du secteur privé, de la population civile, des ONG et des autorités. Les ministères et le Parlement y sont également représentés. En Suisse, on peut par exemple citer l’exemple de la table ronde sur l’énergie hydraulique convoquée en 2020 par Simonetta Sommaruga, au cours de laquelle les principales parties prenantes de ce secteur (associations environnementales telles que Pro Natura, représentations cantonales telles que la Conférence des directeurs cantonaux de l’énergie (EnDK), Office fédéral de l’énergie, Office fédéral de l’environnement, etc.) ont rédigé une déclaration à titre de recommandation destinée à l’instance législative.220 Le succès de ce type d’assemblées et de processus dépend toutefois beaucoup de la clarté de leur mandat, de la légitimité de l’instance qui les convoque (par ex. Conseil fédéral, Parlement) et de la limitation de leur durée, autrement dit du degré d’institutionnalisation de leur travail.219, 221, 222, 223, 224 Leur intégration non seulement horizontale mais aussi verticale (échelons communal, cantonal, national et international) est par ailleurs déterminante pour leur efficacité219, tandis que la mise à disposition d’un budget de fonctionnement constitue un autre facteur de réussite important219. La qualité de l’animation du processus de dialogue et de négociation multilatéral influence également les chances de succès et l’acceptation des résultats des négociations.219, 221, 222, 223, 224 Enfin, le comité scientifique estime que, pour avoir le succès escompté, un tel processus de négociation doit être convoqué officiellement et doté d’une structure institutionnalisée permanente. Les propositions suivantes relatives au développement de la gouvernance du système alimentaire suisse sont basées sur des exemples relevant du secteur alimentaire de pays tiers qui peuvent être considérés comme concluants (voir encadré p. 61). EXEMPLES INTERNATIONAUX CONCLUANTS DE FORUMS DE NÉGOCIATION ET DE COMMISSIONS SUR L’AVENIR DE LA POLITIQUE ALIMENTAIRE ET AGRICOLE de nombreux·ses professionnel·les du secteur agricole. Depuis, des analyses d’impact et des études de faisabilité ont été réalisées. En septembre 2022, la commission a accepté à l’unanimité le mandat du BMEL relatif à la poursuite des travaux, à condition que le gouvernement trouve préalablement une solution pour financer la transformation du système d’élevage des animaux de rente en s’inspirant de ses propres propositions. En effet, la Commission Borchert avait formulé des propositions de financement concrètes largement soutenues par les principaux acteurs et principales actrices politiques, économiques et de la société civile actif·ves dans la filière alimentaire allemande. ALLEMAGNE Commission sur l’avenir de l’agriculture : Convoquée en 2020, la « Zukunftskommission Landwirtschaft » (ZKL) avait pour mission concrète de rédiger un rapport contenant des recommandations pour un système agricole et alimentaire durable en Allemagne. Celui-ci a été adopté à l’unanimité en 2021 et sert désormais de base au gouvernement pour transformer le système alimentaire allemand. Les membres de cette Commission sur l’avenir de l’agriculture sont issu·es des domaines de l’agriculture, de l’économie, de la science et de la protection de l’environnement et des animaux.230 En 2022, la commission a repris ses activités pour accompagner le travail du gouvernement en proposant des compromis négociés sur des pistes d’action politique concrètes destinées à transformer le système alimentaire.230 Commission Borchert : Le réseau de compétences sur l’avenir de l’élevage, également appelé Commission Borchert en référence à son président, Jochen Borchert231, se compose de représentant·es des milieux politique, scientifique, pratique, économique et associatif. En 2019, la Commission Borchert a été chargée par le Ministère fédéral de l’Alimentation et de l’Agriculture (BMEL) d’élaborer des recommandations sur le développement et la mise en œuvre de la stratégie allemande en matière d’élevage. Publiées en février 2020, celles-ci ont reçu un large écho auprès du Parlement fédéral, des Länder et DANEMARK Organic Food Advisory Council / Organic Denmark : Créé en 1987, l’« Organic Food Advisory Council » (Conseil consultatif danois pour l’alimentation biologique) est responsable du développement de la filière bio au Danemark. Grâce à l’implication de plusieurs parties prenantes, un premier plan d’action ambitieux, l’« Organic Action Plan » a pu être élaboré et mis en œuvre en 1995, puis continuellement développé.232, 233 Dans le cadre de ce processus, le conseil est devenu en 1998 une coalition réunissant des représentant·es du monde agricole, des entreprises et des services du secteur alimentaire, des professionnel·les de la restauration (par ex. employé·es de cuisine, cuisinier·ères), des ONG et des consommateurs et consommatrices.219 Il a permis au Danemark de devenir jusqu’à aujourd’hui le pays dans lequel la part de marché des aliments biologiques est la plus forte au monde. Au Danemark, 80 % de la population achète des produits issus de l’agriculture biologique.233 En 2018, l’Organic Action Plan a reçu le Future Policy Award. 61 4.2 Gouvernance du système alimentaire suisse 2030 : Une proposition Recommandations d’objectifs et de mesures Groupe de parties prenantes 1 Groupe de parties prenantes 3 Groupe de parties prenantes 2 Groupe de parties prenantes 4 Conflits d’objectifs et d’intérêts Exécutif fédéral Processus de consultation formel Exécutif cantonal Exécutif communal Fonction de conseil et d’observation Commission sur l’avenir du système alimentaire Échange de données Comité scientifique Système alimentaire Suivi scientifique, conseil et analyse d’impact Suivi scientifique, conseil et analyse d’impact Recommandation fondée sur des données probantes Participation citoyenne par ex. assemblées citoyennes Recommandations d’objectifs et de mesures Convocation officielle par le pouvoir législatif Fonction de conseil et d’observation Compromis négocié Principaux et principales destinataires Législatif fédéral Législatif cantonal Législatif communal Législation Accord sectoriel Convocation officielle du Comité scientifique Création d’un Office fédéral de l’alimentation et de l’agriculture Convocation officielle de la Commission sur l’avenir du système alimentaire Examen de l’opportunité d’une loi sur le système alimentaire Création d’une Commission parlementaire de l’alimentation et de l’agriculture Renforcement de la collaboration entre les offices Recommandations à l’horizon 2025 Participations citoyennes régulières Recommandations à l’horizon 2030 Dans ce qui suit, nous présentons quelques recommandations relatives à l’adaptation de la gouvernance destinées à réduire les conflits d’objectifs, d’intérêts et de valeurs existants dans la politique alimentaire et agricole suisse et à créer ainsi les conditions préalables à une transformation durable du système alimentaire suisse. La figure 7 esquisse un projet de gouvernance du système alimentaire à l’horizon 2030. Nous suggérons en outre par la figure 7 des étapes intermédiaires à l’horizon 2025 pour adapter les structures institutionnelles. Les propositions relatives à la gouvernance du système alimentaire ont été élaborées dans le cadre du processus itératif DELPHI par les auteur·e·s du chapitre et les membres du comité (voir annexes en ligne). 62 Ordonnance Fig. 7 : Fesenfeld et al. (2023)237 / Telek 4.2.1 Commission sur l’avenir du système alimentaire Les propositions s’articulent autour de la mise en place dans les meilleurs délais d’une Commission sur l’avenir du système alimentaire. Celle-ci réunira autour d’une table de négociation multilatérale toutes les parties prenantes du système alimentaire suisse (pas uniquement les acteurs et actrices de la filière agricole au sens strict) afin qu’elles élaborent ensemble des recommandations destinées aux instances législatives. Pour garantir sa légitimité politique, elle devrait idéalement être convoquée officiellement d’ici 2025, par exemple par le Parlement fédéral. Les pouvoirs législatifs communal, cantonal et fédéral exerceront une fonction d’observation au sein de la commission en leur qualité de destinataires directs des recommandations en matière d’objectifs et de mesures formulées par l’organe. L’exécutif, c’est-à-dire les offices fédéraux, la Conférence des gouvernements cantonaux (CdC) et les communes, disposera également d’une fonction d’observation et de conseil lors des séances de la commission. La Commission sur l’avenir du système alimentaire pourrait se baser sur la Commission consultative pour l’agriculture (CCA), une commission extraparlementaire instituée en 2011. Le champ de recommandations de la commission ne devra pas se limiter au développement de la politique agricole, mais couvrir tous les aspects du système alimentaire afin que des propositions concrètes pour une politique intégrée du système alimentaire puissent être formulées. De cette manière, la commission pourra renforcer la cohérence entre les politiques agricole, alimentaire, sanitaire, commerciale et environnementale et ainsi contribuer à réduire la complexité et rendre plus efficace le travail des nombreuses commissions extraparlementaires ayant des interfaces avec la thématique de l’alimentation et de l’agriculture (par ex. la CCA, le Conseil de la recherche agronomique ainsi que le Comité national suisse de la FAO). La Commission sur l’avenir du système alimentaire ouvrira un nouvel espace de négociation multilatéral, une perspective systémique et à plus long terme. Cela permettra aux différentes parties prenantes de former de nouvelles coalitions (potentiellement inattendues) et de négocier des compromis concrets sous forme de recommandations pour le processus législatif. Des compromis que l’ensemble des groupes d’intérêt de la chaîne de valeur considèrent comme acceptables seront plus faciles à mettre en œuvre politiquement et plus efficaces. Pour que le processus de transformation soit concluant, le dialogue multilatéral doit céder le pas à un véritable processus de négociation multilatéral. En effet, plus les groupes d’intérêt du système alimentaire représentés dans ce processus seront nombreux, plus la légitimité politique des recommandations élaborées en commun sera élevée. CRITÈRES DE SÉLECTION DES MEMBRES DE LA COMMISSION SUR L’AVENIR DU SYSTÈME ALIMENTAIRE Il est essentiel que soient représenté·e·s au sein de la commission tou·te·s les acteurs et actrices clés du système alimentaire suisse. Dans un premier temps, il convient donc d’établir sous supervision scientifique une cartographie des parties prenantes en se basant sur l’approche à la fois réputationnelle, décisionnelle et positionnelle. Celle-ci consiste à classer les parties prenantes selon leur compétence décisionnelle (aspect positionnel) et leur participation aux précédentes décisions politiques (aspect décisionnel)234 puis à soumettre la cartographie qui en découle à des spécialistes du système alimentaire suisse pour qu’il·elles la complètent (aspect réputationnel) (ibid.). Dans un deuxième temps, les parties prenantes recensées sont regroupées en clusters le long de la chaîne de création de valeur. Tout au long du processus, il faut également veiller à intégrer les personnes directement et indirectement concernées (par ex. associations locales et internationales de protection de l’environnement et de la santé) à chaque étape de la chaîne de création de valeur du secteur agroalimentaire. Selon les critères de McGinnis et Ostrom (2014), les personnes concernées sont à considérer comme des parties prenantes, car leur intégration dans le processus de décision et de négociation est essentielle pour l’élabora- En ce qui concerne la composition de la Commission sur l’avenir du système alimentaire, il est essentiel que soient représenté·e·s des acteurs et actrices important·e·s des différents domaines du système alimentaire (agriculture, négoce, transformation, commerce de détail, restauration ; consommateurs et consommatrices, pouvoirs publics, caisses-maladie, banques alimentaires ; associations de protection de l’environnement, associations sociales, médecins et représentant·e·s de patient·e·s) dans un rapport équilibré entre les acteurs et actrices qui défendent en priorité des intérêts d’utilité publique et ceux et celles dont les intérêts relèvent avant tout de l’économie privée (voir encadré p. 63 pour le processus de sélection détaillé des membres de la commission). Pour entamer sur les meilleures bases les négociations en vue de l’élaboration d’un compromis commun, l’ensemble des représentant·e·s signent une déclaration dans laquelle ils et elles s’engagent à adopter une démarche constructive et axée sur l’intérêt général dans le cadre de leur participation à la Commission sur l’avenir du système alimentaire. Les négociations doivent en outre se dérouler selon la règle de Chatham House.225 Celle-ci vise à promouvoir un processus de négociation inclusif et ouvert en permettant aux participant·e·s d’utiliser et de diffuser le contenu des discussions sans toutefois révéler l’identité des personnes ou des organisations qui en sont à l’origine ou qui l’ont relayé. La modération et la médiation indépendantes et professionnelles dans le cadre de la commission sont également essentielles pour une bonne collaboration en vue de l’élaboration tion de compromis inclusifs et durables.235 Il s’agit là d’une condition primordiale pour surmonter les clivages sectoriels et opérer une transformation du système dans le sens du développement durable.236 L’objectif de ce processus est de garantir un rapport équilibré entre les acteurs et actrices qui défendent en priorité des intérêts d’utilité publique et ceux et celles dont les intérêts relèvent avant tout de l’économie privée. Au sein de chaque cluster, ces deux catégories d’acteurs et actrices doivent avoir le même poids afin d’éviter tout déséquilibre et de garantir que les différents intérêts sont représentés de manière équitable. Une fois cette cartographie des parties prenantes achevée, les acteurs et actrices sélectionné·e·s sont regroupé·e·s par types. Les types d’acteurs et actrices homogènes se voient attribuer un siège au sein de la commission, tandis que les types hétérogènes décrochent deux sièges, un pour chaque extrémité du spectre qu’ils couvrent. Une procédure de candidature est ensuite lancée pour sélectionner les représentant·e·s des différent·e·s acteurs et actrices clés identifié·e·s qui siégeront à la commission. Cette procédure permet de s’assurer que le petit nombre de représentant·e·s des différents groupes de parties prenantes disposent effectivement du temps nécessaire pour assumer ce rôle et sont disposé·e·s à œuvrer dans le sens d’un compromis commun. 63 d’un compromis que l’ensemble des parties prenantes puissent accepter et appliquer. Les négociations au sein de la commission doivent porter en particulier sur la formulation d’une vision commune, d’objectifs mesurables pour la transformation durable du système alimentaire ainsi que de pistes de mesures politiques concrètes pour atteindre ces objectifs dans tous les sous-secteurs pertinents du système alimentaire. Les destinataires direct·e·s des recommandations qui découleront de ce processus sont les instances législatives nationale (Parlement fédéral), cantonales (parlements cantonaux et Landsgemeinden) et communales (parlements et conseils communaux). La commission devra également accompagner activement la mise en œuvre et l’évaluation des mesures adoptées. À l’avenir, l’implication des groupes d’intérêt et le processus de dialogue et de négociation multilatéral qui en résulte devront être assurés à toutes les étapes du parcours législatif, de la définition du problème à la mise en œuvre concrète et à l’évaluation, en passant par l’élaboration du projet de loi. Néanmoins, la Commission sur l’avenir du système alimentaire doit surtout se concentrer sur les phases d’établissement de l’agenda et de conception de la politique. Ces premières phases du processus politique détermineront l’orientation générale de la transformation du système alimentaire et la méthode adoptée. Or le soutien des groupes d’intérêt est indispensable pour la bonne mise en œuvre de ce projet. C’est au début du cycle politique que les recommandations d’objectifs et de mesures doivent être soumises au pouvoir législatif pour qu’elles puissent être prises en considération et ainsi contribuer à l’établissement de l’agenda et à la formulation de la politique. Un projet de transformation coordonné par les principales parties prenantes avec des recommanda- 64 tions concrètes peut considérablement simplifier le processus législatif et éviter le risque d’impasse politique. Ce processus de négociation multilatéral devra idéalement se dérouler à tous les niveaux du système politique, de l’échelon communal au contexte national et international. En résumé, l’objectif central de la Commission sur l’avenir du système alimentaire est de servir de vecteur à un processus de négociation multi-acteurs et -actrices, intersectoriel et systémique. Cette nouvelle structure de gouvernance ouvre de nouvelles perspectives de coalition et de solutions pour toutes les parties prenantes, et pose ainsi les jalons d’une transformation réussie du système alimentaire, conformément à la Stratégie pour le développement durable 2030. Des exemples à l’étranger montrent que de tels processus de négociation multi-acteurs et -actrices peuvent contribuer avec succès à la transformation, à l’image de l’Organic Food Council au Danemark ou de la Commission Borchert en Allemagne (plus d’informations à ce sujet dans l’encadré p. 61). 4.2.2 Comité scientifique et participation citoyenne Les connaissances et les recommandations du comité scientifique ainsi que l’avis de la population doivent également être pris en compte à titre d’apports et de soutiens supplémentaires pour la Commission sur l’avenir du système alimentaire et le processus législatif général. Au plus tard d’ici 2025, un Comité scientifique Système alimentaire doit être convoqué officiellement et solidement institutionnalisé. Il s’agira d’un groupe interdisciplinaire de chercheurs et chercheuses de tous les domaines pertinents qui offriront un aperçu systémique de l’état actuel de la recherche sur la transformation du système alimentaire. Les analyses scientifiques fournies par le Comité scientifique Système alimentaire serviront de base de travail à la Commission sur l’avenir du système alimentaire, mais également aux instances exécutives et législatives nationales, cantonales et communales. Le Comité scientifique Système alimentaire entretiendra par ailleurs des contacts étroits avec les offices fédéraux, en particulier avec l’Office fédéral de la statistique, afin de mettre les données collectées et les résultats de recherche à la disposition de toutes les parties intéressées de manière transparente. Il aura également pour mission d’examiner la faisabilité des mesures et des objectifs formulés et d’évaluer scientifiquement leur mise en œuvre. Le rôle du monde scientifique au sein de la Commission sur l’avenir du système alimentaire et dans le processus législatif doit avant tout être consultatif, pour garantir une évaluation et un suivi indépendants des recommandations. Une évaluation scientifique annuelle des objectifs pourrait par ailleurs être envisagée. Ainsi, si la réalisation des objectifs devait se révéler insuffisante, l’instance législative et les offices fédéraux compétents pourraient automatiquement mettre en place des mesures immédiates contraignantes pour atteindre les objectifs définis. D’ici 2030, il faudrait en outre lancer davantage de processus de participation citoyenne afin d’impliquer la population dans la transformation du système alimentaire, et ce à intervalles réguliers et dans une démarche délibérative. La méthode participative recommandée par l’OCDE est celle des assemblées citoyennes, qui regroupe un échantillon représentatif de citoyen·ne·s sélectionné·e·s de manière aléatoire.226 Les personnes sélectionnées se réuniront pour discuter de mesures et d’objectifs communs autour de la transformation du système alimentaire suisse et formuler ensemble des propositions représentatives de l’avis de la population. Cette approche de démocratie participative directe présente le potentiel d’attirer en premier lieu des citoyen·e·s critiques envers la démocratie et le gouvernement227 et ainsi de prévenir activement la polarisation des discussions.228 Il est toutefois essentiel de communiquer dès le début et de manière transparente que les processus de participation citoyenne n’ont pas vocation à faire office de parlements de substitution227 mais à mettre en évidence un certain nombre de perspectives devant être prises en compte dans le processus politique. Le projet « Avenir Alimentaire Suisse » mené à bien en 2022 constitue une première expérience de structure de gouvernance telle que nous la proposons ici. Dans le cadre de ce projet, accompagné et soutenu par l’OFAG, l’OSAV et l’OFEV, une assemblée citoyenne et un comité scientifique ont été mis en place. Le présent rapport est l’un des résultats du travail fourni par le comité scientifique créé dans le cadre d’« Avenir Alimentaire Suisse ». 4.2.3 Renforcement de la collaboration entre les offices et création d’une nouvelle commission thématique D’ici 2025, il conviendrait de renforcer la collaboration entre les offices (surtout entre l’OFAG, l’OFAG et l’OFEV) au sein de l’exécutif national. D’ici 2030, nous recommandons d’envisager la possibilité de faire évoluer l’actuel Office fédéral de l’agriculture en un Office fédéral de l’alimentation et de l’agriculture. La réorientation de cette structure exécutive garantirait à moyen terme l’adoption d’une approche systémique dans le processus de transformation. Au sein d’un Office fédéral de l’alimentation et de l’agriculture, la notion de système alimentaire pourrait être prise en compte de manière plus globale et efficace dans le cadre des processus législatifs et administratifs. Le risque de conflits de compétences entre les offices s’en trouverait par ailleurs réduit. Ce nouvel office pourrait par exemple jouer un rôle moteur dans l’élaboration et la mise en œuvre d’une loi intégrée sur le système alimentaire. Aux échelons cantonal et communal, la collaboration entre les offices devrait également être renforcée et, de plus en plus, planifiée, coordonnée et mise en œuvre selon une approche systémique. D’ici 2030, une Commission de l’alimentation et de l’agriculture pourrait constituer le pendant législatif de l’Office fédéral de l’alimentation et de l’agriculture. Les intérêts seraient ainsi regroupés de manière systémique au sein de ces instances. Ces nouvelles structures simplifieraient les échanges et les procédures et créeraient de nouvelles capacités pour se concentrer sur les questions de fond plutôt que sur les aspects formels et administratifs. Une meilleure intégration du travail des offices, du législatif et des acteurs et actrices tout au long de la chaîne de valeur dans une perspective de système alimentaire permettra de mettre en évidence les conflits d’objectifs, d’identifier des pistes de solutions communes et ainsi de relever les défis actuels de manière plus ciblée et efficace. Pour transformer durablement le système alimentaire suisse et atteindre les objectifs de la Stratégie pour le développement durable 2030, il est donc essentiel de réfléchir en termes de système alimentaire global. Cette perspective systémique doit se refléter dans une structure de gouvernance adéquate et, en particulier, une Commission sur l’avenir du système alimentaire, un Comité scientifique, un Office fédéral de l’alimentation et de l’agriculture ainsi qu’une Commission parlementaire de l’alimentation et de l’agriculture. En résumé, nous pouvons établir qu’une négociation réussie sur la future politique du système alimentaire suppose la mise en place des changements institutionnels adéquats. 65 CHAPITRE FINAL Les événements mondiaux de ces trois dernières années ont montré de manière éclatante que notre système alimentaire ne résiste pas à la crise. Les guerres, les pandémies, l’avancée du changement climatique et la perte de biodiversité exercent une pression accrue sur les chaînes d’approvisionnement mondialisées. L’insécurité de l’approvisionnement et les troubles sociaux en sont les conséquences. En même temps, le système alimentaire actuel est un moteur déterminant de ces crises. Il est par exemple à l’origine de près de 30 % des gaz à effet de serre dans le monde et contribue de manière significative à la perte de la biodiversité. Alors que de nombreuses denrées alimentaires sont gaspillées et que l’obésité constitue un problème de santé majeur, de plus en plus de personnes dans le monde souffrent à nouveau de la faim. Une transformation du système alimentaire en accord avec les objectifs de durabilité est donc nécessaire et urgente pour garantir durablement la sécurité alimentaire mondiale, y compris celle de la Suisse. Les Objectifs de développement durable (ODD) des Nations Unies, qui prennent en compte les trois dimensions de la durabilité (sociale, environnementale et économique), constituent le cadre de cette transformation. Ceci est particulièrement important pour assurer une mutation socialement équitable du système alimentaire. La transformation engendre des coûts à court terme, pour lesquels les perdant·e·s du changement doivent recevoir une compensation adéquate. Ce n’est qu’ainsi que les mesures politiques pourront jouir d’une large légitimité. À moyen et long terme, les avantages de la transformation du système alimentaire l’emportent clairement, en permettant par exemple de réaliser des économies considérables dans le domaine de la santé et de l’adaptation au changement climatique. La Suisse peut particulièrement profiter de ce changement. Les adaptations politiques du système alimentaire suisse ont une importance internationale et peuvent réduire les effets négatifs de la production alimentaire à l’échelle mondiale. Ceci est dans l’intérêt de la Suisse, car en tant que pays importateur et place économique d’importance mondiale, elle est tributaire d’écosystèmes fonctionnels et de chaînes d’approvisionnement stables. Son empreinte élevée de gaz à effet de serre liée à l’alimentation et le fait que 80 % des surfaces utilisées pour l’alimentation le sont à l’étranger montrent que la Suisse influence fortement les écosystèmes, même au-delà de ses frontières. Il ne faut pas non plus sous-estimer le rôle de la Suisse en tant que place politique, scientifique et économique interconnectée au niveau mondial. La Suisse a la chance particulière de pouvoir agir en tant que pionnière pour la transformation du système alimentaire et de jouer un rôle de modèle au niveau mondial. 66 Les efforts consentis jusqu’à présent par l’industrie, comme par exemple les objectifs de réduction des émissions des grandes entreprises alimentaires ou les labels biologiques relativement stricts, ne suffisent toutefois pas à la transformation vers un système alimentaire plus durable. Les mesures actuelles ne vont pas assez loin pour atteindre les objectifs de durabilité d’ici 2030. Ainsi, actuellement, la part de la population suisse touchée par le surpoids ou l’obésité est toujours en augmentation. De plus, la teneur en nitrates de la nappe phréatique se maintient depuis 2002 à un niveau très élevé en comparaison internationale, dépassant largement les limites planétaires. La consommation alimentaire suisse n’est pas non plus en accord avec les objectifs de durabilité. Afin de changer cela et de développer des solutions pour une transformation durable du système alimentaire suisse, SDSN Suisse, l’antenne nationale du Sustainable Development Solutions Network de l’ONU, a mis sur pied le comité scientifique Avenir Alimentaire Suisse. Ce rapport a été rédigé sur la base d’un échange scientifique de plus d’un an entre 42 expert·e·s interdisciplinaires dans le domaine du système alimentaire. Le rapport doit servir de guide pour la transformation nécessaire du système alimentaire suisse et précise que celle-ci est urgente et constitue une tâche pour l’ensemble de la société. Une perspective systémique est indispensable pour réaliser un approvisionnement alimentaire selon les principes de l’agroécologie, qui combine des objectifs économiques, sociaux et écologiques sur l’ensemble du système alimentaire. Après une description de la situation de départ au chapitre 1, ce guide formule au chapitre 2 un objectif clair de transformation du système alimentaire suisse en accord avec les ODD. Pour ce faire, les auteur·e·s du rapport ont défini onze indicateurs cibles prioritaires et mesurables. Il s’agit notamment d’adapter le régime alimentaire suisse d’après les principes du Planetary Health Diet, de réduire les pertes d’azote et de phosphore ainsi que les effets écotoxiques des pesticides, d’utiliser de manière durable les sols exploités par l’agriculture et de réduire la déforestation, les pertes de biodiversité, les émissions de gaz à effet de serre, les déchets alimentaires et l’empreinte hydrique liée à l’alimentation. S’y ajoutent les objectifs sociaux de réduction des risques psychosociaux et physiques pour la santé dans l’agriculture suisse, ainsi que la garantie de revenus suffisants pour vivre et le respect des normes de l’OIT sur le travail des enfants tout au long de la chaîne d’approvisionnement. En comparant le statu quo avec ces indicateurs cibles prioritaires pour la transformation du système alimentaire, les expert·e·s montrent que la Suisse a un grand retard à rattraper pour mettre son système alimentaire en conformité avec les ODD d’ici 2030. Le chapitre 3 montre comment ces objectifs pourraient être atteints concrètement. Le comité scientifique y propose une voie d’action politique concrète. En effet, ce n’est pas seulement la formulation de mesures individuelles qui est déterminante pour le succès, mais surtout leur regroupement, leur coordination stratégique et leur ordre chronologique. Ainsi, des points de basculement dits positifs peuvent être déclenchés. Il s’agit de points critiques dans les systèmes socio-techniques et socio-naturels, où des interventions ciblées peuvent avoir de grands effets positifs et à long terme sur le développement durable du système, et modifier ainsi profondément son fonctionnement. Pour y parvenir, le comité scientifique propose quatre séries de mesures à mettre en œuvre entre 2023 et 2030. La première prévoit un fonds de transformation. Outre des programmes de formation initiale et continue destinés à l’ensemble des actrices et acteurs de la chaîne de valeur alimentaire, ce fonds comprend également l’encouragement financier de la recherche et de l’innovation technologique, ainsi que des programmes de compensation et de promotion des processus de conversion dans l’agriculture et la restauration hors domicile. Le fonds de transformation doit ainsi constituer la base et les connaissances nécessaires à une transformation socialement équitable et durable du système alimentaire suisse. Il peut également permettre à terme l’introduction de mesures plus profondes, car de nouveaux modèles commerciaux deviennent rentables, de nouvelles compétences sont renforcées et le changement social est encouragé. La deuxième série de mesures, qui s’appuie sur le fonds de transformation, se compose de mesures réglementaires et de taxes d’incitation. Dans un troisième temps, il faut mettre en œuvre des mesures de politique agricole tout en soutenant davantage la population rurale. Dans la dernière série de mesures, on essaie, grâce aux progrès réalisés, d’aménager les mesures réglementaires de manière plus stricte. Cette conception stratégique de la politique permet d’enclencher des dynamiques positives de points de basculement dans le système. Grâce à une évolution des normes sociales, de nouvelles technologies, des infrastructures et des opportunités de création de valeur, la transformation peut s’accélérer. Une transformation aussi profonde du système alimentaire suisse nécessite des possibilités de négociation institutionnalisées pour les divers·e·s actrices et acteurs du système. Dans le chapitre 4, le comité scientifique esquisse donc une nouvelle orientation de la gouvernance du système alimentaire à l’horizon 2030. L’élément clé est la mise en place d’une commission sur l’avenir du système alimentaire. Son objectif est de réunir autour d’une table de négociation multilatérale les représentant·e·s de tous les groupes d’intérêt importants du système alimentaire suisse. Ce processus de négociation sur le système alimentaire, basé sur la confiance et une bonne médiation, doit permettre d’établir de nouvelles coalitions d’intérêts et d’élaborer des recommandations viables pour le pouvoir législatif. Le comité scientifique Avenir Alimentaire Suisse recommande en outre de multiplier les procédures d’implication des citoyen·ne·s, d’institutionnaliser un suivi scientifique des processus et de renforcer la collaboration entre les autorités, ce qui déboucherait idéalement sur le développement d’un Office fédéral de l’alimentation et de l’agriculture. La perspective du système alimentaire devrait également être reflétée au niveau législatif par une commission spécialisée dédiée. À moyen terme, le comité scientifique recommande en outre d’envisager l’adoption d’une loi générale sur le système alimentaire. En résumé, le comité scientifique est convaincu que des objectifs mesurables et fondés sur des données probantes, une élaboration stratégique des politiques et une gouvernance générale du système alimentaire peuvent déclencher des points de basculement positifs en accord avec les ODD. Si nous unissons nos forces, la transformation durable du système alimentaire suisse est possible d’un point de vue scientifique. 67 68 ÉQUIPE DE PROJET DU COMITÉ SCIENTIFIQUE AVENIR ALIMENTAIRE SUISSE Les principales autrices et principaux auteurs du guide scientifique sont : Fesenfeld Lukas – Université de Berne / ETH Zurich Mann Stefan – Agroscope Meier Matthias – BFH-HAFL Nemecek Thomas – Agroscope Scharrer Bettina – Université de Berne Bornemann Basil – Université de Bâle Brombach Christine – ZHAW Beretta Claudio – ZHAW Bürgi Elisabeth – Université de Berne Grabs Janina – ESADE Business School / ETH Zurich Ingold Karin – Université de Berne / EAWAG Jeanneret Philippe – Agroscope Kislig Sol – Université de Berne Lieberherr Eva – ETH Zurich Müller Adrian – FiBL Pfister Stephan – ETH Zurich Schader Christian – FiBL Schönberg Sonja – BFH Sonnevelt Martijn – ETH Zurich / World Food System Center Zähringer Julie – Université de Berne / Wyss Academy Le guide doit être cité comme suit : Fesenfeld, L. ; Mann, S. ; Meier, M. ; Nemecek, T. ; Scharrer, B. ; Bornemann, B., Brombach, C. ; Beretta, C. ; Bürgi, E. ; Grabs, J. ; Ingold, K. ; Jeanneret, P. ; Kislig, S. ; Lieberherr, E. ; Müller, A. ; Pfister, S. ; Schader, C. ; Schönberg, S. ; Sonnevelt, M. ; Barjolle, D. ; Boivin, P. ; Brunner, T. ; Contzen, S. ; Espa, I. ; Estève, M. ; Forney, J. ; Häberli C. ; Hediger, W. ; Hilbeck, A. ; Kopainsky, B. ; Lehmann, B. ; Mack, G. ; Markoni, E. ; Meier, B. ; Paccaud, F. ; Rohrmann, S. ; Schindler, M. ; Schwab, C. ; Tribaldos, T. ; Waibel, P. ; Zähringer, J. (2023). L’avenir de l’alimentation en Suisse : Guide des principaux leviers et axes politiques pour établir un système alimentaire durable. SDSN Suisse – https://doi.org/10.5281/zenodo.7585685 Le guide se base sur le travail et les échanges professionnels du comité scientifique interdisciplinaire Avenir Alimentaire Suisse, qui a été officiellement convoqué par SDSN Suisse en février 2022. Le travail technique du comité scientifique et la rédaction du guide ont été coordonnés par le Dr Lukas Fesenfeld. Le comité scientifique a travaillé en grande partie sur la base du bénévolat. Cette publication a été réalisée dans le cadre du projet Avenir Alimentaire Suisse et a bénéficié du soutien financier de SDSN Suisse. SDSN Suisse a fourni un soutien organisationnel. Les personnes suivantes ont joué un rôle important dans ce processus : Carole Küng, Niklaus Salzmann, Sol Kislig, Jorge Tamayo et Nathalie Dubler. Le comité scientifique se compose des expert·e·s scientifiques suivant·e·s: Bornemann Basil – Université de Bâle Brombach Christine – ZHAW Barjolle Dominique – Université de Lausanne Beretta Claudio – ZHAW Bernauer Thomas – ETH Zurich Boivin Pascal HEPIA – HES-SO Genève Brunner Thomas – BFH-HAFL Bürgi Elisabeth – Université de Berne Contzen Sandra – BFH-HAFL Espa Ilaria – Université de Berne / Università della Svizzera italiana Estève Magali – Agridea Fesenfeld Lukas – Université de Berne / ETH Zurich Forney Jérémie – Université de Neuchâtel Grabs Janina – ESADE Business School / ETH Zurich Häberli Christian – Université de Berne Hediger Werner – HES des Grisons Hilbeck Angelika – ETH Zurich Ingold Karin – Université de Berne / EAWAG Jeanneret Philippe – Agroscope Kopainsky Birgit – University of Bergen Lehmann Bernard – HLPE-CFS Lieberherr Eva – ETH Zurich Mack Gabriele – Agroscope Mann Stefan – Agroscope Markoni Evelyn – BFH-HAFL Meier Beat – Indépendant Meier Matthias – BFH-HAFL Müller Adrian – FiBL Nemecek Thomas – Agroscope Paccaud Fred – Hôpital universitaire de Lausanne, CHUV Pfister Stephan – ETH Zurich Pidoux Martin – BFH-HAFL Rohrmann Sabine – Université de Zurich Schader Christian – FiBL Scharrer Bettina – Université de Berne Schindler Magdalena – Indépendante Schönberg Sonja – BFH Schwab Christian Nils – EPFL Lausanne Sonnevelt Martijn – ETH Zurich / World Food System Center Tribaldos Theresa – Université de Berne Waibel Piera – Indépendante Zähringer Julie – Université de Berne / Wyss Academy REMERCIEMENTS Ce projet a été rendu possible grâce à l’engagement des personnes susmentionnées et au soutien de nombreuses autres organisations et personnalités. 69 BIBLIOGRAPHIE ​ Appendice en ligne : https://doi.org/10.5281/zenodo.7543736 1. Rockström, J., Edenhofer, O., Gaertner, J. & DeClerck, F. Planet-proofing the global food system. Nat Food 1, 3–5 (2020). ​2. Poore, J. & Nemecek, T. Reducing food’s environmental impacts through producers and consumers. Science (1979) 360, 987–992 (2018). ​3. Armstrong McKay, D. I. et al. Exceeding 1.5°C global warming could trigger multiple climate tipping points. Science (1979) 377, 1135 (2022). ​ . Perotti, A. Moving Towards a Sustainable Swiss Food System: An Estima4 tion of the True Cost of Food in Switzerland and Implications for Stakeholders. doi:10.3929/ethz-b-000473289 (ETH Zürich, World Food System Center, 2020). ​5. Murray, C. J. L. et al. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet 396, 1223–1249 (2020). ​6. Springmann, M., Clark, M. A., Rayner, M., Scarborough, P. & Webb, P. The global and regional costs of healthy and sustainable dietary patterns: a modelling study. Lancet Planet Health 5, e797–e807 (2021). ​7. FAO, IFAD, UNICEF, WFP & WHO. The State of Food Security and Nutrition in the World 2022. Repurposing food and agricultural policies to make healthy diets more affordable. doi:10.4060/cc0639en (2022). ​8. Rao, M., Afshin, A., Singh, G. & Mozaffarian, D. Do healthier foods and diet patterns cost more than less healthy options? A systematic review and meta-analysis. BMJ Open 3, (2013). ​9. Contzen, S. & Crettaz, E. Being a poor farmer in a wealthy country: A Swiss case study. Sociol Ruralis 59, 393–418 (2019). ​10. Whitfield, S. et al. A framework for examining justice in food system transformations research. Nat Food 2, 383–385 (2021). 11. High Level Panel of Experts (HLPE). Food security and nutrition – ​ Building a global narrative towards 2030. https://www.fao.org/3/ca9731en/ ca9731en.pdf (2020). ​12. Schweizerischer Bundesrat. Zukünftige Ausrichtung der Agrarpolitik: Bericht des Bundesrates in Erfüllung der Postulate 20.3931 der WAK-S vom 20. August 2020 und 21.3015 der WAK-N vom 2. Februar 2021. https://www.blw.admin.ch/ dam/blw/de/bilder/website/Politik/postulat.pdf.download.pdf/Bericht%20 in%20Erf%C3%BCllung%20Postulat%20WAK-S%2020.3931_21.3015_d.pdf (2022). ​ 3. United Nations (UN). Transforming our World: The 2030 Agenda for Sustain1 able Development. https://sdgs.un.org/publications/transforming-our-world2030-agenda-sustainable-development-17981 (2015). ​28. Schweizerischer Bundesrat. Die Umsetzung der Agenda 2030 für nachhaltige Entwicklung: Länderbericht der Schweiz 2022. https://www.eda.admin.ch/ content/dam/agenda2030/de/documents/laenderbericht2022_DE.pdf (2022). ​29. Schweizerischer Bundesrat. Statistischer Anhang zum Länderbericht 2022 der Schweiz. https://www.eda.admin.ch/content/dam/agenda2030/de/documents/laenderbericht2022-statistischer-anhang_DE.pdf (2022). ​30. Bundesamt für Gesundheit (BAG) & Schweizerische Konferenz der kantonalen Gesundheitsdirektorinnen und -direktoren (GDK). Nationale Strategie Prävention nichtübertragbarer Krankheiten (NCD-Strategie): 2017–2024. https:// www.bag.admin.ch/dam/bag/de/dokumente/nat-gesundheitsstrategien/ncdstrategie/ncd-strategie.pdf.download.pdf/ncd-strategie.pdf (2016). 31. Bundesamt für Statistik (BFS). Schweizerische Gesundheitsbefragung ​ 2017: Übergewicht und Adipositas. https://www.bfs.admin.ch/bfs/de/home/ aktuell/neue-veroeffentlichungen.assetdetail.14147705.html (2020). ​32. Bundesamt für Umwelt (BAFU). Zustand und Entwicklung Grundwasser Schweiz. https://www.bafu.admin.ch/dam/bafu/de/dokumente/wasser/uz-umwelt-zustand/zustandundentwicklunggrundwasserschweiz.pdf.download.pdf/ UZ-1901-D_NAQUA.pdf (2019). ​33. Bundesamt für Statistik (BFS). Das MONET 2030-Indikatorensystem. https://www.bfs.admin.ch/bfs/de/home/statistiken/nachhaltige-entwicklung/ monet-2030.html (2022). ​34. Bundesamt für Umwelt (BAFU). Indikator Wasser. https://www.bafu.admin.ch/ bafu/de/ home/themen/thema-wasser/wasser--daten--indikatoren-und-karten/ wasser--indikatoren/indikator-wasser.pt.html/aHR0cHM6Ly93d3cuaW5kaW thdG9yZW4uYWRtaW4uY2gvUHVibG/ljL0FlbURldGFpbD9pbmQ9V1MwNDg mbG 5nPWRlJlN1Ymo9Tg%3d%3d.html (2022). ​35. Fesenfeld, L. P., Schmidt, T. & Schrode, A. Climate policy for short- and longlived pollutants. Nat Clim Chang 8, 933–936 (2018). ​36. Bundesamt für Umwelt (BAFU). Ökofaktoren Schweiz 2021 gemäss der Methode der ökologischen Knappheit Methodische Grundlagen und Anwendung auf die Schweiz. https://www.bafu.admin.ch/dam/bafu/de/dokumente/ wirtschaft-konsum/uw-umwelt-wissen/UW-2121-D_UBP.pdf.download.pdf/ UW-2121-D_UBP.pdf (2021). ​37. Nathani, C.; O’Connor, I.; Frischknecht, R.; Schwehr, T.; Zumwald, J.; Peyronne, J. Umwelt-Fussabdrücke der Schweiz: Entwicklung zwischen 2000 und 2018. https://www.bafu.admin.ch/dam/bafu/de/dokumente/wirtschaft-konsum/ externe-studien-berichte/umwelt-fussabdruecke-der-schweiz-entwicklungzwischen-2000-2018.pdf.download.pdf/EBP_(2022)_Umweltfussabdruecke. pdf Im Auftrag: Bundesamt für Umwelt. (2022).​38. Huber, R. Einführung in die Schweizer Agrarpolitik. (vdf Hochschulverlag AG an der ETH Zürich, 2021). doi:10.3929/ethz-b-000530354. ​39. Eidgenössisches Departement für auswärtige Angelegenheiten (EDA). Agenda 2030 für nachhaltige Entwicklung: Wechselwirkungen. https://www.sdgital2030.ch/interactions#summary (2022). ​40. Fesenfeld, L. P., Schmid, N., Finger, R., Mathys, A. & Schmidt, T. S. The politics of enabling tipping points for sustainable development. One Earth 5, 1100–1108 (2022). ​ 4. Schweizerische Eidgenossenschaft. Agenda 2030 und SDGs. 1 https://www.eda.admin.ch/agenda2030/de/home/agenda-2030.html (2021). ​41. Bundesamt für Lebensmittelsicherheit und Veterinärwesen (BLV). Aktionsplan der Schweizer Ernährungsstrategie. https://www.aktionsplanernaehrung. ch/ (2018). 15. United Nations Framework Convention on Climate Change (UNFCCC). ​ The Paris Agreement. http://unfccc.int/files/essential_background/convention/ application/pdf/english_paris_agreement.pdf (2016). 42. Schweizerischer Bundesrat. Langfristige Klimastrategie der Schweiz. ​ https://www.newsd.admin.ch/newsd/message/attachments/65874.pdf (2021). ​ 6. Niles, M. T. et al. Climate change mitigation beyond agriculture: a review of 1 food system opportunities and implications. Renewable Agriculture and Food Systems 33, 297–308 (2018). ​43. Bundesamt für Umwelt (BAFU). Bodenstrategie Schweiz – für einen nachhaltigen Umgang mit dem Boden. https://www.bafu.admin.ch/dam/bafu/de/ dokumente/boden/ud-umwelt-diverses/bodenstrategie-schweiz.pdf.download.pdf/Bodenstrategie_2020-05-01.pdf (2020). ​17. Allievi, F., Antonelli, M., Dembska, K. & Principato, L. Understanding the global food system. in Achieving the Sustainable Development Goals Through Sustainable Food Systems (eds. Valentini, R., Sievenpiper, J. L., Antonelli, M. & Dembska, K.) 3–23 (Springer International Publishing, 2019). ​44. Rockström, J. & Sukhdev, P. How food connects all the SDGs. Opening keynote speech at the 2016 EAT Forum. https://www.youtube.com/watch?time_continue=704&v=tah8QlhQLeQ&feature=emb_title (2016). ​18. Fassio, F. & Tecco, N. Circular Economy for Food: A Systemic Interpretation of 40 Case Histories in the Food System in Their Relationships with SDGs. Systems 7, 43 (2019). ​ 9. Lindgren, E. et al. Sustainable food systems—a health perspective. 1 Sustain Sci 13, 1505–1517 (2018). ​45. Steffen, W. et al. Planetary boundaries: Guiding human development on a changing planet. Science (1979) 347, 1259855 (2015). ​46. Schweizerische Gesellschaft für Ernährung (SGE). Schweizer Lebensmittelpyramide – Hintergrundinformationen. www.sge-ssn.ch/lebensmittelpyramide (2011). ​20. Herrero, M. et al. Articulating the effect of food systems innovation on the Sustainable Development Goals. Lancet Planet Health 5, e50–e62 (2021). ​47. von Ow, A., Waldvogel, T. & Nemecek, T. Environmental optimization of the Swiss population’s diet using domestic production resources. J Clean Prod 248, 119241 (2020). ​21. Strzepek, K. & Boehlert, B. Competition for water for the food system. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 2927–2940 (2010). ​48. Ernstoff, A. et al. Towards Win–Win Policies for Healthy and Sustainable Diets in Switzerland. Nutrients 12, 2745 (2020). ​22. Springmann, M. et al. Options for keeping the food system within environmental limits. Nature 562, 519–525 (2018). ​49. Willet, J., Wetser, K., Vreeburg, J. & Rijnaarts, H. H. M. Review of methods to assess sustainability of industrial water use. Water Resour Ind 21, 100110 (2019). 23. Schweizerischer Bundesrat. Umwelt Schweiz 2022. https://www.bafu. ​ admin.ch/dam/bafu/de/dokumente/allgemein/uz-umwelt-zustand/umweltbericht2022.pdf.download.pdf/umweltbericht2022.pdf (2022). ​50. Zimmermann, A., Nemecek, T. & Waldvogel, T. Umwelt-und ressourcenschonende Ernährung: Detaillierte Analyse für die Schweiz. Agroscope Science 55, 1–170 (2017). ​24. Chaudhary, A. & Kastner, T. Land use biodiversity impacts embodied in international food trade. Global Environmental Change 38, 195–204 (2016). ​51. Jungbluth, N. et al. Analysis of food and environmental impacts as a scientific basis for Swiss dietary recommendations. ESU-services GmbH (2022) doi:10.13140/RG.2.2.28446.41287 (2022). ​25. Liu, J. et al. Framing Sustainability in a Telecoupled World. Ecology and Society 18, (2013). ​ 6. Schleicher, J. et al. Protecting half of the planet could directly affect over 2 one billion people. Nat Sustain 2, 1094–1096 (2019). 70 27. Sachs, J., Kroll, C., Lafortune, G., Fuller, G. & Woelm, F. Sustain​ able Development Report 2022. Sustainable Development Report 2022. doi:10.1017/9781009210058 (2022). ​ 52. Bundesamt für Lebensmittelsicherheit und Veterinärwesen (BLV). Nationale Ernährungserhebung menuCH 2014/15: Realisierte und geplante Auswertungen. https://www.blv.admin.ch/dam/blv/de/dokumente/lebensmittel-und-ernaehrung/ernaehrung/nationale-ernaehrungserhebung-menuch-2014-2015.pdf. download.pdf/Uebersicht%20Auswertungen%20menuCH2014-15%20-%20 Stand%20Sept%202022.pdf (2022). ​ 3. Bundesamt für Lebensmittelsicherheit und Veterinärwesen (BLV). Schweizer 5 Ernährungsbulletin 2021. doi:10.24444/blv-2021-0111 (2021). ​ 4. Schweizer Bauernverband & Agristat. Nahrungsmittelbilanz. https://www. 5 sbv-usp.ch/fileadmin/user_upload/07_SES2021_Nahrungsmittelbilanz.pdf (2022). ​ 5. World Health Organization (WHO) & Food and Agriculture Organization of 5 the United Nations (FAO). Sustainable healthy diets: guiding principles. https:// apps.who.int/iris/handle/10665/329409 (World Health Organization (WHO), 2019). ​6. Renner, B. et al. DGE-Positionspapier zur nachhaltigeren Ernährung. 5 doi:10.4455/eu.2021.030 (2021). ​ 7. Garnett, T. et al. What is a sustainable healthy diet? A discussion paper. 5 https://hdl.handle.net/10568/35584 (2014). ​ 8. Tobi, R. C. A. et al. Sustainable Diet Dimensions. Comparing Consumer 5 Preference for Nutrition, Environmental and Social Responsibility Food Labelling: A Systematic Review. Sustainability 11, 6575 (2019). ​ 9. Triches, R. M. Sustainable diets: definition, state of the art and perspectives 5 for a new research agenda in Brazil. Cien Saude Colet 26, 1833–1846 (2021). ​ 0. Ahmed, S., Downs, S. & Fanzo, J. Advancing an Integrative Framework to 6 Evaluate Sustainability in National Dietary Guidelines. Front Sustain Food Syst 3, 76 (2019). ​ 1. Willett, W. et al. Food in the Anthropocene: the EAT–Lancet Commission on 6 healthy diets from sustainable food systems. The Lancet 393, 447–492 (2019). ​ 2. Bundesamt für Landwirtschaft (BLW). Agrarbericht 2019: Fleisch und Eier. 6 https://2019.agrarbericht.ch/de/markt/tierische-produkte/fleisch-und-eier (2019). ​ 3. Fesenfeld, L. A question of perspective. ETH Zürich Zukunftsblog 6 https://ethz.ch/en/news-and-events/eth-news/news/2018/12/zk-fesenfeld-lukasklimagase.html (2018). ​ 4. Schader, C. et al. Impacts of feeding less food-competing feedstuffs to 6 livestock on global food system sustainability. J R Soc Interface 12, 20150891 (2015). ​ 5. de Vries, F. T. et al. Soil food web properties explain ecosystem services 6 across European land use systems. Proc Natl Acad Sci 110, 14296–14301 (2013). ​ 6. Hürdler, J., Prasuhn, V. & Spiess, E. Abschätzung diffuser Stickstoff- und 6 Phosphoreinträge in die Gewässer der Schweiz: MODIFFUS 3.0. https://ira.agroscope.ch/de-CH/publication/35052 (2015). 79. Schweizerischer Bundesrat. Aktionsplan zur Risikoreduktion und nachhaltigen Anwendung von Pflanzenschutzmitteln. https://www.blw.admin.ch/dam/ blw/de/dokumente/Nachhaltige%20Produktion/Pflanzenschutz/AktionsplanPflanzenschutzmittel/Aktionsplan_Pflanzenschutzmittel_de.pdf.download.pdf/ Aktionsplan_Pflanzenschutzmittel_de.pdf (2017). ​80. Motta, E. V. S., Raymann, K. & Moran, N. A. Glyphosate perturbs the gut microbiota of honey bees. Proc Natl Acad Sci U S A 115, 10305–10310 (2018). ​ 1. Fischer, M. et al. Summary for policymakers of the regional assessment 8 report on biodiversity and ecosystem services for Europe and Central Asia of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. (2018). ​ 2. Baldi, I. et al. Pesticides : Effets sur la santé. https://www.hal.inserm.fr/in8 serm-02102981/document (2013). ​ 3. Kim, K.-H., Kabir, E. & Jahan, S. A. Exposure to pesticides and the asso8 ciated human health effects. Science of The Total Environment 575, 525–535 (2017). ​ 4. Parrón, T., Requena, M., Hernández, A. F. & Alarcón, R. Environmental ex8 posure to pesticides and cancer risk in multiple human organ systems. Toxicol Lett 230, 157–165 (2014). ​ 5. Möhring, N. et al. Pathways for advancing pesticide policies. Nat Food 1, 8 535–540 (2020). ​86. Mathis, M. et al. Comparison of exemplary crop protection strategies in Swiss apple production: Multi-criteria assessment of pesticide use, ecotoxicological risks, environmental and economic impacts. Sustain Prod Consum 31, 512–528 (2022). ​87. de Baan, L. Sensitivity analysis of the aquatic pesticide fate models in SYNOPS and their parametrization for Switzerland. Sci Total Environ 715, 136881 (2020). ​88. Waldvogel, T., Mathis, M., de Baan, L., Haupt, C. & Nemecek, T. Bewertung der Umweltwirkungen und Risiken verschiedener Pflanzenschutzstrategien für fünf Kulturen in der Schweiz. Agroscope Science 64, 174 (2018). ​89. Bystricky, M., Nemecek, T., Krause, S., Gaillard, G. Potenzielle Umweltfolgen einer Umsetzung der Trinkwasserinitiative. Agroscope Science 99, 1-221 (2020). ​ 0. Furrer, C., Stüssi, M. & Bystricky, M. Einfluss von Import- Herkunftsländern 9 und Nahrungsmittelverlusten auf die Umweltwirkungen des Schweizer Agrarsektors. Agroscope Science 114, 1-29 (2021). ​91. Rosenbaum, R. K. et al. The Glasgow consensus on the delineation between pesticide emission inventory and impact assessment for LCA. International Journal of Life Cycle Assessment 20, 765–776 (2015). ​92. Nemecek, T. et al. Operationalising emission and toxicity modelling of pesticides in LCA: the OLCA-Pest project contribution. International Journal of Life Cycle Assessment 27, 527–542 (2022). ​ 7. Malone, T. C. & Newton, A. The Globalization of Cultural Eutrophication in 6 the Coastal Ocean: Causes and Consequences. Front Mar Sci 7, (2020). ​93. Bircher, P., Liniger, H. & Prasuhn, V. Aktualisierung und Optimierung der Erosionsrisikokarte (ERK2): Die neue ERK2 (2019) für das Ackerland der Schweiz. https://ira.agroscope.ch/de-CH/publication/41816 (2019). ​ 8. Rihm, B. & Achermann, B. Critical Loads of Nitrogen and their Exceedances 6 - Swiss contribution to the effects-oriented work under the Convention on Longrange Transboundary Air Pollution (UNECE). doi:10.13140/RG.2.2.16281.01124 (2016). ​ 4. Schmidt, S., Alewell, C. & Meusburger, K. Mapping spatio-temporal dynam9 ics of the cover and management factor (C-factor) for grasslands in Switzerland. Remote Sens Environ 211, 89–104 (2018). ​ 9. Rihm, B. & Künzle, T. Mapping Nitrogen Deposition 2015 for Switzerland. 6 https://www.bafu.admin.ch/dam/bafu/en/dokumente/luft/externe-studien-berichte/mapping-nitrogen-deposition-2015-for-switzerland.pdf.download.pdf/ Mapping%20Nitrogen%20Deposition%202015%20for%20Switzerland.pdf (2019). ​ 0. Roth, T., Kohli, L., Rihm, B., Amrhein, V. & Achermann, B. Nitrogen deposi7 tion and multi-dimensional plant diversity at the landscape scale. R Soc Open Sci 2, 150017 (2015). ​ 5. Beuttler, C., Charles, L. & Wurzbacher, J. The Role of Direct Air Capture in 9 Mitigation of Anthropogenic Greenhouse Gas Emissions. Frontiers in Climate 1, 10 (2019). ​96. European Commission & Directorate-General for Environment. Review of existing information on the interrelations between soil and climate change. doi:https://doi.org /10.2779/12723 (2012). ​ 7. Heller, O., Peter, N., Garland, G. & Weisskopf, P. EJP SOIL Stocktake-Bei9 trag der Schweiz: Tasks 2.1, 2.2 und 2.3. doi:10.34776/as117g (2021). ​ 1. Roth, T., Kohli, L., Rihm, B., Meier, R. & Amrhein, V. Negative effects of 7 nitrogen deposition on Swiss butterflies. Conservation Biology 35, 1766–1776 (2021). ​8. Bundesamt für Umwelt (BAFU). Boden in der Schweiz. https://www. 9 bafu.admin.ch/dam/bafu/de/dokumente/boden/uz-umwelt-zustand/bodenschweiz.pdf.download.pdf/UZ-1721-D_Boden2017.pdf (2017). ​ 2. Guntern, J. et al. Übermässige Stickstoff- und Phosphoreinträge schädigen 7 Biodiversität, Wald und Gewässer. Swiss Academies Factsheet 15, (2020). ​ 9. Autret, B. et al. Alternative arable cropping systems: A key to increase soil 9 organic carbon storage? Results from a 16 year field experiment. Agric Ecosyst Environ 232, 150–164 (2016). ​ 3. Sutton, M. A. et al. The European Nitrogen Assessment: Sources, Effects 7 and Policy Perspectives. doi:10.1017/CBO9780511976988 (Cambridge University Press, 2011). ​4. Augustin, S. & Achermann, B. Deposition von Luftschadstoffen in der 7 Schweiz: Entwicklung, aktueller Stand und Bewertung. Schweizerische Zeitschrift für Forstwesen 163, 323–330 (2012). ​ 5. Bundesamt für Umwelt (BAFU). Kenngrössen zur Entwicklung der Treib7 hausgasemissionen in der Schweiz 1990–2020. https://www.bafu.admin.ch/ dam/bafu/en/dokumente/klima/klima-climatereporting/National_Inventory_Report_CHE.pdf.download.pdf/National_Inventory_Report_CHE_2022.pdf (2022). ​ 6. Schweizerischer Bundesrat. Botschaft zur Weiterentwicklung der Agrar7 politik ab 2022 (AP22+). https://fedlex.data.admin.ch/filestore/fedlex.data.admin.ch/eli/fga/2020/881/de/pdf-a/fedlex-data-admin-ch-eli-fga-2020-881-depdf-a.pdf (2020). ​ 7. Reutimann, J.; Ehrler, A; Schäppi, B. Aktualisierung Stoffflussanalyse Stick7 stoff für das Jahr 2018. Im Auftrag: Bundesamt für Landwirtschaft. (2022) https://www.aramis.admin.ch/Dokument.aspx?DocumentID=69014 ​ 00. Keel, S. G. et al. Loss of soil organic carbon in Swiss long-term agricul1 tural experiments over a wide range of management practices. Agric Ecosyst Environ 286, 106654 (2019). ​ 01. Díaz, S. et al. Summary for policymakers of the global assessment report 1 on biodiversity and ecosystem services. doi:10.5281/ZENODO.3553579 (2019). ​102. Tollefson, J. Why deforestation and extinctions make pandemics more likely. Nature 584, 175–176 (2020). ​ 03. Bundesamt für Umwelt (BAFU). Biodiversität: Das Wichtigste in Kürze. 1 https://www.bafu.admin.ch/bafu/de/home/themen/biodiversitaet/inkuerze. html (2022). ​ 04. Bundesamt für Umwelt (BAFU). Biodiversität in der Schweiz: Zustand und 1 Entwicklung. https://www.bafu.admin.ch/dam/bafu/de/dokumente/biodiversitaet/uz-umwelt-zustand/biodiversitaet-schweiz-zustand-entwicklung.pdf. download.pdf/UZ-1630-D_2017-06-20.pdf (2017). ​ 8. Sánchez-Bayo, F. & Wyckhuys, K. A. G. Worldwide decline of the entomo7 fauna: A review of its drivers. Biol Conserv 232, 8–27 (2019). ​105. Geiger, W., Göttin, T., Guignet, E. M., Hofmann, C. & Wild, F. Stratégie Biodiversité Suisse. https://www.bafu.admin.ch/dam/bafu/fr/dokumente/biodiversitaet/ud-umwelt-diverses/strategie_biodiversitaetschweiz.pdf.download. pdf/strategie_biodiversitesuisse.pdf (2012). ​ ​ 71 106. Bundesamt für Umwelt (BAFU) & Bundesamt für Landwirtschaft (BLW). Umweltziele Landwirtschaft. Statusbericht 2016. https://www.bafu.admin.ch/ dam/bafu/de/dokumente/wirtschaft-konsum/uz-umwelt-zustand/uz-1811-d.pdf. download.pdf/uz-1811-d.pdf (2016). ​ 07. Secretariat of the Convention on Biological Diversity (CBD). 1 Global Biodiversity Outlook 5. https://www.cbd.int/gbo/gbo5/publication/gbo5-en.pdf (2020). ​ 36. Musselli, I. & Bürgi Bonanomi, E. Countering Commodity Trade Mispric1 ing in Low-Income Countries: A Prescriptive Approach. Journal of International Economic Law 25, 447–463 (2022). ​108. Cardinale, B. J. et al. Biodiversity loss and its impact on humanity. Nature 486, 59–67 (2012). ​ 37. Leinert, L., Brand, F. S. & Duma, F. Food Sector Study: Figures, Struc1 ture, Forces of Change, and Development Potential of the Swiss Food Industry. https://digitalcollection.zhaw.ch/bitstream/11475/1128/1/2016_Jul_Branchenbericht%20Food%20%28e%29_Digitalcollection.pdf (2016). ​ 09. Díaz, S., Fargione, J., Chapin, F. S. & Tilman, D. Biodiversity Loss Threat1 ens Human Well-Being. PLoS Biol 4, e277 (2006). ​ 38. Bundesamt für Landwirtschaft (BLW). Agrarbericht 2021. 1 https://2021.agrarbericht.ch/de (2021). ​ 10. Allan, E. et al. A comparison of the strength of biodiversity effects across 1 multiple functions. Oecologia 173, 223–237 (2013). ​ 39. Reissig, L., Crameri, A. & von Wyl, A. Prevalence and predictors of burnout 1 in Swiss farmers – Burnout in the context of interrelation of work and household. Ment Health Prev 14, 200157 (2019). ​ 11. Kremen, C. Managing ecosystem services: what do we need to know 1 about their ecology? Ecol Lett 8, 468–479 (2005). ​12. Bundesamt für Umwelt (BAFU). Umwelt-Fussabdrücke der Schweiz. 1 https://www.bafu.admin.ch/dam/bafu/de/dokumente/wirtschaft-konsum/uzumwelt-zustand/uz-1811-d.pdf.download.pdf/uz-1811-d.pdf (2018). ​ 13. Rockström, J. et al. A safe operating space for humanity. Nature 461, 1 472–475 (2009). ​ 14. Ridoutt, B. G. & Pfister, S. A revised approach to water footprinting to 1 make transparent the impacts of consumption and production on global freshwater scarcity. Global Environmental Change 20, 113–120 (2010). ​ 15. Mekonnen, M. M. & Hoekstra, A. Y. Sustainability: Four billion people fac1 ing severe water scarcity. Sci Adv 2, (2016). ​ 16. Zipper, S. C. et al. Integrating the Water Planetary Boundary With Water 1 Management From Local to Global Scales. Earths Future 8, e2019EF001377 (2020). ​ 17. Bjørn, A. et al. A planetary boundary-based method for freshwater use in 1 life cycle assessment: Development and application to a tomato production case study. Ecol Indic 110, 105865 (2020). ​ 18. Boulay, A.-M. et al. The WULCA consensus characterization model for 1 water scarcity footprints: assessing impacts of water consumption based on available water remaining (AWARE). Int J Life Cycle Assess 23, 368–378 (2018). ​ 19. Food and Agriculture Organization of the United Nations (FAO). Water use 1 in livestock production systems and supply chains: Guidelines for assessment (Version 1). doi:10.4060/ca5685en (2019). ​120. Nathani, C. et al. Environmental hotspots in the supply chain of Swiss companies. https://treeze.ch/fileadmin/user_upload/downloads/Publications/ Case_Studies/Lifestyles/629_UHU_FinalReport_EN_v1.8.pdf (2019). ​ 21. Kulionis, V. & Pfister, S. A planetary boundary-based method to assess 1 freshwater use at the global and local scales. Environmental Research Letters 17, 094031 (2022). ​ 22. Kulionis, V., Froemelt, A. & Pfister, S. Multiscale Orientation Values for Bio1 diversity Climate and Water. doi:10.13140/RG.2.2.32072.44806 (2021). ​ 23. Schweizerischer Bundesrat. Bundesrat will bis 2050 eine klimaneutrale 1 Schweiz. https://www.admin.ch/gov/de/start/dokumentation/medienmitteilungen.msg-id-76206.html (2019). ​ 40. Steck, N., Junker, C., Bopp, M., Egger, M. & Zwahlen, M. Time trend of 1 suicide in Swiss male farmers and comparison with other men: A cohort study. Swiss Med Wkly 150, 21–22 (2020). ​ 41. Noguchi. ILO Convention No. 182 on the worst forms of child labour and 1 the Convention on the Rights of the Child. The International Journal of Children’s Rights 10, 355–369 (2002). ​ 42. Spiess, E. & Liebisch, F. Nährstoffbilanz der schweizerischen Landwirt1 schaft für die Jahre 1975 bis 2018. Agroscope Science 100, 1-30 (2020). ​ 43. Bundesamt für Landwirtschaft (BLW). Übersicht Verordnungspaket 1 Parlamentarische Initiative 19.475 «Das Risiko beim Einsatz von Pestiziden reduzieren». https://www.blw.admin.ch/dam/blw/de/dokumente/Politik/ Agrarpolitik/Parlamentarische_Initiative__Pestizide_reduzieren/uebersicht. pdf.download.pdf/%C3%9Cbersicht%20Verordnungspaket%20Pa.Iv.%20 19.475_d.pdf (2021). ​144. Social Hotspots Database. http://www.socialhotspot.org/ (2022). ​ 45. The Food and Land Use Coalition (FOLU). Accelerating the 10 Critical Tran1 sitions: Positive Tipping Points for Food and Land Use Systems Transformation. https://www.foodandlandusecoalition.org/wp-content/uploads/2021/07/Positive-Tipping-Points-for-Food-and-Land-Use-Systems-Transformation.pdf (2021). ​ 46. Fesenfeld, L. P., Rudolph, L. & Bernauer, T. Policy framing, design and 1 feedback can increase public support for costly food waste regulation. Nature Food 3, 227–235 (2022). ​ 47. Fesenfeld, L. Glimmers of hope: a global Green New Deal is feasible. 1 GAIA 30, 150–155 (2021). ​ 48. Schmidt, T. S. & Sewerin, S. Technology as a driver of climate and energy 1 politics. Nat Energy 2, 17084 (2017). ​ 49. Sharpe, S. & Lenton, T. M. Upward-scaling tipping cascades to meet climate 1 goals: plausible grounds for hope. Climate Policy 21, 421–433 (2021). ​ 50. Abson, D. J. et al. Leverage points for sustainability transformation. 1 Ambio 46, 30–39 (2017). ​ 51. Lenton, T. M. et al. Operationalising positive tipping points towards global 1 sustainability. Global Sustainability 5, e1 (2022). ​ 52. Fesenfeld, L. P., Wicki, M., Sun, Y. & Bernauer, T. Policy packaging can 1 make food system transformation feasible. Nature Food 1, 173–182 (2020). ​ 24. Fankhauser, S. et al. The meaning of net zero and how to get it right. Nat Clim 1 Chang 12, 15–21 (2022). ​ 53. Meckling, J., Sterner, T. & Wagner, G. Policy sequencing toward decarbon1 ization. Nat Energy 2, 918–922 (2017). ​ 25. Rogelj, J., Geden, O., Cowie, A. & Reisinger, A. Net-zero emissions targets 1 are vague: three ways to fix. Nature 591, 365–368 (2021). ​ 54. Meckling, J., Kelsey, N., Biber, E. & Zysman, J. Winning coalitions for climate 1 policy. Science 349, 1170–1171 (2015). ​ 26. van Soest, H. L., den Elzen, M. G. J. & van Vuuren, D. P. Net-zero emission 1 targets for major emitting countries consistent with the Paris Agreement. Nat Commun 12, 2140 (2021). ​ 55. Pahle, M. et al. Sequencing to ratchet up climate policy stringency. Nat Clim 1 Chang 8, 861–867 (2018). ​ 27. Clark, M. A. et al. Global food system emissions could preclude achieving 1 the 1.5° and 2°C climate change targets. Science (1979) 370, 705–708 (2020). ​ 28. Hayek, M. N., Harwatt, H., Ripple, W. J. & Mueller, N. D. The carbon op1 portunity cost of animal-sourced food production on land. Nat Sustain 4, 21–24 (2021). ​ 29. Cheng, C. H. & Redfern, S. A. T. Impact of interannual and multidecadal 1 trends on methane-climate feedbacks and sensitivity. Nature Communications 13, 1–11 (2022). ​30. Schweizerischer Bundesrat. Aktionsplan gegen die Lebensmittelver1 schwendung: Bericht des Bundesrates in Erfüllung des Postulates 18.3829 Chevalley vom 25. September 2018. https://www.bafu.admin.ch/dam/bafu/de/ dokumente/abfall/externe-studien-berichte/aktionsplan_gegen_die_lebensmittelverschwendung.pdf.download.pdf/Aktionsplan%20gegen%20die%20Lebensmittelverschwendung.pdf (2022). ​ 31. Bundesamt für Umwelt (BAFU). Lebensmittelabfälle. https://www.bafu. 1 admin.ch/bafu/de/home/themen/abfall/abfallwegweiser-a-z/biogene-abfaelle/ abfallarten/lebensmittelabfaelle.html (2022). ​ 32. Beretta, C. & Hellweg, S. Lebensmittelverluste in der Schweiz: Mengen 1 und Umweltbelastung. Wissenschaftlicher Schlussbericht. https://www.newsd. admin.ch/newsd/message/attachments/58769.pdf (2019). ​33. Schweizerischer Bundesrat. Strategie Nachhaltige Entwicklung 2030. 1 https://www.eda.admin.ch/agenda2030/de/home/strategie/strategie-nachhaltige-entwicklung.html (2021). ​ 34. Jan, P. et al. Die wirtschaftliche Entwicklung der schweizerischen Land1 wirtschaft 2020: Zentrale Auswertung von Buchhaltungsdaten, Stichprobe Einkommenssituation. https://www.newsd.admin.ch/newsd/message/attachments/68446.pdf (2021). 72 ​ 35. Meemken, E. M. & Bellemare, M. F. Smallholder farmers and contract farm1 ing in developing countries. Proc Natl Acad Sci U S A 117, 259–264 (2020). ​ 56. Fesenfeld, L. P. et al. Enabling Tipping Dynamics in Food System Trans1 formation: How Information and Experience with Novel Meat Substitutes Can Create Positive Political Feedbacks. SSRN Electronic Journal. doi:10.2139/SSRN.3911306 (2021). ​ 57. Leipprand, A., Flachsland, C. & Pahle, M. Starting low, reaching high? 1 Sequencing in EU climate and energy policies. Environ Innov Soc Transit 37, 140–155 (2020). ​ 58. Schmidt, T. & Sewerin, S. Technology as a driver of climate and energy 1 politics. Nat Energy 2, 17084 (2017). ​ 59. Good Food Institute Europe (GFI). Denmark announces 1 billion kroner 1 for plant-based foods in historic climate agreement. https://gfieurope.org/blog/ denmark-plant-based-investment-in-climate-agreement/ (2021). ​ 60. Fesenfeld, L. P., Sun, Y., Wicki, M. & Bernauer, T. The role and limits of 1 strategic framing for promoting sustainable consumption and policy. Global Environmental Change 68, 102266 (2021). ​ 61. Biasini, B. et al. Understanding, promoting and predicting sustainable diets: 1 A systematic review. Trends Food Sci Technol 111, 191–207 (2021). ​ 62. GFI Israel. Israel State of Alternative Protein Innovation Report 2021. 1 https://innovationisrael.org.il/sites/default/files/GFI%20Israel%20State%20 of%20Alternative%20Protein%20Innovation%20Report%202021_0.pdf (2021). ​ 63. ETH Zürich. Novel Proteins for Food and Feed. https://worldfoodsystem. 1 ethz.ch/research/flagship-projects/NovelProteins.html (2022). ​ 64. Bundesamt für Lebensmittelsicherheit und Veterinärwesen (BLV). Gemein1 schaftsgastronomie. https://www.blv.admin.ch/blv/de/home/lebensmittel-undernaehrung/ernaehrung/gemeinschaftsgastronomie.html (2022). ​ 65. Garnett, E. E., Balmford, A., Sandbrook, C., Pilling, M. A. & Marteau, T. 1 M. Impact of increasing vegetarian availability on meal selection and sales in cafeterias. Proc Natl Acad Sci 116, 20923–20929 (2019). ​ 66. Bucher, D. et al. Exploiting Fitness Apps for Sustainable Mobility - Chal1 lenges Deploying the GoEco! App. Proceedings of ICT for Sustainability, 89–98 (2016). ​ 93. Bundesamt für Statistik (BFS). Luftemissionskonten der Haushalte und 1 der Wirtschaft, nach Wirtschaftssektoren. https://www.bfs.admin.ch/bfs/de/ home/statistiken/raum-umwelt/umweltgesamtrechnung/luftemissionen.assetdetail.23464616.html (2022). ​ 68. Walter, A., Finger, R., Huber, R. & Buchmann, N. Smart farming is key to 1 developing sustainable agriculture. Proc Natl Acad Sci 114, 6148–6150 (2017). ​ 94. Eidgenössisches Departement des Innern (EDI). Informationsschreiben 1 2021/9.11: Abgabe von Lebensmitteln nach Ablauf des Mindesthaltbarkeitsdatums (MHD). https://www.blv.admin.ch/dam/blv/de/dokumente/lebensmittel-und-ernaehrung/rechts-und-vollzugsgrundlagen/hilfsmittel-vollzugsgrundlagen/informationsschreiben-neu/informationsschreiben-2021-9-mhd.pdf. download.pdf/d_Informationsschreiben%202021_9%20Abgabe%20von%20 Lebensmittel%20nach%20Ablauf%20des%20Mindesthaltbarkeitsdatums%20 (MHD).pdf (2022). ​ 69. Späti, K., Huber, R., Logar, I. & Finger, R. Incentivizing the adoption of 1 precision agricultural technologies in small-scaled farming systems: A choice experiment approach. Journal of the Agricultural and Applied Economics Association 1, 236–253 (2022). ​195. Deloitte. Nachhaltige Lebensmittel: Was Schweizer Konsumenten von Unternehmen und vom Staat erwarten. https://www2.deloitte.com/content/dam/ Deloitte/ch/Documents/public-sector/deloitte-ch-de-sustainable%20food-final.pdf (2021). ​ 70. Späti, K., Huber, R. & Finger, R. Benefits of Increasing Information Accu1 racy in Variable Rate Technologies. Ecological Economics 185, 107047 (2021). ​ 96. Hartmann, C., Hieke, S., Taper, C. & Siegrist, M. European consumer 1 healthiness evaluation of ‘Free-from’ labelled food products. Food Qual Prefer 68, 377–388 (2018). ​ 67. Finger, R., Swinton, S. M., el Benni, N. & Walter, A. Precision Farming at 1 the Nexus of Agricultural Production and the Environment. Annual Review of Resource Economics 11, 313–335 (2019). ​ 71. Swiss Food & Nutrition Valley. About - Swiss Food & Nutrition Valley. 1 https://swissfoodnutritionvalley.com/about/ (2022). ​ 72. Schmidt, H.-P., Hagemann, N., Abächerli, F., Leifeld, J. & Bucheli, T. Pflan1 zenkohle in der Landwirtschaft. Agroscope Science 112, 1–71 (2021). ​ 73. Altieri, M. A., Nicholls, C. I., Henao, A. & Lana, M. A. Agroecology and 1 the design of climate change-resilient farming systems. Agron Sustain Dev 35, 869–890 (2015). ​ 74. Woods, J., Williams, A., Hughes, J. K., Black, M. & Murphy, R. Energy and 1 the food system. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 2991–3006 (2010). ​ 75. Cunningham, S. A. et al. To close the yield-gap while saving biodiversi1 ty will require multiple locally relevant strategies. Agric Ecosyst Environ 173, 20–27 (2013). ​ 76. Fesenfeld, L. P. Governing urban food systems in the long run: Comparing 1 best practices in sustainable food procurement regulations. GAIA - Ecological Perspectives for Science and Society 25, 260–270 (2016). ​ 77. Moschitz, H., Scharrer, B. & Bigler, G. Diskussionspapier: Stadt und Land1 wirtschaft im Dialog. Schweizerische Akademische Gesellschaft für Umweltforschung und Ökologie (saguf) Diskussionspapier https://orgprints.org/id/ eprint/36580/1/saguf_Diskussionspapier_2019.pdf (2019). ​ 78. Scharrer, B. ‘Dem Sempachersee kommt die Gülle hoch’ _ Das Span1 nungsfeld zwischen intensiver Tierhaltung und Gewässerschutz im Kanton Luzern 1976 - 2003. Berner Forschungen zur Neuesten Allgemeinen und Schweizer Geschichte 12 (2013). ​ 79. Scharrer, B. Fördert die Schweizer Agrarpolitik die Entwicklung einer nach1 haltigen und standortgerechten Landwirtschaft ausreichend? Eine Analyse der aktuellen Instrumente, Zielkonflikte und des Handlungsbedarfs (Working Paper NRP73 NRP73-WPO4-2022). doi:10.48350/172674 (2022). ​ 80. Europäische Kommission. BEKANNTMACHUNG DER KOMMISSION: EU1 Leitlinien für Lebensmittelspenden https://eur-lex.europa.eu/legal-content/DE/ TXT/PDF/?uri=CELEX:52017XC1025(01)&from=PT (2017). ​ 81. Faccioli, M. et al. Combined carbon and health taxes outperform sin1 gle-purpose information or fiscal measures in designing sustainable food policies. Nat Food 3, 331–340 (2022). ​ 82. Springmann, M. et al. Mitigation potential and global health impacts from 1 emissions pricing of food commodities. Nat Clim Chang 7, 69–74 (2016). ​ 97. Lazzarini, G. A., Visschers, V. H. M. & Siegrist, M. How to improve con1 sumers’ environmental sustainability judgements of foods. J Clean Prod 198, 564–574 (2018). 198. Ministère de la Santé et Prévention. Etudes et rapports scientifiques. ​ https://solidarites-sante.gouv.fr/prevention-en-sante/preserver-sa-sante/nutrition/nutri-score/etudes-et-rapports-scientifiques/ (2022). ​ 99. Engels, S. V., Hansmann, R. & Scholz, R. W. Toward a sustainability label 1 for food products: An analysis of experts’ and consumers’ acceptance. Ecol Food Nutr 49, 30–60 (2010). ​ 00. Bundesamt für Lebensmittelsicherheit und Veterinärwesen (BLV). Nutri2 Score. https://www.blv.admin.ch/nutri-score (2019). ​ 01. Fesenfeld, L. & Maier, M. How climate- and sustainability-related food la2 bels change individuals’ attitudes, knowledge, norms, and behaviors: A field and survey experiment. AEA RCT Registry (2022) doi:10.1257/rct.9331-1.0. ​ 02. Hélias, A. et al. Implementing environmental labelling of food products in 2 France. Int J Life Cycle Assess 27, 926–931 (2022). ​ 03. Uebersax, A., Jenni, S., Koch, B., Richner, W. & Huguenin-Elie, O. Fach2 liche Überprüfung der im Gewässerschutzgesetz festgelegten Limitierung der Düngerausbringung pro Hektare Nutzfläche - PDF Kostenfreier Download. https://docplayer.org/115588593-Fachliche-ueberpruefung-der-im-gewaesserschutzgesetz-festgelegten-limitierung-der-duengerausbringung-pro-hektare-nutzflaeche.html (2017). ​ 04. European Coalition for Corporate Justice. Corporate due diligence laws 2 and legislative proposals in Europe. https://corporatejustice.org/wp-content/ uploads/2022/03/Corporate-due-diligence-laws-and-legislative-proposals-in-Europe-March-2022.pdf (2022). ​ 05. United Nations (UN). Guiding Principles on Business and Human Rights: 2 Implementing the United Nations “Protect, Respect and Remedy” Framework. https://www.ohchr.org/sites/default/files/documents/publications/guidingprinciplesbusinesshr_en.pdf (2021). ​ 06. Bundesministerium für Arbeit und Soziales (BMAS). Gesetz über die unter2 nehmerischen Sorgfaltspflichten in Lieferketten. https://wirtschaft-entwicklung. de/fileadmin/user_upload/5_Wirtschaft_und_Menschenrechte/Downloads/Lieferkettensorgfaltspflichtengesetz.pdf (2021). ​ 83. Funke, F. et al. Toward Optimal Meat Pricing: Is It Time to Tax Meat Con1 sumption? Rev Environ Econ Policy 16, 219-240 (2022). ​ 07. European Commission. DIRECTIVE OF THE EUROPEAN PARLIAMENT 2 AND OF THE COUNCIL on Corporate Sustainability Due Diligence and amending Directive (EU) 2019/1937. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52022PC0071 (OECD, 2022). ​ 84. Caillavet, F., de Fadhuile, A. & Nichèle, V. Assessing the distributional ef1 fects of carbon taxes on food: Inequalities and nutritional insights in France. Ecological Economics 163, 20–31 (2019). ​ 08. Möhring, A., Mack, G., Zimmermann, A., Mann, S. & Ferjani, A. Versor2 gungssicherheitsbeiträge: Mittel effizienter einsetzen. Agrarforsch Schweiz 9, 348–355 (2018). ​ 85. Moberg, E., Walker Andersson, M., Säll, S., Hansson, P. A. & Röös, E. 1 Determining the climate impact of food for use in a climate tax—design of a consistent and transparent model. International Journal of Life Cycle Assessment 24, 1715–1728 (2019). ​ 09. Mann, S. Implicit Social Policy in Agriculture. Social Policy and Society 4, 2 271–281 (2005). ​ 86. European Commission. The Environmental Footprint Pilots. https://ec.eu1 ropa.eu/environment/eussd/smgp/ef_pilots.htm (2022). ​ 11. Mann, S., Zaharia, I. & Reissig, L. Social Policy and Burnout: A Bi-national 2 Comparison. Int J Health Wellness Soc 9, 63–75 (2019). ​ 87. Lansche, J., Koch, P., Mouron, P. & Gaillard, G. Methodische Herausfor1 derungen bei der Umweltproduktdeklaration von Landwirtschaftsprodukten. Agrarforsch Schweiz 7, 144–149 (2016). ​12. Bosshard, C., Spiess, E. & Richner, W. Überprüfung der Methode 2 Suisse-Bilanz: Schlussbericht. https://ub.unibas.ch/digi/a125/sachdok/2014/ BAU_1_6291392.pdf (2012). ​ 88. Gligen, A. et al. Indikatorbasierte Direktzahlungen im Agrarumweltbereich: 1 Schlussbericht ans Bundesamt für Landwirtschaft. Agroscope Science 136, 1-101 (2022). ​ 13. Posso, A. Child Labour’s effect on long-run earnings: An analysis of co2 horts. Econ Model 64, 465–472 (2017). ​ 89. Schmidt, A. et al. Direct and Indirect Economic Incentives to Mitigate Ni1 trogen Surpluses: A Sensitivity Analysis. Journal of Artificial Societies and Social Simulation 20, 7 (2017). ​ 90. Paul, S., Ammann, C., Alewell, C. & Leifeld, J. Carbon budget response of 1 an agriculturally used fen to different soil moisture conditions. Agric For Meteorol 300, 108319 (2021). ​ 91. Moberg, E., Säll, S., Hansson, P. A. & Röös, E. Taxing food consumption to 1 reduce environmental impacts – Identification of synergies and goal conflicts. Food Policy 101, 102090 (2021). ​192. Agroscope. Inbetriebnahme einer Agro-Photovoltaik-Versuchsanlage bei Agroscope Conthey. https://www.agroscope.admin.ch/agroscope/de/home/ aktuell/newsroom/2021/10-07_photovoltaik-versuchsanlage.html (2021). ​ 10. Tondani, D. Universal Basic Income and Negative Income Tax: Two differ2 ent ways of thinking redistribution. J Socio Econ 38, 246–255 (2009). ​ 14. Baumgartner, U. & Bürgi Bonanomi, E. Drawing the line between sustain2 able and unsustainable fish: product differentiation that supports sustainable development through trade measures. Environ Sci Eur 33, 1–13 (2021). ​ 15. Bundesamt für Landwirtschaft (BLW). Bedeutung von Verkaufsaktionen im 2 Schweizer Fleischmarkt. https://www.blw.admin.ch/dam/blw/de/dokumente/ Markt/Marktbeobachtung/Fleisch/Marktberichte/mbf_2021_11.pdf.download. pdf/mbf_2021_11_d.pdf (2021). ​ 16. Schweizerischer Bundesrat. Aktionsplan 2021-2023 zur Strategie Nach2 haltige Entwicklung 2030. https://www.are.admin.ch/dam/are/de/dokumente/ nachhaltige_entwicklung/publikationen/aktionsplan2021-2023.pdf.download. pdf/Aktionsplan%202021-2023%20zur%20Strategie%20Nachhaltige%20Entwicklung%202030.pdf ​ 73 217. Bundesamt für Landwirtschaft (BLW). Strategie Nachhaltige Entwicklung 2030. Nationaler Dialog der Schweiz für den UNO Ernährungssystemgipfel 2021: Von Herausforderungen zu Engagements (2021). https://www.blw.admin. ch/dam/blw/de/dokumente/International/Nachhaltigkeit/2030%20Agenda%20 f%C3%BCr%20Nachhaltige%20Entwicklung/fssd_workshop_bericht_3-5-21. pdf.download.pdf/Nationaler_FSSD_Workshop_1_Bericht_DE.pdf INDEX DES ILLUSTRATIONS Figures ​ 19. One Planet. National and Sub-national Food Systems Multi-Stakeholder 2 Mechanisms: An Assessment of Experiences. https://spaces.oneplanetnetwork.org/system/files/strategy_one_planet.pdf (2021). p. 11, fig. 1 (Système d’alimentation) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 ​ 20. Generalsekretariat Eidgenössisches Departement für Umwelt (UVEK). Run2 der Tisch Wasserkraft mit gemeinsamer Erklärung abgeschlossen. https://www. admin.ch/gov/de/start/dokumentation/medienmitteilungen.msg-id-86432.html (2021). p. 12, fig. 2 (SDG Wheel) United Nations (UN). Transforming our World: The 2030 Agenda for Sustainable Development. https://sdgs.un.org/publications/transforming-our-world-2030-agendasustainable-development-17981 (2015). ​ 21. Karanth, A. & Archer, D. Institutionalising mechanisms for building urban 2 climate resilience: Experiences from India. Dev Pract 24, 514–526 (2014). p. 16, fig. 3 (Points de basculement) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 ​ 18. Federal Department of Economic Affairs, Education and Research (EAER) 2 & Federal office for Agriculture, International Affairs and Food Security (FOAG). National Pathway for Food Systems Transformation in Support of the 2030 Agenda - Switzerland. https://summitdialogues.org/wp-content/uploads/2021/09/ National-Pathway-for-Food-Systems-Transformation-in-Support-of-the-2030Agenda-Switzerland_cleared-for-distribution.pdf (2021). ​22. Herens, M., Pittore, K. & Oosterveer, P. Transforming food systems: 2 Multi-stakeholder platforms driven by consumer concerns and public demands. Glob Food Sec 32, 100592 (2022). ​ 23. Kochskämper, E., Challies, E., Newig, J. & Jager, N. W. Participation for 2 effective environmental governance? Evidence from Water Framework Directive implementation in Germany, Spain and the United Kingdom. J Environ Manage 181, 737–748 (2016). ​ 24. Tamara, A. R. et al. Trust-building and leadership in multi-stakeholder fo2 rums: lessons from Indonesia. International Forestry Review 23, 43–58 (2021). ​ 25. Chatham House. Chatham House Rule. https://www.chathamhouse.org/ 2 about-us/chatham-house-rule (2022). ​ 26. Deconinck, K., Giner, C., Jackson, L. A. & Toyama, L. Making better poli2 cies for food systems will require reducing evidence gaps. Glob Food Sec 33, 100621 (2022). ​ 27. Meyer-Ohlendorf, N. Ultimative Bürgerbeteiligung? https://www.ipg-jour2 nal.de/rubriken/demokratie-und-gesellschaft/artikel/ultimative-buergerbeteiligung-5997/ (2022). ​ 28. Goldberg, S. & Bächtiger, A. Catching the ‘Deliberative Wave’? How (Dis2 affected) Citizens Assess Deliberative Citizen Forums. Br J Polit Sci, 1–9 (2022). ​ 29. Food and Agriculture Organization of the United Nations (FAO). Trans2 forming food and agriculture to achieve the SDGs. 20 interconnected actions to guide decision-makers - Revised edition. https://www.fao.org/3/ca1612en/ ca1612en.pdf (2019). ​ 30. Bundesministerium für Ernährung und Landwirtschaft (BMEL). Zukunfts2 kommission Landwirtschaft. https://www.bmel.de/DE/themen/landwirtschaft/ zukunftskommission-landwirtschaft.html (2022). ​ 31. Bundesministerium für Ernährung und Landwirtschaft (BMEL). Umbau der 2 Nutztierhaltung in Deutschland – der Borchert-Prozess. https://www.bmel.de/ DE/themen/tiere/nutztiere/umbau-nutztierhaltung.html (2022). ​ 32. Food Nation. Organic: A vision and a mindset in the Danish food cluster. 2 https://csr.dk/sites/default/files/Food%20Nation%20white%20paper%20Organic_Web_PlanA.pdf (2019). ​33. Future Policy. Denmark’s Organic Action Plan. https://www.futurepoli2 cy.org/healthy-ecosystems/denmarks-organic-action-plan-working-together-for-more-organics/ (2022). ​ 34. Knoke, D., Pappi, F. U., Broadbent, J. & Tsujinaka, Y. Comparing Policy 2 Networks: Labor Politics in the U.S., Germany, and Japan. (Cambridge University Press, 1996). ​ 35. McGinnis, M. D. & Ostrom, E. Social-ecological system framework: initial 2 changes and continuing challenges. Ecology and Society 19, 30 (2014). ​ 36. Brundtland, G. H. Our Common Future—Call for Action. Environ Conserv 14, 2 291–294 (1987). ​ 37. Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., 2 Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/ zenodo.7543576 p. 19, fig. 4 (Wedding Cake Model) Rockström, J. & Sukhdev, P. How food connects all the SDGs. Opening keynote speech at the 2016 EAT Forum. https://www.youtube.com/watch?time_continue=704&v=tah8QlhQLeQ&feature=emb_title (2016). p. 21, fig. 5 (Modèle pour un système alimentaire durable) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 p. 42, fig. 6 (Une voie d’action pour la transformation du système alimentaire suisse) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 p. 62, fig. 7 (Proposition gouvernance du système alimentaire suisse) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 Graphiques p. 13, graphique 1 (Surpoids et obésité) Bundesamt für Statistik (BFS). Schweizerische Gesundheitsbefragung 2017: Übergewicht und Adipositas. https://www.bfs.admin.ch/bfs/de/home/aktuell/ neue-veroeffentlichungen.assetdetail.14147705.html (2020). p. 14, graphique 2 (Nitrates dans les eaux souterraines) Bundesamt für Umwelt (BAFU). Indikator Wasser. https://www.bafu.admin.ch/bafu/ de/home/themen/thema-wasser/wasser--daten--indikatoren-undkarten/wasser-indikatoren/indikator-wasser.pt.html/aHR0cHM6Ly93d-3cuaW5kaWthdG9yZW4u YWRtaW4uY2gvUHVibG/ljL0FlbURldGFpbD9pbmQ9V1MwNDgmbG5nPWR lJlN1Ymo9Tg% 3d%3d.html (2022). p. 14, graphique 3 (Empreinte gaz à effet de serre) Bundesamt für Umwelt (BAFU). Umwelt-Fussabdrücke der Schweiz: Entwicklung zwischen 2000 und 2018. https://www.bafu.admin.ch/dam/bafu/de/dokumente/wirtschaft-konsum/externestudien-berichte/umwelt-fussabdruecke-der-schweiz-entwicklung-zwischen-2000-2018. pdf.download.pdf/EBP_(2022)_Umweltfussabdruecke.pdf (2022). p. 34, graphique 4 (Objectifs pour une alimentation adaptée d’ici 2030) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; 74 Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 p. 35, graphique 5 (Objectifs de réduction d’ici 2030) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz –https://doi.org/10.5281/zenodo.7543576 Tableaux p. 23, tableau 1 (Alimentation adaptée) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 p. 30, tableau 2 (Émissions de gaz à effet de serre) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 p. 43 – 45, tableau 3 (Paquets de mesures) Fesenfeld, L.; Mann, S.; Meier, M; Nemecek, T.; Scharrer, B.; Bornemann, B., Brombach, C.; Beretta, C.; Bürgi, E.; Grabs, J.; Ingold, K.; Jeanneret, P.; Kislig, S.; Lieberherr, E.; Müller, A.; Pfister, S.; Schader, C.; Schönberg, S.; Sonnevelt, M.; Barjolle, D.; Boivin, P.; Brunner, T.; Contzen, S.; Espa, I.; Estève, M.; Forney, J.; Häberli C.; Hediger, W.; Hilbeck, A.; Kopainsky, B.; Lehmann, B.; Mack, G.; Markoni, E.; Meier, B.; Paccaud, F.; Rohrmann, S.; Schindler, M.; Schwab, C.; Tribaldos, T.; Waibel, P.; Zähringer, J. (2023). Wege in die Ernährungszukunft der Schweiz: Leitfaden zu den grössten Hebeln und politischen Pfaden für ein nachhaltiges Ernährungssystem. SDSN Schweiz – https://doi.org/10.5281/zenodo.7543576 Photos Image de couverture : KEYSTONE/Gian Ehrenzeller p. 2, S. 15, S. 67 : istockphoto – minemero/ Daniel Balakov/ makasana p. 6, S. 27, S. 37 : unsplash – Pille R. Priske/ Gabriel Jimenez/ Markus Spiske p. 57 : Lisa Meienberg Illustrations Aline Telek, www.telek.ch MENTIONS LÉGALES Création graphique Aline Telek Traduction Comtexto ; Vanja Guérin / GTGE Impression Vögeli AG (Cradle-to-Cradle certified Silver) © Février 2023, SDSN Suisse 75 www.sdsn.ch L’AVENIR DE L’ALIMENTATION VOUS INTÉRESSE? Scannez le code QR pour obtenir la version numérique du guide. 76
W3196399567.txt
https://link.springer.com/content/pdf/10.1007/s11067-021-09551-8.pdf
en
Leontief Meets Markov: Sectoral Vulnerabilities Through Circular Connectivity
Networks and spatial economics
2,021
cc-by
12,536
Networks and Spatial Economics (2022)22:659–690 https://doi.org/10.1007/s11067-021-09551-8 Leontief Meets Markov: Sectoral Vulnerabilities Through Circular Connectivity Ariel L. Wirkierman1 · Monica Bianchi2 · Anna Torriero2 Accepted: 7 July 2021 / Published online: 30 August 2021 © The Author(s) 2021 Abstract Economists have been aware of the mapping between an Input-Output (I-O, hereinafter) table and the adjacency matrix of a weighted digraph for several decades (Solow, Econometrica 20(1):29–46, 1952). An I-O table may be interpreted as a network in which edges measure money flows to purchase inputs that go into production, whilst vertices represent economic industries. However, only recently the language and concepts of complex networks (Newman 2010) have been more intensively applied to the study of interindustry relations (McNerney et al. Physica A Stat Mech Appl, 392(24):6427–6441, 2013). The aim of this paper is to study sectoral vulnerabilities in I-O networks, by connecting the formal structure of a closed I-O model (Leontief, Rev Econ Stat, 19(3):109–132, 1937) to the constituent elements of an ergodic, regular Markov chain (Kemeny and Snell 1976) and its chance process specification as a random walk on a graph. We provide an economic interpretation to a local, sector-specific vulnerability index based on mean first passage times, computed by means of the Moore-Penrose inverse of the asymmetric graph Laplacian (Boley et al. Linear Algebra Appl, 435(2):224–242, 2011). Traversing from the most central to the most peripheral sector of the economy in 60 countries between 2005 and 2015, we uncover cross-country salient roles for certain industries, pervasive features of structural change and (dis)similarities between national economies, in terms of their sectoral vulnerabilities. Keywords Input-output analysis · Markov chains · Betweenness centrality 1 Introduction Wassily W. Leontief (1905–1999) and Andrey A. Markov (1856–1922) probably never met. Probably, because Leontief’s first year as a student at Saint Petersburg  Ariel L. Wirkierman a.wirkierman@gold.ac.uk Extended author information available on the last page of the article. 660 A.L. Wirkierman et al. State University (SPbU) in 1921 coincided with Markov’s last year of teaching activities affiliated to that same institution, just before his death. And while their paths may not have crossed, intertwining their contributions reveals foundational insights to study the connectivity in networked structures of socio-economic relations.1 Leontief’s Input-Output (I-O, hereinafter) method (Leontief 1937) became the cornerstone for analysing the general interdependence and circularity between sectors of an economy (Leontief 1986; Miller and Blair 2009). An I-O table is the matrix representation of the bilateral flows of commodities in terms of monetary units between industries. It may be interpreted as the adjacency matrix of a network whose edges measure input flows whilst vertices represent economic sectors (Olsen 1992). Self-loops (i.e. payments of an industry to itself) are allowed, and in addition to intermediate input flows (i.e. a transaction between two industries) each sector sells its output to final demand (i.e. final consumption, investment and exports) and faces primary cost elements (i.e. wages, taxes, imports), obtaining a gross operating surplus as a residual (UN 2009). While analyses of the network structure of I-O tables may be found in recent literature (e.g. Xu et al. 2011; McNerney et al. 2013; Tsekeris Tsekeris), the graphtheoretic representation of an I-O system dates back to the work of Solow (1952) and contributions to Morgenstern (1954). More recently, explicit connections between centrality indicators in the emerging field of complex networks (such as Google’s PageRank algorithm) and Leontief’s I-O model(s) have been drawn (Franceschet 2011). The extent to which primary incomes and final expenditures are connected distinguishes a closed from an open I-O model. In the former, circularity is complete, to the point that every element of primary costs is mapped into a component of final expenditure, and vice-versa. In the latter model, final expenditure decisions are autonomous, so that primary costs represent a leakage from money flow circularity. The analytical device of a Markov chain provides a chance process interpretation of the emerging connectivity between nodes in a network (Grinstead and Snell 1997). Superposing such an interpretation to an input-output structure has been thoroughly worked out by Kemeny and Snell (1976, p. 200), mapping an open I-O system into an absorbing Markov chain, i.e. a chain in which intermediate industries are transient states and a primary income component represents the absorbing state. This mapping has been explored in different directions (e.g. Duchin and Levine 2010; Moosavi and Isacchini 2017; Xing et al. 2017; Xing et al. 2018; Kostoska et al. 2020). In particular, Blöchl et al. (2011) devise two centrality measures based on random walks to interpret the propagation of supply shocks through the economy, i.e. shocks that depart from exogenous changes to primary cost components. Differently from Blöchl et al. (2011), to study betweenness centrality and node vulnerabilities in I-O networks, this paper maps a closed I-O system into an ergodic, regular Markov chain (Kemeny and Snell 1976, p. 37). Such a mapping has been recently used to study the community structure of a global network of inter-industry 1 By connectivity, we mean the number or strength of (independent) paths between any two nodes of a network (Newman 2010, p. 147). Leontief meets Markov... 661 flows with an aggregated, unique final sector (Piccardi et al. 2018). Instead, a key contribution of this paper is to extend the specification of the Markov chain to three different final sectors: (i) the foreign sector, i.e. imports and exports; (ii) the households and government sector and (iii) the profits and investment sector. This has important theoretical implications for the logic of economic circularity, which we explore below. Departing from the OECD Input-Output Tables (OECD-IOTs) database, we derive a time-series of non-negative, irreducible and row stochastic empirical transition probability matrices, each associated to a regular Markov chain representing the income side of a closed I-O system. Interpreting the traversal across sectors of the economy as a random walk on an I-O graph, we use the Moore-Penrose inverse of the associated asymmetric graph Laplacian to quantify the betweenness centrality of each sector in the economy (Boley et al. 2011; Ranjan and Zhang 2013). Such a notion of topological centrality captures local, node-level vulnerabilities. In particular, we consider vulnerability to be defined as “the system’s susceptibility to negative [random and/or targeted] shocks” (Cardinale 2019, p. 6), and operationalised — at a local level — by a ‘vulnerability gap’ (Adger 2006, Table A1, p. 279) between constrained and unconstrained network flows. Conceptually, betweenness centrality conveys the extent to which a node has the potential for exerting control over the communication between any two other vertices (Freeman 1979). By the same token, a high value of such potential leaves a node in a “state of susceptibility to harm from exposure to stresses” (Adger 2006, p. 268). In fact, in line with previous studies, “for sequential attack the networks are most effectively degraded by removing vertices in decreasing order of betweenness centrality” (Iyer et al. 2013, p. 16, italics added). We apply this framework to devise a ranking from the most central to the most peripheral sector of the economy for 60 countries between 2005 and 2015, uncovering pervasive features of structural change and (dis)similarities between national economies, in terms of their sectoral vulnerabilities. After this brief introduction, the rest of the paper is organised as follows. Section 2 recalls preliminary concepts and specifies the mapping of a closed Input-Output system into a regular Markov chain, obtaining a betweenness centrality indicator that quantifies local vulnerabilities. Section 3 describes the path that goes from interindustry data to empirical probability transition matrices, develops an empirical strategy to study cross-country sectoral vulnerabilities, reporting the results obtained. Finally, Section 4 presents a summary of findings and some concluding remarks. 2 Leontief Meets Markov: Methods 2.1 Graphs, Markov Chains and Random Walks We quickly recall some standard definitions and results about graph theory and random walks on graphs; for more details the reader is referred to Wilson (1972), Grinstead and Snell (1997), and Kemeny and Snell (1976). 662 A.L. Wirkierman et al. A graph G = (V , E) is a pair of sets (V , E), where V is the set of n nodes (or vertices) and E is the set of edges (or arcs), consisting of m pairs of nodes of V . A directed graph (or digraph) is a graph in which each edge (arc) is an ordered pair (i, j ) of vertices. Moreover, a weight wij may be associated to each edge (i, j ), having a weighted (or valued) graph. A non-negative n-square matrix W = [wij ], representing the adjacency relationships between vertices of G, can be associated to the graph (the adjacency matrix); the off-diagonal elements wij state the strength of the relationship from node i to node j , whereas diagonal elements wii represent the strength of the self-loop of node i. The adjacency matrix of a weighted digraph is, in general, asymmetric. (in) In a directed, weighted graph, the in-degree di of node i is the total strength (out) of arcs directed from other nodes to i and the out-degree di of node i is the total strength of the arcs directed from i to other vertices. If eT = [1, . . . , 1] is a (out) (in) sum vector, then d (out) = W e and d T(in) = eT W . In general, di = di , but eT d (out) = d T(in) e. Throughout the paper we deal with the general case of directed, weighted graphs admitting self-loops and we focus on out-degrees, noting that all the results can be carried out also for in-degrees, by taking the transpose of the adjacency matrix. Let us also assume that every node has at least one out-going edge which can (out) include self-loops, i.e di = 0 for every i. In this case, the matrix D =  d (out) (containing out-degrees in its main diagonal, and zero elsewhere) is non singular (out) and we can define the matrix P = D −1 W . Each element of P , pij = wij /di represents the relative strength exerted from i to j (in proportion to the total strength exerted by node i to others). While matrix P provides us with a structural representation of connectivity for graph G, it is possible to superpose a chance process interpretation of emerging connectivity patterns through the device of a finite Markov chain (Grinstead and Snell 1997,p. 405). Each non-negative element pij may be interpreted as the probability of transitioning from node i (row i in P ) to node j (column j in P ) in the upcoming iteration of the chance process. This process of traversing can be interpreted as a random walk: a sequence of vertices generated from a source node i selecting an edge, traversing the edge to a target vertex j (according to the transition rules codified in P ), and iteratively repeating the transition to another (or the same) node, starting from its immediately previous point of arrival. Thus, matrix P is the transition probability matrix of the Markov chain associated to a random walk on graph G, and it is row-stochastic, i.e. P e = e. As we iterate step-wise over this chance process, the probabilities of being on each node as the process unfolds (say, from t = 0 to t = 1) are given by pT(1) = p T(0) P , where p is a probability vector. This iteration process continues (p T(2) = pT(1) P = p T(0) P 2 ) until we (may) reach a fixed point: π T = π T P , where π specifies the vector of stationary probabilities. If the graph G is strongly connected, i.e. for any pair of vertices there is a directed path leading from one vertex to the other, matrix P is irreducible (P n has only positive entries for some n). In this case, the associated Markov chain is said to be regular 663 Leontief meets Markov... (Kemeny and Snell 1976), being possible to be in any state after n steps, no matter what the starting state is. Moreover, by the Perron-Frobenius theorem (Meyer 2000, p. 693), there exists a unique positive vector of stationary probabilities π = [πi ], describing the long-run proportion of (iteration) periods that a random walker will spend in each node. But while πi provides us with an idea of the relative systemic importance of node i, many interesting structural features of connectivity concern relational magnitudes. In particular, (i) what is the average number of steps required to reach node j for the first time, starting from node i and, (ii) what constraint imposes on (i) having to necessarily pass by node k in traversing from i to j . To obtain (i), it is possible to compute the mean first passage time, i.e. the expected number of transitions needed by a random walker starting in source node i to reach target node j for the first time, denoted by H (i, j ). By convention H (i, i) = 0, ∀i while for i = j , H (i, j ) is usually obtained using an ‘absorbing-chain technique’ (for example, see Blöchl et al. 2011).2 In particular, if P (−j ) is the (n − 1) × (n − 1) matrix obtained from deleting the j -th row and column of P , we have:   H (i, j ) = (I − P (−j ) )−1 e(−j ) i where e(−j ) is an (n − 1) sum vector. Note that, computationally, this approach requires to invert n − 1 matrices, each of dimension (n − 1) × (n − 1). Instead, a key departure of this paper consists in an alternative computation of H (i, j ) based on obtaining the Moore-Penrose inverse of (only) one matrix of dimension n × n.3 To see this, besides the transition probability matrix P , it is possible to define the (ordinary) asymmetric Laplacian matrix L = Π(I − P ), where Π =  π, and which satisfies Le = LT e = 0. Then, if we denote by M = [mij ] the MoorePenrose inverse of L, the mean first passage time from source i to target j is given by (Boley et al. 2011, Theorem 15): H (i, j ) = mjj n  − mij + (mil − mj l )πl (1) l=1 In order to quantify (in number of steps) “how much the restriction of passing through a given vertex [k] represents a detour in going from an arbitrary vertex i to another arbitrary vertex j ” (Boley et al. 2011, p. 236), we combine H (i, k) and H (k, j ): Hk (i, j ) = H (i, k) + H (k, j ) 2 The technique consists in applying the same principles used for absorbing (rather than ergodic) Markov chains to an irreducible chain that does not contain absorbing states (see Grinstead and Snell (1997) and Palacios (1990) and the references therein). 3 Details on how to compute the Moore-Penrose inverse of a matrix are provided in Meyer (2000, p. 423). 664 A.L. Wirkierman et al. i.e. we need to go from i to k and, then, from k to j . By applying Eq. 1, summing over all pairs of source and target nodes and rearranging, we obtain: n  n  Hk (i, j ) − i=1 j =1 (a) n  n  H (i, j ) = n2 mkk (2) i=1 j =1 (b) where term (a) in Eq. 2 is the average length of all walks between any pair of nodes restricted to passing by vertex k, while term (b) in Eq. 2 is the average length of all walks between any pair of nodes. The difference (a) - (b) represents the “extra distance travelled between two vertices when forced to pass through vertex [k], summed over all n2 pairs of source/destination vertices” (Boley et al. 2011, p. 237). Note that (a) - (b) in Eq. 2 represents the difference between constrained — of passing by k — and unconstrained network flows across node pairs, respectively. This difference may be interpreted as a ‘vulnerability gap’ (in the sense of Adger 2006, Table A1, p. 279), where the constrained magnitude captures a node-specific average traversal distance, whilst the unconstrained term captures a ‘threshold’ average distance between any two nodes.4 Given that, throughout this work, we consider graphs with the same number of nodes, (the reciprocal of) mkk in Eq. 2 captures the betweenness centrality of node k. In fact, the higher mkk the higher the extra distance imposed by forcing the passage through that node. Thus, a very high value of mkk indicates that k must be a peripheral node of the graph, otherwise the detour would not imply such a large number of steps. In contrast, the lower mkk the lower the required extra distance, so k must be a central node, implying that i frequently passes by k in order to reach j . Hence, the lower mkk , the lower the extra distance imposed and the higher betweenness centrality of vertex k. Moreover, note that mkk for k = 1, . . . , n can be straightforwardly obtained by extracting the main diagonal of the Moore-Penrose inverse matrix M. More importantly, this notion of betweenness centrality captures local, node-level vulnerabilities for graph G. Vertices with (relatively) low values of mkk reveal those nodes which are crucial for the connectedness of the system. If they experience a negative shock, it may be difficult for the system to find alternative, back-up paths to reach other parts of the network. This leads to a key conceptual distinction: the relative systemic importance of node k (measured by its stationary probability πk ) may not be particularly high, but if mkk is instead relatively low, a shock to this vertex may severely disrupt the system. An illustration of this distinction for a directed graph G of n = 7 nodes is reported below.5 The network structure in Fig. 1 implies that nodes 1, 2, 3 must necessarily pass by node 4 to reach nodes 5, 6, 7 and vice-versa. Removing node 4 from the graph would imply an unsolvable disruption to the network flow. However, all nodes have 4 With respect to the well-being vulnerability gap in Adger (2006), our formulation has the signs reversed, given that a higher distance would correspond to a lower well-being and vice-versa. 5 Please note that this is a directed but unweighted graph, to render the example more immediate to grasp. Leontief meets Markov... 665 Fig. 1 Graph G and associated adjacency matrix W , transition probability matrix P , vector of stationary probabilities π and vector [mkk ], with the main diagonal of Moore-Penrose inverse M of the asymmetric Laplacian L = Π(I − P ). Nodes in matrices and vectors go from 1 to 7, nodes in graph are numbered counter-clockwise, starting from node 1 (labelled) the same stationary probability, π = 0.143, suggesting that they all share the same systemic importance. Instead, the values of vector [mkk ] suggest a clear hierarchy of vulnerabilities. For node k = 4, mkk = 7.143, implying it is the most central node (as expected), whereas vertices 1,2,6 and 7 are the peripheral ones (with mkk = 16.143). As the example illustrates, whilst conveying different structural properties of network nodes, eigenvector (πk ) and betweenness centrality (mkk ) may be shown to be formally related, as well as connected to yet a third notion of node importance, i.e. closeness centrality. Appendix A relates the three centrality concepts by exploring their formulation in terms of mean first passage times using the elements of the Moore-Penrose inverse. By being formulated within the context of a random walk on graph G, the notion of betweenness centrality captured by mkk considers all — rather than only the shortest — network paths, when measuring the influence of a node (Newman 2005). Moreover, as Appendix A shows, mkk incorporates the systemic importance (πk ) of all nodes, weighted by the corresponding row k of the Moore-Penrose inverse. Conceptually, eigenvector centrality reflects the comprehensive effect of a node on other vertices. Closeness centrality indicates how immediately a node is reached, whereas betweenness centrality captures the degree to which a node mediates the comprehensive effects of other vertices (Friedkin 1991). Hence, each indicator captures a different structural property of a node. Our choice of betweenness centrality does not entail an idea of superiority over the other two centrality concepts. We consider that identifying which vertices are more strongly capable of affecting flow transmission between any other two nodes is particularly relevant for analysing local 666 A.L. Wirkierman et al. vulnerabilities, which occur due to random as well as targeted shocks. In essence, “a point that falls on the communication paths between other points exhibits a potential for control of their communication” (Freeman 1979, p. 221, italics added). In what follows, this criterion of betweenness centrality highlighting local vulnerabilities will be applied to the analysis of the productive structure of national economies. 2.2 Closed Input-Output Systems as Regular Markov Chains A distinguishing feature of a national economy consists in the comprehensive network of flows of goods and services that are required to produce the final outputs and generate the net incomes ensuring its reproduction. Analytically, this network may be articulated in a dual system of accounting identities, known as the Input-Output (I-O, hereinafter) system (Olsen 1992):      X f e x = (Expenditure) (3) 1 Y yT 0    T  X f = xT F (Income) (4) e 1 yT 0 which, developing the matrix products for a system with n sectors (n−1 intermediate industries and one final sector), may be written as: Each element xij of the square matrix X represents the flow of commodities from industry i to industry j (i.e. inter-industry flows), element fi of column vector f is the flow of commodity i addressed to final uses (final consumption by households and government, investment and foreign demand), while element yj of row vector y T stands for the net incomes, imports of inputs and net taxes paid by industry j in the course of producing its gross output xj . Moreover, note that the total final uses F equals total net incomes Y , though this is not true at the industry level, i.e. fi = yi , precisely because inter-industry sales and purchases at the sectoral level will, generally, differ. The specification above indicates that, when read by row, we describe the demand sources for the products of an industry i, i.e. any (other) industries or final uses. Instead, when read by column, we describe the cost structure of an industry j . The duality in expenditure/income accounting relations corresponds to a dual set of flows occurring in the system: every sell of commodities from i to j implies an equivalent purchase from j to i, so that to every commodity flow corresponds an equivalent money flow in exchange. Hence, if X represents the matrix of product Leontief meets Markov... 667 flows, XT will stand for the matrix of money flows paid for the acquisition of inputs by every industry. A characterising feature of the I-O system (3)–(4) is that it is closed, in the sense that there are no leakages for a monetary unit which is circulating through the economy: every monetary unit spent by final uses is earned by net incomes, and every monetary unit that has been earned by net incomes is spent as final uses. This was the original formulation of the I-O system advanced by Leontief (1949, pp. 214-5): “In a closed system, for example, the level of labor supply would have been directly connected with the level of real income; that is, the quantities of consumers’ goods absorbed by the households”. Income and expenditure are mutually determining each other. To see this, we can specify system (3)–(4) in ‘intensive’ terms, i.e. per unit of gross output of each buying industry:      A af x x = (Activity levels) (5) a Ty 0 Y Y    T  A af = eT 1 (Price Indices) (6) e 1 a Ty 0 x −1 is a matrix of technical coefficients (element aij = xij /xj where A = [aij ] = X specifies the monetary units of the product of industry i needed to produce a monetary unit of gross output of industry j ), a y = [yj /xj ] is the vector of net incomes induced per unit of gross output, and a f = [fi /Y ] is the distribution of final expenditure categories per unit of net income (a vector of ‘propensities’ to spend). When expressed in ‘intensive’ terms, the accounting identity for expenditure flows (3) becomes the equation system to determine activity levels (5), and the accounting identity for income flows (4) becomes the equation system to determine price indices (6).6   A af Key to this transition is the interpretation of coefficient matrix as a a Ty 0 given, structural description of the economy. Systems (5) and (6) may be read as the right and left eigensystems associated to the eigenvalue λ = 1. Then, by assuming that the coefficient matrix is non-negative (i.e. no negative money flows, including net incomes) and irreducible (i.e. it is possible to reach, even indirectly, any industry from every other industry, including the final expenditure/net income sector), the Perron-Frobenius theorem (Meyer 2000, p. 673) applies: the only positive right and left eigenvectors will be those associated to the maximal (in absolute value) eigenvalue. Given that, as can be seen from Eq. 6, the coefficient matrix is column stochastic (i.e. the sum across rows for each column equals one), the maximal eigen- 6 Each entry xij in X might be considered to be measured in terms of the amount of physical output of the product of industry i that can be purchased for one monetary unit (Leontief 1986, p. 22). Thus, in the current accounting period, the price of gross output for each industry will be equal to one, as seen in equation system (6). 668 A.L. Wirkierman et al.   value will be 1. Therefore, activity levels x T Y and price indices eT 1 are simultaneously determined as the right and left eigenvectors associated to the unit eigenvalue of the coefficient matrix of structural coefficients. The essential aspect of connectivity being emphasised here is that of circularity: to determine income we need to know expenditure, and vice-versa. Activity levels and price indices are defined in terms of themselves, as “self-contained (closed) systems require circular definitions” (Brody 1970, p. 84). Circularity implies that there is no beginning nor end to the production process of a national economy (Leontief 1991[1928]). The circular description of interdependent production provided by the closed Input-Output system (5)–(6) may be interpreted as a regular Markov chain (Brody 1970). Given that (6) implies that the structural coefficient matrix is column stochastic, it is possible to transpose (6):  AT a y a Tf 0     e e = 1 1 (7) so that the transposed coefficient matrix in Eq. 7 is a non-negative, irreducible, row stochastic matrix. Therefore, within this context, we can define for an economy with n − 1 industries and a final sector:  P (n×n) = AT a y a Tf 0  (8) as a probability transition matrix derived from the income side of the closed I-O system (3)–(4). Elements of P show the dual side of every transaction in the system: pij = aj i = xj i /xi (Payment from i to j for inputs per unit of monetary output) pin = ayi = yi /xi (Payment from i to final sector as a cost component) pnj = afj = fj /Y (Payment from final sector to industry j for its output) i.e. commodity flows from j to i (as captured by aj i ) correspond to money flows from i to j (as captured by pij ). Rather than required inputs per unit of gross output, the transition probabilities pij “indicate the liabilities incurred per unit of production” (Leontief and Brody 1993, p. 227). Therefore, each row of P represents the cost structure required by each sector to produce its output.7 But what is the chance process interpretation of the probability of transition pij in this context? It may be framed as follows (Brody 1970, p. 108): “Sector [i] goes to market and will buy one day from one sector and another day from others. Its purchases may have an apparently irregular pattern. Some days 7 Note that the inputs of net income components are commodities for final uses. Leontief meets Markov... 669 it may not buy anything because inventories are full, to be depleted at random. Nevertheless the probabilities of spending will be allotted to the other sectors as the coefficients [pij ] — and the real frequency of purchases, followed through, say, a year, will approach this probability”. Thus, the random walker on this Input-Output graph will be one monetary unit (e.g. US$ 1) that circulates through the system, being iteratively spent into different sectors and  earned as income in each production round. The vector of stationary probabilities π T πn would represent the distribution of US$ 1 across sectors as a fixed point of this iterative For the case of observed, unit prices of  circular process.  system (6), we have that: π T πn = x T Y /(x T e + Y ), i.e. stationary probabilities would coincide with (normalised) observed activity levels. In fact, the potential interest of this analysis resorts in mapping structural changes in the probability transition matrix P (due to, e.g. technological progress or new consumption patterns) into changes in the sectoral distribution of activity levels and relative prices (i.e. price indices) in the economy. This represents a relevant methodological difference between a regular and an absorbing Markov chain (Kemeny and Snell 1976, p. 37), characterising closed and open I-O models, respectively. In the latter, a shock to the system is introduced by altering the absorbing state of the chain (i.e. net incomes or final demand). Instead, in the former, a shock can be introduced by perturbations to the transition probability matrix P in Eq. 8 (see, e.g. Moosavi and Isacchini 2017). Given that betweenness centrality aims to quantify local vulnerabilities in the event of a random or targeted attack, the first case may be modelled by introducing small random disturbances adding up to zero (to keep matrix P row stochastic), whereas an attack targeted at a specific node k may be implemented as a reduction of transition probabilities pik of column k by a given factor (αk ), normalising all affected rows (i.e. redistributing αk ·pik to other columns for each affected row i). If the targeted node k is an industry, this normalisation could be done by allocating αk · pik to the corresponding entry pin of the vector of net incomes a y in Eq. 8, which includes imported inputs. Hence, a targeted attack on domestic industry k would imply that other industries would need to source a higher share of input k from abroad. In this targeted case, the procedure would, to a certain extent, resemble an exercise of ‘hypothetical extraction’ within I-O literature (Miller and Blair 2009, p. 563). For the empirical exploration below, we will analyse node vulnerabilities by means of the betweenness centrality indicator mkk , obtained in Eq. 2. Within a random walk on the I-O graph, each node represents a sector, so the extra distance in traversing from i to j — imposed by forcing the passage through sector k — quantifies the intermediary role of an industry (or final sector) in the economy. The lower the value of mkk , the more vulnerable the economic system will be to a reduction in the productive capacity of sector k. Thus, it is possible to set up a ranking from the most central (lowest mkk ) to the most peripheral (highest mkk ) sectors. 670 A.L. Wirkierman et al. 3 An Empirical Exploration: Results and Discussion 3.1 Dataset Characteristics and Preparation: From Inter-Industry Data to Probability Transition Matrices Our data source is the OECD Input-Output Tables (OECD-IOTs) database.8 We consider data for years covering the period 2005-2015. The original database has I-O tables for 64 countries, each with a 36 industry-level disaggregation based on the standard classification ISIC Rev. 4. A key data requirement of the exercise is that the structural matrices considered should be non-negative and irreducible. In order to comply with these requirements, 4 countries had to be discarded.9 Moreover, the 36 original industries had to be aggregated into 32 activities.10 Appendix C provides a dictionary with the ISO3 country codes and the sectoral classification used throughout the paper. Besides the 32 intermediate industries, a closed Input-Output system needs to carefully specify its final sector(s). This is a crucial aspect of the data preparation process, if consistency is to be obtained between theoretical magnitudes and empirical figures. In particular, our probability transition matrices are required to be non-negative. However, taking into account an Input-Output table (IOT, hereinafter) in full detail (i.e. including taxes net of subsidies, separate income components and inventories) involves dealing with cells that, a priori, may contain negative values. We describe below the procedure to map extracted IOTs into a series of empirical transition probability matrices. For each national economy, we specified an empirical closed system which contains three final sectors: (i) the foreign sector, i.e. imports and exports (labelled 991XM); (ii) the households and government sector (labelled 992YC) and (iii) the profits and investment sector (labelled 993PI). This has important theoretical implications for the logic of economic circularity. Exports become the ‘inputs’ to finance the acquisition of imports (i.e. exports will represent the ‘cost structure’ of imports). Earned wages and taxes (net of subsidies) become (private and public) final consumption spent. Finally, profits (i.e. gross operating surplus) obtained by industries become investment (i.e. gross fixed capital formation). The distinction between different income-expenditure connections allows for a richer analysis of general interdependence in a closed model. In order to make these distinctions, it was crucial to map detailed national accounting items contained in each IOT to the analytical categories just defined. Table 1 8 Accessed at: http://www.oecd.org/sti/ind/input-outputtables.htm. countries excluded from the analysis were: BRN (Brunei Darussalam), LVA (Latvia), MLT (Malta) and SGP (Singapore). Exclusions were due to the fact that some industries in these countries had negative gross value added. 10 Three aggregations were performed: (i) the three original mining industries (mining of energy products, non-energy products and mining support services) were combined into a single mining sector, (ii) Coke and refined petroleum, chemicals and pharmaceutical products were combined into a single sector and (iii) the service activities of private households were added to other personal services. 9 The Leontief meets Markov... 671 Table 1 Mapping between OECD-IOT final sector items and model analytical categories (Each analytical category (991XM, 992YC, 993PI) may include several OECD-IOT items) Source: Authors’ elaboration based on OECD-IOTs Database, 2018 edition below reports the connection for both income (i.e. IOT rows) and expenditure (i.e. IOT columns) sides of the I-O system. Besides the aggregation of accounting items into the three analytical final sectors, two further adjustments were done to obtain a non-negative matrix. First, for those industries with substantial agricultural subsidies (e.g. selected EU countries), other taxes less subsidies on production (item OTXS in Table 1) were negative to the extent of rendering the entire income of the Households and Government sector (992YC) negative. For those cases, OTXS were added to the gross operating surplus (GOPS) of the industry (so moved to analytical category 993PI), as subsidies artificially inflate the operating surplus of the firm. Secondly, for those industries with a negative gross operating surplus (GOPS), this accounting item was netted from the Households and Government sector (992YC) as, when losses by firms are distributed, it would be households owning them and government collecting corporate taxes that would face the ultimate consequences. In those sectors where such a correction was made, GOPS was allocated a value equal to zero. The last aspect of the data preparation procedure concerns the treatment of inventories (accounting item INVNT in the OECD-IOT database). These were removed from the analysis, so all levels of income items were recalculated — using the current technical coefficients — for a final demand which excludes inventories. To illustrate the outcome of this data preparation process, Panel (a) of Table 2 reports an articulated selection of cells of the empirical I-O table for the USA (year 2015), arranged into analytical categories, whereas Panel (b) establishes the correspondence with the matrix, vector and scalar symbols to be used in the theoretical operations that follow. 672 A.L. Wirkierman et al. Table 2 Mapping between an empirical IOT and analytical matrix categories Source: Authors’ elaboration based on OECD-IOTs Database, 2018 edition To derive a non-negative, irreducible and row stochastic empirical transition probability matrix, we consider the income side of the I-O system (in an analogous way to system (4) above): ⎡  eT X ⎢ mTx 111 ⎢ ⎣ wT sT fz 0 τz 0 fc mc τc 0 ⎤ fk  mk ⎥ ⎥ = xT Z C I ⎦ τk 0 (9) Developing the matrix products we obtain expressions for gross output and final expenditure categories: x T = eT X + mTx + w T + s T (Gross output by industry) Z = e f z + τz (Exports) C = e f c + mc + τc I = eT f k + mk + τk (Consumption) (Investment) T T 673 Leontief meets Markov... This system of accounting identities can be expressed in intensive terms, to derive a system of price indices: ⎡  eT X fz ⎢ mTx 0 ⎢ 1 1 1 ⎣ T w τz sT 0 fc mc τc 0 ⎤ ⎡ −1  x fk 0 0 0 −1 ⎢ T mk ⎥ 0 0 ⎥⎢ 0 Z τk ⎦ ⎣ 0 T 0 C −1 0 0 0T 0 0 I −1 ⎡ −1  x 0 0 0 −1 ⎢ T ⎥  T 0 0 ⎥= x Z C I ⎢ 0 Z ⎣ 0T ⎦ 0 C −1 0 T 0 0 0 I −1 ⎤ therefore obtaining (analogously to system (6) above): ⎡ ⎤ A az ac ak c ak ⎥ ⎢ a Tm 0 am  T  T m⎥ e 111 ⎢ T z ⎣ a w aτ aτc aτk ⎦ = e 1 1 1 a Ts 0 0 0 (Price Indices) ⎤ ⎥ ⎥ ⎦ (10) The empirical matrix of structural coefficients in Eq. 10 is non-negative, column stochastic and — as has been verified when carrying out the empirical computations — irreducible for all 60 countries and years (covering the 2005-2015 period) considered. Thus, its transpose defines the empirical transition probability matrix P of a Markov chain for a closed Input-Output system: ⎡ T ⎤ A am aw as ⎢ aT 0 az 0 ⎥ z τ ⎥ P =⎢ (Empirical transition probability matrix) (11) ⎣ aT ac ac 0 ⎦ c m τ k ak 0 a Tk am τ Every state of transition matrix P corresponds to an economic sector of the economy (intermediate or final), whereas each cell represents the probability that sector i pays to sector j , in exchange for inputs to carry out its production. Complementing the interpretation of transition probabilities given in Section 2, there is a nuanced richness of extending the specification of the Markov chain to three final sectors. Crucially, each of them will have a greater weight in the economy than (almost) any intermediate industry on its own. Therefore, cross-country differences in the incomeexpenditure connection of imports with exports, wages-cum-taxes with consumption or profits with investment, may hint at specific structural features of the process of economic development. If we consider the foreign sector (991XM), high transition probabilities across the ‘imports’ column would indicate that industries, final consumers and firms heavily rely on imports to carry out production. But once a random monetary unit circulating in the economy reaches an import requirement, it will become a unit of exports in the following iteration of the chance process. And the more diversified the export structure of a country is, the easier it will be to reach other industries. Hence, in a small open economy with very high import transition probabilities but a very concentrated export basket, sectors may keep getting into the foreign sector, but there are very few options to traverse from the foreign sector to other nodes of the network. On the contrary, in a large country with relatively lower import propensities and a very diversified export basket, the random walker is less likely to reach the foreign sector but, once it has, there are more options to traverse to other industries through exports. 674 A.L. Wirkierman et al. This connection between imports and exports is particularly relevant in a context of international fragmentation of production (Hummels et al. 2001). An increase in the import content of exports which originates in higher intermediate import requirements — a Tm in Eq. 10 — implies, ceteris paribus, a higher probability that the random walker traverses to imports, potentially reducing the betweenness centrality of domestic industries. In many cases, imported inputs are re-exported without any added domestic value. A possible way to account for this leakage from the domestic economy would be to add an ‘import per unit of exports’ coefficient in the intersection between the ‘imports’ row and the ‘exports’ column. But while “re-exports and re-imports” are present in the OECD-IOTs database methodology, almost no country reports this disaggregation.11 A direction for further research to fully take into account inter-country input linkages would be to extend our framework to a global Markov chain (Piccardi et al. 2018). A similar reasoning to that of the foreign sector may be applied to the connection between profits and investment (993PI). High profit margins concentrated in a restricted set of industries (i.e. high transition probabilities in the ‘profits’ column), imply that the economy might be highly dependant on few activities to trigger investment demand in the next iteration. Instead, a diversified set of lower profit margins may redistribute purchasing power across industries to activate investment. However, if the sectoral composition of investment is highly concentrated towards few products (such as construction during a speculative housing bubble), whenever the random walker reaches the profit component of costs, traversing to construction will be highly likely in the next iteration (when profits become investment). And if the economy keeps traversing through very similar paths, other industries in the economy might be very difficult to reach, increasing the length of walks on the I-O graph. Finally, an analogous interpretation may be given to the connection between wages-cum-taxes and consumption. A Markov chain approach to analysing a closed Input-Output system evinces the crucial relevance of the compositional structure of final consumption. This is because the wages-cum-taxes income component will generally be the highest transition probability in most industries across countries. But once the random walker has reached the Household and Government node, it will need to traverse to other vertices through final consumption. Thus, the consumption coefficients crucially redistribute the bulk of monetary circulation between industries. Having obtained yearly series (for the period 2005-2015) of transition probability matrices P = [pij ] (each of dimension 35 × 35) for 60 countries, for each of them we first computed its vector of stationary probabilities π = [πi ]. Then, we derived the associated asymmetric Laplacian matrix L =  π (I −P ) and computed its MoorePenrose inverse M = [mij ].12 11 As documented in Timmer (2012, pp. 20-1), countries widely differ in their treatment of processing trade (i.e. re-imports/re-exports) within their national accounts. 12 The Moore-Penrose generalised inverse is obtained by applying the Singular Value Decomposition (SVD) to matrix L with the R programming language function ginv of package MASS. For implementation details, please see Venables and Ripley (1999, p. 100). Leontief meets Markov... 675 We extracted the main diagonal of M, containing coefficients mkk in Eq. 2, each representing the extra distance traversed between two sectors when forced to pass through sector k. Each of these coefficients captures the (reciprocal) betweenness centrality of industry k, allowing us to compare sectoral vulnerabilities across countries and through time. Our empirical strategy consisted in: (i) assess the (cross-country) time trend of the ranking of sectoral vulnerabilities, (ii) illustrate (time-averaged) country differences in their sectoral vulnerabilities, and (iii) perform a clustering exercise to agglomerate countries based on the similarity of their sectoral rankings, devising a country typology of structural vulnerabilities across the globe. 3.2 Generalised Structural Change Takes Time The first dimension we consider emphasises the time evolution of the ranking of sectoral vulnerabilities. We compute the median (rather than the average) of mjj across countries for each sector j , to avoid outliers biasing the results. Table 3 reports levels of mjj for selected years, the time-average, absolute difference and proportional rate of change between 2005 and 2015. On the left panel of the table, sectors are ordered by their position in the industry classification, i.e. primary, manufacturing and service industries (from 01TO03AGR to 90TO98OTS), followed by the three final sectors: foreign trade, households/government and profits/investment (991XM, 992YC and 993PI, respectively). On the right panel of the table, a plot depicts the transitions in the sectoral ranking through time, sectors being displayed in increasing order of mjj , i.e. in decreasing order of betweenness centrality. From the right panel of the table it emerges that structural change takes time: for the first, most central 17 (out of 35) sectors, there are only two temporary and one permanent (adjacent) switchovers in ranking positions. That is, for the median economic system, the relative position of key sectoral vulnerabilities persists through time. As expected, the three final sectors have the smallest extra distance associated to the paths traversed between any two other industries. This is due to the high share of imports, wages-cum-taxes and profits in the unit cost structure of production activities. Besides final sectors, industries with persistently high betweenness centrality across countries and through time belong to either: (i) the final consumption core of the economy, comprising food (10T12FOD, 01T03AGR), housing (41T43CON, 68REA), government services (84GOV), health (86T88HTH) and education (85EDU), or (ii) the infrastructure core of the economy, comprising trade (45T47WRT), KIBS (69T82OBZ),13 logistics (49T53TRN), energy (35T39EGW, 19T21PECH) and financial intermediation (64T66FIN). On the contrary, activities in the lower half of the ranking evince changes in their relative position between 2005 and 2015. Some variations are temporary, e.g. the motor vehicle industry (29MTR) and some of its key inputs (22RBP), probably triggered by the global financial crisis unfolding during 2009. However, some 13 The ‘KIBS’ acronym stands for: Knowledge Intensive Business Services. 676 A.L. Wirkierman et al. Table 3 Extra distance traversed between two sectors when forced to pass through sector j (in number of sectoral transitions) Indicator mjj from Eq. 2; median across 60 countries; sectoral descriptors in Appendix C, Table 7 trajectories identify persistent structural changes across countries, e.g. the increasingly relevant intermediary role of Information Technology (IT, hereinafter) services (62T63ITS) in contrast to the decreasing role of textiles and apparel (13T15TEX). Higher volatility of positions in the lower half of the ranking may be explained by, at least, two factors. Firstly, productive techniques are expected to change at a faster pace than essential consumption needs and core productive infrastructure. For example, while the content of mild steel in cars is projected to decrease (reducing the relative position of 24MET), transportation and mobility services (19T53TRN) still remain crucial. Secondly, position changes may be reflecting processes of international fragmentation of production across countries. This is because the ranking depicts the betweenness centrality of domestic output by industries. To see this, consider the example of textiles and apparel (13T15TEX), in which several European countries have outsourced industry segments to selected Asian economies. By Leontief meets Markov... 677 Fig. 2 Central tendency (Median) and spread (MAD) across countries of mjj for each sector j (MAD: median absolute deviation). Due to their extreme values, industries 62T63ITS and 30TRQ are not displayed, but considered in the analysis of results. Full description of industry codes is available in Table 7 of Appendix C. Source: Authors’ elaboration based on OECD-IOTs Database, 2018 edition increasingly relying on imports to satisfy textile input requirements or final products, the likelihood with which other sectors traverse to the domestic textile industry is reduced. Hence, when forcing the path between any two sectors through industry 13T15TEX, the extra distance travelled is likely to have increased, reducing the betweenness centrality of the industry. Levels of mjj indicate the extra number of sectoral transitions in the path between any two sectors forced to traverse through node j . Therefore, a reduction in the level of mjj corresponds to an increase in the betweenness centrality of node j . Inspection of the left panel of Table 3 suggests that, between 2005 and 2015, highest increases in mjj concentrated around selected primary and manufacturing industries, whereas decreases mostly involved service activities. To better understand this pattern of changes, Fig. 2 depicts initial levels (x-axis) and dynamics (y-axis) of the central tendency and spread of mjj , for each sector j . Panel (A) in Fig. 2 evinces that less central sectors in 2005 became more peripheral by 2015. That is, for sectors with initially high betweenness centrality (i.e. closer to the origin of the x-axis), the absolute change in centrality between 2005 and 2015 has been contained. From Panel (B) it emerges that sectors with a more dispersed cross-country level of mjj in 2005 (i.e. further away from the origin of the x-axis), experienced a sharper increase in dispersion between 2005 and 2015.14 Thus, the general pattern suggested is that central sectors become more central across most countries, whereas peripheral sectors became even more remote but to an uneven 14 In correspondence with adopting the median as the measure of central tendency, we computed the deviation (MAD) across countries (index r) for each sector j , i.e. MADj =  median absolute r r r median(|mjj − m̃jj |), where m̃jj = median(mjj ). 678 A.L. Wirkierman et al. extent depending on the country considered. In a nutshell, between 2005 and 2015, the cross-country evolution of sectoral vulnerabilities has experienced a phenomenon of ‘convergence at the top, divergence at the bottom’ of the betweenness centrality scale. It would be interesting to compare the sectoral ranking emerging from Table 3 with more traditional I-O indicators, such as simple output multipliers (Miller and Blair 2009, p. 246) and even a sector’s gross output — xj in Eq. 9. The results of such comparison are reported in Appendix B, and allow to contextualise and highlight the additional circularity captured by a fully closed I-O scheme. 3.3 Countries are Structurally Different While the upper-half of the ranking of sectoral vulnerabilities for the cross-country median (reported in the right panel of Table 3) may show persistence and contained changes in dispersion, differences between individual economies are still pervasive. Interpreting such differences may lead to a qualitative understanding of the relationship between country-specific aspects of the productive structure and changing sectoral vulnerabilities along a development path. To illustrate this, Table 4 reports (time-averaged) country differences in sectoral vulnerabilities — as measured by mjj — for China, the median world economy and the United States. The right-panel of the table depicts the transitions in the sectoral ranking of vulnerabilities between these three economies. We have highlighted industries whose ranking position experiences a sharp change in direction, both when transitioning from China to the median world economy, as well as from the latter to the USA. Given that China is, by several metrics, the key emerging economy in the period considered, whereas the United States is still the advanced industrial economy with highest share in global income, ranking transitions exemplify the changing role of sectoral vulnerabilities along (possible) paths of economic growth. While the households/government (992YC) and profits/investment (993PI) final sectors remain crucial across countries, the foreign sector (991XM) becomes less central for the US than it is for China and the median world economy. Intuitively, industry cost structures in larger advanced industrial economies tend to be relatively less dependent on foreign inputs. However, given the extent of its involvement across inter-country value chains and its developmental stage, the foreign sector is still the third source of vulnerability for China. Sectors with a sharp decrease in their relevance as intermediary nodes when transitioning from China to the US (and through the median world economy) comprise, first and foremost, agricultural and food processed products (01T03AGR, 10T12FOD). This evinces the ongoing process of population and labour force expansion in the Chinese economy. A similar point could be made about textiles and wearing apparel (13T15TEX), though this may also be linked to China’s role in global value chains (GVCs) for these products. Finally, the case of basic metals and mechanical machinery equipment (24MET and 28MEQ, respectively) evinces the ongoing process of expanding industrialisation in China, with respect to the US and the median world economy. Leontief meets Markov... 679 Table 4 Extra distance traversed between two sectors when forced to pass through sector j (in number of sectoral transitions) Indicator mjj from Eq. 2; countries: China (CHN), World median (WLD) and the United States (USA); time average for years 2005, 2009 and 2015; sectoral descriptors in Appendix C, Table 7 Instead, the increasing betweenness centrality of (knowledge-intensive) business services in the USA (69T82OBZ) — being the third most relevant sector — illustrates the prominence of its ‘service economy’. Moreover, the US technological leadership may be perceived by the high relative ranking position of IT services (62T63ITS). This process of ever-increasing centrality of service sectors includes technologically advanced personal services (86T88HTH) but also activities leveraging the financialisation process of the US economy (68REA, 64T66FIN). Such a high concentration of node vulnerabilities in employment-intensive service sectors, alerts on the high potential impact of a shock to the US labour force. 3.4 Structural Vulnerabilities: A Typology The comparison between China and the USA evinced contrasting sectoral vulnerabilities between them, as well as with respect to the median world economy. 680 A.L. Wirkierman et al. Therefore, to make sense of cross-country differences in node vulnerabilities, we report the results of a clustering exercise in which we agglomerate countries based on the similarity of their sectoral rankings, devising a country typology of structural vulnerabilities across the globe. In particular, we departed from a sector × country matrix of the (time-averaged) values for mjj and converted each column into a ranking of node vulnerabilities. Then, we computed the Spearman rank correlation between each pair of columns and articulated all (bilateral) correlation coefficients into a resulting symmetric (country matrix was then converted into × country) similarity matrix S = [sij ]. The similarity  a dissimilarity matrix, by computing: dij = 2 × (1 − sij ), and a hierarchical clustering analysis (using the complete agglomeration method) was performed, obtaining the dendrogram displayed in Fig. 3. As a heuristic device, we articulated results of the cluster analysis into 6 groups (G01 to G06 in Fig. 3). We provide a qualitative characterisation of each group based on the ranking of node vulnerabilities across countries within each country set, reported in Table 5. In counter-clockwise order, group G01 comprises mainly China and South East Asian countries (with the exception of MEX and TUN). Economies of ‘factory Asia’ Fig. 3 Hierarchical clustering of countries according to the relative distance in their sectoral rankings for mjj . (Indicator mjj from Eq. 2; Clustering agglomeration method: complete; 60 countries; averaged through time for years 2005, 2009 and 2015; country descriptors in Appendix C, Table 6). Source: Authors’ elaboration based on OECD-IOTs Database, 2018 edition Leontief meets Markov... 681 Table 5 Extra distance traversed between two sectors when forced to pass through sector j (in number of sectoral transitions) Indicator mjj from Eq. 2; median across countries within each group: G01-G06, detailed in Fig. 3; averaged through time for years 2005, 2009 and 2015; sectoral descriptors in Appendix C, Table 7 Source: Authors’ elaboration based on OECD-IOTs Database, 2018 edition evince high inter-country trade integration, as can be seen from the high vulnerability associated to the foreign trade sector (991XM), as well as industries which articulate key global value chains, such as wearing apparel (13T15TEX) and computing equipment (26CEQ). Being a group of fast growing, emerging countries, their developmental stage is characterised by an expanding population (01T03AGR, 10T12FOD) and natural resource exploitation (05T09MIN, 19T21PECH). Countries of group G02 are (mostly) middle-income economies of South America, the Black Sea area, India and Morocco. The important intermediary role of the construction sector (41T43CON) should be highlighted, as well as the extremely low ranking positions of high-tech industries (26CEQ, 62T63ITS), suggesting these economies are far from the technological frontier. The characterising feature of the third group (G03) is that it consists of naturalresource based economies. This is evinced by the relatively high vulnerability of mining and refined petroleum (05T09MIN, 19T21PECH), agricultural activities (01T03AGR) and food processing (10T12FOD). Countries at different development stages are further distinguished within the group: emerging (SAU, RUS, BRA, ZAF) 682 A.L. Wirkierman et al. and advanced (NOR, AUS, CAN) economies with a high reliance on mining resources agglomerate into different sub-clusters. Instead, for countries of group G04, mining is the least central activity of their domestic productive structure. This group represents a core set of manufacturingintensive economies comprising advanced East Asian countries, Sweden, Germany and its associated (through strong value chain linkages) Central and Eastern European countries. Heavy industry and high-tech manufacturing sectors feature prominently among node vulnerabilities: motor vehicles (29MTR), mechanical machinery (28MEQ), computing equipment (26CEQ) and fabricated metal products (25FBM). Countries in group G05 may be described as advanced service economies: it comprises the USA, the UK and high-income small open economies featuring the prominence of knowledge-intensive (69T82OBZ) and financial (68REA, 64T66FIN) services, but also a higher relative betweenness centrality of high-tech industries (62T63ITS, 26CEQ). Finally, group G06 includes a set of European countries (with the exception of CRI) featuring (relatively) high node vulnerabilities in the service infrastructure of the economy (e.g. 69T82OBZ, 49T53TRN, 68REA, 64T66FIN), but also in traditional manufacturing sectors (10T12FOD), and further away from the technological frontier (62T63ITS, 26CEQ) than countries of group G05. Overall, results from the cluster analysis suggest a country typology with differentiated structural profiles. At the risk of oversimplifying, clusters depict emerging economies undergoing fast-paced structural transformation (G01), an array of natural-resource based economies (G03), a high-tech manufacturing core (G04), a high-tech services core (G05), and high-income European (G06) / middle-income peripheral (G02) countries with varying degrees of distance from the technological frontier. A direction for further research would be to study the inter-country trade complementarities between these clusters. 4 Summary of Findings and Concluding Remarks Input-Output (I-O, hereinafter) analyses of structural vulnerabilities are generally performed using the open I-O model (Galbusera and Giannopoulos 2018). Instead, a key contribution of this paper has been to specify and empirically implement the precise mapping between a regular Markov chain and a closed Input-Output system, with multiple categories for the final sector. Amongst different methods to assess I-O network resilience, identification of key sectors is frequently used (e.g. Kelly et al. 2016). But rather than focusing on traditional indicators (such as backward/forward linkages), we applied a notion of betweenness centrality in complex networks (Ranjan and Zhang 2013) to characterise sectoral vulnerabilities in national economies, evincing structural (dis)similarities. Empirical findings suggest that, for the median economic system across the 60 countries considered, the relative position of key sectoral vulnerabilities persists through time. However, between 2005 and 2015, the cross-country evolution of relative vulnerabilities experienced a phenomenon of ‘convergence at the top, divergence at the bottom’ of the betweenness centrality scale. Moreover, a generalised feature of 683 Leontief meets Markov... structural change has been the increasingly relevant intermediary role of Information Technology services. Whilst persistence through time of the intermediary role for key network nodes has been observed, there are pervasive differences between individual economies. A comparison between China and the US reveals changing sectoral vulnerabilities along development paths, traversing from high vulnerability in agricultural and food processed products (in China) to a crucial role for business services (in the US). Clustering economies — based on the similarity of their sectoral rankings — suggests a country typology with differentiated structural profiles, highlighting the distinctions between emerging, industrialisation-intensive economies and natural resource-based countries, as well as between an advanced manufacturing core with respect to a high-tech services core of economies. Further research could study the inter-country trade complementarities between these clusters from a network perspective (Caschili et al. 2015). In fact, a limitation emerging from these results is that our analysis is not framed in terms of a set of interdependent economies, articulated in an inter-country, global Markov chain (as in Piccardi et al. 2018). This notwithstanding, we acknowledge the relations each country has with the rest of the world through the ‘foreign trade’ sector. Finally, an interesting direction for further research would be to simulate targeted attacks to each national (or a global) I-O network through the hypothetical extraction method (Dietzenbacher et al. 2019), iteratively measuring how the sectoral distribution of vulnerabilities changes as more central nodes are removed from the network. Such an approach may hint at economy-wide resilience, complementing the identification of local vulnerabilities. Appendix A: Relationship Between Network Centrality Indices Within the context of random walks on graphs, the notions of eigenvector (Friedkin 1991), closeness (Noh and Rieger 2004) and betweenness (Newman 2005) centrality are amongst the most widely used. Not only for abstract descriptions of network structure, but also when it comes to the analysis of economic, inter-industry flows (Blöchl et al. 2011). In this appendix we formally relate the three centrality notions, in order to clarify how the betweenness centrality indicator for node j , mjj , defined in Eq. 2, considers paths connecting all network nodes. To begin with, recall Eq. 1: H (i, j ) = mjj − mij + n  (mil − mj l )πl l=1 Aggregating over i, we obtain: n  i=1 H (i, j ) = n · mjj − n  i=1 mij + n  l=1 πl n  i=1 mil − n n  l=1 mj l π l 684 A.L. Wirkierman et al. But due to the properties  of the Moore-Penrose inverse (Boley et al. 2011, p. 232), we have ni=1 mij = ni=1 mil = 0, so we may write: mjj = n n  1  · H (i, j ) − mj l π l n i=1 (12) l=1 Note that the first term of the right-hand side of Eq. 12: MF T P (j ) = n 1  · H (i, j ) n (13) i=1 represents the mean first passage time of target node j , i.e. the average number of state transitions needed to reach node j for the first time, averaged across all possible source nodes, known as an indicator of closeness centrality for node j (Blöchl et al. 2011, eq. (5)). Hence, our betweenness centrality indicator mjj may be written as: mjj = MF P T (j ) − n  mj l π l (14) l=1 i.e. it is a measure of closeness centrality, MF P T (j ), corrected by a weighted average of industries’ eigenvector centralities (πl ), the weights being the corresponding elements of row j of the Moore-Penrose inverse (mj l ). In this way, expression (14) renders explicit how our betweenness centrality indicator mjj is connected with the notions of closeness and eigenvector centrality, within the context of random walks on graphs. Appendix B: Comparison with Traditional Input-Output Indicators The interpretation of mjj in Eq. 2 as an indicator of betweenness centrality for a node in a closed Input-Output (I-O, hereinafter) system is, to our knowledge, novel. Hence, it would be useful to compare the (cross-country) sectoral ranking implied by this measure with traditional I-O measures of industry centrality. In particular, two such measures are considered, namely the simple output multiplier and gross output for industry j , respectively: output mult j = eT (I − A)−1 ej go j = xj (15) (16) where A is the technical coefficients matrix defined in Eqs. 5–6, e = [1 . . . 1 . . . 1]T is a sum vector, and ej = [0 . . . 1 . . . 0]T is a column selector vector (with 1 in the j th. position and zeros elsewhere). The simple output multiplier measures the intensity of economy-wide backward linkages activated by final demand addressed to industry j . In this sense, “comparison of output multipliers would show where [an additional unit of final] spending would have the greatest impact in terms of total dollar value of output generated Leontief meets Markov... 685 throughout the economy” (Miller and Blair 2009, p. 246). As such, it may be interpreted as an industry centrality indicator. However, being formulated within the logic of an open I-O model, it only considers network paths amongst industries, whilst mjj accounts also for network paths connecting final sectors. This conceptual difference is expected to generate a mismatch in centrality rankings, given the crucial influence of final sectors in structuring circularity. A second comparison of interest is that with industry gross output xj which, within the network implied by system (9), represents a node’s degree. Note that xj includes both inter-industry as well as final sector transactions. Figure 4 reports the industry ranking comparison between alternative centrality indicators. To allow for comparability with Eq. 15, the analysis is restricted to the set of industries, i.e. excluding final sectors. The left panel depicts industry rankings for cross-country, time-averaged median values of each centrality indicator (m jj, go j and output mult j), whereas the right panels of the figure depict the correlation between indicators for pooled country-level (time-averaged) data points. Two main features emerge. First, there is a neat negative correlation between mjj and gross output of industry j , i.e. lower extra distance traversed indicating a higher betweenness centrality generally corresponds to a higher node degree. In fact, for the top 15 industries, there are only adjacent switchovers in ranking positions when moving from betweenness centrality (m jj) to gross output (go j). Recall that these top industries correspond to the final consumption/infrastructure core of the economy identified from Table 3, and feature prominently in the industry composition of final sectors of the economy. Notably, though, the (log-scaled) scatter plot evinces that the negative relationship between these two indicators is subject to considerable variability. Hence, while an industry ranking based on gross output may be a good predictor of the relative ordering of top industries according to betweenness centrality, there are relevant non-linearities which our centrality indicator mjj is capturing, with respect to a node’s degree. These non-linearities are particularly relevant when assessing local vulnerabilities across countries. Second, the correlation between mjj and the simple output multiplier of industry j is close to zero. As expected, the notorious leaps in ranking positions when moving from betweenness centrality (m jj) to output multipliers (output mult j) evinces the contrast between a closed and an open I-O model, respectively. By not accounting for the circularity intertwining final sectors with intermediate industries, activities ranking high in terms of mjj may have relatively lower output multipliers (e.g. 45T47WRT, 69T82OBZ, 68REA, 84GOV, 85EDU). On the contrary, industries mostly producing (and intensively using) intermediate inputs may rank high in terms of output multipliers, but have relatively lower betweenness centrality (e.g. 23NMM, 22RBP, 58T60PAB, 17T18PAP, 16WOD). 686 A.L. Wirkierman et al. 10 Corr. = −0.651 41T43CON 45T47WRT 69T82OBZ 8 49T53TRN 68REA log(m_jj) 19T21PECH 10T12FOD 84GOV 6 64T66FIN 35T39EGW 86T88HTH 4 Industry Ranking (in decreasing order) 01T03AGR 85EDU 55T56HTR 90T98OTS 2 5 24MET 10 15 log(Gross output industry j) 28MEQ 25FBM 10 Corr. = −0.021 29MTR 31T33OTM 62T63ITS 8 05T09MIN 23NMM log(m_jj) 61TLC 26CEQ 27ELQ 6 22RBP 58T60PAB 17T18PAP 4 13T15TEX 16WOD 30TRQ m_jj go_j Centrality indicator output_mult_j 2 1.0 1.5 2.0 2.5 Output multiplier industry j Fig. 4 Industry ranking comparison between mjj , gross output and I-O output multiplier. (Indicator mjj — m jj— from Eq. 2; Gross output — go j — from Eq. 16; I-O output multiplier — output mult j — from Eq. 15; cross-country median averaged through time for years 2005, 2009 and 2015; sectoral descriptors in Appendix C, Table 7). Source: Authors’ elaboration based on OECD-IOTs Database, 2018 edition. Notes: To facilitate interpretation, gross output and I-O output multipliers in the left panel have been multiplied by −1, so that the logic of their industry ranking in decreasing order coincides with that of mjj Leontief meets Markov... Appendix C: OECD Input-Output Tables Database: Country List and Industry Classification Table 6 OECD Input-Output Tables (IOTs) database country dictionary (List of 60 countries considered for the computations in the paper) Source: OECD Input-Output Tables (IOTs) Database, 2018 edition 687 688 A.L. Wirkierman et al. Table 7 OECD Input-Output Tables (IOTs) sector classification dictionary (35 sectors: 32 industries and 3 final sectors considered for the computations in the paper) Source: Authors’ elaboration based on OECD Input-Output Tables (IOTs) Database, 2018 edition Acknowledgements We gratefully acknowledge the comments and suggestions made by the anonymous referees and guest editors. Responsibility for any errors in the resulting work remains our own. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Leontief meets Markov... 689 References Adger WN (2006) Vulnerability. Glob Environ Chang 16(3):268–281 Blöchl F, Theis FJ, Vega-Redondo F, Fisher EO (2011) Vertex centralities in input-output networks reveal the structure of modern economies. Phys Rev E 83(4):046127 Boley D, Ranjan G, Zhang ZL (2011) Commute times for a directed graph using an asymmetric Laplacian. Linear Algebra Appl 435(2):224–242 Brody A (1970) Prices proportions and planning. Akademiai Kiado, Budapest Cardinale I (2019) Vulnerability, resilience and ‘systemic interest’: A connectivity approach. Netw Spat Econ. https://doi.org/10.1007/s11067-019-09462-9 Caschili S, Medda FR, Wilson A (2015) An interdependent multi-layer model: resilience of international networks. Netw Spat Econ 15(2):313–335 Dietzenbacher E, van Burken B, Kondo Y (2019) Hypothetical extractions from a global perspective. Econ Syst Res 31(4):505–519 Duchin F, Levine SH (2010) Embodied resource flows and product flows: Combining the absorbing Markov chain with the input-output model. J Ind Ecol 14(4):586–597 Franceschet M (2011) Pagerank: Standing on the shoulders of giants. Commun ACM 54(6):92–101 Freeman L (1979) Centrality in social networks - conceptual clarification. Soc Netw 1:215–239 Friedkin NE (1991) Theoretical foundations for centrality measures. Am J Sociol 96:1478–1504 Galbusera L, Giannopoulos G (2018) On input-output economic models in disaster impact assessment. Int J Disaster Risk Reduction 30:186–198 Grinstead CM, Snell JL (1997) Introduction to probability. Am Math Soc Hummels D, Ishii J, Yi KM (2001) The nature and growth of vertical specialization in world trade. J Int Econ 54:75–96 Iyer S, Killingback T, Sundaram B, Wang Z (2013) Attack robustness and centrality of complex networks. PloS ONE 8(4) Kelly S, Tyler P, Crawford-Brown D (2016) Exploring vulnerability and interdependency of UK infrastructure using key-linkages analysis. Netw Spat Econ 16(3):865–892 Kemeny JG, Snell JL (1976) Finite Markov chains. Springer, New York Kostoska O, Stojkoski V, Kocarev L (2020) On the structure of the world economy: An absorbing Markov chain approach. Entropy 22(4):482 Leontief WW (1937) Interrelation of prices, output, savings, and investment. Rev Econ Stat 19(3):109–132 Leontief WW (1949) Recent developments in the study of interindustrial relationships. Am Econ Rev 39(3):211–225 Leontief WW (1986) Input-output economics, 2nd edn. Oxford University Press, Oxford Leontief WW (1991[1928]) The economy as a circular flow. Struct Chang Econ Dyn 2(1):181–212 Leontief WW, Brody A (1993) Money-flow computations. Econ Syst Res 5(3):225–233 McNerney J, Fath BD, Silverberg G (2013) Network structure of inter-industry flows. Physica A Stat Mech Appl 392(24):6427–6441 Meyer CD (2000) Matrix analysis and applied linear algebra. SIAM Miller RE, Blair PD (2009) Input-output analysis: foundations and extensions, 2nd edn. Cambridge University Press, Cambridge Moosavi V, Isacchini G (2017) A Markovian model of evolving world input-output network. PloS ONE 12(10) Morgenstern O (ed) (1954) Economic activity analysis. John Wiley & Sons, New York Newman ME (2005) A measure of betweenneess centrality based on random walks. Soc Netw 27(1):39–54 Newman ME (2010) Networks: An introduction. Oxford University Press, New York Noh JD, Rieger H (2004) Random walks on complex networks. Phys Rev Lett 92(11):118701–1–4 Olsen JA (1992) Input-output models, directed graphs and flows in networks. Econ Model 9(4):365–384 Palacios JL (1990) Bounds on expected hitting times for a random walk on a connected graph. Linear Algebra Appl 141:241–252 Piccardi C, Riccaboni M, Tajoli L, Zhu Z (2018) Random walks on the world input-output network. J Complex Netw 6(2):187–205 Ranjan G, Zhang ZL (2013) Geometry of complex networks and topological centrality. Phisica A 392:3833–3845 690 A.L. Wirkierman et al. Solow R (1952) On the structure of linear models. Econometrica 20(1):29–46 Timmer M (2012) The World Input-Output Database (WIOD): Contents, Sources and Methods, wIOD Working Paper Nro. 10, available at: http://www.wiod.org Tsekeris T (2017) Network analysis of inter-sectoral relationships and key sectors in the Greek economy. J Econ Interac Coord 12(2):413–435 UN (2009) System of National Accounts 2008 ST/ESA/STAT/SER.F/2/Rev.5. United Nations, New York Venables WN, Ripley BD (1999) Modern applied statistics with S-PLUS, 3rd edn. Springer-Verlag, New York Wilson RJ (1972) Introduction to graph theory. Oliver & Boyd, Edinburgh Xing L, Dong X, Guan J (2017) Global industrial impact coefficient based on random walk process and inter-country input-output table. Physica A Stat Mech Appl 471:576–591 Xing L, Guan J, Wu S (2018) Measuring the impact of final demand on global production system based on Markov process. Physica A Stat Mech Appl 502:148–163 Xu M, Allenby BR, Crittenden JC (2011) Interconnectedness and resilience of the US economy. Adv Complex Syst 14(05):649–672 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Affiliations Ariel L. Wirkierman1 · Monica Bianchi2 · Anna Torriero2 Monica Bianchi monica.bianchi@unicatt.it Anna Torriero anna.torriero@unicatt.it 1 Institute of Management Studies (IMS), Goldsmiths, University of London, London, UK 2 Dipartimento di Matematica per le Scienze Economiche, Finanziarie ed Attuariali, Università Cattolica del Sacro Cuore, Milano, Italy
https://openalex.org/W4287185003
https://hal.archives-ouvertes.fr/hal-03503666/document
French
null
Proceedings of the 16th SAGEO Conference
HAL (Le Centre pour la Communication Scientifique Directe)
2,021
public-domain
120,978
To cite this version: Paule-Annick Davoine, Didier Josselin, François Pinet. Actes de la 16ème Conférence SAGEO. Spatial Analysis and GEOmatics, May 2021, La Rochelle, France. 2021, 978-2-910545-12-1. ￿hal-03503666￿ Paule-Annick Davoine, Didier Josselin, François Pinet. Actes de la 16ème Conférence SAGEO. Spatial Analysis and GEOmatics, May 2021, La Rochelle, France. 2021, 978-2-910545-12-1. ￿hal-03503666￿               HAL Id: hal-03503666 https://hal.science/hal-03503666v1 Submitted on 28 Dec 2021 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Public Domain SAGE 2021 SAGE 2021 La Rochelle, 5-7 mai 2021 Actes de la conférence La Rochelle, 5-7 mai 2021 Actes de la conférence SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Avant propos / Foreword Avant propos / Foreword La conférence internationale francophone SAGEO, soutenue par le GDR MAGIS (http://gdr-ma­ gis.imag.fr/), est un événement annuel majeur dans le paysage de la Géomatique, de l’Analyse Spatiale et des Sciences de l’Information Géographique. Ses objectifs sont de présenter des travaux de recherche récents et de réunir des chercheurs de disciplines diverses (informatique, géographie, archéologie, urbanisme, sciences de l’environ­ nement, agronomie, humanités spatiales, etc.). Il s’agit également de promouvoir les échanges entre les différents acteurs du monde de la géomatique (chercheurs et enseignants-chercheurs, ingénieurs, gestionnaires, aménageurs, entrepreneurs...) qui œuvrent au développement et à la valorisation des recherches au sein et auprès des organismes publics, des entreprises et des col­ lectivités territoriales. Nous étions heureux à l’idée de vous accueillir à la Rochelle, la grande ville du littoral charentais à l’interface de deux mondes fertiles, la terre et la mer, comme peut l’être la géomatique. Il ne faut en effet rien moins que toute l’interdisciplinarité au cœur de la géomatique pour saisir la com­ plexité d’un tel territoire, très attractif et en pleine croissance, mais également à haut risque en ces temps de changements et transition climatique. Le littoral urbain durable et intelligent que se propose de concevoir l’Université de la Rochelle doit composer avec ces différentes dimensions, géographiques, informatiques mais aussi écologiques, géophysiques, historiques, pour construire un avenir sécurisé et viable aux générations futures. Il s’agit de comprendre, analyser et anticiper la trajectoire d’un socio-éco-système littoral qui doit concilier les enjeux du développement du­ rable, de la gestion du risque et de l’évolution vers les villes et les territoires intelligents à travers des contributions sur les problématiques littorales, urbaines ou environnementales. Ce sont les enjeux au cœur de cette édition 2021. Les articles soumis ont été évalués à travers 106 rapports d’évaluation réalisés par le Comité de Programme de SAGEO’2021 que nous remercions. Dans ce contexte, nous avons souhaité proposer divers types de communications qui mettent en relief la richesse et la complémentarité des travaux et des applications. Dans ce contexte, nous avons souhaité proposer divers types de communications qui mettent en relief la richesse et la complémentarité des travaux et des applications. Avant propos / Foreword • 3 conférences invitées traitant des différentes dimensions de la géomatique : Michela Bertolotto, School of Computer Science de l’University College Dublin (Irlande) « Evaluation, gestion des risques et connaissances pour la gestion des inondations dans les zones urbaines » Françoise Gourmelon, directrice de recherche au CNRS en Géographie de l’Environnement, Plouzané (France) O t d d é é hi t t j d l Environnement, Plouzané (France) « Ouverture des données géographiques : nouveaux contextes, nouveaux enjeux de recherche » Antoine Doucet, professeur des universités au laboratoire L3i de La Rochelle Université (France) ii Antoine Doucet, professeur des universités au laboratoire L3i de La Rochelle Université (France) « Extraction cross-lingue d’informations géographiques depuis des corpus numérisés, Antoine Doucet, professeur des universités au laboratoire L3i de La Rochelle Université (France) « Extraction cross-lingue d’informations géographiques depuis des corpus numérisés, notamment historiques » « Extraction cross-lingue d’informations géographiques depuis des corpus numérisés, notamment historiques » 3 • 19 présentations qui répondent pour moitié aux enjeux du Littoral urbain durable et intel­ ligent à travers les trois premières sessions (analyse d’image pour l’observation et le suivi du littoral, transition environnementale et énergétique, et étude de l’environnement urbain), 3 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 et pour moitié à des questions fondamentales concernant la mobilité, la modélisation spa­ tio-temporelle, et l’exploitation du Web de données. et pour moitié à des questions fondamentales concernant la mobilité, la modélisation spa­ tio-temporelle, et l’exploitation du Web de données. • 25 posters, mis en valeur à travers une galerie en ligne et la possibilité de voter pour le meil­ leur poster durant le temps de la conférence.­ Par ailleurs, trois ateliers ont été sélectionnés afin d’animer les différentes communautés autour de la géomatique. Nous remercions les responsables des ateliers qui participent activement au foisonnement scientifique de ce domaine pluridisciplinaire. Une session organisée avec l’AFIGEO montre que les rencontres et les échanges entre les laboratoires de recherche spécialisés en Géomatique et les autres acteurs de la géomatique (les entreprises, les collectivités territoriales) sont fertiles pour tous. Cette dynamique va se poursuivre, à travers notamment la participation du GdR CNRS MAGIS aux GeoDataDays qui se tiendront à Grenoble les 15 et 16 septembre 2021. Avant propos / Foreword Enfin, nous remercions très chaleureusement les sponsors (GDR CNRS MAGIS, CNRS, l’Université de la Rochelle, la Communauté d’Agglomération de la Rochelle, l’IUT de la Rochelle), et Nathalie Renaudin (Blanchard) qui a piloté de multiples aspects pratiques de SAGEO’2021, Isabelle Héraudeau pour le suivi financier et l’ensemble des membres du comité d’organisation issus de deux unités de recherche (LIENSs et L3I) pour le travail réalisé avec entrain, enthousiasme et professionnalisme. Un grand merci également à l’accompagnement du comité scientifique, et à Nathalie Brachet pour la constitution des actes. Pour finir, cette édition 2021 qui est entièrement virtualisée du fait de la crise sanitaire COVID 19 a largement mobilisé la communauté, inscrite en masse sur les différentes sessions proposées. Nous espérons avoir l’occasion ultérieurement d’accueillir SAGEO et les membres de la communauté en « vrai » dans cette belle ville de la Rochelle autour de fruits de mer et autres délicieux mets que nous offre ce beau littoral. Le Comité d’organisation de SAGEO’2021 Le Comité d’organisation de SAGEO’2021 4 4 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Comité scientifique / Program Committee Chairs Didier Josselin.............................UMR ESPACE 7300 CNRS, Université d’Avignon Paule-Annick Davoine..................Université Grenoble Alpes François Pinet..............................INRAE Members Jean-Michel Follin.......................Laboratoire GeF - Le Cnam Cyril Ray........................................Arts et Metiers Institute of Technology, École Navale, IRENav Eric Desjardin..............................URCA/CReSTIC Cyril de Runz................................LIFAT, Université de Tours Anne Ruas....................................IFSTTAR Giovanni Fusco............................CNRS - UMR 7300 ESPACE Sébastien Rey-Coyrehourcq........UMR IDEES Marion Le Texier..........................Université de Rouen Normandie Mathieu Roche............................Cirad, TETIS Gilles Gesquiere...........................LIRIS Thérèse Libourel.........................Université Montpellier - UMR Espace-Dev Frédéric Flouvat..........................Universiy of New Caledonia Isabelle Mougenot......................Université Montpellier EspaceDev Thomas Louail.............................CNRS Pierre Dias....................................Université Picardie Jules Verne Guillaume Touya..........................IGN Laurent Etienne...........................LABISEN Carmen Brando...........................École des hautes études en sciences sociales (EHESS) Didier Leibovici............................The University of Sheffield Jean-François Girres....................UMR GRED - Université Paul Valéry Montpellier 3 Jean-Pierre Chanet......................INRAE Jérome Gensel.............................Laboratoire d’Informatique de Grenoble Ali Hassan....................................Umanis André Miralles............................Irstea, UMR TETIS Cécile Duchene............................IGN-France, COGIT Laboratory Paule-Annick Davoine..................Université Grenoble Alpes Alain Bouju..................................La Rochelle University Thomas Devogele........................universite Francois rabelais (Tours) Eric Guilbert................................Laval University Sidonie Christophe......................Paris-Est University, IGN-ENSG, LaSTIG David Sheeren..............................UMR DYNAFOR Christophe Claramunt................Naval Academy Research Institute Florence Le Ber............................icube Lucile Sautot................................AgroParisTech (Dép. SIAFEE, UMR TETIS) Roland Billen...............................University of Liege Gaëtan Palka................................UMR 6554 LETG Rennes Helene Mathian...........................CNRS- UMR 5600 EVS Marie-Hélène de Sède-Marceau..laboratoire ThéMA Sandro Bimonte...........................INRAE Géraldine Del Mondo..................LITIS Christian Sallaberry....................University of Pau et Pays de l’Adour Khaled Rezeg................................Mohamed Khider BISKRA UNIVERSITY Florence Sedes.............................I.R.I.T. Univ. P. SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Comité d’organisation / Steering committee Chairs Alain Bouju..................................EA L3I, La Rochelle Christine Plumejeaud-Perreau....UMR 7266 LIENSs, La Rochelle Members Mickaël Coustaty Cyril Faucher Jamal Malki Damien Mondou Muzzamil Luqman Cyrille Suire SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Avant propos / Foreword Sabatier Michel Mainguenaud..................INSA-Rouen Matthieu Noucher.......................CNRS UMR 5319 Passages Mireille Batton-Hubert...............EMSE Aldo Napoli..................................MINES ParisTech - CRC Rodolphe Devillers......................Memorial University of Newfoundland Nathalie Long..............................LIENSs , CNRS-Université La Rochelle Mauro Gaio..................................LMAP-UMR5142 Eric Kergosien..............................Laboratoire GERIICO Maguelonne Teisseire..................INRAE - UMR Tetis Ali Frihida....................................École Nationale d’Ingénieurs de Tunis 5 5 Comité d’organisation / Steering committee 25 Cartographie de la vulnérabilité côtière par l’intégration des méthodes géo-spatiales : cas de Mostaganem (ouest Algérien) Approches géomatiques pour la transition Françoise Gourmelon, LETG, Brest [Chat] Christine Plumejeaud, LIENSS, La Rochelle Benjamin Pillot, Nadeem Al-Kurdi, Carmen Gervet and Laurent Linguet.................................... 39 Optical network ring analogy for effective solar PV site selection using GIS and optimization modelling Antoine Mury, Antoine Collin and Dorothée James........................................................................ 51 Prospective 2050 : étude de cas sur l’évolution du service écosystémique de protection côtière Mathilde Letard, Antoine Collin, Dorothée James and Antoine Mury............................................ 63 Trajectoire socio-écologique d’une commune de la Baie du Mont-Saint-Michel entre 1947 et 2014 Environnement urbain María Jesús Lobo, LASTIG, Paris Jacques Gautier, LASTIG, Paris [Chat] Alain Bouju, L3I, La Rochelle Thomas Le Jan, Myriam Servières, Thomas Leduc and Vincent Tourre......................................... 77 Isovist computation of outdoor environment with semi-dense line SLAM and monocular camera Thomas Leduc, Xenia Stavropulos-Laffaille and Ignacio Requena-Ruiz....................................... 89 Implementation of a solar model and shadow plotting in the context of a 2D GIS: challenges and applications for the cooling effect of tree-covered based greening solutions in urban public spaces Joan Perez, Giovanni Fusco and Yukio Sadahiro........................................................................... 101 Spatial Evolution of Building Types: Methodological proposals Alessandro Araldi, David Emsellem, Giovanni Fusco, Andrea Tettamanzi and Denis Overal..... 113 Exploring building typologies through fast iterative Bayesian clustering Communications par Session Comité d’organisation / Steering committee Alain Bouju..................................EA L3I, La Rochelle Christine Plumejeaud-Perreau....UMR 7266 LIENSs, La Rochelle 6 6 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Table des matières / Contents Analyse d’image pour l’observation et le suivi du littoral Davis Sheeren, DYNAFOR, Toulouse [Chat] Christine Plumejeaud, LIENSS, La Rochelle Léa Paly, Véronique Beaujouan, Nathalie Carcaud and Lucien Rey................................................. 13 Construction des trajectoires paysagères de zones humides littorales – Apport de la géomatique à différentes échelles Soumia Bengoufa, Simona Niculescu, Mustapha Kamel Mihoubi, Rabah Belkessa and Katia Abbad..................................................................................................................................... 25 Cartographie de la vulnérabilité côtière par l’intégration des méthodes géo-spatiales : cas de Mostaganem (ouest Algérien) Approches géomatiques pour la transition Françoise Gourmelon, LETG, Brest [Chat] Christine Plumejeaud, LIENSS, La Rochelle Benjamin Pillot, Nadeem Al-Kurdi, Carmen Gervet and Laurent Linguet.................................... 39 Optical network ring analogy for effective solar PV site selection using GIS and optimization modelling Antoine Mury, Antoine Collin and Dorothée James........................................................................ 51 Prospective 2050 : étude de cas sur l’évolution du service écosystémique de protection côtière Mathilde Letard, Antoine Collin, Dorothée James and Antoine Mury............................................ 63 Trajectoire socio-écologique d’une commune de la Baie du Mont-Saint-Michel entre 1947 et 2014 Environnement urbain María Jesús Lobo, LASTIG, Paris Jacques Gautier, LASTIG, Paris [Chat] Alain Bouju, L3I, La Rochelle Thomas Le Jan, Myriam Servières, Thomas Leduc and Vincent Tourre......................................... 77 Isovist computation of outdoor environment with semi-dense line SLAM and monocular camera Thomas Leduc, Xenia Stavropulos-Laffaille and Ignacio Requena-Ruiz....................................... 89 Implementation of a solar model and shadow plotting in the context of a 2D GIS: challenges and applications for the cooling effect of tree-covered based greening solutions in urban public spaces Joan Perez, Giovanni Fusco and Yukio Sadahiro........................................................................... 101 Spatial Evolution of Building Types: Methodological proposals Alessandro Araldi, David Emsellem, Giovanni Fusco, Andrea Tettamanzi and Denis Overal..... 113 Exploring building typologies through fast iterative Bayesian clustering Communications par Session 7 Table des matières / Contents Analyse d’image pour l’observation et le suivi du littoral Davis Sheeren, DYNAFOR, Toulouse [Chat] Christine Plumejeaud, LIENSS, La Rochelle Léa Paly, Véronique Beaujouan, Nathalie Carcaud and Lucien Rey................................................. 13 Construction des trajectoires paysagères de zones humides littorales – Apport de la géomatique à différentes échelles Soumia Bengoufa, Simona Niculescu, Mustapha Kamel Mihoubi, Rabah Belkessa and Katia Abbad..................................................................................................................................... Environnement urbain María Jesús Lobo, LASTIG, Paris Jacques Gautier, LASTIG, Paris [Chat] Alain Bouju, L3I, La Rochelle 7 7 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Simulation, évaluation en géomatique Jean-Michel Follin, ESGT, Le Mans [Chat] Alain Bouju, L3I, La Rochelle Alice Jacquier and Carole Adam..................................................................................................... 127 Choix de mobilité : une approche SMA Robin Cura...................................................................................................................................... 139 Pour une évaluation visuelle des modèles de simulation, au-delà de la face validation Marie Havret, Marion Le Texier and Sophie de Ruffray............................................................... 151 Localiser le foncier en 1828 : l’exemple du cadastre ancien de Rouen Extraction de contenu pour le GeoWeb Boris Meriskay, Rennes [Chat] Paul-Annick Davoine, PACTE, Université de Grenoble-Alpes Rodrique Kafando, Rémy Decoupes, Lucile Sautot, Maguelonne Teisseire and Christiane Weber.... 165 Constitution de corpus thématique : Pour un meilleur suivi du territoire de la Métropole de Montpellier Méditerranée Marie Gradeler and Jean-Pierre Chery......................................................................................... 177 Géoweb de narration et de modélisation pour les territoires littoraux : la plateforme NAMO Géoweb Jacques Fize, Lucile Sautot, Martin Lentschat, Ludovic Journaux and Mohamed Hilal.............. 189 Grand débat national : une analyse automatique des contributions relatives aux transports Modélisation spatio-temporelle Camille Bernard, LIG, Grenoble [Chat] Paul-Annick Davoine, PACTE, Université de Grenoble-Alpes Kamaldeep Singh Oberoi and Géraldine Del Mondo.................................................................... 203 Graph-based Pattern Detection in Spatio-Temporal Phenomena Ba-Huy Tran, Alain Bouju, Christine Plumejeaud-Perreau and Vincent Bretagnolle................. 215 Une approche sémantique pour l’exploitation de données spatio-temporelles avec annotation Application à l’évolution d’assolement Diego Vinasco-Alvarez, John Samuel, Sylvie Servigne and Gilles Gesquière................................ 227 Towards a semantic web representation from a 3D geospatial urban data model Simulation, évaluation en géomatique Jean-Michel Follin, ESGT, Le Mans [Chat] Alain Bouju, L3I, La Rochelle Alice Jacquier and Carole Adam..................................................................................................... 127 Choix de mobilité : une approche SMA Robin Cura...................................................................................................................................... 139 Pour une évaluation visuelle des modèles de simulation, au-delà de la face validation Marie Havret, Marion Le Texier and Sophie de Ruffray............................................................... 151 Localiser le foncier en 1828 : l’exemple du cadastre ancien de Rouen Extraction de contenu pour le GeoWeb Boris Meriskay, Rennes [Chat] Paul-Annick Davoine, PACTE, Université de Grenoble-Alpes Rodrique Kafando, Rémy Decoupes, Lucile Sautot, Maguelonne Teisseire and Christiane Weber.... 165 Constitution de corpus thématique : Pour un meilleur suivi du territoire de la Métropole de Montpellier Méditerranée Marie Gradeler and Jean-Pierre Chery......................................................................................... 177 Géoweb de narration et de modélisation pour les territoires littoraux : la plateforme NAMO Géoweb Jacques Fize, Lucile Sautot, Martin Lentschat, Ludovic Journaux and Mohamed Hilal.............. Extraction de contenu pour le GeoWeb Extraction de contenu pour le GeoWeb Boris Meriskay, Rennes [Chat] Paul-Annick Davoine, PACTE, Université de Grenoble-Alpes Rodrique Kafando, Rémy Decoupes, Lucile Sautot, Maguelonne Teisseire and Christiane Weber.... 165 Constitution de corpus thématique : Pour un meilleur suivi du territoire de la Métropole de Montpellier Méditerranée Marie Gradeler and Jean-Pierre Chery......................................................................................... 177 Géoweb de narration et de modélisation pour les territoires littoraux : la plateforme NAMO Géoweb Jacques Fize, Lucile Sautot, Martin Lentschat, Ludovic Journaux and Mohamed Hilal.............. 189 Grand débat national : une analyse automatique des contributions relatives aux transports Modélisation spatio-temporelle Camille Bernard, LIG, Grenoble [Chat] Paul-Annick Davoine, PACTE, Université de Grenoble-Alpes Kamaldeep Singh Oberoi and Géraldine Del Mondo.................................................................... 203 Graph-based Pattern Detection in Spatio-Temporal Phenomena Ba-Huy Tran, Alain Bouju, Christine Plumejeaud-Perreau and Vincent Bretagnolle................. 215 Une approche sémantique pour l’exploitation de données spatio-temporelles avec annotation Application à l’évolution d’assolement Diego Vinasco-Alvarez, John Samuel, Sylvie Servigne and Gilles Gesquière................................ 227 Towards a semantic web representation from a 3D geospatial urban data model Environnement urbain 189 Grand débat national : une analyse automatique des contributions relatives aux transports Modélisation spatio-temporelle Camille Bernard, LIG, Grenoble [Chat] Paul-Annick Davoine, PACTE, Université de Grenoble-Alpes Kamaldeep Singh Oberoi and Géraldine Del Mondo.................................................................... 203 Graph-based Pattern Detection in Spatio-Temporal Phenomena Ba-Huy Tran, Alain Bouju, Christine Plumejeaud-Perreau and Vincent Bretagnolle................. 215 Une approche sémantique pour l’exploitation de données spatio-temporelles avec annotation Application à l’évolution d’assolement Diego Vinasco-Alvarez, John Samuel, Sylvie Servigne and Gilles Gesquière................................ 227 Towards a semantic web representation from a 3D geospatial urban data model Simulation, évaluation en géomatique Jean-Michel Follin, ESGT, Le Mans [Chat] Alain Bouju, L3I, La Rochelle Modélisation spatio-temporelle 8 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Heythem Adjeroud, Adrien Lammoglia and Laurent Chapelon.................................................... 241 Etude des pratiques de déplacement dans une ville moyenne algérienne : le cas de Mila Raffaella Balzarini, Estelle Ployon, Mélanie Marcuzzi and Philippe Bourdeau................................. 244 Observations d’évolutions socio-écosystémiques en haute montagne : l’apport de la géomatique au dispositif Refuges Sentinelles Rabia Sarah Cheriguene, Anne Puissant and Habib Mahi............................................................ 247 Apport des images ALSAT-2A pour la cartographie de l’occupation des sols - le cas de la ville d’Oran Antoine Collin, Tony Durozier, Gwenal Hervouet and Stéphane Costa...................................... 250 Détecter le processus de dépoldérisation par l’observation satellite à très haute résolution temporelle Matthieu Dubuis, Rémi Ratajczak, Lény Grassot, Laure Tougne and Béatrice Fervers............... 253 Détermination semi-automatique de l’occupation du sol grâce au logiciel GOURAMIC : évaluation des performances sur données historiques et opérateur-dépendance Amélie Duquesne, Christine Plumejeaud-Perreau and Jean-Michel Carozza.............................. 256 Application d’une approche semi- quantitative pour estimer les impacts cumulés des interven­ tions anthropiques sur les systèmes fluviaux : l’exemple de la Charente entre Angoulême et Saintes (Ouest de la France) Jacques Fize, Ludovic Moncla and Bruno Martins........................................................................ 259 Résolution de toponymes par apprentissage profond à partir de cooccurrences et de relations spatiales Jean-Michel Follin, Rovasoa Voniarinjanahary, Nathalie Thommeret and Marie Fournier...... 262 FarmSIG : une application web pour l’analyse de l’évolution des espaces agricoles en lien avec les politiques publiques environnementales Dorothée James, Antoine Collin and Mathilde Letard.................................................................. 265 Étude spatio-temporelle d’un écosystème littoral : le pré-salé de Sainte-Anne en baie du Mont- Saint-Michel Yousra Hamrouni, Olivia Bernardoff, Nicolas Karasiak and David Sheeren............................... 268 Adaptation de domaine par transport optimal pour la cartographie à large échelle des peupleraies Jérémy Kalsron, Jean-Marie Favreau and Guillaume Touya......................................................... 271 Le carrefour dont vous êtes le héros : description de carrefours pour les personnes déficientes visuelles Cécilia Pignon-Mussaud, Ahmed Laatabi, Marion Amalric, Brice Anselme, Elise Beck, Nicolas Marilleau and Nicolas Becu.............................................................................................. 274 Intégrer des données géospatiales dans un modèle multi-agents : cas de LittoSIM-GEN Louisa Le Rouzic, Bruno Gervais, Jean-François Breilh and Axel Creach..................................... 277 Mesure de la vulnérabilité des constructions résidentielles face au risque de submersion marine : proposition pour un outil opérationnel Arnaud Ménard, Vincent Lambert de Cursay, Cyril Ray, Charles Guenois, Maryam Maslek Elayam and Maxime Dréau......................................................................................................................... 280 Construction d’une méta-base de ports à l’échelle mondiale Posters Heythem Adjeroud, Adrien Lammoglia and Laurent Chapelon.................................................... Modélisation spatio-temporelle 241 Etude des pratiques de déplacement dans une ville moyenne algérienne : le cas de Mila Raffaella Balzarini, Estelle Ployon, Mélanie Marcuzzi and Philippe Bourdeau................................. 244 Observations d’évolutions socio-écosystémiques en haute montagne : l’apport de la géomatique au dispositif Refuges Sentinelles Rabia Sarah Cheriguene, Anne Puissant and Habib Mahi............................................................ 247 Apport des images ALSAT-2A pour la cartographie de l’occupation des sols - le cas de la ville d’Oran Antoine Collin, Tony Durozier, Gwenal Hervouet and Stéphane Costa...................................... 250 Détecter le processus de dépoldérisation par l’observation satellite à très haute résolution temporelle Matthieu Dubuis, Rémi Ratajczak, Lény Grassot, Laure Tougne and Béatrice Fervers............... 253 Détermination semi-automatique de l’occupation du sol grâce au logiciel GOURAMIC : évaluation des performances sur données historiques et opérateur-dépendance Amélie Duquesne, Christine Plumejeaud-Perreau and Jean-Michel Carozza.............................. 256 Application d’une approche semi- quantitative pour estimer les impacts cumulés des interven­ tions anthropiques sur les systèmes fluviaux : l’exemple de la Charente entre Angoulême et Saintes (Ouest de la France) Jacques Fize, Ludovic Moncla and Bruno Martins........................................................................ 259 Résolution de toponymes par apprentissage profond à partir de cooccurrences et de relations spatiales Jean-Michel Follin, Rovasoa Voniarinjanahary, Nathalie Thommeret and Marie Fournier...... 262 FarmSIG : une application web pour l’analyse de l’évolution des espaces agricoles en lien avec les politiques publiques environnementales Dorothée James, Antoine Collin and Mathilde Letard.................................................................. 265 Étude spatio-temporelle d’un écosystème littoral : le pré-salé de Sainte-Anne en baie du Mont- Saint-Michel Yousra Hamrouni, Olivia Bernardoff, Nicolas Karasiak and David Sheeren............................... 268 Adaptation de domaine par transport optimal pour la cartographie à large échelle des peupleraies Jérémy Kalsron, Jean-Marie Favreau and Guillaume Touya......................................................... 271 Le carrefour dont vous êtes le héros : description de carrefours pour les personnes déficientes visuelles Cécilia Pignon-Mussaud, Ahmed Laatabi, Marion Amalric, Brice Anselme, Elise Beck, Nicolas Marilleau and Nicolas Becu.............................................................................................. 274 Intégrer des données géospatiales dans un modèle multi-agents : cas de LittoSIM-GEN Louisa Le Rouzic, Bruno Gervais, Jean-François Breilh and Axel Creach..................................... 277 Mesure de la vulnérabilité des constructions résidentielles face au risque de submersion marine : proposition pour un outil opérationnel Arnaud Ménard, Vincent Lambert de Cursay, Cyril Ray, Charles Guenois, Maryam Maslek Elayam and Maxime Dréau......................................................................................................................... 280 Construction d’une méta-base de ports à l’échelle mondiale Posters Heythem Adjeroud, Adrien Lammoglia and Laurent Chapelon.................................................... 241 Etude des pratiques de déplacement dans une ville moyenne algérienne : le cas de Mila Raffaella Balzarini, Estelle Ployon, Mélanie Marcuzzi and Philippe Bourdeau................................. Modélisation spatio-temporelle 244 Observations d’évolutions socio-écosystémiques en haute montagne : l’apport de la géomatique au dispositif Refuges Sentinelles Rabia Sarah Cheriguene, Anne Puissant and Habib Mahi............................................................ 247 Apport des images ALSAT-2A pour la cartographie de l’occupation des sols - le cas de la ville d’Oran Antoine Collin, Tony Durozier, Gwenal Hervouet and Stéphane Costa...................................... 250 Détecter le processus de dépoldérisation par l’observation satellite à très haute résolution temporelle Matthieu Dubuis, Rémi Ratajczak, Lény Grassot, Laure Tougne and Béatrice Fervers............... 253 Détermination semi-automatique de l’occupation du sol grâce au logiciel GOURAMIC : évaluation des performances sur données historiques et opérateur-dépendance Amélie Duquesne, Christine Plumejeaud-Perreau and Jean-Michel Carozza.............................. 256 Application d’une approche semi- quantitative pour estimer les impacts cumulés des interven­ tions anthropiques sur les systèmes fluviaux : l’exemple de la Charente entre Angoulême et Saintes (Ouest de la France) Jacques Fize, Ludovic Moncla and Bruno Martins........................................................................ 259 Résolution de toponymes par apprentissage profond à partir de cooccurrences et de relations spatiales Jean-Michel Follin, Rovasoa Voniarinjanahary, Nathalie Thommeret and Marie Fournier...... 262 FarmSIG : une application web pour l’analyse de l’évolution des espaces agricoles en lien avec les politiques publiques environnementales Dorothée James, Antoine Collin and Mathilde Letard.................................................................. 265 Étude spatio-temporelle d’un écosystème littoral : le pré-salé de Sainte-Anne en baie du Mont- Saint-Michel Yousra Hamrouni, Olivia Bernardoff, Nicolas Karasiak and David Sheeren............................... 268 Adaptation de domaine par transport optimal pour la cartographie à large échelle des peupleraies Jérémy Kalsron, Jean-Marie Favreau and Guillaume Touya......................................................... 271 Le carrefour dont vous êtes le héros : description de carrefours pour les personnes déficientes visuelles Cécilia Pignon-Mussaud, Ahmed Laatabi, Marion Amalric, Brice Anselme, Elise Beck, Nicolas Marilleau and Nicolas Becu.............................................................................................. 274 Intégrer des données géospatiales dans un modèle multi-agents : cas de LittoSIM-GEN Louisa Le Rouzic, Bruno Gervais, Jean-François Breilh and Axel Creach..................................... 277 Mesure de la vulnérabilité des constructions résidentielles face au risque de submersion marine : proposition pour un outil opérationnel Arnaud Ménard, Vincent Lambert de Cursay, Cyril Ray, Charles Guenois, Maryam Maslek Elayam and Maxime Dréau......................................................................................................................... 280 Construction d’une méta-base de ports à l’échelle mondiale 9 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Jessica Pic, Etienne Cossart and Mathieu Fressard...................................................................... 283 Implications des infrastructures de drainage des versants dans la dynamique de transferts de sédiments et de produits phytosanitaires, le cas du Beaujolais viticole Marie Ramel-Delobel, Thomas Coudon, Lény Grassot, Delphine Praud, Béatrice Fervers and Pietro Salizzoni........................................................................................................................ Modélisation spatio-temporelle 286 Exposition à la pollution atmosphérique et risque de cancer du sein : vers une prise en compte des mobilités quotidiennes dans les études épidémiologiques Samuel Robert, Brieuc Cabioch, Eric Carroll and Marie-Laure Trémélo..................................... 289 Vers une détermination automatique du nombre d’usagers sur une plage urbaine De l’observation au modèle à base d’intelligence artificielle pour la plage du Prophète à Marseille Nicolas Robinet, Gaëtan Jolly, Guy Lempérière and Yves Petit-Berghem.................................... 292 Apport de l’utilisation du drone et de transects écopaysagers pour l’échantillonnage spatial d’habitats : le cas des digues de la vallée de la Dives (Calvados) Loic Salmon, Nicolas Maslov and Christophe Claramunt............................................................ 295 Processus de modélisation et optimisation pour la collecte de déchets plastiques en mer Magalie Técher, Hassan Ait Haddou and Rahim Aguejdad.......................................................... 298 Effets de la planification urbaine sur les températures de surface du sol (Montpellier, France) Nathalie Thommeret, Delphine Grancher, Aurélien Dousseron, Daniel Brunstein and Clément Virmoux.......................................................................................................................................... 301 L’état des piscines, construction d’un indicateur de la reprise du tourisme en situation post-catastrophe. L’île de Saint-Martin après le passage du cyclone Irma (sept 2017) Thi Huyen Tran, Marion Amalric, Frédéric Rousseaux and Benoit Gaudou................................. 304 Sensibilisation des gestionnaires à la prévention des risques liés à l’eau : modélisation, simulation et participation dans le Delta du Mékong Quy Thy Truong, Anne Ruas and Serge Lhomme.......................................................................... 307 Modélisation spatiale et positionnement de zones de dépôt pour la collecte de déchets post-ouragans Véronique Venturini and Caroline Tafani...................................................................................... 310 La planification littorale spatialisée de Balagne à l’épreuve de son opérationnalité : renouveler les méthodes d’identification des espaces vides Narimane Zaabar, Simona Niculescu, Mustapha Kamel Mihoubi and Soumia Bengoufa.......... 313 Méthodes de Machine Learning pour l’analyse de l’occupation du sol des zones exposées au risque d’inondation : la ville côtière d’Ain Témouchent (Ouest Algérien) Jessica Pic, Etienne Cossart and Mathieu Fressard...................................................................... 283 Implications des infrastructures de drainage des versants dans la dynamique de transferts de sédiments et de produits phytosanitaires, le cas du Beaujolais viticole Marie Ramel-Delobel, Thomas Coudon, Lény Grassot, Delphine Praud, Béatrice Fervers and Pietro Salizzoni........................................................................................................................ 286 Exposition à la pollution atmosphérique et risque de cancer du sein : vers une prise en compte des mobilités quotidiennes dans les études épidémiologiques Samuel Robert, Brieuc Cabioch, Eric Carroll and Marie-Laure Trémélo..................................... Modélisation spatio-temporelle 289 Vers une détermination automatique du nombre d’usagers sur une plage urbaine De l’observation au modèle à base d’intelligence artificielle pour la plage du Prophète à Marseille Nicolas Robinet, Gaëtan Jolly, Guy Lempérière and Yves Petit-Berghem.................................... 292 Apport de l’utilisation du drone et de transects écopaysagers pour l’échantillonnage spatial d’habitats : le cas des digues de la vallée de la Dives (Calvados) Loic Salmon, Nicolas Maslov and Christophe Claramunt............................................................ 295 Processus de modélisation et optimisation pour la collecte de déchets plastiques en mer Magalie Técher, Hassan Ait Haddou and Rahim Aguejdad.......................................................... 298 Effets de la planification urbaine sur les températures de surface du sol (Montpellier, France) Nathalie Thommeret, Delphine Grancher, Aurélien Dousseron, Daniel Brunstein and Clément Virmoux.......................................................................................................................................... 301 L’état des piscines, construction d’un indicateur de la reprise du tourisme en situation post-catastrophe. L’île de Saint-Martin après le passage du cyclone Irma (sept 2017) Thi Huyen Tran, Marion Amalric, Frédéric Rousseaux and Benoit Gaudou................................. 304 Sensibilisation des gestionnaires à la prévention des risques liés à l’eau : modélisation, simulation et participation dans le Delta du Mékong Quy Thy Truong, Anne Ruas and Serge Lhomme.......................................................................... 307 Modélisation spatiale et positionnement de zones de dépôt pour la collecte de déchets post-ouragans Véronique Venturini and Caroline Tafani...................................................................................... 310 La planification littorale spatialisée de Balagne à l’épreuve de son opérationnalité : renouveler les méthodes d’identification des espaces vides Narimane Zaabar, Simona Niculescu, Mustapha Kamel Mihoubi and Soumia Bengoufa.......... 313 Méthodes de Machine Learning pour l’analyse de l’occupation du sol des zones exposées au risque d’inondation : la ville côtière d’Ain Témouchent (Ouest Algérien) rticles et les posters ont été évalués par deux relecteurs du comité scientifique de SAGEO’2021. Les articles et les posters ont été évalués par deux relecteurs du comité scientifique de SAGEO’2021. 10 Communications par Session Analyse d’image pour l’observation et le suivi du littoral [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Le littoral, défini comme la zone d’interface entre l’espace continental et l’étendue maritime (Lévy et Lussault, 2013), est un espace flou et évolutif dans le temps court au gré des marées et dans le temps long notamment avec les variations du niveau de la mer (Lageat, 2019). Les espaces littoraux actuels résultent de la dernière transgression nommée transgression Flandrienne ayant eu lieu lors de l’Holocène (Verger, 2009). Aujourd’hui il est communément accepté par la sphère scientifique de l’entrée du monde dans une nouvelle ère géologique, l’Anthropocène, dont le forçage anthropique confère sa particularité (Crutzen, 2002). La date de début fait encore débat chez les scientifiques (années 1950, première révolution industrielle, Néolithique, etc.) (Le Gall et al., 2017) et varie surtout localement (Garnier, 2019). C'est pourquoi les espaces littoraux, notamment caractérisés par un changement de nature et d’intensité des pressions anthropiques, se doivent d’être étudiés au prisme des projets de territoire dont ils ont fait l’objet. En effet, ces espaces anciennement surnommés « territoires du vide » (Corlay, 1995), ont longtemps inspiré crainte et rejet. Cette dévaluation, plus prononcée sur les zones humides littorales, a longtemps justifié des aménagements conséquents (Sajaloli, 1996). A l’issue d’une période hygiéniste, les sociétés se sont finalement rapprochées du rivage (Corbin, 1988). L’économie littorale s’est progressivement tournée vers le développement touristique et portuaire (Miossec, 2001). Les interactions entre la Société et la Nature sont particulièrement bien illustrées sur les zones humides littorales. Ces espaces ont notamment subi depuis le Moyen- Age de nombreux aménagements (drainage, endiguement, poldérisation, etc.) à des fins d’usages anthropiques (agriculture, pisciculture, urbanisation…). Depuis les années 1960-1970, de nouveaux savoirs et regards ont conduit à la protection de ces zones humides (Poncet et al., 1979 ; Goeldner-Gianella et al., 2011). Les trajectoires paysagères de ces espaces sont donc rythmées entre faits de Nature et faits de Société. Aujourd’hui le dérèglement climatique et l’élévation du niveau la mer, estimée à 0,4m – 0,6m d’ici 2100 (Horton et al., 2014), constituent un réel enjeu pour les zones humides littorales et leur paysage, dont les caractéristiques topographiques sont proches du niveau marin. Face à cet enjeu majeur, le développement territorial se doit d’avoir une vision prospective de ces espaces. 1 Baie de Lancieux - adapto, un projet LIFE : https://www.lifeadapto.eu/baie-de- lancieux.html - ©2020 Le Conservatoire du littoral Léa Paly1, Véronique Beaujouan2, Nathalie Carcaud3, Lucien Rey4 1. ESO – UMR 6590 CNRS, BAGAP, Institut Agro, Agrocampus-Ouest 2 rue André Le Nôtre, 491000, Angers, France lea.paly@agrocampus-ouest.fr 2. UMR BAGAP, Institut Agro – Agrocampus-Ouest, centre d’Angers 3. ESO – UMR 6590 CNRS, Institut Agro, Agrocampus-Ouest, centre d’Angers 4. Master 2 GAED, mention Aménagement et Développement des Espaces Insulaires, Université des Antilles RESUME. Les espaces littoraux sont l’objet de nombreux enjeux. Avec le changement climatique, les collectivités se doivent d’entreprendre des approches prospectives du développement territorial. Celles-ci doivent se nourrir de l’étude des territoires, de leurs trajectoires et des temporalités de changements. L’objectif de cette recherche est de construire les trajectoires paysagères du littoral Atlantique à l’aide d’analyses multivariées et de classifications. Les trajectoires à l’échelle régionale sont confrontées aux trajectoires à l’échelle locale, approfondies par l’analyse des changements d’occupation du sol. Cette comparaison a pour objectif de montrer la complémentarité d’analyse des trajectoires de zones humides littorales apportée par l’emboitement des échelles. ABSTRACT. Currently, coastal areas face many challenges. With climate change, local authorities have a duty to undertake forward-looking approaches of territorial development. This must be based on the study of territories, their trajectories, and temporalities of change. The purpose of this research is to construct landscape trajectories of the Atlantic coast using multivariate analysis and classification. Trajectories at the regional scale are compared with trajectories at local scale, deepened by analysis of changes in land use and land cover. The aim of this comparison is to show the complementary analysis provided by the nesting of scales for the study of coastal wetland trajectories. MOTS-CLES : littoral, zone-humide, paysage, trajectoires, analyses multivariees, classification, transition, occupation du sol. MOTS-CLES : littoral, zone-humide, paysage, trajectoires, analyses multivariees, classification, transition, occupation du sol. KEYWORDS: littoral, wetlands, landscape, trajectories, multivariate analysis, classification, transition, land-use / land-cover 13 Communications Communications 13 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 2 La définition du paysage ici reprend celle de la Convention Européenne du Paysage. Celui-ci est défini comme « la partie du territoire telle que perçu par les populations et dont le caractère résulte de la combinaison de facteurs naturels et/ou humains et de leurs interrelations » (art.1a, convention européenne du paysage, 2000). Trajectoires paysagères littorales et géomatique Cette recherche cherche à construire les trajectoires paysagères2 objectives de huit zones humides littorales de 1977 à 2000 à l’aide de l’outil cartographique et de données en libre accès. L’objectif est de pouvoir identifier les rythmes de changement de ces zones humides en partant de l’hypothèse que chacune de ces zones humides a eu des formes de rupture de sa trajectoire paysagère homogènes mais à des temporalités et des intensités différentes et que ces trajectoires homogènes à l’échelle régionale se distinguent à l’échelle locale. Une analyse sur les changements paysagers à l’échelle régionale viendra éclairer les grandes tendances d’évolution de ces espaces des années 1970 à aujourd’hui. Ces évolutions seront ensuite explorées à l’échelle locale par la détermination des trajectoires paysagères du site de Corsept (Loire-Atlantique). 1. Introduction De nombreux acteurs, dont le Conservatoire du littoral, se sont saisi de ce sujet avec par exemple le programme ADAPTO qui préconise une gestion douce du trait de côte et au sein duquel les restitutions de terres prises à la mer sont envisagées comme scénario d’évolution future1. La connaissance de l’évolution passée de chaque espace est de ce fait essentielle pour le développement d’une approche prospective, d’autant plus que chaque territoire possède ses particularités. 14 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Trajectoires paysagères littorales et géomatique Trajectoires paysagères littorales et géomatique 2.1. Contexte de l’étude Les zones humides analysées pour cette étude font partie d’un ensemble de 8 sites pilotes répartis sur la côte Atlantique (FIG 1.), étudiés dans le cadre d’un travail de doctorat. Ils ont pour point commun la présence d’un patrimoine naturel protégé actuellement qui est hérité d’aménagements anciens. FIGURE 1. Situation géographique des huit sites pilotes étudiés Aber en Crozon Anse de Pouldon et polder de Combrit Marais de Mousterlin Etangs et anse de Penfoulic Marais de Pen an Toul Marais de Corsept Marais du Payré FIGURE 1. Situation géographique des huit sites pilotes étudiés La construction des trajectoires paysagères des sites retenus repose sur la mobilisation de trois approches (FIG 2.). L’approche historique à l’aide d’archives et notamment de cartes et plans anciens apporte une compréhension de l’état passé 15 Communications Communications 15 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] d’un site et, couplée à l’approche géoenvironnementale (décrite ci-dessous), permet de construire les trajectoires paysagères. Enfin l’approche émique qui se base sur le discours et les représentations des acteurs sociaux (De Sardan, 1998) utilise les entretiens pour aborder les représentations paysagères qu’ont les usagers des sites. Les rythmes de changements de ces paysages sont ainsi identifiés en partant de l’hypothèse que les moments de rupture sont différents notamment dans leur temporalité. FIGURE 2. Schéma des trois approches utilisées FIGURE 2. Schéma des trois approches utilisées Cette étude présente les premiers résultats issus de l’approche géo- environnementale qui a pour objectif de mettre en lumière les grandes classes d’évolution à l’échelle régionale depuis les années 1970 et de les confronter aux spécificités locales actuelles des sites. Avec le développement numérique, les Systèmes d’Information Géographiques deviennent des outils incontournables pour l’étude du paysage (Brossard & Wieber, 1984). A partir de données d’occupation du sol et l’utilisation d’indicateurs paysagers, de nombreuses études font usage d’analyses statistiques multivariées aboutissant à des classifications de paysages et d’évolution (Guneroglu et al., 2015 ; Weng, 2007) ainsi qu'à la caractérisation de la succession de ces paysages le long de gradients (Modica et al., 2012; Vizzari et al., 2018; Vizzari & Sigura, 2015). Les matrices de transition ont aussi démontré leur intérêt vis-à-vis d’études spatio-temporelles pour décrire et analyser les changements surfaciques de la végétation (Delbosc et al., 2018; Dez et al., 2017) ou d’occupation du sol (Frondoni et al., 2011; Baud et al., 2015). 2.1. Contexte de l’étude Elles apportent en particulier une représentation quantitative et qualitative des trajectoires identifiées. 2.2. Analyse cartographique diachronique Cette recherche s’appuie sur des classifications des types d’évolution du paysage dans le temps à l’échelle régionale et à l’échelle locale ; ainsi que sur l’utilisation de matrices de transition. Chaque analyse réalisée à l’échelle locale est étudiée par le 16 Communications 3 La couche LITTOMOS a été réalisée en 2000 pour le Finistère et Morbihan, 2004 pour la Loire-Atlantique et 2006 pour la Vendée. Communications 16 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Trajectoires paysagères littorales et géomatique Trajectoires paysagères littorales et géomatique site des marais de Corsept. La connaissance historique et locale de ce site permet d’avoir un recul critique sur la détermination des trajectoires paysagères. site des marais de Corsept. La connaissance historique et locale de ce site permet d’avoir un recul critique sur la détermination des trajectoires paysagères. 4 Les cartes d’occupation du sol ont été rastérisées à 12.5m dans le but de garder le détail des éléments de faibles surfaces et d’optimiser les temps de calculs. 2.2.3. Analyse régionale : classification de l’évolution des paysages La première analyse s’est appuyée sur la différence des valeurs des différents indicateurs par maille entre deux dates pour étudier l’évolution du paysage. La réalisation d’une analyse en composante principale (ACP) suivie d’une classification ascendante hiérarchique (CAH) a permis d’obtenir une classification des dynamiques paysagères observées à l’échelle régionale entre 1977 et 2000, 2000 et 2013. La classification utilisée, qualifiée de bottom-up, est choisie pour son opérationnalité. Une classification top-down, où les classes de paysages sont prédéterminées, aurait pu être utilisée mais une connaissance préalable des classes de paysages littoraux aurait été nécessaire ; de plus, la différence des résultats n’est pas conséquente (Van der Zanden et al., 2016). 2.2.2. Calcul des indicateurs paysagers 2.2.2. Calcul des indicateurs paysagers La construction des trajectoires paysagères s’est établie à travers trois analyses opérant à différentes échelles spatiales. Pour chaque analyse, l’outil statistique R, avec les packages FactoMineR pour les ACPs et networkD3 pour les diagrammes de Sankey, a été utilisé ainsi que les logiciels de cartographie QGis et ArcGis. Chaque carte d’occupation du sol a été rastérisée à une résolution de 12,5m4 à l’échelle régionale puis analysée avec le logiciel Chloé (Boussard et Baudry, 2017) pour en extraire des indicateurs paysagers (TABLE 2.). Des indicateurs de structure ont été calculés à l’échelle de mailles de 700m à l’échelle régionale et de 100m à l’échelle locale, tels que la pente moyenne et son écart-type, l’altitude moyenne et son écart-type, l’indice de position topographique (étendue), la distance minimale de la maille au trait de côte, la distance minimale de la maille au cours d’eau et la proportion de la maille depuis laquelle on voit le trait de côte. Ensuite des indicateurs de composition (taille moyenne à l’échelle des taches et du paysage et leur écart-type, pourcentage des différents types d’OCS, nombre de taches en eau, indice de diversité de Shannon, pourcentage de couples de pixels en « milieux humides ») et de forme (longueur d’interface entre différents type d’OCS, indice d’agrégation des taches) ont été calculés avec le plugin Chloé sur des fenêtres d’analyse de 2500m à l’échelle régionale et de 1000m à l’échelle locale. 2.2.1. Homogénéisation et traitement des bases de données spatiales Nomenclature des différents postes d’occupation du sol aux trois dates ANALYSE REGIONALE ANALYSE LOCALE 11 Zone urbanisées et espaces associés 11 Zone urbanisées et espaces associés 62 Milieux ouverts agricoles 20 Cultures 23 Prairies 71 Milieux semi-ouverts 71 Milieux semi-ouverts 31 Milieux boisées 31 Milieux boisés 60 Milieux ouverts minéraux 61 Milieux ouverts 63 Milieux ouverts herbacées 64 Milieux ouverts humides 40 Milieux humides 42 Production maritime 50 Surface en eau 50 Surface en eau 17 17 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 2.2.1. Homogénéisation et traitement des bases de données spatiales 2.2.1. Homogénéisation et traitement des bases de données spatiales Cette étude mobilise trois couches d’occupation du sol en accès libre datant de 1977, 20003 et 2013. Leur comparaison, en raison de nomenclatures thématiques différentes, a nécessité la création d’une nomenclature commune aux trois couches. La couche d’occupation du sol IPLI de 1977 constitue le premier inventaire de l’occupation du sol du littoral sur l’ensemble du territoire national à l'échelle 1/25000ème. Sa mise à jour, LITTOMOS, dans les années 2000 à 2006 reprend la même échelle dans le but de faciliter les analyses temporelles. Ces deux sources de données ont été élaborées par la photo-interprétation de photographies aériennes sur une bande d’environ 7km sur l’ensemble du littoral métropolitain. Enfin, l’occupation du sol à grande échelle produite en 2013 pour la Loire-Atlantique, la Vendée et le Morbihan au 1/5000ème a été exploitée. Les descriptions de la couverture du sol et de l’usage du sol ont été couplées pour obtenir une couche d’occupation du sol (OCS) comparable aux précédentes sources pour ces trois départements. L’utilisation de la couche OSO créée par Copernicus, a permis de discriminer les prairies et cultures. Les couches IPLI et LITTOMOS présentent des disparités régionales en termes de nomenclature avec le regroupement pour IPLI des prairies et cultures sous la catégorie terres cultivées pour le Finistère (Sparfel, 2011). Ainsi la nomenclature de correspondance employée possède deux niveaux de détails. L’un, le plus large, sera adopté à l’échelle régionale, et le second, plus détaillé par suite d’une rectification des couches IPLI et LITTOMOS par photo- interprétation, sera utilisé à l’échelle locale. Cette nomenclature (TABLE 1.), gage de la correspondance entre les trois couches, comprend au maximum 9 postes décrivant les espaces urbanisés, les espaces naturels, les milieux humides, agricoles et forestiers. TABLE 1. Nomenclature des différents postes d’occupation du sol aux trois dates TABLE 1. Nomenclature des différents postes d’occupation du sol aux trois dates ANALYSE REGIONALE ANALYSE LOCALE TABLE 1. 2.2.5. Analyse locale : évolution des paysages à l’échelle des sites : cas de Corsept La dernière analyse consiste en l’examen de l’évolution du paysage à l’échelle des sites et est étudiée par le cas des marais de Corsept. Cette méthode permet de visualiser les trajectoires des sites sur le graphique des individus de l’ACP en isolant les points composant un site en 1977, 2000 et 2013. Cette étude a été entreprise afin de comprendre la trajectoire spécifique de ce site par analogie aux grandes classes d’évolution caractérisée à l’échelle régionale. Une ACP, dite historique, a été réalisée sur un tableau présentant en individus les mailles de la grille correspondant aux sites pour chacune des trois dates. L’ensemble de ces individus a été caractérisé par les mêmes indicateurs paysagers qu’à l’échelle régionale. 3. Premiers résultats Ces résultats sont actuellement en cours de construction et d’interprétation. D’ici mai 2021, l’ensemble des analyses aura été effectuée et approfondie. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Trajectoires paysagères littorales et géomatique Trajectoires paysagères littorales et géomatique Trajectoires paysagères littorales et géomatique d’Intérêt Faunistique et Floristique) de ce site. Après agrégation de ces trois couches, la nature des transitions entre deux dates et pour chacun des postes d’OCS est connue et exprimée en pourcentage absolu de surface passant d’un poste à un autre. La matrice de transition ainsi constituée est ensuite modélisée par un diagramme de Sankey. Ce même procédé a été répété sur un périmètre élargi par des zones tampon de 250m, 500m et 1000m afin de comparer les changements d’occupation du sol au voisinage proche de la zone humide 2.2.4. Etude des changements d’occupation du sol à l’échelle des sites : cas de Corsept Cette analyse vient alimenter les observations des changements de paysage par la détermination des changements d’occupation du sol (OCS) à l’échelle locale. Les OCS aux trois dates ont été découpées selon le périmètre du site de Corsept. Celui-ci a été délimité selon la géomorphologie du site et le zonage de l’espace reconnu pour son caractère naturel, attribué au périmètre des ZNIEFFs 1 et 2 (Zone Naturelle 18 Communications Communications 18 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3.1. Typologie de l’évolution des paysages à l’échelle régionale La première analyse réalisée sur une ancienne nomenclature entre 2000 et 2013 a identifié cinq grands types d’évolution des paysages à l’échelle régionale (FIG 3.). FIGURE 3. Dendrogramme issu de la CAH réalisée sur l’ACP concernant les évolutions entre 2000 et 2013, 5 classes sont retenues FIGURE 3. Dendrogramme issu de la CAH réalisée sur l’ACP concernant les évolutions entre 2000 et 2013, 5 classes sont retenues 19 19 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] FIGURE 4. Typologie des évolutions des paysages entre 2000 et 2013 pour les quatre départements et le site de Corsept (Loire-Atlantique) FIGURE 4. Typologie des évolutions des paysages entre 2000 et 2013 pour les quatre départements et le site de Corsept (Loire-Atlantique) La classification fait ressortir une évolution propre à l’estuaire de la Loire avec la classe 5 entre 2000 et 2013 (FIG 4.). La première lecture des ACPs montre des processus généraux tels que l’intensification agricole et l’urbanisation mais aussi le verdissement des espaces qui deviennent plus « naturels » pouvant révéler une volonté politique de renaturation ou bien une déprise ou spécialisation agricole. Il est difficile à ce jour de plus s’avancer sur les résultats, l’interprétation étant encore en cours. Ces première analyses, d’ici Mai, seront approfondis notamment dans le choix d’indicateurs paysagers spécifiques à l’évolution de ces espaces grâce à l’élaboration des matrices de transition à l’échelle régionale. 3.2. Matrices de transition de l’occupation du sol de Corsept Les représentations des matrices de transition montrent que certains types d’occupation du sol ont plus varié que d’autres entre 1977 et 2013 (FIG 5, a. et b.). Les forêts ont augmenté entre 1977 et 2000 avant de diminuer en 2013 au profit des zones urbanisées. Les espaces naturels ont augmenté. Les zones en eau sont restées stables avec néanmoins une proportion plus importante en 1977. Les espaces urbanisés se sont légèrement accrus à chaque année (FIG 5. a.), et ont doublé entre 1977 et 2000 sur le périmètre élargi par la zone-tampon (FIG 5. b.) Les espaces de culture, en prairie et les zones humides représentent les plus grandes variations. Sur 20 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Trajectoires paysagères littorales et géomatique le périmètre de la zone humide, les prairies ont conséquemment augmenté entre 1977 et 2000 et de façon plus minime de 2000 à 2013 par transformation des zones humides principalement entre 1977 et 2000 et des cultures (1977 à 2013). Après les années 2000 la surface des zones humides reste stable. La même tendance d’évolution des prairies ressort sur le périmètre élargi par la zone tampon avec néanmoins un pourcentage de prairies à chaque année moins important, contrebalancé par une part de culture plus grande. FIGURE 5. Transition entre les postes d’OCS entre 1977, 2000 et 2013 pour (a.) le périmètre du marais du Corsept, (b) étendu avec une zone tampon de 500m FIGURE 5. Transition entre les postes d’OCS entre 1977, 2000 et 2013 pour (a.) le périmètre du marais du Corsept, (b) étendu avec une zone tampon de 500m 3.3. Evolution des paysages à l’échelle du site de Corsept (Loire-Atlantique) L’ACP historique apportera une lecture « paysagère » de ces évolutions entre 1977, 2000 et 2013 à l’aide des différents indicateurs décrits pour l’analyse régionale et adaptés aux spécificités locales du site. Celle-ci est en cours de réalisation. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] humides à la catégorie prairie de 1977 à 2000 avant de se stabiliser. Il est important de prendre en compte le fait que la majeure partie du site en 1977 était composée de prairies humides poste d’OCS qui a été intégré à la catégorie zone humide. Ainsi l’augmentation des prairies entre 1977 et 2000 est à relativiser et peut finalement caractériser l’assèchement des prairies humides présentes en 1977. Cette période correspond à la construction d’une digue et d’une douve collectrice des eaux des étiers en 1988-89 altérant alors les échanges entre la zone soumise aux marées et ces marais estuariens. Enfin, la connaissance du terrain est indispensable pour ce genre d’analyse. La transition des cultures vers les prairies peut traduire un phénomène de déprise ou de spécialisation agricole mais aussi un phénomène de patrimonialisation. En effet, la mise en place de mesures de gestion sur les zones humides implique souvent le maintien d’un paysage ouvert sous forme de prairies de fauche ou de pâturage. Ici, bien que le processus de patrimonialisation de cette zone humide et plus généralement de l’estuaire de la Loire a débuté avant les années 2000, les conséquences de cette patrimonialisation sur le site de Corsept sous forme de pratiques de gestion sont bien plus tardives. Ainsi cette évolution pourrait traduire plutôt un phénomène de déprise ou de spécialisation agricole pour ce site. On le comprend, les matrices de transition ont permis de révéler des erreurs de correspondance dans les nomenclatures des couches cartographiques. Malgré ces contraintes, l’utilisation de couches de données en libre accès permet la reproductibilité de ces analyses. Les premiers résultats suggèrent une trajectoire de la zone humide de Corsept qui s’inscrit dans une trajectoire globale de changements des pratiques culturales (augmentation des prairies au détriment des cultures, majoritairement) mais qui se distingue à l’échelle de l’espace protégé (peu voire pas d’urbanisation, part de prairies beaucoup plus importante, peu de culture), notamment en lien avec ses spécificités historiques (construction d’une digue, patrimonialisation récente). Les matrices de transition, couplées à une connaissance historique du terrain, offrent une analyse fine des trajectoires de ces espaces. Elles permettent à la fois une analyse quantitative en modélisant les pourcentages absolus des transitions d’OCS, qualitative par la connaissance des successions d’OCS mais aussi spatialisée avec l’étude de périmètres différents. 4. Discussion Les premiers résultats de ces analyses (en cours de réalisation et d’interprétation) soulèvent plusieurs questionnements au niveau de l’adéquation de la méthode. Une réduction des indicateurs utilisés est nécessaire afin de sélectionner les indicateurs discriminant les évolutions sur les espaces littoraux. L’usage de couches d’occupation du sol élaborées par différents acteurs et en libre accès apporte la simplicité d’accès mais demeure contraignant quant à leur mise en comparaison. Les méthodes de classification automatiques sont intéressantes, notamment sur de grands jeux de données, cependant leur utilisation doit être précautionneuse. La connaissance des variations des indicateurs utilisés est nécessaire pour mener cette analyse de façon optimale. De plus, l’interprétation doit prendre en compte l’imperfection des tables de correspondance, particulièrement pour les matrices de transition. La figure 5 (a.) a témoigné sur le site de Corsept d’un apport des zones 21 21 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Trajectoires paysagères littorales et géomatique Trajectoires paysagères littorales et géomatique Trajectoires paysagères littorales et géomatique de territoire ont souvent voulu appliquer un même projet de développement aux zones humides littorales comme l’ont montré les nombreux manuscrits techniques sur le dessèchement des marais du XIXème siècle (Dienne, 1891) et début XXème siècle (Lévy-Salvador, 1900). Aujourd’hui encore, les grands projets de restauration/renaturation des zones humides littorales peuvent conduire à banaliser ces paysages. La connaissance des héritages paysagers par la reconstitution des trajectoires paysagères permet de reconnaitre les spécificités de ces territoires. L’outil cartographique et les méthodes de classification sont un bon moyen de déterminer des classes de trajectoires à des échelles spatiales différentes. Il serait néanmoins intéressant d’utiliser des données plus anciennes, notamment du début du XXème siècle avant le grand développement touristique des espaces côtiers. Cela permettrait de mieux cerner les moments de rupture liés à une pression anthropique trop importante et de les comparer aux autres sites. Pour les marais de Corsept ce moment de rupture est récent et peu conséquent en termes paysagers alors que pour les marais du Payré (85), par exemple, ce moment était déjà initié au XIIème siècle par la création des marais salants et à poissons. L’objectif de cette recherche est de poursuivre ces trajectoires du XVIIIème au XXIème siècles grâce à la numérisation de cartes anciennes. 5. Conclusion Au vu de ces premiers résultats, on comprend qu'il est important et nécessaire de prendre en compte l’histoire du paysage. Ce recul permet d’avoir un regard critique sur l’observation de phénomènes actuels. Ce regard est d’autant plus important sur l’espace littoral qui a été profondément modifié par l’Homme. Les sites pilotes identifiés peuvent à travers leur typologie présenter un paysage similaire. Or, en se penchant sur leur trajectoire paysagère, à travers l’évolution des paysages à l’échelle du site et son couplage à l’étude des transitions d’usage du sol, on s’aperçoit de la diversité dynamique de ces espaces. Ce travail montre aussi l’intérêt de varier les échelles d’analyse et la complémentarité des échelles régionales et locales pour comprendre les trajectoires paysagères des espaces littoraux. Hier, les grands projets 22 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Bibliographie Boussard H, Baudry J. 2017 Chloe4.0: a software forlandscape pattern analysis. See https://www6.rennes.inra.fr/bagap/PRODUCTIONS/Logiciels. Brossard, T., & Wieber, J.-C. (1984). Le paysage : Trois définitions, un mode d’analyse et de cartographie. Espace géographique, vol. 13, n°1, 5‑12. Corbin A., 1988. Le territoire du vide. L’occident et le désir de rivage. Flammarion, Champs histoire, 2018, 399p Corlay, J.-P. (1995). Géographie sociale, géographie du littoral. Norois, vol. 165, n°1, 247‑265. Crutzen, P. J. (2002). Geology of mankind The Anthropocene. Nature, vol. 415, n°23, 1 Delbosc, P., Tanné, M., Bioret, F., Panaïotis, C., & Sawtschuk, J. (2018). Analyse spatio- temporelle du paysage végétal du massif du Haut-Vénacais (Corse). Cybergeo, n°846, 1- 30 De Sardan, J.-P. O. (1998). Émique. L’Homme, 38(147), 151‑166. https://doi.org/10.3406/hom.1998.370510 Dez, M. L., Sawtschuk, J., & Bioret, F. (2017). Les prairies de l’estuaire de la Loire : Étude de la dynamique de la végétation de 1982 à 2014. Mappemonde, n°119. Dienne, É. de (Cte). (1891). Histoire du dessèchement des lacs et marais en France avant 1789. Bibliothèque Nationale de France. Frondoni, R., Mollo, B., & Capotorti, G. (2011). A landscape analysis of land cover change in the Municipality of Rome (Italy) : Spatio-temporal characteristics and ecological implications of land cover transitions from 1954 to 2001. Landscape and Urban Planning, vol. 100, 117‑128. 23 23 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Garnier, E. (2019). Trajectoires d’une anthropisation dans la vallée du Doubs, XVIe-XXIe siècles. Revue scientifique Bourgogne-Franche-Comté Nature, vol. 30, 116‑139. Goeldner-Gianella, L., Feiss-Jehel, C., & Decroix, G. (2011). Les oubliées du « désir du rivage » ? L’image des zones humides littorales dans la peinture et la société françaises depuis le XVIIIe siècle. Cybergeo. Guneroglu, N., Acar, C., Guneroglu, A., Dihkan, M., & Karsli, F. (2015). Coastal land degradation and character assessment of Southern Black Sea landscape. Ocean & Coastal Management, vol. 118, 282‑289. Horton, B. P., Rahmstorf, S., Engelhart, S. E., & Kemp, A. C. (2014). Expert assessment of sea-level rise by AD 2100 and AD 2300. Quaternary Science Reviews, vol. 84, 1‑6. Le Gall, J., Hamant, O., & Bouron, J.-B. (2017). Notion en débat : Anthropocène - Géoconfluences [Document]. Géoconfluences. http://geoconfluences.ens- lyon.fr/informations-scientifiques/a-la-une/notion-a-la-une/anthropocene Lévy-Salvador, P. (1900). Hydraulique agricole—LIVRE III - 4e à 8e PARTIE - Assainissements et dessèchements. Colmatages. Polders. Drainage. Utilisation agricole des eaux d’égout. Annexes. Vve CH., EDITEUR, 3(4‑8), 574. Miossec, A. (2001). Bibliographie L’évolution de la géographie des océans et des littoraux face aux perspectives du développement durable au XXIe siècle. Quelles hypothèses envisager?//The geography of coasts and seas facing sustainable development : What could be the future? Annales de Géographie, vol. 110, n°621, 509‑526. Modica, G., Vizzari, M., Pollino, M., Fichera, C. R., Zoccali, P., & Di Fazio, S. (2012). Spatio-temporal analysis of the urban-rural gradient structure : An application in a Mediterranean mountainous landscape (Serra San Bruno, Italy). Earth System Dynamics Discussions, vol. 3, n°2, 827‑870. Sajaloli, B. (1996). Les zones humides : Une nouvelle vitrine pour l’environnement (Wetlands : a new showcase for environment). Bulletin de l’Association de géographes français, vol. 73, n°2, 132‑144. Sparfel, L. (2011). Etude des changements d’occupation des sols dans la zone côtière à partir de données hétérogènes : Application au pays de Brest. [Sciences de l’Homme et Société / Géographie]. Université de Bretagne occidentale - Brest. Van der Zanden, E. H., Levers, C., Verburg, P. H., & Kuemmerle, T. (2016). Representing composition, spatial structure and management intensity of European agricultural landscapes : A new typology. Landscape and Urban Planning, vol. 150, 36‑49. Verger F. (2009). Zones humides du littoral français. BELIN, Paris, 447p. Vizzari, M., Hilal, M., Sigura, M., Antognelli, S., & Joly, D. (2018). Urban-rural-natural gradient analysis with CORINE data : An application to the metropolitan France. Landscape and Urban Planning, vol. 171, 18‑29. Vizzari, M., & Sigura, M. (2015). Landscape sequences along the urban–rural–natural gradient : A novel geospatial approach for identification and analysis. Landscape and Urban Planning, vol. 140, 42‑55. Weng, Y.-C. (2007). Spatiotemporal changes of landscape pattern in response to urbanization. Landscape and Urban Planning, vol. 81, n°4, 341‑353. 24 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 KEYWORDS: Vulnerability Index, West-Algeria, Machine Learning, GIS, Erosion, Soumia BENGOUFA 1, 2,3*, Simona NICULESCU 2, Mustapha Kamel MIHOUBI 2, Rabah BELKESSA 1, Katia ABBAD 1 1. Ecole Nationale Supérieure des Sciences de la Mer et de l’Aménagement du Littoral, Dely Ibrahim, Alger, Algérie, bengoufasoumia@gmail.com, belkessa.r.littoral@gmail.com, abbadkatia@gmail.com, 1. Ecole Nationale Supérieure des Sciences de la Mer et de l’Aménagement du Littoral, Dely Ibrahim, Alger, Algérie, bengoufasoumia@gmail.com, belkessa.r.littoral@gmail.com, abbadkatia@gmail.com, RESUME. Dans le contexte de la croissance d'activités socio-économiques littorales et l’accentuation des risques côtiers, ce travail vise à évaluer la vulnérabilité d’un linéaire côtier de 32 km de la ville de Mostaganem (l’Ouest-Algérien). En utilisant des méthodes de télédétection et des Système d'Information Géographique (SIG), cinq variables physiques et socio-économiques, à savoir le taux d'érosion (trait de côte), le type du relief, la pente, la hauteur significative des vagues (Hs) et le type d’occupation du sol, sont traitées. Un modèle de SIG a été créé pour calculer un Indice de Vulnérabilité Côtière (IVC) en utilisant une méthode de pondération. Sur la base des valeurs de l'IVC, la côte est catégorisée en quatre classes de vulnérabilité, à savoir : faible (7,5 km), modérée (4.08 km), élevée (11 km), et très élevée (9,74 km). Cette étude pourrait contribuer à la cartographie de la vulnérabilité de la zone d’étude dans une optique de mitigation et prévention du risque de l’érosion côtière. ABSTRACT. In the context of the increasing rate of coastal socio-economic activities and the exacerbating of the coastal hazards, this work aims to assess the coastal vulnerability of 32 km of the Mostaganem coastline (Algeria), using remote sensing methods and Geographic Information Systems (GIS). To achieve that, five variables, namely erosion rate (shoreline), relief type, slope, significant wave height (Hs) and land use type, are considered. A GIS model was created to calculate a Coastal Vulnerability Index (CVI) based on weighting method. According to the CVI values, the shoreline is classified into four vulnerability classes, LOW (42.5 km), MODERATE (23.46 km), HIGH (34.61 km), and VERY HIGH (7.5 km). Thus, this study is useful for risk mitigation and prevention of coastal erosion hazard. MOTS-CLES : Indice de Vulnérabilité, Ouest-algérien, Machine Learning, SIG, Erosion 1. Introduction Les côtes sont des reliefs dynamiques, dont les changements dépendent des facteurs physiques qui les caractérisent (géologie, géomorphologie, etc.) et/ou du forçage de la mer (l'action des vagues, niveau de la mer, etc.). Compte tenu du fait que 21 % de la population mondiale y vit (Brooks et al., 2006), ces côtes sont sujettes à des concentrations importantes et croissantes d'activités et d’enjeux socio- économiques. En effet, le dynamisme naturel, les et les aléas météorologiques extrêmes rendent ces zones côtières particulièrement vulnérables. Ceci a motivé la recherche sur le concept de la vulnérabilité afin de la rendre mesurable. Le but étant de fournir des méthodes permettant aux acteurs engagés sur les questions littorales de traiter les causes et prévenir les conséquences pour une gouvernance efficace. Au vu de la complexité de cette zone, qui se trouve au carrefour de plusieurs facteurs, ces méthodes d’évaluations requièrent une approche systémique. Une des premières tentatives d’évaluations la vulnérabilité côtière par un calcul d'Indice de Vulnérabilité Côtière (IVC) était celle de Gornitz (1991). Suite à son étude, de nombreuses versions d’IVC ont été développées à différentes échelles. Le sous-groupe de la gestion des zones côtières du Groupe Intergouvernemental d’Experts sur l'Évolution du Climat (GIEC) a encouragé la mise au point d’une Méthodologie Commune (MC) pour évaluer la vulnérabilité des différentes nations côtières. Cependant, les données nécessaires pour évaluer un ou plusieurs facteurs ne sont pas toujours disponibles, en particulier pour les pays en développement (Awange et al., 2018). De plus, la surveillance de l'environnement côtier est souvent une tâche fastidieuse en raison des techniques traditionnelles chronophages qui ne permettent pas de couvrir un système côtier en entier. La télédétection, les approches de la géomatique et des Systèmes d’Information Géographique (SIG) peuvent fournir des alternatives pour cette surveillance. Les images satellitaires fournissent une couverture synaptique et sont utiles pour les applications de surveillance détaillée (Mujabar & Chandrasekar, 2013). Ainsi, leurs propriétés spectrales peuvent être utilisées pour extraire plusieurs caractéristiques notamment celles du relief côtier (Boak & Turner, 2005) et de l’occupation du sol. Dans cette étude, une version modifiée de l'IVC est proposée en traitant 5 variables propices aux côtes micro-tidales, en se basant sur l’apport de la télédétection et des SIG. 1. Ecole Nationale Supérieure des Sciences de la Mer et de l’Aménagement du Littoral, Dely Ibrahim, Alger, Algérie, bengoufasoumia@gmail.com, belkessa.r.littoral@gmail.com, abbadkatia@gmail.com, 2. UBO, Université de Bretagne Occidentale, Laboratoire LETG Brest UMR 6554 CNRS, Brest, France simona.niculescu@univ-brest.fr 3. Ecole National Supérieure de l’Hydraulique, Mobilisation et Valorisation des Ressources en Eau, NR 29; BP 31 BLIDA, Algérie, 09000 ; k.mihoubi@ensh.dz 3. Ecole National Supérieure de l’Hydraulique, Mobilisation et Valorisation des Ressources en Eau, NR 29; BP 31 BLIDA, Algérie, 09000 ; k.mihoubi@ensh.dz RESUME. Dans le contexte de la croissance d'activités socio-économiques littorales et l’accentuation des risques côtiers, ce travail vise à évaluer la vulnérabilité d’un linéaire côtier de 32 km de la ville de Mostaganem (l’Ouest-Algérien). En utilisant des méthodes de télédétection et des Système d'Information Géographique (SIG), cinq variables physiques et socio-économiques, à savoir le taux d'érosion (trait de côte), le type du relief, la pente, la hauteur significative des vagues (Hs) et le type d’occupation du sol, sont traitées. Un modèle de SIG a été créé pour calculer un Indice de Vulnérabilité Côtière (IVC) en utilisant une méthode de pondération. Sur la base des valeurs de l'IVC, la côte est catégorisée en quatre classes de vulnérabilité, à savoir : faible (7,5 km), modérée (4.08 km), élevée (11 km), et très élevée (9,74 km). Cette étude pourrait contribuer à la cartographie de la vulnérabilité de la zone d’étude dans une optique de mitigation et prévention du risque de l’érosion côtière. p q g p q ABSTRACT. In the context of the increasing rate of coastal socio-economic activities and the exacerbating of the coastal hazards, this work aims to assess the coastal vulnerability of 32 km of the Mostaganem coastline (Algeria), using remote sensing methods and Geographic Information Systems (GIS). To achieve that, five variables, namely erosion rate (shoreline), relief type, slope, significant wave height (Hs) and land use type, are considered. A GIS model was created to calculate a Coastal Vulnerability Index (CVI) based on weighting method. According to the CVI values, the shoreline is classified into four vulnerability classes, LOW (42.5 km), MODERATE (23.46 km), HIGH (34.61 km), and VERY HIGH (7.5 km). Thus, this study is useful for risk mitigation and prevention of coastal erosion hazard. MOTS-CLES : Indice de Vulnérabilité Ouest-algérien Machine Learning SIG Erosion 25 25 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction L'objectif principal de cette étude est d’appliquer une méthode d'évaluation de la vulnérabilité côtière face au risque d’érosion, qui sera à la fois pratique et scientifiquement fondée sur des sources et de méthodes de traitement de données fiables. Le but est de dériver un outil reproductible d’aide à la planification, la prise de décisions et l’aménagement des villes côtières. 2. Zone d’étude La zone d’étude est une des trois grandes Wilayas du Nord-ouest algérien. Située à 400 km de la capitale, Mostaganem est une zone touristique par excellence (Taibi, 2016). 26 Communications 3. Matériel et méthodes Le présent travail formule un modèle SIG de l'IVC dans un contexte socio-éco- systémique, en prenant en considération les variables propices aux côtes micro- tidales, à l’instar des côtes du bassin ouest méditerranéen. Le modèle développé est composé de cinq variables géophysiques et socio-économiques : a) le taux d'érosion/accrétion (m), b) les formes de relief (géomorphologie), c) pente (%), d) hauteur moyenne significative des vagues (m), e) l'occupation du sol (fig. 2). Ce sont des variables de différentes dimensions. Cette hétérogénéité est éliminée par un système de pondération de 1 à 5 (fig. 2) selon les intervalles définis par McLaughlin & Cooper, (2010). Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Cartographie de la vulnérabilité côtière Cartographie de la vulnérabilité côtière La demande croissante ainsi que le nombre insuffisant d’infrastructures ont conduit les autorités à augmenter le nombre de structures d’accueil. En effet, 16 zones d'extension touristique côtières sont prévues et sont en cours d’aménagement (Ghodbani et al., 2016). Compte tenu des enjeux sociaux, économiques et naturels, des constituent un sujet de préoccupation de plus en plus prégnant pour la Wilaya de Mostaganem. De ce fait, il convient de promouvoir une approche pragmatique afin d’évaluer son état de vulnérabilité actuel, pour un aménagement et une planification spatiale idoine. Figure 1. Situation géographique de la zone d'étude Figure 1. Situation géographique de la zone d'étude 3.1. Traitement des données Depuis une synthèse cartographique, jusqu’au recours à la modélisation passant par l’analyse des images satellitaires (fig. 2) divers types de sources de données ont été utilisées afin de traiter les variables retenues pour le modèle IVC : 27 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] a- La variable : le taux d'érosion/accrétion (Trait de côte) Figure 2. Méthodologie de travail et approche de traitement de donnée Figure 2. Méthodologie de travail et approche de traitement de donnée a- La variable : le taux d'érosion/accrétion (Trait de côte) Une analyse spatio-temporelle de l’évolution du trait de côte (TC) a été élaborée afin de calculer le taux d’érosion, en comparant l’ancienne position du TC issue des photos aériennes de 1992 avec sa position récente extraite automatiquement à partir des images de type Pléiades (0.5 m). L’acquisition de ces images a été programmée en juillet 2019 pour couvrir la zone d’étude le jour même de l’acquisition des données In-Situ par GPS utilisées afin de valider le modèle d’extraction automatique du TC. Ce modèle est basé sur une classification supervisée des images par une approche orientée objet. La méthode d’extraction automatique comprend trois étapes principales : (a) la segmentation de l'image (b) l'entraînement et la classification et (c) Extraction du trait de côte (fig. 3). L’algorithme de segmentation de type Multirésolution a été appliqué en utilisant le logiciel ECognition. L’algorithme Random Forest (RF) a été utilisé pour la classification de l’image afin de détecter l'indicateur du TC retenu pour l'évaluation du taux érosion, qui est la limite supérieure du sable humide. C’est est un algorithme d'apprentissage automatique (Machine Learning) non paramétrique (Breiman, 2001). 28 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Cartographie de la vulnérabilité côtière Cartographie de la vulnérabilité côtière Après un processus d'optimisation par validation croisée, les valeurs du nombre d'arbres dans la forêt (Ntree) et le nombre de variables à chaque nœud (Mtry) ont été fixées à 500 et 60 respectivement. La classification a permis de détecter le TC qui a été extrait automatiquement sous SIG. Après un processus d'optimisation par validation croisée, les valeurs du nombre d'arbres dans la forêt (Ntree) et le nombre de variables à chaque nœud (Mtry) ont été fixées à 500 et 60 respectivement. La classification a permis de détecter le TC qui a été extrait automatiquement sous SIG. Une étape de lissage s’est avérée nécessaire afin de le comparer à celui extrait des photos aériennes (fig. 3). Les TC extraits ont été intégrés dans le modèle Digital Shoreline Analysis System (DSAS). Les valeurs de End Point Rate (EPR) ont été calculées sur des transects d’une équidistance de 20 m. Il s'agit d'un taux calculé en divisant la distance entre la position du TC initial (1992) et celle du plus récent (2019) par le temps écoulé (58 ans). a- La variable : le taux d'érosion/accrétion (Trait de côte) L'EPR a l’avantage d’une exigence minimale pour les entrées (deux traits de côtes) (To et Thao, 2008). Figure 3. Extraction du trait de côte a) Segmentation, b) Classification, c) Extraction Figure 3. Extraction du trait de côte a) Segmentation, b) Classification, c) Extraction b- La variable : relief côtier (géomorphologie) Les données permettant d’analyser le degré de vulnérabilité liée à la géomorphologie de la côte étudiée sont tirées des Modèles Numériques de Terrain (MNT), obtenus via le portail en ligne de l'United States Geological Survey (USGS), ainsi que des images Pléiades. Les MNT ont été fusionnés avec les données issues des cartes topographiques et traitées sous SIG (ArcGis® 10.1). Le traitement des images Pléiades a permis de reconnaître les faciès géomorphologiques du littoral. A noter que les connaissances des traits géomorphologiques du site acquises lors des sorties terrain effectuées, ont facilité l’évaluation. 3.2. Calcul d’indice de vulnérabilité côtière (IVC) Les résultats du traitement relatif à chaque variable ont été intégrés dans un modèle SIG pour une application d’un système de pondération de 5 degré de cotation (un code de 1 jusqu’à 5) (Fig.1) selon une grille de calcul par cellule de 20m*20m. A chaque variable est affectée un degré en fonction de l'ampleur potentielle de son exposition et sa contribution à des changements physiques de la côte. Les 5 degrés obtenus font l’objet de calculs en utilisant l’équation (1) développée par McLaughlin et al. (2001). L'éventail complet de l'IVC a été divisé en quatre, à parts égales, et le quart supérieur a été pris comme "côte à très haut risque". CVI = (1) CVI = (1) a = Taux d'érosion ; b = géomorphologie ; c = pente ; d = hauteurs des vagues e = occupation du sol ; e- La variable : occupation du sol Le type d’occupation du sol a été déterminé via un traitement des images Pléiades par une approche de classification orientée objet. Une segmentation par l’algorithme Multiresolution a été effectuée avec le logiciel Ecognition. Une classification supervisée des objets résultants de la segmentation a été réalisée par l’algorithme RF avec une optimisation préalable des hyperparamètres (nombre d’arbres Ntree et profondeur maximale Mtry). L’optimisation a été effectuée par la méthode de validation croisée afin de fixer les valeurs des paramètres de la segmentation Multirésolution et de l’algorithme RF. En effet, ces valeurs ont été choisies sur la base des modèles de classification qui ont obtenu la plus grande précision globale. Ainsi, les valeurs des paramètres de segmentation étaient : Sp=10, Sp= 0.5, Cp= 01 et les valeurs de RF étaient : Mtry=25 et Ntree=500. Une étape de validation a été entreprise pour calculer la matrice de confusion et les indices de pertinence (indice de précision globale et indice Kappa). c- La variable : pente La pente régionale de la zone côtière a été calculée à partir d'une grille d’élévations topographiques, élaborée par le traitement du MNT et la grille bathymétrique, dont les données numériques sont extraites de la carte marine établie par le Service Hydrographique des Forces Navales algériennes (SHFN). Ces données ont été fusionnées et traitées à l’aide du logiciel SIG (ArcGis® 10.1) afin d’établir une carte de pente. 29 29 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 d- La variable : la hauteur significative des vagues à la côte (Hs) Basée sur des modèles mathématiques, la modélisation numérique décrit de manière suffisamment fiable les phénomènes physiques. En effet, un modèle numérique du MIKE21-DHI a été utilisé dans le cadre de cette étude afin de calculer les hauteurs significatives des vagues à la côte (Hs). Il s’agit dans un premier temps de recueillir l’ensemble des données nécessaires à la simulation (des données bathymétriques, météorologiques et le niveau de la mer). Les modèles en sorties ont montré que le degré d’exposition augmente lorsque les vagues proviennent du secteur NW (fig. 4), avec un temps de retour de 8s et une Hs au large de 2.5m qui sont les paramètres choisis pour l'évaluation de la sensibilité liée à cette variable. Cartographie de la vulnérabilité côtière Cartographie de la vulnérabilité côtière a. La variable trait de côte Le modèle d'extraction automatique du TC a été validé par des données terrain. En observant la distance moyenne entre le TC extrait et celui du terrain, il est apparu que le TC produit par le modèle d’extraction présente une bonne précision (Distance max de 0.7m). La figure 4 illustre le taux d’évolution du TC, de la zone d'étude, estimé à partir des données historiques et récentes (1992 et 2019). Le taux d'évolution pendant la période étudiée révèle une érosion importante dans plusieurs parties de la côte de Mostaganem (plus de 2m/an), par conséquent, une valeur pondérée de 5 leur a été attribuée (fig.9). Cette forte érosion est notée au niveau des plages sableuses. Ceci pourrait être lié aux constructions massives sur les cordons dunaires et l’extraction du sable non réglementée (le pillage) (Senouci & Taibi, 2019). Les côtes marquant une érosion modérée à faible (moins de 2m/an), principalement les côtes rocheuses, ont une résistance considérable face aux forçages de la mer. Graph of Transects_25m_rates_20201018_115416 TransectID 2.27. 65. 99. 142. 193. 244. 296. 347. 398. 450. 501. 552. 604. 655. 707. 758. 809. 861. 912. 963. 1,025. 1,097. 1,169. 1,241. 1,313. 1,385. 1,456. 1,528. 1,600. EPR 3 2 1 0 -1 -2 -3 -4 Figure 4. Taux net d’évolution du trait de côte par an EPR (1992-2019) Graph of Transects_25m_rates_20201018_115416 Figure 4. Taux net d’évolution du trait de côte par an EPR (1992-2019) b. La variable : pente La figure 5 montre que le gradient de la pente du tronçon côtier étudié est estimé entre 0,2 % et 9,3 %. Les zones ayant les gradients de pente inférieurs à 1,5%, à savoir les plages, sont plus susceptibles de subir un lessivage des sédiments à contre- courant (Jeyabal & Ramasamy, 2014). A l'opposé d'une côte en pente douce, où toute élévation du niveau de la mer inonderait de grandes étendues de terre, la conséquence de l’action de la mer sur une côte abrupte serait insignifiante (Rao et al., 2008). Par conséquent, à faible pente la côte devrait reculer plus vite que les côtes à pente raides, par conséquent, ces zones se voient attribuer les valeurs pondérées de 5. 4. Résultats et discussion Les résultats du traitement des données relatifs à chaque variable ainsi que le degré de leur vulnérabilité sont présentés ci-dessous. 30 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 c. La variable : relief côtier Le littoral étudié présente une alternance de plages, falaises, baies rocheuses, dunes et cordons dunaires. En fonction du faciès rencontré, les risques d’érosion côtière sont plus ou moins importants. Selon McLauglin et Cooper (2010) les reliefs côtiers qui sont à basse altitude (<10 m), à l’instar des plages sableuses, se voient attribuer la valeur 5 (très forte vulnérabilité) (fig.9), ces côtes ont un potentiel élevé aux inondations ou aux submersions marines. Les côtes à falaise et de rivages rocheux de hautes altitudes (>de 20 m) sont considérées moins vulnérables parce qu'elles sont plus résistantes au forçage de la mer et à la remontée du niveau marin. De ce fait, la valeur pondérée de 1 leur a été attribuée (fig.9). 31 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 5. Carte des pentes (%) Figure 6. Carte topographique de la zone (dérivée du MNT) Figure 6. Carte topographique de la zone (dérivée du MNT) Figure 5. Carte des pentes (%) Figure 6. Carte topographique de la zone (dérivée du MNT) Figure 5. Carte des pentes (%) [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 d. La variable : hauteur des vagues Les modèles en sorties montrent que la quasi-totalité du littoral de Mostaganem est exposée à l’action des vagues et des houles. Cette forte exposition est liée au fait que la morphologie côtière et sous-marine dissipe peu l’énergie marine. Les zones côtières du secteur ouest de la zone d’étude montrent une augmentation de la hauteur des vagues de plus de 1,4 m, libèrent plus l'énergie sur la côte, ce qui entraîne une érosion ou un affaissement important des plages. De ce fait, une valeur pondérée de 5 (très forte vulnérabilité) lui a été attribuée (fig.11). Les ondes de faible énergie produisent une hauteur d'onde plus faible (<1,0 m), ce qui entraîne des processus d'accrétion dus à l’accumulation des sédiments. Par conséquent, une valeur pondérée de 1 a été attribuée à ces zones de faible hauteur de vagues. Figure 7. Hauteurs des vagues à la cote calculées par le model de Mike-21 Figure 7. Hauteurs des vagues à la cote calculées par le model de Mike-21 32 Communications Communications [ISBN : 978-2-910545-12-1] e. La variable : occupation du sol Les résultats de l’analyse de l’image Pléiade par une approche de classification orientée objet et l’algorithme RF sont représentés dans la figure 12, qui illustre la répartition spatiale du type d’occupation du sol du littoral de Mostaganem. Cette analyse a permis de détecter cinq classes d’occupation du sol. La classification par l’algorithme RF utilisant la segmentation Multirésolution a montré peu d'indications d'erreurs. Une évaluation de la précision a été effectuée afin de montrer l'efficacité de la classification basée sur les modèles optimisés. L'analyse de la matrice de confusion révèle que les indices de précision pour plusieurs classes étaient relativement cohérents avec un indice kappa de l’ordre de 93%. L’analyse globale a révélé que la côte étudiée est fortement urbanisée au niveau de certains secteurs, principalement dans sa partie centre (ville de Mostaganem). L’activité agricole, quant à elle, est très présente, particulièrement dans le secteur ouest. En raison de l'impact de ces actions anthropiques et suite à la suppression du tissu végétal et aux empiètements, les cordons dunaires, qui sont considérés comme barrière naturelle protectrice, connaissent un taux de diminution de plus de 2,0 km2 (Senouci & Taibi, 2019). Ces mêmes types d’occupation du sol, responsables de la destruction de l’écosystème dunaire, se trouvent sans protection et directement exposées à l’action de la mer. En effet, ils sont considérés à la fois comme enjeux socio-économiques à protéger et facteurs anthropiques accentuant la vulnérabilité côtière, la raison pour laquelle une valeur pondérée de 5 leur a été attribuée. 4.1. Indice de Vulnérabilité Côtière (IVC) Figure 8. Carte d’occupation du sol de la frange côtière de Mostaganem ndice de Vulnérabilité Côtière (IVC) Figure 8. Carte d’occupation du sol de la frange côtière de Mostaganem Figure 8. Carte d’occupation du sol de la frange côtière de Mostaganem 4.1. Indice de Vulnérabilité Côtière (IVC) Figure 8. Carte d’occupation du sol de la frange côtière de Mostaganem 4.1. Indice de Vulnérabilité Côtière (IVC) Le résultat des valeurs de l’IVC varient de de 3.09 à 19.36, avec une valeur moyenne de 10,41. Sur la base de cette fourchette de valeurs, les zones côtières sont catégorisées en quatre classes : vulnérabilité, faible, modérée, forte et très forte. La 33 33 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] carte de l’IVC (fig. 9.f) montre le degré de vulnérabilités côtières le long du littoral de Mostaganem. 5. Discussion Plusieurs études (ex. Ayadi, et al., 2016 ; Kermani, et al., 2016 ; Bengoufa, et al., 2018) ont rapporté que le littoral algérien est soumis à une érosion considérable mettant en péril des enjeux socio-économiques. En effet, il est essentiel que les scientifiques contribuent à développer des méthodes d'évaluation des risques et d'assurer leur applicabilité et ainsi produire des données fiables et adaptées aux pays en développement, qui disposent peu de données (Awange, et al., 2018). Les variables du modèle de l’IVC peuvent être choisies en fonction de la localisation et les caractéristiques des côtes (Koroglu et al. 2019). Dans le modèle ici développé 5 variables ont été traitées par des méthodes de la géomatique et de la télédétection. Ces méthodes intégrées permettent d’optimiser le temps et l’effort et offrent les solutions technologiques pour générer des bases de données sur le relief côtier et ses processus dynamiques (Kaliraj, S., & Chandrasekar, N. 2012). Ainsi, leur efficacité pour rendre mesurable la vulnérabilité côtière, a été prouvée grâce au système de cotation de 5 niveaux selon les intervalles d’intensité de chaque variable qui ont des unités hétérogènes, en homogénéisant leurs dimensions. La performance de l'algorithme d’apprentissage automatique RF pour la détection du trait de côte, a été démontrée et validée par des mesures in-situ. En fait, un nombre limité d'études ont évalué la précision de ces algorithmes pour la détection du trait de côte. Ainsi, rares sont celles qui ont validé leurs résultats avec des données de référence (Dada, et al., 2016). En outre, pour analyser le niveau d’exposition des activités socio-économiques au risque d'érosion, le modèle d’IVC ici développé a intégré le type d’occupation du sol. Des études qui intègrent la composante socio-économique, ce qui était le cas de ce modèle d’IVC, permettront une planification urbanistique plus efficace et une définition les règles qui en découlent, y compris l’interdiction des constructions dans les zones exposées. [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 e. La variable : occupation du sol Le graphe (fig.10) montre les étendues en linéaire côtier (Km). La côte étudiée s’étend sur un linéaire de 32 km, dont 30.13 % à 34.03 % présente un fort à très fort degré de vulnérabilité respectivement (IVC de 12.13 à 19.36). 12.62% de ce linéaire est caractérisée par une intensité modérée (IVC de 8.51 à 2.13). La carte de synthèse de l’IVC révèle que la forte intensité caractérise surtout le secteur central de Mostaganem, fortement urbanisé, ainsi que les reliefs de faible altitude tels que l’estuaire d’Oued Cheliff et les criques sableuses. Par ailleurs, 23.2% de ce linéaire présente un faible degré de vulnérabilité (IVC<8.51). Ces zones connaissent un phénomène d'accrétion dues aux affleurements rocheux au large des côtes (barrières naturelles). De même pour les côtes rocheuses et à falaise (élévation supérieure à 10,0 m) qui sont mieux protégées. Figure 9. Carte la vulnérabilité côtière a) vulnérabilité par rapport à la variable tra ôte, b) vulnérabilité par rapport à pente, c) vulnérabilité par rapport à la variable r côtier, d) vulnérabilité par rapport à la variable Hs, e) vulnérabilité par rapport à variable type d’occupation du sol, f) Indice de Vulnérabilité Côtière de Mostagane Figure 10. Présentation graphique du degré de vulnérabilité en termes de pourcentage du linéaire côtier Figure 10. Présentation graphique du degré de vulnérabilité en termes de pourcentage du linéaire côtier Figure 10. Présentation graphique du degré de vulnérabilité en termes de pourcentage du linéaire côtier Figure 9. Carte la vulnérabilité côtière a) vulnérabilité par rapport à la variable trait de côte, b) vulnérabilité par rapport à pente, c) vulnérabilité par rapport à la variable relief côtier, d) vulnérabilité par rapport à la variable Hs, e) vulnérabilité par rapport à la variable type d’occupation du sol, f) Indice de Vulnérabilité Côtière de Mostaganem 34 Communications [ISBN : 978-2-910545-12-1] Financement Ce travail de recherche est soutenu par le Partenariat Hubert Curien PHC-Tassili. Ce travail de recherche est soutenu par le Partenariat Hubert Curien PHC-Tassil [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Financement Bibliographie Awange, J. L., Saleem, A., Konneh, S. S., Goncalves, R. M., Kiema, J. B. K., & Hu, K. X. (2018). Liberia’s coastal erosion vulnerability and LULC change analysis : Post-civil war and Ebola epidemic. Applied Geography, 101, 56–67. Boak, E. H., & Turner, I. L. (2005). Shoreline definition and detection : A review. Journal of coastal research, 21(4 (214)), 688–703. Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32. Brooks, N., Nicholls, R., & Hall, J. (2006). Sea-level rise : Coastal impacts and responses. Norwich WBGU ISBN, 3–936191. Chandrasekar, V., Keränen, R., Lim, S., & Moisseev, D. (2013). Recent advances in classification of observations from dual polarization weather radars. Atmospheric Research, 119, 97–111. Dada, OA., Guangxue, L., Qiao, L., Ding, D., Yanyan, M., Jishang X. 2016. Seasonal shoreline behaviours along the arcuate Niger Delta coast: Complex interaction between fluvial and marine processes. Continental Shelf Research 122: 51–67. Ghodbani, T., Kansab, O., & Kouti, A. (2016). Développement du tourisme balnéaire en Algérie face à la problématique de protection des espaces littoraux. Le cas des côtes mostaganemoises. Études caribéennes, 33‑34. Gornitz, V. (1991). Global coastal hazards from future sea level rise. Global and Planetary Change, 3(4), 379‑398. Jeyabal, G., & Ramasamy, S. (2014). Down core variation in sediment characteristics and trace element geochemistry of a core sample in Pichavaram Mangrove area, Tamil Nadu, Southeast Coast of India. Enviro Geo Chimica Acta, 1(3), 206–214. Kaliraj, S., & Chandrasekar, N. (2012). Geo-processing model on coastal vulnerability index to explore risk zone along the south west coast of Tamilnadu, India. Int JEarth Sci Eng, 5. Koroglu, A., Ranasinghe, R., Jiménez, J. A., & Dastgheib, A. (2019). Comparison of coastal vulnerability index. Ocean & coastal management, 178, 104799. McLaughlin, S., & Cooper, J. A. G. (2010). A multi-scale coastal vulnerability index : A tool for coastal managers? Environmental Hazards, 9(3), 233–248. Mujabar, P. S., & Chandrasekar, N. (2013). Shoreline change analysis along the coast between Kanyakumari and Tuticorin of India using remote sensing and GIS. Arabian Journal of Geosciences. 6 (3), 647-664 Senouci, R., & Taibi, N.-E. (2019). IMPACT OF URBANIZATION ON COASTAL DUNE : CASE OF KHARROUBA, WEST ALGERIA. Journal of Sedimentary Envi, 4(1) Taibi, N.-E. (2016). Conflict Between Coastal Tourism Development and Sustainability : Case of Mostaganem, Western Algeria. Eur Jrnl of Sust Development, 5(4), 13–13. To, D. V., & Thao, P. T. P. (2008). 6. Conclusion Dans une optique de spatialisation des risques, cette étude a permis de développer une méthode d'évaluation de la vulnérabilité face au risque d’érosion selon une approche géospatiale. D'après les résultats, plusieurs points chauds (forte vulnérabilité) ont été identifiés à différents endroits. Cette identification des zones permet de mieux tracer les plans d’aménagement côtier dans une optique de développement durable et une gestion d’un système de potentialités incontournables. L’utilisation des méthodes intégrées de télédétection et de SIG a permis de générer une base de données relative au relief côtier, aux processus dynamiques et aux facteurs socio-économiques. La classification supervisée des images Pléiades a montré un grand potentiel dans le traitement des variables, notamment pour l’extraction du trait de côte et le type de l’occupation du sol. Ainsi, ces méthodes ont permis de rendre la vulnérabilité mesurable malgré l’hétérogénéité des variables qui entrent en jeu en les intégrant dans un seul modèle SIG. Ce modèle est une plate- forme efficace de cartographie des zones potentiellement vulnérables aux dommages dus à l’érosion et il est utile aux scientifiques et décideurs pour une planification à long terme de la gestion des zones côtières. 35 Communications [ISBN : 978-2-910545-12-1] Bibliographie A shoreline analysis using DSAS in Nam Dinh Coastal Area. International Journal of Geoinformatics, 4(1). 36 Communications Communications par Session Approches géomatiques pour la transition Approches géomatiques pour la transition SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 1 Introduction Nowadays, the main goal of most countries’ energy planning policy (Hache, Palle, 2019) in terms of energy transition, relies on the integration of renewable- based generation in power networks. However, increasing the share of renew- able energy (RE) sources is still challenging, as a result of their inherent in- termittent and geographical dispersion (Hache, Palle, 2019). Specific planning strategies (Gorsevski et al., 2013; Ramirez Camargo, Stoeglehner, 2018) must be developed accordingly, in order to reach energy transition targets without threatening the existing infrastructures. Recent works propose an integrated model framework, combining Geo- graphic Information Systems and Robust Optimization, such as the GREECE- OPSPV (Geographical REnewable Energy Candidate Extraction - Optimal Plan- ning and Sizing of PV parks) system (Al-Kurdi et al., 2019; Pillot et al., 2020). It identifies the optimal sites for solar photovoltaic (PV) generation at utility scale, according to specific energy planning targets. The approach connects both geographical information system (GIS) and robust linear optimization (RO) through the pipe depicted in Figure 1. The GIS component gathers large heterogeneous sets of spatiotemporal data, and allows for location and size of the best solar PV sites to be retrieved with respect to geographical constraints (restricted areas, land use, distance to grid, etc.), spatial disper- sion of the resource, hourly global energy demand and generation, predefined planning scenarios, and the degree of risk adversity authorized by the decision maker. Through a set of spatio-temporal data layers and control parameters, the GIS module (GREECE) converts spatial constraints and parcels into items characterized by de-spatialized attributes and solar resource time series. Based on current electricity generation and demand time series as well as projection scenarios, the RO model (OPSPV) then looks for optimal site candidates (loca- tion, size and power), with respect to given temporal constraints (parcel size, hourly electricity demand, maximum penetration of intermittent RE power) and objectives (maximize energy generation and minimize total costs). The model eventually gives an estimate of the risk associated with solar PV invest- ment at utility scale to the decision and policy maker, by means of a Pareto approach (cost vs. energy generation) and according to best and worst case scenarios. In this integrated GIS and decision support model, when a solar PV site (location and size) is proposed, a new substation is considered for connection to the power network. Optical network ring analogy for effective solar PV site selection using GIS and optimization modelling Benjamin Pillot 1, Nadeem Al-Kurdi 1, Carmen Gervet 1, Laurent Linguet 1 ESPACE-DEV, Univ Montpellier, IRD, Univ Guyane, Univ Réunion, Univ Antilles, Montpellier, France ABSTRACT. Energy transition requires a holistic approach, involving land, resources, environmental and economical data and constraints. The core purpose of energy tran- sition, is to migrate power systems towards renewable energy usage (solar, hydro, biomass, wind), in a technical and economical viable manner. In this article we address this challenge as a spatio-temporal analysis problem combined with decision support targeted for policy makers and investors. We show how the selection of po- tential PV sites, can efficiently be combined with the optimization of the connection to the grid in terms of operational costs. Our contribution lies in bringing forward a modeling analogy with the SONET problem, addressed in fiber n etwork d esigns. Im- proved results compared to an existing GIS-optimization PV site placement approach are illustrated on a real case study and data from the French Guiana. RÉSUMÉ. La transition énergétique nécessite une approche holistique, impliquant la gestion des sols, ressources potentielles, les données et contraintes environnemen- tales et économiques. L’objectif central est de migrer les systèmes électriques vers l’utilisation des énergies renouvelables (solaire, hydroélectrique, biomasse, éolienne), de manière techniquement et économiquement viable. Dans cet article, nous abordons ce défi comme u n p roblème d ’analyse s patio-temporelle i ntégré à u n m odèle d ’aide à la décision conçu pour les décideurs politiques et les investisseurs. Nous montrons comment la recherche de sites photovoltaïques peut être combinée efficacement avec l’optimisation du raccordement au réseau en termes de coûts. Notre contribution ré- side dans la mise en place d’une analogie de modélisation avec le problème SONET, défini d ans l es conceptions d e réseaux d e fibre op tique. La qu alité de s ré sultats com- parés à une approche existante de placement de parcs photovoltaïques est présentée sur un cas d’étude en Guyane française. KEYWORDS: Energy planning, Spatial Decision support, Spatio-temporal data MOTS-CLÉS : transition énergétique, aide à la décision spatiale, données spatio- temporelles 39 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1 Introduction This approach was justified as the model was designed for private investors in RE plants, thus each solar PV site required its own connection to the grid. However, in island networks (i.e. not interconnected), the network manager might have the authority to decide which site should be turned on and off, depending on the share of intermittent power (Notton 40 Communications Communications 40 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] GIS and optimization modelling Integrated model framework GREECE module Optimization module Subset of power stations Size Capacity Set of candidates Topography Land use Resource dataset Restricted area layers Base layer (county, etc.) Road, grid, etc. Partition larger polygons Remove smaller polygons and polygons beyond distance threshold Remove restricted areas Distance thresholds Land surface area Buffers Geographical data layers Control parameters Compute de-spatialized features Location & Features Aggregate PV power time series Parcel size maximal size PV existing + new 35 % of total power demand Existing production + new PV hourly demand Maximize energy generation Minimize total costs Figure 1. Schematic view of the original GREECE-OPSPV integrated model framework (Al-Kurdi et al., 2019) Figure 1. Schematic view of the original GREECE-OPSPV integrated model framework (Al-Kurdi et al., 2019) et al., n.d.; Tapachès et al., 2019). Accordingly, some investors have moved on to diversifying their RE investments (hydro, biomass, etc.) or exploiting power plants with added storage capacity (Spaes, 2019; Tapachès et al., 2019). Another solution though is to rely on co-investment, whereby several sites belonging to multiple investors would be connected to the same substation. To our knowledge this approach hasn’t been investigated in terms of simulation models. This would be a cost-effective solution as the maximum power per RE site is constrained by the technical features of the power network. In fact, for multiple small-scale solar PV facilities, the final cost would grow with the number of corresponding substations. In this paper we address this problem by deriving a model that would mitigate the cost by aggregating various power plants around one unique substation. This objective comes with new challenges on the GIS and optimization model. et al., n.d.; Tapachès et al., 2019). Accordingly, some investors have moved on to diversifying their RE investments (hydro, biomass, etc.) or exploiting power plants with added storage capacity (Spaes, 2019; Tapachès et al., 2019). Another solution though is to rely on co-investment, whereby several sites belonging to multiple investors would be connected to the same substation. To our knowledge this approach hasn’t been investigated in terms of simulation models. This would be a cost-effective solution as the maximum power per RE site is constrained by the technical features of the power network. In fact, for multiple small-scale solar PV facilities, the final cost would grow with the number of corresponding substations. Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] model of the framework. Since this raises the computational complexity of the combinatorial optimization, the data range of those constraints is first pruned by enhancing the GREECE model with respect to the spatial constraints that apply to the potential sites. It ensures keeping the CPU time reasonably low without loss of solution quality. With the proposed architecture, the best solar PV sites are now gathered into rings, i.e. local aggregates of plants, connected to one sole substation. It results in lower energy unit cost (Me/GWh) as well as means for the decision makers and grid managers to manage their risk adversity, compared to the previous architecture. To describe the resulting benefits, we compare b oth a pproaches, with and without considering constraints from the SONET problem, by applying the GREECE-OPSPV framework to the real case of French Guiana. The article is structured as follows: section 2 presents the revisited version of GREECE- OPSPV based on the SONET analogy; section 3 is the experimental section based on a real-world case study and compare both results from the former and the new version of GREECE-OPSPV; section 5 concludes the paper. 2 Applying the SONET design to RE sources in power systems 2 Applying the SONET design to RE sources in power systems 2.1 Planning small-scale PV sites for safer grid operation and improved risk management Safer grid operation. When too much intermittent power is injected into the grid, it might be necessary for the network manager to disconnect some of the solar PV sites (Tapachès et al., 2019; Notton et al., n.d.). When there are only large power plants, the loss of one of them could affect grid stability. The use of smaller facilities therefore gives more flexibility w hen h andling t he energy supplied from intermittent sources, and eventually ensures safer grid operation once solar PV sites have been commissioned. This idea of targeting smaller solar PV sites is even more relevant in small not interconnected electricity networks, whose low inertia actually implies higher impact from intermittent RE sources on frequency variability and grid stability (Notton et al., n.d.; Tapachès et al., 2019). Typically, the GREECE-OPSPV model allows risk adversity to be assessed upstream using Pareto optimisation through best and worst case scenarios (Al-Kurdi et al., 2019; Pillot et al., 2020). Communications In this paper we address this problem by deriving a model that would mitigate the cost by aggregating various power plants around one unique substation. This objective comes with new challenges on the GIS and optimization model. The problem tackled is to maximize PV energy supply and minimizing costs through small-scale power facilities. Our new framework is based on an anal- ogy with the Synchronous Optical NETwork (SONET) (Pelleau et al., 2009; Goldschmidt et al., 2003), which is a popular network design in the field of fiber-optic technology. Typically, we first tailor the constraints of the SONET optimization problem to our own problem and then integrate them into the RO 41 41 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] In this paper, we consider smaller PV sites, and look for handling the risk downstream: whether the riskier solution is chosen or not, the grid manager must remain capable of eventually switching off a plant without jeopardizing the whole network. Risk adversity management. When there is one unique investor deploying mul- tiple sites, it is also more interesting to invest in various small-scale facilities rather than in few bigger ones. As a matter of fact, the grid manager will not care about the economical risk but rather about the technical risk: he will eventually shut down any solar PV site in case of too much intermittent power injected into the network (Notton et al., n.d.; Tapachès et al., 2019). This technical liability might not be related to technical constraints, as this is the 42 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] GIS and optimization modelling GIS and optimization modelling GIS and optimization modelling case for EDF in French Guiana which disconnects plants in the chronological order of their connection to the grid (CTG, 2017; EDF, 2019). As a result, while the investor may ensure for instance to still have 6 or 7 out of 8 small- scale PV plants running when some sites are eventually disconnected, he could loose half of the production if he only runs two large-scale facilities. This is an economical risk that some investors might not be willing to take. Clusters of RE sources: towards the SONET analogy. Previous works on the GREECE-OPSPV model, sought each solar PV power plant to be built along its own substation (Al-Kurdi et al., 2019; Pillot et al., 2020). As mentioned, this approximation is no longer valid when the number of sites dramatically increases, which is the case if we only allow small-scale plants to be eventually built. In order to keep the solutions economically relevant, we revisit this model so that several power plants might actually be connected to the same substation. We present the underlying methodology in the following section. Communications 42 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 2.2 Adapting SONET constraints to the RE planning problem In the field of fiber-optic technology, one of the most popular network designs is the Synchronous Optical NETwork (SONET) (Pelleau et al., 2009; Gold- schmidt et al., 2003). In one possible topology, each customer is connected to exactly one local ring through add-drop multiplexers (ADM), and those local rings are all connected to one federal ring through a digital cross connector (DXC) (Pelleau et al., 2009). This topology is depicted in Figure 2a. The cost Customer DXC ADM Federal Ring (a) SONET with DXC Customer DXC ADM Federal Ring (a) SONET with DXC Power plant Substation Connection point Electricity network (b) Power network analogy Figure 2. Fiber optical rings (Pelleau et al., 2009) and power network analogy Customer DXC ADM Federal Ring (a) SONET with DXC Power plant Substation Connection point Electricity network (b) Power network analogy Power plant Substation Connection point Electricity network (b) Power network analogy g (a) SONET with DXC Figure 2. Fiber optical rings (Pelleau et al., 2009) and power network analogy Figure 2. Fiber optical rings (Pelleau et al., 2009) and power network analogy of DXCs is much higher than that of ADMs, so the number of DXCs must be minimized, that is the number of rings. This is known as the SONET Ring Assignment Problem (SRAP) with capacity constraints. This can be formally depicted as a node-partitioning problem for a given graph G (Goldschmidt et al., 2003). Nodes of G stand for the customers to be connected and the edge weights represent the traffic demand between sites. The analogy with power networks is depicted in Figure 2b and can be summarized as follows: we may think of the federal ring as the electricity network and of the DXCs as the differ- ent substations connected to it. Various power plants can then be connected to each substation/local ring with respect to the available hosting capacity. Those power plants can be of different type, such as dispatchable (biomass, hydroelec- tricity, geothermal, etc.) and non-dispatchable (solar, wind, etc.) RE sources 43 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] for example. Each ring may be seen as an aggregate of power plants that are spatially close enough to each other to be connected to the same substation. 2.2 Adapting SONET constraints to the RE planning problem We now show how constraints for this problem can be derived and adapted from the SRAP (Goldschmidt et al., 2003), as well as developed to answer specific n eeds. In order to avoid combinatorial explosion, the spatio-temporal GIS model is used to determine pairwise distance between all solar PV potential sites belonging to the set P S. For each ring k, we will therefore limit the available sites to the set Bk, that is the set of site indices for which sites P Sj have distance dkj from site P Sk below a threshold D (including P Sk): Bk = {j | j ∈N = {1, . . . , n}, dkj ⩽D} ∀k ∈N = {1, . . . , n} (1) (1) We then apply the constraints from the SONET problem to revisit the former optimisation problem (see Table 1 for the nomenclature): > k  i  h SAik × Ppvh,i (2a) max < b  k  i  h SAik × Ppvh,i (2a) (2a) s.t. i∈Bk SAik × Pnom ⩽C ∀k ∈N = {1, . . . , n}, (2b)  k xik ⩽1 ∀i ∈N = {1, . . . , n}, (2c)  i∈Bk xik ⩽M × yk ∀k ∈N = {1, . . . , n}, (2d) M ×  i∈Bk xik ⩾yk ∀k ∈N = {1, . . . , n}, (2e)  i∈Bk  k SAik × Ppvh,i + Einth ⩽0.35 × Demh ∀h, (2f)  i∈Bk  k SAik × Ppvh,i + Einth + Eph ⩽Demh ∀h, (2g) SAik ⩽Smaxi × xik, ∀i ∈Bk, ∀k ∈N = {1, . . . , n}, (2h) xik × Smin ⩽SAik, ∀i ∈Bk, ∀k ∈N = {1, . . . , n}, (2i) xik, yk ∈{0, 1}. s.t. i∈Bk SAik × Pnom ⩽C ∀k ∈N = {1, . . . , n}, (2b)  k xik ⩽1 ∀i ∈N = {1, . . . , n}, (2c)  i∈Bk xik ⩽M × yk ∀k ∈N = {1, . . . , n}, (2d) M ×  ⩾ ∀k ∈N {1 } (2 ) M ×  i∈Bk xik ⩾yk ∀k ∈N = {1, . . . , n}, (2e) Let xik = 1 if site PSi is selected and belongs to ring k and xik = 0 otherwise. Let yk = 1 if power plants are connected to ring k (i.e. ring is active) and yk = 0 otherwise. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] GIS and optimization modelling GIS and optimization modelling GIS and optimization modelling Satisfaction of the forecast energy demand is defined by constraint (2g), using existing resources augmented with new PV generation. Constraints (2h) and (2i) relate the size of the PV sites, lying between Smin and Smaxi, to whether they are selected or not. This relationship is required to link both the energy production and the different costs. If a plant size is not null then the site is selected, and conversely if a site is not selected then its size is forced to be null. Satisfaction of the forecast energy demand is defined by constraint (2g), using existing resources augmented with new PV generation. Constraints (2h) and (2i) relate the size of the PV sites, lying between Smin and Smaxi, to whether they are selected or not. This relationship is required to link both the energy production and the different costs. If a plant size is not null then the site is selected, and conversely if a site is not selected then its size is forced to be null. Essentially the strategic energy planning over some given time horizon has two main objective functions: 1) to maximize the total hourly energy produc- tion over the year through new PV energy generation (2a), 2) to minimize the total costs related to PV installation, connection to the grid, etc. Since these functions are in different units they are not combined into a single weighted function that would not be meaningful, but instead solved by seeking the Pareto frontier, i.e. optimizing each function while constraining the other one. As a result, the optimization problem (2a)-(2i) is applied for every given constrained cost value in the range of the Pareto. Minimize costs: Modeling non-linear functions. The total cost Cost cor- responds to the sum of the costs in every ring k. Each cost in ring k is the aggregate of the capital cost Capk of all the PV sites built in the ring, plus the connection cost to the grid Cconk, the substation cost Cstak and the opera- tional & maintenance costs Copk: Cost =  k Capk + Copk + Cconk + Cstak (3) (3) Table 1. Nomenclature for the OPSPV problem formulation Table 1. Nomenclature for the OPSPV problem formulation h ∈H = {1, . . . , 8760} Hour (per year) i ∈N = {1, . . . [ISBN : 978-2-910545-12-1] , n} Site index k ∈N = {1, . . . , n} Ring index n Number of candidate sites derived from GREECE xik, yk Boolean decision variables Bk Boundary set of nearest PV sites corresponding to ring k C Substation maximum hosting capacity (kW) M Big number to enforce Boolean inference Einth Current hourly production from intermittent energy sources (kWh) Eph Current hourly production from dispatchable energy sources (kWh) Demh Estimated global (forecasted) hourly power demand (kWh) P nom Nominal power per unit area (kW/m2) SAik Surface area of new selected PV site belonging to ring k (m2) Smin, Smaxi Minimum and maximum area for each candidate parcel (m2) P Si Potential Site P pvh,i Estimated hourly production per PV unit (kWh/m2) dgi Minimal distance from the grid to PV site centroid (m) dij Distance between site P Si and site P Sj Clan Transmission line unit cost (e/m) Ccapk Capital cost of implementation of new PV power plants (e) Copk Annual operational cost per PV power plant (e) Cconk Connection costs for each ring, transmission lines (e) Cstak Capital cost for new substation per ring (e) Table 1. Nomenclature for the OPSPV problem formulation h ∈H = {1, . . . , 8760} Hour (per year) i ∈N = {1, . . . , n} Site index k ∈N = {1, . . . 2.2 Adapting SONET constraints to the RE planning problem Aggregate of PV power per ring is kept below a threshold C corresponding to the substation maximum hosting capacity (2b). Constraint (2c) ensures that a selected site only belongs to one ring. Constraints (2d) and (2e) guarantee that a ring with one or more selected PV sites is active. Constraint (2f) prevents the amount of intermittent energy from being over 35 % of the total forecast energy demand (CTG, 2017; EDF, 2017). 44 Communications 44 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] , n} Ring index n Number of candidate sites derived from GREECE xik, yk Boolean decision variables Bk Boundary set of nearest PV sites corresponding to ring k C Substation maximum hosting capacity (kW) M Big number to enforce Boolean inference Einth Current hourly production from intermittent energy sources (kWh) Eph Current hourly production from dispatchable energy sources (kWh) Demh Estimated global (forecasted) hourly power demand (kWh) P nom Nominal power per unit area (kW/m2) SAik Surface area of new selected PV site belonging to ring k (m2) Smin, Smaxi Minimum and maximum area for each candidate parcel (m2) P Si Potential Site P pvh,i Estimated hourly production per PV unit (kWh/m2) dgi Minimal distance from the grid to PV site centroid (m) dij Distance between site P Si and site P Sj Clan Transmission line unit cost (e/m) Ccapk Capital cost of implementation of new PV power plants (e) Copk Annual operational cost per PV power plant (e) Cconk Connection costs for each ring, transmission lines (e) Cstak Capital cost for new substation per ring (e) 45 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Capital cost Capk and operational & maintenance costs Copk are defined as piecewise linear functions (Pillot et al., 2020), and depend on the nominal power range of the PV sites implemented in ring k. Regarding connection costs Cconk, using the centroid of all sites as the connection bridge to the sites would in fact make the problem nonlinear. To keep linearity, we have thus considered in first approximation the maximum distance to the grid among PV sites implemented in ring k: Cconk = Clan × max i∈Bk(dgi × xik) ∀k (4) (4) where Clan stands for the unit cost of transmission lines (e/km). Finally, following the same idea as for the plant capital cost, we also define the sub- station cost Cstak as a piecewise function. The piecewise model allows for the problem to remain linear and for economies of scale in building substations to be included into the analysis. The substation cost depends on the aggregated PV nominal power in ring k, which actually determines the final substation hosting capacity: where Clan stands for the unit cost of transmission lines (e/km). [ISBN : 978-2-910545-12-1] Finally, following the same idea as for the plant capital cost, we also define the sub- station cost Cstak as a piecewise function. The piecewise model allows for the problem to remain linear and for economies of scale in building substations to be included into the analysis. The substation cost depends on the aggregated PV nominal power in ring k, which actually determines the final substation hosting capacity: Cstak =      a4 × Pnom × ( i∈Bk SAik) + y4 if 0MW ≤( i∈Bk SAik) × Pnom ≤10MW a5 × Pnom × ( i∈Bk SAik) + y5 if 10MW ≤( i∈Bk SAik) × Pnom ≤50MW a6 × Pnom × ( i∈Bk SAik) + y6 if 50MW ≤( i∈Bk SAik) × Pnom (5) (5) 3 Case study: ring vs. site approach in French Guiana In this section, we compare the SONET-based ring approach with the site approach (Al-Kurdi et al., 2019; Pillot et al., 2020). We have applied both methods in French Guiana, according to characteristics and policy targets for the horizon 2030 in the region. In the context of the French Energy Transition Act, it is expected to triple RE source capacities by 2023 (CTG, 2017), first by increasing solar PV, then biomass. 3.1 Data and processing The spatio-temporal GIS GREECE model first extracts the potential sites that will feed the OPSPV module depending on geographical constraints and land management. The model prunes the unrestricted territory in order to finally get suitable parcels with respect to minimum and maximum surface area. We refer the reader to (Pillot et al., 2020) for further details about the methodology. It resulted in a set PS of 133 land parcels between 1.5 and 50 ha, depicted in Figure 3. They correspond to all the potential sites where solar PV plants could be built, such that all the constraints hold. We will rely on those plots to compare both ring and site approaches. Beyond geographic location, each potential site generated by GREECE also comes along with specific features and resource time series (e.g. solar irradia- tion). Those features are extracted from heterogeneous data layers and remote sensing data. Essentially, GREECE allows for de-spatialization and discretiza- 46 Communications 46 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] GIS and optimization modelling 5°0′N 5°24′N 54°0′W 53°36′W 53°12′W 52°48′W 5°1′N 5°3′N 52°38′W 52°36′W 52°34′W Figure 3. Set of potential sites derived by the GREECE model in Fr. Guiana. GIS and optimization modelling Figure 3. Set of potential sites derived by the GREECE model in Fr. Guiana. tion of the potential sites in order to feed the combinatorial optimization model with items defined as tuples of digitalized attributes. In the present study, for simplicity purposes, items have been defined as tuples of 3 attributes: land sur- face area, distance to the grid, and time series of solar irradiation values. Our optimization model then converts those values into maximum nominal power, connection costs, and solar PV plant output power profiles. Finally, hourly energy demand for the 2030 horizon has been projected according to 2016 records (EDF, 2019) and EDF estimations for worst case (5% annual growth) and best case (2% annual growth). 3.2 Results and analysis In this paper, we compare two main aspects of relevance to the decision maker and network manager (power plant energy investors in French Guiana and EDF) between the GREECE-OPSPV model (site approach) and our proposed SONET-based model (ring approach). These aspects are related to what we previously stated as safer grid operations and better risk adversity manage- ment (see section 2.1). The comparison is based on the worst case scenario implemented in the site approach. In all cases, we only consider rings made of small-scale facilities whose nominal power is not greater than 5 MW. We first compare Pareto solutions for both approaches (Figure 4) in two fair distinct cases: (a) only small-scale facilities up to 5 MW can be built in the site ap- proach, while no limitation is set on the ring hosting capacity (i.e maximum power that can be injected into the corresponding substation, sum of all PV sites’ power attached to that ring); (b) solar PV sites up to 20 MW can be built in the site approach, and so the same applies to ring hosting capacity in the ring approach, which cannot exceed 20 MW as well. The 20 MW limit corresponds to the specific technical constraints of the French Guiana power network (CTG, 2017; EDF, 2017). This comparison investigates the relative contribution of each approach with respect to economical benefits, and risk management in terms of spatial distribution of the sites. 47 47 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 0 20 40 60 80 100 120 140 160 0 10 20 30 40 50 60 70 80 Energy (GWh) Park approach Ring approach Cost (M€) (a) site capacity ⩽5 MW, no limitation on ring hosting capacity 0 20 40 60 80 100 120 140 160 Cost (M€) 0 10 20 30 40 50 60 70 80 Energy (GWh) Park approach Ring approach (b) Site capacity ⩽20 MW, Ring hosting capacity ⩽20 MW Figure 4. Pareto solutions for ring and site approach. Table 2. Characteristics of the solar PV sites depicted in Figure 5a Table 2. Characteristics of the solar PV sites depicted in Figure 5a Table 2. Characteristics of the solar PV sites depicted in Figure 5a Table 2. Characteristics of the solar PV sites depicted in Figure 5a site ID Nominal power (MW) site 1 6.87 site 2 20 site 3 18.52 Total 45.39 site ID Nominal power (MW) site 1 6.87 site 2 20 site 3 18.52 Total 45.39 Table 3. Characteristics of the rings depicted in Figure 5b Ring ID Number of sites Installed power (MW) site capacities (MW) Ring 1 2 5.56 {4.72, 0.84} Ring 2 4 19.3 {5, 5, 4.74, 4.56} Ring 3 5 19.99 {5, 5, 5, 2.8, 2.19} Total 11 44.85 — 3.2 Results and analysis 0 20 40 60 80 100 120 140 160 0 10 20 30 40 50 60 70 80 Energy (GWh) Park approach Ring approach Cost (M€) (a) site capacity ⩽5 MW, no limitation on ring hosting capacity 0 20 40 60 80 100 120 140 160 Cost (M€) 0 10 20 30 40 50 60 70 80 Energy (GWh) Park approach Ring approach (b) Site capacity ⩽20 MW, Ring hosting capacity ⩽20 MW Figure 4. Pareto solutions for ring and site approach. Selected PV sites Power network Potential PV sites Legend Park 1 Park 2 Park 3 (a) Selected PV sites in site approach (site capacity ⩽20 MW) Ring 1 (2 parks) Ring 2 (4 parks) Ring 3 (5 parks) Power network Potential PV sites Legend (b) Selected rings in ring approach (ring hosting capacity ⩽20 MW) Figure 5. site vs. ring approach spatial results for Pareto solution of Figure 4b (selected cost value = 80 Me). 0 20 40 60 80 100 120 140 160 Cost (M€) 0 10 20 30 40 50 60 70 80 Energy (GWh) Park approach Ring approach (b) Site capacity ⩽20 MW, Ring hosting capacity ⩽20 MW 0 20 40 60 80 100 120 140 160 0 10 20 30 40 50 60 70 80 Energy (GWh) Park approach Ring approach Cost (M€) (a) site capacity ⩽5 MW, no limitation on ring hosting capacity 80 10 Cost (M€) (a) site capacity ⩽5 MW, no limitation on ring hosting capacity (a) site capacity ⩽5 MW, no limitation on ring hosting capacity (b) Site capacity ⩽20 MW, Ring hosting capacity ⩽20 MW (b) Site capacity ⩽20 MW, Ring hosting capacity ⩽20 MW g g p y p y Figure 4. Pareto solutions for ring and site approach. Selected PV sites Power network Potential PV sites Legend Park 1 Park 2 Park 3 (a) Selected PV sites in site approach (site capacity ⩽20 MW) Ring 1 (2 parks) Ring 2 (4 parks) Ring 3 (5 parks) Power network Potential PV sites Legend (b) Selected rings in ring approach (ring hosting capacity ⩽20 MW) Ring 1 (2 parks) Ring 2 (4 parks) Ring 3 (5 parks) Power network Potential PV sites Legend (b) Selected rings in ring approach (ring hosting capacity ⩽20 MW) Figure 5. site vs. ring approach spatial results for Pareto solution of Figure 4b (selected cost value = 80 Me). 3.2 Results and analysis Communications 48 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 GIS and optimization modelling GIS and optimization modelling When only small-scale solar PV sites can be integrated into a power network (a), the Pareto solutions depicted in Figure 4a show that the ring approach gives better results, with for instance 10 MW of extra installed power and 16 GWh of extra power generation compared to the site approach for the same overall cost (60 Me). When both approaches are subject to the same constraints (b), the resulting Pareto solutions remain similar in both cases regardless of the cost value (see Figure 4b). However, as stated in section 2.1, small-scale power stations allow for more manageable risk adversity once it is operating. Hence the ring approach gives the best energy planning strategy with respect to both the decision maker and the network manager needs. To describe the resulting spatial difference between both models, we plot in Figure 5 the selected PV sites against the selected rings regarding the second case (Figure 4b) and with respect to the same Pareto solution (cost value = 80 Me). In both examples, the selected PV sites are scattered throughout the Northern shore of French Guiana, along the power network (i.e. cost is lower for sites close to the grid). The total installed power is similar (45.4 MW against 44.9 MW) but is allocated among 3 facilities in the site approach (Figure 5a), and among 3 rings and 11 small-scale facilities in the ring approach (Figure 5b). The characteristics of the PV sites in both examples are given in Table 2 and Table 3. 4 Conclusion In this paper, we exploited the strengths of spatial decision support, applied to the field of energy transition. In particular, we studied the integration of spatial analysis on spatio-temporal data (solar radiation) as a pre-processing step to an optimization module. This module introduces a novel approach to planning the placement of solar PV sites, given the GIS computation of potential sites, such that the costs are further optimized compared to a more traditional approach. We showed and modelled an analogy with the SONET model, to aggregate Communications 49 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] potential sites around one shared substation that feeds the power grid. The results show substantial gain in costs per KWh produced, as well as enhanced energy stability and risk management given a larger number of selected sites. Further and ongoing work include the extension of our GIS and optimization model, to assess biomass potentials and its energy power embedded into the grid. References Al-Kurdi N., Pillot B., Gervet C., Linguet L. (2019). Towards robust scenarios of spatio-temporal renewable energy planning: A gis-ro approach. In International conference on principles and practice of constraint programming, pp. 729–747. CTG. (2017). Programmation pluriannuelle de l’énergie (PPE) 2016-2018 et 2019- 2023 de la Guyane. Technical report. Collectivité Territoriale de Guyane. EDF. (2017). Systèmes énergétiques insulaires Guyane - bilan prévisionnel de l’équilibre offre / demande d’électricité. Technical report. Paris, France. EDF. (2019). Open Data EDF Guyane. Retrieved from https://opendata-guyane.edf .fr/pages/home/ (Accessed: 2019-04-30) Goldschmidt O., Laugier A., Olinick E. V. (2003). SONET/SDH ring assignment with capacity constraints. Discrete Applied Mathematics, Vol. 129, No. 1, pp. 99–128. Gorsevski P. V., Cathcart S. C., Mirzaei G., Jamali M. M., Ye X., Gomezdelcampo E. (2013). A group-based spatial decision support system for wind farm site selection in Northwest Ohio. Energy Policy, Vol. 55, pp. 374–385. Hache E., Palle A. (2019). Renewable energy source integration into power networks, research trends and policy implications: A bibliometric and research actors survey analysis. Energy Policy, Vol. 124, No. September 2018, pp. 23–35. Notton G., Voyant C., Duchaud J. L. (n.d.). Difficulties of solar PV integration in island electrical networks - Case study in the French islands. Pelleau M., Van Hentenryck P., Truchet C. (2009). SONET network design problems. Electronic Proceedings in EPTCS, Vol. 5, pp. 81–95. Pillot B., Al-Kurdi N., Gervet C., Linguet L. (2020). An integrated GIS and robust optimization framework for solar PV plant planning scenarios at utility scale. Applied Energy. Ramirez Camargo L., Stoeglehner G. (2018). Spatiotemporal modelling for integrated spatial and energy planning. Energy, Sustainability and Society, Vol. 8, No. 1, pp. 29. Spaes J. (2019). Inauguration de la centrale solaire avec stockage de Savane des Pères en Guyane. https://www.pv-magazine.fr/2019/10/17/inauguration-de-la-centrale -solaire-avec-stockage-de-savane-des-peres-en-guyane/. Tapachès E., Perez R., Lauret P., Perez M., David M. (2019). Mitigation of the variability of a PV fleet via geographical dispersion and energy storage systems on the Reunion island non-interconnected grid. In 2019 ieee 46th photovoltaic specialists conference (pvsc), p. 2082-2086. 50 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Les marais littoraux sont des systèmes écogéomorphologiques complexes, remplissant de nombreuses fonctions écosystémiques, notamment en matière de support pour la biodiversité (McOwen et al., 2017), de fourniture de matière première, de séquestration du dioxyde de carbone atmosphérique (Mcowen et al., 2017) ou encore de régulation face aux risques de submersion marine (Mury et al., 2020) et d’érosion côtière. Or ces systèmes, et les fonctions qui y sont associées, se trouvent, de par leur situation géographique, à l’interface entre terres et mers, menacés par les changements environnementaux globaux, parmi lesquels la hausse du niveau moyen des mers (IPCC, 2014). Néanmoins, les marais littoraux disposent d’un atout non-négligeable pour faire face à ces changements et ainsi s’adapter à ces modifications des conditions environnementales : leur capacité naturelle à s’exhausser en captant les sédiments (Stumpf, 1983 ; Allen, 2000 ; French, 2006 ; Détriché et al., 2011 ; McKee et al., 2012). En effet, si le phénomène d’accrétion est un processus naturel au sein des baies, en raison de la dissipation de l’énergie qui s’y produit et de l’apport en matériel sédimentaire d’origine continentale par les fleuves côtiers, ce phénomène s’accroît en présence de surfaces de marais littoraux. Ceci s’explique par les caractéristiques éco-physiologiques des espèces végétales qui constituent les marais, et qui, par l’intermédiaire de leurs systèmes caulinaires et racinaires, présentent des capacités de piégeage et de fixation des sédiments en suspension lors des épisodes d’immersion qui les affectent, permettant ainsi l’élévation des sols (Stumpf, 1983). Ce phénomène d’accrétion a notamment été quantifié en baie du Mont-Saint- Michel, par Détriché et al. (2011) pour les surfaces de marais littoraux, sur la période 1999-2005 et par Berger et Caline (1991), ainsi que par Larsonneur (1989), pour les surfaces sablo-vaseuses. Ces études réalisées en baie du Mont-Saint-Michel, font état de taux de sédimentation variables en fonction du gradient altitudinal et de la nature biogéomorphologique des sites d’observation. Ainsi, pour les surfaces de marais littoraux, trois taux ont été quantifiés, dépendant de la physiographie des observations : le bas schorre (3,45-38,11 mm/an), le moyen schorre (> 4,91 mm/an), et le haut schorre (>1,35 mm/an). Pour les surfaces sablo-vaseuses, deux taux moyens ont été déterminés par des méthodes de datation au 210Pb (Berger et Caline, 1991), venant corroborer les analyses cartographiques réalisées par Larsonneur (1989), avec respectivement environ 4,5 mm/an et 15 mm/an, pour la slikke (vase) et la frange sablo-vaseuse du marais. Prospective 2050 : étude de cas sur l’évolution du service écosystémique de protection côtière Antoine Mury1, Antoine Collin1, Dorothée James1 1. EPHE, PSL Université Paris, CNRS LETG, 15 Boulevard de la Mer, 35800 Dinard, France antoine.mury@etu.ephe.psl.eu 1. EPHE, PSL Université Paris, CNRS LETG, 15 Boulevard de la Mer, 35800 Dinard, France antoine.mury@etu.ephe.psl.eu RÉSUMÉ. Face aux changements environnementaux, induits par le changement climatique global, et notamment face la hausse du niveau moyen des mers, prévue et déjà observé à travers le monde par le groupe d’experts intergouvernemental sur l’évolution du climat (GIEC), certains systèmes écogéomorphologiques font preuve de résilience et d’adaptation. C’est le cas notamment des marais littoraux, qui grâce à leur capacité à piéger les sédiments, parviennent à s’exhausser progressivement, en suivant le rythme de hausse du niveau marin. Cependant, il convient de s’interroger au sujet de l’aptitude de ces derniers à maintenir leurs niveaux de services écosystémiques face à ces changements environnementaux, et tout particulièrement leur capacité de protection contre le risque naturel de submersion marine. Ainsi, cette étude vise à modéliser de manière heuristique et prospective le service écosystémique de protection côtière assuré par les systèmes écogéomorphologiques de la baie du Mont-Saint-Michel à l’horizon 2050, dont la topographie future a été simulée à partir des taux de sédimentation recensés dans la littérature. ABSTRACT. Facing the environmental changes resulting from global climate change, and particularly the rise in mean sea levels, predicted and already observed worldwide by the Intergovernmental Panel on Climate Change (IPCC), some ecogeomorphological systems are showing resilience and adaptation. This is especially the case, in certain geographical contexts, of salt marshes, which, thanks to their capacity to trap sediments, manage to rise gradually, following the sea level rise rhythm. However, questions need to be asked about their ability to maintain their levels of ecosystem services in the face of these environmental changes, particularly their ability to protect against the natural risk of marine flooding. Thus, this study aims to model, in a heuristic and prospective manner, the ecosystem service of coastal protection provided by the ecogeomorphological systems of the Bay of Mont-Saint- Michel by 2050, whose future topography was simulated from the sedimentation rates identified in the literature. MOTS-CLÉS : Submersion marine, service écosystémique, modélisation, adaptation, prospective. KEYWORDS: Marine flooding, ecosystem service, modelling, adaptation, prospective KEYWORDS: Marine flooding, ecosystem service, modelling, adaptation, prospective 51 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Ces taux de sédimentation, peuvent être comparés aux tendances régionales mesurées entre 1970 et 2016 en matière de hausse du niveau marin, qui font état d’une hausse annuelle comprise entre 2 et 3 mm, inférieure donc aux 3,3 mm/an prédits sur cette même période à l’échelle globale (IPCC, 2014). La comparaison de ces taux d’élévation, du niveau marin, d’une part, et des systèmes 52 Communications 2.1. Site d’étude Cette étude a été réalisée sur un site d’environ 3 km² dans la partie occidentale de la baie du Mont-Saint-Michel, présentant des enjeux humains (bâtiments, axes de communications) ainsi que des systèmes écogéomorphologiques (marais littoraux, cheniers coquilliers, estran sablo-vaseux) représentatifs du paysage de la baie et susceptibles de fournir un service écosystémique de protection contre la submersion marine (FIGURE 1). FIGURE 1. Situation géographique générale du site d’étude et localisation des stations de mesure des hauteurs significatives des vagues. FIGURE 1. Situation géographique générale du site d’étude et localisation des stations de mesure des hauteurs significatives des vagues. Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Prospective 2050 écogéomorphologiques, d’autre part, mettant en évidence une vitesse d’accrétion du marais supérieure à la vitesse de hausse du niveau marin, permet de dresser l’hypothèse d’une résilience des systèmes écogéomorphologiques littoraux de la baie du Mont-Saint-Michel face à cette hausse du niveau marin. Se pose néanmoins la question de l’évolution des fonctions écosystémiques, notamment de protection côtière, dans ce futur contexte local. Cette étude vise donc à modéliser de manière exploratoire et prospective, à l’horizon 2050, la capacité d’atténuation des vagues par les systèmes écogéomorphologiques côtiers de la baie du Mont-Saint-Michel, dont l’élévation future a été simulée à partir des taux de sédimentation précédemment évoqués. L’objectif étant in fine d’observer les effets de cette modification topographique sur l’évolution du service écosystémique de protection. 2.3.1. Données originales Les 16 bandes spectrales de l’imagerie passive satellite WorldView-3 (WV-3) (acquise le 20 février 2019), ainsi que les produits dérivés de l’imagerie active LiDAR (Light Detection And Ranging, acquise entre le 2 mai et le 4 juillet 2018) sont utilisées comme prédicteurs de la modélisation de l’atténuation des Hm0, qui permet d’estimer la protection offerte par les systèmes écogéomorphologiques littoraux contre le risque de submersion marine. Les 16 bandes spectrales de l’imagerie satellite, d’une résolution spatiale originale de 1.20 m, ont été corrigées géométriquement et radiométriquement avant d’être affinées à une résolution spatiale de 0,30 m par l’intermédiaire d’une procédure de pan-sharpening (Gram-Schmidt) basée sur la finesse de la bande panchromatique (résolution spatiale 0,30 m × 0,30 m, cf. Mury et al., 2020b). Des modèles numériques de surface (MNS) et d’élévation (MNE) sont réalisés à partir des données de l’imagerie LiDAR avec une résolution spatiale de 0,30 m × 0,30 m et une précision verticale inférieure à 0,2 m. 2.2.2. Méthodologie de calcul des atténuations 2.2.2. Méthodologie de calcul des atténuations À partir des valeurs de Hm0 acquises lors de la campagne de mesure in situ, des valeurs d’atténuation sont calculées en %/m pour chaque paire de stations de mesure. Ces valeurs d’atténuation constituent les réponses statistiques à la modélisation de l’atténuation des vagues, induite par les systèmes écogéomorphologiques littoraux. 2.3. Données d’imageries spatioportées superspectrales WorldView-3 et aéroportées LiDAR 2.3.2. Données simulées Afin d’observer les effets de l’évolution de la topographie du site sur le service d’atténuation des vagues, à l’horizon 2050, les valeurs des MNS et MNE dérivés de l’imagerie LiDAR ont été modifiées de manière à simuler l’élévation des systèmes écogéomorphologiques à partir des taux annuels recensées au sein de la littérature. Pour chaque secteur du site, préalablement identifié par photo-interprétation, a été attribué un taux d’accrétion annuelle (FIGURE 2), lequel est ensuite multiplié par 32 (nombre d’années entre les deux dates) pour simuler l’élévation prévisionnelle entre 2018 et 2050. 54 Communications 2.2.1. Méthodologie d’acquisition des mesures de vagues Les hauteurs significatives des vagues (Hm0) ont été acquises in situ à l’occasion d’une campagne de mesure de quatre cycles de marées consécutifs entre les 22 et 23 53 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] janvier 2019, à l’aide de 17 capteurs de pressions (NKE SP2T10) réparties sur 4 transects transversaux (FIGURE 1) et enregistrant à une fréquence de 2 Hz. janvier 2019, à l’aide de 17 capteurs de pressions (NKE SP2T10) réparties sur 4 transects transversaux (FIGURE 1) et enregistrant à une fréquence de 2 Hz. 2.2.2. Méthodologie de calcul des atténuations Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Prospective 2050 FIGURE 2. Valeurs d’accrétion sédimentaire par secteurs géographiques. Prospective 2050 FIGURE 2. Valeurs d’accrétion sédimentaire par secteurs géographiques. FIGURE 2. Valeurs d’accrétion sédimentaire par secteurs géographiques. FIGURE 2. Valeurs d’accrétion sédimentaire par secteurs géographiques. 2.4.1. Corrélation des prédicteurs avec les valeurs d’atténuation 2.4.1. Corrélation des prédicteurs avec les valeurs d’atténuation Préalablement à la réalisation des modélisations spatiales, une sélection des cinq prédicteurs les plus corrélés avec les valeurs d’atténuation est réalisée afin de conserver un équilibre statistique, au sein de la modélisation, entre le nombre de prédicteurs et le nombre de réponses statistiques (les valeurs d’atténuation). 2.4. Modélisation spatiale de l’atténuation des Hm0 De manière à évaluer le service écosystémique de protection contre la submersion marine, deux modélisations spatiales de l’atténuation des Hm0 sont réalisées, à partir des valeurs d’atténuation calculées précédemment pour chaque paire de capteurs (cf. 2.2.) : utilisant (1) des prédicteurs spectraux et topographiques originaux, puis (2) utilisant les modèles numériques simulés (cf. 2.3.), par l’intermédiaire de régressions linéaires multiples, tel que : Y = β0+ β1X1+ β2X2+…+ βnXn (1) (1) Où, Y est la valeur d’atténuation, β0 une constante, β1 la valeur du coefficient qui est propre à X, la valeur du prédicteur. (2) (2) Cette manipulation a pour objectif de mettre en évidence les compartiments dont les valeurs d’atténuation moyennes offertes par les éco- et géo-systèmes sont les plus affectées par les changements de topographie attendus d’ici 2050. Cela vise ainsi à observer la variabilité dans le temps du service d’atténuation des Hm 0 induit par les éco- et géo-systèmes du site d’étude. 2.4.2. Conception des compartiments d’évaluation 2.4.2. Conception des compartiments d’évaluation 2.4.2. Conception des compartiments d’évaluation Afin de pouvoir comparer aisément les capacités d’atténuation des Hm0 quantifiées à travers les deux modélisations spatiales, les valeurs d’atténuation en %/ 55 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] m sont moyennées au sein des compartiments de 30 m × 600 m orientés en fonction des houles dominantes affectant le site d’étude. m sont moyennées au sein des compartiments de 30 m × 600 m orientés en fonction des houles dominantes affectant le site d’étude. 2.4.3. Taux global d’évolution du service de protection côtière 2.4.3. Taux global d’évolution du service de protection côtière L’évolution du service d’atténuation des vagues, sous-jacente à l’évolution de la topographie, à l’horizon 2050, peut être étudiée par l’intermédiaire du taux global d’évolution calculé tel que : 2.4.4. Évolution de la capacité adaptative des enjeux La capacité protectrice des éco- et géo-systèmes littoraux peut être abordée en tant que proxy de la capacité adaptative, qui constitue l’une des composantes de la vulnérabilité des enjeux (Mury et al., 2020). La transposition de cette composante, depuis les modèles moyennés, aux éléments du cadre bâti (identifiés préalablement par un algorithme de classification supervisée (« maximum de vraisemblance »), permet donc d’estimer un premier degré de vulnérabilité des enjeux face à l’aléa submersion marine. De manière à pouvoir être comparées de manière diachronique, ces valeurs de capacité adaptative des enjeux font l’objet d’une standardisation entre 0 et 1. 3.1. Corrélation des prédicteurs avec les valeurs d’atténuation Cinq prédicteurs des imageries ont été sélectionnés pour la modélisation spatiale de manière que le nombre de prédicteurs de la modélisation soit inférieur à la moitié du nombre de valeur d’atténuations calculées à partir des mesures in situ. Les cinq prédicteurs des imageries affichant les coefficients de corrélations les plus proches de 1 sont, en ordre croissant, SWIR1, PIR1, PIR2, MNE et MNS (TABLE 2). C’est donc la combinaison de ces prédicteurs, spectraux et topographiques, qui est utilisée pour la modélisation spatiale de l’atténuation des Hm0. TABLE 1. Coefficients de corrélation entre les prédicteurs de la modélisation et les valeurs d’atténuation. Prédicteurs Coefficient de corrélation (Pearson) 56 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Prospective 2050 PIR 1 0,66 PIR 2 0,69 SWIR 1 0,65 MNE 0,71 MNS 0,73 3.2.1. Modélisation originale La modélisation spatiale de l’atténuation des Hm0 réalisée à partir des prédicteurs spectraux PIR1, PIR2, SWIR1 et des prédicteurs topographiques MNE et MNS originaux issus de l’imagerie LiDAR, affiche un coefficient de détermination R² de 0,70 et fait état de valeurs d’atténuation comprises entre -1,25 et 1,88%/m (FIGURE 3a). Cette modélisation permet l’élaboration d’un modèle présentant des valeurs d’atténuation moyennées comprises entre 0,02 et 0,77%/m (FIGURE 3b). FIGURE 3. a) Modélisation spatiale originale de l’atténuation des Hm0 et b) modèle moyenné associé. FIGURE 3. a) Modélisation spatiale originale de l’atténuation des Hm0 et b) modèle moyenné associé. 3.2.2. Modélisation prospective 57 57 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 La modélisation spatiale prospective de l’atténuation des Hm0, réalisées à partir des prédicteurs spectraux PIR1, PIR2, SWIR1 et des MNS et MNE simulés, présente quant à elle des valeurs d’atténuation entre -1,35 et 1,79%/m (FIGURE 4a). Le modèle « à compartiments » réalisé à partir de la modélisation prospective présente des valeurs moyennes d’atténuation des Hm0 comprises entre -0,10 et 0,64%/m (FIGURE 4b). FIGURE 4. a) Modélisation spatiale prospective de l’atténuation des Hm0 et b) modèle moyenné associé. FIGURE 4. a) Modélisation spatiale prospective de l’atténuation des Hm0 et b) modèle moyenné associé. 3.3. Évolution 2018-2050 du service de protection côtière L’étude du taux global d’évolution entre les deux modélisations, montre des valeurs comprises entre -2,60 et 2,60% sur le site d’étude (FIGURE 5a). Lorsque ce taux est étudié par l’intermédiaire du modèle basé sur les valeurs moyennées, les valeurs d’évolution contenues dans les compartiments entre 2018 et 2050 sont comprises entre -1,12% et 0,78% (FIGURE 5b). 58 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Prospective 2050 FIGURE 5. a) Modélisation spatiale du taux global d’évolution et b) modèle moyenné associé. Prospective 2050 FIGURE 5. a) Modélisation spatiale du taux global d’évolution et b) modèle moyenné associé. 3.4. Évolution de la vulnérabilité des enjeux L’application des différents modèles « à compartiments » dont les valeurs ont été standardisées, de 2018 et 2050, successivement, sur les enjeux situés en retrait du trait de côte, fait état de valeurs comprises entre 0,18 et 0,92 pour 2018 (FIGURE 6a), entre 0,18 et 1 pour 2050 (FIGURE 6b). Le taux global d’évolution entre les modélisations, appliqué aux enjeux, est également standardisé et présente des valeurs entre 0 et 1 (FIGURE 6c). 59 59 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] FIGURE 6. Application des modélisations sur les enjeux, à partir a) du modèle original, b) du modèle prospectif, et c) du modèle de taux global d’évolution. FIGURE 6. Application des modélisations sur les enjeux, à partir a) du modèle original, b) du modèle prospectif, et c) du modèle de taux global d’évolution. 4.1. Corrélation des prédicteurs avec les valeurs d’atténuations La présence des bandes spectrales PIR1, PIR2 et SWIR1 de l’imagerie WV-3 et des MNE et MNS dérivés de l’imagerie LiDAR, parmi les prédicteurs présentant les meilleures corrélations avec les valeurs d’atténuation des Hm0, s’explique par leurs capacités respectives à fournir des informations sur les caractéristiques éco- physiologiques des communautés végétales de notre site d’étude, ainsi que sur les caractéristiques topographiques du site, qui sont les principaux éléments régissant la circulation des eaux. En effet, les bandes spectrales PIR1 et PIR2 sont particulièrement adaptées pour caractériser la structure anatomique des feuilles et sont utilisées, comme un indicateur du volume de la végétation (Collin et al., 2018). La bande spectrale SWIR1, quant à elle, révèle l'absorption de la lumière par l'eau, et permet de déterminer l'état de santé et de développement de la végétation (Collin et al., 2018). S’agissant des modèles numériques de surface et d’élévation dérivés de l’imagerie LiDAR, leurs fortes corrélations (0,73 et 0,71, respectivement) avec les 60 Communications 60 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Prospective 2050 valeurs d’atténuation, s’expliquent par leur capacité à fournir des informations attenantes à la topographie du site, la rugosité de la surface et les pentes. 4.2. Modélisation spatiale de l’atténuation des Hm0 Les modélisations spatiales de l’atténuation des Hm0, réalisées sur le site d’étude, mettent en évidence la capacité de protection qu’offrent les systèmes écogéomorphologiques littoraux face aux vagues et à l’aléa submersion marine. Ainsi, pour le modèle réalisé à partir des données originales, les valeurs d’atténuation des Hm0 sont comprises entre -1,25%/m et 1,88%/m sur l’ensemble du site d’étude (FIGURE 3a). La modélisation prospective présente, quant à elle, des valeurs sensiblement similaires, entre -1,35%/m et 1,79% (FIGURE 4a), venant ainsi confirmer l’hypothèse d’un maintien de la capacité de protection offerte par les systèmes écogéomorphologiques, à l’horizon 2050. Ces modélisations spatiales permettent la mise en lumière du rôle protecteur joué par les marais littoraux qui présentent les valeurs d’atténuations les plus importantes du site d’étude. Ce rôle des systèmes écogéomorphologiques, et spécialement du marais littoral, dans l’atténuation des vagues, est tout particulièrement observable à travers les modèles « à compartiments », avec des compartiments présentant des valeurs d’atténuation supérieures sur les secteurs présentant systèmes écogéomorphologiques. Cette représentation des valeurs d’atténuation moyennées, permet ainsi, au-delà des valeurs, d’identifier spatialement les secteurs du site d’étude qui bénéficient d’une capacité de protection moindre de la part des systèmes écogéomorphologiques. Ainsi, sur le site d’étude, il apparaît, à l’analyse des différents modèles, originaux comme prospectifs, que les secteurs offrant un service d’atténuation des vagues plus restreint se localisent sur les secteurs présentant peu ou pas de végétation et une faible topographie, et qui ne présentent donc pas de marais littoral développé. 4.3. Évolution 2018-2050 du service de protection côtière L’étude du taux global d’évolution du service d’atténuation des Hm0, entre les modélisations originales et prospectives, notamment par l’intermédiaire du modèle « à compartiments », met en relief deux constats. En premier lieu, ce taux global d’évolution montre la capacité d’adaptation des systèmes écogéomorphologiques, qui par leur potentiel à s’exhausser naturellement, parviennent à maintenir un service écosystémique performant, bien qu’en léger recul, en matière d’atténuation des Hm0. Ceci est tout particulièrement tangible sur les secteurs de marais littoraux à l’Est et à l’Ouest de la zone d’étude. En second lieu, ce taux global d’évolution permet d’observer l’accroissement des inégalités, à l’horizon 2050, entre les secteurs bénéficiant d’un service de protection performant de la part des systèmes écogéomorphologiques et ceux bénéficiant d’un service moindre, et dont les taux globaux d’évolution sont les plus négatifs. Ainsi les enjeux identifiés comme étant ceux présentant une capacité adaptative moindre avec la modélisation originale (FIGURE 6a), le restent avec la modélisation 61 Communications Communications 61 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] prospective 2050 (FIGURE 6b). Néanmoins les taux globaux d’évolution entre les deux modélisations, originale et prospective, font état d’un recul plus important du service de protection dont bénéficient les enjeux situés derrière les secteurs ne présentant pas de marais littoraux et de cheniers développés (FIGURE 6c). Remerciements Les auteurs remercient l’Agence spatiale européenne pour la fourniture des données d’imagerie satellite superspectrale WorldView-3, ainsi que le Service hydrographique et océanographique de la marine pour la fourniture des données d’imagerie LiDAR. Stumpf, R. (1983). The process of sedimentation on the surface of a salt marsh. Estuarine, Coastal and shelf science 17, 5, p495-508. Références Allen, J.R.L. (2000). Morphodynamics of Holocene salt marshes: a review sketch from the Atlantic and Southern North Sea coasts of Europe. Quaternary Science Reviews 19, p. 1155–1231. Berger, G., Caline, B. (1991). Estimation of sedimentation rates in the Bay of Mont-Saint- Michel (France) by 210 Pb dating technique. A pilot study. Ocenianologica acta, Proceedings of the International Colloquium on the environment of epicontinental seas, Lille, 20-22 March 1990, vol. sp. n°1, p. 137-142. Détriché, S., Susperregui, A-S., Feunteun, E., Lefeuvre, J-C, Jigorel, A. (2011). Interranual (1999-2005) morphodynamic evolution of macro-tidal salt marshes in Mont-Saint-Michel Bay (France). Continental Shelf Research, 31, p. 611-630. French, J. (2006). Tidal marsh sedimentation and resilience to environmental change: exploratory modelling of tidal, sea-level and sediment supply forcing in predominantly allochtonous systems. Marine Geology 235, p. 119–136. Collin, A. ; Lambert, N. ; Etienne, S. (2018). Satellite-based salt marsh elevation, vegetation height, and species composition mapping using the superspectral WorldView-3 imagery. International Journal of Remote Sensing, 39(17), p. 5619-5637. McKee, K.; Rogers, K.; Saintilan, N. (2012). Response of salt marsh and mangrove wetlands to changes in atmospheric CO2, Climate, and sea level. Global change and the function and distribution of wetlands, p. 63-96. Mcowen C.J.; Weartherdon, L.V.; Van Bochove J-W.; Sullivan, E., Blyth S.; Zockler, C.; Stanwell-Smith, D.; Kingston, N.; Martin, C.S.; Spalding, M.; Fletcher, S. (2017). A global map of saltmarshes. Biodiversity data journal 5, e11764. Mury, A., Collin, A., Jeanson, M., James, D., Gloria, H., Pastol, Y., Etienne, S. (2020a), Mapping nature-based marine flooding risk using VHR wave, airborne LiDAR and satellite imagery: the case study of the Dol March (Bay of Mont-Saint-Michel, France), Journal of Coastal Research, Special Issue No. 95, p. 743-747. Mury, A., Collin, A., James, D. (2020b), Protection naturelle contre la submersion, apport de l’intelligence artificielle, Comité français de cartographie, numéro spécial « Cartorisk », soumis. Stumpf, R. (1983). The process of sedimentation on the surface of a salt marsh. Estuarine, Coastal and shelf science 17, 5, p495-508. 62 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Trajectoire socio-écologique d’une commune de la Baie du Mont-Saint-Michel entre 1947 et 2014 Mathilde Letard1, Antoine Collin1, Dorothée James1, Antoine Mury1 Mathilde Letard1, Antoine Collin1, Dorothée James1, Antoine Mury1 1. Introduction Les prés salés sont des écotones à végétation basse situés en zone intertidale. Ils fournissent des services importants aux sociétés côtières (Barbier et al. 2011), dont un support à la biodiversité locale (McOwen et al. 2017), le stockage de carbone atmosphérique (Johnson et al. 2014 ; McOwen et al. 2017) ou une protection naturelle des inondations et de l’érosion (Mury et al. 2020). Régis par des processus complexes, leur suivi est nécessaire pour mieux les préserver des contraintes anthropiques et climatiques (Wiberg et al. 2019) auxquelles ils sont exposés. En effet, facteurs naturels et anthropiques se combinent pour déterminer l’évolution de la surface, de la biodiversité et de la morphologie des prés salés (Townend et al. 2010). Parmi eux, l’élévation, la végétation, la dynamique hydrosédimentaire (Townend et al. 2010 ; Fagherazzi et al. 2004 , 2012), l’eutrophisation (Deegan et al. 2012), les aménagements humains (Bouchard et al. 1995), l’invasion par divers usages interférant avec leur régime naturel (Allen 2000), ou encore les changements climatiques comme la montée du niveau marin (IPCC 2018). Ces mécanismes variés contrôlant l’apport et la fixation sédimentaires essentiels à l’existence des prés salés (Townend et al. 2010 ; Fagherazzi et al. 2004 , 2012) expliquent la diversité des profils d’évolution de ces écosystèmes dans le monde. Si plusieurs études récentes font état d’un recul conséquent de leur surface globale (World Resources Institute 2005), les prés salés de certaines régions tendent au contraire à prograder (Détriché et al. 2011 ; Bouchard et al. 1995). C’est le cas en Baie du Mont Saint Michel, où la diversité des interactions qui conditionnent les marais littoraux est bien illustrée. Le fort marnage et le régime sédimentaire de la baie qui favorise la faible énergie des courants et le comblement sédimentaire leur permettent de progresser vers le large depuis plusieurs décennies. Cependant, la pression anthropique, l’exploitation des ressources naturelles et l’aménagement des côtes influencent également le développement des prés salés (Bouchard et al. 1995, Bromberg Gedan et al. 2009). Le comportement des écotones côtiers mettant en jeu de nombreux paramètres, un raisonnement reconnaissant l’interconnexion de la société et de la biosphère et prenant en compte des forçages exogènes permettrait certainement une meilleure approche des changements qu’ils connaissent (Folke et al. 2011). Mathilde Letard1, Antoine Collin1, Dorothée James1, Antoine Mury1 1. EPHE, PSL Université Paris, CNRS LETG 15 Boulevard de la Mer, 35800 Dinard, France mathilde.letard@ephe.psl.eu; 1. EPHE, PSL Université Paris, CNRS LETG 15 Boulevard de la Mer, 35800 Dinard, France mathilde.letard@ephe.psl.eu; RESUME. Les conséquences environnementales du changement climatique global nous renvoient à la nécessité de prendre en compte l’interconnexion des sociétés et des écosystèmes. Dans cette optique, cet article appréhende l’évolution d’une commune de la Baie du Mont-Saint-Michel comme celle d’un socio-écosystème, dont l’état est déterminé par les interactions entre ses composantes mais aussi par des forçages exogènes. Des séries temporelles de l’occupation du sol sont constituées à partir d’images aériennes anciennes et actuelles et permettent d’analyser la trajectoire socio-écologique de la commune. Les modèles de régressions linéaires attribuent en premier lieu l’anthropisation croissante de la zone à la croissance démographique, puis à des paramètres climatiques (R² = 0.87). Le développement de la population entraîne également la progradation du pré salé à proximité directe du bourg (R² = 0.97). Cependant, les prés salés de la zone semblent approcher d’un changement d’état : après une croissance parallèle à celle du bâti, ils connaissant un net recul qui semble lié à l’augmentation de la surface bâtie au-delà d’un certain seuil. ABSTRACT. Climate change is bringing us back to considering the connection between societies and ecosystems. In this respect, this study aims at approaching the evolution of a small town of the Mont-Saint-Michel bay as the trajectory of a social-ecosystem influenced both by the interactions of its components and by external forcing. Airborne imagery is used to constitute land use time series in order to analyse the social-ecological trajectory of the town. Linear regression models show that population growth is the first explanation to increasing anthropization of the area, before climate factors (R² = 0.87). It is also the main factor of the progression of the salt marsh located right next to the city centre (R² =0.97). However, salt marshes of the sector seem to come close to change of state: after having followed a growing trend parallel to that of the urban area, they are shrinking significantly, which seems to be linked to the augmentation of the urban area beyond a given threshold. MOTS-CLES : socio-écosystème, séries temporelles, imagerie aérienne, modélisation, trajectoires KEYWORDS: social-ecosystems, time series, airborne imagery, modelling, trajectories 63 Communications 63 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 2.1. Site d’étude La Baie du Mont Saint Michel est comprise entre la Péninsule du Cotentin et la Côte d’Émeraude, dans le Golfe Normand-Breton. Elle est caractérisée par un régime macro-tidal, son marnage faisant partie des plus élevés au monde (Archer 2013). La baie est délimitée au sud par le Marais de Dol, polder situé derrière la digue de la Duchesse Anne aux pieds de laquelle s’étendent d’importants prés salés recouvrant environ 40 km² (Valéry et Radureau 2014). Le site d’étude choisi s’étend sur 6.5 km² autour de la commune du Vivier-sur- Mer, située dans la partie Ouest de la Baie du Mont Saint Michel, où les conditions hydrodynamiques sont plus calmes que dans la partie estuarienne de la baie. Délaissée par ses habitants dans la première partie du XXème siècle, la commune a connu un renouveau socio-économique avec l’essor de l’activité mytilicole et, plus largement, conchylicole, matérialisée par le développement du port « Le Vivier-sur- Mer/Cherrueix » qui est aujourd’hui le plus important complexe conchylicole d’Europe. Le socio-écosystème comprend un village, des champs, des prés salés ainsi qu’une partie de la zone intertidale. Un fleuve, dont l’estuaire se trouve au niveau du port, assure un apport sédimentaire aux prés salés depuis les terres, en complément de l’apport de sédiments des courants de la baie. FIGURE 1. Localisation du site d’étude autour du Vivier-sur-Mer FIGURE 1. Localisation du site d’étude autour du Vivier-sur-Mer [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Trajectoire socio-écologique d’une commune ente 1947 et 2014 1. Introduction Cet article s’attache donc à évaluer l’évolution socio-écologique d’une commune côtière de la Baie du Mont Saint Michel via des séries temporelles d’évolution de l’occupation des sols constituées à partir d’images aériennes à haute résolution spatiale. Deux hypothèses sont émises et testées : (1) la démographie, la température, les précipitations, la vitesse des vents et le niveau marin conditionnent l’évolution du pré salé situé dans la zone et (2) expliquent l’anthropisation du site. 64 Communications 64 [ISBN : 978-2-910545-12-1] 2.2.3. Données démographiques et historiques 2.2.3. Données démographiques et historiques 2.2.3. Données démographiques et historiques Pour étudier l’influence de la démographie sur l’évolution paysagère, les recensements de population effectués par l’INSEE tous les 5 à 9 ans entre 1968 et 2007 sur la commune du Vivier-sur-Mer ainsi que les taux de variation de la population – solde naturel et migratoire – entre chaque relevé sont utilisés. D’autres informations sur la démographie, ainsi que sur l’aménagement progressif du port sont extraites de travaux sur l’histoire du Vivier-sur-Mer. 2.2.1. Images aériennes 2.2.1. Images aériennes Dix prises de vues aériennes de sources diverses mises à disposition par l’IGN sont utilisées pour cette étude. Leurs caractéristiques sont résumées dans la Table 1. 65 Communications Communications 65 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] TABLE 1. Prises de vues aériennes utilisées pour l’étude et caractéristiques TABLE 1. Prises de vues aériennes utilisées pour l’étude et caractéristiques TABLE 1. Prises de vues aériennes utilisées pour l’étude et caractéristiques Date Bande d’observation Résolution spatiale 1947 Panchromatique 0.37 m 1952 Panchromatique 0.37 m 1961 Panchromatique 2.72 m 1966 Panchromatique 3.56 m 1978 Panchromatique 1.11 m 1986 Panchromatique 0.73 m 1989 Panchromatique 0.46 m 1996 Panchromatique 0.15 m 2002 Rouge + Vert + Bleu 0.5 m 2014 Rouge + Vert + Bleu + Infrarouge 0.5 m 2.2.2. Données climatologiques 2.2.2. Données climatologiques 2.2.2. Données climatologiques Pour étudier l’influence du climat sur l’évolution paysagère, plusieurs variables extraites des simulations du climat entre 1951 et 2005 réalisées par Météo France, l’Institut Pierre Simon Laplace et le CERFACS sont exploitées. On dispose ainsi, pour chaque jour entre le 1er Janvier 1951 et le 31 Décembre 2005, de températures (en K), de précipitations totales (en kg/m2/s), et de la vitesse des vents (en m/s). 2.3.1. Préparation des imageries aériennes Dix orthomosaïques sont générées à partir des images de chaque campagne via le logiciel Metashape. Elles sont ensuite géoréférencées sur la base de l’Ortholittorale V2 (2013-2014) et découpées selon l’emprise de la zone d’étude (logiciel ENVI). Pour assurer une continuité dans les données employées, une bande panchromatique composite est calculée en sommant les informations des bandes rouge, verte et bleue pour les dates de 2002 et 2014. Toutes les images sont ensuite 66 Communications 66 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Trajectoire socio-écologique d’une commune ente 1947 et 2014 ré-échantillonnées à 1 mètre de résolution spatiale, moyenne raisonnable pour détecter les objets socio-écologiques à très haute résolution spatiale afin de suivre leur dynamique, avec la méthode du plus proche voisin (logiciel ENVI). 2.3.3. Création de séries temporelles d’occupation du sol La surface de chacune des classes est calculée pour chaque date à partir de la carte correspondante, afin d’obtenir des séries temporelles de surface. Pour certains types d’occupation, il est nécessaire de disposer d’une série de valeurs annuelles sur tout l’intervalle de 1947 à 2014. Des interpolations sont donc réalisées en ajustant une courbe de tendance issue de modèles statistiques aux observations. Pour le pré salé et le bâti, les interpolations sont faites avec un polynôme de degré 4. La somme des surfaces des champs et des arbres ainsi que la classe « routes/chemins » sont interpolées linéairement. A chaque fois, le modèle retenu pour l’interpolation – polynômial ou linéaire – est celui pour lequel la valeur du coefficient de Pearson obtenu était la plus élevée. L’eau présente dans la scène ne représente que les lagunes et bassins portuaires créés progressivement au fil du développement du port. Les valeurs intermédiaires sont ainsi ajustées en fonction des connaissances de l’histoire de l’aménagement du port. 2.3.2. Cartographie de l’occupation du sol Une cartographie de l’occupation du sol est ensuite produite à partir de chaque image. Pour cela, une classification par maximum de vraisemblance est effectuée sur une combinaison de 8 attributs de texture obtenus par l’application de filtres sur l’image panchromatique. En effet, la bande panchromatique seule ne suffit pas à distinguer les différentes occupations du sol efficacement. La variance, l’homogénéité, le contraste, la dissimilarité, l’entropie, le moment d’ordre 2, la corrélation, et la valeur moyenne sont donc calculés pour multiplier l’information disponible pour le classifieur. La détection obtenue est ensuite corrigée localement par photo-interprétation, pour éliminer la confusion persistant entre les classes dont la signature spectrale est très proche sur les images panchromatiques anciennes. Les classes définies pour la classification correspondent aux types d’occupation du sol cartographiés : pré salé, sédiments vaseux, sédiments coquilliers, bâti, routes et chemins, arbustive et arborescente, champs (pelouses et cultures), sol nu, eau. 2.3.6. Modèles explicatifs à partir de régressions linéaires Des régressions linéaires multiples sont réalisées pour tenter d’expliquer l’évolution de l’anthropisation du site d’étude et l’évolution de la surface du pré salé. Elles sont menées sur les données normalisées, afin de permettre la comparaison des coefficients de corrélation obtenus sans biais d’échelle. (2) (2) A est l’indice d’anthropisation. La surface de sol artificiel correspond à la somme des surfaces occupées par le bâti, les routes et chemins et l’eau (qui correspond aux bassins portuaires aménagés). La surface de sol naturel est quant à elle obtenue en additionnant les surfaces des champs, des arbres et arbustes, et du pré salé. Le sol nu n’est pas inclus car il est compliqué de caractériser sa présence comme un signe d’anthropisation ou non. 3.1. Cartes de l’occupation du sol pour chaque millésime Les cartes obtenues par classification des 10 orthomosaïques sont présentées en Figure 2. 9 types d’objets sont distingués : les sédiments vaseux, les sédiments coquilliers, le pré salé, les routes et chemins, le bâti, le sol nu, l’eau, les champs (pelouses et cultures) et la végétation arborescente et arbustive. 68 Communications 2.3.4. Création de séries temporelles climatologiques et démographiques Les données climatiques sont moyennées annuellement pour obtenir une série d’une valeur par an pour chaque paramètre entre 1951 et 2005. Les recensements de l’INSEE ne renseignent la population qu’à certaines dates. A partir des données d’évolution démographique, la population est estimée entre 1968 et 2014. Pour les dates antérieures, des données historiques sur la population du Vivier-sur-Mer en 1936, 1957, 1962, 1965 et 1967 sont utilisées et reliées par des valeurs interpolées linéairement. 67 Communications Communications 67 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 2.3.5. Création de séries temporelles de statistiques paysagères Outre la surface de chaque occupation du sol identifiée, l’indice de Shannon est employé pour évaluer la diversité paysagère. Cet indice est défini comme suit : (1) (1) Où H’ est l’indice de diversité de Shannon, S le nombre de classes dans la scène et pi la proportion de la classe i par rapport à l’ensemble des classes. L’indice suivant quantifiant l’anthropisation du paysage sur la partie « terrestre » de la zone (excluant sédiments vaseux et coquilliers) est également utilisé : (2) Communications 68 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Trajectoire socio-écologique d’une commune ente 1947 et 2014 FIGURE 2. Cartes de l’occupation du sol autour du Vivier Sur Mer de 1947 à 2014 : a)1947, b)1952, c)1961, d)1966, e)1978, f)1986, g)1989, h)1996, i)2002, j)2014 FIGURE 2. Cartes de l’occupation du sol autour du Vivier Sur Mer de 1947 à 2014 : a)1947, b)1952, c)1961, d)1966, e)1978, f)1986, g)1989, h)1996, i)2002, j)2014 3.2. Séries temporelles de l’évolution de la surface des différentes occupations du sol 3.2. Séries temporelles de l’évolution de la surface des différentes occupations du sol Ces cartes permettent de dresser l’évolution de la surface des différentes catégories de sol présentée en Figure 3. Les valeurs s’étendent de 0 à 2 506 716 m². 69 Communications 69 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] FIGURE 3. Évolution de la surface des classes d’occupation du sol en mètres carrés de 1947 à 2014 FIGURE 3. Évolution de la surface des classes d’occupation du sol en mètres carrés de 1947 à 2014 Les tendances d’évolution des classes principales sont présentées en Table 2. TABLE 2. Tendances d’évolution des classes d’occupation du sol de 1947 à 2014 Sujet Tendance Coefficient de détermination Pré salé (surface) Positive (polynôme ordre 4) 0.98 Bâti (surface) Positive (polynôme ordre 4) 0.97 Routes/chemins (surface) Positive (linéaire) 0.83 Végétation (surface) Négative (linéaire) 0.95 Diversité du paysage Positive (polynôme ordre 2) 0.65 Anthropisation (indicateur) Positive (polynôme ordre 4) 0.98 TABLE 2. Tendances d’évolution des classes d’occupation du sol de 1947 à 2014 3.3. Modèles explicatifs de l’évolution socio-écologique Les deux hypothèses testées s’avèrent pertinentes sur le plan statistique. L’hypothèse 1 est corroborée par le coefficient de corrélation de Pearson (évaluant 70 70 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Trajectoire socio-écologique d’une commune ente 1947 et 2014 Trajectoire socio-écologique d’une commune ente 1947 et 2014 la pertinence du modèle) dont la valeur atteint 0.93 et par un coefficient de détermination R² (évaluant le pouvoir prédictif du modèle) de 0.87. Pour l’hypothèse 2, le coefficient de corrélation de Pearson vaut 0.97 et le coefficient de détermination R² atteint 0.94. Celle-ci est donc également validée. Les prédicteurs climatiques et démographiques n’ont pas tous le même impact sur l’anthropisation ou la surface du pré salé, comme l’illustre la Figure 4. FIGURE 4. Influence des prédicteurs sur l’anthropisation (a) et sur la surface de pré salé (b) (les valeurs présentées sont celles des coefficients de régression linéaire) FIGURE 4. Influence des prédicteurs sur l’anthropisation (a) et sur la surface de pré salé (b) (les valeurs présentées sont celles des coefficients de régression linéaire) Finalement, on trouve que l’anthropisation et la surface de pré salé peuvent être prédites avec les équations suivantes : A = -0,10 + 1,26 Pop + 0,08 T + 0,07 Nm + 0,03 V – 0,006 P (3) (3) Spré salé = 0,42 + 1,44 Pop – 0,15 Nm – 0,06 V – 0,04 T – 0,01 P (4) (4) Où A correspond à l’indice d’anthropisation à un instant donné, Spré salé à la surface de pré salé à un instant donné et Pop, T, Nm, V et P sont respectivement la population, la température, le niveau marin, la vitesse du vent et les précipitations à ce même instant. 4. Discussion L’étude des séries temporelles de l’occupation du sol permet d’identifier les trajectoires d’évolution du socio-écosystème de la zone du Vivier-sur-Mer. Les cartes et les séries de surfaces obtenues montrent un changement radical du socio- écosystème. En deçà de la digue, on passe d’un bourg entouré d’un paysage bocager à une petite ville au bâti organisé (quartiers résidentiels à voirie curviligne, zone industrielle portuaire) entourée d’un openfield. Les séries temporelles corroborent cette observation : l’évolution de la surface des arbres et celle de la surface des champs sont symétriques. Lorsque les champs s’étendent, l’étendue des arbres et des haies diminue et inversement. L’augmentation significative de la surface bâtie traduit quant à elle le développement de l’activité portuaire et de la ville, et donc l’essor socio-économique à la suite de l’installation du port conchylicole. 71 71 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Au-delà de la digue, le pré salé se développe : sa surface de 2014 représente plus du double de sa surface en 1947, témoignant d’un apport sous-jacent en sédiments important (Valéry et Radureau 2014). Ces changements paysagers sont matérialisés par la croissance de la diversité du paysage, mais aussi et surtout par l’anthropisation, qui connaît une croissance polynomiale depuis 1947. Les résultats des régressions linéaires multiples montrent que l’anthropisation croissante et la progradation du pré salé sont la conséquence de forçages exercés par des paramètres exogènes sur le socio-écosystème. La pression anthropique – traduite par la croissance démographique – est le paramètre influençant le plus l’évolution du paysage. Ces résultats constituent non seulement un modèle prédictif rétrospectif mais aussi prédictif. Le coefficient de détermination élevé obtenu pour les deux modélisations laisse penser que l’évolution future du socio écosystème pourrait être modélisée à partir des projections climatiques du GIEC (IPCC 2018) et des projections de croissance démographique de l’INSEE ou des Nations Unies. L’étude de l’interaction des composantes du socio-écosystème renseigne également sur les trajectoires d’évolution observables. Le graphique de la Figure 4 lie la surface du pré salé (en ordonnée) et celle du bâti (en abscisse). Les valeurs s’étendent de 151 551 pixels à 1 586 695 mètres carrés. FIGURE 5. Surface des prés salés en fonction de la surface du bâti en mètres carrés de 1947 à 2014 FIGURE 5. 4. Discussion Surface des prés salés en fonction de la surface du bâti en mètres carrés de 1947 à 2014 L’étude de l’évolution de la surface du pré salé en fonction de celle du bâti – intrinsèquement liées comme le montrent les résultats des régressions – montre que le développement du bâti entraîne celui du pré salé, les courbes évoluant quasiment parallèlement entre 1960 et 2002. Ceci est cohérent, puisque l’artificialisation des sols change la dynamique d’écoulement des eaux, permettant un apport de sédiments au pré salé via le fleuve. Il semblerait toutefois qu’il existe un pallier d’augmentation du bâti au-delà duquel l’effet inverse se produit : depuis 2002, le 72 Communications 72 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Trajectoire socio-écologique d’une commune ente 1947 et 2014 Trajectoire socio-écologique d’une commune ente 1947 et 2014 bâti continue d’augmenter mais le pré salé connaît un net recul. Ceci peut s’expliquer par l’invasion croissante du pré salé par les activités humaines : entre 2002 et 2014, les infrastructures portuaires s’agrandissent, grignotant toujours plus sur le pré salé, et la zone de pré salé défrichée pour des usages loisir à l’Ouest de la zone est plus étendue. Les facteurs naturels, matérialisés par une tempête en 2014, ont également un rôle-clé dans ce changement de trajectoire. Cependant, on peut supposer que cela laisse présager un changement d’état du pré salé, qui pourrait subir de manière plus marquée les forçages anthropiques (Folke et al. 2011). Il pourrait être intéressant de pousser l’analyse de résilience du socio-écosystème plus loin, afin de mieux anticiper les changements à venir. Pour cela, une analyse de l’évolution des taux de croissance de chacune des classes pourrait être utile, et confirmerait peut-être l’arrivée d’une bifurcation dans la trajectoire d’évolution du pré salé (Scheffer 2010). Il serait également opportun d’exprimer statistiquement le phénomène de débocagisation observable par la simple observation des cartes. En effet, celui-ci a des conséquences évidentes sur le ruissellement des eaux et donc des sédiments terrestres vers le pré salé. Finalement, l’étude montre que l’analyse des tendances d’évolution des différentes occupations du sol d’un socio-écosystème permet d’expliquer les dynamiques d’évolution des prés salés. Il est cependant important de noter que la méthode de classification retenue et le recours à une correction par photo- interprétation ont un impact sur la précision des valeurs de surfaces calculées. 4. Discussion Si les données chiffrées pouvaient être précisées par une méthode de classification plus performante (ex., apprentissage profond, Letard et al. 2020), l’allure des séries temporelles obtenues permettrait toutefois bien d’envisager les trajectoires socio- écologiques rétrospectives et prospectives du site d’étude du Vivier-sur-Mer. Celui- ci semble se rapprocher d’une bifurcation dans son régime d’évolution : la pression anthropique, force motrice principale de la progradation du pré salé, pourrait avoir franchi un seuil et devenir un frein à celle-ci. Bibliographie et références Allen J.R.L. (2000). Morphodynamics of Holocene salt marshes : a review sketch from the Atlantic and Southern North Sea coasts of Europe. Quaternary Science Reviews, vol. 18, n°12, p. 1155-1231. Archer A.W. (2013). World’s highest tides: Hypertidal coastal systems in North America, South America and Europe. Sediment. Geol., vol. 1, n°25, p. 284-285. Barbier E., Hacker S.D., Kennedy C., Koch E.W, Stier A.C. et al. (2011). The value of estuarine and coastal ecosystem services. Ecological Monographs, vol. 81, n°2, p. 169- 193. Bouchard V., Digaire F., Lefeuvre, J-C., Guillon L-M. (1995). Progression des marais salés à l’ouest du Mont-Saint-Michel entre 1984 et 1994. Mappemonde, vol. 4, p.28-34. Deegan, L.A., Johnson D.S., Warren R.S., Peterson B.J., Fleeger J.W. et al. (2012). Coastal eutophication as a driver of salt marsh loss. Nature, vol. 490, n°7420, p. 388-392. 73 73 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Détriché S., Susperregui A-S., Feunteun E., et al. (2011). Interannual (1995-2005) morphodynamic evolution of macro-tidal salt marshes in Mont-Saint-Michel Bay (France). Continental Shelf Research, vol. 3, n°6, p. 611-630. Fagherazzi S., Marani M., Blum L.K. (2004). The ecogeomorphology of tidal marshes. American Geophysical Union, Washington. Fagherazzi S., Kirwan M.L., Mudd S.M., Guntenspergen G.R, Termmerman S. et al. (2012). Numerical models of salt marsh evolution: ecological, geomorphic, and climatic factors. Reviews of Geophysics, vol. 50, n°1. Folke C., Jansson A., Rockström J. et al. (2011). Reconnecting to the Biosphere. AMBIO, vol. 40, n°719. Gedan K.B., Silliman B.R., Bertness M.D. (2009). Centuries of hulman-driven change in salt marsh ecosystems. Annual Review of Marine Science, vol.1, p.117-141. IPCC (2018). Climate Change 2013: The physical science basis. Contribution of Working group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge United Kingdom and New York, NY, USA, 1588 p. Johnson B.J., Lovelock C.E., Herr D. (2018). Climate regulation: salt marshes and blue carbon. The Wetland Book. Springer, Dordrecht, p. 1-12. Letard M., Collin A., Pastol Y. (2020). Improving mangrove and salt marsh mapping using superpixel-based deep classification of VHR superspectral satellite imagery and topobathymetric LiDAR data, Sea Tech Week 2020, 12-15 Oct. 2020, Brest. McOwen C.J., Weatherdon L.V., Van Bochove J-W., Sullivan E., Blyth S. et al. (2017). A global map of saltmarshes. Biodiversity Data Journal, vol. 5. Mury A., Collin A., Etienne S., Jeanson M. (2020). Bibliographie et références Wave attenuation service by intertidal coastal ecogeosystems in the Bay of Mont-Saint-Michel, France: Review and Meta- Analysis. Estuaries and Coastal Zones in Times of Global Change. Proceedings of ICEC- 2018, p. 555-572. Scheffer M. (2010). Foreseeing tipping points. Nature, vol. 467, p. 411-412. Townend I., Fletcher C., Knappen M., Rossington K. (2011). A review of salt marsh dynamics. Water and Environment Journal, vol. 25, p. 477-488. Valéry A., Radureau, A. (2014). Evolution de la végétation des marais salés de la baie du Mont-Saint-Michel—Analyse cartographique 1984–2013. Rapport Convention CREBS/Université de Rennes1/Conservatoire du Littoral/Inter-SAGE baie du Mont-Saint- Michel/Agence de l’Eau Seine-Normandie/Agence de l’eau Loire-Bretagne. Rennes, France, 15p. Wiberg P.L., Fagherazzi S., Kirwan M.L. (2020). Improving predictions of salt marsh evolution through better integration of data and models. Annual Review of Marine Science, vol. 12, n°1, p. 389-413. Millennium Ecosystem Assessment (2005). Ecosystems and Human Well-being: Synthesis. Island Press, Washington, DC. 74 74 Communications Communications par Session Communications par Session Environnement urbain [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Thomas Le Jan 1, Myriam Servières 1, Thomas Leduc 2, Vincent Tourre 1 Thomas Le Jan 1, Myriam Servières 1, Thomas Leduc 2, Vincent Tourre 1 1. AAU, Ecole Centrale de Nantes, Nantes thomas.le-jan@eleves.ec-nantes.fr,firstname.lastname@ec-nantes.fr 2. AAU, CNRS, ENSA Nantes, Nantes thomas.leduc@crenau.archi.fr 1. AAU, Ecole Centrale de Nantes, Nantes thomas.le-jan@eleves.ec-nantes.fr,firstname.lastname@ec-nantes.fr 2. AAU, CNRS, ENSA Nantes, Nantes thomas.leduc@crenau.archi.fr 1. AAU, Ecole Centrale de Nantes, Nantes thomas.le-jan@eleves.ec-nantes.fr,firstname.lastname@ec-nantes.fr ABSTRACT. Isovists are powerful tools for the morphological analysis of urban spaces, allowing us to account for the characteristics of the visible space from a given point of view. However, computing isovist field from digital model is limited by the quality of the model, and computing isovist field of the real environment with LiDaR sensors is limited by its cost and the setup of cumbersome equipment (eg. on a car). To allow the use of lightweight devices, we propose a method to compute isovists from a single monocular camera, catching a path in an urban environment. This method uses SLAM (Simultaneous Localization and Mapping) and ray casting algorithms. Our re- sults are compared against digital model’s isovists. RÉSUMÉ. Les isovists sont des outils performants d’analyse morphologique des es- paces urbains permettant de rendre compte des caractéristiques de l’espace visible à partir d’un point de vue donné. Cependant, le calcul des champs d’isovists à partir d’un modèle numérique est limité par la qualité de ce modèle, et le calcul des champs d’isovists d’environnement réel à l’aide de capteurs LiDaR reste limité par son coût et la mise en place d’un équipement encombrant (par exemple, sur un véhicule). Pour permettre l’utilisation d’équipement portable, nous proposons d’établir une méthode d’obtention d’isovists à partir d’une caméra monoculaire captant un parcours dans un environnement urbain. Cette méthode est basée sur les algorithmes de SLAM (Simul- taneous Localization And Mapping) et de lancer de rayons. Les résultats obtenus par notre méthode sont comparés à ceux obtenus sur un ensemble d’isovists calculés sur un modèle numérique 3D simplifié de la ville. KEYWORDS: Isovists, camera, space reconstruction KEYWORDS: Isovists, camera, space reconstruction MOTS-CLÉS : Isovists, caméra, reconstruction de l’espace KEYWORDS: Isovists, camera, space reconstruction MOTS-CLÉS : Isovists, caméra, reconstruction de l’espace KEYWORDS: Isovists, camera, space reconstruction MOTS-CLÉS : Isovists, caméra, reconstruction de l’espace 77 77 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Isovists and isovist fields are conceptual tools used to carry out morpho- logical analyses of architectural and urban spaces. They were developed and theorized in the 70s (Benedikt, 1979). An isovist determines the available space in a direct line of sight from a given position in an architectural or urban en- vironment. As the visibility of the urban space is part of the urban design qualities identified by urban planners (Ewing, Handy, 2009), isovists are par- ticularly relevant to evaluate the quality of an urban space. Depending on the dimension of the space, these isovists have a polygonal (2D) or polyhedral (3D) shape. They’re usually computed from a digital urban model extracted from a GIS, which is a simplified representation of the urban environment. Some methods can calculate isovists along a route caught with sensors, such as LiDaR (Schmid, Stülpnagel, 2018), but the use of these powerful sensors is not accessible to as many people as possible and this may be very inconvenient because of the cum- bersome equipment that requires additional transport (e.g. on a car). Although there is more consumer LiDaR equipment available, here we focus on sensors that are embedded in everyone’s smartphones. It is possible to reconstruct a map of the environment with lightweight sensors, for example monocular cameras, using SLAM algorithms (Simultaneous Localization And Mapping). SLAM addresses the computational issue of constructing or updating a map of an unknown environment while simultaneously keeping track of the observer’s location. Time-of-flight camera are also available on several smartphones, but their actual depth performances and sensitivity to lighting conditions, among others drawbacks, prevent us to use them to reconstruct outdoor environments. In this article, we propose a method to delineate an isovist field from a video caught on a simple monocular camera (video-based isovists), using SLAM algorithms. To check the consistency of our results, we will confront them to the isovists generated from a digital urban model (mockup-based isovists). [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Outdoor isovist computation with SLAM areas of interest in keyframes and assign a depth to every pixel of these areas. Unfortunately, these methods each present some weaknesses. Indeed, a sparse point cloud may be fairly accurate but, due to its lack of density, it does not permit to reconstruct real obstacles (there are not enough feature points de- tected). Dense and semi-dense methods can return much denser point clouds but the returned maps will be very noisy. Some recent methods (He et al., 2018) can reconstruct a map of the en- vironment with lines instead of points, which makes it easier to reconstruct accurate surfaces using Delaunay triangulation. The method implemented in this paper uses a semi-dense version of ORB-SLAM2 (Mur-Artal, Tardos, 2015) to generate a semi-dense point cloud and extract lines from it. SLAM methods often include loop closure which improves tracking and mapping by recognizing places already visited by the user. Then the associated map points and camera poses are adjusted according to their previous values. There are some methods that allow isovist computation from point clouds (Díaz-Vilariño et al., 2018). These methods are based on space discretization. Once space is subdivided into voxels, we determine if the voxels are occupied or empty by the number of points they contain. Then by ray-casting, we determine if the voxels are visible or obstructed which gives us our isovist. There are limits to these kinds of methods. The first is that the returned isovists won’t be a vector object, as it is computed in a discretized space. The second is that the method expects a dense and accurate point cloud to clearly model the environment, which is not the case of the previous common visual SLAM methods. The other simple way to compute isovists as vector objects is to apply ray- casting on surfaces, instead of points. That is possible with semi-dense-line SLAM (He et al., 2018) and its surface reconstruction method. 2. Previous Works SLAM algorithms provide a fairly accurate estimation of the trajectory of a camera while reconstructing the environment in which the camera evolves. Usually, SLAM algorithms return a point cloud mapping the environment, so- called reconstructed map. Depending on the SLAM method, the point cloud can be sparse (e.g. ORB-SLAM2 (Mur-Artal, Tardos, 2017), PTAM (Klein, Murray, 2007)), dense (e.g. DTAM (Newcombe et al., 2011)) or semi-dense (e.g. (Engel et al., 2014)). The sparse methods detect feature points on keyframes and they are assigned 3D coordinates in the reconstructed map. The dense methods rather give depth to almost every pixel of a keyframe. The map is then reconstructed by overlapping the different depth maps. Finally, semi- dense methods fall midway between sparse and dense methods. They detect 78 Communications 78 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3. Proposal – LIghtweight Real Isovist: LIRI Our method’s workflow is shown in Fig. 1. It is based on semi-dense-line SLAM algorithm (He et al., 2018), which can reconstruct surfaces from semi- dense point cloud obtained by video processing (Fig. 2). We use a smartphone monocular camera but this method is standard enough to be applied to any type of monocular camera provided that we know its intrinsic parameters, which can be computed by camera calibration. So we have a 3D model of the environment made of triangles and an estimation of the sensor trajectory in the reconstructed map. The different poses returned by the algorithm are located in the coordinate system of the first detected keyframe. The global orientation of the reconstructed map then depends on the orientation of the camera lens during the detection of the first keyframe. In order to ease the comparison between video-based isovists and mockup-based isovists and to stabilize the 79 Communications 79 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 1. Workflow of our method. Figure 1. Workflow of our method. skyline, we use an additional phone stabilizer, minimizing the camera orienta- tion error. The phone stabilizer doesn’t impact the standard and lightweight setup hypothesis. The resulting model has some outlier triangles that don’t represent real assets so we apply a filter upon the area and the perimeter of the triangles to eliminate the outliers. This filter was set up experimentally. We use a ray-tracing based method to compute isovists from the recon- structed map of the environment and estimated trajectory. We scan the envi- ronment using a panoptic approach in the horizontal plane of vision. To com- pute ray-tracing, we use the Möller-Trumbore intersection algorithm (Möller, Trumbore, 2005) that can quickly calculate the intersection point between a triangle and a ray in 3D space, without using plane equations. For each ray, we look at every triangle and determine the intersection point, if it exists, and we only keep the nearest intersection point from the ray origin. By connecting the intersection points obtained in the different directions we can reconstruct the isovist for a given position. Finally, if we aggregate the different isovists obtained by the positions along a trajectory, we have the isovists field related to that trajectory. To compare our method to the topographic-based model, we have to trans- form our data in a geographical context. 1. To calculate the mockup-based isovists, we use standard topographic data sets such as the ones provided by the IGN BD TOPO® database (June 2020 edition) and the t4gpd Python plugin (Leduc, Leduc, 2020). 3. Proposal – LIghtweight Real Isovist: LIRI So we have to apply scaling, trans- lation, and homothety (dilation) but also extract geographical coordinates of 80 Communications 80 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Outdoor isovist computation with SLAM Figure 2. Surfaces reconstructed with semidense-line SLAM method. The red line represents the trajectory of the sensor. Figure 2. Surfaces reconstructed with semidense-line SLAM method. The red line represents the trajectory of the sensor. one of the points of the trajectory. In our case we focus on the first point which corresponds to the first moment a keyframe is detected. The position of the camera associated with this first keyframe is predominant; we determine it empirically with the help of orthophotographs. We use a similar method to get the scale factor: we determine manually the geographical location of two points of the trajectory, using a GIS and orthophotographs, and we calculate the dis- tance between them in the reconstructed map and in the topographic-based model. By applying this factor to our data, it is now possible to implement the estimation of the trajectory in the digital model to compute mockup-based isovists1. We can also compare the isovists obtained by both models. one of the points of the trajectory. In our case we focus on the first point which corresponds to the first moment a keyframe is detected. The position of the camera associated with this first keyframe is predominant; we determine it empirically with the help of orthophotographs. We use a similar method to get the scale factor: we determine manually the geographical location of two points of the trajectory, using a GIS and orthophotographs, and we calculate the dis- tance between them in the reconstructed map and in the topographic-based model. By applying this factor to our data, it is now possible to implement the estimation of the trajectory in the digital model to compute mockup-based isovists1. We can also compare the isovists obtained by both models. 81 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 4. Results 82 Communications Communications 82 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Outdoor isovist computation with SLAM To go beyond visual comparison of the different isovists, we used four mor- phological indicators: – Perimeter : P – Area : A – Ellipticity factor : ϵ = π × L2/4A – Convexity defect : econv = A/Ahull – Perimeter : P – Area : A – Ellipticity factor : ϵ = π × L2/4A – Convexity defect : econv = A/Ahull – Convexity defect : econv = A/Ahull With L, the largest distance between two vertices of the polygon and Ahull the area of the convex hull of the polygon. The ellipticity factor reflects the compression of a circle to form an ellip- soid and the flattening of the isovist. The convexity defect reflects the convex aspects of an isovist, especially the presence of concavity. All these four parameters give information upon the global form of an isovist. We then compute these parameters for one in ten of the isovists. By observing the Fig. 4, 5 and 6 we can separate the isovists into three main groups whose limits are indicated by green dashed lines in the figures. The first group may contain the isovists corresponding to the first poses of the trajectory. In this group, the video-based and mockup-based isovists show some differences. Indeed the first video-based isovists of the trajectory are almost circular, because of the lack of information of the reconstructed map in the first poses of the trajectory. This lack of information is due to the camera’s limited field of view. This period during which the SLAM algorithm has not reconstructed the environment with enough information yet could be considered as an initialization phase. The second group contains the next isovists. Here, the video-based and mockup-based isovists show much more similarities than in the first group. This is because there is enough information in the video to fully map the environment and the digital model is similar to what we can see in the video. Finally, the third group contains the isovists of the last poses of the trajec- tory. Here the video-based and mockup-based isovists show differences. 4. Results We present here the results on one dataset collected in the city of Nantes. We suppose that the user made a single one-way journey. We also implement an arbitrary limit distance for the ray-casting, based on the size of the re- constructed map. The computation has been conducted on a PC with 8GB RAM, 8x Intel Core i7 CPU 2.60 GHz, using Ubuntu 16.04 64 bits. With these characteristics, semi-dense-line SLAM ran for half an hour to fully map the environment from a one minute video. Then our method takes about two minutes to generate the 164 isovists related to the video. Figure 3. Comparison of isovists field. Left: video-based isovists. Right: mockup-based isovists. Figure 3. Comparison of isovists field. Left: video-based isovists. Right: mockup-based isovists. A quick comparison (Fig. 3) of the video-based and mockup-based isovists shows the similarities and the differences between the two models. Thus, the video-based isovists from the first poses of the trajectory are very circular, because of the limited field of view of the camera that doesn’t allow the SLAM method to reconstruct the environment behind the observer. As seen in Fig. 3 left, as the scene behind the observer is not reconstructed, there is no mask (i.e. no building facades) to stop the rays. This lack of mask casts the rays to the limits of the artificial horizon (which is here of very limited range) which explains the circularity of the resulting shape. The video-based and mockup- based isovists attached to the following poses seem pretty similar. Finally, the last poses show again some differences. This is caused by the fact that video- based isovists take into account walls delineating private plots while mockup- based isovists only consider buildings surfaces. The fences detected in the video are not present in the 3D model derived from standard datasets. 4. Results This may be explained by the fact that the video shows the reality of the envi- ronment, with additional occlusions such as walls or fences delineating pri- vate plots, while the digital model only considers buildings as polygons. The environment around the latter part of the trajectory contains a lot of these additional occlusions. It can also be seen from Fig. 7 that the convexity defect graph of video- based isovist is less constant and overall higher than the mockup-based isovist convexity defect graph. This indicates the presence of more occlusion leading to more concavity, which corresponds to the less smooth real environment. 83 Communications 83 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 0 20 40 60 80 60 70 80 90 Curvilinear abscissa (m) Perimeter (m) Figure 4. Perimeters (m) of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 100 200 300 400 Curvilinear abscissa (m) Area (m2) Figure 5. Areas (m2) of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 60 70 80 90 Curvilinear abscissa (m) Perimeter (m) Figure 4. Perimeters (m) of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 60 70 80 90 Curvilinear abscissa (m) Perimeter (m) Curvilinear abscissa (m) Figure 4. Perimeters (m) of video-based (blue) and mockup-based (red) isovists. Figure 4. Perimeters (m) of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 100 200 300 400 Curvilinear abscissa (m) Area (m2) Figure 5. Areas (m2) of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 100 200 300 400 Curvilinear abscissa (m) Area (m2) Curvilinear abscissa (m) Figure 5. Areas (m2) of video-based (blue) and mockup-based (red) isovists. 84 Communications 5. Conclusion, Discussion and Perspectives In this paper, we presented a process to generate isovists from a video recorded with a monocular camera, in an urban street. If the video shows enough information, the video-based isovists are pretty similar to the ones generated using standard topographic datasets, provided the viewing distance. There is even evidence to suggest that our method gives additional details upon the real environment, compared to the digital urban model, by taking into account some obstacles that are not houses or buildings. Our method can be used to further improve the digital model by taking into account real assets obstructing the user’s view represented by video-based isovists; thus tackling the well known updating problem in GIS. Moreover, the video-based reconstruction of the shape of the immediate surroundings could help several applications, eg. the display of geolocalised information in augmented reality. Our method cannot run in real-time due to the important amount of cal- culation during the mapping of the environment in the semi-dense-line SLAM algorithm. This forced us to use short videos to test our method and thus to work on small environments, and this is why we limited the range of the isovists. This issue could be solved using a more powerful computer and longer videos of the environment. We could also use the loop closure of the SLAM algorithm to optimize the reconstructed map by recognizing places already visited by the user. Another axis of improvement would be to automatize the acquisition of world coordinates and scale factor of the reconstructed map of the environment and the estimated trajectory. This may be done by getting the camera’s meta- data giving GPS coordinates and orientation of the sensor. This would also allow an automatic correspondence of video-based and mockup-based isovists. Our method is not dependant on any kind of device, and any monocular video – if we know the intrinsic parameters of the camera – can be taken as an entry. Considering that a limit of the video-based isovists is the limited field of view of a simple monocular camera, we may wonder about the possibility to extend this method to omnidirectional cameras. Indeed omnidirectional cameras can catch videos with a very large field of view and adapting this method to this kind of camera only demands to adapt the mathematical model of the SLAM algorithm from pinhole model to unified model (Caruso et al., 2015). Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Outdoor isovist computation with SLAM 0 20 40 60 80 2 3 4 5 6 Curvilinear abscissa (m) Ellipticity (-) Figure 6. Ellipticity of video-based (blue) and mockup-based (red) isovists. Outdoor isovist computation with SLAM Outdoor isovist computation with SLAM 0 20 40 60 80 2 3 4 5 6 Curvilinear abscissa (m) Ellipticity (-) Figure 6. Ellipticity of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 0.6 0.8 1 1.2 Curvilinear abscissa (m) Convexity defect (-) Figure 7. Convexity defect of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 2 3 4 5 6 Curvilinear abscissa (m) Ellipticity (-) Curvilinear abscissa (m) Figure 6. Ellipticity of video-based (blue) and mockup-based (red) isovists. 0 20 40 60 80 0.6 0.8 1 1.2 Curvilinear abscissa (m) Convexity defect (-) Figure 7. Convexity defect of video-based (blue) and mockup-based (red) isovists. Figure 7. Convexity defect of video-based (blue) and mockup-based (red) isovists. 85 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] In this way, video-based isovists generated by our method more accurately reflects the reality of the environment, as long as there are enough data provided by the video. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Outdoor isovist computation with SLAM may make the calculations heavier but we would have complete video-based isovists rather than half video-based isovists for the first poses of the trajectory. Furthermore, it would be interesting to add semantic treatment to our video in order to optimize the surface reconstruction. It’s an approach which is already used by MIT-SPARK (Rosinol et al., 2020) but with stereo cameras. This approach may soon be extended to monocular cameras which could also be inserted in our method. Another very recent method (Yang et al., 2020) allows user to reconstruct in real-time a dense surface mesh of the environment with a mobile device equipped with an embedded monocular camera. This method has given inter- esting results in indoor environments. We could imagine applying it to outdoor environments to replace the semidense-line SLAM. As future work, we could directly compare isovist field extracted from LiDaR point cloud with mock-up based isovist. Such a comparison could validate our method, taking the point cloud as a reference. 5. Conclusion, Discussion and Perspectives Using omnidirectional cameras implies processing more pixels which Communications 86 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] References Benedikt M. L. (1979). To take hold of space: isovists and isovist fields. Environment and Planning B: Planning and Design, Vol. 6, No. 1, pp. 47–65. Caruso D., Engel J., Cremers D. (2015, sep). Large-scale direct SLAM for om- nidirectional cameras. In 2015 IEEE/rsj international conference on intelli- gent robots and systems (iros), pp. 141–148. IEEE. Retrieved from http:// {IEEE}xplore.{IEEE}.org/document/7353366/ Díaz-Vilariño L., González-deSantos L., Verbree E., Michailidou G., Zlatanova S. (2018, 9). From point clouds to 3D isovists in indoor environments. IS- PRS - International Archives of the Photogrammetry, Remote Sensing and Spa- tial Information Sciences, Vol. XLII-4, pp. 149-154. Retrieved from https:// www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-4/149/2018/ Engel J., Schöps T., Cremers D. (2014). LSD-SLAM: Large-Scale Direct Monocular SLAM. In European conference on computer vision (eccv), pp. 834–849. Retrieved from http://link.springer.com/10.1007/978-3-319-10605-2_54 Ewing R., Handy S. (2009). Measuring the unmeasurable: Urban design qualities related to walkability. Journal of Urban Design, Vol. 14, pp. 65-84. He S., Qin X., Zhang Z., Jagersand M. (2018). Incremental 3D Line Segment Extrac- tion from Semi-dense SLAM. Proceedings - International Conference on Pattern Recognition, Vol. 2018-August, pp. 1658–1663. Klein G., Murray D. (2007, nov). Parallel Tracking and Mapping for Small AR Workspaces. In IEEE international symposium on mixed and augmented reality (ismar). IEEE. Retrieved from http://ieeexplore.ieee.org/lpdocs/epic03/wrapper .htm?arnumber=4538852 87 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Leduc T., Leduc M. (2020). Minimum-area ellipse bounding an isovist: towards a 2D GIS-based efficient implementation. International Journal of Geographical Information Science, pp. 1–24. Möller T., Trumbore B. (2005). Fast, minimum storage ray/triangle intersection. ACM SIGGRAPH 2005 Courses, SIGGRAPH 2005, No. 1, pp. 1–7. Mur-Artal R., Tardos J. (2015, jul). Probabilistic Semi-Dense Mapping from Highly Accurate Feature-Based Monocular SLAM. In Robotics: Science and systems xi. Robotics: Science and Systems Foundation. Retrieved from http:// www.roboticsproceedings.org/rss11/p41.pdf Mur-Artal R., Tardos J. D. (2017, 10). Orb-SLAM2: An open-source SLAM sys- tem for monocular, stereo, and rgb-d cameras. IEEE Transactions on Robotics, Vol. 33, pp. 1255-1262. Retrieved from http://IEEExplore.IEEE.org/document/ 7946260/ Newcombe R., Lovegrove S., Davison A. (2011). DTAM: Dense tracking and mapping in real-time. In Proc. of the intl. conf. on computer vision (iccv), barcelona, spain, Vol. 1. Retrieved from http://www.doc.ic.ac.uk/~ajd/Publications/newcombe _etal_iccv2011.pdf Rosinol A., Abate M., Chang Y., Carlone L. (2020, may). Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping. In 2020 ieee international conference on robotics and automation (icra), pp. 1689–1696. IEEE. Retrieved from https://ieeexplore.ieee.org/document/9196885/ Schmid K., Stülpnagel R. von. (2018). References Quantifying local spatial properties through lidar-based isovists for an evaluation of opinion-based vgi in a vr setup. In P. Kiefer, H. Huang, N. V. de Weghe, M. Raubal (Eds.),, p. 173-178. ETH Zurich. Retrieved from https://www.research-collection.ethz.ch/handle/20.500 .11850/225613 Yang X., Zhou L., Jiang H., Tang Z., Wang Y., Bao H. et al. (2020). Mobile3Drecon: Real-time monocular 3D reconstruction on a mobile phone. IEEE Transactions on Visualization and Computer Graphics, pp. 3446-3456. 88 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 Communications 88 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Thomas Leduc1, Xenia Stavropulos-Laffaille1, Ignacio Requena- Ruiz1 Thomas Leduc1, Xenia Stavropulos-Laffaille1, Ignacio Requena- Ruiz1 1. AAU-CRENAU, UMR 1563 CNRS/ECN/ENSAG/ENSAN ENSA Nantes, 6 quai François Mitterrand, BP 16202, 44262 Nantes, France thomas.leduc@crenau.archi.fr RESUME. L'adaptation des espaces publics aux épisodes de chaleur intense est désormais un défi majeur pour la ville. Dans cette optique, cet article présente une contribution qui vise à délimiter et à traiter l'ombre au sol, qu'elle provienne des bâtiments ou du couvert arboré. Après une comparaison aux tracés des ombres obtenues via deux outils de référence, nous présentons un site urbain qui mélange des ombres d'origines différentes ainsi que divers indicateurs. L'objectif est ici d'aider les concepteurs urbains à trouver des solutions permettant de distinguer les contributions respectives des ombres, leur évolution annuelle et les éventuelles continuités d’ombres spatiales ou temporelles. ABSTRACT. The adaptation of public spaces to episodes of intense heat is now a major challenge for the city. With this in mind, this article presents a contribution aimed at delineating and handling the ground shadow, whether it comes from buildings or from the tree canopy. After a comparison with shadows obtained via two reference tools, we present a real urban site that mixes shadows of different origins and, in addition, different indicators. The aim of these indicators is to assist urban designers with solutions that make it possible to distinguish the respective shadow contributions, their annual evolution and potential spatial or temporal continuities. MOTS-CLES : carte des ombres, ombre urbaine, ombre du couvert arboré, microclimat urbain, conception basée sur l'ombre KEYWORDS: shade map, urban shade, tree canopy shade, urban microclimate, shade-based design 89 Communications 1. Introduction The urban heat island (UHI)1 phenomenon (Sobstyl et al., 2018) is a bioclimatic response specific to compact high-rise urban districts, amplifies these overheating, exposes populations to high health risks, and significantly alters living conditions in the built environment (Davis and Gertler, 2015). Both climate and demographic pressures constraint decision-makers to organise urban lifestyles according to mitigation and adaptation strategies, which nowadays focus on how to cool down the urban space in short and mid-term. (Aram et al., 2019) consider four groups of urban cooling strategies: i) vegetative cover, ii) stack night ventilation, iii) water bodies and iv) modification of surface albedos. (Ruefenacht and Acero, 2017) highlight the relevance of shading as a “key measure to improve Outdoor Thermal Comfort (OTC) and mitigate UHI because it leads to the reduction of air and surface temperature and can therefore result in cooling benefits”. In this regard, (Middel et al., 2016) enhance “the importance of solar access active management”. Considering outdoor air temperature difficult to control, (Aleksandrowicz et al., 2020) observe that sunlight exposure in streets is easier to regulate since it is directly linked to design decisions; therefore, they consider “outdoor shade as a primary comfort indicator”. We argue that a tool developed in the context of a standard GIS (based on shade index, tree canopy cover, etc.) can help in quantifying the microclimatic qualities of urban spaces revealing, for instance, sunlighted areas requiring shade intensification or, conversely, strategic shaded areas that deserve to be protected. The strictly geometric aspects of shading is essential in answering simple questions like: where are located the “shadow fences”2 generated by the urban environment at a given date and time? What is the fraction of space, for a predefined study area - which is in shadow for a given date and time? What is the tree canopy contribution to the total shade cover? What is the evolution of this contribution over the summer period, the crown remaining unchanged? What areas are shaded for at least 3 hours on a hot summer day? Or even, to satisfy a constraint of temporal continuity, what areas are shaded for at least 3 consecutive hours on a hot summer day? Or even, do these areas guarantee spatial continuity in order to allow comfortable pedestrian mobility? Simulation softwares such as Solene (Miguet and Groleau, 2002), Ladybug (Roudsari and Pak, 2013), UMEP-SOLWEIG (Lindberg et al., 2018) or ENVI-met3, already allow shadow studies. Communications 89 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1 As (Ruefenacht and Acero, 2017) summarizes: UHI “occurs because cities consume huge amounts of energy in electricity and fuel, have less vegetation to provide shade and cooling, and are built of materials that absorb and store energy from the sun”. 2 We borrow this expression from (Knowles, 2003) to underline the fact that shadowing can also, in itself, generate constraints useful to the urban planner during a design phase (in reference to the solar envelope concept). 3 ENVI-met is a German trademark (No. 304 73 896), https://www.envi-met.com/ 2 We borrow this expression from (Knowles, 2003) to underline the fact that shadowing can also, in itself, generate constraints useful to the urban planner during a design phase (in reference to the solar envelope concept). 1 As (Ruefenacht and Acero, 2017) summarizes: UHI “occurs because cities consume huge amounts of energy in electricity and fuel, have less vegetation to provide shade and cooling, and are built of materials that absorb and store energy from the sun”. 3 ENVI-met is a German trademark (No. 304 73 896), https://www.envi-met.com/ 1. Introduction Some of them (Solene and Ladybug) run as plugins of computer-aided design programs that do not offer the functionalities of GIS (power of synthesis of 2D maps, capacity to process large geographic areas, etc.) and require 90 Communications 90 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Solar model and shadow plotting in the context of a 2D GIS a particular expertise in architectural design. Therefore, their implementation by the geographic information departments of municipalities is difficult. In the contrary, UMEP-SOLWEIG has been embedded into a GIS. Nevertheless, they require digital surface and terrain models. In addition, the raster-based processes do not allow the shadow fences to be handled as simply as standard vector entities. This article aims to highlight the interest of a specific vectorial development in a standard GIS to answer the questions mentioned above. After a presentation of the solar model and the treatment of mineral and vegetal shadows that we have implemented, we will present the mechanism for calculating multiple shadow overlaps. In the next section, we will highlight the interest of these developments in one case study of the Coolscapes research program. Finally, we will discuss the contributions of our tool to a microclimate sensitive urban space design. 2.1. 3D processing in the context of 2D GIS 2.1. 3D processing in the context of 2D GIS The sun path geometry and the delineation of the cast shadow generated by buildings and tree canopies are intrinsically three-dimensional urban issues. Our aim is to identify shaded areas that allow the city dweller to cool down; therefore, we simply delimit these areas on a horizontal plane at a pedestrian level (in this case, we choose the ground plane). This simplifying hypothesis allows us, at the end of a 3D processing presented later on, to produce a classical 2D cartography in the context of a standard GIS. 2.2. Implementation of the sun path model Pysolar and PyEphem are Python implementations of ephemerides that allow estimating the trajectories and positions of the sun over time. To ensure consistency with the laboratory’s simulation tools, we have implemented the ephemeris model of the Solene software in the context of our GIS simulation. A systematic comparison of the positions returned by Pysolar with those of our simulation shows that the respective solar heights differ by less than 1% on annual average. 2. Method Two main principles guide our methodological development: replicability and transportability on the one hand, and ease of implementation on the other. To guarantee these two principles, we use standard topographic data sets such as the ones provided by the IGN BD TOPO® database (June 2020 edition) and a widely distributed free and open source software (QGIS Development Team, 2020). It is noteworthy that the presented method of shadow simulation and mapping is potentially applicable at larger scales, e.g. district or city scale. 2.5. Management of shadow overlaps over several (consecutive) hours To identify areas in the shade for several hours (possibly consecutively), we need to make multiple geometric intersections, the combinatory of which are potentially substantial (for example, being in the shade for 2 hours can be translated multiply: 10 am-11 am and 3 pm-4 pm, 1 pm-3 pm, etc.). To deal efficiently with this issue and bypass these numerous intersections, we proceed in three steps: i) geometric union (via the unary_union operator) of the shadow contours, ii) polygonization of the network of polylines thus produced, iii) for each polygon resulting from the polygonization, calculation of the number of enclosing shadow zones. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] description consists in a set of polygons (sort of flat roof footprints) completed with an attribute equal to the height of the corresponding building. This twofold information allows us to generate a collection of right prisms each one representing a building in the urban model. The direction of the sun's beams being predetermined, it is then possible to i) calculate the ground shadow of each right prism, ii) and to proceed to a geometric union of all shadows. 2.4. Shadow cast from the tree canopies Nantes Metropole’s geographic information services provided us with a database of 75,156 trees in the conurbation. In addition to the point position of each trunk, this data includes various attributes such as the height of the tree, the circumference of the trunk, the species of the tree, etc. In the first version of tree modelling, we developed three different crown models: the first is spherical, the second cylindrical and the third conical. Based on the tree attributes and positions as well as the crown model and the sunbeam direction, we have then implemented a ground shadow model (the projection of a sphere being an ellipse for example). As in the case of the building shadows, at the end of the process we proceed to the geometric union of all tree shadows in order to deal with possible overlaps. 2.3. Shadow cast from a collection of right prisms The topographic data sets employed here provide a 3D metric and vector description (structured in objects) of the elements of the territory and its infrastructures, usable on scales from 1:5,000 to 1:50,000. More precisely, this 91 Communications 91 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3.2. Experimental setup In order to test our methodological proposal and respond to the research questions, we have defined several parameters as follows. Regarding the choice of representative dates and times for the annual sunshine simulation, we have chosen the 1st, 11th and 21st of each month. For each of these days, we have made a calculation with a hourly time step from 10 am to 4 pm. Regarding the choice of a representative summer day, we opted for July 21st. Regarding the tree modelling in this area, after empirical evaluation on site, we chose a cylindrical crown model with two distinct parameterizations. Eight trees in the Courtyard have a total height of 15 metres, a crown height of 10 metres and a crown radius equal to 5 metres. The rest of the trees are modelled with a total height of 9 metres, a crown height of 6 metres and a crown radius of 3 metres. It is noteworthy that although the trees in the area are deciduous, we do not consider the change in the crown due to seasonality. 3.1. Case study: Graslin district in Nantes (France) The study site is located in the centre of Nantes (France), including parts of the Graslin district. This area consists of 13.6 ha, 293 building footprints representing a total floor area of about 6.4 ha, and 98 trees. Its altimetry varies from 6 to 30 meters and the average height of the buildings is about 15.7 meters (with a standard deviation set at 5.8 meters). The explained method is applied to this site with a special interest in the urban garden called “Cours Cambronne”. This classic French garden, lined with buildings and trees, covers about 0.9 ha and presents various shaded areas (from both surrounding buildings and 64 trees) that modify comfort indexes and citizens practices (i.e., sitting places and trajectories). It is roughly 180 metres long and 50 metres wide. It also has the advantage of crossing shadows from the tree cover (rows of trees) and the building environment, all on a relatively flat ground. 92 Communications 92 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Solar model and shadow plotting in the context of a 2D GIS Solar model and shadow plotting in the context of a 2D GIS 4.2. Comparison with the Urban Multi-scale Environmental Predictor (UMEP As the verification proposed in the previous section is purely visual, we suggest comparing our simulation results with those of the UMEP reference tool. Indeed, the latter embeds the shadow generator plugin capable of generating shadow plots from building, ground or even vegetation digital surface models, in the QGIS context. In order to use UMEP’s “Daily Shadow Pattern” functionality, we first had to generate a digital terrain model, a digital model of the built-up surfaces, and a raster model of the tree canopy. We used a metric precision raster model such as the one proposed by the IGN’s RGE Alti® to stick as closely as possible to the building layouts. Fig. 2 shows the relatively good correlation of the raster shadows produced by UMEP with the vector traces from our simulations. FIGURE 2. Comparison of the shadows from the UMEP Shadow generator (black pixels) with those from the t4gpd plugin (red polylines). FIGURE 2. Comparison of the shadows from the UMEP Shadow generator (black pixels) with those from the t4gpd plugin (red polylines). To go beyond this visual comparison, we measured the shadow-covered areas respectively produced by UMEP and our simulation tool at different times on 21st July. Table 1 shows variations ranging from 0.1% to 5.2%. Although this comparative analysis does not formally validate our simulation tool, it reinforces our ability to produce consistent results with the UMEP reference tool. TABLE 1. Comparison of shadow-covered area measurements (in m²) assessed via the UMEP software vs. via the t4gpd plugin, on July 21st. TABLE 1. Comparison of shadow-covered area measurements (in m²) assessed via the UMEP software vs. via the t4gpd plugin, on July 21st. TABLE 1. Comparison of shadow-covered area measurements (in m²) assessed via the UMEP software vs. via the t4gpd plugin, on July 21st. 10 am 11 am 12 am 1 pm 2 pm 3 pm UMEP 35,498.6 29,062.9 27,838.8 27,677.2 28,900.2 36,852.8 T4gpd 33,694.2 29,093.6 28,036.8 27,533.8 28,560.2 36,161.6 Deviation 5.2 % 0.1 % 0.7 % 5.2 % 1.2 % 1.9 % 10 am 11 am 12 am 1 pm 2 pm 3 pm UMEP 35,498.6 29,062.9 27,838.8 27,677.2 28,900.2 36,852.8 T4gpd 33,694.2 29,093.6 28,036.8 27,533.8 28,560.2 36,161.6 Deviation 5.2 % 0.1 % 0.7 % 5.2 % 1.2 % 1.9 % Communications 4.1. Comparison of shadows in the context of a CAAD software The objective of this section is to use the solar model and the shadow engine implemented natively in the SketchUp software, to compare the shadows computed by the t4gpd plugin in the context of a GIS with those of a tool widely used by architectural practitioners. For that, we first import the building footprints and tree positions to build the corresponding 3D model using the SketchUp pushpull mechanism. We then import the shadow polygons produced by our simulations (through the SHP file format) and compare them visually to the shadows produced by the SketchUp shadow engine. Fig. 1 shows that these two shadow features match satisfactorily. y FIGURE 1. Comparison of the native SketchUp shadow features (in grey) with those from the t4gpd plugin (red polylines), on July 21 at 10 am UTC. FIGURE 1. Comparison of the native SketchUp shadow features (in grey) with those from the t4gpd plugin (red polylines), on July 21 at 10 am UTC. 93 Communications 93 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] 4.2. Comparison with the Urban Multi-scale Environmental Predictor (UMEP) Communications 94 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Solar model and shadow plotting in the context of a 2D GIS Solar model and shadow plotting in the context of a 2D GIS 5.1.2. Shade contribution of the tree canopy In the context of urban greening policies, it is essential to distinguish the shadows generated by the built environment (i.e. the buildings) from those produced by the tree canopy. For the configuration shown in Fig. 3 (right), the shade produced by the tree canopy represents about 16% of the total shade on the ground. This contribution is naturally conditioned by the tree crown geometries as well as by their spatial distribution within the urban space (they may indeed be hidden by possible taller buildings). 5.1.1. Shade ratio at a given date and time The first very simple application consists, in delineating the ground shadow due to the tree canopy and the built environment for a given territory, date and time. On 21 July at 12 am (UTC), the 293 buildings and 98 trees in the study site thus produce a ground shade of approximately 2.8 ha, i.e. approximately 20% of the total area. Furthermore, as shown in Fig. 3 (left), few streets oriented almost orthogonal to the sun’s beams benefit from shade. FIGURE 3. (left) The plugin allows evaluating both shadows of buildings and trees on a horizontal plane according to the date and time chosen. The result is a multi-polygon geometry (light grey area). (right) The plugin is able to distinguish, in the shadow cast on the ground, the contribution of trees (light green polygons) from that of buildings (light grey polygons). FIGURE 3. (left) The plugin allows evaluating both shadows of buildings and trees on a horizontal plane according to the date and time chosen. The result is a multi-polygon geometry (light grey area). (right) The plugin is able to distinguish, in the shadow cast on the ground, the contribution of trees (light green polygons) from that of buildings (light grey polygons). 5.1.3. Shade ratio evolution over the summer period In this section, the yearly evolution of the shadow ratio for different times a day (between 10 am and 4 pm) is studied (Fig. 4). The considered period is chosen during 95 Communications 95 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 summertime, when the sun is at its highest path in the sky and, consequently, the solar energy input is the greatest. Considering the results, for the period between spring equinox and autumn equinox, it can be seen that, the shade provided by the tree canopy is significant. In fact, the tree shades cover on average one third of the courtyard. This contribution is almost systematically higher than the one of the surrounding buildings. At its peak (at 2 pm UTC), it is even around 6 times higher than the shade provided by the buildings. he year for a few predetermined hours FIGURE 4. Evolution of shade ratios over the year for a few predetermined hours of daylight. The fall of foliage in autumn and winter is likely to modify the geometry of the crown and reduce the amount of shade. FIGURE 4. Evolution of shade ratios over the year for a few predetermined hours f daylight. The fall of foliage in autumn and winter is likely to modify the geometry of the crown and reduce the amount of shade. 5.2.1. Delineation of shaded areas over several hours The annual evolution of the ground shade presented in the previous section can also be studied for a given day. This allows distinguishing potential cool spots, during a heat wave event. Thus, the shade zones are cut out according to the number of hours of shade. With the special case of 21st July, Fig. 5 (left) shows the excessively fragmented nature of this cut within the shadow zone. It can be seen, however, that although the central strip situated halfway between the two rows of trees (longitudinal centre line of the courtyard) is finally fairly exposed, the tree roots are very well protected. 96 Communications 5.2.2. Delineation of shaded areas over several consecutive hours The accumulation of shadow hours shown in Fig. 5 (left) does not include any notion of temporal continuity. However, it could be interesting to detect areas that are permanently in the shade (in order, for example, to design and deploy appropriate street furniture). Fig. 5 (right) shows a revised version of the indicator incorporating this dimension of continuity. The idea here is to represent areas as a function of their exposure to shade for several consecutive hours. As one can note, the overall fragmentation is much less, and patterns emerge that guarantee a certain spatial continuity. Communications 96 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Solar model and shadow plotting in the context of a 2D GIS Solar model and shadow plotting in the context of a 2D GIS FIGURE 5. (left) Non-uniform distribution of shade in the courtyard during the hours when the sun is at its highest in summer. (right) Representation of areas in shade over several consecutive hours. el and shadow plotting in the context of a 2D GIS Solar model and shadow plotting in the context of a 2D GIS FIGURE 5. (left) Non-uniform distribution of shade in the courtyard during the hours when the sun is at its highest in summer. (right) Representation of areas in shade over several consecutive hours. FIGURE 5. (left) Non-uniform distribution of shade in the courtyard during the hours when the sun is at its highest in summer. (right) Representation of areas in shade over several consecutive hours. 6.1. Pros: beyond descriptive analysis, shadow as a design object While it is possible to vectorise the raster shadows obtained via UMEP, the underlying pixelisation is nevertheless problematic when seeking to use shadows for the design of urban spaces as given, for example, in the solar envelope concept (Knowles, 2003). From this point of view, our proposal is innovative in the specific context of geomatics as these vector shadows enable to delineate potential cool-paths and urban oases or cools spots in the urban fabric… all useful features for citizens during heat waves. 7. Conclusion Our work contributes to an estimation, in the geomatics context, of tree-shaded zones to building ones, which is crucial for summer thermal comfort. It also makes it possible to go beyond the calculation and descriptive analysis of the shadows cast in the urban space and to use them as standard polygonal shapes. Continuing its development by integrating a comprehensive energy and a more realistic geometric model will make it operational for urban planners. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] with significant elevation changes. Furthermore, the simplified and strictly geometric approach is not sufficient to properly inform the complex notion of pedestrian thermal comfort in heterogeneous urban spaces. On the one hand, particular building geometries (e.g., walkways, porches, steep rooftops) are not considered. On the other hand, thermal comfort simulation tools such as ENVI-met allow studying the spatial distribution of thermal comfort indexes by integrating an energy-based approach. Supplementing our tool with a more representative geometric approach, based on 3D city models (ex. CityGML (Biljecki et al., 2018)) and a thermo-radiative (or even aeraulic) model would make it possible to transform this prototype into a more relevant tool for city designers. Such an assessment will require considering the radiative models of the facades in a full 3D approach. Acknowledgment This study is supported by the French National Research Agency (ANR) under reference ANR-18-CE22-0003 (COOLSCAPES project). 6.2. Contras: need for an energy-based approach It should be noted, however, that the current method is not based on Digital Elevation Models (DEMs); thus, the interactions between the topography and the sun path model are not considered. This is a potential application limit especially for cities 97 Communications Communications 97 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] References Abdel-Aziz D M, Shboul A Al, Al-Kurdi N Y, 2015, “Effects of Tree Shading on Building’s Energy Consumption -The Case of Residential Buildings in a Mediterranean Climate” American Journal of Environmental Engineering 5(5) 131–140 Aleksandrowicz O, Zur S, Lebendiger Y, Lerman Y, 2020, “Shade maps for prioritizing municipal microclimatic action in hot climates: Learning from Tel Aviv-Yafo” Sustainable Cities and Society 53(September 2019) 101931, https://doi.org/10.1016/j.scs.2019.101931 Aram F, Higueras García E, Solgi E, Mansournia S, 2019, “Urban green space cooling effect in cities” Heliyon 5(4) e01339, https://doi.org/10.1016/j.heliyon.2019.e01339 Biljecki F, Kumar K, Nagel C, 2018, “CityGML Application Domain Extension (ADE): overview of developments” Open Geospatial Data, Software and Standards 3(1) 13, https://opengeospatialdata.springeropen.com/articles/10.1186/s40965-018-0055-6 Coutts A M, White E C, Tapper N J, Beringer J, Livesley S J, 2016, “Temperature and human thermal comfort effects of street trees across three contrasting street canyon environments” Theoretical and Applied Climatology 124(1–2) 55–68, http://link.springer.com/10.1007/s00704-015-1409-y 98 Communications Communications 98 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Solar model and shadow plotting in the context of a 2D GIS Davis L W, Gertler P J, 2015, “Contribution of air conditioning adoption to future energy use under global warming” Proceedings of the National Academy of Sciences 112(19) 5962–5967, http://www.pnas.org/lookup/doi/10.1073/pnas.1423558112 Demuzere M, Orru K, Heidrich O, Olazabal E, Geneletti D, Orru H, Bhave A G, Mittal N, Feliu E, Faehnle M, 2014, “Mitigating and adapting to climate change: Multi- functional and multi-scale assessment of green urban infrastructure” Journal of Environmental Management 146 107–115, https://linkinghub.elsevier.com/retrieve/pii/S0301479714003740 Emmanuel R, Loconsole A, 2015, “Green infrastructure as an adaptation approach to tackling urban overheating in the Glasgow Clyde Valley Region, UK” Landscape and Urban Planning 138 71–86, https://linkinghub.elsevier.com/retrieve/pii/S0169204615000432 Ewing R, Handy S, Brownson R C, Clemente O, Winston E, 2006, “Identifying and Measuring Urban Design Qualities Related to Walkability” Journal of Physical Activity and Health 3(Suppl 1) 223–240, http://forum.activelivingresearch.com/sites/default/files/JPAH_15_Ewing.pdf Feyisa G L, Dons K, Meilby H, 2014, “Efficiency of parks in mitigating urban heat island effect: An example from Addis Ababa” Landscape and Urban Planning 123 87–95, http://dx.doi.org/10.1016/j.landurbplan.2013.12.008 Gehl J, 2011 Life Between Buildings: Using Public Space (Island Press/Center for Resource Economics) Gillner S, Vogt J, Tharang A, Dettmann S, Roloff A, 2015, “Role of street trees in mitigating effects of heat and drought at highly sealed urban sites” Landscape and Urban Planning 143 33–42, http://dx.doi.org/10.1016/j.landurbplan.2015.06.005 Gunawardena K R, Wells M J, Kershaw T, 2017, “Utilising green and bluespace to mitigate urban heat island intensity” Science of The Total Environment 584–585 1040–1055, http://dx.doi.org/10.1016/j.scitotenv.2017.01.158 Kabisch N, van den Bosch M, Lafortezza R, 2017, “The health benefits of nature-based solutions to urbanization challenges for children and the elderly – A systematic review” Environmental Research 159(July) 362–373, https://linkinghub.elsevier.com/retrieve/pii/S0013935117315396 Knowles R L, 2003, “The solar envelope: its meaning for energy and buildings” Energy and Buildings 35(1) 15–25, http://linkinghub.elsevier.com/retrieve/pii/S0378778802000762 Lafortezza R, Chen J, van den Bosch C K, Randrup T B, 2018, “Nature-based solutions for resilient landscapes and cities” Environmental Research 165(December 2017) 431–441, https://linkinghub.elsevier.com/retrieve/pii/S0013935117317115 Lindberg F, Grimmond C S B, Gabey A, Huang B, Kent C W, Sun T, Theeuwes N E, Järvi L, Ward H C, Capel-Timms I, Chang Y, Jonsson P, Krave N, Liu D, Meyer D, Olofson K F G, Tan J, Wästberg D, Xue L, Zhang Z, 2018, “Urban Multi-scale Environmental Predictor (UMEP): An integrated tool for city-based climate services” Environmental Modelling & Software 99 70–87, https://linkinghub elsevier com/retrieve/pii/S1364815217304140 ghub.elsevier.com/retrieve/pii/S1364815217304140 Martilli A, Krayenhoff E S, Nazarian N, 2020, “Is the Urban Heat Island intensity relevant for heat mitigation studies?” Urban Climate 31(January 2019) 100541, https://linkinghub.elsevier.com/retrieve/pii/S2212095519300070 99 Communications Communications 99 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Middel A, Selover N, Hagen B, Chhetri N, 2016, “Impact of shade on outdoor thermal comfort—a seasonal field study in Tempe, Arizona” International Journal of Biometeorology 60(12) 1849–1861, http://dx.doi.org/10.1007/s00484-016-1172-5 Miguet F, Groleau D, 2002, “A daylight simulation tool for urban and architectural spaces— application to transmitted direct and diffuse light through glazing” Building and environment 37 833–843, h // i di / i / i l / ii/S0360132302000495 Mullaney J, Lucke T, Trueman S J, 2015, “A review of benefits and challenges in growing street trees in paved urban environments” Landscape and Urban Planning 134 157–166, https://linkinghub.elsevier.com/retrieve/pii/S016920461400245X QGIS Development Team, 2020, “QGIS Geographic Information System. Open Source Geospatial Foundation Project.”, http://qgis.osgeo.org Redon E, Lemonsu A, Masson V, 2020, “An urban trees parameterization for modeling microclimatic variables and thermal comfort conditions at street level with the Town Energy Balance model (TEB-SURFEX v8.0)” Geoscientific Model Development 13(2) 385–399, https://gmd.copernicus.org/articles/13/385/2020/ Roudsari M S, Pak M, 2013, “Ladybug: A parametric environmental plugin for grasshopper to help designers create an environmentally-conscious design” Proceedings of BS 2013: 13th Conference of the International Building Performance Simulation Association 3128–3135 Ruefenacht L A, Acero J A, 2017, “Strategies for Cooling Singapore: A catalogue of 80+ measures to mitigate urban heat island and improve outdoor thermal comfort”, https://www.research-collection.ethz.ch/handle/20.500.11850/258216 Sobstyl J M, Emig T, Qomi M J A, Ulm F-J, Pellenq R J-M, 2018, “Role of City Texture in Urban Heat Islands at Nighttime” Physical Review Letters 120(10) 108701, https://link.aps.org/doi/10.1103/PhysRevLett.120.108701 Taha H, Akbari H, Rosenfeld A, Org E, 1991, “Heat Island and Oasis Effects of Vegetative Canopies: Micro-Meteorological Field-Measurements” Theoretical and Applied ClimatologyTheoretical and Applied Climatology 44 123–138, https://escholarship.org/uc/item/5wt7072h Whyte W H, 1980 The Social Life of Small Urban Spaces (Washington, D.C: Conservation Foundation) Yin L, 2017, “Street level urban design qualities for walkability: Combining 2D and 3D GIS measures” Computers, Environment and Urban Systems 64 288–296, http://dx.doi.org/10.1016/j.compenvurbsys.2017.04.001 100 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Perez Joan1, Fusco Giovanni1, Sadahiro Yukio2 Perez Joan1, Fusco Giovanni1, Sadahiro Yukio2 1. Université Côte d’Azur, CNRS, ESPACE, Nice, France. 2. Department of Urban Engineering, University of Tokyo, Tokyo, Japan. 1. Université Côte d’Azur, CNRS, ESPACE, Nice, France. 2 D f U b E U f T k T k J . Unive sité Côte d u , CN S, S C , Nice, ance. 2. Department of Urban Engineering, University of Tokyo, Tokyo, Japan. ABSTRACT. This paper presents different methods implemented with the aim of studying urban perforation and regeneration dynamics at the building level. The minimum requirement to perform these methods is a building footprint available for two different periods (time depth of at least 10 years) and heights (or floors) as attribute data. First, basic morphometric indicators are computed for each building: area, floor-area, number of neighbors, elongation, and convexity. Based on the availability of expert knowledge, different types of classifications and clustering are performed to obtain building types. A grid is superimposed on the test region of Osaka (Japan) and the number of building types per cell and for each period is computed, as well as the differences between each period. Mappings are then performed, showing that building types have specific locations and evolution trajectories. In some extreme cases, a specific building type can even gradually replace a type on a declining dynamic. KEYWORDS: Spatial analysis, urban perforation, urban regeneration, building type, classification, clustering. 101 KEYWORDS: Spatial analysis, urban perforation, urban regeneration, building type, classification, clustering. Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Cities are in a perpetual flow of transformations with urban areas that constantly make, unmake and remake themselves. Different models are discussed in the literature, mostly focusing on the benefits and/or drawbacks of the increase (urban sprawl, compact cities, regeneration, etc.) or decrease (shrinking cities, perforation, etc.) of building densities. A factor that is often overlooked regarding the evolution of densities is the role of inherited urban forms. Occurrence of areas having difficulties in redefining themselves are often closely related in term of urban form. Of course, even if the form factor is often overlooked, this issue is nothing new. The numerous criticisms of modernist architecture (Jacobs 1961, Salingaros 2005), which brought large-sized and specialized buildings, unable to evolve with the needs of the cities, are a good illustration of the link between urban life and urban form. What has changed more recently however, are the increasing possibilities to explore this link through: (1) the existence of large-sized datasets related to urban form and available at different dates, such as the French BD TOPO ® or the Japanese Zenrin Maps ® and (2) an increase of computing power that allows running complex algorithms on the aforementioned datasets. Examples of such work in the literature include the classification of building footprints by a Random Forest Classifier (Hecht et al., 2015), unsupervised Bayesian clustering for building types identification (Perez et al., 2019a), etc. This paper seeks to emphasize on different needs and precautions that shall be taken while working on the link between urban form and space-time evolutions of urban areas. Suggestions and recommendations while using building footprints as the primary source of data are also provided throughout the paper. Section 2 presents the data requirement, the test region and the computation of basic indicators of built-up form. Section 3 performs three different classifications, through a decision tree if reliable expert knowledge is available, a double k-means if only limited expert knowledge is available and finally, an unsupervised Self- Organizing Map if no expert knowledge is available. Section 4 creates a grid of a predetermined size over the test region in order to count the numerosity of each building type per cell and their evolution. Thematic maps are also displayed in this section. Section 5 concludes the paper. 1 ZENRIN is a private map information company that holds the top share in the Japanese market for local residential and car navigation maps. https://www.zenrin.co.jp/product/category/gis/basemap/zmaptown/index.html [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types changes in the distribution of building types. Each building shall be digitized as a single unit. Corrections should thus be made to GIS layers which aggregate contiguous buildings (like row houses) into single units. In addition, one attribute is required: building height, and another one is valuable, but not mandatory: building specialization. Specialization attribute allows filtering non-residential buildings to solely focus on residential patterns and dynamics of regeneration and perforation. Yet, due to mixed-use buildings, the distinction between residential and non- residential buildings is hardly ever straightforward. Thus, according to the source of data, attention should be paid to how specialization is encoded. 2.1. Minimum Data Requirement The bare minimum dataset required to implement the different methods discussed in this paper is a GIS layer of building footprints with multiple years of consistency (at least two different periods). The gap in years between the layers is dependent upon the scope of the analysis, but also on specificities of the location under study. A gap of 10 years is enough in countries following a deconstruction/reconstruction model of buildings (some countries in Asia, Africa, etc.) while in countries where buildings are usually renovated or rehabilitated, such as in Europe, a minimum time depth of 20 to 30 years is needed to observe structural Communications 102 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 b.). The original GeoPackage file only contains the two aforementioned building layers, respectively named BU0304 and BU1314. BU0304 is made of 760.067 inputs and BU1314 of 739.536, thus showing a decrease of 2.7% of the raw number of buildings in 10 years. ZENRIN Residential Maps possess several attribute data from which two are of interest in this research: the height of the buildings, expressed in number of floors, and building specialization. Building specialization is an encoded attribute, with codes as follows: 1200 for official and religious buildings (schools, administrative buildings, temple, etc.), 1363 for collective housing, 1364 for single-family homes, 1365 for private offices and mixed-use buildings, etc. In this paper, we focus on residential buildings, both collective and individual housings (1363 and 1364), which concern 469.519 buildings in 2003/04 and 428.875 in 2013/14. Each layer within the GeoPackage file originally contains 3 attribute data: an identifier, the number of floors and specialization. b.). The original GeoPackage file only contains the two aforementioned building layers, respectively named BU0304 and BU1314. BU0304 is made of 760.067 inputs and BU1314 of 739.536, thus showing a decrease of 2.7% of the raw number of buildings in 10 years. ZENRIN Residential Maps possess several attribute data from which two are of interest in this research: the height of the buildings, expressed in number of floors, and building specialization. Building specialization is an encoded attribute, with codes as follows: 1200 for official and religious buildings (schools, administrative buildings, temple, etc.), 1363 for collective housing, 1364 for single-family homes, 1365 for private offices and mixed-use buildings, etc. In this paper, we focus on residential buildings, both collective and individual housings (1363 and 1364), which concern 469.519 buildings in 2003/04 and 428.875 in 2013/14. Each layer within the GeoPackage file originally contains 3 attribute data: an identifier, the number of floors and specialization. 2.3. Indicator Computation The first step consists in computing a basic series of morphometric indicators for each building made of: the building footprint surface (area), the total amount of usable floor area, elongation, convexity and the number of adjoining neighbors. Elongation, convexity and the number of adjoining neighbors are detailed in Perez et al., (2019a). They respectively provide a measure of how buildings are elongated compared to the most compact equivalent shape (a circle), how buildings have intricated or squared shapes, and finally, if buildings are free-standing detached structures or possess adjoining neighbors. The final set of variables is made of the newly calculated morphometric indicators plus number of floors and specialization. Supplementary materials algorithm 1 shows how indicators are easily computed within R using the “sf” and “lwgeom” packages. This algorithm is applied to both BU0304 and BU1314. For the number of adjoining neighbors, we operate a small buffer (algorithm 1, # neighbors 1/2) with the aim of correcting buildings that shall be considered as adjoining, but are not due to low geolocation accuracy. Even if such occurrences are negligible within the ZENRIN Residential Maps, this precautionary step ought to be taken before computing the number of adjoining neighbors. Lastly, range01 is a user defined function that allows normalizing elongation and convexity. 2.2. Test Region and Data Presentation The test region is an area of 15 by 18 km in Japan containing central Osaka and its surroundings (Fig.1 a.). The peculiarities of Japan are that houses and small collective residential complexes are easily demolished to be reconstructed or to make way to new urban projects (Shelton, 2012) and that urban areas are experiencing urban shrinkage phenomena (Buhnik, 2010). For example, single- family houses have a lifespan of only 30 years (MLIT, 2007), which is in sharp contrast with European countries where houses are renovated and passed down from generation to generation. A short lifespan for buildings is an undeniable advantage for the study of the evolution of building types. Indeed, even now, high-quality datasets of building footprints are uneasy to access and, if they are, historic data of equivalent quality are usually not. Regarding urban shrinkage, demolitions without reconstructions is another interesting phenomenon to monitor over time. FIGURE 1. a. Osaka, OpenStreetMap. b. Buildings in Osaka in 2013-14 (Zmap TOWN II). FIGURE 1. a. Osaka, OpenStreetMap. b. Buildings in Osaka in 2013-14 (Zmap TOWN II). ZENRIN Residential Maps (Zmap TOWN II1), which are digital maps focusing on the building footprints throughout Japan, are extracted and compiled into a GeoPackage file for the extent of the test region in 2003/04 and in 2013/14 (Fig 1. Communications 103 Communications [ISBN : 978-2-910545-12-1] 3.1. Classification with Expert Knowledge When expert knowledge is available, a simple and efficient method of classification is found in a manually built decision tree. Successive tests on attributes split the original dataset into different class labels. When a solution is found and deemed robust by the expert, the conditional control statements can be used as benchmarks for other datasets possessing similar attributes. Supplementary materials algorithm 2 is a user defined function named DTs which allows classifying each row of a given dataset using the aforementioned morphometric indicators. Each test follows a if condition then outcome structure. Several conditions per test are possible, and the outcome is always a label attribution. Using the apply function, DTs can be used on any dataset possessing the relevant indicators, in our case BU0304 and BU1314. FIGURE 2. Visualization of the tests implemented in the DTs function FIGURE 2. Visualization of the tests implemented in the DTs function ualization of the tests implemented in the DTs functio Fig.2 is a flowchart visualization of the successive tests implemented in DTs. They are 9 different outcomes, in which 4 are for collective housings (C1 to C4) and 5 are for single-family homes (S1 to S5). Each label attribution is based on expert knowledge. For example, C1 are high rise residential buildings following a “tower in the park” model, C2 are elongated residential complexes representative, amongst other structures, of Danchi housing (団地) which are complexes of apartment buildings built following western models after the Second World War, S3 are small and elongated row houses, typical of traditional wooden Nagaya (長屋), etc. The relevance of these building types in Japanese urban areas is derived from literature (e.g. Shelton, 2012; Bonnin et al., 2014; Perez et al., 2019b). Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types 3. Residential Buildings: Classification of Types Once a basic set of indicators has been calculated for each input layer, the next step is to perform a classification of building types that can be applied to different time points. Three options arise. First, reliable expert knowledge is available, and as such, the number and the characteristics of the relevant building types are known in advance. Second, limited expert knowledge is available, which leads to partial information known in advance, such as the number of types sought, but not their characteristics. Lastly, no expert knowledge is available, thus leading to a situation in which both the number and the content of the building types are unknown. Communications Communications 104 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 3.2. Classification with limited Expert Knowledge If expert knowledge is limited, but still available in some extent, it is possible to make several decisions prior to a series of unsupervised cluster analyses. For 105 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types The left side of Fig.3 is a possible representation of the two clustering within two-dimensional spaces made by extracting the first two principal components (package “factoextra”) along which the variation in the data is maximal. Quality metrics in these clustering analyses are the share of the total variance in the data set that is explained by the clustering, given by dividing the between sum of square (BSS) by the total sum of square (TSS). These values are 94.13% for 4-class collective housing clustering and 93.82% for the 5-class, respectively. The right side of Fig.3 shows radar charts for each variable. It is possible to add variables in the radar charts that have not been used during the clustering processes, such as area and floor in Fig.3. These plots are made with the package “fmsb”. The charts point at highly differentiated profiles, especially for collective housing, with profiles that are partially matching the DTs model. C3 are for example regrouping both high rise residential buildings and Danchi housings (団地), C4 appears to be the equivalent of DTs C4: small-sized complexes, etc. Cluster concordances with the DTs model are less clear for single-family houses, except for S2: articulated detached houses (equivalent to DTs S2) or S1: large villas (DTs S5). S3 and S5 are row houses with close characteristics, also close in the cluster map. In fact, even if literature or partial expert knowledge say that we shall seek for 5 different profiles of single-family homes, it appears that 5 is not a robust number of clusters for the k-means algorithm. In this respect it is possible to either (1) add/remove variables, such as using floor and area, instead of floor-area, or add newly computed morphometric indicators, (2) re-run the procedure (k-means is sensitive to the location of starting centroids) or try another algorithm, such as one of the algorithm discussed at the beginning of this section (3) re-run the algorithm with a different number of sought clusters (4) manually merge the clusters that have close characteristics. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] example, we could know the number of required clusters, as well as the fact that some inputs must not be grouped together. This could for instance be the case for collective and single-family homes, for specialized and residential buildings, or for building possessing large and small footprints. In such cases, standard unsupervised clustering techniques that minimize intra-cluster distances and maximize inter- cluster distances can be separately used for each subset, such as K-Nearest Neighbors, hierarchical clustering, k-means, classification trees, random forests, neural networks, etc. To illustrate this point, one of the simplest technique, k-means clustering, which aims at partitioning inputs into clusters such that the sum of squares from inputs to the assigned cluster centers is minimized, is applied as follows: Supplementary materials algorithm 3 is a R protocol performing two different k-means, one for collective (attribute 1663) and one for single-family homes (attribute 1664). Two random subsets are draw and merged from BU0304 and BU1314 for collective buildings only (sample size of 20% of the inputs), and, once again, from BU0304 and BU1314 for single-family homes only (20% again). Drawing and merging random samples made of both periods altogether allows avoiding giving more weight to one period over the other. A clustering analysis is then performed for each subset (collective housing and single-family homes), with a number of clusters per model chosen a priori. In algorithm 3, we select the same number of clusters as within the DTs model (section 3.1: 4 clusters for collective housing and 5 clusters for single-family homes). Data are scaled and, in order to avoid redundancy, floor and area are removed in favor of floor-area. The two clustering analyses are thus performed on convexity, elongation, number of direct neighbors and floor-area. Once the two models are trained, for both periods, and for both collective and single-family home inputs, the whole datasets are mapped to the nearest clusters. This mapping is made using the k-nearest neighbor searching algorithm of the “FNN” package. FIGURE 3. Cluster Map and Profiles for individual Clusters for Collective Housing (top) and Single-Family Homes (bottom) FIGURE 3. Cluster Map and Profiles for individual Clusters for Collective Housing (top) and Single-Family Homes (bottom) 106 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Regeneration and Perforation of Building Types 3.3. Classification with no Expert Knowledge Finally, in the case of a total lack of expert knowledge, any widely known clustering algorithms can be used, such as k-means, neural networks, or Bayesian networks. The main issues with unsupervised clustering are the assessment of the quality of the partitions, as well as the number of “interesting” partitions (Haldiki et al., 2001). Since there is no way to a priori determine the most suitable number of clusters, researchers tend to use heuristics, such as looking at a cutoff point regarding the improvement of the explained variance (or log-likelihood in the case of Bayesian clustering) for each additional partition (elbow method). An unsupervised Bayesian clustering on BU1314, based on the indicators detailed in section 2.3 (plus specialization) has been performed in Perez et al., (2019a). This clustering used an expectation-maximization algorithm (Dempster et al., 1977) to perform one thousand clustering analyses and a MDL score (combining log- likelihood and a penalization function for the growing number of clusters) to define the optimal number of partitions. This application yields seven interesting clusters, from detached small houses to isolated high-rise buildings. In what follows, we will propose an alternative approach to unsupervised clustering based on Self-Organizing 107 107 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Maps (SOM) (Kohonen, 1982). Bayesian and SOM clustering have already been compared in unsupervised approaches (Fusco and Perez, 2019). Maps (SOM) (Kohonen, 1982). Bayesian and SOM clustering have already been compared in unsupervised approaches (Fusco and Perez, 2019). Maps (SOM) (Kohonen, 1982). Bayesian and SOM clustering have already been compared in unsupervised approaches (Fusco and Perez, 2019). Supplementary materials algorithm 4 is a R protocol performing SOM clustering on a sample made of BU0304 and BU1314 (sample size of 20% of the inputs). This time, we do not distinguish collective from single-family homes buildings. Once again, the sample is made with inputs draw from both periods altogether. The same indicators than in section 3.2 are used. SOM use a neighborhood function for each input to find the Best Matching unit (competitive training). Units, also called nodes, are distributed on a two-dimensional space (map). Once the best matching unit is found, a radius parameter allows updating the neighboring nodes, thus giving topological properties to the map that can be investigated through their proximity. There are two ways to identify clusters within a SOM. 4.1. Automated Grid Creation and Evolution Count In the previous section, we performed three different classifications that have been applied to the different periods associated with our datasets: 2003/04 and in 2013/14. To summarize, algorithm 2, 3 and 4 are individually able to provide a new variable: a class label, that can be compared between two periods. This is the only input required in this section: a class label associated to at least two spatial datasets describing the same study area at different time points. Algorithm 5 creates a grid with predetermined cell size, 250 meters, that are superimposed on the extent of the test region. Cells that are not intersecting any buildings are filtered out of the grid. Then, using a tapply structure, a recursive intersect is performed for each remaining cell to count the buildings grouped by labels. Once the tapply function has been ran for both datasets, a final line of code groups the results of both periods in a single dataframe. This whole process takes only a few seconds with an Intel® Core™ i7- 8700 CPU @ 3.20GHz, 16gb RAM, while, on a traditional GIS software, for each and every class label, an intersect with a grid of this extent takes several minutes to run. For each cell, it is then possible to calculate the difference per building type, in order to map the temporal evolutions of each type. [ISBN : 978-2-910545-12-1] Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types rlen parameter in algorithm 4 must thus be set to a bigger value. Fig.4 b. to e. are the different heatmaps, showing that similar values are aggregating in the same areas. f. is the cluster map, where each color is a different building type. In this example, we set the number of clusters to seven. The black lines are the cluster boundaries, which have been retroactively mapped to the heatmaps. As in k-means, the ratio between BSS and TSS can be calculated for different numbers of clusters. For the 7- class solution, it reaches 84.01%. The “blue” cluster is made of different kinds of small sized single-family houses, the “pink” is for elongated row houses such as wooden Nagaya (長屋), the “purple” is for Danchi housing (団地), the “red” is representative of small collective residential complexes, etc. Once the SOM is trained with the sample, BU0304 and BU1314 data are mapped into the model using the map.kohonen function. 3.3. Classification with no Expert Knowledge First, the number of nodes within the two-dimensional space can be set to a small value. Each node is then considered as a cluster. The second approach, which is the one used in algorithm 4, is to parameterize many nodes for the map (225, i.e. 15 by 15 in our application). The distances between couple of nodes form a distance matrix (U-Matrix). The advantage of this approach is that it allows visualizing each variable distribution across the map. Depicted by colors, these graphical representations are called heatmaps. Since similar values are aggregated in the same areas, heatmaps provide relevant information that can be used prior to the segmentation of the U-Matrix. The segmentation can then be performed using a simple hierarchical clustering, as in algorithm 4, in which the number of divisions is set manually. FIGURE 4. Self-Organizing Map outputs for a sample of BU0304 and BU1314 a. Training Progress b. Heatmap for direct neighbors c. Heatmap for floor-area d. Heatmap for Elongation e. Heatmap for Convexity f. Clustering results. FIGURE 4. Self-Organizing Map outputs for a sample of BU0304 and BU1314 a. Training Progress b. Heatmap for direct neighbors c. Heatmap for floor-area d. Heatmap for Elongation e. Heatmap for Convexity f. Clustering results. Fig.4 displays the outputs of a SOM clustering performed on a sample made of BU0304 and BU1314 using the “kohonen” package. The first thing to control is that the training curve flattens, as in Fig.4 a. The training is not over if it doesn’t, the Communications 108 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] 4.2. Mapping To map the count and evolution of building types, we use the “tmap” package, which allows building thematic maps following a layer-based structure. Supplementary materials algorithm 6 shows a layered structure leading to the map of the count of the number of building for the C3 profile of the DTs model (Fig.5 a.). Before setting the maps, we first import an OpenStreetMap background. C3 values are then discretized into several categories (natural breaks), sequentially added as new layers to the map. Algorithm 6 is interesting on both counts. First, for each category, cells are dissolved into a single multipart polygon before being Communications 109 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] plotted. It allows plotting only the geometries thus substantially reducing computation time. Second, for each category, polygons are smoothed in order to make localized aggregates stand out. FIGURE 5. 2013-14 Count and 2003-4/2013-14 Evolution of building types for selected Profiles of the DTs model a. Count for C3 b. Evolution for C3 c. Count for S1 d. Evolution for S1 e. Count for S3 f. Evolution for S3 FIGURE 5. 2013-14 Count and 2003-4/2013-14 Evolution of building types for selected Profiles of the DTs model a. Count for C3 b. Evolution for C3 c. Count for S1 d. Evolution for S1 e. Count for S3 f. Evolution for S3 Fig.5 shows the count and evolution of three building types from the DTs model. Mappings within the above figure can be realized for any building type and for any of the three models discussed in section 3 using the structure of algorithm 6. C3 are mid to high rise compact/narrow buildings, mostly located in the central part of Osaka. The evolution map shows different hot and cold spots within this central area, thus pointing at a building type which is self-regenerating. S1 are convex detached houses mostly located outside of the center of Osaka. The evolution map shows that this kind of single-family homes is increasing, which is consistent with the literature pointing at a spread of small-sized prefabricated homes in Japan (Buntrock, 2017). S3 are small and elongated row houses, typical of traditional wooden Nagaya (長屋). This type is also located outside central Osaka, but also highly concentrated in specific neighborhoods. 4.2. Mapping The evolution map points at a gradual disappearance of this traditional building type, even perhaps at a replacement if we link this disappearance with the increase of modern prefabricated homes. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types Regeneration and Perforation of Building Types layers of building footprints with heights (or floors) as attribute data are available for at least two different time points. Another attribute is valuable, but not mandatory: building specialization. Attention should be paid to the time depth between the layers, as it must be sufficient to analyze structural changes in the spatial organization of building types. The temporality behind these structural changes is also dependent on cultural specificities regarding planning, building norms and urban development traditions (deconstruction, dismantlement, renovation, etc.). If the prerequisites are fulfilled, then it is possible to compute a basic set of indicators for each building, such as the one presented in section 2.3. Based on the kind of available expert knowledge, different types of classifications have been performed, from manually built decision trees to fully unsupervised clustering. Regardless of the method, quality metrics show that significantly different clusters can be found. It is possible to check the validity of each classification by drawing random samples and go on the field, or use a street-based urban imagery (such as Google Street View), then calculate the usual sensitivity and specificity metrics. The automated superimposition of a grid then allows counting occurrences of each type, as well as quantify the difference per cell between the different time points. For each category, mappings are performed, with cells that are discretized in the same category dissolved into a single multipart polygon, thus highlighting localized aggregates and, in the case of temporal evolution, hot and cold spots of perforation and regeneration of building types. This paper focused on methodological suggestions, proposals and recommendations for studying building evolutions. Regarding the clustering applications, the methodological propositions could be improved, especially by focusing on maintaining output coherence over different settings or initializations. A matrix of similarity indexes can for example be calculated among the clustering results to evaluate the robustness of the protocols to pseudorandom number generation (seed), and cross-validation methods, such a k-fold or Jackknife resampling, can be used to evaluate the robustness of the outputs to different random samples. Yet, interesting preliminary results already stand out regarding specific locations and evolution trajectories of certain building types in Osaka, Japan. [ISBN : 978-2-910545-12-1] It shows that the perpetual mutation that characterize urban areas is linked, at least to some extent and to different degrees, to urban form. This research opens interesting perspectives, such as the study of the disappearance of some building types, or the gradual replacement of some types by other. Several methods, from standard correlation matrices to geographically weighted regressions, could allow exploring these dynamics in conjunction with other factors, such as population evolution, socio-demographic characteristics, planning policies, etc. These results must be analyzed together with literature and fieldwork related to the metropolitan area of Osaka. Bonnin, P., Nishida, M. and Inaga, S. (2014) Vocabulaire de la spatialité japonaise. CNRS Editions, 560 p. 5. Conclusion and Discussion This paper put forward several recommendations and suggestions with the aim of studying the spatial location and temporal evolution of building types in large metropolitan areas. The first and most important condition is that high quality GIS 110 Communications 110 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 6. References 111 Communications 111 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Buhnik, S. (2010) From Shrinking Cities to Toshi no Shukushō: Identifying Patterns of Urban Shrinkage in the Osaka Metropolitan Area. Berkeley Planning Journal, 23(1), pp. 132-155. Buntrock, D. (2017) “Prefabricated housing in Japan”, in Smith, E. and Quale, J. (Eds.), Offsite Architecture Constructing the Future, Routledge: New York, NY, USA, pp. 190–213. Dempster, A., Laird, N. and Rubin, D (1977) "Maximum likelihood from incomplete data via the EM algorithm", Journal of the Royal Statistical Society. Series B, 39(1), pp. 1-38. Fusco G. and Perez J. (2019) Bayesian Network Clustering and Self-Organizing Maps under the Test of Indian Districts. A comparison, Cybergeo: European Journal of Geography, paper 887. Hecht, R., Meinel, G. and Buchroithner, M. (2015) Automatic identification of building types based on topographic databases – a comparison of different data sources. International Journal of Cartography, 1(1), pp. 18-31. Haldiki, M., Batistakis, Y. and Vazirgiannis, M. (2001) "On Clustering Validation Techniques", Journal of Intelligent Information Systems, Vol.17, pp. 107-145. Jacobs J. (1961) The Death and Life of Great American Cities, Random House, New York, 458 p. Kohonen, T. (1982) “Self-organized formation of topologically correct feature maps”, Biological Cybernetics, volume 43, pp. 59-69. MLIT. (2007) “White Paper on Land, Infrastructure, Transport and Tourism in Japan”, Ministry of Land, Infrastructure, Transport and Tourism Publication, Ministry of Land: Tokyo, Japan, 68 p. Perez, J., Fusco, G., Araldi, A. and Fuse, T. (2019a) “Identifying building typologies and their spatial patterns in the metropolitan areas of Marseille and Osaka”, Asia- Pacific Journal of Regional Science, 4, pp. 193-217. Perez, J., Araldi, A., Fusco, G. and Fuse, T. (2019b) “The Character of Urban Japan: Overview of Osaka-Kobe’s Cityscapes”, Urban Science, 3(4), 105. Salingaros, N. (2005) Principles of Urban Structure, Techne Press, Delft, 254 p. Shelton, B. (2012) Learning from the Japanese City: West Meets East in Urban Design. Taylor & Francis, 224 p. https://zenodo.org/record/4551462 Acknowledgment: This study is supported by Joint Research Program no. 774 at CSIS, UTokyo (Zmap TOWN II 2013/14 Shapefile Osaka prefecture). 112 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Université Côte d’Azur, CNRS, ESPACE, 98, Bd Herriot, BP 3209, 06200 Nice, France. [alessandro.araldi, giovanni.fusco]@univ-cotedazur.fr 2. KCITYLABS - Kinaxia Group, 80, route des Lucioles, 06560 Valbonne, France. [david.emsellem, denis.overal]@kcitylabs.fr 3. Université Côte d’Azur, I3S 2000, route des Lucioles, 06900 Sophia Antipolis, France. d i@ i d f Alessandro Araldi1, David Emsellem2, Giovanni Fusco1, Andrea Tettamanzi3, Denis Overal2 1. Université Côte d’Azur, CNRS, ESPACE, 98, Bd Herriot, BP 3209, 06200 Nice, France. [alessandro.araldi, giovanni.fusco]@univ-coteda 1. Université Côte d’Azur, CNRS, ESPACE, 98, Bd Herriot, BP 3209, 06200 Nice, France. [alessandro.araldi, giovanni.fusco]@univ-cotedazur 1. Introduction An interdisciplinary and growing body of research investigates the relationship occurring between the form and the functioning of cities (Carmona, 2019). In the typo-morphological tradition, the properties of the urban form are defined by specific spatial combinations of its constitutive elements: streets, plots and buildings (Moudon, 1997). While streets and plots show a higher inertia over time, buildings change at a faster pace depending on the specific historical, socio-economic and technological context. Moreover, while some buildings (and building types) are easily torn down and replaced, others can endure for several decades or centuries showing higher resilience to urban transformation. Both geography and urban typo- morphology developed multiple approaches and protocols for the identification and analysis of building types based on the comparative analysis of buildings. As defined by Sheer (2017, p.171) “a building type is an abstraction, a pattern, where we observe formal similarities between one building and another even though the buildings may have different architectural expressions. [] buildings might share many common formal characteristics, but are very different in color, materials, style, and expressiveness.” Building types can result from a given time period, in a specific regional context with certain easily recognizable stylistic patterns (i.e. the English terraced house; the American bungalow house; the Parisian Haussmann apartment-building). They could also be more overarching, grouping together similar buildings produced in different time periods, but sharing consistent common features. Different information at different description levels can be used for the identification and description of building types. The combination of specific features can thus result in a comprehensive building typology, i.e. an organization of building types having a given logic and inner coherence. Three levels of details can be distinguished: i) aesthetical and stylistic features (such as façades materials and composition, colours, etc.), ii) the internal organisation of the building (including some structural considerations) iii) the overall external hull of a building (shape, footprint area, height etc.). Nonetheless, formal definitions of building types, categories and structures describing the whole building stock of a large urban region are extremely rare (Orford and Readcliffe 2007). Similarly, there is no agreement about the definition of which combination of formal characteristics is required for the identification of building types and their differentiation, since different sets of features underlie the definition of each building type. andrea.tettamanzi@univ-cotedazur.fr RESUME. L'identification et la description des typologies de bâtiments jouent un rôle fondamental dans la compréhension de la forme de l’espace bâti. Un nombre croissant de travaux développe et implémente de nouveaux protocoles sophistiqués de géomatique pour l'identification des typologies de bâtiments et leur organisation. Cet article partage le même objectif. Une procédure innovante, basée sur l’analyse quantitative des données est présentée ici avec, comme objectif, l'identification et la description non supervisée des types de bâtiments et de leur organisation. Après une procédure de pré-traitement spécifiquement adaptée à notre donnée source, nous développons un protocole de clustering non supervisé combinant un algorithme novateur d’inférence Bayésienne Naïve avec des approches ascendantes hiérarchiques; le tout, reposant sur cinq caractéristiques morphométriques intrinsèques de chaque bâtiment. Ce protocole permet d'identifier des groupes de bâtiments partageant des caractéristiques morphologiques similaires spécifiques ainsi que leur structure globale à différents niveaux d'agrégation. La méthodologie proposée est implémentée et évaluée dans l'espace d’étude du Département des Alpes-Maritimes, France. ABSTRACT. The identification and description of building typologies play a fundamental role in the understanding of the overall built-up form. A growing body of works is developing and implementing sophisticated, computer-aided protocols for the identification of building typologies and their organisation. This paper shares the same goal. An innovative data-driven procedure for the unsupervised identification and description of building types and organisation is here presented. After a specific pre-processing procedure, we develop an unsupervised clustering combining a new algorithm of Naive Bayes inference and hierarchical ascendant approaches relying on five morphometric features of buildings. This protocol allows us to identify groups of buildings sharing specific similar morphological characteristics and their overall structure at different aggregation levels. The proposed methodology is implemented and evaluated in the case study of the Alpes-Maritimes Department, France. MOTS-CLES : Bâtiments, Typo-morphologie, Clustering Bayésien Naïve, CAH KEYWORDS: Building, Typo-morphology, Bayesian Naïve Clustering, HCA 113 113 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction When the goal of the study is the identification of building types over a large urban region, databases encompass few features, often limited to the simple geometrical form. The study of the external building hulls allows typifying what might be called the skeletal form of the building. Two main approaches are currently used for the identification of building types (Hecht et al. 2015). Knowledge-based approaches have been proposed based on the study of small but exhaustive datasets including building features at each of the three detail levels. These approaches are based on expert knowledge of the relevant building types and features in a given study area but are hardly adaptable to other regions, or in the presence of a poorer description of buildings. Computer-aided protocols have been 114 114 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] developed since the 70s, especially by the Centre for Land Use and Built Form Studies (LUBFS) at Cambridge University (Steadman 2016). Thanks to more recent data analysis developments, together with the increasing computing power availability, systematic and quantitative protocols have been developed: Data-based approaches have been proposed especially from digital cartography, with the goal of cartographic generalisation or for the identification of urban structures, building detection and building pattern recognition. While the study of building typologies with expert-based protocols and based on highly detailed and historical datasets finds its origin in the urban typo-morphology tradition since the 1950s, the identification of building types from footprint-based data is a more recent field of study (Hecht 2015, 2016). Two subgroups might be further specified in data-based approaches. Supervised protocols (similarly to knowledge-based analysis) require prior knowledge of the target groups we want to identify within the dataset: features are attributed to each group based on similarity rules. Examples of classificatory approaches can be found in Orford and Radcliffe 2007, Hech et al 2015, Hartmann 2016, etc. Unsupervised approaches encompass clustering protocols where the identification of groups is based on algorithms looking for internal similarity among features and without prior knowledge of the target groups or user intervention. Clustering protocols automatically determine natural partitions (clusters) from the input arising from the specific data structure without imposing a predefined identification of the classes. Examples can be found in the works of Schirmer and Axhausen 2015, Perez et al. 2019, etc. In fact, expert-based knowledge is never completely absent. 2.1. Data preprocessing The protocol presented in this paper is implemented on French Department of Alpes- Maritimes (Section 3.1). In France, the BD TOPO® by the National Geographic Institute (IGN) is an exhaustive dataset of metrical precision providing the information on building footprint and height, retrieved from satellite or aerial imagery. Since April 2019, a new version of this dataset has been released (BD TOPO®, V3.0): its main novelty consists in the combination of the original data with information from the national computerized cadastral plan (MAJIC). Several new features are made available such as the number of dwellings, the age of construction and the number of stories. Nonetheless, this dataset has some limitations, among which three are of concern for our research, requiring specific pre-processing protocols. The first issue is related to the building footprint definition. Part of the building dataset, enriched by the cadastral plan information, have a detailed definition of their footprint: while one (or more) polygon(s) corresponds to the main built-up body(ies), a number of extensions (such as terraces, loggias, porches, etc.) are separately modelled as adjoining polygons. These extensions are identified by a specific feature attribute, namely “light structure”. The same attribute is also associated with independent structures such as greenhouses, garages, small and large industrial shads. Another share of buildings is defined from satellite/aerial based methods: in this case, the delineation of the footprint is defined with imagery detection algorithms where the footprint corresponds to the external demarcation of the overall built-up structure, therefore including both the main building body and all extensions. Thus, a harmonization protocol has been specifically developed in order to re-aggregate buildings made by several constituent parts. The second issue is related to the definition of the functional specialization attribute. The feature “Building Nature” provides the distinction between several specialized structures (i.e. religious, industrial, agricultural, etc.). When the ‘overall architecture or aspect of a building does not reveal with exactitude its function’ (ibid.) the building is classified as undifferentiated. After a manual assessment of this field on a subspace encompassing about 15 thousand buildings, we observed that the specialization attribute is always correctly assigned, although limited to 5.1% of the actual functional specialized buildings (overall accuracy 54.2%). Thus, this feature is enriched through another IGN BD TOPO layer, namely ‘Activity Zones’, where specialized buildings are retrieved and collected from other national authoritative sources. 2. Method 2. Method 2. Method 1. Introduction In supervised protocols, it is required at the beginning of the analysis to define the target groups, their numerosity and their overall organisation. In clustering approaches, it is needed for the interpretation of outcomes. The former allows the analyst to better identify specific predefined building typologies, while the latter allows a more exploratory analysis where natural groups emerge from the data structure and are later interpreted and related to the specific characteristics of the study region. When focusing on clustering approaches, group identification is also influenced by the underlying algorithmic rules. As discussed in Fusco and Perez (2019) most of the traditional approaches (such as K-means) impose the sphericity of clusters (i.e. intra-cluster homogeneity) on all the descriptive variables, which could not always be coherent with the complexity of the context under study. Bayesian Clustering (BC) allows us to overcome these limitations. Still, as for most of the clustering approaches, BC, even when using Naïve clustering models, imposes other kinds of constraints and can be particularly time-consuming when exploring possible solutions in parameter space. This paper presents a specific protocol aiming to identify building typologies of a large urban region and their overall organisation combining unsupervised, incremental Naïve Bayesian and Hierarchical clustering approaches. The protocol is implemented on a few selected features describing the morphological characteristics of building hulls. The paper is organised as follows. Section 2 describes the proposed clustering protocol. In Section 3, the protocol is implemented to the study area of the Alpes-Maritimes Department, France, and its results are described and interpreted. Section 4 closes the paper with a discussion on limitations of the protocol, improvements and future perspectives. 115 115 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 2.1. Data preprocessing A set of specific rules and filters are defined and implemented to associate the specialized function to the original building dataset: on the same 15-thousand features subset, the resulting enriched definition of the field shows 86.8% accuracy score in identifying specialized buildings. This allows us to filter them and implement our clustering protocols only on ordinary residential or mixed function buildings. Indeed, as discussed in the typo- morphological literature, specialized buildings (industrial, commercial, public, etc.) have often specific and extreme morphological properties values (Caniggia et Maffei 2001) introducing important outliers in our dataset, biasing the final outcomes. Communications 116 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Once the subset of non-specialized buildings is redefined, morphometric descriptors can be implemented. Three indicators are directly computed from building footprint: Surface (S), Elongation (E) and Convexity (C). One indicator, Topological Contiguity (TC), is defined as the number of neighbours within a continuous built-up unit of adjoining neighbours. Finally, the building Height (H) is provided by the original dataset. These five indicators represent a set of minimal description of the building form obtained by a simplified 3D dataset (LOD0+ of the CityGML data model in Biljecki et al., 2016). Nonetheless, building Height (as for several features of the latest BD Topo V0.3) can be unknown. As we will see in the next section, the clustering protocol developed in this work is specifically conceived to deal with partial missing information. As previously introduced, BC algorithm is based on a probabilistic framework. This approach requires qualitative or categorical data. A discretization of the five morphometric descriptors is therefore necessary: for our case study the discretisation was obtained through a mix of univariate data analysis and expert knowledge. This last pre-processing step produces a first reduction of the overall complexity of the original data: in our specific case study, for example, it allows us to pass from about 300 thousand building units to 2.2 thousand building tuples, each one corresponding to a specific combination of our features modalities. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] ordinal structure of our data, we used Manhattan distance instead of Hamming distance, after normalizing for the cardinality of the ordinal values. Secondly, the usual Laplacian smoothing in the initialization of conditional probability tables, is reinforced by a further smoothing on the ordinal values which are contiguous to those of the seeds. After seed selection, all remaining records are assigned one by one to the cluster having maximum log-likelihood, and the cluster conditional probability tables will be updated after each record assignment. Iterative random removal of individual records from their current clusters to be re-evaluated and reassigned to a better cluster can thus begin. Given the Naïve model architecture and the resulting additive formula of model log-likelihood, the local optimization of log-likelihood in record assignment to clusters produces a global log-likelihood improvement for the whole model. More precisely, iterative record clustering is done in two phases. In the first phase, cluster priors are considered equal for all k clusters. In a second phase, cluster priors are determined from the current cluster probability distributions, allowing a more precise calculus of posterior probabilities in Bayes’ formula. A further improvement of McCaffrey’s original INBIAC algorithm has been the treatment of missing values. Under the Missing at Random assumption, likelihoods and posterior probabilities of cluster assignments are calculated only on the observed values, but missing value imputation is later performed based on the most probable values within the assigned cluster. Imputed values are iteratively erased and re- imputed within the INBIAC procedure, and the final log-likelihood of the clustering solution includes the contribution of imputed values. The clustering iterations within INBIAC stop when no record can be reassigned to a different cluster. Finally, buildings are weighted by their footprint surface in the algorithm, giving the same importance to each m2 of built-up surface (the clustering solution would otherwise be biased by the overrepresentation of small buildings). The implementation of the protocol for several numbers of clusters k, ranging in a user-defined interval, allows to explore different clustering solutions. The optimal clustering solution(s) can be then selected through a simple graphic method: similarly to the well-known elbow method, the log-likelihood loss scores are plotted across different numbers of clusters k (Fig.1.a-b). The presence of strong variations might reveal structural changes in the data clustering corresponding to suboptimal solutions at a given number of classes k. 2.2. Iterative Naive Bayesian Inference Agglomerative Clustering Bayesian inference is a powerful probabilistic option for quantitative and qualitative multivariate data clustering using simple model architectures as the Naïve Bayesian Classifier, where the cluster variable is conceived as the common parent of all the other variables, and conditional independence among them is assumed knowing the cluster variable (Duda and Hart 1973). EM algorithm is normally used to identify an optimal clustering solution in terms of log-likelihood, for a given number of clusters (Dempster et al. 1977). Exploring solutions with varying number of clusters can be done with a random walk in solution space, using a clustering score combining log- likelihood and a penalization for model complexity, i.e. number of clusters (like in Fusco and Perez 2019). McCaffrey (2013) offers an interesting alternative to the EM algorithm for BC: Iterative Naive Bayesian Inference Agglomerative Clustering (INBIAC). So far, the implementation of the INBIAC algorithm can be only found in Carneiro et al. (2015) for credit card fraud detection. INBIAC is a much faster algorithm than EM as it replaces recursive batch inference of cluster assignments for all records to an iterative assignment of individual records which are randomly extracted from the database and assigned to the highest likelihood cluster at that given moment of the clustering procedure. The higher speed of the INBIAC algorithm can be used to perform a higher number of clustering solutions. Just like EM, INBIAC results are sensitive to the clustering initialization. In EM, initialization implies randomly assigning all records to clusters. In INBIAC, a k- cluster solution needs the use of k records as initial cluster seeds. McCaffrey (2013) proposes a preliminary phase of seed initialization, randomly choosing k seeds and finally keeping the set of seeds with maximum Hamming distance. In our algorithm we improved McCaffrey’s protocol in several respects. Firstly, to better represent the 117 117 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 2.3. HCA This approach shares the same underlying hypothesis of consensus clustering protocols (Monti et al., 2003), where several cluster solutions are combined in order to achieve a more robust solution. Nonetheless while consensus clustering looks for similarities within a larger number of clustering solutions for a given number of clusters k, in our case we use a more ‘controlled’ subset of n suboptimal solutions at different levels k. Moreover, instead of implementing the same clustering protocol at two different stages of the analysis, we combine a non-hierarchical and a hierarchical protocol in the first and second stage of the analysis, respectively. Finally, our protocol provides a profile for each cluster as probability distributions of its members over the values of each clustering variable. These profiles are later used for cluster interpretation (section 3.2). 2.3. HCA The implementation of the INBIAC protocol, allows us to identify one (or more) optimal solution(s) based on the optimization of specific scores and descriptive parameters. Nonetheless, independently from the specific set of parameters used for the model evaluation, the selected clustering(s) solution(s) would always provide a specific partition of the original dataset defined for a given number of groups k. It is thus interesting to study the variation of building clustering across k. Buildings constantly grouped in the same clusters could reveal stronger structural patterns within the data. We thus implement an agglomerative Hierarchical Clustering Analysis (HCA) using as input the INBIAC best outcomes for each k clustering solution in the interval explored. The rationale underlying this methodological choice is the 118 Communications 118 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] following: the subset of n best clustering solutions can be used to partition our original dataset (or similarly, the 2.2 thousand tuples) in smaller subgroups of elements (kernels) always being clustered together independently of the number of clusters k. These kernels represent the finest partition for which the highest inter-level consensus is observed: no cluster at any level further divides the element in finer groups. Within our specific context, these kernels correspond to a highly detailed meta-cluster solution of specific building sub-types; few kernels gather most of the buildings (more precisely of the built-up surface, given our weighting scheme), and vice versa a large number of kernels encompass less built-up surface with less recurrent shapes. HCA will be here implemented with Gower's dissimilarity metric for cluster distance and Ward-linkage agglomerative principle among clusters. Implementing an HCA allows us to produce hierarchically nested groupings based on the similarities within this elementary kernel partition. Kernels are arranged in a hierarchical manner. Thus, the combination of the INBIAC and HCA protocol combines the advantages of the two protocols. On the one side, we keep the ability of probabilistic Bayesian inference to select non-spherical clusters defined with a maximum log-likelihood approach on subgroups of features. On the other, an overall hierarchical structure allows the analyst to observe the overall data clustering organization similarly to knowledge- and ontological-based classifications. Moreover, the main advantage compared to regular HCA applied on raw data, is that the outcome variability produced by the high sensitivity to the initial clustering settings is strongly reduced. 3.2 Implementation and results From the original 500 thousand polygons, the data pre-processing protocol allows us to identify about 300 thousand buildings: 30% are specialized and the remaining 70% have a residential or mixed use. Our clustering protocol will be applied here only to this larger stock of ordinary buildings. The segmentation of the five features (S, H, C, E, TC) further reduces the variability of our dataset resulting in 2.2 thousand observed combinations. Then, the INBIAC protocol is implemented exploring 100 solutions with different initial seeds, for each number of clusters k between 5 and 15 producing a total of 1.1 thousand models (Fig.1.a). The quality parameters of these solutions vary with k: loglikelihood loss is maximal for the 5-cluster solution and minimal for the 15-cluster solution. Fig.1.a shows the scatter plot of the log-likelihood loss vs the category utility score (Gluck and Corter, 1985) for all the models implemented: important gains are observed for the 6-, 8- and 11-cluster solutions, further confirmed by the elbow method applied on the log-likelihood loss scores (Fig.1.b). According to our approach, we will nevertheless use the information derived from all the 11 best clustering solutions. Their combination further defines 280 kernels, 15 of which encompass 57% of the overall built-up surface. The implementation of the agglomerative HCA on these kernels, allows us to identify and describe the overall organization of the building types in the study area (Fig.1.c-e). The dendrogram in Fig.1.c shows the succession of cluster agglomerations along a distance axis, starting from the 280 kernels (below) and arriving to the complete amalgamation of clusters (top). The length of the segment on the distance axis during which a given k-cluster solution is present is indicative of its importance in structuring the building typology in the study area. The first four solutions showing the longest segments on the distance axis of the dendrogram correspond to 4, 10, 21 and 8 clusters (Fig.1.d). The ten-cluster solution is described as follows. A.1 corresponds to compact small detached single-family houses and duplexes while A.2 to compact small detached two and three stories houses; they account for 23.5 and 12.6 % of the overall built-up surface of the study area, respectively. These groups correspond to two of the most common building types of the French Riviera, known as the niçois house. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] hinterland represents over one third of the overall departmental surface but has a population of only 15 thousand inhabitants (1% of the departmental population). The building stock of the Alpes-Maritimes shows a hyper-concentrated urban fabric along the coast, a strong sprawl in the western sector, and an urban development in the eastern sector strongly influenced by its higher topographic constraints. With the exclusion of Grasse, the other five main urban centres are located along the coastline, with other 11 municipalities, gathering 42% of the overall departmental buildings (51% building surface). When including the whole of the coastal conurbation 33% of the departmental surface encompasses almost 97% of the building stock. 3.1 Study Area The protocol presented in this work, is implemented on the building stock of the Alpes-Maritimes Department in southern France. This study area is made up of a large coastal conurbation and of its alpine hinterland. The coastal conurbation of the French Riviera stretches over 60 km from the French-Italian border to the Esterel mountains. Its western section includes the cities of Cannes, Grasse and Antibes, counting 74.2, 51 and 73.8 thousand inhabitants, respectively. Nice, with its 343 thousand inhabitants represents the largest municipality of the French Riviera and the administrative centre of the Department. The enclave of Monaco and the border city of Menton have respectively 38 and 28 thousand inhabitants. Spread around these main centres, 295 thousand people find their home in smaller cities, villages and hamlets. The alpine 119 119 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3.2 Implementation and results These building types often do not exceed 150 m2 and three stories; they represent a specific production of the early XX century and in the second after war. While A.1 one corresponds to a single dwelling house, A.2 might also include few dwellings initially intended for different members of the same family group. In 120 Communications 120 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] the same building type, we also find some more recent small real estate development projects, often used as second homes and for touristic purposes. FIGURE 1. Outcomes of the building classification: a) Scatter plot of the 1000 models scores obtained with the INBIAC clustering: log-likelihood loss vs. category utility (CU); b) Box-plots of the top 20 INBIAC models scoring the highest loglikelihood-loss scores for each number of cluster k in the interval [5:15]; c) HCA dendrogram implemented on the outcomes of the 11 best scoring INBIAC models; d) identification of the partitions corresponding to the highest dendrogram depths; e) interpretation of the 10-cluster HCA solution: the dendrogram partitions between 2 and 10 are weighted by the building surface assigned to each group. c a b d e b c a b d e e e FIGURE 1. Outcomes of the building classification: a) Scatter plot of the 1000 models scores obtained with the INBIAC clustering: log-likelihood loss vs. category utility (CU); b) Box-plots of the top 20 INBIAC models scoring the highest loglikelihood-loss scores for each number of cluster k in the interval [5:15]; c) HCA dendrogram implemented on the outcomes of the 11 best scoring INBIAC models; d) identification of the partitions corresponding to the highest dendrogram depths; e) interpretation of the 10-cluster HCA solution: the dendrogram partitions between 2 and 10 are weighted by the building surface assigned to each group. 121 Communications 121 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 B.1 Identifies articulated homes and villas while B.2 corresponds to very articulated large villas (18.7 and 2.8 % of the overall built-up surface, respectively). Both groups combine large villas and (semi) detached houses with a footprint surface mainly between 150 and 300 m2. On the one side, we find large individual compact homes and villas produced before the ‘50s which have been enriched by later functional additions/extensions (garages, porches, etc.). 3.2 Implementation and results On the other side, more recent large articulated villas resulting from the suburban sprawl of end of the XX century and the beginning of 2000. What mainly differentiate the B.2 from B.1, is a more complex shape together with higher building heights (2-3 stories). High standing villas often occupy geographic locations with specific peculiar positions such as the top of hills, capes etc.; large gated communities are mainly made of these two building types. C1. and C.2 are small-sized (<300 m2) mid-rise adjoining buildings (1 and 6.9% of the total building surface, respectively). Most of the medieval centres of cities and villages are made up of these groups, as well as more informal settlements along connective axes. C1 differs from C2 for its lower height (1-2 stories) and its larger surface; moreover, C1 can often be found in more recent urban fabric as a residual of a previous settlement, surrounded (and often substituted) by mid-rise buildings D. D. Adjoining mid-size and mid-rise apartment buildings represents the most common building type characterizing city centres outside their historical core (8.3% of the total building surface). Despite their different aesthetical properties (ranging from belle epoque, art deco or more modern and contemporary stylistic features), these buildings share similar footprint size (between 300 and 600 m2), elevated contiguity and height (above 4 stories, but for long time limited at 6 stories by cityscape rules). E. Two building types can be found within cluster E (7.1% of the building surface). On the one side we find mid-sized and mid-rise detached and semi-detached large apartment buildings. On the other we find some huge (historical) villas and manors, often occupying the most priced locations of the French Riviera (such as capes) and immersed in residential areas made by large villas of type B. F. Free standing mid-to-high rise towers (8.6% of the total building surface). Towers are apartment buildings whose simple compact footprint is relatively small in comparison to their vertical development. More precisely, cluster F encompasses two different building types: traditional small mid-rise towers (3 to 6 storeys and building footprint between 300 and 600 m2) and more modern, larger high-rise towers (more than 6 storey-tall but rarely skyscrapers in our study area). G. Freestanding very large and high-rise apartment buildings, with articulated shapes, (10% of the total building surface). 3.2 Implementation and results Buildings of G type have a footprint surface above 1200 m2; they have very elongated and/or composite shaped buildings; they can be freestanding or belonging to large and complex urban residential development projects. Most of the housing production is located in the close peripheries; their production only started from the late 60s and it encompasses both high standing apartment buildings and large projects developed by local and national housing programs. Some of these exceptional buildings can also be found in more central and 122 122 Communications 4. Conclusion and perspective In this paper we presented an innovative protocol for the identification and description of building types and their overall organisation combining Naïve Bayesian and Hierarchical clustering protocols. The outcomes of this systematic and quantitative analysis allow the data-driven derivation of a system of typologies of buildings, hierarchically organised. The protocol is described and implemented for the real- world contemporary case study of the Alpes-Maritimes Department. Ten building types have been identified and described. Several research perspectives can be outlined. From the methodological point of view, sensitivity analyses should be implemented to assess the robustness of the three main steps of the protocol presented in this paper. The first step considers the role of the variable discretisation: the same protocol should be evaluated both with a general binning method and with segmentation approaches based on the specific statistical distribution of variables observed in the (sub)region under analysis. The second and third phases correspond to the two clustering protocols: both INBIAC and HCA can be assessed considering different distance measures and evaluating their validity under different parametric conditions. Independently of the specific parametric choices, a comparative analysis should also be carried out with more traditional approaches (i.e. k-means, DBSCAN) in order to further identify relative strengths and weaknesses of the protocol here proposed, both in the specific context of building type identification but also within other thematic fields. From the thematic-related point of view four major directions for further developments can be outlined. Firstly, this same protocol can be tested with an incremental number of descriptors of the building envelope aiming at testing and identifying the role played by individual morphometrics into the building typologies. This work might contribute to the debate on the definition of a reliable and universally accepted set of characters and variables for the identification of building envelope typologies. Moreover, internal layout and details of style, facade, roof coverage might also be included: the implementation of the same clustering approach with different levels and granularity of information can shed a new light on the relative role played by skeletal, internal and stylistic features in the identification and definition of building typologies. Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 compact contexts (i.e. demolition and reconstruction of a large articulated buildings built over an entire block). compact contexts (i.e. demolition and reconstruction of a large articulated buildings built over an entire block). 4. Conclusion and perspective Secondly the protocol proposed in this work should be tested on different study areas and at larger scales, to allow a better appreciation of the methodological robustness, its computational and geographical scalability; more important, the reproducibility will permit systematic, quantitative and fine grained comparative analysis on a large number of case studies (both with national-wide datasets and with volunteered geographic large dataset such as OpenStreetMap at the international level). Thirdly, the analysis of the spatial organisation of buildings types and their relative cooccurrences represent a key factor in the definition of streetscapes, urban fabrics and morphological regions, allowing to investigate the multi-scalar 123 123 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] nature of cities based on the finest level of the building unit. The work here described represents indeed only the first step of an undergoing larger research aiming at understanding the building types and the urban fabrics of French contemporary cities. nature of cities based on the finest level of the building unit. The work here described represents indeed only the first step of an undergoing larger research aiming at understanding the building types and the urban fabrics of French contemporary cities. Acknowledgment: This research was funded by the IDEX UCA JEDI, within the AAP Partenariat 2019 (action 6.5). References Caniggia G, Maffei GL (1979). Lettura dell’edilizia di base. Firenze: Alinea. Carmona M, (2019) Place value: place quality and its impact on health, social, economic and environmental outcomes, Journal of Urban Design, 24:1, 1-48. Carneiro EM, et al. (2015) Cluster analysis and artificial neural networks: A case study in credit card fraud detection. 2015 12th Int. Conference on Information Technology. IEEE, 2015. Dempster AP, Laird NM, Rubin DB (1977). Maximum Likelihood from Incomplete Data via the EM Algorithm. Journal of the Royal Statistical Society, Series B. 39 (1): 1–38 Duda RO, Hart PE (1973) Pattern classification and scene analysis. NY: John Wiley & Sons. Fusco G, Perez J (2019) Bayesian Clustering and SOM Neural Networks under the Test of Indian Districts. A comparison. Cybergeo, n°887, .https://journals.openedition.org/cybergeo/31909 Gluck MA, Corter JE (1985), Information, uncertainty, and the utility of categories, Program of the Seventh Annual Conference of the Cognitive Science Society, pp. 283–287 Hartmann A, Meinel G, Hecht R, et Behnisch M (2016). A workflow for automatic quantification of structure and dynamic of the German building stock using official spatial data. ISPRS International Journal of Geo-Information, 5(8), 142. Hecht R., Meinel G. & Buchroithner M. (2015). Automatic identification of building types based on topographic databases – a comparison of different data sources, International Journal of Cartography, 1:1, 18-31. IGN (2020) BD TOPO®, Version 3.0. Descriptif de contenu. https://geoservices.ign.fr/ressources_documentaires/Espace_documentaire/BASES_VECTOR IELLES/BDTOPO/DC_BDTOPO_3-0.pdf McCaffrey J. (2013) Data Clustering Using Naive Bayes Inference. http://msdn.microsoft.com/en-us/magazine/jj991980.aspx. Monti S, Tamayo P, Mesirov J et Golub T, (2003). Consensus clustering: a resampling-based method for class discovery and visualization of gene expression microarray data. Machine learning, 52(1-2), pp.91-118. Moudon, A. V. (1997). Urban morphology as an emerging interdisciplinary field. Urban morphology, 1(1), 3-10. Orford, S., Radcliffe, J. (2007). Modelling UK residential dwelling types using OS Mastermap data: A comparison to the 2001 census. Computers, Env. & Urban Systems, 31(2), 206–227. Perez J., Fusco, G., Araldi, A., & Fuse, T. (2020). Identifying building typologies and their spatial patterns in the metropolitan areas of Marseille and Osaka. Asia-Pacific Journal of Regional Science, 4(1), 193-217. Scheer, B. C. (2018). Urban Morphology as a Research Method. Planning Research and Knowledge. Abingdon, UK: Routledge, 167-181. Steadman P (2016). Research in architecture and urban studies at Cambridge in the 1960s and 1970s: what really happened. The Journal of Architecture 21, no. 2 (2016): 291-306 124 124 Communications Communications Communications par Session Communications par Session Simulation, évaluation en géomatique SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Choix de mobilité : une approche SMA Alice Jacquier 1, Carole Adam 1 Univ. Grenoble-Alpes, Laboratoire d’Informatique de Grenoble (LIG) prenom.nom@univ-grenoble-alpes.fr Univ. Grenoble-Alpes, Laboratoire d’Informatique de Grenoble (LIG) prenom.nom@univ-grenoble-alpes.fr RÉSUMÉ. La mobilité a beaucoup changé récemment, et va continuer à évoluer pour faire face à de nouveaux défis, en particulier environnementaux. Le projet SwITCh vise à créer un outil pour aider les planificateurs urbains à concevoir des villes intelligentes et durables. Cet article y contribue en proposant un modèle agent des décisions de la population concernant leur modalité de transport pour leur trajet quotidien domicile- travail. Nous avons implémenté ce modèle avec la plateforme de simulation GAMA, et utilisé ce simulateur pour tester 2 hypothèses. Nos premiers résultats montrent que l’amélioration des infrastructures cyclables conduit à une augmentation du nombre de cyclistes, mais que par contre une simple augmentation du prix de l’essence ne dé- courage pas les automobilistes. Des travaux futurs étudieront donc d’autres stratégies pour augmenter la part des mobilités douces dans les trajets quotidiens. ABSTRACT. Means of mobility have changed a lot recently and will continue to evolve, in order to meet new environmental challenges. The SwITCh project aims to create a tool to help urban planners designing smart and sustainable cities. This paper contri- butes a first step by providing an agent-based model of the population’s decisions re- garding transport modality for their daily home-work commute. We implemented our model with the GAMA simulation platform, and used this simulator to test 2 hypo- theses. Our first results show that better cycling infrastructures result in an increase in the number of cyclists; and that a simple petrol price rise is not enough to discou- rage car users. Future work will thus investigate better strategies to improve the part of sustainable mobility in daily commutes. MOTS-CLÉS : Simulation multi-agents, modèle de décision, mobilité urbaine O S A t b d i l ti d i i d l b bilit MOTS-CLÉS : Simulation multi-agents, modèle de décision, mobilité urbaine KEYWORDS: Agent-based simulation, decision model, urban mobility MOTS-CLÉS : Simulation multi-agents, modèle de décision, mobilité urbaine KEYWORDS: Agent-based simulation, decision model, urban mobility 1. Introduction Nous faisons face à une crise environnementale, et la question de la durabilité de nos pratiques quotidiennes se place de plus en plus au centre de nos pré-occupations. Durant ces dernières décennies, la distance moyenne quotidienne 127 Communications Communications Communications 127 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] parcourue par les français a augmenté considérablement (5 km en moyenne dans les années 50, contre 45 km en moyenne de nos jours (Viard, 2011)), faisant de la mobilité une problématique importante dans cette recherche de durabilité. Nous voulons donc proposer un outil pédagogique à destination du grand public pour leur permettre de comprendre les raisons et l’impact de leurs choix de mobilité. La simulation sociale à base d’agents est très utile pour étudier des phéno- mènes sociaux et proposer de tels outils. Elle permet en particulier de répéter et contrôler exactement les paramètres de scénarios définis, afin d’en explorer les implications, et de tester des stratégies. Dans ce contexte, nous présentons donc un modèle à base d’agents de la population d’une ville (ici Castanet- Tolosan), et de ses décisions concernant le mode de transport domicile-travail. Ce modèle attribue à chaque agent, représentant un habitant, une modalité préférée, en fonction des caractéristiques de l’environnement et de ses attributs sociaux et psychologiques. Il permet ensuite de simuler différents scénarios en faisant varier les paramètres environnementaux comme le prix de l’essence, les infrastructures de la ville, ou la météo du jour. Dans cet article nous testons dans notre simulateur deux interventions pour essayer de faire baisser la part de la voiture. A terme, ce premier modèle servira de base à un jeu sérieux dans lequel les citoyens pourront explorer diverses interventions. La Section 2 introduit un état de l’art pluridisciplinaire des travaux sur le choix modal, en sociologie et en informatique. La Section 3 décrit notre modèle conceptuel basé sur ces travaux. La Section 4 décrit son implémentation avec GAMA ainsi que les premières expérimentations et leurs résultats. Enfin la Section 5 conclut cet article et discute des perspectives d’amélioration. 2. Choix du mode de transport : un état de l’art pluridisciplinaire Le choix modal a été étudié en sociologie sous plusieurs axes : définition de profils d’usagers (Section 2.1), ou analyse des critères de décision (Section 2.2). Divers modèles informatiques en ont aussi été proposés (Section 2.3). [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Approche SMA de la mobilité Approche SMA de la mobilité l’école nécessite d’avoir un siège passager sur son mode de transport) et la force de l’habitude (certains usagers ayant tendance à reproduire la même routine). Rocci (2007) propose 6 autres profils d’usagers : les automobilistes passion- nés en adhésion à l’automobile ; les automobilistes passionnés en opposition à un autre mode ; les multimodaux rationnels, en adhérence à la voiture mais qui utilisent les transports en commun par moment ; les multimodaux en op- position à la voiture, mais contraints de la prendre par moment ; les alternatifs passionnés en adhérence à leur mode ; et les alternatifs en opposition à leur mode. Cette classification aborde de manière moins tranchée la dualité entre voiture et autres modes, et est aussi plus détaillée, allant au-delà des seuls au- tomobilistes. On y retrouve cependant des similitudes, notamment cette idée d’usage contraint de la voiture. Rocci détaille d’ailleurs une autre raison à cet usage contraint : ce qu’elle appelle le “capital mobilité”, ou la capacité à uti- liser un mode de transport (posséder un vélo ou une voiture, savoir lire une carte de métro, avoir le permis...). Cela recouvre aussi la notion d’accessibilité (tous les transports en commun ne sont pas -facilement- accessibles). Enfin elle montre que chaque personne a une vision personnelle des différents modes de transport : les automobilistes convaincus, par exemple, vont avoir tendance à sous-estimer le prix de la voiture, et surestimer le prix des transports. De plus, les transports en commun ont aussi une image associée au stress, agression et dépendance (Kaufmann et al., 2010). Par contre, Rocci ne fournit aucune statistique de répartition de ces profils dans la population fournie. Nous pouvons trouver dans la littérature d’autres études qui ont créé des profils, par exemple en utilisant des modèles de mélange (Briand et al., 2016). Cette étude ne concerne que les transports en commun, et souligne différents profils concernant leur utilisation. Ces modèles sont intéressants sur la métho- dologie, mais pour notre travail nous nous baserons sur les profils cités ci-dessus, moins complexes et qui explicitent mieux les raisons des choix de chaque profil. 2.2. Sociologie: facteurs de décision Il existe une infinité de facteurs que nous pourrions prendre en compte dans notre modèle. Cependant, nous désirons suivre le paradigme KISS ("Keep it simple, stupid", (Axelrod, 1997 ; Adam, Gaudou, 2016)) et concevoir un modèle simple, afin de mieux en comprendre les résultats. Comme il est destiné à servir de base pour un jeu sérieux, nous voulons aborder de manière simple les mécanismes de prise de décision. Nous présentons donc ici une sélection non-exhaustive des principaux facteurs de décision. 2.1. Sociologie : profils d’usagers L’étude Mobil’Air (Chalabaev et al., 2018) présente 4 profils d’usagers de la voiture par opposition aux autres modes de transport : automobilistes ou- verts à tout mode ; automobilistes pro mode individuel (préférant un mode de transport assurant l’indépendance) ; automobilistes à contre coeur (contraints d’utiliser la voiture) ; automobilistes convaincus (aimant conduire et prenant la voiture pour tous leurs trajets). L’avantage de ce projet est de fournir les pour- centages de la population dans chaque catégorie. Par contre, son point négatif est de ne considérer que les automobilistes. Ces profils illustrent le principe de contrainte de mode de déplacement selon son motif (e.g. emmener un enfant à Communications 128 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 2.2.2. Le contexte de l’environnement L’environnement impose des contraintes dans la décision du mode de trans- port (Rocci, 2007). La voiture est utilisée pour des longs trajets, et pour combler une carence en transports en commun (pour les activités nocturnes). La voi- ture est aussi utilisée pour diminuer la charge mentale car elle permet un trajet direct. En effet, la réduction de la charge mentale est pour certains un argu- ment pour ne pas prendre la voiture, mais seulement si le trajet de transport en communs est facile ou direct (i.e un seul métro, tram, bus à prendre, pas de changement à faire). Si le trajet est trop compliqué, alors la préférence va à la voiture. Enfin, c e c hoix d épend a ussi d u t emps t otal d e t rajet, incluant l’accès aux arrêts de transport en commun, et donc de la répartition du réseau sur l’agglomération. Un voyageur qui n’a pas d’arrêt à distance raisonnable de chez lui et/ou de sa destination va aussi préférer la voiture. Ces informations montrent que l’environnement joue un rôle dans la décision du mode de transport. Il nous faudra donc modéliser assez précisément ces aspects de l’environnement : position des arrêts, fréquence des bus, etc. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] sont bouleversées (changement d’emploi du temps, déménagement, naissance d’un enfant, emménagement avec un conjoint, etc..), sont des moments propices pour briser une ancienne habitude ou en intégrer une nouvelle, en ciblant les interventions sur ces populations (Rothman et al., 2015). Enfin, les habitudes peuvent modifier les perceptions (temps, coût, etc). Ainsi un usager habitué de la voiture va surestimer le temps de trajet en transport public et minimiser le temps de trajet en voiture (Rocci, 2007) (Betsch et al., 2001). Il nous semble donc essentiel de prendre en compte l’habitude dans notre modélisation du comportement humain de choix modal. Par ailleurs, l’instru- mentalisation des changements de cycle de vie est une piste intéressante pour les scénarios d’interventions à tester dans notre simulateur. 2.2.1. L’habitude Les habitudes sont au coeur des décisions de mobilité. Elles peuvent se changer dans les moments décisifs des cycles de vie des personnes (Brette et al., 2014). Les changements de contexte, c’est-à-dire quand les autres habitudes 129 Communications 129 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3. Modèle de décision du mode de transport Cette section décrit notre modèle conceptuel de la décision modale des usa- gers : le modèle de l’environnement urbain (Section 3.1) ; et le modèle des agents humains, chacun représentant un usager (Section 3.2). [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Approche SMA de la mobilité Approche SMA de la mobilité critère). Mais cette approche est très complexe et ne laisse pas comprendre ce qu’il se passe, car elle prédit le résultat du processus de décision mais n’ex- plique pas ce processus lui-même. Cette approche ne convient donc pas pour concevoir notre jeu sérieux, qui doit faire comprendre au public le cheminement de pensée des agents simulés pour avoir un impact sur leur propre raisonne- ment. D’autres projets ont d’ailleurs préféré l’approche SMA pour ces mêmes objectifs : comprendre le processus de formation du résultat, et non pas trou- ver le résultat optimal. C’est le cas de Wester (Wester, 2015) qui cherche à comprendre la construction du réseau de transports en commun de Brazzaville. Nous choisissons donc nous aussi une approche SMA. 2.4. Notre proposition Notre but est de créer un jeu sérieux pédagogique à destination du grand public, afin d’expliquer et explorer des scénarios d’aménagement urbain. De plus, le jeu pourrait permettre de mettre en lumière des comportements indési- rables (comme la prise de la voiture, par exemple). L’approche par simulation multi-agents est donc une bonne option. En effet, la modélisation à l’échelle microscopique et individuelle nous offre une vision différente et réaliste du pro- cessus du choix modal. De ce fait, les conséquences des actions menées dans le jeu sont compréhensibles, et explicables à un public non expert. Cependant nous n’avons trouvé aucun modèle à base d’agents décrivant le processus de décision d’un mode de transport, à l’échelle individuelle. Nous décrivons donc dans la section suivante notre propre modèle de choix modal individuel, basé sur les facteurs de choix identifiés dans l’état de l’art ci-dessus. 2.3. Informatique: modèles de décision du mode de transport Il existe déjà des simulateurs de trafic ( par e xemple l e s imulateur SUMO, (Behrisch et al., 2011)). Différents modèles de décision de mode de transport ont aussi été proposés selon différentes approches. D’un côté, le champ de la recherche opérationnelle sur les MCDA (multi-criteria decision aid) propose des méthodes comme Promethée (Behzadian et al., 2010) permettant de trouver une solution optimale. Gandit (Gandit, 2007) propose un modèle de décision de mode de transport (uniquement entre voiture et transports en commun) avec une approche par apprentissage automatique. Il réalise une régression logistique sur la base de trois types de critères : socio-démographiques, structurels, et psychologiques. Sa classification d es c ritères p our chaque i ndividu e st i ntéressante, e t e st ba- sée sur 3 études menées dans le cadre de sa thèse (une pour chaque type de Communications 130 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3.1. L’environnement, une variable à ne pas négliger Éléments géométriques : nous commençons par définir l’environnement dans lequel les usagers évoluent, qui contient les éléments suivants : – Road (route), qui composent les arêtes d’un graphe représentant le réseau de routes, et qui a un attribut "has_bike_lane" qui informe de la présence d’une piste cyclable ou non. – Building : id pour identifier les bâtiments. type le type du bâtiment. sub_area, le lieu du bâtiment. nb_households, le nombre de ménage dans ce bâtiment. size, la taille du bâtiment. 131 Communications 131 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 – Crossroad : type, le type du carrefour. crossing, qui donne des informa- tions sur ce carrefour. Nous utilisons des graphes pour représenter les réseaux de transports (un graphe par type de transport) ainsi que le réseau routier. Les bâtiments sont représentés par des polygones, selon les informations obtenues sur Open Street Map. Nous n’avons pas considéré le relief des trajets, mais pour l’instant uni- quement leur distance. Cependant, nous n’excluons pas la possibilité d’intégrer ultérieurement un attribut "dénivelé" qui permettrait de prendre en compte le relief, grâce à l’intégration d’un modèle numérique de terrain. Dans l’implémen- tation du simulateur, ces éléments pourront être importés depuis Open Street Map pour n’importe quelle ville. Dans nos résultats (cf Section 4) nous avons choisi par exemple la ville de Castanet-Tolosan. Éléments contextuels : Dans l’environnement, nous prenons aussi en compte des éléments contextuels pertinents dans le choix modal (météo, infor- mations sur le réseau de transport public, prix de l’essence, infrastructures pour les cyclistes). Cependant, nous avons dû simplifier dans un premier temps beau- coup de ces éléments. Ainsi notre modèle des transports en commun comporte uniquement des informations de confort (affluence) et temporelles (fréquence) mais pas encore d’informations spatiales (distance à l’arrêt). Pour la météo, nous faisons simplement un tirage aléatoire entre 3 types de temps (ensoleillé, nuageux, pluvieux), pondéré par les informations relatives à la ville modélisée (temps d’ensoleillement, informations pluviométriques). Cela est suffisant pour expérimenter des scénarios réalistes (moins de cyclistes par temps pluvieux), et pourra être complexifié dans le futur si nécessaire. Tous ces attributs jouent un rôle dans le choix du mode de transport, et sont pris en compte dans le modèle de décision décrit dans la section 3.2. 3.2. Agents usagers Dans notre modèle, chaque usager est représenté par un agent, doté d’un cer- tain nombre d’attributs : un âge, un genre, une catégorie socio-professionnelle, un lieu de travail, un lieu de résidence, un niveau de sportivité et trois listes d’agents représentant respectivement sa famille, ses amis, et ses collègues. Chaque agent a également une liste contenant les notes associées à chaque mode de transport. Le calcul de ces notes est expliqué dans le paragraphe suivant 3.2. Ils ont de plus une liste de critères auxquels ils apportent plus ou moins d’im- portance, voir 3.2. L’habitude est représentée par un coefficient qui augmente la note du mode de transport préféré courant. Nous avons aussi un attribut qui permet de déterminer si l’agent possède un vélo, ou encore une voiture. Certains de ces attributs ne sont pas encore utilisés dans le modèle, mais sont déjà implémentés. Les listes représentant la famille, les amis, et les collègues par exemple serviront dans le futur à modéliser la pression sociale, ou encore la diffusion de pratiques parmi les pairs. La catégorie socio-professionnelle n’est 132 Communications 132 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Approche SMA de la mob Figure 1. Schéma récapitulatif du processus de prise de décision Approche SMA de la mobilité Figure 1. Schéma récapitulatif du processus de prise de décision pas non plus utilisée ; elle le sera ultérieurement pour déterminer la possession ou non d’une voiture. Calcul des priorités des modalités : Chaque agent calcule ensuite sa note de préférence pour chaque mode de transport, en fonction des attributs ci-dessus. En se basant sur les études de (Rocci, 2007) et (Kaufmann et al., 2010), nous avons pu définir 6 critères : prix, temps, confort, sécurité, simplicité, écologie. Chaque agent accorde une importance plus ou moins grande à chacun de ces critères. En fonction de l’environnement de l’agent et de l’importance donnée à chaque critère, nous calculons alors une note individuelle de priorité pi de chaque mode de transport i, pour chaque agent, avec la formule suivante : pi = sumc∈critères(vali,c × prefc) + habit_coeff × pi nombre de critères + 1 Vélo Bus Voiture Marche Confort Distance sportivité météo Affluence 1 Distance sportivité météo Prix 0 Prix abonnement Prix essence distance 0 Ecologie 1 0.8 0 1 Simplicité Distance #correspondances (pour l’instant ≈0.7) 1 Distance Sécurité Ratio routes avec pistes cyclables Genre, heure du jour Niveau de trafic Genre heure du jour Temps Distance vitesse moyenne vélo (pondérée par sportivité) Distance Vitesse moyenne bus (pondéré par trafic) Distance Vitesse moyenne voiture (pondéré par trafic) Distance, vitesse moyenne marche selon l’ancienneté de la valeur d’une priorité (ancienneté notée n), de manière logarithmique : coeff_habit = log n. La croissance de ce coefficient est donc de plus en plus faible, à mesure que n augmente à chaque pas de temps. Dès que l’utilisateur change de mode de transport préféré, n est réinitialisé à 0. selon l’ancienneté de la valeur d’une priorité (ancienneté notée n), de manière logarithmique : coeff_habit = log n. La croissance de ce coefficient est donc de plus en plus faible, à mesure que n augmente à chaque pas de temps. Dès que l’utilisateur change de mode de transport préféré, n est réinitialisé à 0. 1. https://www.comses.net/codebases/ pi = sumc∈critères(vali,c × prefc) + habit_coeff × pi nombre de critères + 1 La valeur vali,c, comprise entre 0 et 1, correspond à une évaluation par l’agent de la valeur du critère c pour le mode de transport i, selon l’état de l’environnement et les caractéristiques de l’agent. Par exemple le vélo pourrait obtenir une note de 1 sur le critère écologie, mais sa note de confort dépendra de la distance à parcourir et du niveau de sportivité de l’agent, et sa note de sécurité dépendra des infrastructures cyclables de la ville. La note de la voiture sur le critère de prix dépendra entre autres du prix de l’essence, alors la note de simplicité dépendra de la possession ou non d’un véhicule, etc. Ce calcul est résumé au Tableau 1. Ce calcul est par ailleurs pondéré par un coefficient d’habitude coeff_habit, expliqué dans le paragraphe suivant. Prise en compte de l’habitude : Lorsqu’un agent décide quel moyen de transport il préfère utiliser, le coefficient d’habitude va pondérer de manière plus ou moins forte la valeur de la priorité. Cela permet de modéliser le fait qu’un choix ”routinier” n’est plus réévalué quotidiennement, au contraire il faut un changement majeur pour remettre en question un choix. Ainsi, si cela fait plusieurs années que l’agent utilise la voiture, il va de moins en moins remettre en question son choix ou analyser les autres moyens de transports autour de lui, et il lui sera plus difficile de réaliser quand un autre mode de transport correspond mieux à ses critères. Ce coefficient d’habitude est calculé 133 Communications 133 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Tableau 1. Facteurs entrant en compte dans le calcul de la note de chaque mode de transport sur chaque critère. Si la valeur est fixe, elle est indiquée directement dans la case. Tableau 1. Facteurs entrant en compte dans le calcul de la note de chaque mode de transport sur chaque critère. Si la valeur est fixe, elle est indiquée directement dans la case. directement dans la case. 4.1. Implémentation et méthodologie La plateforme GAMA. GAMA (Taillandier et al., 2019) est une plate- forme de simulation multi-agent open source, permettant d’implémenter des modèles agents de haut-niveau, avec un langage simple, accessible aux non- informaticiens. GAMA fournit différentes architectures d’agents (automate, réflexes, architecture cognitive, émotions...). Elle permet aussi d’inclure des données de divers types, y compris des données issues de Systèmes d’Informa- tion Géographique. Le code de notre simulation sera partagé sur OpenABM 1 ultérieurement, une fois stable. Création de l’environnement. Nous avons récupéré les données relatives à la ville de Castanet-Tolosan via Open Street Map (graphe des routes, infor- mations relatives aux bâtiments, etc) afin de générer la carte utilisée dans notre modèle de déplacement. Nous fixons le ratio de pistes cyclables à 0.5, et une météo ensoleillée. Nous fixons également : le prix de l’essence à 1,5e au litre ; le prix mensuel de l’abonnement de transports en commun à 30e ; la capacité des bus à 50 personnes, et leur fréquence de passage toutes les 7 minutes. Création de la population. Nous avons lancé la simulation avec une population synthétique de 10 000 habitants. Il existe déjà des travaux sur la Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Approche SMA de la mobilité Approche SMA de la mobilité génération de populations synthétiques réalistes (par exemple le projet Genstar (Chapuis et al., 2019)), qui seront réutilisés ultérieurement pour améliorer le réalisme de notre simulateur. Pour l’instant, dans un souci de simplification, nous considérons qu’un agent sur deux possède une voiture, un agent sur deux possède un vélo, et l’initialisation des autres attributs (lieu de résidence et de travail, priorité de chacun des critères) se fait aléatoirement. Avec ces informa- tions, l’agent choisit un mode de transport (voiture, vélo, bus ou marche) pour son trajet domicile-travail. Méthodologie. Nous mesurons en sortie la distribution (statique) des agents par mode de transport à une date donnée, sur 2 scénarios : 1) en variant le ratio de pistes cyclables ; et 2) en variant le prix de l’essence (tous les autres paramètres fixés). Les autres paramètres ne changent pas (météo ensoleillée, initialisation des agents avec la même graine). Chaque situation a été simulée 10 fois et nous avons fait la moyenne des distributions de mode de transport résultantes, pour lisser la part d’aléatoire. Scénario 1 : part de cyclistes selon les infrastructures. Nous voulons tester l’hypothèse que le taux d’infrastructures cyclables influence le nombre d’agents choisissant le vélo : plus il y a de pistes cyclables, plus il y a de cy- clistes, et vice-versa. Pour cela, nous avons simulé deux situations : 1) aucune infrastructure pour le vélo, et 2) toutes les routes équipées d’infrastructures cyclables. La figure 2 illustre la distribution moyenne des usagers sur les dif- férents modes de transport dans ces 2 situations. Ces résultats valident notre hypothèse : la quantité d’infrastructures influe bien sur le taux de cyclistes. Figure 2. Distribution selon ratio de pistes cyclables Figure 2. Distribution selon ratio de pistes cyclables Scénario 2 : part d’automobilistes selon le coût de l’essence. La deuxième hypothèse que nous émettons est que le prix de l’essence influence le nombre d’automobilistes : plus l’essence est chère, moins il y a d’automobilistes. Nous testons pour cela 2 situations : 1) prix de l’essence à 0.2 e par litre ; et 2) prix de l’essence augmenté à 5 euros par Litre. La figure 3 montre les résultats de ces deux simulations. Nous constatons un résultat assez mitigé : le taux d’automobilistes ne change pas, peu importe le prix de l’essence. Communications 134 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Ce résultat semble incorrect, et pourtant il reste crédible. La voiture coûte cher, 135 Communications 135 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] et les personnes ayant pour critère important le prix ne choisiraient pas la voiture dans un premier temps ; donc lorsque le prix augmente, cela a peu d’incidence. Cela correspond aussi aux observations dans le monde réel : le prix de l’essence augmente régulièrement depuis des années, pourtant la part de la voiture ne diminue pas ; c’est d’ailleurs encore plus le cas actuellement avec la crise sanitaire qui décourage le recours aux transports en commun. Il faut donc chercher d’autres moyens pour faire augmenter la part des mobilités douces en milieu urbain. Figure 3. Distribution selon prix de l’essence. Figure 3. Distribution selon prix de l’essence. 4.2. Discussion La principale limitation de ce modèle est sa validation difficile. Les résultats sont crédibles mais il est difficile de vérifier le raisonnement réel des habitants. Par ailleurs une fois le modèle plus stable, il sera intéressant d’utiliser OpenMole pour en faire une analyse de sensibilité. Ce modèle est aussi très simplifié, du point de vue de l’environnement comme des agents. Dans le futur, nous envisageons donc de donner plus de réa- lisme à l’environnement (météo basée sur des relevés pluviométriques, comme fait dans Vigiflood (Adam, Andonoff, 2019) ; pollution de l’air, comme dans HoanKiemAir (Duc et al., 2020) ; prise en compte du dénivelé des trajets grâce à un Modèle Numérique de Terrain...) et de tester d’autres scénarios (ajout d’un mode de transport, gratuité des transports publics, piétonnisation du centre...). Des travaux sont aussi en cours pour générer et intégrer une population synthétique réaliste, incluant par exemple la provenance sociale. Pour l’instant nous avons supposé que les habitants vivent et travaillent dans la même ville, ce qui augmente artificiellement la part de la marche dans les déplacements. Les décisions des agents sont aussi basées uniquement sur des critères ration- nels pour l’instant, mais nous allons maintenant étudier d’autres facteurs plus subjectifs : influence de l’entourage ou de leaders d’opinion (par analogie avec la diffusion d’innovation, largement documentée dans la littérature (Vodopivec Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Approche SMA de la mobilité Approche SMA de la mobilité et al., 2021)) ; émotions (Bourgais et al., 2018) ; biais cognitifs (Arnaud et al., 2017) ; ou encore habitude, dont le rôle est central dans l’inertie des pratiques modales (Brette et al., 2014). 5. Conclusion Nous avons présenté dans cet article un modèle du choix du mode de dé- placement domicile-travail. Notre modèle prend en entrée des caractéristiques liées à l’environnement (météo, réseau de transports en commun, infrastruc- tures cyclables, prix de l’essence), et d’autres liées aux usagers (genre, âge, préférences, lieu de résidence et de travail, capital mobilité). En sortie, le mo- dèle donne pour chaque agent un mode de transport préféré. Nos résultats ont montré que l’amélioration des infrastructures cyclables augmentait la part des cyclistes (mais plutôt aux dépends du bus), et que l’augmentation du prix de l’essence ne décourageait pas les automobilistes. Malgré certaines limitations, ces résultats sont crédibles. Nous prévoyons dans le futur d’améliorer le réa- lisme, que ce soit de l’environnement, ou bien du processus cognitif des agents. A terme, ce modèle sera utilisé dans un jeu sérieux afin d’impliquer et de sensi- biliser les citoyens de manière ludique à l’impact de leurs choix individuels sur la collectivité. Ils pourront en particulier participer à l’élaboration de nouveaux scénarios. En effet la transition écologique ne peut se faire sans la participation active de tous. Remerciements. Ce travail résulte d’un stage de M2R et se poursuit en thèse. Il est financé dans le cadre du projet SwITCh, qui a bénéficié d’une aide de l’Agence Nationale de la Recherche portant la référence ANR-19-CE22-0003. Communications 136 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Bibliographie Adam C., AndonoffE. (2019). Vigiflood: a serious game for understanding the chal- lenges of risk communication. In International conference on information systems for crisis response and management. Valencia, Espagne, ISCRAM association. Adam C., Gaudou B. (2016). Bdi agents in social simulations: a survey. Knowledge Engineering Review, vol. 31, no 3, p. 207–238. Arnaud M., Adam C., Dugdale J. (2017, May). The role of cognitive biases in reactions to bushfires. In Iscram. Albi, ISCRAM. Axelrod R. (1997). The complexity of cooperation: Agent-based models of competition and collaboration (vol. 3). Princeton university press. Behrisch M., Bieker L., Erdmann J., Krajzewicz D. (2011). Sumo–simulation of urban mobility: an overview. In Third international conference on advances in system simulation (simul). Barcelona, Spain, IARIA. Behzadian M., Kazemzadeh R. B., Albadvi A., Aghdasi M. (2010). Promethee: A comprehensive literature review on methodologies and applications. European journal of Operational research, vol. 200, no 1, p. 198–215. 137 Communications 137 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Betsch T., Haberstroh S., Glöckner A., Haar T., Fiedler K. (2001). The effects of routine strength on adaptation and information search in recurrent decision making. Organizational behavior and human decision processes, vol. 84, no 1. Bourgais M., Taillandier P., Vercouter L., Adam C. (2018, March). Emotion modeling in social simulation: a survey. JASSS, vol. 21, no 2. Brette O., Buhler T., Lazaric N., Marechal K. (2014). Reconsidering the nature and effects of habits in urban transportation behavior. Journal of Institutional Economics, vol. 10, no 3, p. 399–426. Briand A. S., Come E., El Mahrsi M. K., Oukhellou L. (2016). Classification à base de modèle de mélange pour l’identification de profils temporels types d’usagers de transport public. In Fr. int. conf. on data science. Maroc, AAFD & SFC. Chalabaev A., Chardonnel S., Bouscasse H., Duché S., Isoard-Gatheur S., Mathy S. et al. (2018). Les apports de la littérature et des analyses préalables pour définir une intervention visant un report modal durable. In MSFS. Tours, France, MSFS. Chapuis K., Thiriot S., Taillandier P., Gaudou B., Amblard F. (2019). Gen*: an integrated tool for realistic agent population synthesis. In SSC. Germany, ESSA. Duc P. M., Chapuis K., Drogoul A., Gaudou B., Grignard A., Marilleau N. et al. (2020). Hoankiemair: simulating impacts of urban management practices on traffic and air pollution using a tangible agent-based model. In Int. conf. on computing and communication technologies (rivf), p. 1–7. IEEE. Gandit M. (2007). Bibliographie Déterminants psychosociaux du changement de comportement dans le choix du mode de transport : le cas de l’intermodalité. Thèse de doctorat non publiée, Univ. Grenoble. (Thèse de doctorat) Kaufmann V., Tabaka K., Guidez J.-M., Louvet N. (2010). Et si les français n’avaient plus seulement une voiture dans la tête?. Rapport technique. France, C. d’études sur les réseaux, les transports, l’urbanisme et les constructions publiques. Rocci A. (2007). De l’automobilité à la multimodalité? analyse sociologique des freins et leviers au changement de comportements vers une réduction de l’usage de la voiture. Thèse de doctorat non publiée, Sociologie, Univ. R. Descartes - Paris V. Rothman A. J., Gollwitzer P. M., Grant A. M., Neal D. T., Sheeran P., Wood W. (2015). Hale and hearty policies: How psychological science can create and main- tain healthy habits. Perspectives on Psychological Science, vol. 10, no 6, p. 701–5. Taillandier P., Gaudou B., Grignard A., Huynh Q.-N., Marilleau N., Caillou P. et al. (2019). Building, composing and experimenting complex spatial models with the gama platform. GeoInformatica, vol. 23, no 2, p. 299–322. Viard J. (2011). Éloge de la mobilité: essai sur le capital temps libre et la valeur travail. France, Editions de l’Aube. Vodopivec N., Adam C., Chanteau J.-P. (2021). Modeling opinion leader’s role in the diffusion of innovation. arXiv preprint arXiv:2101.11260. Wester L. (2015). Modélisation multi-agents de transports collectifs artisanaux : structures émergentes et stratégies individuelles. In SAGEO. Tunisia. 138 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Cet article s’appuie sur une thèse de doctorat, et plus spécifiquement sur le matériel conceptuel, méthodologique et technique qui y est développé dans les chapitres 3 et 4 (Cura, 2020a, pp. 107–180, 181–262). 1.1 Contexte et cas d’étude La proposition méthodologique ici développée se veut générique, mais a été développée dans le cadre d’un projet interdisciplinaire de modélisation de transformations des structures spatiales sur le temps long, l’ANR TransMon- Dyn. Au sein de ce projet, un groupe composé de géographes-modélisateurs, d’archéologues et d’historiens a conçu et développé un modèle de simulation, SimFeodal, cherchant à reproduire les dynamiques de polarisation, de hié- rarchisation et de fixation d e l ’habitat r ural d e l ’Europe d u N ord-Ouest au tournant du 1er millénaire de notre ère, en modélisant les mécanismes de mi- grations et de concentration de foyers paysans. Le modèle et sa démarche de construction ont fait l’œuvre de publications spécifiques ( Cura e t a l., 2017 ; Tannier et al., 2020), et nous en donnerons donc ici un descriptif très suc- cinct. SimFeodal est un modèle de simulation à base d’agents stochastique, à visée exploratoire, qui suit l’approche « KIDS » formalisée par Edmonds et Moss (2005). Il se classe ainsi dans la catégorie des modèles descriptifs, avec de nombreux types d’agents (foyers paysans, seigneurs, églises paroissiales, pôles d’attraction, etc.) dont les interactions multiscalaires font émerger les dyna- miques étudiées à l’aide de mécanismes complexes et nombreux sur une longue durée simulée de 400 ans. Pour une évaluation visuelle des modèles de simulation, au-delà de la face validation Robin Cura UMR Géographie-cités, CNRS robin.cura@parisgeo.cnrs.fr RÉSUMÉ. Nous proposons une approche méthodologique dédiée à l’évaluation de mo- dèles de simulation stochastiques descriptifs caractérisés par une faible possibilité de recours à des données empiriques. En l’absence de données structurées de contrôle – ou quand celles-ci sont trop peu nombreuses, incertaines ou lacunaires –, l’évaluation peut être réalisée d’après les connaissances expertes, de manière qualitative. Pour que ce processus soit mené de manière rigoureuse, robuste et reproductible, il ne doit pas être uniquement basé sur des intuitions ou premières impressions. L’évaluation peut prendre la forme d’une comparaison systématique entre les sorties du modèle et les hypothèses thématiques sous-jacentes à la modélisation, tirées des connaissances ex- pertes. Cette comparaison est menée à l’aide d’une méthode d’« évaluation visuelle » que nous définissons et appliquons ici. L’approche méthodologique proposée doit ainsi permettre et faciliter l’évaluation de modèles de simulations par le biais de connais- sances expertes explicitées, en s’appuyant sur l’utilisation d’un environnement d’ex- ploration interactive des données de sortie de simulation. ABSTRACT. In this talk, we propose a new methodological framework, the “visual evaluation”, that enables or facilitates the evaluation of complex and stochastic si- mulation models that mostly rely on expert knowledge rather than empirical data for their hypotheses. This method has been developped for the modelling of space-time processes and relies on the thorough use of qualitative visual comparison between the model’s outputs and the underlying empirical knowledge. Visual evaluation is based on a remediation of face validation for massive simulation datasets, and thus requires an interactive exploratory data analysis platform dedicated to simulation outputs. MOTS-CLÉS : Simulation multi-agents, méthodes d’évaluation, évaluation visuelle, analyse interactive de données, modélisation en interdisciplinarité KEYWORDS: Agent-based modelling, evaluation methods, visual evaluation, face va- lidation, Exploratory Data Analysis, interdisciplinary modelling 139 139 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1 Introduction Pour comprendre et modéliser une structure spatiale modelée par l’héritage de ses états antérieurs, il est nécessaire d’en comprendre l’émergence, souvent bien antérieure. Comme l’écrit Durand-Dastès (1999, p. 23), l’explication d’un « objet spatial » contemporain n’est en effet possible qu’en montrant « à la fois comment il est né et comment il a persisté ». La modélisation de proces- sus spatiaux génériques en géographie implique donc de comprendre les ori- gines et raisons de leur apparition, et donc de faire appel à des connaissances en dehors du champ strict de la modélisation en géographie. Si, dans cette discipline, la modélisation de dynamiques spatiales à l’aide de modèles de si- mulation multi-agents est désormais bien ancrée, la « culture modélisatrice » est moins présente dans les champs disciplinaires historiques et archéologiques La conception d’un modèle de simulation spatial requiert alors d’une part la capacité d’intégrer les connaissances expertes nécessaires via un processus de co-construction (Cura, 2020a, pp. 26–29, 340–346) 1 et d’autre part, de manière inhérente à une démarche de modélisation, la capacité de soumettre le modèle conçu à une évaluation rigoureuse. 2.1 Évaluer un modèle Depuis les travaux précurseurs en simulation informatique (Naylor, Fin- ger, 1967 ; Hermann, 1967), la plupart des chercheurs mettent en avant qu’un modèle de simulation non évalué n’a ni utilité – pour Naylor et Finger (1967) notamment –, ni validité. Sans caricaturer ces écrits, on peut noter que tous can- tonnent les modèles non évalués à des « jeux » ou encore, pour les plus modérés, à des outils pédagogiques. Nous ambitionnons de rendre possible l’évaluation de modèles pour lesquelles les méthodes plus classiques sont soit inadaptées, soit insatisfaisantes. Pour une évaluation visuelle Pour une évaluation visuelle thode proposée est particulièrement pensée pour l’évaluation de modèles qui s’appuient avant tout sur des connaissances à dire d’experts plus que sur des données structurées. Cette méthode d’évaluation, que nous nommons « évalua- tion visuelle », constitue une extension d’une étape classique de l’évaluation de modèle : la face validation, c’est-à-dire l’évaluation de la plausibilité des sorties d’un modèle. Cette méthode d’évaluation est fondée sur une analyse visuelle et subjective de la plausibilité du modèle, en demandant à l’évaluateur de se conforter à des intuitions sur le comportement attendu. L’utilisation de cette méthode est souvent considérée comme une mesure préalable relevant du bon sens plus que de l’évaluation strictement dite. L’évaluation visuelle étend le processus de face validation en systématisant l’analyse des sorties du modèle de manière qualitative. Il s’agit ainsi de définir en amont une grille d’analyse exhaustive, couvrant tous les aspects du modèle. C’est au prisme d’une compa- raison systématique entre cette grille de critères attendus et des « indicateurs de sortie de simulation » (indicateurs visuels, quantitatifs ou qualitatifs) que le modèle sera évalué. Pour permettre à un public d’experts diversifié de mener cette évaluation visuelle directement, sans faire appel à un expert modélisateur ou à un analyste de données, nous proposons un outil interactif d’analyse vi- suelle et de comparaison des simulations effectuées par le biais de leurs données en sortie. 1.2 Objectif Nous proposons ici une méthode d’évaluation dédiée à des modèles de si- mulation à base d’agents co-construits selon un paradigme descriptif. La mé- Communications Communications 140 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Pour une évaluation visuelle Pour une évaluation visuelle 2. « Face validity can be seen as the result of face validation. Under this paradigm I want to subsume all methods that rely on natural human intelligence. Examples are structured walk-through, expert assessments of descriptions, animations or results. Thus, face validity shows that processes and outcomes are reasonable and plausible within the frame of theoretic basis and implicit knowledge of system experts or stake-holder », Klügl (2008, pp. 39–41). 2.1.2 Quelle évaluation pour SimFeodal ? 2.1.2 Quelle évaluation pour SimFeodal ? Toutes les méthodes ne sont pas pour autant applicables à tous les modèles. Par exemple, les méthodes formelles sont souvent peu adaptées à l’évaluation de modèle de simulation à base d’agents. On emploie en effet généralement ces modèles parce qu’ils encouragent une approche anthropomorphique, plus aisément compréhensible et requérant moins de connaissances mathématiques que d’autres approches, mais aussi car il est extrêmement difficile d’exprimer des systèmes dotés de multiples interactions, qui plus est multi-scalaires, sous forme de réseaux d’équations. Les méthodes statiques ne sont aussi que partiel- lement applicables, et renvoient davantage à la validation interne des modèles, via une vérification logique des interactions entre les mécanismes des modèles sur un plan algorithmique. Pour la validation externe, le plus classique est de faire appel aux méthodes « dynamiques » au sein d’une étape communé- ment nommée validation statistique. Il s’agit de confronter quantitativement les données produites par le modèle aux données empiriques qu’elles cherchent à reproduire, c’est-à-dire à s’assurer de la qualité de l’ajustement des données simulées. Ces méthodes dynamiques s’appuient donc intrinsèquement sur la présence de données empiriques directement comparables avec les sorties du modèle, chose commune dans la modélisation de phénomènes contemporains ou récents. Dans le cas de SimFeodal, et plus généralement des modèles de processus anciens et inscrits dans la longue durée, les méthodes dynamiques sont difficile- ment mobilisables. Les données empiriques disponibles sont en effet des traces, incertaines et peu précises de par leur ancienneté, et extrêmement incomplètes (absentes et lacunaires) de par la durée de 400 ans modélisée – d’après la typo- logie de De Runz (2008). SimFeodal est fondé avant tout sur des hypothèses thématiques et des connaissances expertes, et les méthodes dynamiques ne sont alors pas exploitables. Dans cette situation, les théoriciens de l’évaluation re- commandent, à défaut, de tout de même faire appel aux méthodes informelles (Petty, 2010), telle que la « face validation » qui est la plus couramment mo- bilisée. 2.1.1 Différentes approches d’évaluation Le terme d’évaluation en lui-même porte une certaine polysémie et peut être désambiguïsé, en reprenant les termes de Amblard et al. (2006), en « validation interne » et « validation externe ». Dans le premier cas, il s’agit de s’assurer de la bonne conception du modèle – « building the model right » (Balci, 1994, p. 123) – et dans le second, de montrer que le modèle est adapté à ce qu’il cherche à représenter – « building the right model » (Balci, 1994, p. 121). Ces deux composantes permettent de définir une procédure générale d’évaluation qui peut être résumée entre des cycles d’allers-retours et d’itérations entre les hypothèses du modèle, la validation interne de l’implémentation qui en est faite, et la validation externe des sorties produites (Klügl, 2008). Pour chacune de ces composantes, de nombreuses méthodes existent, sont généralement préconisée dans l’une ou l’autre des étapes de l’évaluation, et peuvent être classées en 141 Communications 141 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] différentes catégories, selon qu’elles utilisent une approche formelle, statique, dynamique ou encore informelle (Balci, 1998). 2.2.1 Limites de la face validation et propositions de résolution. C’est toutefois une méthode assez critiquée par les théoriciens de l’évaluation (Hermann, 1967 ; Balci, 1994), qui considèrent que la face validation ne peut qu’être une étape préalable à des méthodes d’évaluation plus quantitatives et formelles. Parmi ces auteurs, Hermann (1967, p. 222) en pointe plusieurs limites – à partir desquelles nous construisons notre proposition d’évaluation visuelle – que l’on peut résumer ainsi. Un manque de connaissance experte. Pour Hermann, la face validation a le défaut d’être trop accessible et de de pouvoir être menée par n’importe qui, y compris par des gens qui n’auraient qu’une expérience limitée des phénomènes modélisés. Cette remarque est généralisable, et s’applique d’ailleurs quelle que soit la méthode d’évaluation : une expertise ne vaut que par la qualité de l’ex- pert impliqué. Hermann considère ainsi que c’est au modélisateur uniquement de mener cette phase d’évaluation. Il se positionne en cela dans une logique de séparation forte entre les acteurs de l’élaboration d’un modèle : les modé- lisateurs conçoivent et évaluent le modèle seuls, et les thématiciens n’en sont qu’utilisateurs. L’évaluation visuelle est conçue en opposition à ce principe, et ne peut être pensée sans une collaboration étroite entre les porteurs de connais- sances thématiques expertes sur le phénomène modélisé et les porteurs d’une connaissance méthodologique experte sur les manières de le modéliser. Invraisemblance de certains comportements. Hermann met ensuite en avant le fait que dans un modèle, tous les mécanismes n’ont pas vocation à être vraisemblables et qu’une recherche de plausibilité de tels mécanismes fausserait l’analyse du modèle. Cela revient à ignorer l’importance du dialogue entre modélisateur et évaluateur. Si le modélisateur connaît les « substitutions » opérées dans le modèle, il pourra se garder de chercher à juger de leur vrai- semblance. À l’inverse, un expert thématicien pourrait être étonné par certains comportements microscopiques si et seulement si les correspondances entre élé- ments du modèle et éléments du système modélisé n’étaient pas explicitées. Plus généralement, cette limite est présente s’il n’y a pas d’explicitation des « substitutions » implémentées, c’est-à-dire de la correspondance entre le do- maine conceptuel et le domaine du modèle implémenté : le niveau attendu de plausibilité individuelle de chaque comportement doit être décidé pendant la construction du modèle. 2.2 De la face validation à l’évaluation visuelle La face validity a été définie dès les débuts de la simulation comme une « im- pression initiale du réalisme d’un [modèle] » (Hermann, 1967, p. 221). Klügl (2008) en donne une définition plus explicite 2 et dresse une typologie des temps Communications 142 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Pour une évaluation visuelle Pour une évaluation visuelle et modes d’intervention de la face validation, c’est-à-dire du processus d’éva- luation permettant de juger de la face validity d’un modèle. Cette méthode a l’avantage de pouvoir être mobilisée dans toutes les étapes du cycle d’évalua- tion (Balci, 1998) et de systématiquement mettre en avant les connaissances expertes des modélisateurs et thématiciens sur l’objet modélisé. 2.2.2 Définition et pré-requis de l’évaluation visuelle C’est à partir de ces remarques que l’on peut proposer une définition de l’évaluation visuelle, qui consiste alors à évaluer la plausibilité du comporte- ment d’un modèle à partir des données qu’il produit. Cette évaluation s’appuie sur une comparaison systématique menée entre 1) les sorties du modèle et 2) les connaissances expertes sur les processus modélisés. Les critères de cette compa- raison sont définis en amont de l’évaluation, et doivent expliciter pour chaque sortie analysée la ou les formes qu’elle devrait prendre. Ces éléments, les critères d’évaluation, ne peuvent être formulés par n’importe qui : si le modélisateur autant qu’un expert externe peuvent les spécifier, i l c onvient d e s ’assurer de l’expertise – thématique et de la connaissance des processus tels que modélisés – de l’évaluateur. L’ensemble des critères d’évaluation forment alors une grille d’analyse qualitative, et la visualisation des différentes sorties du modèle per- met de la compléter. Il devient alors possible d’apprécier globalement l’écart entre le modèle et les processus empiriques qu’il représente, sans chercher pour autant à quantifier ni à mesurer cet écart. 2.2.1 Limites de la face validation et propositions de résolution. L’explicitation des choix de modélisation, et l’assu- rance que leurs enjeux et effets sont bien compris de tous les membres d’un projet de modélisation, sont donc des préalables indispensables à une évalua- tion visuelle. 143 143 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Besoin d’explicitation préalable des objectifs. Hermann note enfin que la face validation ne peut constituer une méthode d’évaluation adaptée si l’on ne spécifie pas, en amont, les critères qu’elle doit s’attacher à examiner. Le risque est alors d’introduire encore plus de subjectivité dans cette analyse, et en particulier de briser la capacité de reproductibilité ou de justification d’une éva- luation. Quand un modèle résulte d’un travail collaboratif les évaluations d’un même résultat peuvent en effet varier, et doivent alors être explicitées autant que possible, ne serait-ce que pour se doter d’un outil de discussion et de débat commun. Il apparaît alors primordial de définir e n a mont d e l a modélisation les critères d’évaluation. Cela ne limite aucunement la nécessaire subjectivité et complémentarité des évaluateurs experts, mais permet au contraire d’ins- crire leurs discours dans un référentiel de comparabilité. Dès lors, en matière d’évaluation visuelle, la mise au point d’une grille d’analyse systématique nous semble indispensable. Celle-ci doit expliciter, en amont des analyses, l’ensemble des éléments qui seront analysés dans le modèle, les « substitutions » dans le domaine empirique auxquelles ils ont trait, et dès lors, les comportements at- tendus pour chacun. 2.2.3 Composition de l’évaluation visuelle Pour établir la correspondance entre le domaine empirique et celui du mo- dèle, nous proposons une dichotomie simple en systématisant l’usage d’un vo- cabulaire souvent employé de manière plurielle. Nous caractérisons ces éléments en deux grands ensembles : les indices empiriques et les indicateurs de sortie. Afin d’évaluer la capacité du modèle à reproduire un phénomène observé, il est nécessaire de disposer dans le domaine empirique de « points de repère », qui doivent relever de l’ensemble des échelles spatiales et temporelles que l’on choisit de mettre en scène. Chacun doit pouvoir être estimé, c’est-à-dire être en capacité d’être reproduits et comparables avec d’autres mesures. Nous les qua- Communications 144 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Pour une évaluation visuelle lifions d’« indices empiriques » et les regroupons en deux catégories basées sur la précision avec laquelle ils peuvent être décrits : les « ordres de grandeur », mono-dimensionnels et d’une certitude variable, et les « formes stylisées », qui reposent sur l’agrégation ou la composition de plusieurs ordres de grandeur. Dans le modèle de simulation, il est nécessaire de définir des « indicateurs de sortie », c’est-à-dire des variables informatiques produites par le modèle que l’on pourra ensuite comparer aux « indices empiriques » définis. Comme pour les indices empiriques, les indicateurs de sortie de simulation peuvent être distingués en des formes numériques simples (des scalaires), et en des plus complexes, les indicateurs multi-dimensionnels, qui peuvent être mis en correspondance avec les « formes stylisées » du domaine empirique. On peut illustrer cette typologie et sa mise en correspondance avec un exemple tiré de SimFeodal . Dans le domaine empirique, on sait que la part de l’habitat rural dispersé a diminué au cours de la période étudiée, depuis environ 95% en 800 jusqu’à environ 20% en 1200. Ces deux valeurs sont des indices empiriques, de type « ordre de grandeur ». Dans le modèle de simulation, un indicateur de sortie comparable est calculé et correspond au taux de foyers paysans isolés (fig. 2, C). Le taux de foyers paysans isolés en début de simulation est un indicateur de sortie scalaire, de même que celui en fin de simulation. 2.2.3 Composition de l’évaluation visuelle Dans le domaine empirique, la part de l’habitat dispersé a diminué régulièrement au cours du temps (forme stylisée), et on évaluera donc la capacité du modèle à reproduire un processus de concentration semblable. La courbe d’évolution du taux de foyers paysans isolés au cours de la simulation – constituée d’une série d’indicateurs de sortie « scalaires » – constitue alors, dans le domaine du modèle, un indicateur de sortie multi-dimensionnel que l’on pourra analyser : si la courbe part de 95% en début de simulation pour atteindre 20% en fin de simulation, avec une diminution régulière, alors l’indicateur de sortie multi- dimensionnel sera en adéquation avec l’indice empirique de type forme stylisée « évolution de la part de l’habitat rural dispersé ». En définissant en amont de la modélisation un ensemble d’indices empiriques, d’indicateurs de sortie correspondants ainsi que des propriétés attendues des ces derniers, on constitue alors une « grille d’analyse » qui permettra de mener l’évaluation visuelle du modèle. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] lages, etc.), l’évolution du maillage des paroisses ou encore le suivi du nombre de châteaux créés. Ces différents éléments permettent d’appréhender globale- ment la multi-dimensionnalité de la transformation de la société entre 800 et 1200, et sont nécessaires à la bonne compréhension de sa composante spatiale qui constitue le cœur d’interêt de SimFeodal. La multiplication des indica- teurs de sortie, nécessaire à une bonne appréhension de la diversité des com- portements d’un modèle, pose alors, en premier lieu, une difficulté concrète ne serait-ce que pour les visualiser. 3.1.1 Une collecte de toutes les données de sortie de simulation Cette collecte doit être aussi large que possible, afin d ’être e n m esure de fournir une base de calcul à l’ensemble des indicateurs de sortie de simulation convenus. Au-delà même des indicateurs de sortie définis, et afin de garantir une reproductibilité des analyses, il est nécessaire de veiller à généraliser la collecte des données produites par le modèle, c’est-à-dire d’enregistrer aussi les données qui ne devraient pas être étudiées. En effet, pour faire face à de potentielles futures modifications du modèle ou assurer une rétro-compatibilité complète, il est obligatoire de disposer des données fines dont l’agrégation permet le calcul des indicateurs et leur visualisation. De plus, au fur et à mesure de l’évolution du modèle, le nombre d’exécution du modèle dont il faut pouvoir conserver les données augmente. Par conséquent, la masse de données produites augmente d’autant, et devient ainsi rapidement trop importante pour être gérée de ma- nière manuelle. Dans le cas de SimFeodal par exemple, la masse totale de données collectées sur le modèle au cours des nombreuses étapes de son évo- lution – plusieurs milliers de simulations – représente des dizaines de millions d’enregistrements en base de données. 3.1 Visualiser de nombreux indicateurs de sortie de simulation La masse des indicateurs de sortie (voir 2.2.3) rend, en effet, peu évidente la consultation « en direct » de l’exécution du modèle. De plus, en raison de la stochasticité du modèle, il est nécessaire d’effectuer plusieurs réplications de chaque simulation afin d ’en v isualiser l a t endance a grégée. L e n ombre d’indi- cateurs ainsi généré est important, et nous avons choisi, pour en garantir une consultation aisée, d’organiser ces indicateurs dans un outil interactif dédié à la mise en œuvre de l’évaluation visuelle. La visualisation doit donc se faire a posteriori, et implique dès lors de mener une collecte importante des données générées. 3 Mise en œuvre pratique de la démarche d’évaluation visuelle Dans le cas de SimFeodal, nous avons identifié une cinquantaine d’indices empiriques utiles à mobiliser, c’est-à-dire à comparer systématiquement avec les indicateurs de sortie de simulation correspondant – ces indices sont pré- sentés, ainsi que leur forme attendue, dans Cura (2020a, pp. 142–150). Cette masse s’explique notamment par le fait que, dans ce modèle descriptif, près d’une dizaine de types d’agents coexistent et interagissent. Il est dès lors impor- tant d’observer les comportements (au moins agrégés) de ces différents types d’agents. On peut citer comme exemple le suivi temporel du taux de foyers paysans dispersés, la répartition spatiale des agrégats de population (villes, vil- 145 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 3.1.2 Gestion et organisation d’une importante masse de données Pour assurer l’universalité et la simplicité de l’interrogation de ces données massives, nous avons choisi de les organiser au sein d’une base de données re- lationnelle. La masse de données imposait des contraintes fortes en matière de performance des requêtes, notamment en raison de notre choix de calculer les indicateurs « à la volée ». L’interrogation et l’agrégation de millions d’enregis- trements dans un temps acceptable pour une visualisation interactive nous a Communications Communications 146 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Pour une évaluation visuelle Pour une évaluation visuelle poussé à réaliser un benchmark de systèmes de gestion de bases de données (SGBD) relationnels performants (Cura, 2020a, fig. 4.10, p. 226). En raison de la généricité d’interrogation (SQL) et de l’extrême rapidité de traitement des requêtes qu’il permet, nous avons choisi de mobiliser un SGBD « analytique » libre, orienté colonne et haute-performance, OmniSciDB (Root, Mostak, 2016). 3.1.3 Mise en place d’un pipeline de visualisation Afin de garantir une certaine reproductibilité à nos analyses, nous avons opté pour une chaîne de traitement basée sur des solutions libres et programmables plutôt que sur des outils graphiques, afin que le partage des codes-sources (et des données) permette de reproduire les visualisations de manière stricte. La production graphique des indicateurs de sortie de simulation s’inscrit ainsi dans un « pipeline de visualisation » (fig. 1) formalisé avec le langage libre R, qui orchestre l’interrogation des données, leur filtrage, la production des visualisations (carto)graphiques et tabulaires, ainsi que leur organisation et présentation au sein d’une plateforme interactive dédié, SimEDB (fig. 2). OmnisciDB (SQL) Orchestration : Shiny (R/js) Données en sortie de simulation Calcul des indicateurs de sortie Représentation des indicateurs de sortie Figure 1. Le « pipeline » de visualisation – d’après Keim et al. (2010) – et son implémentation dans SimEDB. Figure 1. Le « pipeline » de visualisation – d’après Keim et al. (2010) – et son implémentation dans SimEDB. 3.2 Comparer visuellement des simulations La co-construction d’un modèle de simulation s’appuie sur une itération fréquente entre évolutions du modèle conceptuel, implémentation et évaluation des sorties. Chaque version du modèle entraîne en effet des modifications dans les valeurs de paramètres ou dans le détail des mécanismes, et il est donc im- portant de pouvoir comparer les effets de ces changements. Pour appliquer une démarche d’évaluation visuelle, il est alors nécessaire de procéder régulièrement à une comparaison visuelle entre les simulations. 3.2.1 Visualiser les différences entre simulations Dans la définition de l’évaluation visuelle (2.2.2), nous explicitions le be- soin de recourir à une comparaison entre les indicateurs de sortie du modèle et des formes attendues d’indices empiriques exprimées dans une grille d’analyse. 147 Communications Communications 147 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] A1 A2 B C Figure 2. Interface de la plateforme SimEDB. Figure 2. Interface de la plateforme SimEDB. À mesure que la construction du modèle évolue (calibration notamment), les écarts entre ces valeurs diminuent, jusqu’à atteindre des valeurs trop faibles pour être jugées thématiquement significatives. Il devient alors nécessaire de mener une évaluation plus fine, en comparant directement les indicateurs de sor- tie des simulations les unes avec les autres (évaluation relative) plutôt qu’avec le référentiel absolu que constituent les indices empiriques. SimEDB est donc orientée vers une approche comparative plus que normative : si une partie des indices empiriques sont figurés dans certaines visualisations, l’accent est majo- ritairement placé sur la visualisation des différences entre simulations. 3.2.2 De la visualisation à la comparaison interactive Au cours de son « cycle de vie », le modèle SimFeodal est passé par sept versions majeures et près d’une cinquantaine de sous-versions, sans compter les essais de variations de jeux de paramètres au sein même de ces sous-versions. L’identification des configurations du modèle (versions concurrentes, valeurs des jeux de paramètres mobilisés) devenait alors un problème en elle-même. Nous avons alors choisi de renforcer l’interactivité de SimEDB, en permettant notamment une sélection des simulations à comparer par un outil graphique. Dans l’interface de SimEDB, le choix des simulations dont on souhaite visuali- ser les indicateurs de sortie se fait ainsi via une sélection graphique des valeurs de paramètres à isoler, via un brushing sur des graphiques interactifs en coor- données parallèles (fig. 2, A1 et A2). En répétant ces opérations de « brushing » Communications 148 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Pour une évaluation visuelle Pour une évaluation visuelle dans l’interface , on peut ainsi comparer rapidement les indicateurs de sortie des simulations (accessibles via le double-menu d’onglets B). L’outil étant interac- tif, nous préférons le présenter sous forme de vidéo plutôt que d’en réaliser une description textuelle forcément moins adaptée. Une présentation de SimEDB est ainsi disponible dans un dépôt dédié (Cura, 2020b). 4 Conclusion Nous proposons une méthode d’évaluation de modèles de simulation sto- chastiques, descriptifs, et fondés sur la connaissance experte plus que sur des données empiriques. Cette méthode en elle-même a été conçue, raffinée et mise en œuvre collectivement dans le cadre de la co-construction d’un modèle de si- mulation ancré dans un passé lointain et sur le temps long. Au vu des évolutions du modèle SimFeodal que l’évaluation visuelle a accompagné, nous pouvons dire que le résultat de cette démarche est concluant. Le modèle produit des résultats comparables aux attentes que nous en avions et donc conforme aux connaissances empiriques sur lesquelles il s’appuie. L’exécution systématique de ce processus d’évaluation visuelle a facilité la réalisation d’allers-retours nombreux et réguliers entre le modèle conceptuel, son implémentation géné- rale, le détail technique des mécanismes et les valeurs de paramètres testées. Les dernières versions du modèle, les mieux calibrées, nous permettent ainsi de valider la plausibilité des hypothèses thématiques que le modèle cherchait à expérimenter. La robustesse de SimFeodal a enfin été testée par l’application d’une analyse de sensibilité, elle aussi menée de manière visuelle (Cura, 2020a, pp. 290–322). Il nous semble donc que l’évaluation visuelle, enrichissant l’étape préalable et peu formalisée qu’est la face validation, est une méthode tout à fait valable pour l’évaluation de modèles de simulations qui se prêteraient mal aux méthodes plus formelles. Plus généralement, nous gageons que la construc- tion et l’évaluation de tous les types de modèles de simulation gagneraient à mobiliser plus fréquemment des méthodes visuelles. Cura R. (2020b). Présentation de SimEDB. DOI: 10.5281/zenodo.4554338. Bibliographie Amblard F., Rouchier J., Bommel P. (2006). Evaluation et validation de mo- dèles multi-agents. Modélisation et simulation multi-agents. Applications pour les Sciences de l’Homme et de la Société, Hermès, p. 103–140. Balci O. (1994). Validation, verification, and testing techniques throughout the life cycle of a simulation study. Annals of operations research, vol. 53, no 1, p. 121– 173. Balci O. (1994). Validation, verification, and testing techniques throughout the life cycle of a simulation study. Annals of operations research, vol. 53, no 1, p. 121– 173. Balci O. (1998). Verification, validation, and accreditation. In Proceedings of the 1998 WSC, vol. 1, p. 41–48. Cura R. (2020a). Modéliser des systèmes de peuplement en interdisciplinarité : co- construction et exploration visuelle d’un modèle de simulation. Thèse de doctorat non publiée, Université Panthéon-Sorbonne. (https://tel.archives-ouvertes.fr/tel -02944405) Cura R. (2020b). Présentation de SimEDB. DOI: 10.5281/zenodo.4554338. Communications 149 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Cura R., Tannier C., Leturcq S., Zadora-Rio E., Lorans E., Rodier X. (2017). Tran- sition 8 : 800-1100. Fixation, polarisation et hiérarchisation de l’habitat rural en Europe du Nord-Ouest. In L. Sanders (Ed.), Peupler la terre - De la préhistoire à l’ère des métropoles, p. 301–332. Presses universitaires François Rabelais. De Runz C. (2008). Imperfection, temps et espace : modélisation, analyse et visuali- sation dans un SIG archéologique. Thèse de doctorat non publiée, Université de Reims - Champagne Ardenne. (https://tel.archives-ouvertes.fr/tel-00560668) Durand-Dastès F. (1999). Jamais deux fois... Ou : de quelques précautions à prendre avec le temps. Travaux de l’Institut de Géographie de Reims, vol. 26, no 101, p. 5–23. Edmonds B., Moss S. (2005). From KISS to KIDS – An ‘Anti-simplistic’ Modelling Approach. In P. Davidsson, B. Logan, K. Takadama (Eds.), International work- shop on multi-agent systems and agent-based simulation, p. 130–144. Springer Berlin Heidelberg. Hermann C. (1967). Validation Problems in Games and Simulations with Special Reference to Models of International Politics. Behavioral Science, vol. 12, no 3. Keim D., Kohlhammer J., Ellis G., Mansmann F. (2010). Mastering the Information Age Solving Problems with Visual Analytics. Eurographics Association. Klügl F. (2008). A Validation Methodology for Agent-based Simulations. In Procee- dings of the 2008 ACM Symposium on Applied Computing, p. 39–43. New York, NY, USA, ACM. Naylor T. H., Finger J. M. (1967). Verification of Computer Simulation Models. Management Science, vol. 14, no 2, p. B92-B106. Petty M. D. (2010). Verification, validation, and accreditation. Bibliographie Modeling and simula- tion fundamentals: Theoretical underpinnings and practical domains, p. 325–372. Root C., Mostak T. (2016). MapD: A GPU-powered Big Data Analytics and Visua- lization Platform. In ACM SIGGRAPH 2016 Talks, p. 73:1–73:2. New York, NY, USA, ACM. Tannier C., Cura R., Leturcq S., Zadora-Rio E. (2020). An agent-based model for exploring the combined effects of social and demographic changes on the concen- tration and hierarchy of rural settlement patterns in North-Western Europe during the Middle Ages (800–1200 CE). Journal of Anthropological Archaeology, vol. 59, p. 101204. 150 Communications 150 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Marie Havret 1, Marion Le Texier 2, Sophie de Ruffray 3 Marie Havret 1, Marion Le Texier 2, Sophie de Ruffray 3 1. UMR 6266 IDEES, Université de Rouen Normandie marie.havret@univ-rouen.fr 1. UMR 6266 IDEES, Université de Rouen Normandie marie.havret@univ-rouen.fr 2. UMR 6266 IDEES, Université de Rouen Normandie marion.le-texier@univ-rouen.fr 3. UMR 6266 IDEES, Université de Rouen Normandie sophie.deruffray@univ-rouen.fr RÉSUMÉ. Le développement des outils informatiques a permis de porter un nouveau re- gard sur de nombreuses sources historiques, tant dans le recueil des données que dans l’usage de nouvelles approches pour les analyser. C’est notamment le cas du cadastre ancien qui permet de collecter des données foncières fiables et homogènes, à l’échelle intra-urbaine pour l’ensemble du territoire français dès le début du XIXe siècle. Cet article compare deux méthodes de localisation automatique des adresses permettant de répondre à l’absence de plan de référence. Il se base sur l’étude du cadastre ancien de la ville de Rouen, réalisé à partir de 1827. ABSTRACT. The use of historical sources is renewed by the latest developments in computer tools, both in the collection of data and in the approaches used to analyse them. This is notably the case of the Napoleonic cadastre, a major historical source to gather reliable and homogenous intra-urban land data (dwelling) for the whole of France since the beginning of the 19th century. This article describes two methods to locate cadastre data in the absence of a reference plan. It is based on a study of the Napoleonic cadastre of the city of Rouen (France), carried out from 1827. MOTS-CLÉS : base de données géohistorique, cadastre, localisation intra-urbaine, repro- ductibilité, géocodage KEYWORDS: geohistorical database, cadastre, intra-urban location, reproducibility, geo- coding 151 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Les plans sont consultables à l’adresse suivante : http://www.archivesdepartementales76 .net/rechercher/archives-en-ligne/plans-du-cadastre/ 2. http://www.inventaire.culture.gouv.fr/pdf/2015_Haute_Normandie.pdf 1. Introduction Elle est composée de deux parcelles, la maison et le jardin (respectivement no1015 et no1014) - Première ligne : Le jardin, d’une superficie de 252 m2 (colonne "Contenance"), est évalué à 2,52 francs (colonne "Revenu"). La quatrième classe (colonne "Classement") correspond aux jardins d’agréments. - Première ligne : Le jardin, d’une superficie de 252 m2 (colonne "Contenance"), est évalué à 2,52 francs (colonne "Revenu"). La quatrième classe (colonne "Classement") correspond aux jardins d’agréments. - Deuxième ligne : La maison est évaluée à 165 francs et dispose de 22 portes et fenêtres (colonne "Nombre d’ouvertures imposables"). Sa superficie au sol est connue grâce à la ligne suivante. - Deuxième ligne : La maison est évaluée à 165 francs et dispose de 22 portes et fenêtres (colonne "Nombre d’ouvertures imposables"). Sa superficie au sol est connue grâce à la ligne suivante. - Troisième ligne : La superficie au sol de la maison est de 61 m2, correspondant à un revenu de 0,61 francs. - Troisième ligne : La superficie au sol de la maison est de 61 m2, correspondant à un revenu de 0,61 francs. Figure 2 – Extrait d’une matrice cadastrale (4e section, Rouen, 1828) Figure 2 – Extrait d’une matrice cadastrale (4e section, Rouen, 1828) 1. Introduction Elaboré à partir du début du XIXe siècle en France, le cadastre est une source de données extrêmement riche pour les chercheurs, puisqu’il permet de recueillir des informations fiables, exhaustives et homogènes sur les propriétés foncières pour l’ensemble du territoire national, de manière comparable dans le temps. Il est composé de deux documents distincts : le plan cadastral (FIG. 1), qui cartographie la géométrie des parcelles, et la matrice cadastrale (FIG. 2), qui recueille les informations attributaires : propriétaire (nom, profession, ...), propriété (adresse, superficie, ouvertures ...) et annexes (jardin, ...). Source: Archives départementales de la Seine-Maritime, 1827 Figure 1 Plan cadastral de Rouen Source: Archives départementales de la Seine-Maritime, 1827 Figure 1 – Plan cadastral de Rouen Figure 1 – Plan cadastral de Rouen Les matrices cadastrales sont mises à jour à mesure des changements de propriétaires, des modifications du bâti et des réévaluations périodiques. Ce- pendant, le plan parcellaire réalisé lors de l’introduction du cadastre (entre 1808 et 1848 selon les territoires infra-nationaux) a rarement été mis à jour avant 1970 et ne permet donc pas de suivre les modifications de propriétés (agréga- tion ou morcellement du parcellaire ; création, disparition ou modification du bâti). La seule information permettant de localiser les parcelles recensées dans le cadastre au cours du temps est l’adresse, renseignée par un numéro et un nom de rue dans les matrices. La littérature rend compte d’un ensemble d’approches localisant des objets historiques à partir d’adresses (TAB. 1), dont la mise en oeuvre rencontre des pré-requis et des difficultés techniques plus ou moins importants et surmon- tables selon le contexte de recherche. Dans cet article, nous évaluons la possi- bilité de localiser des parcelles à partir de leurs adresses sur un plan de rues dans un temps limité à partir de l’exemple du cadastral historique rouennais de 1828. 152 Communications 152 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Localiser le foncier en 1828 Source: Archives départementales de la Seine-Maritime, 1828 Localiser le foncier en 1828 Localiser le foncier en 1828 Source: Archives départementales de la Seine-Maritime, 1828 Source: Archives départementales de la Seine-Maritime, 1828 Source: Archives départementales de la Seine-Maritime, 1828 Monsieur Juquin, tailleur de profession et résidant au 11 Rue de l’Hôpital à Rouen, détient une propriété située au 19 Bis Rue Bihorel. 2. Le cadastre rouennais de 1828 La ville de Rouen est cadastrée au cours des années 1827 et 1828. Le plan est d’abord levé, puis les matrices cadastrales sont rédigées. Un registre est tenu par section cadastrale. Seul le plan cadastral de 1827 est disponible à Rouen, mais les matrices sont mises à jour, et ce jusqu’en 1971. L’ensemble des documents cadastraux de la ville sont disponibles aux Ar- chives départementales de Seine-Maritime (AD76). Les données des matrices cadastrales de 1828 ont été saisies manuellement dans un tableur, représentant 13 911 parcelles bâties. Le plan cadastral de Rouen a été scanné par les AD76 1 puis vectorisé dans le cadre du projet de recherche Morphologie urbaine, porté par l’Université de Rouen et la Région Haute-Normandie (2012-2015) 2. Lors de 153 153 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Les deux méthodes présentées dans l’article s’inspirent des méthodes apparaissant en orange Table 1 – Méthodes de localisation des adresses anciennes Les deux méthodes présentées dans l’article s’inspirent des méthodes apparaissant en orange Table 1 – Méthodes de localisation des adresses anciennes Table 1 – Méthodes de localisation des adresses anciennes Table 1 – Méthodes de localisation des adresses anciennes la vectorisation du plan, les membres du projet n’ont pas renseigné l’identifiant unique des parcelles dans la table attributaire, élément qui aurait permis de passer automatiquement du plan à la base de données. Nous avons constitué un échantillon de contrôle afin de comparer les posi- tions réelles et les positions estimées par les méthodes de localisation automa- tique. Les numéros de parcelle de cet échantillon ont été saisis dans la table attributaire du shapefile reprenant le plan cadastral. Trois feuilles de section situées dans des zones variées en termes de morphologie urbaine et d’ancienneté du bâti ont été sélectionnées, pour un total de 733 parcelles (FIG. 1): – Une feuille de la section 6 : située dans l’hypercentre, cette section est caractérisée par un bâti dense et ancien. – Une feuille de la section 4 : elle correspond à l’extension urbaine du début du XIXe siècle, à l’extérieur des boulevards intra-urbains. Elle est majoritaire- ment constituée de maisons individuelles disposant de jardins privés. – Une feuille de la section 11 : située sur la rive gauche de la ville, cette feuille parcellaire est caractérisée par une faible urbanisation, concentrée le long des rues principales. 154 Communications 154 3.1. Positionnement par géocodage à partir d’une base de données actuelles 3.1. Positionnement par géocodage à partir d’une base de données actuelles La méthode de positionnement par géocodage utilise les adresses actuelles pour définir la localisation des adresses anciennes (FIG. 3). Elle postule donc une relative stabilité entre les adresses anciennes et actuelles, en dépit des modifications de la structure urbaine. Afin d’attribuer des coordonnées géographiques aux parcelles bâties à par- tir des adresses recensées dans les matrices, nous avons utilisé la librairie R "banR", créée par Joël Gombin et Paul-Antoine Chevalier, qui permet d’inter- roger directement l’API de la Base Adresse Nationale (BAN), mise à jour de façon hebdomadaire, et de récupérer les coordonnées géographiques de l’adresse demandée (Giraud et al., 2019). Les parcelles sans numéro sont positionnées aléatoirement dans la rue. Aucune localisation n’est attribuée aux parcelles dont les rues ne figurent pas dans la BAN. Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Localiser le foncier en 1828 Localiser le foncier en 1828 Enfin, afin de localiser les numéros et rues des propriétés bâties, nous avons vectorisé et renseigné les 674 rues figurant sur le plan cadastral dans un SIG. 3. L’ensemble des données et des codes R utilisés dans ce papier seront publiés prochainement sur https://gitlab.huma-num.fr/mhavret/ 3.2. Positionnement proportionnel La méthode de positionnement proportionnel déduit la position de chaque parcelle le long de la rue à partir de son numéro et nom de rue, du nombre de parcelles bâties et de la longueur de la rue (FIG. 4). Cette méthode positionne les numéros de façon régulière le long de la rue, en faisant correspondre le début de la rue au sens géométrique (startpoint) au premier numéro de la rue 3. Cette méthode postule donc une régularité de la distribution des habitations le long des rues. Cette méthode de positionnement proportionnel ne fonctionne donc que si le premier et le dernier numéros de rue ainsi que le sens de la numérotation sont connus. Nous avons localisé le startpoint suivant la logique édifiée pour Paris en 1805 et généralisée à l’ensemble des communes françaises par l’édit royal du 23 avril 1823 (Duvergier, 1823) : pour les rues parallèles à la Seine, la numérotation suit le sens de l’écoulement de l’eau ; pour les rues perpendiculaires ou obliques, les premiers numéros commencent du côté de la rue qui est le plus proche de la Seine. Cette méthode nécessite toutefois d’identifier au préalable les premiers et derniers numéros de chaque rue, soit par un plan, soit par un dictionnaire 155 155 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 3 – Méthode d’attribution d’une localisation à partir du géocodage des adresses Figure 3 – Méthode d’attribution d’une localisation à partir du géocodage des adresses recensant la localisation des numéros de rue comme dans le cas parisien (Varet- Vitu et al., 2020), par exemple. 3.3. Critères d’évaluation Trois critères sont pris en compte pour évaluer les méthodes : – la complétude, mesurée par le nombre de maisons que l’algorithme par- vient à positionner (correctement ou non) sur le plan de rues, – la précision, mesurée par la distance entre la position calculée et la posi- tion réelle que nous étudions de façon agrégée et par la comparaison cartogra- phique, – la distorsion, mesurée par le différentiel analytique obtenu selon la mé- thode de localisation, que nous illustrons par la mesure de l’autocorrélation spatiale des valeurs foncières (prix au mètre carré) calculée à partir des posi- tions réelles et des positions estimées. Communications Communications 156 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Localiser le foncier en 1828 Figure 4 – Méthode de positionnement proportionnel Figure 4 – Méthode de positionnement proportionnel Les classes de distance choisies pour discrétiser les distances d’erreur se basent sur celles établies par Cura et al (2018) dans leur étude sur le géo- codage des adresses parisiennes aux XIXe et XXe siècles afin de faciliter les comparaisons. A partir d’un échantillon d’adresses géocodées et de leurs posi- tions réelles définies manuellement, les auteurs proposent les classes suivantes : entre 0 et 15m, le positionnement est correct ; entre 15 et 55m, la rue est cor- recte, mais les numéros des bâtiments sont légèrement mal placés ; entre 55 et 155m, la rue est correcte dans la plupart des cas, mais les numéros des bâti- ments sont loin de leur position correcte ; au-dessus de 155m, les positions sont mauvaises. 4.1. Complétude A l’issue des traitements, les deux méthodes de localisation parviennent cha- cune à localiser 714 parcelles (97% de l’échantillon). La méthode de position- 157 Communications Communications Communications 157 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 nement proportionnel ne parvient pas à localiser les parcelles dont le numéro de rue est inconnu. Cette situation concerne les sections les plus récemment urbanisées en 1828 (5 parcelles pour la section 4, 14 parcelles pour la section 11). La méthode de géocodage ne parvient pas à localiser les parcelles de deux rues de la section 11, toutes deux disparues aujourd’hui. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Localiser le foncier en 1828 actuels correspondent à ceux de rues disparues ou renommées depuis. C’est par exemple le cas de la "Rue du Barbet", anciennement située dans l’hypercentre de Rouen, associée à la "Rue Henri Barbet", actuellement située dans l’extrême ouest de la ville. Logiquement, les jointures par proximité sémantique partielle sont d’autant plus sources d’erreur que le seuil de ressemblance accepté est bas. Avec les paramètres par défaut de la fonction BanR, l’adresse "5 Rue du Corbeau" est par exemple associée à la "Rue du Fardeau". Au regard de ce critère, la méthode d’attribution d’une localisation par géocodage n’apparaît pas satisfaisante, en l’état actuel. Ces résultats soulignent la nécessité d’associer cette méthode à une validation manuelle ou à une base de données historiques définie en amont. 4.3. Distorsion Pour la méthode de positionnement proportionnel dont les résultats sont satisfaisants au regard des deux premiers critères d’évaluation, une évaluation thématique a également été conduite. La comparaison des valeurs locales d’au- tocorrélation spatiale 4 du prix du foncier ramené au mètre carré (TAB. 2 et FIG. 7) montre que les résultats obtenus par une localisation manuelle ou au- tomatique des parcelles sont relativement homogènes pour les sections 4 et 11, quelque soit la taille du voisinage pris en compte. Cependant, la section 6 fait apparaître de forts écarts entre les deux échantillons, notamment avec un voisinage restreint. 1 voisin 5 voisins 10 voisins Section 4 3.92 0.56 0.32 Section 6 100.5 (***) 11.04 (**) 5.41 (*) Section 11 4.62 (.) 3.59 2.17 Les valeurs correspondent aux valeurs du Chi2. Signif. p-value : 0.001 ’***’; 0.01 ’**’; 0.05 ’*’ ; 0.1 ’.’; 1 ’ ’ T 2 Ré lt t d t t d Chi2 d’h é éité 1 voisin 5 voisins 10 voisins Section 4 3.92 0.56 0.32 Section 6 100.5 (***) 11.04 (**) 5.41 (*) Section 11 4.62 (.) 3.59 2.17 Les valeurs correspondent aux valeurs du Chi2. Signif. p-value : 0.001 ’***’; 0.01 ’**’; 0.05 ’*’ ; 0.1 ’.’; 1 ’ ’ T 2 Ré l d d Chi2 d’h é éi é Les valeurs correspondent aux valeurs du Chi2. Signif. p-value : 0.001 ’***’; 0.01 ’**’; 0.05 ’*’ ; 0.1 ’.’; 1 ’ ’ Les valeurs correspondent aux valeurs du Chi2. Signif. p-value : 0.001 ’***’; 0.01 ’**’; 0.05 ’*’ ; 0.1 ’.’; 1 ’ ’ Table 2 – Résultats du test du Chi2 d’homogénéité. 4. L’indice I de Moran local a été utilisé (Anselin, 1995). 4.2. Précision Les modifications d e t oponymie e t l e r éordonnancement d es n uméros de rue ayant eu lieu entre 1828 et aujourd’hui expliquent l’importance des écarts de distance obtenus par la méthode de géocodage (FIG. 5). La méthode de positionnement proportionnel, n’étant pas vulnérable à ces changements, per- met d’obtenir, globalement, une plus grande précision dans la localisation des parcelles historiques. Figure 5 – Distance entre les positions réelles et les positions estimées selon les deux méthodes (n=713) Figure 5 – Distance entre les positions réelles et les positions estimées selon les deux méthodes (n=713) Le regroupement spatial des écarts de distance les plus importants ren- seignent sur les facteurs d’erreur de localisation (FIG. 6). On remarque ainsi que la méthode de positionnement proportionnel est moins précise lorsque l’ur- banisation n’est pas homogène, notamment lorsque un tronçon de rue est très densément construit, alors que le reste ne l’est pas, à l’image de la rue Saint- Julien (section 11). C’est également le cas de la rue Eau de Robec (section 6) et du chemin aux Boeufs (section 4). Plus la rue est longue, plus le problème est exacerbé. Ces rues concentrent la quasi totalité des positions incorrectes (>155m). Les résultats obtenus par la méthode de géocodage sont beaucoup plus hé- térogènes spatialement. Cette méthode fonctionne mieux dans l’hypercentre de Rouen qu’en périphérie, notamment pour les rues anciennes dont la morpholo- gie, le nom, et la numérotation sont stabilisés depuis deux siècles. Les résultats sont moins satisfaisants pour les sections 4 et 11, où plusieurs noms de rue Communications 158 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 5. Conclusion et perspectives Comme nous le rappelions en introduction, le cadastre constitue une source unique pour l’analyse comparée du foncier dans le temps et dans l’espace. Cependant, le format des archives (papier ou image pixelisée) en rend la mani- pulation difficile. Nous avons comparé la qualité de l’information géographique obtenue avec deux méthodes de positionnement automatique des parcelles à 159 159 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 6 – Distance euclidienne entre les positions réelles et estimées par la méthode de positionnement proportionnel (à gauche) et par géocodage (à droite) Figure 6 – Distance euclidienne entre les positions réelles et estimées par la méthode de positionnement proportionnel (à gauche) et par géocodage (à droite) 160 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Localiser le foncier en 1828 Figure 7 – Clusters issus des mesures d’autocorrélation spatiale de la variable "Prix au m2", avec les positions réelles (à gauche) et estimées (à droite) - 1 voisin Figure 7 – Clusters issus des mesures d’autocorrélation spatiale de la variable "Prix au m2", avec les positions réelles (à gauche) et estimées (à droite) - 1 voisin 161 Communications Communications 161 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] partir des adresses renseignées dans les matrices cadastrales (seul élément mis à jour pendant plus de 100 ans). Nous avons montré qu’une évaluation basée sur des critères géométriques seuls (complétude, précision) conduirait à considérer que le positionnement proportionnel des numéros de rue -tel que défini dans cet article- produit d’ex- cellents résultats. Cependant, l’utilisation des données ainsi définies dans une analyse thématique montre l’étendue des distorsions des résultats obtenus en comparaison de ceux issus d’une localisation manuelle des parcelles. Une des principales perspectives de ce travail est donc d’améliorer la méthode de po- sitionnement proportionnel en définissant p ar e xemple d es p oints d e contrôle (bâti remarquable) comme le font Anne Varet-Vitu et al (2020) pour affiner les localisations utilisées. La méthode de géocodage pourrait, quant à elle, être perfectionnée en constituant un répertoire d’adresses historiques en amont. (W lf d 2019)(Y t l 2016)(L L l 2019)(O f d t l 2002) Bibliographie Anselin L. (1995). Local indicators of spatial association—lisa. Geographical analysis, vol. 27, no 2, p. 93–115. Cura R., Dumenieu B., Abadie N., Costes B., Perret J., Gribaudi M. (2018). Historical collaborative geocoding. ISPRS International Journal of Geo-Information, vol. 7, no 7, p. 262. Duvergier J.-B. (1823). Collection complète des lois, décrets, ordonnances, règlements et avis du conseil d’état (vol. 24). imprimé par Charles Noblet. Giraud T., Pecout H., Ysebaert R. (2019). Accessibilité et données openstreetmap sur r : exemple pratique sur les maternités de la niévre et de la seine-saint-denis. Séminaire CIST de juin 2019 - Atelier calculs d’accessibilités avec des données en libre accès. Lan T., Longley P. (2019). Geo-referencing and mapping 1901 census addresses for england and wales. ISPRS International Journal of Geo-Information, vol. 8, no 8, p. 320. Orford S., Dorling D., Mitchell R., Shaw M., Smith G. D. (2002). Life and death of the people of london: a historical gis of charles booth’s inquiry. Health & place, vol. 8, no 1, p. 25–35. Varet-Vitu A., Marraud M., Mermet É. (2020). Spatialités sociales à paris à la veille de la révolution. Histoire urbaine, no 2, p. 157–186. Walford N. S. (2019). Bringing historical british population census records into the 21st century: A method for geocoding households and individuals at their early- 20th-century addresses. Population, Space and Place, vol. 25, no 4, p. e2227. Yu R., Luo Z., Chiang Y.-Y. (2016). Recognizing text in historical maps using maps from multiple time periods. In 2016 23rd international conference on pattern recognition (icpr), p. 3993–3998. 162 Communications Communications 162 Communications par Session Extraction de contenu pour le GeoWeb [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Constitution de corpus thématique : Pour un meilleur suivi du territoire de la Métropole de Montpellier Méditerranée Rodrique Kafando 1,4, Rémy Decoupes 1, Lucile Sautot 1,2, Maguelonne Teisseire 1, Christiane Weber 1,3 Rodrique Kafando 1,4, Rémy Decoupes 1, Lucile Sautot 1,2, Maguelonne Teisseire 1, Christiane Weber 1,3 1. TETIS, Univ Montpellier, AgroParisTech, CIRAD, CNRS, INRAE, Montpellier, France prenom.nom@inrae.fr 2. AgroParisTech, Montpellier, France lucile.sautot@agroparistech.fr 2. AgroParisTech, Montpellier, France lucile.sautot@agroparistech.fr 3. CNRS, France christiane.weber@cnrs.fr 4. Montpellier Méditerranée Métropole, France 4. Montpellier Méditerranée Métropole, France RÉSUMÉ. Dans le cadre d’un partenariat avec la métropole de Montpellier (3M) et son service ville intelligente, nous établissons une plateforme permettant l’intégration et l’analyse de données hétérogènes massives pour une observation intelligente du ter- ritoire concerné. Dans cet article, nous décrivons un processus de récolte automatique de documents de sources diverses, selon trois thématiques choisies par 3M qui sont l’urbanisation, l’agriculture et l’hydrologie. A partir de termes graines à dire d’experts enrichies grâce à des articles Wikipédia puis Google, un vocabulaire de concepts est automatiquement constitué. Les documents obtenus à partir d’un moteur de recherche sont ensuite évalués à l’aide d’une mesure sémantique spécifique. Nous détaillons les premiers résultats obtenus et présentons les avantages du protocole proposé. ABSTRACT. In collaboration with the metropolis of Montpellier (3M) and its smart city service, we are establishing a platform allowing the integration and analysis of massive heterogeneous data for intelligent observation of the territory concerned. In this article, we describe the process implemented to ensure an automatic collection of documents according to three themes chosen by 3M, which are urbanization, agricul- ture and hydrology. Results are detailed and the benefits o f t his s calable s olution are highlited. 165 Communications 165 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 MOTS-CLÉS : Observation du territoire, Constitution automatique de corpus KEYWORDS: Land Planning, automatic corpus constitution 1. Introduction La Métropole de Montpellier Méditerranée (3M) développe depuis plusieurs années une politique pro-active en faveur de développements numériques inno- vant par la mise en œuvre d’approches fondées sur l’hétérogénéité des sources, l’extraction de connaissances à partir de données massives dans le but de pro- duire des outils d’aide à la décision. L’intégration de données hétérogènes, en prenant en compte l’écosystème « ville intelligente » de la Métropole (labos, start up, collectifs, etc.), se fonde sur des informations quantitatives (démogra- phiques, économiques, climatiques, etc.) mais aussi sur des informations venant d’autres sources de données, officielles et non officielles. La finalité est de mon- trer le gain et l’efficacité de la mise en relation de données gérées jusqu’ici en silo et de l’organisation des interactions qui peuvent en découler. C’est dans ce contexte que se positionnent les travaux présentés dans cet article sur un processus de constitution de corpus thématique (ou corpus dédié) à partir du Web sur le territoire concerné. Cette étude rentre dans le cadre de travaux qui visent la mise en relation et l’analyse de données textuelles et hétérogènes. Le but principal est de pouvoir collecter par couple de thématique et territoire, des données textuelles pertinentes qui couvrent une grande diversité de sources. Le challenge et le verrou associé est de ne pas se limiter aux documents purement officiels (souvent disponibles dans l’open data de la métropole). Il devient in- dispensable de disposer d’une terminologie moins académique que celle utilisée par les différentes collectivités territoriales afin de pouvoir accéder à l’expres- sion de la composante sociétale, indispensable à une véritable perception de la dynamique du territoire dans sa globalité. Dans la suite de cet article, en Section 2, nous listons un panorama des travaux sur la constitution de corpus thématiques, en particulier ceux qui ont inspiré les bases de notre démarche. Ensuite, la méthodologie que nous propo- sons pour la constitution d’un corpus propre à notre cas d’étude est détaillée dans la Section 3. Dans la Section 4, nous présentons les premiers résultats quantitatifs suite aux différentes expérimentations réalisées. Enfin, nous dres- sons un bilan et les perspectives associées en Section 5. 2. État de l’art La mise en oeuvre des concepts de la ville intelligente devrait permettre aux villes et métropoles d’améliorer l’efficacité de leurs services de gestion (Guéran- ger, Mathieu-Fritz, 2019), en offrant, notamment, davantage de transparence. Communications Communications 166 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique Cette efficacité peut être évaluée en analysant les opinions exprimées sur les différents projets et réalisations en lien avec l’aménagement du territoire (Ker- gosien et al., 2014). Les ressources disponibles sur le web, et en particulier celles de la presse locale, permettent de suivre les débats que suscitent les pro- jets d’urbanisme. Dans (Kergosien. et al., 2015), les auteurs proposent, notam- ment, une méthodologie semi-automatisée pour enrichir les scènes issues d’une série temporelle d’images satellites, focalisées sur un projet d’aménagement. Les genres, ou catégories textuelles, des ressources indexées par les moteurs de recherche sont très variés : blog, publicité, rapport officiels, articles scien- tifique etc. Plusieurs travaux proposent différentes hiérarchies de genre (Vi- dulin et al., 2009; Santini, 2011) afin de faciliter la classification de nouvelles ressources. (Madjarov et al., 2019) exploite ces hiérarchies pour renforcer les capacités d’un moteur de recherche à proposer des résultats pertinents selon les genres souhaités. L’évaluation de la pertinence des ressources collectées sur le web est, bien entendu, un enjeu majeur. Une approche classique vise à calculer les co-occurrences de termes entre la requête et les documents rapatriés pour ensuite classer les résultats en fonction du poids du terme dans le document, en utilisant, notamment la mesure TF-IDF(Aizawa, 2003). Cette méthode a cepen- dant des limites. Cette approche par mot clé ne peut retranscrire la proximité sémantique entre deux termes comme, par exemple, vélo et mobilité douce. Pour remédier ce problème, (Schaeffer. et al., 2020) propose de filtrer les ressources pertinentes en calculant la proximité sémantique, via le modèle de prolongement lexical word2vec (Mikolov et al., 2013), entre des termes qui composent le document et un thésaurus thématique. Dans un contexte de surveillance en épidémiologie animale, (Arsevska. et al., 2016) montre que la construction de requête avec deux mots clés, symptôme et animal, permet de récolter des ressources avec davantage de pertinence. Il est possible, enfin, de combiner ces deux dernières approches, prolongement lexical et requête multi- termes, pour enrichir la sémantique du terme par l’ajout d’un contexte apporté par la phrase, ou le groupe de mots, qui le contient en utilisant le modèle BERT (Devlin et al., 2018). Communications De la même façon, nous proposons la mise en place d’un ensemble de techniques prenant en compte des stratégies d’extrac- tion de concepts (officiels et non officiels) et de constitution automatique de corpus thématiques provenant de sources diversifiées, validées par des mesures de similarité sémantique. 3.1. Constitution des vocabulaires de concepts thématique Le processus pour la constitution de vocabulaire de concepts thématique est basé sur le principe de construction de corpus proposé dans (Kilgarriff, Grefenstette, 2003; Kilgarriff et al., 2010; Sharoff, 2006). Il consiste à utiliser un ensemble de graines ou une liste de termes d’un domaine caractéristique pour requêter des documents sur le web. Les résultats sont ensuite utilisés pour étendre la liste des termes et donc le corpus, et ainsi de suite. L’approche proposée se décline en deux phases : la constitution de l’ensemble de termes graines noté TG, et la constitution du vocabulaire de concepts noté VC. Figure 1 – Choix de termes initiaux - TG Figure 1 – Choix de termes initiaux - TG 3. Méthodologie L’approche méthodologique que nous proposons porte sur deux grands points. Le premier est la constitution d’un vocabulaire de concepts thématique. Le second concerne la collecte de documents en utilisant l’ensemble de termes précédemment obtenus. 167 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 1. https://fr.wikipedia.org/wiki/Wikip%C3%A9dia:Accueil_principal 3.1.2. Constitution du vocabulaire de concepts VC La constitution du vocabulaire de concepts est basée sur une première étape qui consiste à utiliser la liste TG comme une liste de graines permettant de collecter un grand nombre de documents via un moteur de recherche comme par exemple Google. L’objectif est d’obtenir des variantes de termes pour faire évoluer la terminologie de TG, afin de récolter des documents parlant de la thématique mais avec des mots et des termes qui ne seraient pas présents dans Wikipédia. Ensuite, grâce à BioTex, le corpus obtenu est utilisé pour extraire un grand ensemble de termes noté TB (termes BioTex), à partir duquel, nous allons réduire le bruit éventuel pour former le vocabulaire VC. Pour éviter de prendre en considération les termes généraux ou les termes qui ne permettent pas de définir la thématique étudiée, nous avons mis en place un filtre basé sur une approche sémantique en utilisant DistilBERT (Sanh et al., 2019), reconnue pour l’évaluation de la Similarité Sémantique Textuelle (STS). La valeur de similarité varie entre -1 pour de très faible similarité et 1 lorsque les termes sont très proches ou identiques. La mesure sémantique est évaluée à partir 1) des termes extraits TB qui sont les nouveaux termes, et 2) les termes graines ou experts TG. Pour chaque terme de TB, nous calculons sa similarité sémantique avec chacun des termes de l’ensemble TG. Ce qui nous donne un nombre de valeurs de similarité égal au nombre de termes dans l’ensemble TG pour chaque terme t de TB. Pour obtenir la valeur de similarité finale pour un terme, nous faisons la moyenne des TOP@n (n correspond aux n termes les plus pertinents) en tenant compte des plus grandes valeurs de similarités pour garder une certaine exhaustivité vis-à- vis de l’ensemble des termes experts. À la fin du processus, nous obtenons un nouvel ensemble ordonné de termes, constitué des mêmes termes de l’ensemble TB, mais avec des scores de pertinence différents. Les vocabulaires de concepts seront par la suite sélectionnés à partir de ces ensembles de termes pour chaque thématique, en partant de ceux disposant d’un plus grand score. La Figure 2 décrit les différentes étapes pour la construction de vocabulaire de concepts thématique dans son ensemble et le calcul de similarité entre termes est illustré sur la Figure 3. Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique de termes ainsi constitué forme la liste de termes experts que nous appelons ici TG. de termes ainsi constitué forme la liste de termes experts que nous appelons ici TG. 3.1.1. Choix des termes graines TG La proposition de l’ensemble TG comprend deux étapes (Figure 1). La première consiste à extraire un ensemble de termes à partir d’un mini corpus obtenus de Wikipédia 1 en utilisant l’outil BioTex de (Lossio-Ventura et al., 2014). Ce corpus est construit en utilisant un mot clé thématique (ex : urba- nisation), et les vingt premières pages renvoyées par Wikipédia sont utilisées pour former le corpus. Pour BioTex, nous avons principalement utilisé deux mesures statistiques qui sont C_Value et F-TFIDF-C_M. C_Value donne de l’importance aux termes apparaissant plusieurs fois dans le même document et aux termes com- posés permettant de valoriser les expressions. F-TFIDF-C_M représente la moyenne harmonique de C_Value et TF-IDF, qui permet de classer les termes en fonction de leur pertinence vis-à-vis du document en prenant en compte l’ensemble du corpus. F-TFIDF-C_M présente l’avantage d’utiliser toutes les valeurs de la distribution et de réduire le bruit tout au long du processus d’ex- traction. C-Value et F-TFIDF-C_M sont complémentaires, car la première favorise l’extraction des termes composés pertinents et la seconde privilégie les termes discriminants. Discriminants dans le sens ou elle permet de capturer les termes pertinents qui majoritairement ne sont pas reconnus par C_Value et TF-IDF prise individuellement. La seconde étape consiste à faire choisir un ensemble de termes par des ex- perts du domaine parmi les termes récupérés à partir de Wikipédia. L’ensemble 1. https://fr.wikipedia.org/wiki/Wikip%C3%A9dia:Accueil_principal Communications Communications 168 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] 3.2.1. Collecte de documents Une fois le vocabulaire de concepts thématique construit, nous avons pour chaque thématique, un ensemble de concepts permettant de la décrire au mieux. La phase de collecte (voir Figure 4) consiste à utiliser un module, qui prend en entrée un vocabulaire de concepts VC et une emprise spatiale, pour retourner Communications 169 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 2 – Constitution de vocabulaire de concepts thématique - VC Figure 3 – Exemple de calcul de similarité sémantique - score final calculé en faisant la moyenne des 2 plus grandes valeurs Figure 2 – Constitution de vocabulaire de concepts thématique - VC Figure 2 – Constitution de vocabulaire de concepts thématique - VC p q Figure 3 – Exemple de calcul de similarité sémantique - score final calculé en faisant la moyenne des 2 plus grandes valeurs des documents en relation avec la thématique concernée pour former un corpus de taille importante. Nous pouvons ainsi constituer pour chaque thématique, un corpus le plus en cohérence possible avec son contenu. La prochaine étape permet de filtrer les documents les moins pertinents pour chaque corpus thé- matique. Lorsque le document obtenu est pertinent, une fiche de méta-données contenant les informations sur le document est automatiquement générée avant d’être stockée dans un lac de données (Kafando et al., 2020). Figure 4 – Collecte et évaluation des documents Figure 4 – Collecte et évaluation des documents Les différentes étapes, illustrées sur la Figure 5, constituent le protocole de collecte proposé. 4. Expérimentation Nous cherchons des documents traitant de la Métropole de Montpellier et relatifs à l’une des trois thématiques qui sont l’Agriculture, l’Urbanisation et l’Hydrologie. Les codes sources de la solution mise en place seront disponibles et accessibles dans notre répertoire partagé 2. 2. https://github.com/aidmoit 3.2.2. Évaluation automatique des documents par mesure sémantique Dans le but de ne considérer que les documents qui sont pertinents pour chaque thématique, nous avons introduit une évaluation par mesure de simila- rité. Identique à la précédente, l’évaluation des documents se fait entre un do- cument collecté par rapport à l’ensemble des termes de son vocabulaire étendu de concepts thématique VEC obtenu avec WordNet (Miller, 1998). L’objectif principal de cette étape est de disposer 1) d’une couverture plus large sur les termes en introduisant leur synonyme grâce à WordNet, et 2) d’obtenir des termes moins experts et/ou administratifs qui sont importants pour l’évalua- tion des documents moins administratifs tels que les blogs, les annonces, etc. Cette technique permet de prendre en compte ces synonymes lors de l’évalua- tion des documents. Comme décrite précédemment, la valeur de similarité avec Communications 170 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique DistilBERT varie entre -1 et 1. Un document de la thématique urbanisation sera évalué avec le vocabulaire étendu de concepts de la thématique urbani- sation. Plus la valeur de similarité est forte, plus le document sera considéré comme pertinent vis-à-vis de la thématique, et vice-versa. Le processus pour l’évaluation des documents est illustré sur la Figure 4. Afin d’éliminer les documents qui ne sont pas pertinents, nous définissons lors des expérimentations, une valeur seuil à partir de laquelle, nous décidons de garder le document ou de l’exclure du corpus. Figure 4 – Collecte et évaluation des documents Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique Protocole de constitution de corpus thématique Une fois les termes graines obtenus, nous procédons à la constitution des vocabulaires de concepts VC. Pour chaque thématique, la liste TG est utilisée pour servir de mots clés aux requêtes pour constituer de corpus plus importants, sur le web, sans distinction de site. Ceci nous permet d’avoir une forte diversité des données tant dans leurs contenus que dans leurs sources. Nous avons récolté respectivement 105, 120 et 117 documents pour former les corpus Urbanisation, Agriculture et Hydrologie à partir desquels seront extraits les ensembles de termes que nous nommons termes Biotex (TB). Afin de pouvoir constituer les vocabulaires de concepts thématiques finaux, nous avons utilisé la mesure sémantique proposée dans le Section 3.1.2 qui nous a permis de limiter le bruit, en donnant des poids très faibles aux termes dont la sémantique est éloignée de la thématique. Dans le tableau 3, nous présentons les TOP@10 des termes de l’ensemble TB avant et après l’application de la mesure sémantique pour la thématique Agri- culture. Tout d’abord, nous constatons que l’ordre de pertinence des termes Table 3 – TOP@10 des termes de la thématique Agriculture avant et après la mesure sémantique (MS) Thématique Agriculture TOP@10 avant la MS TOP@10 après la MS dégradation des terres production agricole agriculture urbaine exploitation agricole etats membres agricole production changement climatique oeuvre agricole google scholar entreprise agricole occupation du sol service agricole field sizes techniques de production agricole lutte contre la désertification production agricole perdu production agricole agricole perdu matière de dégradation des terres gestion de la production agricole Table 3 – TOP@10 des termes de la thématique Agriculture avant et après la mesure sémantique (MS) Thématique Agriculture TOP@10 avant la MS TOP@10 après la MS dégradation des terres production agricole agriculture urbaine exploitation agricole etats membres agricole production changement climatique oeuvre agricole google scholar entreprise agricole occupation du sol service agricole field sizes techniques de production agricole lutte contre la désertification production agricole perdu production agricole agricole perdu matière de dégradation des terres gestion de la production agricole Table 3 – TOP@10 des termes de la thématique Agriculture avant et après la mesure sémantique (MS) a changé entre les deux étapes. 4.1. Vocabulaire de concepts thématique La première étape est le choix de termes graines TG décrit dans la sec- tion 3.1. La seconde étape consiste à soumettre cet ensemble de termes à un groupe d’experts dont le rôle est de sélectionner un sous-ensemble (liste finale de TG) de termes permettant de décrire au mieux la thématique. Pour chaque requête, nous utilisons le libellé de la thématique comme mot clé principal de 171 Communications Communications 171 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 5 – Les étapes du protocole de collecte Figure 5 – Les étapes du protocole de collecte la recherche sur Wikipédia, soit ’urbanisation’, ’agriculture’, ou ’hydrologie’. Les corpus ainsi constitués de chaque thématique sont ensuite utilisés pour l’extraction des termes parmi lesquels seront issus les termes graines. Table 1 – Statistiques sur la constitution de la liste de termes graines TG Termes requêtes Urbanisation Agriculture Hydrologie Taille Corpus (Nb docs) 20 20 20 Nombre de termes retenu 100 100 100 Table 2 – Extrait de termes experts par thématique Urbanisation TG-1 Agriculture TG-2 Hydrologie TG-3 Politiques urbaines Récolte agricole Hydrogéologie Aménagement territoriale Agriculture Eau pluviale Aménagement foncier Aménagement agricole Mesure hydrologique Aménagement touristique Production agricole Cycle de l’eau Architectonique Exploitation agricole Réseau hydrologique Les tableaux 1 et 2 présentent respectivement les statistiques et les TOP@5 Table 1 – Statistiques sur la constitution de la liste de termes graines TG Termes requêtes Urbanisation Agriculture Hydrologie Taille Corpus (Nb docs) 20 20 20 Nombre de termes retenu 100 100 100 Table 1 – Statistiques sur la constitution de la liste de termes graines TG Termes requêtes Urbanisation Agriculture Hydrologie Taille Corpus (Nb docs) 20 20 20 Nombre de termes retenu 100 100 100 Table 2 – Extrait de termes experts par thématique Urbanisation TG-1 Agriculture TG-2 Hydrologie TG-3 Politiques urbaines Récolte agricole Hydrogéologie Aménagement territoriale Agriculture Eau pluviale Aménagement foncier Aménagement agricole Mesure hydrologique Aménagement touristique Production agricole Cycle de l’eau Architectonique Exploitation agricole Réseau hydrologique Les tableaux 1 et 2 présentent respectivement les statistiques et les TOP@5 de la liste de termes graines TG de chaque thématique. 172 Communications 172 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Protocole de constitution de corpus thématique Dans le tableau 3, première colonne, de la thématique Agriculture, nous remarquons la présence de terme comme ’google scholar’, qui n’a pas un sens sémantique proche à l’agriculture, mais qui occupe un rang important dans les classements de BioTex (c’est-à-dire dans l’ensemble TB). Après avoir appliqué le calcul de similarité sémantique entre ces termes et les termes experts (conf Table 2), nous constatons dans la seconde colonne, l’apparition de nouveaux termes plus pertinents dans les TOP@10. Cela signi- fie que les nouveaux termes (ex: production agricole au 9ème rang) présents dans la deuxième étape occupaient un rang de pertinence moins fort avant la mesure de la similarité. Cette remarque est aussi valable pour les thématiques Urbanisation et Hydrologie. a changé entre les deux étapes. Dans le tableau 3, première colonne, de la thématique Agriculture, nous remarquons la présence de terme comme ’google scholar’, qui n’a pas un sens sémantique proche à l’agriculture, mais qui occupe un rang important dans les classements de BioTex (c’est-à-dire dans l’ensemble TB). Après avoir appliqué le calcul de similarité sémantique entre ces termes et les termes experts (conf Table 2), nous constatons dans la seconde colonne, l’apparition de nouveaux termes plus pertinents dans les TOP@10. Cela signi- fie que les nouveaux termes (ex: production agricole au 9ème rang) présents dans la deuxième étape occupaient un rang de pertinence moins fort avant la mesure de la similarité. Cette remarque est aussi valable pour les thématiques Urbanisation et Hydrologie. 173 Communications 173 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 4.2.1. Vocabulaire de concepts thématique final 2.1. Vocabulaire de concepts thématique final Le choix du vocabulaire de concepts thématique final est effectué en fixant un seuil minimal spécifique à chaque thématique. Pour ce faire, nous regardons donc à partir de quelle valeur du score, les termes commencent à être éloignés de la thématique. Pour la suite de notre étude, nous avons considéré les TOP@1000 termes de chaque thématique. Le tableau 4 indique le seuil correspondant pour chaque thématique. Table 4 – Statistiques sur la constitution du vocabulaire de concepts théma- tique VC Thématique Urbanisation Agriculture Hydrologie Taille Corpus (Nb documents) 105 120 117 Taille du VC TOP@1000 TOP@1000 TOP@1000 Valeur du seuil de similarité 0.80 0.75 0.77 Table 4 – Statistiques sur la constitution du vocabulaire de concepts théma- ti VC Une fois le seuil fixé, la liste de termes est étendue en utilisant WordNet. Les synonymes obtenus permettent de sortir du cadre de langage officiel ou académique, et d’utiliser des expressions moins techniques, qui permettront par exemple d’évaluer des documents en relation avec des réseaux sociaux, des blogs ou encore des annonces telles que les offres d’emplois. En exemple, certains synonymes obtenus avec le terme "zone d’aménagement concerté" de la thématique Urbanisation sont : district d’aménagement concerté, domaine d’aménagement concerté, aire d’aménagement concerté, quartier d’aménage- ment concerté, etc.. 4.2.2. Évaluation automatique des documents par mesure sémantique Cette évaluation est faite par calcul de similarité sémantique entre chaque document collecté avec le vocabulaire étendu de concepts thématique de la thématique concernée. Pour chaque thématique, nous avons considéré les TOP@1000 termes de son vocabulaire étendu de concepts thématique pour l’évaluation des docu- ments la concernant. Ce choix est fixé suite aux différentes expérimentations qui montrent une très faible variation, voir nulle au delà des TOP@1000. À la fin du processus, pour chaque thématique, un score est affecté à chaque do- cument, représentant sa proximité sémantique avec le vocabulaire étendu de concepts thématique précédemment obtenu. Nous avons défini une valeur seuil de 0.5 comme valeur minimale. Un document dont la valeur de la proximité sémantique est inférieur à 0.5 ne sera donc pas considéré dans le corpus, et ce pour l’ensemble des thématiques. Cette valeur est fixée après une vérification des résultats obtenus. Le tableau 5 indique les caractéristiques pour chaque corpus thématique. 174 174 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Protocole de constitution de corpus thématique Table 5 – Statistiques sur les corpus constitués par thématique Thématique Urbanisation Agriculture Hydrologie Taille du VC utilisé TOP@500 TOP@500 TOP@500 Taille du VEC utilisé TOP@1000 TOP@1000 TOP@1000 Seuil de similarité 0.5 0.5 0.5 Taille Corpus 867 1.400 1380 5. Conclusion Le protocole que nous avons élaboré dans cette étude vise à proposer une démarche générique permettant de constituer 1) des vocabulaires thématiques et 2) de constituer des corpus thématiques spécifiques (l’agriculture, l’urbani- sation et l’hydrologie pour la métropole de Montpellier) pour permettre une meilleure observation du territoire. Elle s’appuie principalement sur le principe de faire évoluer une terminologie allant de la constitution de graines à la for- mation d’un vocabulaire de concepts thématique bien spécifique. Quant à la constitution des corpus, elle se fait en utilisant chaque vocabulaire thématique comme ensemble de mots clés de recherche et une spécification de l’emprise spatiale d’intérêt. Chacune des étapes est soumise à une évaluation séman- tique, dans le but de limiter le bruit aussi bien dans le vocabulaire de concepts que dans les corpus thématiques. Nos travaux futurs porteront sur l’intégration de l’ensemble de ces documents au sein du lac de données de la métropole de Montpellier et leur mise en lien selon les dimensions spatiale, temporelle et thématique. Notre contribution s’inscrit en sciences de l’environnement avec le développement de méthodes permettant de gérer une grande masse de don- nées pour assurer le suivi d’un territoire donné et d’en extraire une meilleure connaissance. Ces travaux ont été soutenus par l’Agence nationale française de la recherche dans le cadre du programme Investissements d’avenir #DigitAg, référencé ANR-16-CONV-0004. Guéranger D., Mathieu-Fritz A. (2019, novembre). The smart city at work. Reseaux, vol. No 218, no 6, p. 41–75. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Kafando R., Decoupes R., Sautot L., Teisseire M. (2020). Spatial Data Lake for Smart Cities: From Design to Implementation. AGILE: GIScience Series, vol. 1, p. 1-15. Kergosien. E., Alatrista-Salas. H., Gaio. M., Güttler. F., Roche. M., Teisseire. M. (2015). When textual information becomes spatial information compatible with satellite images. In Proceedings of the 7th international joint conference on know- ledge discovery, knowledge engineering and knowledge management - volume 1: Kdir, (ic3k 2015), p. 301-306. SciTePress. Kergosien E., Laval B., Roche M., Teisseire M. (2014). Are opinions expressed in land-use planning documents? International Journal of Geographical Information Science, vol. 28, no 4, p. 739-762. KilgarriffA., Grefenstette G. (2003). Introduction to the special issue on the web as corpus. Computational linguistics, vol. 29, no 3, p. 333–347. KilgarriffA., Reddy S., Pomikálek J., Avinesh P. (2010). A corpus factory for many languages. In Lrec. Lossio-Ventura J. A., Jonquet C., Roche M., Teisseire M. (2014). Biotex: A system for biomedical terminology extraction, ranking, and validation. In Proceedings of the 2014 international conference on posters demonstrations track - volume 1272, p. 157–160. Aachen, DEU, CEUR-WS.org. Madjarov G., Vidulin V., Dimitrovski I., Kocev D. (2019). Web genre classification with methods for structured output prediction. Information Sciences, vol. 503, p. 551 - 573. Mikolov T., Sutskever I., Chen K., Corrado G. S., Dean J. (2013). Distributed representations of words and phrases and their compositionality. In C. J. C. Burges, L. Bottou, M. Welling, Z. Ghahramani, K. Q. Weinberger (Eds.), Advances in neural information processing systems, vol. 26, p. 3111–3119. Curran Associates, Inc. Miller G. A. (1998). Wordnet: An electronic lexical database. MIT press. Sanh V., Debut L., Chaumond J., Wolf T. (2019). Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108. Santini M. (2011). Cross-testing a genre classification model for the web. In A. Meh- ler, S. Sharoff, M. Santini (Eds.), Genres on the web: Computational models and empirical studies, p. 87–128. Dordrecht, Springer Netherlands. Schaeffer. C., Interdonato. R., Roche. M. (2020). Construction d’un corpus sur la problématique de la sécurité alimentaire guidée par un lexique et des approches de fouilles de textes, in : Toth 2020 - terminologie et ontologie : théories et applications. roche christophe (ed.). chambéry : Presses universitaires savoie mont blanc. SharoffS. (2006). Creating general-purpose corpora using automated search engine queries. WaCky, p. 63–98. Vidulin V., Lustrek M., Gams M. (2009, 01). Bibliographie Aizawa A. (2003, 01). An information theoretic perspective of tf-idf measures. In- formation Processing Management, vol. 39, p. 45-65. Arsevska. E., Roche. M., Hendrikx. P., Chavernac. D., Falala. S., Lancelot. R. et al. (2016). Identification of associations between clinical signs and hosts to monitor the web for detection of animal disease outbreaks. International Journal of Agricultural and Environmental Information Systems. Devlin J., Chang M.-W., Lee K., Toutanova K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. Guéranger D., Mathieu-Fritz A. (2019, novembre). The smart city at work. Reseaux, vol. No 218, no 6, p. 41–75. 175 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Multi-label approaches to web genre identification. JLCL, vol. 24, p. 97-114. 176 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Le présent article porte sur le développement d’une plateforme géoweb dans le cadre du projet de recherche RestEAUr’Lag, qui traite de l’évolution de territoires littoraux méditerranéens, en particulier en Languedoc et en Corse. Il traite de la question de la durabilité de systèmes lagunaires, concentrant des enjeux de ressources, d’usages, de capacités de résilience face à des aménagements déployés sur le temps long, entre les XVIIIe et XXIe siècles. Partant d’un besoin générique de production et de représentation de l’évolution de l’occupation et de l’utilisation des sols de ces territoires sur une période longue, le projet retient l’option d’une plateforme de cartographie en ligne, relevant du géoweb. Le géoweb est « l’organisation de l’information d’Internet à travers un géoréférencement direct ou indirect sur la surface terrestre de tout contenu informationnel » (Joliveau, 2011). Cette plateforme vise à répondre à deux besoins différents et complémentaires : (i) le partage des données et informations issues des travaux des membres du projet de recherche (géographes, historiens, politologues, écologues, etc.) sous une forme cartographique et (ii) la diffusion des résultats de la recherche auprès des acteurs des territoires concernés (gestionnaires et grand public), selon les principes prônés par le bailleur, la Fondation de France. La plateforme doit constituer un outil de sciences participatives permettant au grand public (associations, habitants, scolaires) une appropriation facilitée et de contribuer via des fonctionnalités, dans la logique du géoweb 2.0. Celui-ci peut être défini comme le geoweb qui « hérite des nouveaux principes du Web 2.0 tels que collaboration, participation, partage, échange, ouverture, interaction, communication, et plus encore » (Batita, 2019). Est attendu par exemple un usage en classes de collège ou de lycée afin de préparer des sorties de terrain sur le littoral, pour ensuite compléter ou enrichir les informations déjà présentes par celles récoltées durant ces sorties, avec des exercices narratifs qui assurent l’association du temps et de l’espace des observations. Pour les gestionnaires, il y a l’exemple de services chargés de la gestion des aménagements hydrauliques très intéressés par des modes de valorisations d’informations et d’apports argumentaires spatialisés dans des réunions de concertation, comme dans le cas de l’évaluation des chemins de l’eau en zones inondables et de chemins « oubliés » et redevenus fonctionnels lors d’événement de crues. Marie Gradeler1, Jean-Pierre Chery2 Marie Gradeler1, Jean-Pierre Chery2 1. TETIS, AgroParisTech, Cirad, CNRS, INRAE, Univ Montpellier, Montpellier, France marie.grapdeler@ gmail.com 2. TETIS, AgroParisTech, Cirad, CNRS, INRAE, Univ Montpellier, Montpellier, France 2. TETIS, AgroParisTech, Cirad, CNRS, INRAE, Univ Montpellier, Montpellier, France j i h @ i t h f 2. TETIS, AgroParisTech, Cirad, CNRS, INRAE, Univ Montpellier, Montpellier, France jean-pierre.chery @agroparistech.f r RÉSUMÉ. Un géoweb concernant le littoral méditerranéen français est développé pour valoriser des travaux et résultats de recherche portant en particulier sur les changements d’occupation et d’usage de sols sur le temps long. La conception de cette plateforme géoweb, nommée NAMO (narration et modélisation), mobilise des fonctionnalités ouvertes, libres et gratuites suite à un état de l’art. C’est en particulier le développement de deux dimensions de valorisation et de positionnement en géoweb 2.0 qui est souligné : la cartographie narrative et la modélisation systémique. Cela permet de coupler les facteurs de temporalités variés des territoires considérés et la complexité des interactions entre sociétés et environnement, ainsi que de placer le dispositif en capacité d’assurer un croisement des connaissances scientifiques expertes et des usages citoyens. ABSTRACT. To highlight the work and results of a research project on the French Mediterranean coastline, especially those regarding the changes in land occupation and use over time, a geoweb platform called NAMO (narration and modelling) is being developed. This platforms’ design mobilizes free and open-source functionalities following a state-of-the- art. This paper focuses on the development of two dimensions of valorisation and positioning in geoweb 2.0: narrative mapping and systemic modelling. This makes it possible to couple the various temporality factors of the considered territories with the complexity of the interactions between society and environment, as well as the right incorporation of the platform to ensure a crossover of expert knowledge and citizen use. MOTS-CLÉS : Géoweb, Cartographie narrative, modèle systémique, DPSIR, littoral KEYWORDS: Geoweb, Narrative cartography, system modeling, DPSIR, coastal area 177 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1 L’Évolution de Paris du Bas Empire à 1900 ; OREME – Portail cartographique ; Picto – Occitanie – D6DAMC : Analyse multicritères ; DATA Portraits Paris / Grand Paris ; USGS Historical Topographic Map Explorer ; ALPAGE – AnaLyse diachronique de l’espace PArisien : approche GEomatique ; AHL – Atlas Historique du Limousin ; IGN – Remonter le temps ; Chronas ; CEREMA – Rivages ; Géolittoral – Carte dynamique (CEREMA) ; Lost Rivers Walks ; Observatoire des Territoires ; IGN – Géoportail ; SURFRIDER – Fondation Europe ; Global Forest Watch ; Mobiliscope. 2. Méthode : conception de la plateforme géoweb Les besoins du projet de restituer des aspects de la complexité territoriale et des évolutions dans le temps appellent à dépasser le simple déploiement d’un service cartographique de type SIG en ligne. La méthodologie générale retenue s’appuie d’abord sur un état de l’art des fonctionnalités existantes dans différents sites proposant une cartographie en ligne. Les fonctionnalités attendues doivent traiter des deux besoins méthodologiques particuliers de valorisation des connaissances dans le temps, avec des cartes narratives, et de modélisation de la complexité des territoires littoraux. La sélection de fonctionnalités permet alors la conception d’une maquette pour validation. 1. Introduction Du fait du double usage scientifique et grand public, la plateforme doit proposer une aide à la modélisation qui soit compatible avec un niveau d’exigence experte et un niveau d’approche de néophyte. La notion d’indicateur possède a priori ces deux dimensions afin de traiter des enjeux entre la société et l’environnement, et en particulier les indicateurs conçus dans la perspective du développement durable (Decamps et Vicard, 2010). Le géoweb offre des possibilités de communication et d’interactivité explorées et adoptées par de nombreuses plateformes d’institutions et de projets publics et privés (Laurini et Servigne, 2011). Beaucoup de sites de cartographie en ligne relèvent des technologies directement proposées dans les applications SIG et l’ergonomie en hérite directement : vue cartographique 178 178 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] NAMO Géoweb : narration et modélisation pour les territoires littoraux principale, volet de liste et d’ordre des couches utilisées, paramétrage des propriétés de visibilité et d’apparence en symboles graphiques de ces dernières, etc. (Mericskay, 2016). Des outils de mesure, de sélection, de saisie accompagnent ces fonctionnalités génériques. La plupart des plateformes géographiques de collectivités territoriales, par exemple en France, donnent à voir tout ou partie de leur patrimoine de données géographiques associées aux couches de référentiels mises à disposition par les organismes assurant la couverture géographique standardisée comme l’IGN. Ces plateformes sont d’abord des interfaces de consultation des infrastructures de données géographiques (IDG) et permettent généralement des téléchargements de couches après la vérification de leurs caractéristiques topologiques et tabulaires, ainsi que des sorties cartographiques adaptées au besoin particulier de l’utilisateur. Les choix méthodologiques de conception et de réalisation de la plateforme sont présentés, avec les principales fonctionnalités permettant d’assurer les usages liés aux questions de temporalité et de complexité du territoire littoral selon les possibilités d’usages de type géoweb 2.0. 2.1. Etat de l’art des critères de choix du type de plateforme Un état de l’art a été réalisé sur la base d’un échantillon de 17 cartographies web interactives consultables en ligne en 20201. Il a permis de lister les principales caractéristiques de ces plateformes selon différents critères. Deux critères portent sur les aspects de complexité, via le traitement multi-thématique, et de temporalité. Trois autres critères relèvent d’aspects d’ergonomie, de partage liés au positionnement en sciences participatives. Ces cinq critères sont : 179 Communications 179 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 – le critère de dimension thématique. Il porte sur la représentation cartographique de territoires de types régionaux, dans leurs aspects anthropiques et environnementaux, sur le sujet des littoraux particulièrement, avec l’entrée privilégiée de l’occupation et l’utilisation du sol et des ressources liées à l’eau ; – le critère de dimension temporelle. Celui-ci vise à estimer les possibilités et contraintes pour la représentation diachronique, multi-dates et les évolutions, stagnations et changements dans le temps, avec importance marquée pour l’approche historique. La représentation sous forme de frises chronologiques et d’interactivité de la plateforme par la dimension temporelle sont particulièrement considérés ; – le critère de la dimension fonctionnelle. Il relève d’une part du besoin assez générique d’une accessibilité, d’une gestion et d’une représentation cartographique de type SIG (multi-couches, vecteurs et rasters, sémiologie graphique adaptée), et d’autre part de choix technologiques de développement de type libre, ouvert, gratuit et référencé ; – le critère d’interface. Il propose de s’intéresser à la capacité de la plateforme d’offrir un usage intuitif, une ergonomie dans l’accès et la manipulation des éléments de configuration, d’identité éditoriale et de charte graphique et d’équilibre général des fonctions et éléments affichés ; – le critère d’accessibilité. Ce dernier concerne la capacité de faciliter l’utilisation de l’interface par un utilisateur non averti, en soulignant des fonctionnalités pédagogiques ; Le choix des plateformes analysées est issu d’une recherche par mots-clés sur un moteur de recherche et de liens éventuels repérés en références dans les revues ou sites spécialisés dans le domaine. En l’absence d’un observatoire du géoweb, même si des travaux commencent à structurer un panorama, en considérant d’abord les IDG (Noucher et al., 2019), c’est donc une approche partielle qui est adoptée. 2.1. Etat de l’art des critères de choix du type de plateforme La variété des fonctionnalités repérées permet de retenir celles considérées comme pertinentes pour le projet, et d’estimer les priorités de développement et d’ajouts selon les contraintes de temps, de sécurité et de gestion de la plateforme. La difficulté apparait dans l’équilibre de ces fonctionnalités à retenir, afin d’assurer une ergonomie efficace. Ces fonctionnalités ont permis de définir l’architecture générale de la plateforme pour réaliser une maquette fonctionnelle à l’issue d’un travail de six mois dont trois de développement. 2.2. Architecture et fonctionnalités retenues Suite à l’état de l’art, quatre volets sont retenus pour l’interface afin assurer un accès structuré à l’information géographique disponible selon les principes classiques de la cartographie en ligne, et une logique éditoriale issue des atlas : les zones d’études, les fonds de cartes, les couches et les graphiques. Les atlas généralistes édités sur papier au XXe siècle ou les atlas scolaires actuels possèdent ces éléments : les zones d’études relèvent du plan d’un atlas, en continents, pays, régions comme autant de cas particuliers à distinguer dans une page, une vue Communications 2 Projet ANR COMMUNES à partir de 2020 ; Projet « Forêts de Cassini » depuis 2012 ; BD Carto Etat-major version 1.0 de l’IGN, depuis 2019. 2.3. Intégration des dimensions temporelle et systémique Deux volets spécifiques ont été développés pour proposer deux options qui relèvent du géoweb 2.0 dans ses dimensions multimédia, dynamique et d’interaction avec l’utilisateur : un volet de cartographie narrative et un volet de modélisation selon un cadre conceptuel systémique. Ces volets ont été conçus avec les mêmes méthodes et outils que les quatre volets précédents. Communications 180 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] NAMO Géoweb : narration et modélisation pour les territoires littoraux cartographique. Les fonds de cartes de ces atlas sont généralement des cartes de relief et d’hydrographie, avec les toponymes et des villes et routes principales et des lieux remarquables. Les couches du géoweb sont autant de couches SIG permettant des combinaisons afin de produire des cartes thématiques et des sélections tandis que les graphiques sont des éléments d’accompagnement importants de la cartographie (diagrammes divers, histogrammes, etc.). Une partie de ces éléments sont ceux des SIG en ligne habituels : l’utilisateur gère l’information par sélection et paramétrages d’affichage (zoom, visibilité et transparence). Ces quatre volets sont conçus via des méthodes et outils de structure et de style (HTML et CSS), d’interactivité (JavaScript, JQuery, JQuery UI, OpenLayers et D3.js) et de développement de la base de données et du serveur (QGIS, PostgreSQL et GeoServer). 3 http://www.lostrivers.ca/ [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] géoweb renforce l’intérêt d’un cheminement narratif, depuis un récit dans le temps et l’espace avec l’association dynamique de différentes vues (de la carte, de textes et d’autres produits multimédias). Les éléments associés sont coordonnés à la vue cartographique principale affichée (Caquard et Dimitrovas, 2017). Le projet « Lost Rivers Walks »3 est un exemple de l’intérêt des cartes narratives pour aborder la question des relations entre les aspects anthropiques et environnementaux d’un territoire. Ce projet documente des cheminements dans la métropole de Toronto avec des affichages cartographiques s’adaptant aux étapes d’un itinéraire préalablement établi, avec des éléments significatifs d’aménagement, de règlements d’urbanisme et autres. Ceux-ci expliquent ou illustrent les changements dans les écoulements urbains de l’eau, et finalement la disparition des rivières du paysage, par des aménagements de couverture, dérivation ou comblement. Cela invite à des actions de type « sorties de terrain », préalables ou postérieures à une consultation de la plateforme. Dans notre projet, cette perspective de fonctionnalité est bien retenue comme élément de contribution aux sciences participatives. Cette même fonctionnalité, appliquée dans le cadre de réunions au sein des collectivités pour présenter la logique d’un projet d’aménagement ou de renaturation dans le temps long apparait pertinente. L’alimentation d’une cartographie narrative dans la plateforme consiste d’une part à fournir un service de dépôt de contenus que l’auteur de la narration souhaite combiner (textes du récit, images, cartes, schémas, photographies qui ponctuent la narration), avec les paramètres d’emprise cartographique du géoweb hébergeur (centre de la carte et niveau de zoom) et à intégrer d’autre part cette base de données dans le serveur pour son accès depuis le volet de carte narrative. Le service de dépôt permet de proposer des activités de contributions variées, avec des productions scientifiques d’historiens et autres experts qui replacent les étapes d’une évolution de relations homme-environnement selon les concepts, objets et données de leurs disciplines. 2.3.2. Cartographier les éléments du système territorial via le cadre conceptuel DPSIR 2.3.2. Cartographier les éléments du système territorial via le cadre conceptuel DPSIR Dans le projet, la dimension systémique du territoire littoral est traduite sous forme cartographique. Chaque élément du système étudié disponible dans les couches géographiques du géoweb doit se voir affecté d’un attribut de sémiologie graphique (couleur, trame, symbole) qui renvoie à son statut dans les chaines causales. La représentation cartographique vise à restituer les lieux selon leur rôle (localisation de phénomènes impactant ou localisation d’entités impactées). Par exemple, on peut différencier la distribution spatiale d’éléments émetteurs de pollution de l’eau (rejets d’eau usées de maisons individuelles non raccordées ou émetteur en mer d’une station d’épuration) et d’éléments impactés par ces pollutions (parties de lagunes ou portions du littoral maritime). Les notions de concentration ou de diffusion spatiale des conséquences d’un système d’eaux usées peuvent ainsi être représentées. Communications 2.3.1. Aborder la variété des temporalités : la carte narrative comme approche Dans le temps d’un projet de recherche, une bonne partie des données sont souvent en cours de production et leur exploitation ne peut être envisagée que dans une seconde phase pour la cartographie en ligne. Les pistes méthodologiques et techniques pour traiter les dynamiques et les changements peuvent relever du domaine de la cartographie animée (Cunty et al., 2017). Depuis quelques années, des plateformes émergent dans les domaines de l’histoire et de l’archéologie, aux besoins marqués en bases spatio-temporelles (Costa, 2012). En France, différents programmes de standardisation de couches d’informations géographiques issues de sources des XIXe et XXe siècles en particulier, sont en cours2. Ils donnent une perspective de mise à disposition de ces informations afin de les traiter selon les méthodes d’analyse et de représentation des changements. En attendant ces bases de données historiques et leur exploitation en cartographie animée des changements, une autre méthodologie est retenue : la cartographie narrative, approche assez souple pour alimenter la plateforme géoweb de manière autonome. La plateforme doit valoriser les travaux d’historiens spécialisés dans l’environnement et les aménagements régionaux et locaux. Ces historiens produisent des récits qui reconstituent des situations temporelles, avec l’appui de sources primaires, issues d’archives et des travaux dans lesquels le texte est prédominant. La cartographie narrative possède des fonctionnalités appropriées pour mettre en valeur les dimensions spatiale et temporelle des récits (Caquard et Cartwright, 2014). Le 181 Communications 181 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] NAMO Géoweb : narration et modélisation pour les territoires littoraux NAMO Géoweb : narration et modélisation pour les territoires littoraux NAMO Géoweb : narration et modélisation pour les territoires littoraux Une relation systémique oblige d’une part à déclarer un ensemble causal minimal (cause et effet) et nécessite d’autre part d’avoir une cohérence causale (une cause ne peut être qu’antérieure ou contemporaine à sa ou ses effets). Dans la plateforme, cette relation vise à être traitée par la constitution de tables de relations maillées, qui permettent d’utiliser les métadonnées des dates ou périodes de validité des informations contenues dans les couches mises en relation. Le volet de modélisation systémique doit avoir une cohérence avec les autres volets et aux usages qu’ils proposent. Afin d’établir pratiquement les structures causales, le cadre conceptuel « Driver, Pressure, State, Impact, Response » (DPSIR) est retenu. Il est utilisé par la recherche et les politiques publiques pour identifier des indicateurs selon les principes systémiques causaux : c’est un cadre de modélisation à la fois répandu, régulièrement utilisé et bien documenté (Binder et al., 2013). Les trois éléments centraux de la chaîne de causalité DPSIR sont (i) les pressions qui pèsent sur (ii) l’état du système que l’on souhaite représenter, et (iii) les réponses qui y sont apportées. Les pressions engendrent des modifications de l’état du système. On pense généralement à l’augmentation des activités humaines, la densification de la population, et dans le cas du projet de recherche, le développement de l’habitat pavillonnaire sur le littoral ou bien l’augmentation des rejets d’eaux usées qui en découle. La concentration en éléments polluants est ainsi plus importante. La société met alors en place des réponses correctives pour atténuer les effets non désirés des pressions sur l’environnement. Elles prennent généralement la forme de législations, d’aménagement du territoire, d’espaces de protection, etc. Dans le cas d’illustration, une des réponses apportées est le développement d’un réseau d’assainissement des eaux. En complément, les forces motrices permettent d’améliorer la notion de pression en renseignant celles qui sont indirectes ou globales. Elles correspondent généralement aux grandes tendances sociétales dont les effets sont soit d’accroître soit de mitiger les pressions exercées sur l’environnement. Dans le cas de l’exemple, une force motrice à l’origine de la densification de la population et du développement de l’habitat pavillonnaire sur le littoral est l’attrait pour le littoral méditerranéen et pour son climat. Communications 182 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 4 On peut citer : IGN (BDCarto Etat-Major, BDOrtho Historique, BDOrtho, BDParcellaire, BDTopo, France Raster, OCS Grande Échelle, PVA Historiques, RPG, Scan 50 Historique, Scan25, Scans Cassini, Scans EM40K) ; SHOM (Litto 3D, MNT Bathy), Sandre (BDCarthage, ROE, Stations hydrographiques, Système de traitements) ; IFREMER (Photos anciennes, Surveillance littorale) ; Géolittoral (IPLI, LITTOMOS, Ortho Littoral) ; AD (Cadastre napoléonien, Plan terrier) ; ESA Copernicus (Corine Land Cover, Images Sentinel 2). [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] sélectionner une représentation de la concentration en polluants dans l’eau des lagunes. Il est libre de construire son modèle systémique autour de cet élément en sélectionnant d’autres couches de données et leur adjoignant un attribut DPSIR. Comme le modèle est produit par l’utilisateur, une fonctionnalité doit assurer qu’il ne peut pas définir plus d’une couche comme étant l’état du système, et qu’une contrainte temporelle s’applique sur son modèle. Une couche A ayant une date ou une période strictement plus récente que celle d’une couche B ne pourra pas être déclarée avec un attribut en amont dans la chaîne causale DPSIR. Par exemple, l’indicateur d’eutrophisation de 2000 de la couche des points de surveillance de l’étang de l’Or ne pourra pas être un impact si la quantité d’azote consommée par les exploitations agricoles du bassin versant de l’Or en 2010 est déclarée comme pression. La perspective est de permettre aux utilisateurs d’enregistrer leurs modèles en ligne dans le but de créer une bibliothèque distante. Elle pourrait fonctionner comme une base de données capable d’alimenter les travaux des chercheurs souhaitant proposer un modèle global par contributions d’acteurs pour le territoire littoral. NAMO Géoweb : narration et modélisation pour les territoires littoraux Les impacts permettent quant à eux de préciser l’état du système en lui affectant une notion d’enjeux : les concentrations importantes en polluants (comme état de système) dans l’eau des lagunes modifient entre autres la santé des êtres vivants qui consomment cette eau, directement ou indirectement. Ce risque ou ce dégât pour la santé et l’environnement conduit à d’éventuelles réponses correctives par la société. La méthode retenue pour développer le volet de modélisation est la suivante : chaque attribut DPSIR (forces motrices ; pressions ; état ; impacts ; réponses) est considéré comme un conteneur pouvant accueillir un ou plusieurs indicateurs. Ces indicateurs sont dans notre cas représentés par les jeux de données géographiques, intégrés à l’application en tant que couches. Ces éléments, mis à disposition de l’utilisateur, lui permettent de constituer son propre modèle DPSIR à partir des données auxquelles il a accès, sur la base d’un graphe causal vierge préétabli. Il peut ainsi définir un état unique autour duquel construire son modèle. Cet état prend la forme d’une couche de données : pour reprendre l’exemple précédent, il peut 183 Communications 183 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3. Résultat : la plateforme NAMO Géoweb, à dimensions narrative et de modélisation systémique La plateforme Géoweb est appelée NAMO, pour « Narration et Modélisation ». La plateforme a été testée d’abord en accès local pour des raisons de sécurité. On ne détaille pas ici la solution déployée sur le serveur et la configuration serveur-client, qui relève des dispositions standards pour les services de cartographie en ligne et de bases de données distantes. NAMO Géoweb est développée en utilisant les langages HTML, CSS et JavaScript. L’emploi du JavaScript apporte de l’interactivité principalement via l’API DOM et la bibliothèque JQuery. Par ailleurs, la librairie D3 facilite l’affichage de graphiques en permettant d’écrire dynamiquement un SVG. La figure 1 présente les cinq volets de l’interface géoweb, avec les éléments affichés pour la consultation, le choix et l’activation de traitements associés. Le premier volet du choix de la zone d’étude est construit en proposant des miniatures d’emprises auxquelles sont associées un couple coordonnées / niveau de zoom dans le script de l’application. Le deuxième volet, pour le choix du fond de carte, utilise les flux WMS pour chaque source retenue : OpenStreetMap et l’imagerie satellite d’ESRI accessibles librement, ainsi que les fonds IGN qui sont récupérés via une clé d’identification pour le bouquet de flux utiles au projet. Différents scripts permettent de récupérer ces flux cartographiques et, associés aux miniatures affichées dans le menu, de contrôler leur affichage. Le volet des couches géographiques permet la gestion de leur affichage et mobilise de nombreuses fonctionnalités ainsi que des scripts d’automatisation pour simplifier l’alimentation de l’application en données de sources diverses. L’approche orientée objet d’OpenLayers permet d’intégrer différents widgets, comme par exemple l’opacité des couches, la légende des couches issues de flux WMS ou le choix de dates pour afficher des séries temporelles. Le volet des graphiques et diagrammes a été développé en permettant une lecture diversifiée des données présentées dans les couches affichées. Par exemple, une couche des populations communales par année de recensement couvre une trentaine de millésimes entre 1793 et 2016 et permet l’affichage d’un graphique représentant la variation temporelle de la population pour une commune donnée. Le volet de la carte narrative a été testé en traitant un ensemble d’éléments fournis par un chercheur historien, à propos d’étapes de l’histoire de la construction des canaux sur les étangs palavasiens au XVIIIe siècle et des conséquences négatives qui ont pu en découler ultérieurement pour l’environnement et la société locale. 2.4. Base de données et collecte des données La majorité des données concernées dans la plateforme géoweb sont celles des organismes publics français et de services et observatoires d’intérêt pour la thématique littorale4. Elles visent à être consultables en flux externes via les fonctionnalités retenues. Ces données peuvent être en formats tableurs, vectoriels et raster, flux externes WFS et WMS. Le stockage des différentes sources de données a été conçu en fonction de l’architecture serveur en place dans l’organisme qui accueille ce travail de développement et de recherche, à savoir une base de données PostgreSQL et GeoServer, en prenant en compte la migration des données lorsque la nouvelle IDG locale concernée sera mise en place. Les données originales du projet de recherche n’ont pas encore été mises à disposition pour la plateforme, considérant que la plupart seront offertes en open data, selon les principes de la science ouverte. Elles portent par exemple sur la représentation des différentes classes d’occupation et utilisation du sol en mode vectoriel, à différentes dates. Par ailleurs, les informations du cadastre napoléonien sont traduites par la vectorisation et le géoréférencement de 18600 parcelles pour 11 communes littorales languedociennes. 184 Communications 184 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 NAMO Géoweb : narration et modélisation pour les territoires littoraux 4. Discussion Les possibilités offertes par les méthodes et outils du géoweb, en termes de facilité de conception et d’interactions entre membres d’un projet de cartographie en ligne, soulignent l’intérêt du design géoweb. Ici, il semble important de souligner les possibilités du webdesign dans des conceptions libres, évolutives et adaptables pour conserver une même plateforme de cartographie en ligne en lui assurant des propriétés de capitalisation d’expériences d’usage. Les propositions méthodologiques et techniques de notre plateforme visent à souligner l’importance du lien entre l’information et la connaissance. Les deux développements de cartographie narrative et de modélisation systémique suggèrent cette intégration, via les couches d’information généralement utilisées en mode « SIG en ligne ». Cela ouvre à notre avis des possibilités de travaux d’expérimentation et d’analyse pour des situations d’usages multi-acteurs, tant experts scientifiques que représentants de la société civile et du grand public en général. Les contributions qui peuvent être déposées sur la même plateforme alimentent la co-construction de la connaissance et de son partage. Des dispositifs d’expérimentation sont alors à engager dans le cadre des sciences citoyennes et pourront contribuer aux débats et enjeux de ce type de dispositifs engagés depuis le début de la décennie 2010 (Newmann et al., 2010 ; Mericskay et Roche, 2011). Les enjeux de connaissances thématiques des milieux littoraux, et dans le cas de notre projet, des territoires lagunaires méditerranéens, constituent un domaine riche pour mettre en œuvre un géoweb des relations entre les sphères anthropiques et environnementales. Pour vérifier la robustesse conceptuelle et technique de la plateforme, celle-ci est en cours de réplication dans un autre domaine de recherche, concernant un système territorial marqué par les enjeux de pression de l’agriculture sur l’environnement. Des ateliers d’animation et de débat déjà réalisés pourrons être mis en valeur et discutés en soulignant l’importance des localisations des parties prenantes et des éléments du territoire qui les placent en interaction. 3. Résultat : la plateforme NAMO Géoweb, à dimensions narrative et de modélisation systémique Il est possible de modifier le fond de carte et les couches affichées pour chaque étape de l’histoire. Un script d’automatisation de l’affichage des histoires a été développé afin d’épargner une gestion lourde d’alimentation de contenus par les administrateurs de l’application. 185 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] FIGURE 1. Principe multi-volets de NAMO géoweb : (a) volets « Zones d’études », « Fonds de cartes » et « Couches » ; (b) volet « Graphiques » avec la vue cartographique thématique associée ; (c) Consultation des informations d’une carte narrative (volet multimédia de textes et figures) avec vue cartographique associée et (d) volet de modélisation (source : auteurs) FIGURE 1. Principe multi-volets de NAMO géoweb : (a) volets « Zones d’études », « Fonds de cartes » et « Couches » ; (b) volet « Graphiques » avec la vue cartographique thématique associée ; (c) Consultation des informations d’une carte narrative (volet multimédia de textes et figures) avec vue cartographique associée et (d) volet de modélisation (source : auteurs) Enfin, le volet de modélisation offre une interface permettant d’élaborer une carte thématique composée d’éléments selon les cinq catégories du modèle DPSIR. Ce volet est composé de deux parties. La première est un graphe causal du modèle DPSIR avec ses cinq catégories représentées par défaut avec des cases de couleurs différentes. L’utilisateur choisit des couches ouvertes en session, avec les champs des attributs associés, pour chaque case du modèle afin de produire la carte thématique DPSIR. La deuxième partie du volet est un bandeau d’affichage du test de la cohérence causale donne le résultat des affectations par un message de 186 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] NAMO Géoweb : narration et modélisation pour les territoires littoraux validation ou d’invalidation du modèle avec le diagnostic des erreurs repérées. Ce volet n’est pas encore testé : il constitue une dimension exploratoire de la plateforme NAMO Géoweb. validation ou d’invalidation du modèle avec le diagnostic des erreurs repérées. Ce volet n’est pas encore testé : il constitue une dimension exploratoire de la plateforme NAMO Géoweb. Remerciements Le projet RestEAUr’Lag, dans lequel se positionne ce travail, est soutenu financièrement par la Fondation de France, sur la période 2019-2021, dans le cadre de son programme Littoral. 5. Conclusion Dans le projet de recherche, la question d’un passage à un niveau de mise à disposition d’une information construite par les différents axes scientifiques, sur des expertises et des exploitations particulières des données, ainsi que la question d’un meilleur accès aux informations par le plus grand nombre a permis de retenir une posture éditoriale contraignante. Elle vise à fournir des fonctionnalités qui permettent de cheminer dans les étapes d’un raisonnement causal, systémique, spatial et temporel, des thématiques associées aux enjeux des relations entre société et environnement, pour les territoires littoraux et applicables dans d’autres contextes. 187 Communications 187 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Les perspectives offertes par le géoweb 2.0 sont celles de la contribution et de la science participative, entre des enrichissements au fil du temps, la mise en vis-à-vis de controverses et des possibilités d’animations au sein de collectifs variés. Bibliographie Batita W. (2019). GeoWeb 2.0 applications could support the Geodesign process. Revue Int. Geomat., vol 29, n° 3-4, p. 263-285. Binder C.R., Hinkel J., Bots P.W.G., Pahl-Wostl C. (2013). Comparison of frameworks for analyzing social-ecological systems. Ecology and Society, n° 18, p. 26. Caquard S., Cartwright W. (2014). Narrative Cartography: From Mapping Stories to the Narrative of Maps and Mapping. The cartographic Journal, vol. 51, n° 2, p. 101-106. Caquard S., Dimitrovas S. (2017). Story Maps & Co. Un état de l’art de la cartographie des récits sur Internet. Mappemonde, n° 121, 17 p. Costa L. (2012). La construction de référentiels géohistoriques : un enjeu pour l’interdisciplinarité dans les sciences historiques. L’Espace géographique, vol. 41, n° 4, p. 340-351. Cunty C., Mathian H., Groupe Cartomouv. (2017). Les pratiques de cartographie animée pour représenter le changement. Mappemonde, n° 120, 21 p. Decamps M., Vicard F. (2010). Mesurer le développement durable : jeux d’indicateurs et enjeux locaux. Revue d’Économie Régionale & Urbaine, p. 749-771. Joliveau T. (2011). Le géoweb, un nouveau défi pour les bases de données géographiques. L’Espace géographique, vol. 40, n° 2, p. 154-163. Laurini R., Servigne S. (2011). Potentialité du géoweb : vers une nouvelle génération de Sig participatifs. L’Espace géographique, vol. 40, n° 2, p. 109-116. Mericskay B., Roche S. (2011). Cartographie 2.0 : le grand public, producteur de contenus et de savoirs géographiques avec le web 2.0. Cybergeo : European Journal of Geography, Science et Toile, document 552, mis en ligne le 20/10/2011, consulté le 21/02/2021. Mericskay B. (2016). La cartographie à l’heure du géoweb : retour sur les nouveaux modes de représentation spatiale des données numériques. Cartes et géomatique, n° 229-230, p. 37- 50. Newman G., Zimmerman D., Crall A., Laituri M., Graham J., Stapel L. (2010). User-friendly web mapping: lessons from a citizen science website. International Journal of Geographical Information Science, vol. 12, n° 24, p. 1851-1869. Noucher M., Gourmelon F., Claramunt C. (2019). Pour un observatoire des données géographiques du Web. Expérimentation à partir des infrastructures de données géographiques françaises. Revue Int. Geomat., vol. 29, n° 1, p. 9-30. 188 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction En octobre 2018, une hausse du prix du carburant automobile déclenche en France un mouvement spontané de protestations sociales dénommé "mou- vement des gilets jaunes". Le mouvement s’exprime par des blocages de routes, de ronds-points et de nombreuses manifestations. La première manifestation réunit 280 000 personnes le 17 novembre 2018 (acte 1) et surprend par son ampleur (Boyer et al., 2020). Elle sera suivie par de multiples formes de mo- bilisations accompagnées de revendications sociales et politiques. En décembre 2018, face à ce mouvement, le gouvernement français saisit la Commission na- tionale du débat public (CNDP 1) aÞn de lancer une consultation citoyenne sur divers sujets : le Grand Débat National . Cette consultation s’est officiellement tenue du 15 janvier au 15 mars 2019 autour des quatre grands thèmes suivants : transition écologique, Þscalité, démocratie et citoyenneté, organisation de l’État et des services publics. La consultation s’est déroulée sous différentes formes : réunions publiques locales, cahiers de doléances en mairie, conférences avec des citoyens tirés au sort, contributions en lignes sur une plateforme dédiée, etc. L’analyse et la synthèse des contributions en ligne posent plusieurs déÞs. Tout d’abord, elles sont nombreuses. Sur la plateforme en ligne, pour le ques- tionnaire de la Transition Écologique , plus de 134 000 personnes ont rempli 12 champs de texte libre, soit un total de 1,6 millions de textes potentiels 2 à ana- lyser. De plus, les textes recueillis sont de longueurs différentes et présentent une grande diversité (style, syntaxe, orthographe, grammaire, lexique). À la demande du gouvernement français, une première analyse des contributions en ligne est effectuée par l’entreprise OpinionWay 3 immédiatement après la clô- ture du débat (Opinion Way, 2019). Les synthèses proposées offrent une vue d’ensemble du contenu des contributions mais présentent plusieurs limitations. Tout d’abord, la méthodologie utilisée n’est pas complètement documentée (Bellet et al., 2020), ce qui rend difficile sa reproductibilité et la comparai- son avec d’autres travaux. Ensuite, selon les questions analysées, un nombre important de réponses (entre 10 % et 30 %) est écarté. EnÞn, la dimension territoriale inhérente tant au mouvement social à l’origine de la consultation qu’aux thématiques de la consultation a été très peu abordée. Par ailleurs, peu de travaux scientiÞques ont été publiés sur l’analyse de ces contributions. Parmi ces communications, Ploux et al. Grand débat national : une analyse automatique des contributions relatives aux transports Jacques Fize 1, Lucile Sautot 2, Martin Lentschat 3, Ludovic Journaux 4, Mohamed Hilal 5 Jacques Fize 1, Lucile Sautot 2, Martin Lentschat 3, Ludovic Journaux 4, Mohamed Hilal 5 1. INSA Lyon - LIRIS - jacques.Þze@insa-lyon.fr 2. AgroParisTech - UMR TETIS - lucile.sautot@agroparistech.fr 3. Univ. de Montpellier - UMR TETIS - martin.lentschat@umontpellier.fr 4. AgroSupDijon - Lab. Info. Bourgogne - ludovic.journaux@agrosupdijon.fr 5. INRAE - UMR CESAER - mohamed.hilal@inrae.fr RÉSUMÉ. Le Grand Débat National, lancé par Emmanuel Macron début 2019 pour répondre au mouvement social des Gilets Jaunes, a permis de collecter les contribu- tions de citoyens sur la transition écologique via une plateforme en ligne. Dans cet article, nous exploitons le corpus constitué par ces contributions pour identiÞer des zones où les participants demandent le développement de pistes cyclables et d’équipe- ments ferroviaires. Pour cela, nous avons créé un modèle de classiÞcation permettant d’identiÞer les contributions traitant de la thématique du transport et proposé une méthode d’extraction de motifs traduisant les propositions des contributeurs. A l’aide des codes postaux donnés par les contributeurs, nous présentons les motifs extraits sur plusieurs cartes. ABSTRACT. The Great National Debate, launched by Emmanuel Macron at the begin- ning of 2019 to respond to the Yellow Vests social movement, allowed the collection of citizens’ contributions on the ecological transition via an online platform. In this article, we use the corpus constituted by these contributions to identify areas where participants are asking for the development of bicycle paths and railway facilities. For this purpose, we have created a classiÞcation model to identify contributions dealing with the theme of transportation and proposed a method for extracting patterns that reßect the contributors’ proposals. We then represented these patterns on maps, using the contributors’ postal codes. MOTS-CLÉS : Grand Débat National, Motifs, Transport KEYWORDS: Grand Débat National, Patterns, Transport MOTS-CLÉS : Grand Débat National, Motifs, Transport KEYWORDS: Grand Débat National, Patterns, Transport 189 Communications 189 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. https://www.debatpublic.fr/ 2. Les contributeurs ne répondent pas nécessairement à l’ensemble des questions. 3. Les synthèses OpinionWay sont disponibles ici : https://granddebat.fr/pages/syntheses- du-grand-debat 3. Les synthèses OpinionWay sont disponibles ici : https://granddebat.fr/pages/syntheses- du-grand-debat 1. https://www.debatpublic.fr/ 1. Introduction (2020) ont proposé une analyse automatique sur la base de motifs et d’une comparaison du lexique employé dans différentes consultations incluant celles du Grand Débat National. Cet article n’a a priori pas encore été publié dans une revue ou ouvrage à comité de lecture. Messer et al. (2019) ont également proposé une approche qualitative Communications 190 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Grand débat national et transports Grand débat national et transports des contributions relatives à la mobilité. Ce rapport, très riche sur la description des propositions formulées par les contributeurs, ne dit rien sur la dimension spatiale. des contributions relatives à la mobilité. Ce rapport, très riche sur la description des propositions formulées par les contributeurs, ne dit rien sur la dimension spatiale. Notre objectif est de proposer une approche permettant de cartographier les opinions des contributeurs sur une thématique donnée. Cette approche doit être automatique (aÞn de traiter l’ensemble des contributions) et doit capter l’avis des personnes ayant proposé leur contribution. EnÞn, cette approche doit générer des cartes permettant de localiser les opinions exprimées. Pour répondre à ces problématiques, nous proposons un traitement des don- nées reproductible et une analyse des résultats tenant compte de la dimension géographique. Dans cet article, nous présentons les travaux suivants: – Focus sur les réponses de la "transition écologique" et sur une sous- thématique cruciale : les transports. – Création d’un modèle de classiÞcation pour identiÞer les réponses parlant de cette sous-thématique. – L’extraction de motifs dans les données traduisant les propositions émises par les contributeurs. – L’extraction de motifs dans les données traduisant les propositions émises par les contributeurs. – Cartographie des motifs extraits au travers de deux objets : le train et les pistes cyclables. Cet article s’organise de la manière suivante. La section 2, présente l’ap- proche utilisée pour classiÞer les réponses et extraire les motifs d’opinions. La section 3, présente les motifs extraits selon leur fréquence et leur localisation géographique. EnÞn, nous proposons plusieurs discussions en section 4 sur les résultats obtenus, avant de conclure en section 5. Figure 1. Processus d’extraction des motifs d’opinions Figure 1. Processus d’extraction des motifs d’opinions 191 Communications 4. https://granddebat.fr/pages/donnees-ouvertes 5. https://www.data.gouv.fr/fr/datasets/base-officielle-des-codes-postaux/ 6. Données extraites sur OpenStreetMap https://download.geofabrik.de/europe/france.html 7. https://www.insee.fr/fr/information/2115011 8. Données accessibles ici : https://grandeannotation.fr 9. 31 mai 2020 Communications 191 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 2. Matériel et méthodes Dans ces travaux, nous réalisons une analyse des opinions sur la question du transport dans les réponses de la transition écologique. Ce processus s’effectue en deux étapes : la première (c.f. Section 2.2) identiÞe les réponses mentionnant le transport ; la seconde (c.f. Section 2.3) en extrait les motifs en s’appuyant sur des règles d’extraction et plusieurs lexiques. Le processus est illustré dans la Figure 1. Avant de présenter les méthodes proposées pour chaque étape, nous présen- tons les données utilisées dans l’ensemble du processus. 2.1. Données Dans cet article, nous utilisons les contributions de la plateforme web du Grand Débat National 4. La plateforme proposait deux modalités de contribu- tion : libre ou questionnaire à choix multiples. Seules les contributions libres sont exploitées dans cet article. Plus particulièrement, nous avons uniquement retenu les contributions au volet portant sur la Transition Écologique . Concernant l’analyse géographique, nous exploitons le code postal indiqué par chaque contributeur, sachant que 97 % de ses codes sont exploitables. Les coordonnées géographiques (latitude, longitude) sont extraites de la base offi- cielle des codes postaux 5 et correspondent aux centroïdes. Les réponses sont contextualisées avec des informations externes liées à l’offre de transport. Plus particulièrement, la présence des infrastructures 6 (les pistes cyclables, les gares, les voies ferrées) et l’organisation du territoire par les déplacements domicile-travail (base communale des aires urbaines déÞnies par l’INSEE 7). 2.2. ClassiÞcation des réponses liées aux transports Dans ce travail, nous nous concentrons sur une thématique centrale dans le cadre de la transition écologique : le transport. Pour cela, nous isolons les réponses liées à ce thème en utilisant les annotations réalisées, parallèlement à l’appel à contribution du grand débat, par le collectif nommé Grande An- notation 8. Les réponses du Grand Débat sont annotées à l’aide d’une liste de tags. Au moment du téléchargement des données 9, 280 321 réponses étaient Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Grand débat national et transports Grand débat national et transports annotées. Ce nombre d’annotations est insuffisant pour une analyse globale des réponses du Grand Débat, toutefois, elles sont en nombre suffisant pour entraîner un modèle de classiÞcation. L’objectif du modèle est d’indiquer si une réponse est associée ou pas à la thématique du transport. Pour entraîner ce modèle, nous avons construit un jeu de données d’entraînement/validation composé de deux types de réponses, l’une évoquant le transport et l’autre non. L’identiÞcation des réponses annotées parlant de transport s’opère en sélection- nant celles ayant un des tags faisant partie de la liste présentée dans la Table 1. Une fois les données d’entraînement et de validation extraites, nous avons testé différents algorithmes de classiÞcation. Parmi ceux-ci, nous comparons des approches classiques : Support Vector Machine (SVM) (Vapnik, 2013); Naive-Bayes (Manning et al., 2008); KNN (Altman, 1992); Decision Tree (Loh, 2011) et Random Forest (Breiman, 2001). De plus, nous ajoutons des approches plus récentes venant du deep learning : Bi-LSTM et BERT (Devlin et al., 2019). Table 1. Tags de la Grande Annotation associée à la thématique du transport Transports doux ou moins nombreux Pistes cyclables, sécurisées Commerce de proximité, circuit-court Déménagement ou change- ment travail Obligation, incitation au transport collectif Transport commun gratuit Télétravail Transport commun plus Þable, rapide Covoiturage, auto-partage, stop Pistes cyclables Transport commun moins cher Transport commun plus sûr Transports en commun meilleurs Télétravail, horaires ßexibles Parkings (vélo, auto) à proxi- mité des gares Transports en commun moins chers Transport commun plus proche Développer le réseau ferro- viaire Taxer les gros pollueurs Transport commun plus fré- quent Ramassage scolaire Vélo, trottinette (électrique) Voiture électrique Transports en commun (bus, tram) Voiture plus propre Marche à pied Voiture à hydrogène Voiture autonome Train Voiture hybride A Déplaçons-nous moins Transport ßuvial Transport à la demande Biocarburant (E85, ...) Pour la représentation des données, nous utilisons une représentation bag of words (sac de mots) pour les approches classiques. Pour résumer, une repré- sentation bag of words encode un texte dans un vecteur v = {v0, . . . , vN} de dimension N, où N correspond au nombre de mots uniques présents. Chaque valeur vi du vecteur correspond à la présence d’un terme ou au nombre d’oc- currences de celui-ci dans la réponse. Ici, une valeur correspond au nombre d’occurrences d’un mot pondéré par la formule TF-IDF de (Gerard Salton, 1983). Communications 192 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 2.3. Extraction de motifs associés à une opinion Dans ces travaux, nous nous appuyons sur la déÞnition d’une opinion de (Liu, Zhang, 2012). Une opinion oi est déÞnie selon le quintuplet {oe, op, oc, oa, od}; oi ∈O, où O correspond à l’ensemble des opinions présentes dans une unité de texte (phrase, paragraphe, etc.), oe correspond à l’émetteur de l’opinion, op est le polarisateur, oc la cible de l’opinion, oa est l’attribut de la cible de l’opinion et od est la date à laquelle l’opinion est émise. Par exemple, dans la phrase, "Je trouve que le prix du billet de train est hors de prix.", l’opinion o1 est {"Je", "hors de prix", "billet de train", "prix", ∅}. Dans ces travaux, nous ajoutons à ce quintuplet la localisation géographique de l’émetteur, ol. [ISBN : 978-2-910545-12-1] Avant de transformer les textes en bag of words, ceux-ci sont lemmati- sés. La lemmatisation consiste à remplacer chaque mot par son lemme, c’est à dire la forme canonique d’un mot (e.g aimé →aimer). Dans la Table 2, nous présentons les performances obtenues par les dif- férents algorithmes de classiÞcations sélectionnées. Si les valeurs d’exactitude Communications 193 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] sont proches, ce sont les algorithmes Naive-Bayes et BERT qui présentent les scores les plus élevés. Une fois l’algorithme de classiÞcation ayant produit un modèle, nous utilisons celui-ci sur le reste des réponses des contributions sur la transition écologique. Table 2. Résultat de classiÞcation selon diffŽrents algorithmes Modèle Paramètres Exactitude SVM g=0.1 & kernel=rbf 0.86 Naive-Bayes alpha = 1 0.87 SGD loss=log & penalty = L1 0.86 KNN n = 5 & p =2 0.76 Decision Tree criterion = gini 0.83 Random Forest g=100 & kernel=rbf 0.84 Bi-LSTM 32 LSTM cells 0.69 BERT BERT multilingual base model (cased) 0.87 Table 2. Résultat de classiÞcation selon diffŽrents algorithmes Table 2. Résultat de classiÞcation selon diffŽrents algorithmes Modèle Paramètres Exactitude SVM g=0.1 & kernel=rbf 0.86 Naive-Bayes alpha = 1 0.87 SGD loss=log & penalty = L1 0.86 KNN n = 5 & p =2 0.76 Decision Tree criterion = gini 0.83 Random Forest g=100 & kernel=rbf 0.84 Bi-LSTM 32 LSTM cells 0.69 BERT BERT multilingual base model (cased) 0.87 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Grand débat national et transports Grand débat national et transports changement. La cible de l’opinion oc correspond à un objet appartenant à la catégorie de transport (moyen de locomotion, infrastructure, etc.). Si celle-ci peut être absente, l’attribut oa de l’objet cible oc est caractérisé par un nom (e.g. prix, fonctionnement, rapidité, etc.). La date d’émission od correspond à la date de publication sur la plateforme. La localisation ol est indiquée par le code postal donné par le contributeur. AÞn de traiter de la thématique du transport, les objets cibles oc, les attri- buts oa mais aussi les polarisateurs op sont délimités au sein des trois lexiques suivants : 1. Le lexique du transport. Ce lexique contient les termes désignant des moyens de locomotions (train, vélo, bus, voiture, etc.), des infrastructures (gare, voix de chemin de fer). 1. Le lexique du transport. Ce lexique contient les termes désignant des moyens de locomotions (train, vélo, bus, voiture, etc.), des infrastructures (gare, voix de chemin de fer). 2. Le lexique du changement. Ce lexique contient des termes impliquant une modiÞcation (plus de, changer, augmenter, etc.). 3. Le lexique d’attribut. Ce lexique contient les caractéristiques associées aux termes de transports (prix, accessibilité, quantité, etc.). 3. Le lexique d’attribut. Ce lexique contient les caractéristiques associées aux termes de transports (prix, accessibilité, quantité, etc.). L’extraction de motifs de proposition s’effectue de la manière suivante. Dans un premier temps, nous lemmatisons l’ensemble du document. Ensuite, nous identiÞons les termes des différents lexiques ainsi que leur position dans le document. En s’appuyant sur ces mêmes positions, nous identiÞons les termes qui appartiennent à une même fenêtre dont la taille est déÞnie par la constante tfen. En conséquence, si un terme appartenant au lexique du changement et au lexique du transport apparaît dans une même fenêtre, une opinion suivant un premier motif est considérée (polarisateur + cible). Si un terme du lexique des attributs se trouve entre les deux termes formant ce motif, alors l’opinion suit un second motif (polarisateur + attribut + cible). Dans la section suivante, nous présentons les résultats de l’extraction de motifs obtenus en s’appuyant sur notre méthode et les données utilisées. 2.3.1. Un format d’opinion : la proposition Il existe plusieurs manières d’exprimer une opinion. Parmi celles-ci, on peut trouver le constat et la proposition. Par exemple, la phrase "J’utilise les trans- ports en commun" exprime un constat et "Il faut augmenter le nombre de train" exprime une proposition. Dans la première phrase, la cible sont "les transports en commun", le verbe utiliser est le polarisateur. Dans la seconde phrase, la proposition est d’"augmenter"(polarisateur) le "nombre"(attribut de la cible) de "train"(cible de l’opinion). Le problème du constat réside dans l’ambiguïté du sentiment qu’éprouve un acteur concernant la cible de l’opinion. Reprenant la phrase d’exemple, le contributeur utilise peut-être les transports en commun par contrainte Þnancière et souhaite dans un avenir proche passer à un moyen de transport individuel. Étant moins ambiguë, nous préférons nous focaliser sur la forme d’opinion qu’est la proposition. Dans ces travaux, reprenant le quintuplet de (Liu, Zhang, 2012), l’émetteur oe correspond à l’auteur de la contribution. Le polarisateur op est porté par un verbe (modiÞer, développer, augmenter, etc.) ou adverbe (plus, moins, etc.) de Communications 194 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 3.1. Répartition des motifs obtenus La distribution des motifs selon le polarisateur et objet cible des opinions détectées est présentée dans la Figure 2. Dans ces résultats, plusieurs tendances sont identiÞables. La première concerne le développement des infrastructures telles que les pistes cyclables ou encore les réseaux de transport mais aussi de moyens de locomotions comme le covoitu- rage. Inversement, on observe une tendance négative envers les transports indi- viduels et plus particulièrement la voiture, par exemple, associée très largement au verbe limiter, réduire et diminuer. Cependant, celle-ci est contrebalancée par une troisième tendance en lien avec le remplacement d’un véhicule souvent plus 195 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 économique (diesel) et/ou plus écologique (hydrogène, électrique). On peut ob- server cette dernière tendance avec les verbes tels que "acheter", "changer" et "remplacer". Figure 2. Répartition des motifs selon leur polarisateur et objet ciblé Figure 2. Répartition des motifs selon leur polarisateur et objet ciblé 10. Code. https://github.com/Make-the-Debat-Great-Again/map_generation; Données. https://zenodo.org/record/4270184 3.2. Répartition géographique des motifs obtenus Notre analyse permet de mettre en évidence la localisation des contributeurs qui demandent un développement des pistes cyclables, et des contributeurs qui demandent un développement du réseau ferroviaire et/ou une augmentation de la fréquence des trains. Une échelle d’analyse plus locale semble plus appropriée pour exploiter au mieux ces données. La conférence ayant lieu à La Rochelle, nous avons choisi, à titre d’exemple, de présenter les mêmes données à l’échelle de la Charente- Maritime (Figure 3 et Figure 4). Les contributeurs étant localisés grâce à leurs codes postaux, une analyse Þne à l’échelle de la commune ou de la communauté de commune est impossible. En revanche, il est possible de faire des recommandations à l’échelle des aires urbaines et des départements. Ainsi, en Charente-Maritime, il semble que le développement de pistes cyclables soit demandé par des contributeurs dans les aires urbaines de La Rochelle, Saintes et Saint Jean d’Angély, ainsi que dans les communes de l’axe Rochefort/Saintes/Cognac. Concernant les demandes de développement du réseau ferroviaire et/ou une augmentation de la fréquence des trains, les besoins sont exprimés au centre des aires urbaines de Sainte et de Communications 196 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Grand débat national et transports Grand débat national et transports La Rochelle . On remarque également une demande dans les communes isolées au nord de Saint Jean d’Angély , entre les gares de Saint Jean d’Angély et de Villeneuve-la-Comtesse . Les cartes produites par le croisement des contributions en ligne au Grand Débat National, des aires urbaines et des équipements recensés par OpenS- treetMap permettent donc de proposer des zones de demandes pour certains équipements spéciÞques. Les données et le code nécessaire à la réalisation des cartes sont mis à disposition en ligne. 10 Figure 3. Contributeurs demandant un développement des pistes cyclables en Charente-Maritime et en Charente Figure 3. Contributeurs demandant un développement des pistes cyclables en Charente-Maritime et en Charente 4. Discussions 4. Discussions 4.1. Représentativité des contributeurs Les contributions du Grand Débat National ne sont pas représentatives de la population française. En effet, les résultats que nous présentons dans cet article reßètent l’opinion des contributeurs, mais ne peuvent en aucun cas être considé- rés comme un reßet Þdèle de l’opinion de la population. La méthode d’enquête de ce débat le rend plus proche de la production participative (crowdsour- cing) ou du réseau social, que du sondage. La différence entre les statistiques 197 Communications 197 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 4. Contributeurs demandant un développement du réseau ferroviaire et/ou une augmentation de la fréquence des trains en Charente et Charente Maritime Figure 4. Contributeurs demandant un développement du réseau ferroviaire et/ou une augmentation de la fréquence des trains en Charente et Charente Maritime classiques et l’analyse des réseaux sociaux (et plus généralement l’analyse de contenu généré par les utilisateurs) est bien connue, mais son importance a néanmoins été reconnue même en présence de différents types de biais (Mors- tatter, Liu, 2017). Ainsi, les caractéristiques sociologiques des contributeurs (âge, sexe, catégorie socio-professionnelle, ...) n’ont pas été demandées et sont absentes dans le jeu de données. Il n’est donc pas possible de s’assurer que les contributeurs constituent un échantillon représentatif de la population. Seul le code postal, renseigné par le contributeur, permet d’avoir une idée de la répartition spatiale des contributeurs. La participation n’a pas été homogène sur le territoire national (Bennani et al., 2019). On constate par exemple une sous-représentation des grands pôles urbains (58,4 % de la population contre seulement 27 % des contributeurs) et une surreprésentation du périurbain (24,6 % contre 47,6 %) et des autres communes (17 % contre 25,5 %). De plus, il faut noter que les cartes qui ont été présentées ne montrent pas le nombre de contributions dans lesquelles les motifs ont été détectés. classiques et l’analyse des réseaux sociaux (et plus généralement l’analyse de contenu généré par les utilisateurs) est bien connue, mais son importance a néanmoins été reconnue même en présence de différents types de biais (Mors- tatter, Liu, 2017). Ainsi, les caractéristiques sociologiques des contributeurs (âge, sexe, catégorie socio-professionnelle, ...) n’ont pas été demandées et sont absentes dans le jeu de données. Il n’est donc pas possible de s’assurer que les contributeurs constituent un échantillon représentatif de la population. 4.1. Représentativité des contributeurs Seul le code postal, renseigné par le contributeur, permet d’avoir une idée de la répartition spatiale des contributeurs. La participation n’a pas été homogène sur le territoire national (Bennani et al., 2019). On constate par exemple une sous-représentation des grands pôles urbains (58,4 % de la population contre seulement 27 % des contributeurs) et une surreprésentation du périurbain (24,6 % contre 47,6 %) et des autres communes (17 % contre 25,5 %). De plus, il faut noter que les cartes qui ont été présentées ne montrent pas le nombre de contributions dans lesquelles les motifs ont été détectés. Grand débat national et transports Grand débat national et transports Grand débat national et transports mement, l’utilisation d’une fenêtre glissante n’est pas suffisante pour détecter des motifs plus complexes. Troisièmement, nous ne détectons pas l’emploi de la négation. Au-delà de l’aspect de la méthode, le modèle de la "proposition" s’appuyant sur (Liu, Zhang, 2012) et les motifs choisis ne couvre pas l’ensemble des motifs disponibles. Par conséquent, un travail d’annotation et d’explora- tion des données plus profond sont nécessaires à l’amélioration de ce travail. Des travaux sont menés pour explorer deux nouvelles méthodes d’extractions. Les premiers s’appuient sur l’utilisation de la grammaire de dépendance (ou dependency parsing) (Kübler et al., 2009) dans la formulation de règles d’ex- traction. Les seconds se concentrent sur la construction d’un corpus annoté pour la création d’un modèle de classiÞcation. 5. Conclusion L’extraction de motifs d’opinion des contributions en ligne au Grand Dé- bat National et la cartographie de ces opinions, enrichie de données publiques, permettent d’identiÞer des zones dans lesquelles une demande d’ajout, de déve- loppement d’équipements et d’infrastructures est proposée. Ces contributions recueillies à l’échelle nationale sont donc porteuses d’indications pour les déci- deurs locaux. Elles permettent en effet d’identiÞer des zones où l’investissement dans certaines infrastructures pourrait permettre de répondre aux attentes de la population. La précision spatiale de la localisation des contributeurs, les échelles spatiales les plus intéressantes nous semblent être l’aire urbaine et le département. Ainsi, ces premiers résultats sont encourageants et montrent le potentiel des contributions au Grand Débat National comme une source d’in- formation pour les collectivités locales. Néanmoins, les limites inhérentes aux données (la non représentativité et le faible nombre de contributeurs) doivent nous conduire à utiliser les résultats avec prudence. Ainsi, les zones identiÞées avec notre approche sont des zones où il y a apparemment une demande mais cette demande doit être conÞrmée par une enquête plus approfondie. Par ailleurs, nous avons identiÞé des limites méthodologiques. Les futurs travaux porteront sur l’amélioration de l’extraction d’opinions. Plus particuliè- rement, par la mise en place d’une campagne d’annotation, puis par la construc- tion d’une ressource lexicale (ontologie) plus complète sur la thématique du transport. L’objectif est de produire un corpus annoté permettant d’extraire d’identiÞer de nouveaux motifs et/ou de produire des modèles automatiques en s’appuyant sur des méthodes d’apprentissage automatique. Altman N. S. (1992). An introduction to kernel and nearest-neighbor nonparametric regression. The American Statistician, vol. 46, no 3, p. 175–185. Bellet A., Denis P., Gilleron R., Keller M., Vauquier N. (2020, juin). Pour plus de transparence dans l’analyse automatique des consultations ouvertes : leçons de la 4.2. Couverture des motifs détectés dans les contributions Plusieurs limitations importantes du travail proposé dans cet article se rap- porte à la couverture de la méthode de détection de motifs. Premièrement, notre méthode s’appuie sur un ensemble de lexiques non-exhaustifs. Deuxiè- 198 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] synthèse du Grand Débat National. Consulté sur https://hal.inria.fr/hal-02860659 (working paper or preprint) synthèse du Grand Débat National. Consulté sur https://hal.inria.fr/hal-02860659 (working paper or preprint) Bennani H., Gandré P., Monnery B. (2019). Les déterminants locaux de la partici- pation numérique au grand débat national : une analyse économétrique. Consulté sur https://economix.fr/pdf/dt/2019/WP_EcoX_2019-7.pdf?1.0 Boyer P. C., Delemotte T., Gauthier G., Rollet V., Schmutz B. (2020). Les déter- minants de la mobilisation des Gilets jaunes. Revue économique, vol. 71, no 1, p. 109–138. Breiman L. (2001). Random forests. Machine learning, vol. 45, no 1, p. 5–32. Devlin J., Chang M.-W., Lee K., Toutanova K. (2019). Bert: Pre-training of deep bidirectional transformers for language understanding. Gerard Salton M. M. (1983). Introduction to modern information retrieval. McGraw- Hill. Kübler S., McDonald R., Nivre J. (2009). Dependency parsing. Synthesis lectures on human language technologies, vol. 1, no 1, p. 1–127. Liu B., Zhang L. (2012). A survey of opinion mining and sentiment analysis. In Mining text data, p. 415–463. Springer. Loh W.-Y. (2011). ClassiÞcation and regression trees. Wiley Interdisciplinary Re- views: Data Mining and Knowledge Discovery, vol. 1, no 1, p. 14–23. Manning C. D., Schütze H., Raghavan P. (2008). Introduction to information retrieval. Cambridge university press. Messer M. A., Utz S., Moreau D. (2019). Analyse exploratoire des contributions au grand débat national, volet mobilité. techreport. Forum Vies Mobiles, Paris (France). Consulté sur https://fr.forumviesmobiles.org/sites/default/Þles/editor/ granddebat_rapport-Þnal.pdf Morstatter F., Liu H. (2017, 1er 6). Discovering, assessing, and mitigating data bias in social media. Online Social Networks and Media, vol. 1, p. 1–13. Opinion Way. (2019). Traitement des données issues du grand débat national. Rapport technique. Auteur. Consulté sur https://granddebat.fr/pages/syntheses-du-grand -debat Ploux S., Genay M., Ploux-Chillès L. (2020, janvier). Les mots du Grand débat national : quelques résultats et réßexions préliminaires. Consulté sur https://hal .archives-ouvertes.fr/hal-02441477 (working paper or preprint) Vapnik V. (2013). The nature of statistical learning theory. Springer science & busi- ness media. 200 Communications Vapnik V. (2013). The nature of statistical learning theory. Springer science & busi- ness media. Bibliographie Altman N. S. (1992). An introduction to kernel and nearest-neighbor nonparametric regression. The American Statistician, vol. 46, no 3, p. 175–185. 199 Communications 199 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Communications par Session Modélisation spatio-temporelle [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Graph-based Pattern Detection in Spatio-Temporal Phenomena Kamaldeep Singh Oberoi, Géraldine Del Mondo 1. Introduction The field of Spatio-Temporal (ST) Modeling has seen a growing interest in the last decades, most notably in the development of Geographical Informa- tion Science (GIS). GISystems store and analyze data about real-world entities, highlighting their ST evolution as well as the evolution of the overall phenome- non. Taking Leibnizian point of view, the real-world entities act as the basic units of space. In fact, space is defined in terms of the relations between such entities (Chrisman, 1977). From a conceptual point of view, these distinct en- tities and their time-varying spatial relations represent the ST phenomenon. This point of view leads to the development of a graph model of the pheno- menon (Batty, 2003), where the nodes of the graph correspond to the entities and its edges represent the spatial relations between them. With this approach, the evolution of the phenomenon and its entities is explained in terms of the evolution of the graph, i.e., by modeling change in the attributes of the nodes and the edges and by modeling change in its overall structure. This evolu- tion of the graph makes the temporal dimension of the model explicit. Such a modeling approach has been used for modeling dynamics of street networks (Costes et al., 2015), for describing topological relations between places and events (Maduako, Wachowicz, 2019), etc. in the literature. Representing the structure of the phenomenon using a graph encourages the development of analysis techniques which exploit the structural properties of the phenomenon. In other words, using this methodology, graph algorithms can be directly applied to analyse the characteristics of the phenomenon. One such analysis technique is the graph-based pattern detection. Pattern detection, in general, is useful to analyse the evolution or the behaviour of the entities of the phenomenon. Depending on the context, patterns might represent the evolution of environmental data (Tsoukatos, Gunopulos, 2001), spatio-temporal proximity of trajectories of moving objects (Demšar, Virrantaus, 2010) etc. However, in these cases, since the representation of the phenomenon does not take into account its underlying structure, the patterns detected are based on the quantitative data acquired about the phenomenon. Such patterns are called statistical patterns since statistical techniques are used for their definition and detection. When the underlying structure of the phenomenon is modeled, the concept of structural patterns comes into picture. Kamaldeep Singh Oberoi, Géraldine Del Mondo Normandie Univ, INSA Rouen, LITIS, 76000, Rouen (France) kamal.oberoi@insa-rouen.fr,geraldine.del_mondo@insa-rouen.fr RÉSUMÉ. Les modèles spatio-temporels (ST) sont souvent utilisés pour analyser les phénomènes ST. Une de ces techniques d’analyse consiste à détecter des motifs dans le phénomène pour comprendre son évolution et modéliser le comportement de ses en- tités dans l’espace et le temps. Dans cet article, nous nous concentrons sur l’utilisation d’une représentation fondée sur les graphes dynamiques pour la modélisation des phé- nomènes ST, au sein duquel des motifs structurels, également modélisés sous forme de graphes dynamiques, peuvent être détectés. Nous illustrons le concept de motif par deux exemples applicatifs - le trafic routier et les sports collectifs. Pour chacun d’eux, nous présentons une modélisation du graphe ainsi que des motifs associés. Ensuite, nous formalisons le problème de la détection de motifs comme celui de l’isomorphisme de sous-graphe pour les graphes dynamiques. Enfin, nous présentons les résultats de notre algorithme pour résoudre ce problème. Les premiers résultats décrits dans cet article, qui sont obtenus en utilisant les graphes aléatoires, présentent une base de référence pour les futurs tests de l’algorithme. ABSTRACT. Spatio-temporal (ST) models are often used for analyzing ST phenomena. One such analysis technique is to detect patterns in the phenomenon to understand its evolution and model the behaviour of its entities over space-time. In this paper, we focus on using a dynamic graph-based representation for modeling ST phenomena, within which structural patterns, also modeled using dynamic graphs, can be detected. We illustrate the concept of pattern using two applications - road traffic and invasive team sports. For both these applications, we present the graph model as well as the corresponding patterns. Then we formalize the problem of pattern detection as that of subgraph isomorphism for dynamic graphs. Finally, we present the results of our algorithm to solve this problem. The initial results described in this paper, obtained using random graphs, present a baseline for the future tests of the algorithm. MOTS-CLÉS : Modélisation spatio-temporelle, Graphes dynamiques, Isomorphisme de sous-graphe, Trafic routier, Sports collectifs KEYWORDS: Spatio-temporal modeling, Dynamic graphs, Subgraph isomorphism, Road traffic, Invasive team sports 203 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction Such patterns take a qualitative modeling approach and are defined in terms of the distinct entities and their corresponding relations which make up the ST phenomenon. In this paper, we consider the notion of structural patterns and describe a method for their detection in ST phenomena. First, we consider two appli- cations - urban road traffic and invasive team sports (specifically Handball) - to demonstrate the conceptual idea behind the structural patterns within a phenomenon. In case of road traffic, the patterns represent the movement of dynamic entities, like vehicles and pedestrians, and in case of handball, they re- present predefined plays of the attacking team. The considered ST phenomena Communications 204 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Graph-based pattern detection in ST phenomena (road traffic and handball) as well as the structural patterns are modeled using dynamic graphs with time-varying edge sets. For now, we do not consider the change in the attributes of the nodes or the edges. Next, we describe the problem of pattern detection in a ST phenomenon in terms of the well known problem of subgraph isomorphism (SI) in graph theory, where a given pattern graph has to be detected in a given target graph. SI has been well studied in case of static pattern and target graphs (cf. (Conte et al., 2004)). However, we are currently developing an algorithm to extend it for dynamic pattern and target graphs. In this paper, without going into details of the algorithm, we will present some initial results in terms of its runtime, which were computed using random dynamic pattern and target graphs. By using random graphs, we were able to modify various graph parameters, for both pattern and target graphs, to formalize the performance of our algorithm. The results presented in this paper provide us with a baseline which will be helpful when the algorithm will be applied to real data about different ST phenomena in the future. The paper is organised as follows. Section 2 presents the related work. In Section 3, we will describe the graph model for representing road traffic and discuss different kinds of structural traffic patterns which can be defined using this model. In Section 4, we follow the same plan and discuss structural patterns in a game of handball. 1. Introduction In Section 5, we formalize the problem of SI in case of dynamic graphs and then discuss the initial results of our algorithm. Lastly, Section 6 concludes the paper and discusses the future work. 2. Related work Graphs have been used to model different kinds of spatial and spatio- temporal phenomena for the last two decades. However, an interesting point of discussion is the type of analysis that has been done using such graph represen- tations. For example, (B. Jiang et al., 2000) proposed a graph representation, called connectivity graph, to model the spatial structure of a building, using which graph parameters, like node degree, were applied for developing a cogni- tive point of view of space. Similarly, graph representations have been used to model street networks to understand the topological structure of the network and the centrality or the importance of a given street (Porta et al., 2006). Incorporating temporal information in the graph model leads to the defi- nition of temporal version of graph parameters. For example, in (Del Mondo et al., 2010), the authors have proposed a spatio-temporal graph model for representing ST phenomena. Since time is incorporated in the graph in terms of its time-varying structure, the neighborhood of a node is time-dependent. The authors have used the term spatio-temporal neighborhood to describe the neighbors of a node at a given instant (spatial) or during an interval (tem- poral). Using this notion, the authors are able to model the propagation of 205 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] brambles through space-time. Furthermore, for understanding the propagation of a disease within a population, it is important to note the contacts between different people over time. This idea has been applied in (Shirani-Mehr et al., 2012) where a dynamic contact network is used to analyze the propagation of a virus. The authors study reachability queries which verify if a path exists between two nodes of the graph. But since the graph is time-varying, the path between two nodes evolves over time. Graph-based pattern detection is also a very well studied research field. Some existing research focuses on frequent pattern mining, where the pattern is not known a priori and a subgraph which exists a certain number of times is considered pertinent (C. Jiang et al., 2013). Moreover, detecting a given pattern graph in a given target graph also has its particular applications (Conte et al., 2004). However, detection of dynamic patterns in dynamic graphs is not very matured research area. 2. Related work With this paper, we want to, first, h ighlight the semantics behind dynamic graph-based patterns in different spatio-temporal phenomena and, second, formalize the problem of pattern detection as that of subgraph isomorphism for dynamic graphs. The experiments performed on random graphs outline the performance of our algorithm. 3. Structural patterns in urban road traffic In this section, we consider the case of urban road traffic. We will first describe, in brief, the formalization of the graph model representing traffic and semantics behind its nodes and edges and then discuss some examples of structural traffic patterns. 3.1. Graph-based model of road traffic In our previous work (Oberoi et al., 2017), a graph for representing urban road traffic was proposed. One of the objectives of this work was to understand the behaviour of dynamic entities of traffic (vehicles and pedestrians) under dif- ferent traffic situations. As a result, we developed a traffic model which takes into account not only dynamic but static entities (buildings, traffic signs, road markings, footpaths, road segments, intersections) as well which affect traffic flow. In addition, the interactions between these entities, in terms of spatial re- lations between them (like topological, orientation, relative trajectory, relative speed, qualitative distance etc.), were also considered. The entities constitu- ting traffic were represented as the nodes and the relations between them were represented as its edges. Given that traffic is a dynamic phenomenon, the mo- del incorporated the temporal dimension as well. The proposed model had the support for snapshot-based approach, where each graph snapshot represented the state of the road traffic at a given time instant. Hence, the evolution of the graph, and in turn of road traffic, was described by a set of distinct graph snap- shots for a given period of time. The model represents the behaviour of dynamic Communications 206 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Graph-based pattern detection in ST phenomena entities in terms of their movement and their interactions with other entities over time. Given that the evolution of traffic is described using a set of distinct graph snapshots, the movement of dynamic entities can also be modeled using the same methodology. This defines the semantics behind the structural traffic patterns discussed below. Figure 1. Pattern representing the movement of a vehicle and a pedestrian Figure 1. Pattern representing the movement of a vehicle and a pedestrian 3.2.1. Vehicle crossing an intersection A basic traffic situation is that of a vehicle crossing an intersection (the green encircled part of the graphs in Figure 1). In this situation, we consider five graph snapshots which represent the location of the vehicle V 1 at a given instant. When the vehicle is on a road segment (at times t1, t2, t4 and t5), an edge between the vehicle and the corresponding road segment node (R1 or R2) exists in those graph snapshots. At t3, the vehicle is on the intersection I, hence, there is an edge between the vehicle and the intersection nodes. At times t2 and t4 , the vehicle is shared between the road segment and the intersection. 3.2. Structural traffic patterns Here, we present two examples of structural traffic patterns describing the movement of a vehicle and/or a pedestrian and their interactions with other entities over time. It is noteworthy that in the model proposed in (Oberoi et al., 2017) edges represent spatial relations between entities but for defining the structural patterns we do not explicitly consider any spatial relations and focus only on the time-varying graph structure. Communications 4. Structural patterns in team sports In the literature, invasive team sports have been represented using graphs to model player interactions (Gudmundsson, Horton, 2017). In this section, while focusing on the game of handball, we will describe a graph modeling approach which takes into account the "role" of the players as well as their positions on the playing field. Then we will describe two examples of predefined plays for the attacking team, modeled using dynamic graphs. Such predefined actions could represent spatio-temporal patterns in other team sports as well. 3.2.2. Vehicle crossing an intersection while pedestrian crossing a street This situation combines the movement of the pedestrian P1 with that of the vehicle V 1 (Figure 1). At time t1, the pedestrian is on the footpath FR1 and 207 Communications 207 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] the vehicle is approaching the intersection. The vehicle waits for the pedestrian at time t3 and once the pedestrian has crossed the street, the vehicle continues moving at time t4. The road marking (zebra crossing) is represented as M2. This example highlights the fact that the movement of dynamic entities gets affected due to the presence of other entities around them and this effect can be represented using the structural representation of road traffic. 4.1. Graph-based model for handball We are working on developing a graph model to represent the game of handball while taking into account the role of each player and their correspon- ding positions for both defending and attacking teams. For the attacking team, there are six roles: Left and Right Backcourt, Center Backcourt, Left and Right Wingers and Pivot. For the defending team, there are seven roles (six defenders and one goal keeper) and we consider "6-0" defending formation. In addition, we consider two zones on the court which describe the position of each player. Free Throw ("FT") zone represents the area between the six meter line and the free throw line whereas Nine metre ("9m") zone represents the area beyond the free throw line towards the middle of the court. These two zones are added as nodes in the graph which has a total of seventeen nodes, fourteen corresponding to the player roles, one representing the ball and two for the zones. The edges between "role" nodes represent the proximity relation, i.e., two players of same or opposite teams are connected if they are close to each other. The exact distance value to define the relation between two role nodes is not yet fixed. An edge between a role and the ball node represents the possession of the ball by the player having that role. The ball node could also be connected to two role nodes representing the action of pass between corresponding players. And finally, an edge between a role and zone node represents the topological relation between the player and the zone. It is added to represent the position of the player on the court. As in the case of road traffic, this graph represents the underlying structure of the game by describing player interactions. It is evident that this graph evolves with time and its evolution can be modeled in terms of its time-varying structure. The progress of the game is represented as Communications 208 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Graph-based pattern detection in ST phenomena a set of discrete states of the graph. Such a graph model can be used to model and extract the predefined actions of the attacking team as discussed below. Figure 2. Switch between CB and LB players in the attacking team (in blue). The defending team is in red and the green star represents the ball Figure 2. 4.1. Graph-based model for handball Switch between CB and LB players in the attacking team (in blue). The defending team is in red and the green star represents the ball 4.2. Structural patterns in handball Similar to road traffic, structural patterns representing player actions in handball can be defined and modeled using dynamic graphs. 4.2.1. Switch between Center Backcourt and Left Backcourt This tactic involves the players with roles of Center Backcourt CB, Left Backcourt LB and Right Backcourt RB in the attacking team (Figure 2). The tactic begins when Center Backcourt is in the possession of the ball at time t1. Then the ball is passed to the Right Backcourt at time t2 . Then Center Backcourt takes the place of Left Backcourt and Left Backcourt takes the place of Center Backcourt leaving Right Backcourt with the ball possession at time t3. Then, according to the game situation, Right Backcourt may pass the ball or take the shot. Here, the roles of Right and Left Backcourt players can be interchanged. 5. Pattern detection in ST phenomena In Sections 3 and 4, we described the graph-based approach for modeling road traffic and invasive team sports respectively. In both cases, the global phenomenon was modeled using a dynamic (target) graph within which some pertinent patterns, also modeled using dynamic (pattern) graphs, could be detected. In this section, we will the formalize the problem of pattern detection as that of subgraph isomorphism (SI). First, we will formally define the problem and then present initial results of the algorithm under development to solve SI in case of dynamic pattern and target graphs. 4.2.2. Crossover between Center Backcourt and Right Backcourt Here, the Center Backcourt and the Right Backcourt are involved in the action. The tactic begins when Center Backcourt is in the possession of the ball at time t1. He/she runs towards the direction of the Right Backcourt and Right Backcourt runs towards the middle of the playing field. When they are 209 Communications 209 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] close enough, the Center Backcourt makes a pass to the Right Backcourt at time t2, leaving Right Backcourt with the ball possession at time t3. Then, depending on the situation, Right Backcourt either shoots or makes another pass. The same action can also be performed by Center and Left Backcourt players as well. 5.1. Problem formulation Consider a static graph G = (X, E, ψ) with a set of nodes X and set of edges E ⊂X × X. Node classification function ψ : X →C assigns a class to each node from a set of classes C. It is possible for two nodes to belong to the same class. For example, in case of road traffic a node class could mean Vehicle, Road marking etc. whereas in case of handball a node class could mean Zone, Player of the attacking team etc. In addition, unique node labels are assigned to each node. Given a pair of static pattern graph H = (XH, EH, ψH) and static target graph G = (XG, EG, ψG), the problem of SI is to find an injective mapping µ ⊂XH ×XG between the nodes of both graphs, while respecting the following conditions: ∀u ∈XH, ∃v ∈XG | (u, v) ∈µ (1) ∀u, u′ ∈XH, u ̸= u′, ∃{(u, v), (u′, v′)} ∈µ ⇒v ̸= v′ (2) ∀(u, u′) ∈EH, ∃(v, v′) ∈EG (3) ∀u, u′ ∈XH, (v, v′) ∈EG, {(u, v), (u′, v′)} ∈µ ⇒(u, u′) ∈EH (4) ∀u ∈XH, ∃(u, v) ∈µ ⇒ψH(u) = ψG(v) (5) (5) Equations 1 and 2 state that all distinct nodes of the H must be mapped to some (but distinct) nodes of G. Equation 3 states that all edges of H are mapped to some edges of G. Equation 4 verifies the "induced" version of SI by checking that a mapped edge of G has a corresponding edge in H, i.e., the subgraph of G matched with H has no extra edges. Lastly, equation 5 states that only the nodes belonging to the same class in both H and G are mapped. 210 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Graph-based pattern detection in ST phenomena Graph-based pattern detection in ST phenomena Graph-based pattern detection in ST phenomena Graph based pattern detection in ST phenomena Figure 3. Snapshots of target graph G and pattern graph H with detected pattern highlighted in G. Set of nodes in both graphs remains constant over time but the isolated nodes are not shown for better readability Figure 3. Snapshots of target graph G and pattern graph H with detected pattern highlighted in G. 5.2. Experiments and results We are developing an algorithm which takes graphs H and G as input and detects all instances of H in G. Our algorithm is based on another algorithm, VF3 (Carletti et al., 2018), which has been developed for static pattern and target graphs. We extend VF3 to include the temporal order for both graphs. We have tested our algorithm using 6050 random pattern and target graph pairs which were generated with different values of graph parameters - Size (N), Density (D), Number of node classes (C) and Number of snapshots (T ), while making sure that at least one instance of the pattern graph exists in the corresponding target graph. We calculated the average runtime and standard deviation (SD) for fifty g raph p airs w ith d ifferent v alues o f e ach parameter, while considering the total time taken, excluding, however, the time required to load each graph from the files. Some of the results obtained are tabulated in Table 1. For denser target graphs, the average runtime increases. This is due to the fact that for denser target graphs, the number of possible node mappings to be verified i s h igher. S imilarly, i ncreasing t he n umber o f s napshots o f the target graph, the runtime increases since algorithm performs more iterations to find the solution. Increasing, however, the number of node classes of the target graph does not have a significant e ffect o n t he a verage r untime. For denser pattern graphs, the runtime also increases since it takes more time to look for the corresponding edges in the target graph to satisfy the equations 3 and 4 (Section 5.1). Increasing the number of snapshots of the pattern graph shows a peculiar behaviour. The runtime keeps increasing with the increase in the number of pattern snapshots since more graphs have to mapped to find the complete solution. However, when the number of pattern snapshots becomes equal to the number of target snapshots (Tt = Tp = 100 in this case) the algorithm terminates quickly. This is due to the fact that the steps required to look for a future snapshot of the target graph are not performed in this case since a snapshot Hi of the pattern graph gets mapped to the snapshot Gi of the target graph. 5.1. Problem formulation Some target graph snapshots (G4, G6 and G7) represent ”noise”. matched as: V 1 → V 100, R4 → R35, I → I15 and R3 → R25. Some target graph snapshots (G4, G6 and G7) represent ”noise”. matched as: V 1 → V 100, R4 → R35, I → I15 and R3 → R25. Some target graph snapshots (G4, G6 and G7) represent ”noise”. 5.2. Experiments and results Similarly to target graph, increasing the number of node classes of the pattern graph does not have any significant effect on the runtime. 6. Conclusion and future work The objective of this paper is to show the applicability of qualitative graph- based representations for modeling ST phenomena. Graphs represent the un- derlying structure of the phenomenon which can be exploited to analyse its spatio-temporal evolution. We focused on detecting patterns within different phenomena. Taking the examples of road traffic and handball, we highlighted the conceptual idea of structural pattern and then formalized the problem of 212 Communications 5.1. Problem formulation Set of nodes in both graphs remains constant over time but the isolated nodes are not shown for better readability Next, we define a simple dynamic graph G = (XG, EG, TG, ψG) as a graph whose set of edges EG is time-varying. Its discrete time domain TG is represen- ted as a set of time instants TG = {t1, t2, ..., tk}. We consider a snapshot-based model where a graph snapshot Gi at time instant ti represents its state. Fol- lowing this, a dynamic graph can be represented as a set of its discrete states, i.e., G = {G1, G2, ..., Gk} (Figure 3). Given a pair of dynamic pattern graph H = {H1, H2, ..., Hm}, having m snapshots, and dynamic target graph G = {G1, G2, ..., Gn}, having n snapshots, with m ≤n, we extend the problem of SI for dynamic graphs, i.e., (1) Each snapshot of H must be mapped to some snapshot of G while respecting the conditions of subgraph isomorphism (Equations 1-5). (2) The time order of H and G must be respected, i.e., if a snapshot Hi, 1 ≤i ≤m of H is mapped to a snapshot Gj, 1 ≤j ≤n of G, then the next snapshot Hi+1 of H must be mapped to some snapshot Gf of G where f > j. (3) To find a complete matching of H, a node v ∈XH must be mapped to the same node v′ ∈XG at every time instant where the matching is possible and both nodes v and v′ exist. (4) It is possible to have some snapshots of G to which no snapshots of H are matched. Such snapshots represent noise. They are included to be realistic in terms of the sampling rate to generate the target graphs in real applications. In Figure 3, the node classes are represented by the first character of their unique labels, i.e., the node having label V 20 belongs to the class V etc. The pattern and target graph snapshots are matched as follows: H1 →G1, H2 →G2, H3 →G3, H4 →G5 and H5 →G8, while the nodes of the pattern graph are 211 Communications 211 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] matched as: V 1 → V 100, R4 → R35, I → I15 and R3 → R25. Acknowledgements The authors would like to thank François Rioult, Quentin Bourgeais and Jeoffrey Girault for providing the data for handball and for multiple discussions related to the detection of predefined plays in handball as part of NormaSTIC Federation. Communications 212 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Graph-based pattern detection in ST phenomena Table 1. Avg. runtime and standard deviation for different graph parameters Pattern Graph Target Graph Runtime (secs) SD (secs) Np Cp Dp Tp Nt Ct Dt Tt 30 20 0.8 30 700 20 0.2 30 0.15 0.02 30 20 0.8 30 700 20 0.3 30 0.15 0.016 30 20 0.8 30 700 20 0.4 30 0.34 0.06 30 20 0.8 10 700 20 0.3 30 0.63 0.08 30 20 0.8 10 700 20 0.3 60 1.43 0.16 30 20 0.8 10 700 20 0.3 100 2.57 0.3 30 20 0.8 30 700 5 0.3 30 0.14 0.015 30 20 0.8 30 700 10 0.3 30 0.13 0.013 30 20 0.8 30 700 20 0.3 30 0.15 0.016 100 20 0.4 30 300 20 0.3 30 0.71 0.033 100 20 0.6 30 300 20 0.3 30 0.76 0.032 100 20 0.8 30 300 20 0.3 30 0.82 0.034 100 20 0.4 30 300 20 0.3 100 28.02 1.23 100 20 0.4 60 300 20 0.3 100 58.55 2.6 100 20 0.4 100 300 20 0.3 100 2.43 0.12 100 5 0.4 30 300 20 0.3 30 0.71 0.031 100 10 0.4 30 300 20 0.3 30 0.72 0.033 100 20 0.4 30 300 20 0.3 30 0.71 0.034 pattern detection as that of subgraph isomorphism for dynamic graphs. Finally, the results of our algorithm applied to random dynamic graphs were discussed. The next step would be to apply our algorithm on the real data. In fact, we have generated structural traffic patterns using a traffic simulator and we are in the process of applying our algorithm on the generated traffic model. For the case of Handball, the graph has been generated for a single game using video sequences and we are at the stage of defining pertinent structural patterns. Batty M. (2003). Network geography: Relations, interactions, scaling and spatial processes in gis. Centre for Advanced Spatial Analysis Working Paper Series, p. 63:1-63:23. Bibliographie 213 Communications 213 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Carletti V., Foggia P., Saggese A., Vento M. (2018, April). Challenging the time complexity of exact subgraph isomorphism for huge and dense graphs with VF3. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no 4, p. 804-818. Chrisman N. (1977). Concepts of space as a guide to cartographic data structures. In First international advanced study symposium on topological data structures for geographic information systems. Conte D., Foggia P., Sansone C., Vento M. (2004). Thirty years of graph matching in pattern recognition. International Journal of Pattern Recognition and Artificial Intelligence, vol. 18, no 03, p. 265-298. Costes B., Perret J., Bucher B., Gribaudi M. (2015). An aggregated graph to qua- lify historical spatial networks using temporal patterns detection. In 18th agile international conference on geographic information science. Del Mondo G., Stell J. G., Claramunt C., Thibaud R. (2010). A graph model for spatio-temporal evolution. Journal of Universal Computer Science, vol. 16, no 11, p. 1452–1477. Demšar U., Virrantaus K. (2010). Space–time density of trajectories: exploring spatio- temporal patterns in movement data. International Journal of Geographical In- formation Science, vol. 24, no 10, p. 1527-1542. Gudmundsson J., Horton M. (2017, avril). Spatio-temporal analysis of team sports. ACM Comput. Surv., vol. 50, no 2. Jiang B., Claramunt C., Klarqvist B. (2000). Integration of space syntax into gis for modelling urban spaces. International Journal of Applied Earth Observation and Geoinformation, vol. 2, no 3, p. 161 - 171. Jiang C., Coenen F., Zito M. (2013). A survey of frequent subgraph mining algo- rithms. The Knowledge Engineering Review, vol. 28, no 1, p. 75–105. Maduako I., Wachowicz M. (2019). A space-time varying graph for modelling places and events in a network. International Journal of Geographical Information Science, vol. 33, no 10, p. 1915-1935. Oberoi K. S., Del Mondo G., Dupuis Y., Vasseur P. (2017). Towards a qualitative spa- tial model for road traffic in urban environment. In 2017 IEEE 20th international conference on intelligent transportation systems, p. 1724-1729. IEEE. Porta S., Crucitti P., Latora V. (2006). The network analysis of urban streets: A primal approach. Environment and Planning B: Planning and Design, vol. 33, no 5, p. 705-725. Shirani-Mehr H., Banaei-Kashani F., Shahabi C. (2012). Efficient reachability query evaluation in large spatiotemporal contact datasets. Proc. VLDB Endow., vol. 5, no 9, p. 848–859. Tsoukatos I., Gunopulos D. Bibliographie (2001). Efficient mining of spatiotemporal patterns. In C. S. Jensen, M. Schneider, B. Seeger, V. J. Tsotras (Eds.), Advances in spatial and temporal databases, p. 425–442. Springer Berlin Heidelberg. 214 214 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Une approche sémantique pour l’exploitation de données spatio-temporelles avec annotation Application à l’évolution d’assolement Ba-Huy Tran 1, Alain Bouju 2, Christine Plumejeaud-Perreau 3, Vincent Bretagnolle 4 4. Centre d’Etudes Biologiques de Chizé, U.M.R. 7372, CNRS & Université de La Rochelle breta@cebc.cnrs.fr breta@cebc.cnrs.fr RÉSUMÉ. Cet article présente une approche ontologique pour l’exploitation des don- nées d’assolement. Nous présentons une ontologie spatio-temporelle décrivant ces don- nées en réutilisant une ontologie du temps et une ontologie de l’espace. Nous montrons ensuite comment incorporer le processus de fouille de données sémantique et l’anno- tation sémantique afin d e f ournir a ux u tilisateurs u n o util d ’analyse a utomatique et efficace. L’approche proposée a été appliquée aux évolutions des assolements de la Zone Atelier Plaine et Val de Sèvre. ABSTRACT. This article presents an ontology-based approach for exploiting land use data. We present a spatio-temporal ontology describing these data by reusing an on- tology of time and an ontology of space. We then show how to incorporate the process of data mining and annotation on semantic data to provide users an automatic and efficient analysis tool. The proposed approach was applied to the evolution of the land use of the Plaine and Val de Sèvre workshop area. 215 215 Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 MOTS-CLES : ontologies, spatio-temporelle, assolement, ETL, fouille de données, annotation KEYWORDS: ontology, spatio-temporal, land use, ETL, data mining, annotation MOTS-CLES : ontologies, spatio-temporelle, assolement, ETL, fouille de données, annotation KEYWORDS: ontology, spatio-temporal, land use, ETL, data mining, annotation 1. http://za-geminat.cnrs.fr/ 1. Introduction Dans les zones rurales avec une prédominance d’activités agricoles, l’étude des questions environnementalespeuvent bénéficier de l’analyse sur le long terme de la mosaïque des cultures résultant des pratiques agricoles. En effet, les pay- sages agricoles sont principalement le résultat des décisions des agriculteurs portant sur le choix des cultures et de leur répartition à l’échelle de l’exploita- tion. L’agencement, la forme et la nature des cultures structurent l’organisation spatiale du paysage qui a une incidence sur les processus écologiques à diffé- rentes échelles (Schaller et al., 2012). Cette organisation spatiale du paysage évolue dans le temps parce que les agriculteurs modifient l’assolement de leurs parcelles et éventuellement les li- mites de ces parcelles chaque année. En outre, les parcelles peuvent changer de propriétaire ou appartenir à des associations d’exploitants agricoles. Cette information peut être très utile lors de l’étude des liens entre l’environnement socio-économique, les pratiques agricoles et l’organisation spatiale du paysage. L’article présente notre travail étendant celui de (Tran, 2017) réalisé pour le projet GEMINAT 1 visant à étudier ces pratiques agricoles et l’organisation spatial du paysage d’une zone atelier, tout en exploitant ses données d’asso- lement. Nous présentons d’abord les ontologies spatio-temporelles qui servent au développement de notre ontologie montrée par la suite. La section 3 décrit les processus Extract-Transform-Load (ETL) sémantiques qui consistent à ex- traire, transformer et importer les données sources dans un triplestore RDF. La section suivante détaille deux techniques pour exploiter la base construite : la fouille de données sémantiques et l’annotation sémantique. Ensuite, nous démontrons l’utilisation du système appliqué à un cas d’étude sur la base d’as- solement de la zone atelier. Enfin, la conclusion résume notre travail et ouvre des perspectives sur de futurs travaux. 2. Ontologies spatio-temporelles La modélisation spatio-temporelle par une approche ontologique comporte l’ontologie du domaine, l’ontologie du temps et l’ontologie de l’espace pour la représentation des informations sémantiques, temporelles et spatiales res- pectivement. La modélisation spatio-temporelle consiste à coordonner ces trois composantes pour rendre compte des dynamiques spatiales. Nous présentons Communications Communications 216 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Une approche sémantique avec annotation Une approche sémantique avec annotation en premier une étude sur les ontologies existantes et proposons notre choix, ensuite, notre ontologie spatio-temporelle pour l’assolement. en premier une étude sur les ontologies existantes et proposons notre choix, ensuite, notre ontologie spatio-temporelle pour l’assolement. 2.1. Ontologies du temps Les ontologies du temps visent à représenter des concepts et des relations temporelles. Plusieurs ontologies du temps ont fait leur apparition au sein de la communauté du Web sémantique. Les deux travaux majeurs du domaine sont l’ontologie OWL-Time (Hobbs, Pan, 2004) et SWRL Temporal Ontology (O’Connor, Das, 2011). Étant une recommandation du W3C, la première a été choisie pour notre modélisation. L’ontologie repose sur les concepts temporels et les relations temporelles définies dans la théorie d’Allen et bénéficie d’une spécification précise forma- lisée en OWL. Elle a devenu une ontologie de référence pour représenter le temps dans plusieurs domaines, y compris le domaine de la recherche d’infor- mation. L’élément principal de l’ontologie est TemporalEntity qui comprend deux sous-classes : Instant et Interval. Les propriétés hasBeginning et hasEnd sont utilisées pour définir le début et la fin d’une entité temporelle Interval. 2. Geography Markup Language 3. Open Geospatial Consortium 3. Open Geospatial Consortium 2. Geography Markup Language 2.2. Ontologies de l’espace De nombreux travaux ont proposé des ontologies pour définir les concepts et les relations spatiales. Ces ontologies sont majoritairement basées sur le modèle GML 2 de l’OGC 3 et utilisent le langage de formalisation ontologique OWL. Nous nous intéressons à la partie déclarative et à la représentation des don- nées spatiales par des ontologies. Les ontologies de l’espace les plus utilisées sont: Basic Geo Vocabulary, GeoRSS et GeoOWL. Il existe aussi des modèles minimalistes provenant des extensions du langage SPARQL comme le modèle stRDF (Koubarakis, Kyzirakos, 2010) et l’ontologie GeoSPARQL (Battle, Ko- las, 2012). Il faut noter qu’il existe aussi d’autres ontologies comme NeoGeo Geometry Ontology, GeoNames Ontology ou Biological Spatial Ontology ou la transformation d’un modèle spatial en une ontologie spatiale comme utilisé dans (Mefteh et al., 2012). Parmi ces ontologies, nous avons choisi GeoSPARQL, qui est une recomman- dation du W3C, pour pouvoir profiter des fonctions fournies par les triplestore spatiaux pour effectuer du raisonnement spatial. L’ontologie GeoSPARQL pos- sède trois classes : geo:Feature représentant des entités qui peuvent avoir une localisation, geo:Geometry décrivant un emplacement spatial, c’est-à-dire un ensemble de coordonnées, et geo:SpatialObject étant la classe racine des deux 217 217 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] classes. La propriété geo:hasGeometry met en relation une entité à son em- placement. En séparant les entités réelles et leurs emplacements, GeoSPARQL permet à plusieurs emplacements d’être liés à une entité à des fins diverses. La ressource de la géométrie a alors un littéral qui est lié à une propriété en utilisant soit geo:asWKT ou geo:asGML. 2.3. Représentation des changements Les ontologies de temps présentées ci-dessus permettent de représenter des concepts et des relations temporelles. Toutefois, elles ne s’intéressent qu’à dé- crire des éléments de données individuels plutôt qu’à construire un modèle temporel pour décrire systématiquement toutes les informations temporelles dans un système. Autrement dit, de telles ontologies ne suffisent pas à modéli- ser l’évolution des entités. Par ailleurs, comme les deux langages ontologiques du Web sémantique, OWL et RDFS, ne permettent que des relations binaires entre les entités, le traitement des informations qui changent au fil d u temps devient un problème critique. Afin de pallier ces problèmes, plusieurs approches ont été proposées. Nous pouvons distinguer les approches qui ne modifient pas les ontologies comme le versionnement, ou les approches qui doivent introduire de nouveaux éléments dans l’ontologie comme les techniques de réification et l’utilisation des modèles des fluents (ou 4D fluents). Les modèles des fluents a vec l a n otion d e T imeSlice ( Welty, F ikes, 2006) semblent les plus connus. On peut citer des travaux de (Batsakis, Petrakis, 2011) ou (Harbelot et al., 2013) visant à améliorer la modélisation et à l’appli- quer à différents domaines. En effet, le concept TimeSlice joue le rôle central en étant lié à une dimension spatiale, une dimension temporelle et éventuellement une dimension sémantique liée au domaine applicatif. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Une approche sémantique avec annotation Une approche sémantique avec annotation – Composante temporelle : Cette composante représente une unité de temps durant lequel le timeslice est valide. La propriété gem:hasTime associe un ti- meslice à une entité temporelle. – Composante spatiale : Cette composante représente une description géo- métrique de l’entité. Pour cela, nous considérons la classe gem:TimeSlice comme une spécialisation de geo:Feature. Afin de représenter ces associations spatiales dans le temps, nous utilisons la relation filiation qui permet de connecter deux timeslices consécutifs d’un même objet. Un changement sur la composante spatiale ou sur la composante sémantique génère un nouveau timeslice. D’une part, ce timeslice est obligatoi- rement lié par une relation de filiation avec le timeslice d’origine. D’autre part, l’intervalle d’existence du timeslice parent est contigu à l’intervalle du timeslice enfant. La propriété gem:hasEvent, associant deux gem:ParcelTimeSlice, est uti- lisée pour représenter des événements territoriaux (Plumejeaud et al., 2011) survenus entre ces timeslices. Figure 1. Un extrait de l’ontologie GEMINAT utilisée pour l’assolement et son articulation avec les ontologies GeoSPARQL et OWL-Time Figure 1. Un extrait de l’ontologie GEMINAT utilisée pour l’assolement et son articulation avec les ontologies GeoSPARQL et OWL-Time 6. http://mayor2.dia.fi.upm.es/oeg-upm/index.php/en/technologies/315-morph-rdb/ // / / / / / 4. http://d2rq.org/ 5. https://capsenta.com/ultrawrap/ 5. https://capsenta.com/ultrawrap/ 7. https://web.imsi.athenarc.gr/redmine/projects/geoknow_public/wiki/TripleGeo 4. http://d2rq.org/ 5. https://capsenta.com/ultrawrap/ 6. http://mayor2.dia.fi.upm.es/oeg-upm/index.php/en/technologies/315-morph-rdb 7. https://web.imsi.athenarc.gr/redmine/projects/geoknow_public/wiki/TripleGeo 2.4. Ontologie pour l’assolement Nous proposons une ontologie GEMINAT, préfixée g em d ans l es para- graphes qui suivent. Inspiré des modèles de fluents, n ous c onsidérons q ue les parcelles agricoles possèdent une ou plusieurs tranches de temps correspondant aux différents états à travers leur vie. Elles peuvent changer de forme (par des événements territoriaux) ou de culture suivant les saisons. Le changement peut être enregistré à travers des observations ou des relevés. Ainsi, la classe gem:ParcelTimeSlice est introduite. Elle se compose de trois composantes (Fi- gure 1) : – Composante sémantique : Cette composante représente l’ensemble des propriétés décrivant les caractéristiques de la parcelle durant la période de validité du timeslice. Dans un premier temps, seule la culture (gem:LandUse) est utilisée pour cette composante. 218 Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] 5. https://capsenta.com/ultrawrap/ 6. http://mayor2.dia.fi.upm.es/oeg-upm/index.php/en/technologies/315-morph-rdb/ 7 https://web imsi athenarc gr/redmine/projects/geoknow public/wiki/TripleGeo 3. ETL sémantique Dans le peuplement de l’ontologie, lui-même, il existe des techniques pour exposer des données relationnelles en RDF : (i) la matérialisation de données et (ii) la traduction de données à la demande. Quelle que soit la technique choisie, un outil de traduction est nécessaire pour l’exposition de ces données. Nous pouvons distinguer les outils D2RQ 4, Ultrawrap 5, Morph 6, TripleGeo 7 219 Communications Communications 219 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] ou GeoTriples 8 aidant à transformer les données sources en RDF et les outils D2R 9 visant à exposer un triplestore virtuel, certains outils comme Ontop 10 réunissant les deux fonctionnalités. Nous avons choisi la matérialisation de données dans un triplestore géospa- tial pour les raisons suivantes: – Il n’est pas facile de réaliser la traduction à la demande parce que les sources de données sont disponibles dans différents formats, cela requiert une étape de pré-traitement. En effet, comme présenté, nous avons intégré d’autres sources de données (qui ne sont pas décrites ici). – Un triplestore géospatial peut être utilisé comme un entrepôt de données pour stocker les données sémantiques. De cette manière, le raisonnement spatial sera pris en charge par le triplestore et l’enrichissement ou l’annotation de données peuvent être réalisés. Actuellement, plusieurs triplestores prennent en charge le stockage et l’in- terrogation de données spatiales à l’aide de GeoSPARQL. Il existe des triples- tores gratuits comme Parliament 11 (Battle, Kolas, 2012), Strabon 12 (Kyzirakos et al., 2012) ou GraphDB 13; et d’autres payants comme Oracle Spatial and Graph 14 ou Stardog 15. Parmi les triplestores gratuits, Strabon a été choisi car il a de bonnes performances globales grâce à des techniques d’optimisa- tion particulières qui permettent aux opérations spatiales de bénéficier des fonctionnalités de PostGIS au lieu de s’appuyer sur des bibliothèques externes (Patroumpas et al., 2014). Pour les applications complexes qui incluent à la fois des jointures spatiales ou des agrégations spatiales, Strabon est le seul triplestore qui fonctionne bien (Ioannidis et al., 2019). 4. Exploitation de données Une fois les données importées dans l’entrepôt, elles peuvent être exploitées de différentes manières. Nous allons présenter la fouille de données sémantiques et l’annotation sémantique de données. La première permet de détecter des erreurs au sens statistique du terme par exemple, tandis que la seconde fournit un support sémantique pour l’enregistrement des résultats de la fouille dans le modèle. 8. http://geotriples.di.uoa.gr/ 9. http://d2rq.org/d2r-server 10. http://ontop.inf.unibz.it/ 11. http://parliament.semwebcentral.org/ 12. http: //strabon.di.uoa.gr/ 13. http://graphdb.ontotext.com/ 14. https://www.oracle.com/database/technologies/spatialandgraph.html 15. https://www.stardog.com/ Communications 220 4.2. Annotation sémantique Une annotation attache des données à d’autres données : elle établit, dans un certain contexte, une relation (typée) entre les données annotées et les don- nées d’annotation. Nous distinguons trois types d’annotations (Oren et al., 2006) : informelles, formelles (formellement définie) et ontologiques (utilisant uniquement des termes ontologiques). Il existe de nombreuses définitions de l’annotation sémantique mais le point commun de ces définitions est qu’une annotation sémantique est le processus de liaison d’une ressource électronique à une ontologie spécifique. Les ressources électroniques peuvent être des conte- nus textuels, des images, des vidéos, des services, etc. L’ontologie n’est ici qu’un des moyens possibles pour fournir une sémantique formelle (Liao et al., 2011). Les modèles d’annotation les plus connus sont Annotea (Kahan et al., 2002), Annotation Ontology (Ciccarese et al., 2011), ou Open Annotation Data Mo- del (Sanderson et al., 2013). Le dernier, présenté par le W3C, est utilisé pour annoter les ressources de notre système. Le modèle décrit les annotations par la classe Annotation qui est une association entre deux éléments, une ressource de contenu, appelée Body et une ou plusieurs ressources appelée Target. L’an- notation fournit des informations sur le Target via la connexion au Body. Les annotations peuvent être complétées par des méta-données, par exemple, l’au- teur ou la date de création de l’annotation. 4.1. Fouille de données sémantiques Au cours de la dernière décennie, de nombreuses approches ont été propo- sées pour appliquer des algorithmes de fouille aux données sémantiques. Nous pouvons distinguer trois approches : (i) Approches basées sur les logiques de description visant à adapter les algorithmes existants au nouveau format de re- présentation des données, comme dans (Józefowska et al., 2010). (ii) Fouille des graphes consistant à extraire des sous-graphes (Kuramochi, Karypis, 2001) ou des sous-arbres fréquents (Chi et al., 2004). (iii) Approches statistiques visant à transformer les données RDF en structure de données classique utilisée par des algorithmes de fouille comme dans (Nebot, Berlanga, 2012). Comme nous souhaitons implémenter un système simple et efficace pour faciliter l’analyse des données, il est préférable de réutiliser des algorithmes existants. Dès lors, les approches statistiques conviennent à nos besoins car elles ne consistent pas à introduire de nouveaux algorithmes comme dans le cas de la fouille de graphes ou à modifier des algorithmes existants comme dans le cas des approches basées sur la logique de description. Une approche sémantique avec annotation Une approche sémantique avec annotation Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Une approche sémantique avec annotation Communications 5. Exploitation de données d’assolement par l’approche ontologique Notre pipeline pour l’exploitation de données d’assolement par l’approche ontologique se compose de cinq modules: 221 Communications 221 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Figure 2. Le pipeline pour l’exploitation de données sémantiques. Figure 2. Le pipeline pour l’exploitation de données sémantiques. 1. ETL : Les données sources sont extraites, transformées en RDF et im- portées dans le triplestore Strabon. 1. ETL : Les données sources sont extraites, transformées en RDF et im- portées dans le triplestore Strabon. 2. Requêtage : Le module sert à traiter les requêtes SPARQL 1.1. Elles peuvent être des requêtes de mise à jour comme dans le cas de l’annotation ou l’enrichissement; ou des requêtes normales pour les autres cas. 3. Visualisation : Le résultat d’une requête peut être affiché sur une carte grâce à la librairie OpenLayers. 3. Visualisation : Le résultat d’une requête peut être affiché sur une carte grâce à la librairie OpenLayers. 4. Fouille de données : La fouille de données peut être enchaînée après une requête. Pour cela, une sélection d’attribut peut être réalisée pour éliminer les attributs non pertinents comme l’identifiant ou la géométrie contenus dans chaque transaction. Le format de l’entrée du processus est tabulaire. Trois algorithmes de fouille sont implémentés à l’aide de l’outil Weka: (i) Apriori pour la recherche des règles d’association, (ii) Part pour la recherche des règles de classification. (iii) K-moyennes pour la recherche des partitions. 5. Annotation: Basé sur les connaissances obtenues lors de l’étape précé- dente ou sur le résultat des requêtes, on peut annoter des doutes ou anomalies. Par exemple, on peut annoter des timeslices ayant une culture douteuse relati- vement à une règle de succession ou à la période donnée. 5. Annotation: Basé sur les connaissances obtenues lors de l’étape précé- dente ou sur le résultat des requêtes, on peut annoter des doutes ou anomalies. Par exemple, on peut annoter des timeslices ayant une culture douteuse relati- vement à une règle de succession ou à la période donnée. 6. Enrichissement: Les connaissances ou les résultats obtenus peuvent servir à enrichir la base de connaissances. Un cas d’étude sur le regroupement des cultures visant à améliorer le taux de confiance des règles d’association a été démontré dans (Tran, 2017). 6. Enrichissement: Les connaissances ou les résultats obtenus peuvent servir à enrichir la base de connaissances. 5. Exploitation de données d’assolement par l’approche ontologique Un cas d’étude sur le regroupement des cultures visant à améliorer le taux de confiance des règles d’association a été démontré dans (Tran, 2017). 222 Communications 222 Une approche sémantique avec annotation Une approche sémantique avec annotation 6.2. Règle d’association sur la succession de cultures Afin d’examiner les pratiques agricoles, nous avons pris le cas d’étude simple sur la succession de culture de deux années. En formulant une requête SPARQL et en paramétrant l’algorithme Apriori avec le support = 0.01 (plutôt petit pour réserver les successions rares) et la confiance = 0.8, nous avons obtenu une liste des règles ci-dessous (Tableau 1). Les résultats d’analyse détaillés ont été présentés dans (Tran, 2017). 6. Application Pour démontrer la faisabilité de notre proposition, nous avons développé un démonstrateur 16. Celui-ci a été appliqué à la base d’assolement de la Zone Atelier Plaine et Val de Sèvre. Nous présentons d’abord la zone atelier et les données, ensuite, nous résumons le résultat obtenu par une méthode de fouille de données et enfin nous décrivons notre annotation des données. 6.1. La base d’assolement Appartenant au réseau national des Zones Ateliers, la Zone Atelier Plaine et Val-de-Sèvre couvre 450 km2 au sud de Niort, Deux-Sèvres. Il s’agit essentiel- lement d’une plaine de cultures céréalières intensives où les activités d’élevage sont encore présentes mais en forte baisse. Les parcelles agricoles sont encore de taille modeste (4-8 ha) et 15% d’entre elles sont occupées par des prairies. L’organisation spatiale du paysage évolue dans le temps parce que les agri- culteurs modifient l’assolement de leurs parcelles chaque année, mais également recomposent parfois les parcelles entre elles, changeant ainsi leurs formes. De- puis 1994, les occupations du sol sont donc relevées annuellement sur le terrain et numérisées sur les 19 000 parcelles agricoles. Ces données sont centralisées dans une base de données nommée «Assolement». Dans cette base, une parcelle agricole est considérée comme une unité de gestion, un polygone entouré par des entités ayant différentes cultures au cours des années successives. Une parcelle est délimitée par des limites physiques et ne contient qu’un seul type de culture. Elle diffère de la parcelle cadastrale, mais également des blocs stockés dans le registre parcellaire graphique (RPG). 16. Son code source est accessible sur https://github.com/tbhuy/stRDFMining [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Communications [ISBN : 978-2-910545-12-1] 7. Conclusion et perspectives Nous avons présenté dans l’article une approche sémantique pour l’exploita- tion de données d’assolement de la zone atelier Plaine de Val de Sèvre. Pour dé- crire les données sources, nous avons développé une ontologie spatio-temporelle en réutilisant des modèles bien connus. Les processus ETL ont été appliqués pour matérialiser les données dans un triplestore spatial à partir duquel la fouille de données sémantique et l’annotation sémantique de données peuvent être effectuées. Nous pensons appliquer le système sur d’autres jeux de données de la zone atelier. À cet égard, nous souhaitons étendre l’ontologie développée et la rendre la plus générique possible pour que le système s’adapte au mieux à ces sources. 6.3. Annotation des valeurs douteuses À partir des règles obtenues, on peut examiner des observations (ou tran- sactions) douteuses. On peut identifier et annoter les instances concernées (par 223 Communications 223 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Table 1. Les règles d’association trouvées par l’Apriori pour une succession de cultures sur deux années Table 1. Les règles d’association trouvées par l’Apriori pour une succession de cultures sur deux années N° Culture année T Culture année T+1 Confiance 1 Urbain Urbain 0,97 2 Forêt et friches Forêt et friches 0,91 3 Colza Blé 0,87 4 Tournesol Blé 0,82 5 Prairies et luzernes Prairies et luzernes 0,73 exemple les timeslices) afin de les faire ré-examiner et corriger par les experts ou les éliminer dans les futures analyses. Par exemple, pour la règle Bâti -> Bâti, il existe 9177 transactions qui la satisfassent (98%). On pourrait donc ensuite examiner celles (183 transactions équivalent 2%) qui ne la satisfont pas. Le tableau 2 représente un extrait des observations (ou transactions) ne satisfaisant pas la première règle. Le timeslice correspondant peut être marqué comme douteux. Figure 3 en décrit un exemple dans lequel, une annotation est créée pour associer le Ti- meslice/337 à la Suspicion/01 pour marquer un doute sur la culture quand la règle d’association a été lancée sur une succession de cultures sur deux ans. Table 2. Un extrait des transactions ne satisfaisant pas la règle choisie N° Timeslice T Timeslice T+1 Culture T Culture T+1 1 Timeslice/209 Timeslice/210 Péri-village Urbain 2 Timeslice/337 Timeslice/338 Blé Urbain 3 Timesice/1585 Timeslice/1586 Blé Urbain Figure 3. Exemple d’une annotation sémantique sur un timeslice. Table 2. Un extrait des transactions ne satisfaisant pas la règle choisie N° Timeslice T Timeslice T+1 Culture T Culture T+1 1 Timeslice/209 Timeslice/210 Péri-village Urbain 2 Timeslice/337 Timeslice/338 Blé Urbain 3 Timesice/1585 Timeslice/1586 Blé Urbain Figure 3. Exemple d’une annotation sémantique sur un timeslice. Figure 3. Exemple d’une annotation sémantique sur un timeslice. 224 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Kuramochi M., Karypis G. (2001). Frequent subgraph discovery. In Proceedings 2001 ieee international conference on data mining, p. 313-320. Kyzirakos K., Karpathiotakis M., Koubarakis M. (2012). Strabon: A semantic geo- spatial dbms. In Proceedings of the 11th international conference on the semantic web - volume part i, p. 295–311. Berlin, Heidelberg, Springer-Verlag. Liao Y., Lezoche M., Panetto H., Boudjlida N. (2011, octobre). Semantic Annota- tion Model Definition for Systems Interoperability. In OTM (Ed.), OTM 2011 Workshops 2011 - 6th International Workshop on Enterprise Integration, Inter- operability and Networking (EI2N), vol. LNCS 7046, p. 61-70. Hersonissos, Crete, Greece, Springer-Verlag. Mefteh W., Bouju A., Malki J. (2012, mars). Une approche ontologique pour la structuration de données spatio-temporelles de trajectoires. Application à l’étude des déplacements de mammifères marins. Revue Internationale de Géomatique, vol. 22, no 1, p. 55-75. Nebot V., Berlanga R. (2012, février). Finding association rules in semantic web data. Know.-Based Syst., vol. 25, no 1, p. 51–62. O’Connor M. J., Das A. K. (2011). A method for representing and querying tem- poral information in owl. In A. Fred, J. Filipe, H. Gamboa (Eds.), Biomedical engineering systems and technologies, p. 97–110. Berlin, Heidelberg, Springer Berlin Heidelberg. Oren E., Möller K., Scerri S., Handschuh S., Sintek M. (2006). What are semantic annotations. Relatório técnico. DERI Galway, vol. 9, p. 62. Patroumpas K., Giannopoulos G., Athanasiou S. (2014). Towards geospatial semantic data management: Strengths, weaknesses, and challenges ahead. In Sigspatial ’14: 22nd sigspatial international conference on advances in geographic information systems, dallas, texas, november, 2014, p. 301–310. New York, USA, ACM. Plumejeaud C., Mathian H., Gensel J., Grasland C. (2011). Spatio-temporal analysis of territorial changes from a multi-scale perspective. International Journal of Geographical Information Science, vol. 25, no 10, p. 1597-1612. Sanderson R., Ciccarese P., Sompel H. Van de. (2013). Designing the w3c open anno- tation data model. In Proceedings of the 5th annual acm web science conference, p. 366–375. New York, NY, USA, Association for Computing Machinery. Schaller N., Lazrak E., Martin P., Mari J.-F., Aubry C., Benoît M. (2012). Com- bining farmers’ decision rules and landscape stochastic regularities for landscape modelling. Landscape Ecology, vol. 27, no 3, p. 433-446. Tran B.-H. (2017, novembre). Une approche sémantique pour l’exploitation de don- nées environnementales : application aux données d’un observatoire. Welty C., Fikes R. (2006). Bibliographie Batsakis S., Petrakis E. G. M. (2011). SOWL: A Framework for Handling Spatio- temporal Information in OWL 2.0. In Proceedings of the 5th international confe- rence on rule-based reasoning. and Applications, RuleML’2011, Programming. Battle R., Kolas D. (2012). Enabling the Geospatial Semantic Web with Parliament and GeoSPARQL. Semantic Web, vol. 3, no October 2012, p. 355–370. Chi Y., Muntz R. R., Nijssen S., Kok J. N. (2004, novembre). Frequent subtree mining - an overview. Fundam. Inf., vol. 66, no 1–2, p. 161–198. Ciccarese P., Ocana M., Castro L. J. G., Das S., Clark T. (2011). An open annotation ontology for science on web 3.0. In Journal of biomedical semantics, vol. 2, p. S4. Harbelot B., Arenas H., Cruz C. (2013). Continuum: A spatiotemporal data model to represent and qualify filiation relationships. In Proceedings of the 4th acm sigspatial international workshop on geostreaming, p. 76–85. New York, NY, USA, ACM. Hobbs J. R., Pan F. (2004). An ontology of time for the semantic web. ACM Transactions on Asian Language Information Processing, vol. 3, p. 66-85. Ioannidis T., Garbis G., Kyzirakos K., Bereta K., Koubarakis M. (2019). Eva- luating geospatial RDF stores using the benchmark geographica 2. CoRR, vol. abs/1906.01933. Józefowska J., Lawrynowicz A., Lukaszewski T. (2010). The role of semantics in mining frequent patterns from knowledge bases in description logics with rules. CoRR, vol. abs/1003.2700. Kahan J., Koivunen M.-R., Prud’Hommeaux E., Swick R. (2002). Annotea: an open rdf infrastructure for shared web annotations. Computer Networks, vol. 39, no 5, p. 589 - 608. Koubarakis M., Kyzirakos K. (2010). Modeling and querying metadata in the seman- tic sensor web: The model strdf and the query language stsparql. In Proceedings of the 7th international conference on the semantic web: Research and applications - volume part i, p. 425–439. Berlin, Heidelberg, Springer-Verlag. 225 Communications 225 Communications [ISBN : 978-2-910545-12-1] A reusable ontology for fluents in owl. In Fois 2006: The 4th international conference on formal ontology in information systems, november 9-11, 2006, baltimore, maryland, usa, p. 226–236. Amsterdam, NL, IOS Press. 226 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 understand the ever-changing urban landscape. These models may encounter issues with interoperability and data-loss if conversion is required for integration. Today we can base ourselves on conceptual models which can help deal with data losses during conversion and may also help preserve data interoperability. This kind of model-driven approach can be useful as common representation. Recently, a movement towards graph and semantic based data representations has also grown in popularity to respond to these issues. As a first step, we consider CityGML, a common standard that can be used to represent 3D urban information. We propose a strategy for converting the semantics of CityGML conceptual model into ontologies and later to semantic web formats to facilitate integration. In addition, we propose a method for converting and storing CityGML instances into RDF individuals that respect the generated ontology. This proposed approach overcomes the loss of semantic information resulting from the direct translation of different types of data into graphs such as RDF. MOTS-CLES : Modèle conceptuel de données urbaines 3D, Web sémantique, Interopérabilité, CityGML, RDF, GeoSPARQL KEYWORDS: 3D urban data conceptual models, Semantic Web, Interoperability, CityGML, RDF, GeoSPARQL 1. Introduction The urban landscape is a complex and heterogeneous source of information. The anthropization and urbanization of modern cities have brought various actors, issues, and systems that all have complex interconnections and dependencies both on local and global scales. The data-driven approaches used to represent and model these urban environments often depend on multi-source and multidimensional urban data, which consists of a variety of different information coming from multiple actors and organizations (Barbosa et al., 2014). For example, city governments have different departments for handling different subjects or issues like water and sewage, pollution, traffic, energy, etc. (Rochet and Pinzon, 2016). Sometimes these organizations release open datasets using their own internally defined business formats or using formats that conform to a specific use-case as opposed to more generic or widely used standards, thus making it difficult to get an integrated view of the data from multiple sources. In response to the limits of using internal formats, we are now seeing the growing use of geospatial and 2D/3D urban data standards. Different international organizational bodies (OGC, ISO, W3C...) propose standards and release conceptual data models for their standards to ensure that the data is represented uniformly by the users, thereby giving open datasets an aspect of interoperability. Take for example, the CityGML standard1 released by the Open Geospatial Consortium (OGC), a city 1 https://www.ogc.org/standards/citygml Communications 228 Diego Vinasco-Alvarez1, John Samuel2, Sylvie Servigne3, Gilles Gesquière4 1. Univ. Lyon, Université Lumière Lyon 2, LIRIS 5 av. Pierre Mendès-France, 69622, Villeurbanne, France diego.vinasco-alvarez@univ-lyon2.fr 1. Univ. Lyon, Université Lumière Lyon 2, LIRIS 5 av. Pierre Mendès-France, 69622, Villeurbanne, France diego.vinasco-alvarez@univ-lyon2.fr 2. Univ. Lyon, CPE Lyon, LIRIS 43, bd du 11 novembre 1918, 69616, Villeurbanne, France john.samuel@cpe.fr 3. Univ. Lyon, INSA Lyon, LIRIS 7 Avenue Capelle, 69621, Villeurbanne, France sylvie.servigne@insa-lyon.fr 4. Univ. Lyon, Université Lumière Lyon 2, LIRIS 5 av. Pierre Mendès-France, 69622, Villeurbanne, France gilles.gesquiere@univ-lyon2.fr RESUME. L'urbanisation est un processus d'évolution continue qui est actuellement étudié par nombre de chercheurs. Les modèles d'information de la ville, multi-sources et multidimensionnels, sont souvent utilisés pour comprendre le paysage urbain, qui est en constante évolution. Ces modèles peuvent cependant poser des problèmes d'interopérabilité et de perte de données lors de conversions, souvent nécessaires pour permettre leur intégration. Aujourd'hui, nous pouvons nous baser sur des modèles conceptuels qui peuvent aider à traiter les pertes de données lors de conversions, et peuvent également contribuer à préserver l'interopérabilité des données. Ce type d'approche modèle-centrée permet de définir une représentation commune. Récemment, un mouvement vers la représentation de données basée sur les graphes et la sémantique a gagné en popularité pour répondre aux problèmes d’interopérabilité. Dans un premier temps, pour illustrer notre approche, nous considérons CityGML, un standard de représentation des informations urbaines en 3D. Nous proposons une stratégie pour convertir la sémantique du modèle conceptuel de CityGML vers des ontologies et, ensuite, en formats web sémantiques pour faciliter l'intégration. Nous proposons également une méthode pour convertir et stocker les instances CityGML en individus RDF qui respectent l'ontologie générée. L’approche proposée permet de pallier le déficit d’information sémantique résultant de la traduction directe de différents types de données en graphe, tels que RDF. ABSTRACT. Urbanization is a continuous evolution process that is currently studied by a number of researchers. Multi-source and multidimensional city information models are often used to 227 Communications 227 Communications [ISBN : 978-2-910545-12-1] 1 https://www.ogc.org/standards/citygml Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] A semantic web representation of urban data information model used for 3D urban modeling to represent city objects at different levels of detail. Another aspect of urban data is the underlying n-dimensional structure. For a long time, 2D maps, aerial views, and cadastral views were used for the management of the urban lifecycle. With the growing availability of 3D data, thanks to new data acquisition processes (LIDAR, photogrammetry...), numerous applications are now possible with 3D urban data (Biljecki 2015). Recently, some urban data models also take into account the changes of cities as they evolve over time in addition to spatial information (Chaturvedi and Kolbe, 2019; Jaillot et al., 2020; Samuel et al., 2020), thereby adding the temporal dimension (4D). These models help historians and city planners visualize and contextualize the impacts previous key projects have had on the development of a city. Such studies require an integration of the urban models with document corpus consisting of project plans, newspaper articles, archives etc. However, it is challenging to integrate data from different sources with different standards for providing an integrated urban view (Biljecki et al., 2018; Ohori et al., 2018). Semantic web models are now being used increasingly to ensure interoperability among multiple data sources (Malinverni et al., 2020; Nuninger et al., 2020; Tran et al., 2016). Due to its increasing popularity, a number of tools are now available to transform data in legacy formats to semantic web formats but, as said by C. Claramunt in (Claramunt, 2020) “3D models and Building Information Models should be fully integrated.” However, any transformation must conform to the original conceptual model used for developing the standard or risk inconsistent or semantically incomplete representations (Bohring and Auer, 2015). In this work, we take into consideration a model-centric approach for transformation of urban data into ontological and graph formats. We demonstrate our approach with CityGML, particularly making use of open data from the metropole of Lyon. Our approach is detailed in this article. Section 2 presents the problem statement of model-centric urban data transformation in detail. Section 3 presents the existing state of the art. Taking the case study of CityGML, we detail our approach in Section 4. Communications However, the focus of our work is not limited to transforming CityGML, but to link data from other sources, especially those based on open standards. These perspectives are discussed in section 5. Section 6 concludes the article. 2. Problem Statement Managing and analyzing urban data is complex because different stakeholders and different entities produce these data, sometimes in silos. City administrators and urban planners often require an integrated view of these data for analyses and for obtaining meaningful insights for daily routine tasks and future planning. Take for example, urban planners who wish to construct a new high-rise building in a particular sector, need to have information on the other structures present around the proposed building. Such information may help them to study the impact on visibility of historical buildings of interest, analyze the reduced or increased sunshine or shadows because 229 229 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] of the new construction. All these studies require data from multiple sources, such as the terrain information, 3D structures of urban objects, position of celestial bodies at different points of times etc. However, the task of integrating urban data from multiple sources remains very difficult. Additionally, the various standards used to model this data evolve based on the growing requirements of the domain users. New versions of standards need to be released. Sometimes, the data released under previous versions is incompatible with the newer versions. Hence, any semantic web representation solution for an integrated view of urban data must deal with the heterogeneous data from multiple sources as well as their evolution. Concerning 3D geospatial urban data models, many approaches propose directly translating different data formats into graphs, such as RDF, without taking into account the underlying conceptual model (Bohring and Auer, 2015), and consequently omit semantic information during translation, weakening the initial model. The question is therefore to propose an approach that allows the preservation of the semantics of a 3D urban model when converting into a semantic web representation. 2 http://schemas.opengis.net/gml/3.2.1/gml_32_geometries.rdf 3 https://github.com/opengeospatial/CityGML-3.0Encodings 4 https://github.com/opengeospatial/geosemantics-dwg 3. State of the art There are growing efforts to ensure semantic interoperability across heterogeneous data from diverse multiple sources. To obtain an integrated view, one commonly used approach is to transform the data from one data format to a common format by making use of data transformers. Stylesheets like XSLT, JSONT are commonly used for this purpose. However, some of these stylesheets are written on demand basis and may not be available on the internet for use. This means that if the users want to reuse and reproduce some of the results based on the transformed data, they may not be able to achieve it, especially if transformers follow different methodologies. In addition, information loss is another major problem during data transformation (Levina, 2012). It is important to take into consideration the initial conceptual models used for the conception of the modeling. Several information modeling techniques exist in the literature (Bork et al., 2020). One possible approach is to ensure that a common modeling language is used across domains, like the semantic web ontology (OWL) language. However, UML models are often used by domain experts (De Paepe et al., 2017) to conceive and represent information models instead of semantic web technologies like OWL, RDFS etc. Another possible approach is to build ontologies from existing standards, i.e., transforming models conceived using different modeling languages to a semantic web ontology. Researchers are currently exploring how to make the most of the familiar modeling languages like UML (De Paepe et al., 2017) and automatically generate ontologies in OWL from these models. Works like (Bohring and Auer, 2015; Kramer et al., 2015; Usmani et al., 2020) use similar approaches based on XML Schema as a model towards OWL ontologies. Another possible solution to this problem is if a standards body releases the ontologies for their conceived models. Take for example, the GML 3.2 geometries 230 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] A semantic web representation of urban data ontology2 made available by the OGC. Currently, an official ontology for CityGML 2.0 is missing from the standards body. Nevertheless, it is important to state some works like (Brink et al., 2014; Métral and Falquet, 2018) proposed ontologies for CityGML 2.0. 3. State of the art Additionally, there is a GIS working group3,4 looking to ensure interoperability with semantic web technologies such as RDF and OWL for future releases of CityGML (Kutzner et al., 2020). As discussed above, we are looking for integrating heterogeneous data from multiple sources by proposing an approach that preserves the semantics of a 3D urban model when converting it to semantic web representations. Hence, it is important to have standard procedures or guidelines for ensuring the mapping between different models so that it can be easily reproduced by other users. Our approach is to take into consideration the conceptual model of information related to cities and transform the model using these standard procedures so that the data converted is conformance to the original model and can also be reproduced by other users following these guidelines. 4.1. Proposed Approach Design In order to transform the CityGML conceptual model and CityGML instances into semantic web representations, our proposal uses several XSLT-based transformations based on previous works on XML to RDF, RDFS, and OWL conversion (Bedini et al., 2011; Bohring and Auer, 2015; Brink et al., 2014; Kramer et al., 2015; Métral and Falquet, 2018; Vinasco-Alvarez et al., 2020). The general proposed pipeline is described in Figure 1, which can be broken down into 3 main activities: p , p p based on previous works on XML to RDF, RDFS, and OWL conversion (Bedini et al., 2011; Bohring and Auer, 2015; Brink et al., 2014; Kramer et al., 2015; Métral and Falquet, 2018; Vinasco-Alvarez et al., 2020). The general proposed pipeline is described in Figure 1, which can be broken down into 3 main activities: – Creation of an urban data model ontology from the CityGML conceptual model (as described by its application schema) – Creation of an urban data model ontology from the CityGML conceptual model (as described by its application schema) – Generation of a CityGML instance to RDF transformation stylesheet from the CityGML conceptual model (as described by its application schema) – Transformation of CityGML instances to RDF graphs using the stylesheet generated in step 2. – Transformation of CityGML instances to RDF graphs using the stylesheet generated in step 2. FIGURE 1. Overview of proposed pipeline applied to the CityGML conceptual model and instances FIGURE 1. Overview of proposed pipeline applied to the CityGML conceptual model and instances During each transformation, the resulting information must be logically consistent and maintain its interoperability from CityGML and GML. To ensure this, several challenges need to be overcome. For instance, the generated RDF graphs and types must conform to the model described by the ontology. How can mappings be created to ensure this? The CityGML application schema often implements elements that do not have a direct equivalent in OWL or RDFS; how should these elements be represented to best describe CityGML with semantic web formats? In addition, CityGML schema often draws from elements, types, functionality from external schema such as xLinks and xAL addresses. How should these imported elements be addressed to preserve their original functionality? Finally, how can we make use of existing semantic web standards to make geospatial queries? These issues are addressed in sections 4.2 and 4.3. 4. Proposed approach illustrated with the CityGML 2.0 Conceptual Model As previously stated, the information produced by cities and the urban lifecycle can be broken down into hierarchies of concrete and abstract subdomains of information - such as energy, transportation, infrastructure, and many others - and be represented and visualized by 3D urban data models. CityGML is an international OGC standard that is commonly used for this purpose. Taking the CityGML conceptual model as an example, we see frameworks for describing the spatial and thematic information of the more ‘material’ subdomains within CityGML’s modules, e.g. water, buildings, terrain, etc. More immaterial subdomains such as education or public administration are not currently represented by these modules but as a first step the existing conceptual model is more than sufficient to begin transforming open urban data into semantic web formats. To efficiently take advantage of the benefits of semantic web formats, a direct transformation of CityGML instances is not enough, the CityGML conceptual model itself must also be transformed and be used to ‘guide’ instance transformations. This results in meaningful RDF graphs that respect the resulting urban data conceptual model. 231 Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 4.1. Proposed Approach Design In addition, to create holistic transformations, the CityGML conceptual model must be assimilated or imported in its entirety into our pipeline. Because the model is 232 Communications Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] A semantic web representation of urban data divided into several modules which are sometimes interdependent and dependent on the GML standard, the complete model is defined by 43 different XML schema documents: 29 to define the GML 3.1 standard5 and 14 to define the CityGML 2.0 standard6 (not including xAL). Schema cannot be processed in isolation or we risk certain loss of data in the transformation (for example in distinguishing when to generate an owl:ObjectProperty from an owl:DatatypeProperty when the XML element type in question is declared outside the current schema). Our proposed approach chooses to consolidate these components in a composite XML schema, created from the definitions within each relevant schema in the conceptual model. This prerequisite step also requires that all definitions in the composite schema use normalized namespace prefixes to differentiate between references to one module or another. Once this is complete, the composite schema can be passed into the transformation pipelines. 5 http://schemas.opengis.net/gml/3.1.1/ 6 http://schemas.opengis.net/citygml/ 7 https://www.ogc.org/standards/geosparql 4.2. Creating a 3D Urban Data Ontology As an initial approach, we consider that the conceptual model can be expressed by XML schema and in a general simplification, XML schema can use xs:elements, xs:complexTypes, and xs:simpleTypes to define the concepts and semantics of domain specific urban. In addition, when creating the transformation from the CityGML conceptual model into an OWL ontology, it is important to reuse as many existing concepts as possible in the semantic web to enrich and render the model interoperable. Using the mapping transformations proposed in the previously mentioned works (Bedini et al., 2011; Bohring and Auer, 2015; Kramer et al., 2015; Vinasco-Alvarez et al., 2020), this can be efficiently achieved. Once the XML schema is transformed into a domain ontology, we can integrate the GeoSPARQL standard7. In GeoSPARQL, the geo:SpatialObject class represents any real world object with thematic and geospatial properties. Geometric properties of these classes are represented by geo:Geometry, a subclass of geo:SpatialObject, and can be serialized by geo:gmlLiterals. It is disjoint with the geo:Feature class, which represents the thematic properties of a geo:SpatialObject and can be linked to related geometric classes via the geo:hasGeometry property. As suggested in (Battle and Kolas, 2012), to integrate the thematic classes of our generated ontology with GeoSPARQL, we must declare our classes as a subclass of the geo:Feature class. Since our ontology is partially generated from the GML schema, any class which is derived from the gml:_Feature type is implicitly declared an rdfs:subClassOf geo:Feature. This also means that all of the geometry class declarations defined from the GML schema can be removed from the ontology as they are already defined in the 233 Communications Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] GeoSPARQL standard and by ISO 19136:20078 and references to these classes can be changed to http://www.opengis.net/ont/gml. GeoSPARQL standard and by ISO 19136:20078 and references to these classes can be changed to http://www.opengis.net/ont/gml. These proposed transformation strategies to create OWL ontologies from XML schema are an amalgam of previous transformation approaches with several specializations for the CityGML application schema. However, they are largely generalized and can be applied to schema outside of the main GML and CityGML schema, such as the external xAL addressing schema used by CityGML. 4.3. Creating Meaningful Transformations of Urban Data Instances When generating the CityGML instance to RDF transformation, it is important that the patterns created are general enough to be reused and can take advantage of the OWL/RDFS and GeoSPARQL vocabulary whenever possible. Several techniques suggested in (Brink et al., 2014) are implemented to achieve this. In addition, this process uses the general structure of the approaches proposed in (Bohring and Auer, 2015; Kramer et al., 2015; Vinasco-Alvarez et al., 2020) with consideration taken so that these mappings work with the ontology to be generated alongside the RDF data. In general, the following mappings in table 1 are applied to the composite schema to generate the CityGML to RDF transformation stylesheet. The generated CityGML to RDF template also uses gml:id attributes as rdf:ID or rdf:about whenever possible, as proposed in (Brink et al., 2014). If no gml:id is available, a unique id is generated and appended to the local name of the element. Also as suggested in (Brink et al., 2014), the rdf:type of each individual should be generated from its local name in order to reference the generated ontology from section 4.2, or the official GML ontology in the case of geometric elements. Table 1. General XSD to ‘XML to RDF XSLT’ mapping transformations XML Schema Concept Resulting XML to RDF XSLT global xs:element elements based on a xs:complexType xsl:template for generating owl:NamedIndividual of type xs:element that calls the template for its xs:complexType xs:element which are children of xs:complexType or xs:group xsl:template for owl:ObjectProperty or owl:DatatypeProperty based on the type of the xs:element xs:attribute of xs:complexType or xs:attributeGroup xsl:template for owl:DatatypeProperty xs:complexType, xs:attributeGroup, and xs:group xsl:template that calls relevant templates for possible properties or text the element could have Table 1. General XSD to ‘XML to RDF XSLT’ mapping transformations XML Schema Concept Resulting XML to RDF XSLT global xs:element elements based on a xs:complexType xsl:template for generating owl:NamedIndividual of type xs:element that calls the template for its xs:complexType xs:element which are children of xs:complexType or xs:group xsl:template for owl:ObjectProperty or owl:DatatypeProperty based on the type of the xs:element xs:attribute of xs:complexType or xs:attributeGroup xsl:template for owl:DatatypeProperty xs:complexType, xs:attributeGroup, and xs:group xsl:template that calls relevant templates for possible properties or text the element could have Table 1. General XSD to ‘XML to RDF XSLT’ mapping transformations 234 Communications 8 https://def.isotc211.org/ontologies/iso19136/ Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] A semantic web representation of urban data xs:simpleType with an xs:restriction or xs:extension of a native XML datatype xsl:template that selects the text of an element xs:simpleType with an xs:restriction or xs:extension of another xs:simpleType xsl:template that calls the relevant template for the xs:simpleType it is based on These transformations also map the geometry of a CityGML feature to an RDF triple using GeoSPARQL’s geo:asGML datatype property. This is applied whenever a branch of GML elements is detected which has a CityGML parent and is composed of only GML nodes. Such a branch represents a valid geo:gmlLiteral if all of its nodes are in the substitution group gml:_Geometry. This is ensured by placing the geo:gmlLiteral transformation within the template for gml:AbstractGeometryType, as any element within the same substitution group as this type can be serialized as a GML literal. This script also scans the transformed instances for malformed RDF triples and fully qualifies any RDF attributes that contain namespace prefixes such as rdfs:Resource, rdf:type, and rdf:about. These transformations follow the “garbage in, garbage out” concept that poorly formed data input into a program, will produce nonsensical results, and thus assume that the GML and CityGML instance documents provided are well structured and conform to their application schema. If this assumption is met, the resulting data should conform to the ontology transformation discussed in the previous section. 9 https://github.com/SemWebCentral/parliament 10 https://jena.apache.org 4.4. Leveraging the GeoSPARQL Standard Once the dataset was generated, a GeoSPARQL endpoint was installed to perform geospatial queries. The Parliament triple-store9 was used for this purpose as proposed in (Battle and Kolas, 2012) since it contains a SPARQL endpoint with GeoSPARQL support based on the Apache-Jena libraries10. We noted that if a particular coordinate reference system (CRS) is used, it must be declared in the literal value as a gml:srsName attribute in order for Parliament to parse and index the coordinates and geometry as the default reference system is <http://www.opengis.net/def/crs/OGC/ 1.3/CRS84>. Thus any GML data that uses a different reference system must include it in the GML instances themselves during transformation or manually add it to the output geo:gmlLiterals after transformation. This implementation adds these CRS declarations to the data through a Python script after the transformation since they are not declared in the source CityGML data. Through these practices, the geometry of CityGML instances can effectively be generated, stored, and queried through GeoSPARQL. 235 5. Discussion The CityGML conceptual model is a useful starting point for creating a 3D urban data ontology as a city information model. Its domain specific, modular structure provides a solid base for describing several hard domains of the urban ecosystem, which can be easily extended and integrated once in ontological form (Métral and Falquet, 2018). Even ADE extensions of CityGML could be applied to this approach to describe existing sub-domains of information if an XML schema is provided - such as the NoiseADE11 provided by the OGC. Another benefit of this modular approach is that the ontology can easily be pruned into domain specific sub-ontologies as each domain can be referenced by a unique URI namespace. In addition, enriching the ontology after transformation with existing geospatial standards is straightforward. However, a limitation of this approach is its reliance on the semantic limitations of XML Schema. For example, the recently declared ISO 19150-2 standardizes guidelines for UML to OWL transformation of geographic information standards. For GML, (Brink et al., 2014) proposes using Shapechange12, which takes advantage of this standard, to generate ontological models and acknowledges that ontology generation from UML and XML schema conceptual models are not semantically equivalent. When comparing the resulting ontologies between Shapechange and this proposed approach, still focusing on CityGML, initial explorations indicate that the ISO 19150-2:201513 ontology mappings are more expressive and direct than XML Schema to OWL. For example, an aggregation in UML can be converted into an OWL object property, while XML Schema expresses these entities as complex types which may be converted into extraneous OWL classes and properties to represent the same semantic information. Yet, Shapechange does not facilitate the transformation of CityGML or XML data into graph formats, and thus can only improve the city information model generated by our approach without supplemental transformations. Communications Communications Communications [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 11 http://schemas.opengis.net/citygml/examples/2.0/ade/noise-ade/CityGML NoiseADE.xsd 12 https://shapechange.net/ 13 Geographic information - Ontology - Part 2: Rules for developing ontologies in the Web Ontology Language (OWL), https://www.iso.org/standard/57466.html, Publication date: 2015-07 6. Conclusion In this article, we proposed a model centric approach towards semantic 3D urban data representations, where we consider that any data transformation from one data format to another format must conform to the equivalent underlying conceptual model. We implemented this proposed approach to convert the CityGML 2.0 conceptual model into an OWL ontology and to convert CityGML 2.0 data to RDF based on the conceptual model. Subsequently, geospatial queries were made on the resulting data using a GeoSPARQL endpoint. This open-source solution is available 236 Communications 13 Geographic information - Ontology - Part 2: Rules for developing ontologies in the Web Ontology Language (OWL), https://www.iso.org/standard/57466.html, Publication date: 2015-07 Acknowledgements The authors would like to thank University Lumière Lyon 2 for funding the PhD Thesis of Diego Vinasco-Alvarez and the LIRIS UMR 5205 Laboratory for funding his internship as part of a transversal effort with the GOAL team. This work has been done within the Vcity project (https://projet.liris.cnrs.fr/vcity/). Special thanks to Eric Boix for fruitful discussions and kind help. Data and code availability statement Detailed notes and the code for reproducing the results of section 4.2 and 4.3 can be found on software heritage14 under the folders ‘CityGML_Transformations/ XSD-to-OWL’ and ‘CityGML_Transformations/XML-to-RDF’ respectively. Communications SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] A semantic web representation of urban data A semantic web representation of urban data online and has been tested with the data from the 1st borough of Lyon (see the data and code availability statement below). Our future works will explore integrating other information sources - taking into consideration their conceptual models - and integrating a UML based approach. This approach will lead us to facilitate integration of 3D data in city information models and building information models. We are working to test this approach on CityGML 3.0. This will also demonstrate the application of this approach across different versions of standards. Bibliographie et références Barbosa, L., Pham, K., Silva, C., Vieira, M.R., Freire, J., 2014. Structured Open Urban Data: Understanding the Landscape. Big Data 2, 144–154. Battle, R., Kolas, D., 2012. Enabling the geospatial Semantic Web with Parliament and GeoSPARQL. Semantic Web 3, 355–370. https://doi.org/10.3233/SW-2012-0065 Bedini, I., Matheus, C., Patel-Schneider, P.F., Boran, A., Nguyen, B., 2011. Transforming XML Schema to OWL Using Patterns, in: 2011 IEEE Fifth International Conference on Semantic Computing. Presented at the 2011 IEEE Fifth International Conference on Semantic Computing, pp. 102–109. Biljecki, F., Kumar, K., Nagel, C., 2018. CityGML Application Domain Extension (ADE): overview of developments. Open Geospatial Data Softw. Stand. 3, 13. https://doi.org/10.1186/s40965-018-0055-6 Bohring, H., Auer, S., 2015. Mapping XML to OWL ontologies. Presented at the Leipziger Informatik-Tage (LIT 2005), Gesellschaft für Informatik e. V., pp. 147–156. Bohring, H., Auer, S., 2015. Mapping XML to OWL ontologies. Presented at the Leipziger Informatik-Tage (LIT 2005), Gesellschaft für Informatik e. V., pp. 147–156. Bork, D., Karagiannis, D., Pittl, B., 2020. A survey of modeling language specification techniques. Inf. Syst. 87, 101425. https://doi.org/10.1016/j.is.2019.101425 Brink, L., Janssen, P., Quak, W., Stoter, J., 2014. Linking spatial data: automated conversion of geo-information models and GML data to RDF. Int. J. Spat. Data Infrastruct. Res. 9, 59–85. https://doi.org/10.2902/1725-0463.2014.09.art3 Informatik-Tage (LIT 2005), Gesellschaft für Informatik e. V., pp. 147–156. Bork, D., Karagiannis, D., Pittl, B., 2020. A survey of modeling language specification techniques. Inf. Syst. 87, 101425. https://doi.org/10.1016/j.is.2019.101425 Brink, L., Janssen, P., Quak, W., Stoter, J., 2014. Linking spatial data: automated conversion Bork, D., Karagiannis, D., Pittl, B., 2020. A survey of modeling language specification techniques. Inf. Syst. 87, 101425. https://doi.org/10.1016/j.is.2019.101425 Brink, L., Janssen, P., Quak, W., Stoter, J., 2014. Linking spatial data: automated conversion of geo-information models and GML data to RDF. Int. J. Spat. Data Infrastruct. Res. 9, 59–85. https://doi.org/10.2902/1725-0463.2014.09.art3 14https://archive.softwareheritage.org/swh:1:dir:7a2459481066a4f77a433ec9a220ddfe5eaf f279;origin=https://github.com/VCityTeam/UD-Graph;visit=swh:1:snp:d95e3eaaa6422e7 7d63f1531be3901e71fe6744b;anchor=swh:1:rev:22b63103c50d21234961d1854c03cb3ab d4b2abc/ 237 Communications [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Chaturvedi, K., Kolbe, T., 2019. A Requirement Analysis On Extending Semantic 3d City Models For Supporting Time-Dependent Properties. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. IV-4/W9, 19–26. Claramunt, C., 2020. Ontologies for geospatial information: Progress and challenges ahead. J. Spat. Inf. Sci. https://doi.org/10.5311/JOSIS.2020.20.666 p p g De Paepe, D., Thijs, G., Buyle, R., Verborgh, R., Mannens, E., 2017. Automated UML-Based Ontology Generation in OSLO2. pp. 93–97. Ontology Generation in OSLO2. pp. 93–97. Jaillot, V., Servigne, S., Gesquière, G., 2020. Delivering time-evolving 3D city models for web visualization. Int. Bibliographie et références J. Geogr. Inf. Sci. 25 p. https://doi.org/10.1080/13658816.2020.1749637 Kramer, T.R., Marks, B.H., Schlenoff, C.I., Balakirsky, S.B., Kootbally, Z., Pietromartire, A., 2015. Software Tools for XML to OWL Translation. 2015. Software Tools for XML to OWL Translation. Kutzner, T., Chaturvedi, K., Kolbe, T.H., 2020. CityGML 3.0: New Functions Open Up New Applications. PFG – J. Photogramm. Remote Sens. Geoinformation Sci. 88, 43–61. Levina, O., 2012. Assessing Information Loss in EPC to BPMN Business Process Model Transformation, in: 2012 IEEE 16th International Enterprise Distributed Object Computing Conference Workshops. Presented at the 2012 IEEE 16th International Enterprise Distributed Object Computing Conference Workshops, pp. 51–55. Malinverni, E.S., Naticchia, B., Lerma Garcia, J.L., Gorreja, A., Lopez Uriarte, J., Di Stefano, F., 2020. A semantic graph database for the interoperability of 3D GIS data. Appl. Geomat. https://doi.org/10.1007/s12518-020-00334-3 Métral, C., Falquet, G., 2018. Extension And Contextualisation For Linked Semantic 3d Geodata, in: ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Presented at the 13th 3D GeoInfo Conference 2018 (Volume XLII-4/W10) - 1-2 October 2018, Delft, The Netherlands, Copernicus GmbH, pp. 113–118. Netherlands, Copernicus GmbH, pp. 113–118. Nuninger, L., Verhagen, P., Libourel, T., Opitz, R., Rodier, X., Laplaige, C., Fruchart, C., Leturcq, S., Levoguer, N., 2020. Linking Theories, Past Practices, and Archaeological Remains of Movement through Ontological Reasoning. Information 11, 338. https://doi.org/10.3390/info11060338 Ohori, K., Diakité, A., Krijnen, T., Ledoux, H., Stoter, J., 2018. Processing BIM and GIS Models in Practice: Experiences and Recommendations from a GeoBIM Project in The Netherlands. ISPRS Int. J. Geo-Inf. 7, 311. https://doi.org/10.3390/ijgi7080311 Rochet, C., Pinzon, J.D., 2016. Urban lifecycle management: A research program for smart government of smart cities. Rev. Gest. E Secr. 7, 1–20. https://doi.org/10.7769/gesec.v7i2.531 Samuel, J., Servigne, S., Gesquière, G., 2020. Representation of concurrent points of view of urban changes for city models. J. Geogr. Syst. 22, 335–359. https://doi.org/10.1007/s10109-020-00319-1 Tran, B.-H., Plumejeaud, C., Bouju, A., Bretagnolle, V., 2016. Towards a semantic framework for exploiting heterogeneous environmental data. Int. J. Metadata Semant. Ontol. 11. https://doi.org/10.1504/IJMSO.2016.081586 p g Usmani, A.U., Jadidi, M., Sohn, G., 2020. Automatic Ontology Generation Of BIM And GIS Data, in: ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Presented at the XXIV ISPRS Congress, Commission IV (Volume XLIII-B4-2020) - 2020 edition, Copernicus GmbH, pp. 77–80. https://doi.org/10.5194/isprs-archives-XLIII-B4-2020-77-2020 Vinasco-Alvarez, D., Samuel, J.S., Servigne, S., Gesquière, G., 2020. From CityGML to OWL (Technical Report). LIRIS UMR 5205. 238 Communications Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Etude des pratiques de déplacement dans une ville moyenne algérienne : le cas de Mila Heythem Adjeroud1,2, Adrien Lammoglia3, Laurent Chapelon4 1.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France haythem@hotmail.es 2.IGTU, Université Salah Boubnider Constantine 3, Algérie 3.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France lammoglia.adrien@gmail.com 4.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France lammoglia.adrien@gmail.com MOTS-CLES : Ville moyenne, Algérie, Transport artisanal, Enquête déplacement, Planification urbaine MOTS-CLES : Ville moyenne, Algérie, Transport artisanal, Enquête déplacement, Planification urbaine Heythem Adjeroud1,2, Adrien Lammoglia3, Laurent Chapelon4 1.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France haythem@hotmail.es 2.IGTU, Université Salah Boubnider Constantine 3, Algérie 3.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France lammoglia.adrien@gmail.com 4.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France lammoglia.adrien@gmail.com 4.Univ Paul Valéry Montpellier 3, LAGAM, F34000, Montpellier, France lammoglia.adrien@gmail.com 241 Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 RESUME ETENDU : Le paysage du transport urbain dans la plupart des villes algériennes s’est transformé après le désengagement de l’état par la libéralisation du marché des transports en 1988 (Boubakour, 2013), puis son réengagement à travers les EPIC dans les années 2000, il est aujourd’hui caractérisé par la coexistence d’un réseau public structurellement organisé, et d’un réseau artisanal auto-organisé (Lammoglia, 2013). Ce travail fait partie d’une étude portant sur Mila, ville moyenne du nord-est algérien, qui a pour finalité de mettre en exergue des particularités du système de transport très diversifié (Cf. Figure 1) dans cette ville et l’évaluation de sa performance, ainsi que l’identification des enjeux et des perspectives en termes de gouvernance et de planification. FIGURE 1. Réseaux de transport à Mila et les équipements générateurs de flux (Adjeroud 2020) FIGURE 1. Réseaux de transport à Mila et les équipements générateurs de flux (Adjeroud 2020) Cette partie de l’étude, qui vient à la suite d’une analyse territoriale préalable, sert à la construction d’une base de données géolocalisées pour la ville de Mila, se trouvant dans un contexte où l’accès à la donnée et à l’information est plutôt compliqué, et cherche à déterminer, de manière localisée les besoins de mobilité et de déplacement dans la ville, ainsi qu’à évaluer les niveaux de satisfaction des usagers par rapport au service fourni, et leur connaissance de l’offre. Ceci est concrétisé à partir d’une enquête réalisée auprès de 923 citoyens, à travers un échantillonnage empirique par quotas, obtenu après avoir réalisé des estimations de la démographie des différents quartiers à l’aide de chiffres officiels (très incomplets), de travail de terrain et d’exploitation des images satellitaires. Les niveaux de mobilité, de connaissance et de satisfaction, ainsi que la répartition des flux et le taux d’usage des transports en commun montrent les limites du système Posters 242 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Etude des pratiques de déplacement dans une ville moyenne algérienne : le cas de Mila Etude des pratiques de déplacement dans une ville moyenne algérienne : le cas de Mila de transport de la ville, constitué d’un réseau privé artisanal dense, à caractère atomisé (Godard, 2008), et relativement flexible, régi par des logiques de rentabilité, et d’un réseau public récent et peu développé, qui peine à s’élargir. RESUME ETENDU : In fine, malgré l’existence d’une complémentarité, notamment sur le plan spatial, ce double fonctionnement assure une desserte et une qualité de service qui n’est pas en adéquation avec les besoins des usagers. Il semble indispensable de repenser la forme de coexistence public/privé et les modes de gouvernance des transports urbains à Mila. Bibliographie : Adeniji K. (1987). Para-transit modes in Nigeria: problems and prospects. Cities n° 4 (4), 290–392. Adeniji K. (1987). Para-transit modes in Nigeria: problems and prospects. Cities n° 4 (4), 290–392. Adjeroud H., Lammoglia A., Chapelon L., & Kebiche A., (2019). Offre de transport et besoins de mobilité à Mila : les particularités d’une ville moyenne algérienne. Géotransports, n°10, p. 21-39. Bavoux J.J., Beaucire F., Chapelon L., Zembri P., (2005). Géographie des transports, Armand Colin, 232 p. Boubakour F., et Bencherif H., (2013). Évolution du transport urbain en Algérie : du mode artisanal à la régulation par l’État, Géotransports. Transport et développement des territoires, (1-2), p. 91-104. Boubakour F,. Merzaga A., (2006). Planification du transport urbain et développement durable : Quelques réflexions à propos de la ville de Batna (Algérie). Revue Transports, n° 437. Éditions techniques et économiques. Paris. Centre d’études sur les réseaux, les transports, l’urbanisme et les constructions publiques. (2013). Les enquêtes déplacements « standard CERTU », Note de présentation. Cervero R,. Golub A., (2007). Informal transport : A global perspective. Transport Policy, n°14 (6), p. 445-457. EMA/BETUR & CNEAP (2004). Enquête ménages sur les déplacements dans l’agglomération d’Alger. Godard X. (2002). Les transports et la ville en Afrique au sud du Sahara : le temps de la débrouille et du désordre inventif, KARTHALA Editions. Godard X., (2008). Transport artisanal, esquisse de bilan pour la mobilité durable, Codatu Hô Chi Minh Ville (Vietnam). Lammoglia A. (2013). Analyse et modélisation multi-agents de transports flexibles. Comparaison de services français et sénégalais. Thèse de doctorat, Université d’Avignon. 243 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Balzarini Raffaella1, Ployon Estelle1, Marcuzzi Mélanie1, Bourdeau Philippe1 Balzarini Raffaella1, Ployon Estelle1, Marcuzzi Mélanie1, Bourdeau Philippe1 1. Laboratoire Pacte, Univ. Grenoble Alpes, CNRS, Sciences Po Grenoble, 38000 Grenoble, France raffaella.balzarini@univ-grenoble-alpes.fr estelle.ployon@univ-grenoble-alpes.fr melanie.marcuzzi@umrpacte.fr philippe.bourdeau@univ-grenoble-alpes.fr MOTS-CLES : RECHERCHE-ACTION, SOCIO-ECOSYSTEME, MONTAGNE PEU AMENAGEE, REFUGES, PRATIQUES SPORTIVES ET TOURISTIQUES, TRANSITION, INNOVATION TERRITORIALE, OBSERVATIONS IN SITU, BASE DE DONNEES SPATIALISEE, SIG, CARTOGRAPHIE INTERACTIVE, GEO-VISUALISATION. 244 Posters Posters 244 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Les refuges de montagne constituent des observatoires privilégiés des effets croisés des changements environnementaux et culturels. Au-delà de leur fonction première d’hébergement, ils deviennent des laboratoires récréatifs dans lesquels des innovations de niche contribuent à l’émergence de modèles de développement, en phase avec les enjeux de diversification et de transition du tourisme dans « la montagne peu aménagée » (Bourdeau et al, 2019). Or, face à un déficit de connaissances majeur sur la fréquentation et la gouvernance de la montagne « peu aménagée », un programme de recherche-action « Refuges Sentinelles (RS) », a été développé depuis 2017, par le Parc National des Écrins (PNE) et le LabEx Innovation et Transitions Territoriales en Montagne de l’Université Grenoble Alpes (UGA), en partenariat avec l’Agence Française de la Biodiversité, dans le cadre du programme Sentinelles des Alpes (Girard, 2019). Ce dernier vise à co-construire une vision alpine des relations climat-homme- biodiversités sur la base de suivis sur le long-terme. Dans ce contexte, RS met au cœur de son suivi les transformations environnementales et culturelles dont les refuges sont témoins. À l’échelle d’un panel de 16 refuges du PNE destiné à couvrir des configurations diversifiées en matière d’altitude, de processus naturels et de contextes touristiques, les principaux axes de recherche de RS portent sur les fréquentations et pratiques, en lien avec les métiers touristiques et sportifs, la météorologie, la biodiversité, la géomorphologie, les risques et la sécurité (Mourey et Ravanel, 2017a). L’objet d’étude est « le pratiquant » qui est observé à travers la mise en place de protocoles adaptés au contexte de la haute montagne déconnectée (Bourdeau et al., 2018). Les protocoles comportent la collecte de données quantitatives (i.e. : nombre de nuitées par saison dans un refuge) et qualitatives (i.e. : profils sociodémographiques issus d’enquêtes, questionnaires), d’observations in situ (i.e. : carnet de récit), de photo-constats, d’ateliers collaboratifs. Si la démarche scientifique implique l’ensemble des parties prenantes professionnelles et territoriales concernées (fédérations sportives, syndicats professionnels, offices du tourisme), qui participent à l’orientation du programme, les gardien.ne.s, de refuge sont les protagonistes du dispositif car le déploiement des protocoles s’appuie sur leur implication (Bourdeau, 2018). L’exploitation des données et les indicateurs de fréquentation qui en découlent (i.e. Observations d’évolutions socio-écosystémiques en haute montagne Initialement fortement exploratoire, le principal défi du dispositif Refuges Sentinelles a été de passer d’une logique d’expérimentations de terrain, ponctuelles et locales, à la co-construction d’une méthodologie d’observation pérenne et reproductible dans d’autres massifs. C’est grâce au rôle central de la géomatique que ce défi a pu être relevé. La mise en place d’une base de données (PostGres-PostGis) et d’un atlas cartographique interactif (ArcGIS Online) a permis de mettre en cohérence les données hétérogènes pour qu’elles puissent être corrélées aux autres infrastructures de suivi existantes, notamment celles de Sentinelles des Alpes. La logique ‘géomatique’ a eu un impact conceptuel et structurel dans le dispositif RS. Conceptuellement, elle a amené à redéfinir les questions de recherche autour de la dimension géographique de l’objet ‘refuge’ considéré comme ‘point’ cartographique, camp de base et laboratoire avancé d’observations, mais aussi comme catalyseur des flux et d’interactions (les itinéraires empruntés, les zones de divagations, les croisements homme-biodiversité, etc..) et part d’un maillage territorial d’hébergements semblables, à l’échelle d’un massif de montagne. La logique ‘géomatique’ a été révélatrice de l’importance d’une approche spatiale des protocoles d’observation. Structurellement, la base de données et l’atlas interactif de RS ont permis de cadrer les nouveaux protocoles de terrain en standardisant des modèles descriptifs des données, de faciliter l’interopérabilité des données multi- sources et de fédérer les différentes disciplines qui se croisent dans son contexte. Notre poster présentera le processus de cette démarche d’intelligence collective entre chercheurs et praticiens, que Refuges Sentinelles a initié dans la durée ; il mettra en évidence comment les outils de la géomatique qui ont supporté cette action, induisent une valeur ajoutée informationnelle et stratégique à l’échelle du dispositif, en contribuant à la co-construction, participative et pédagogique, de connaissances. Particulièrement, le poster montrera comment les outils de la géomatique ont joué un rôle cognitif dans la visualisation d’informations agrégées (Balzarini et al., 2018) permettant aux gardiens de mieux comprendre comment leur refuge contribue et s’adapte aux changements, à l’échelle locale et du massif. Mourey J. et Ravanel L. (2017a). Évolution des itinéraires d’accès aux refuges du bassin de la Mer de Glace (massif du Mont Blanc, France). Revue de Géographie Alpine. Posters : quota de remplissage des refuges, nombre de pratiquants par randonnée et/ou par sommet à partir d’un refuge, % des typologies des profils des pratiquants) sont restitués aux gardiens à la fin de chaque saison lors du débriefing interprofessionnel qui permet de mutualiser les observations et de croiser le regard des chercheurs et des parties prenantes. RS est caractérisé par une configuration éparse, dans l’espace et dans le temps : des refuges localisés dans des zones très diverses, des saisonnalités différenciées, ce qui implique des protocoles de suivi qui produisent des données multi-source, multi- format et multi-échelle. Ceci explique l’importante approche exploratoire, nécessaire à la validation de la méthodologie qui a abouti au suivi saisonnier des flux de fréquentations et de destinations (présences dans les refuges des pratiquants et leurs randonnées effectuées) et au suivi pluriannuel (pas de temps de 4 ans) des typologies des usagers (des refuges). 245 Posters 245 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Observations d’évolutions socio-écosystémiques en haute montagne Bibliographie non numérotée et références Balzarini R., Ployon E., Marcuzzi M. (2018). La montagne en été : une approche empirique pour étudier spatialement le changement des pratiques, Conférence SIG ESRI 2018, Paris. Balzarini R., Ployon E., Marcuzzi M. (2018). La montagne en été : une approche empirique pour étudier spatialement le changement des pratiques, Conférence SIG ESRI 2018, Paris. Bourdeau Ph., Marcuzzi M., Dentant C., Charron J., Bonet R. (2019). Mountain huts: laboratories for environmental and cultural change, International Mountain Conference, Innsbruck. Bourdeau Ph. (2018). Guides et gardiens de refuge, pionniers de l’adaptation ! In Impacts du changement climatique et transition(s) dans les Alpes du Sud, Les cahiers du GREC-SUD, édités par l’Association pour l’innovation et la recherche au service du climat (AIR), p. 36. Girard V. (2019). "Recherche action" Sentinelles des Alpes, des scientifiques impliqués sur le terrain. Montagne Magazine, Numéro Spécial #469 – Octobre 2019. Mourey J. et Ravanel L. (2017a). Évolution des itinéraires d’accès aux refuges du bassin de la Mer de Glace (massif du Mont Blanc, France). Revue de Géographie Alpine. 246 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Rabia Sarah Chériguène 1,2, Anne Puissant1 et Habib Mahi 2 1. Laboratoire Image, Ville et environnement, UMR 7362 CNRS / Université de Strasbourg, 3 rue de l’Argonne, 67000, Strasbourg, France. Strasbourg, 3 rue de l’Argonne, 67000, Strasbourg, France. 2. Centre des Techniques Spatiales, 1 Avenue de la Palestine, 31200, Arzew, Algérie. 2. Centre des Techniques Spatiales, 1 Avenue de la Palestine, 31200, Arzew, Algérie. rabia-sarah.cheriguene@ live-cnrs.unistra.f r rabia-sarah.cheriguene@ live-cnrs.unistra.f r MOTS-CLES : OCCUPATION DES SOLS, HAUTE RÉSOLUTION SPATIALE, CLASSIFICATION SUPERVISÉE, ALSAT-2A. MOTS-CLES : OCCUPATION DES SOLS, HAUTE RÉSOLUTION SPATIALE, CLASSIFICATION SUPERVISÉE, ALSAT-2A. La croissance démographique rapide et incontrôlée ainsi que le développement économique et industriel, en particulier dans les pays en développement, ont multiplié à plusieurs reprises le taux de changement d’utilisation des sols/ couverture des sols. (Talukdar et al., 2020), la ville d’Oran fait partie des espaces dans lesquels la problématique du développement démographique est très pertinente dans la mesure où les atteintes à l’environnement y sont généralisées. L’objectif de ce travail est d’exploiter les images ALSAT-2A et de cartographier par une approche supervisée les modes d’occupations des sols afin de quantifier les changements entre les années 2010 et 2018. Différents tests ont été effectués sur l’image 2018, d’une part en faisant varier les données en entrée de la classification et d’autre part en comparant deux algorithmes d’apprentissage supervisé : (1) les Forêts Aléatoires (RF) et les Supports à Vecteur de Machines (SVM) introduites par (Birmann, 2001) et (Vapnick, 1995). Les résultats de classification sont évalués à l’aide de différentes métriques quantitatives et qualitatives. Les mêmes tests seront appliqués sur l’image de 2010. Apport des images ALSAT-2A pour la cartographie de l’occupation des sols - le cas de la ville d’Oran Rabia Sarah Chériguène 1,2, Anne Puissant1 et Habib Mahi 2 2. Classification supervisée, protocoles de test et résultats La première étape consiste à définir la typologie des classes thématiques à extraire. Huit classes ont été définies basées sur les connaissances de terrain et des échantillons ont été sélectionnés par photo-interprétation pour les classes d’eau (en 3 classes – Mer, Etang et Sebkha), la forêt, le jachère, le maraîchage, la céréaliculture, la végétation, le bâtis/sol nu ainsi que la classe route issue des données OpenStreetmap. Une analyse de la séparabilité de ces échantillons a permis d’évaluer la qualité de notre jeu de données destiné à l’entraînement et à la validation des modèles de classification et de vérifier leur pertinence. Dans un second temps, les deux algorithmes RF et SVM ont été appliqués sur les résultats de l’ACP. Dans le premier test, la première CP est utilisée en entrée de la classification avec un pourcentage de valeurs propres de 80%. Les deuxièmes et troisièmes tests ont consisté à prendre en entrée de la classification, les deux premières, puis les trois premières avec un taux de 16% et 4%. La CP1 est composée des bandes proche infrarouge, rouge, vert et de l’indice de brillance, la CP2 est composée de la bande proche infrarouge et la CP3 est composée de la bande vert et rouge. Les résultats ont montré que le RF donne de meilleurs résultats (93.22%) par rapport au SVM (77.16%). La performance globale de la classification en appliquant l’algorithme RF est de 92.98% sur les deux composantes comparativement 93.22% lorsque la classification utilise les trois composantes en entrée. (Table 1). TABLE 1. Résultats de classification RF CP1 CP1+CP2 CP1+CP2+CP3 Précision globale 66.96% 92.98% 93.22% F-score 0.67 0.93 0.93 1. Données utilisées et prétraitements Les images satellites utilisées sont des images ALSAT-2A couvrant la région d’Oran de l’année 2018, avec une résolution spatiale de 10m en mode multispectrale (Bleu, Vert, Rouge et Proche Infrarouge) et de 2.5 m de résolution spatiale en mode panchromatique. Les images sont corrigées au niveau 2A et ont été orthorectifiées. Le choix de l’agglomération d’Oran comme terrain de recherche a été fait sur la base d’une connaissance préalable du terrain ainsi que sur la disponibilité des images. 247 Posters 247 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 L’enjeu de cette étude est de discriminer les différents modes d’occupation des sols à 10 m de résolution spatiale à partir des bandes spectrales et de différents indices spectraux de végétation (NDVI), de brillance (IB) ou texturaux. L’intérêt de l’utilisation des indices de ce type pour caractériser l’état des sols particulier dans les zones urbaines, a été démontré dans plusieurs travaux (Schaaf et al.,2003 ; Pelletier, 2017) L’analyse de texture est fondée sur deux paramètres l’homogénéité et l’entropie, calculés à partir de l’indice NDVI sur une fenêtre de 5 x 5 en raison de la résolution spatiale de 10 m (Puissant et al., 2005). Une Analyse en Composantes Principales (ACP) a été aussi appliquée sur les 7 bandes /indices afin de réduire la corrélation statistique entre les bandes. Apport des images ALSAT-2A pour la cartographie Apport des images ALSAT-2A pour la cartographie Tlemcen, en utilisant le même modèle d’apprentissage, afin de vérifier la transférabilité de la méthode sur une ville aux caractéristiques morphologiques assez proche. Tlemcen, en utilisant le même modèle d’apprentissage, afin de vérifier la transférabilité de la méthode sur une ville aux caractéristiques morphologiques assez proche. SAGEO’2021 – La Rochelle, 5-7 mai 2021 3. Conclusion et perspectives Les résultats de différents tests ont montré que le RF donne de bons résultats en utilisant trois composantes principales 93.22% à partir de 7 bandes spectrales/indices. Cette méthode sera appliquée sur l’image de 2010 et les changements d’occupation des sols seront analysés et présentés dans le poster proposé. La méthode sera également appliquée sur notre deuxième ville test 248 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Antoine COLLIN1,2, Tony DUROZIER3, Gwenal HERVOUET4, Stéphane COSTA5 1. EPHE-PSL Université, CNRS LETG 15 boulevard de la mer, 35800 Dinard, France antoine.collin@ephe.psl.eu 1. EPHE-PSL Université, CNRS LETG 15 boulevard de la mer, 35800 Dinard, France antoine.collin@ephe.psl.eu 2. LabEx CORAIL 58 avenue Paul Alduy, 66860 Perpignan, France MOTS-CLES : HAUSSE DU NIVEAU MARIN, STRATEGIE DE RECUL, DEPOLDERISATION, LANCIEUX, SATELLITE TRES HAUTE RESOLUTION TEMPORELLE, PLANETSCOPE. 250 Détecter le processus de dépoldérisation par l’observation satellite à très haute résolution temporelle Antoine COLLIN1,2, Tony DUROZIER3, Gwenal HERVOUET4, Stéphane COSTA5 1. EPHE-PSL Université, CNRS LETG 15 boulevard de la mer, 35800 Dinard, France antoine.collin@ephe.psl.eu 2. LabEx CORAIL 58 avenue Paul Alduy, 66860 Perpignan, France 3. Conservatoire du Littoral et des Rivages Lacustres – Délégation Normandie 5 avenue Tsukuba, 14200 Hérouville-Saint-Clair, France t.durozier@conservatoire-du-littoral.fr 4. Conservatoire du Littoral et des Rivages Lacustres – Délégation Bretagne 8, quai Gabriel Péri, 22194 Plérin, France g.hervouet@conservatoire-du-littoral.fr 5. Université Caen Normandie, CNRS LETG Esplanade de la Paix, 14 000, Caen, France stephane.costa@unicaen.fr MOTS-CLES : HAUSSE DU NIVEAU MARIN, STRATEGIE DE RECUL, DEPOLDERISATION, LANCIEUX, SATELLITE TRES HAUTE RESOLUTION TEMPORELLE, PLANETSCOPE. Détecter le processus de dépoldérisation par l’observation satellite à très haute résolution temporelle Antoine COLLIN1,2, Tony DUROZIER3, Gwenal HERVOUET4, Stéphane COSTA5 1. EPHE-PSL Université, CNRS LETG 15 boulevard de la mer, 35800 Dinard, France antoine.collin@ephe.psl.eu 2. LabEx CORAIL 58 avenue Paul Alduy, 66860 Perpignan, France 3. Conservatoire du Littoral et des Rivages Lacustres – Délégation Normandie 5 avenue Tsukuba, 14200 Hérouville-Saint-Clair, France t.durozier@conservatoire-du-littoral.fr 4. Conservatoire du Littoral et des Rivages Lacustres – Délégation Bretagne 8, quai Gabriel Péri, 22194 Plérin, France g.hervouet@conservatoire-du-littoral.fr 5. Université Caen Normandie, CNRS LETG Esplanade de la Paix, 14 000, Caen, France stephane.costa@unicaen.fr MOTS-CLES : HAUSSE DU NIVEAU MARIN, STRATEGIE DE RECUL, DEPOLDERISATION, LANCIEUX, SATELLITE TRES HAUTE RESOLUTION TEMPORELLE, PLANETSCOPE. Références Breiman L. (2001). Random Forests. Machine Learning, 45, 5-32. Pelletier C. (2017). Cartographie de l’occupation des sols à partir de séries temporelles d’images satellitaires à hautes résolutions. Thèse de doctorat, Université de Toulouse. Puissant A., Hirsh J., Weber C., (2005). The utility of texture analysis to improve per pixel classification for high spatial resolution imagery. International Journal of Remote Sensing, 26 (4), pp. 733-745. Schaaf C. Strahler A. Hodges J. Gao F. Reed B. Huete A. Zhang X., Friedl M. Monitoring vegetation phenology using modis. Remote Sensing of Environment.84, 471 -475, 2003. Talukdar S., Singha p., Mahato S., Shahfahad., Pal S., Liou Y., Rahman A., (2020). Land- Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations, Remote sening review, 12, 1135; doi:10.3390/rs12071135. Vapnick V N. (1995). The Nature of Statistical Learning Theory. Springer Verlag New York. 249 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 RESUME ETENDU. L’interface littorale est le siège de processus complexes, où se combinent les hausses du niveau marin, de la démographie et des enjeux socio-économiques. L’augmentation de ces facteurs contribue à l’aggravation du risque de submersion marine. Plusieurs stratégies existent pour lutter contre ce risque : la plus résiliente est le recul des enjeux, de la mer vers la terre. Ainsi, des zones terrestres gagnées sur la mer au fil des siècles font désormais l’objet de réflexions de dépoldérisation (Bazin et Olivri, 2017). Le Conservatoire du littoral anime le projet « Adapto » en accompagnant sur 10 sites naturels, dont la baie de Lancieux, une gestion souple du trait de côte, pouvant aboutir à des recompositions territoriales significatives. En mars 2020, ce site breton, incluant un polder doté d’une digue érigée au XVIIIème siècle, a connu un basculement notable dans sa gestion avec l’apparition d’une brèche dans l’ouvrage, laissant les hautes mers de vive eau pénétrer sur les parcelles poldérisées, acquises par le Conservatoire. Suivie depuis quatre ans, la baie ne bénéficie pas d’une couverture spatiale continue. En effet, les photographies au sol, dotées d’une résolution millimétrique, ne permettent pas d’embrasser la dynamique du site. D’autre part, le suivi par imagerie aérienne ou satellitaire traditionnelle, pourvue d’une résolution, respectivement, centimétrique et décimétrique, est entaché par une faible résolution temporelle, empêchant l’acquisition durant les marées de vive eau, (Collin et al., 2017). Ce travail propose le premier suivi de dépoldérisation par l’utilisation d’imagerie satellitaire à très haute résolution temporelle. FIGURE 1. Image de réflectance infrarouge du site d’étude pour octobre (a) 2016, (b) 2017, (c) 2018, (d) 2019, et (e) 2020. Croix : brèche ; cercles : stations suivies. FIGURE 1. Image de réflectance infrarouge du site d’étude pour octobre (a) 2016, (b) 2017, (c) 2018, (d) 2019, et (e) 2020. Croix : brèche ; cercles : stations suivies. 251 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Dépolderisation par satellite THR temporelle Dépolderisation par satellite THR temporelle Dépolderisation par satellite THR temporelle La partie méridionale du polder de la Baie de Lancieux a ainsi été suivie après chaque grande marée d’équinoxe, lors de la première quinzaine d’octobre, entre 2016 et 2020 (Figure 1). L’imagerie infrarouge PlanetScope, dotée d’une résolution de 3 m, a été géométriquement et radiométriquement corrigée, résultant en une réflectance de surface comparable pour les cinq années. Collin A., Dubois S., James D., Houet T. (2019). Improving intertidal reef mapping using UAV surface, red edge, and near-infrared data. Drones, vol. 3, n° 3, p. 67-. Collin A., Etienne S., Feunteun E. (2017). VHR coastal bathymetry using WorldView-3 : colour versus learner. Remote Sensing Letters, vol. 8, n° 11, p. 1072-1081. Bazin P., Olivri D. (2017). Le Conservatoire du littoral: un acteur face au changement climatique. Sciences Eaux Territoires, vol. 3, p. 50-53. Bazin P., Olivri D. (2017). Le Conservatoire du littoral: un acteur face au changement climatique. Sciences Eaux Territoires, vol. 3, p. 50-53. Matthieu DUBUIS12, Rémi RATAJCZAK3, Lény GRASSOT12, Laure TOUGNE3, Béatrice FERVERS12 1. Centre Léon Bérard, Département Prévention Cancer Environnement – Lyon 2. INSERM UA8 Radiations : Défense, Santé, Environnement – Lyon 3. CNRS UMR 5205 LIRIS, Université Lumière Lyon 2 – Lyon MOTS-CLÉS : OCCUPATION DU SOL, PRISE DE VUES AÉRIENNES HISTORIQUES, DÉTERMINATION SEMI-AUTOMATIQUE, LOGICIEL SIG RESUME ETENDU. La variance élevée de la réflectance infrarouge du site d’étude (Figure 2a) indique une évolution notable de la station du polder au nord-est de la brèche (cercle rose) en 3 phases : fort déclin de 2016 à 2017, plateau de 2017 à 2019, et faible déclin de 2019 à 2020 (Figure 2b). FIGURE 2. Variance spatiale (a) et évolution pluriannuelle (b) de la réflectance infrarouge sur sept stations d’intérêt, dont la rose est la plus proche de la brèche. FIGURE 2. Variance spatiale (a) et évolution pluriannuelle (b) de la réflectance infrarouge sur sept stations d’intérêt, dont la rose est la plus proche de la brèche. Cette tendance de réduction de la réflectance infrarouge témoigne de l’augmentation de la saturation en eau (Collin et al., 2019) de la station, avant que la brèche ne soit apparue en mars 2020. Il est important de souligner que la brèche a accentué ce processus. La variabilité spatio-temporelle de cette dépoldérisation fera l’objet d’analyses via PlanetScope sur d’autres stations d’intérêt, comme les zones d’accumulations sédimentaires et de colonisation des communautés d’habitats intertidaux. La très haute résolution temporelle, idoine à la constellation PlanetScope, atteint la journée. Couvrant toutes les surfaces terrestres à la résolution de 3 m, cette constellation civile est la seule à offrir cette résolution temporelle, permettant de répondre adéquatement aux exigences de la saisonnalité de la variation de hauteurs d’eau dans des sites mégatidaux. Cet apport technologique est aussi très prometteur pour d’autres suivis littoraux d’intérêt, comme l’évolution des falaises, dunes, prés salés, mangroves, herbiers marins, algueraie, ou encore récifs coralliens. 252 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Détermination semi-automatique de l’occupation du sol grâce au logiciel GOURAMIC : évaluation des performances sur données historiques et opérateur- dépendance Matthieu DUBUIS12, Rémi RATAJCZAK3, Lény GRASSOT12, Laure TOUGNE3, Béatrice FERVERS12 1. Centre Léon Bérard, Département Prévention Cancer Environnement – Lyon 2. INSERM UA8 Radiations : Défense, Santé, Environnement – Lyon 3. CNRS UMR 5205 LIRIS, Université Lumière Lyon 2 – Lyon Matthieu DUBUIS12, Rémi RATAJCZAK3, Lény GRASSOT12, Laure TOUGNE3, Béatrice FERVERS12 1. Introduction L’étude ci-présentée s’intègre au sein du volet environnement de « l’étude prospective multicentrique de type cas-témoins visant à étudier l’association entre expositions environnementales – domestiques – professionnelles et le risque de tumeur germinale du testicule à l’âge adulte » dit TESTIS (Centre Léon Bérard / OMS CIRC / UMR LIRIS). Elle a pour objectif l’analyse de l’association entre différentes expositions, au premier rang desquelles les pesticides, tout au long de la vie et particulièrement pendant les périodes de sensibilité critique du développement du testicule et le risque de développer un cancer à l’âge adulte. Le logiciel GOURAMIC (1) intervient dans le volet évaluation des expositions environnementales afin de déterminer l’occupation du sol autour d’une adresse à une date donnée. Il doit permettre de reconstituer tout au long des parcours de vie des sujets de l’étude, les occupations du sol environnant les lieux de résidence successivement habités. L’objectif ici est d’évaluer les performances du logiciel GOURAMIC tant dans la finesse spatiale de ses résultats que dans la variabilité inhérente à une utilisation qui en est faite par différents utilisateurs. 253 Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 I – la première analyse consiste à comparer les performances du logiciel GOURAMIC au regard d’autres méthodes de détermination de l’occupation du sol. Cet aspect vise à comparer les résultats traitement avec les différentes sources disponibles d’occupation du sol sur des périodes anciennes à l’échelle du territoire national ou d’espaces particuliers. II – la seconde analyse consiste à mesurer la variabilité induite par les utilisateurs du logiciel GOURAMIC : quantifier les différences d’interprétation et donc des résultats spatialisés induites par le traitement par les différents opérateurs d’un même espace dans GOURAMIC. (1) Ratajczak R. (2020). Analyse automatique d'images aériennes historiques : application à une étude épidémiologique. Thèse de doctorat en Informatique, École doctorale en informatique de Lyon. 2.1. Comparer les performances du logiciel GOURAMIC Les matrices de confusion nous ont permis de comprendre quelles étaient les catégories les plus difficiles à repérer et sur lesquelles un travail supplémentaire doit être réalisé. Ainsi, HILDA présente de nombreuses difficultés dues notamment à sa précision (mailles très importantes) ; GOURAMIC et deepGOURAMIC disposent de bonnes tendances par rapport à CLC mais divergent sur certaines catégories (vergers et vignobles) où la finesse de CLC (25Ha minimum) tend à limiter la détection de ces parcelles habituellement de faible surface. 2. MÉTHODOLOGIE Ces résultats sont croisés avec les images de référence : l’analyse des écarts permettant dès lors de quantifier l’impact de l’interprétation dans le processus de détermination de l’occupation du sol. 2. MÉTHODOLOGIE Plusieurs contraintes ont été identifiées : ancienneté des données, non disponibilité d’une vérité terrain, diversité des sources et des méthodes. Considérant ces difficultés, nous avons intégré les différentes sources disponibles à une date relativement ancienne et commune : Corine Land Cover (COPERNICUS, éd.1990), HILDA (HIstoric Land Dynamics Assessment, éd.1990), OCS Picardie (édition 1994). Nous avons de plus, ajouté une version automatisée de GOURAMIC – appelée DeepGOURAMIC – utilisant un réseau de neurones pour reconstituer l’occupation du sol sans intervention opérateur à partir des jeux de données générées par les utilisateurs sur plus de 1000 images. Se rajoutent enfin les résultats issus de GOURAMIC de différents utilisateurs ainsi que des images dites de référence travaillées manuellement par un utilisateur expérimenté. Pour toutes les sources, un travail préliminaire d’harmonisation des items d’occupation du sol a été fait pour agencer 7 classes distinctes : Arable – Eau – Forêt – Prairie – Urbain – Verger – Vignoble. L’interprétation des espaces en eau étant difficile sur les photographies aériennes, notamment dans GOURAMIC et deepGOURAMIC, et le fait que cette classe ne soit pas stratégique dans l’étude des pesticides (non ciblée par l’épandage) il a été décidé de la retirer de la comparaison, à l’aide d’un masque (BDTopo hydrologie de l’IGN) pour l’ensemble des couches étudiées. L’étude croisée des résultats s’intéressera à l’analyse des accords entre sources ou utilisateurs grâce aux kappa de Cohen. Pour ce faire, toutes les couches résultats ont été intersectées pour découper l’ensemble des territoires analysés en polygones disposant d’une unique occupation du sol par source ou utilisateur. La performance de GOURAMIC au regard des autres sources utilisera comme jeu de données initial 206 zones tampon d’environ 7 km² (GOURAMIC – deepGOURAMIC – HILDA – CLC) ; et à cause du périmètre très restreint de cette couche locale seulement 4 avec la source OCS Picardie. La variabilité inter-opérateur aura pour données les résultats issus d’un test d’utilisation par différents opérateurs, préalablement formés sur le logiciel et devant 254 Posters 254 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Détermination semi-automatique de l’occupation du sol grâce au logiciel GOURAMIC 9 images classée par difficulté d’interprétation. Ces résultats sont croisés avec les images de référence : l’analyse des écarts permettant dès lors de quantifier l’impact de l’interprétation dans le processus de détermination de l’occupation du sol. 9 images classée par difficulté d’interprétation. 2.2. Mesurer la variabilité induite par les utilisateurs Comme pressenti, les résultats confirment la faible variabilité générale à l’échelle d’un territoire mais des écarts néanmoins forts pour certaines classes d’occupation du sol induite par des interprétations différentes au sein d’une même image par différents utilisateurs. Ainsi, les catégories Prairies et Arable sont souvent confondues, qui plus est dans les tests sur les images disposant de tous les types de surface et donc où la place à l’interprétation est la plus importante. 255 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Amélie Duquesne1, Christine Plumejeaud-Perreau1, Jean-Michel Carozza1 1. Littoral Environnement et Sociétés (LIENSs), La Rochelle Université 2 rue Olympe de Gouges, 17000 La Rochelle, France amelie.duquesne1@univ-lr.fr 1. Littoral Environnement et Sociétés (LIENSs), La Rochelle Université 2 rue Olympe de Gouges, 17000 La Rochelle, France amelie.duquesne1@univ-lr.fr MOTS-CLES : ANALYSE MULTICRITERE, ANALYSE SPATIALE, SYSTEME D’INFORMATION GEOGRAPHIQUE, CARTOGRAPHIE D’IMPACTS ANTHROPIQUES, RESTAURATION DES COURS D’EAU. 256 Application d’une approche semi- quantitative pour estimer les impacts cumulés des interventions anthropiques sur les systèmes fluviaux : l’exemple de la Charente entre Angoulême et Saintes (Ouest de la France) Amélie Duquesne1, Christine Plumejeaud-Perreau1, Jean-Michel Carozza1 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 La quantification des impacts anthropiques cumulés sur les milieux « naturels » a fait l’objet de travaux en particulier pour les milieux estuariens, côtiers et marins (e.g., Halpern et al., 2007 ; Pereira et al., 2019). Ces approches restent fortement discutées en raison des limites intrinsèques liées à la méthode employée (Toro et al., 2010). Alors que l’anthropisation des cours d’eau est une thématique en vogue (Brown et al., 2018), ils n’ont pas fait l’objet de tentatives d’application de ces approches. Ils ont en effet été sujets à un aménagement massif pour la navigation commerciale, la force motrice de l’eau ou encore la subsistance des populations. Certains de ces usages sont à l’heure actuelle sur le déclin ou à l’arrêt, mais les paysages fluviaux en conservent les traces et impacts. A cela s’ajoute, de nouveaux usages (navigation de plaisance, activités récréatives et sportives de plein air, etc) dont les impacts s’inscrivent dans un milieu déjà très anthropisé. De ce fait, la Directive cadre européenne sur l’eau déclare que 52% des cours d’eau français sont sujets à une dégradation générale de leur état écologique en 2009. La pollution des eaux, la faible continuité écologique et sédimentaire et les altérations morphologiques et hydrologiques sont les principaux obstacles à l’atteinte du bon état écologique des masses d’eau. Les gestionnaires de cours d’eau sont alors à l’origine de projets de restauration dont l’objectif est de rétablir les fonctionnalités des milieux fluviaux ou de les recréer, ou de reconstituer un état antérieur de l’écosystème fluvial dans une optique de gestion durable. Cependant, ces pratiques de restauration se heurtent à un déficit de connaissances liées aux incertitudes sur le fonctionnement écologique, à la faible présence de programmes interdisciplinaires et au manque de retours d’expérience. Les actions de restauration écologique sont priorisées par les acteurs car elles sont plus faciles à mettre en œuvre que des actions de restauration hydromorphologique ou sur la qualité de l’eau (logistique, financement, accès au foncier, etc). Le fleuve Charente ne fait pas exception. L’état des lieux initial du bassin versant de la Charente effectué dans le cadre de l’élaboration du Schéma Directeur d’Aménagement et de Gestion des Eaux (SDAGE 2010-2015) conclue à une dégradation générale de l’ensemble des cours d’eau charentais. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Impacts anthropiques cumulés sur le fleuve Charente Impacts anthropiques cumulés sur le fleuve Charente récents et de leurs impacts cumulés afin de les aider dans la mise en œuvre d’actions adaptées et durables de gestion et de restauration sur la Charente. récents et de leurs impacts cumulés afin de les aider dans la mise en œuvre d’actions adaptées et durables de gestion et de restauration sur la Charente. Ce travail propose alors d’analyser et de quantifier les impacts cumulés des interventions anthropiques sur le fleuve Charente entre Angoulême et Saintes sur la période 1700-2020. Un inventaire des interventions humaines a été réalisé sur la période d’étude à partir de données géohistoriques récupérées auprès des Archives départementales de Charente-Maritime et de Charente et municipales de Saintes et Cognac, et intégré dans une base de données. Les interventions humaines ont ensuite été appréhendées à partir d’une approche globale semi-quantitative et spatialisée. Elle s’appuie sur le calcul d’un indice d’impacts cumulés unitaire estimé pour chaque intervention humaine (ex : dragage…), et d’un indice d’impacts cumulés total évalué pour chaque type d’intervention humaine (ex : entretien du lit…). Le modèle d’impacts cumulés est estimé via une méthode multicritère basée sur une série d’attributs qualitatifs (zone impactée, impact spatialisé, impact fonctionnel…) déterminés par dire d’expert.e.s. Les données obtenues sont spatialisées sur un maillage couvrant 241 km² disposé sur l’ensemble de la plaine alluviale afin d’évaluer les impacts anthropiques selon un double gradient longitudinal et latéral. Les résultats permettent 1) de caractériser et d’analyser l’organisation spatiale et l’intensité de l’anthropisation sur le fleuve Charente ; 2) de proposer une hiérarchisation des interventions humaines les plus impactantes sur le système fluvial et 3) de cartographier les impacts cumulés des interventions humaines et leurs principaux effets sur le fonctionnement du cours d’eau. Au final, ce travail de recherche au caractère exploratoire contribue à mieux identifier l’organisation et l’intensité des impacts anthropiques sur le système fluvial, et à mieux prioriser les enjeux de restauration et le type d’action à mener sur la Charente. Posters Seulement 14% des masses d’eau superficielles du périmètre du Schéma d’Aménagement et de Gestion des Eaux (SAGE) de la Charente sont déclarées en bon état écologique contre 47% pour le bassin Adour-Garonne. A l’échelle même de la Charente, 18% des masses d’eau sont en bon état écologique dans la partie en amont et 6% dans la partie aval. Les principales causes de dégradation de l’état écologique du cours d’eau sont la pollution des eaux, la faible continuité écologique et sédimentaire en raison de la présence de multiples ouvrages en travers, l’altération du régime hydrologique et la prolifération des espèces invasives tant animales que végétales. Elles sont liées aux interventions humaines directes et indirectes sur le système fluvial. Il est par conséquent fondamental de proposer aux gestionnaires de cours d’eau une synthèse des connaissances sur l’organisation spatiale et l’intensité des usages anciens et 257 Posters 257 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Bibliographie Brown A. G., Lespez L., Sear D. A., Macaire J.-J., Houben P., Klimek K., Brazier R. E., Van Oost K., Pears B. (2018). Natural vs anthropogenic streams in Europe : History, ecology and implications for restoration, river-rewilding and riverine ecosystem services. Earth- Science Reviews, vol. 180, p. 185-205. Halpern B. S., Selkoe K. A., Micheli F., Kappel C. V. (2007). Evaluating and Ranking the Vulnerability of Global Marine Ecosystems to Anthropogenic Threats. Conservation Biology, vol. 21, n°5, p. 1301-1315. Pereira C. I., Madrid D. A., Correa I. D., Pranzini E., Botero C. M. (2019). An evaluation of human interventions in the anthropogenically disturbed Caribbean Coast of Colombia. Anthropocene, vol. 27, p. 1-11. Toro J., Requena I., Zamorano M. (2010). Environmental impact assessment in Colombia : Critical analysis and proposals for improvement. Environmental Impact Assessment Review, vol. 30, p. 247-261. 258 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Résolution de toponymes par apprentissage profond à partir de cooccurrences et de relations spatiales Jacques Fize 1, Ludovic Moncla 1, Bruno Martins 2 1. LIRIS UMR 5205, INSA Lyon, France jacques.fize@gmail.com,ludovic.moncla@insa-lyon.fr 1. LIRIS UMR 5205, INSA Lyon, France jacques.fize@gmail.com,ludovic.moncla@insa-lyon.fr 2. Instituto Superior Técnico - INESC-ID, University of Lisbon, Portugal bruno.g.martins@tecnico.ulisboa.pt 1. Une implémentation en Python de ce modèle est disponible en ligne : https://git.liris.cnrs .fr/HExtGEO/toponym-geocoding 2. https://jacobe2169.github.io/mapthetoponymsim/ 1. Introduction En traitement automatique du langage, le processus de geoparsingqui per- met d’identifier les noms de lieux (toponymes) dans un texte et de leur associer des coordonnées géographiques non-ambiguës (latitude-longitude) (Gritta et al., 2018) est une tâche complexe qui s’effectue en deux étapes : (1) l’iden- tification d es t oponymes ( ou g eotagging) e t ( 2) l a r ésolution d es toponymes (ou geocoding). La plupart des approches de résolution de toponymes utilisent des index géographiques (i.e. gazetteer) associant à un nom de lieu plusieurs données : alias, coordonnées géographiques, catégorie (e.g. ville, pays), etc. Plusieurs types d’ambiguïtés rendent complexe la tâche de résolution des to- ponymes (i) plusieurs noms peuvent exister pour un même lieux, (ii) plusieurs lieux peuvent avoir le même noms, (iii) un même nom peut faire référence à un lieu ou à un autre type d’entité. Par ailleurs, pour diverses raisons, certains lieux existants ne sont pas répertoriés et si les gazetteers peuvent être suffi- sants pour l’analyse de documents contemporains, ce n’est pas le cas pour des documents plus anciens ou historiques. Pour contourner ces problèmes, des tra- vaux récents (DeLozier et al., 2015 ; Cardoso et al., 2019) proposent d’entraîner des modèles de geocoding afin d e n e p as avoir à u tiliser d e g azetteers l ors de l’étape de résolution des toponymes. Ces approches sont basées sur des tech- niques d’apprentissage profond avec en entrée les toponymes et leur contexte (e.g. mots cooccurrents). 259 Posters 259 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 3. Expérimentations et résultats Les trois contextes décrits dans la section précédente constituent trois jeux de données pouvant être combinés (cooccurrences, proximité et inclusion) pour l’entraînement d’un modèle. Nous avons constitué des jeux de données sur dif- férentes zones géographiques. Nous utilisons la mesure d’exactitude (avec une précision de 100km) et obtenons les résultats suivants : 0.79 pour la France, 0.88 pour la Grande-Bretagne, 0.21 pour les Etats-Unis, et 0.90 pour Texas + Île de France. Par ailleurs, nous observons que la combinaison des diffé- rents contextes (spatiaux et textuels) permet d’améliorer les résultats et que les cooccurrences de toponymes jouent un rôle très important pour cette tâche de résolution des toponymes. Nous observons des scores très faibles pour les États-Unis contrairement aux autres zones géographiques qui peuvent s’expli- quer par une toponymie particulière sur ce territoire. Ces premiers résultats sont encourageants et plusieurs pistes, telle que l’utilisation de modèles de langue pré-entraînés (e.g. BERT), sont envisagées pour étendre la couverture et améliorer l’exactitude du modèle. 4. Remerciements Ce travail a été réalisé grâce au soutien financier du Projet IDEXLYON de l’Université de Lyon dans le cadre du programme Investissement d’Avenir (ANR-16-IDEX-0005). 2. Méthodologie Dans ce travail, nous proposons une architecture 1 basée sur des réseaux de neurones récurrents pour l’entraînement d’un modèle de résolution des topo- nymes (Fig. 1). Cette architecture prend en entrée des paires de toponymes et les coordonnées géographiques du premier. Le modèle entraîné a pour rôle de prédire la localisation d’un premier toponyme auquel nous associons un deuxième toponyme comme contexte. La phase d’entraînement du modèle se décompose en trois étapes. La première consiste à transformer les toponymes en séquences de n-grammes de caractères (e.g. Paris →par, ari, ris). Cela permet au modèle de s’adapter à différentes variations des noms de lieux ainsi qu’à la régionalité de certains affixes (préfixe, suffix) 2. L’ensemble des n-grammes est ensuite encodé selon un embedding pré-entraîné avec la méthode Word2vec. Les vecteurs obtenus pour chaque n-gramme des toponymes d’entrée sont donnés en entrée de deux réseaux Bi-LSTM (Bidirectional Long Short-Term Memory) (Graves, Schmidhuber, 2005) chargés de l’extraction des caractéristiques puis l’entraînement pour la prédiction des coordonnées est réalisé par des réseaux de neurones. Paris Cherbourg ... pa ar ri is ... LSTM LSTM ... LSTM LSTM ... Ngram Embedding + Longitude Multi-Layer Perceptron Multi-Layer Perceptron Texas WGS84 Coordinates Latitude OUTPUT INPUT LEARN/PREDICT Figure 1. Architecture pour l’entraînement d’un modèle de geocoding. Figure 1. Architecture pour l’entraînement d’un modèle de geocoding. Dans le cadre de nos expérimentations nous avons considéré trois manières différentes de constituer les paires de toponymes : le première basée sur les cooc- currences de toponymes dans des textes et les deux autres basés sur leurs rela- tions spatiales (relations de proximité et relations d’inclusion). Pour le contexte "cooccurrence", nous générons les paires de toponymes à partir des liens ex- primés entre les noms de lieux dans les articles Wikipedia. Pour le contexte ’spatial’ nous construisons les paires de toponymes à partir de relations d’in- clusion et de proximité calculées à partir des données de Geonames. L’inclusion Posters 260 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Résolution de toponymes par apprentissage indique que le premier toponyme est contenue dans le second et la proximité est calculée selon une distance maximale entre les deux lieux (100km dans le cadre de nos expérimentations). Pour éviter des temps de calcul trop élevés, les relations de proximité et de cooccurrences sont échantillonnées aléatoirement. Gritta M., Pilehvar M. T., Limsopatham N., Collier N. (2018). What’s missing in geographical parsing? Language Resources and Evaluation, vol. 52, no 2, p. 603– 623. Bibliographie Cardoso A. B., Martins B., Estima J. (2019). Using Recurrent Neural Networks for Toponym Resolution in Text. In Proceedings of the EPIA Conference on Artificial Intelligence, p. 769–780. Springer International Publishing. DeLozier G., Baldridge J., London L. (2015). Gazetteer-Independent Toponym Reso- lution Using Geographic Word Profiles. In Proceedings of the 29th AAAI Confe- rence on Artificial Intelligence. Graves A., Schmidhuber J. (2005). Framewise phoneme classification with bidirectio- nal LSTM networks. In Proceedings of the IEEE International Joint Conference on Neural Networks., vol. 4, p. 2047-2052 vol. 4. Gritta M., Pilehvar M. T., Limsopatham N., Collier N. (2018). What’s missing in geographical parsing? Language Resources and Evaluation, vol. 52, no 2, p. 603– 623. 261 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Follin Jean-Michel1, Voniarinjanahary Rovasoa1, Thommeret Nathalie1, Fournier Marie1 1. Laboratoire Géomatique et Foncier, Cnam-ESGT 1 boulevard Pythagore, 72000 Le Mans jean-michel.follin@lecnam.net 1. Laboratoire Géomatique et Foncier, Cnam-ESGT 1 boulevard Pythagore, 72000 Le Mans jean-michel.follin@lecnam.net RESUME. Le projet FARMaine, mené dans le cadre du programme national PSDR4 a pour objectif d’éclairer les modalités de mise en œuvre des Politiques Publiques d’Environnement (PPE) dans les territoires de fonds de vallée du bassin versant de la Maine, et leurs conséquences sur le fonctionnement des exploitations agricoles et la gestion du foncier agricole. Pour comprendre les caractéristiques foncières, spatiales et agricoles des fonds de vallée étudiées ainsi que leurs évolutions en lien avec les PPE, des analyses spatiales et statistiques ont été menées. Elles combinent des données abordant différentes thématiques sur la période 2009 à 2018 : le RPG niveau 2, les Mesures Agro-Environnementales et Climatiques (MAEC), et les Fichiers fonciers. Un web-SIG baptisé FarmSIG, encore en cours de développement, permet de partager et diffuser les données rassemblées et produites, en cohérence avec la démarche partenariale du projet impliquant des chercheurs et acteurs des territoires de fond de vallée. MOTS-CLES : WEB MAPPING, DATAVIZ, MESURES AGROENVIRONNEMENTALES, RPG, FICHIERS FONCIERS, ANONYMISATION PAR CARROYAGE, ANALYSE STATISTIQUE 262 Problématique et motivations La question du suivi des territoires agricoles où sont mises en œuvre des politiques publiques d’environnement (PPE), en particulier les conséquences de celles-ci sur le fonctionnement des exploitations agricoles (EA) et la gestion du foncier agricole, est primordiale pour les acteurs publics qui en ont la charge. Le projet FARMaine (Foncier et Régulations dans le bassin de la Maine) mené dans le cadre du programme national PSDR4 (Pour et Sur le Développement Régional) décliné à l’échelle du Grand Ouest, s’est fixé pour objectif d’éclairer les modalités de mise en œuvre des PPE dans les territoires de fonds de vallée du bassin versant de la Maine. Quatre territoires d’étude ont été sélectionnés pour mener nos études en fonction de leurs caractéristiques (enjeux en matière d'environnement, existence de PPE depuis plusieurs années, variété dans les systèmes agricoles observés) : les Basses Vallées Angevines, les Alpes Mancelles, la Vallée du Loir et le bassin versant de l'Oudon. Dans ce cadre, différents traitements (combinaisons, analyses statistiques et anonymisations) ont été menés sur des données spatiales portant sur différentes dates (entre 2009 et 2018), thématiques (foncier, agricole, réglementaire) et échelles (bassin versant et territoires d’études). Ils poursuivent un double objectif qui est d’une part, d’assurer la confidentialité des données produites avant diffusion et d’autre part d’apporter des éléments de réponse aux questions suivantes : observe-t-on des évolutions du profil agricole des zones étudiées ? peut-on établir un lien entre structures spatiales des EA et présence de PPE ? Un web-SIG, baptisé FarmSIG, a été développé pour rassembler et diffuser les données et analyses produites dans le cadre du projet FARMaine. Il vise à proposer une vision cartographique synthétique des zones de fonds de vallée étudiées et des clés pour comprendre leurs caractéristiques et leurs évolutions en lien avec les PPE. FarmSIG s’adresse à des profils variés (exploitants agricoles, agents de Chambres d’Agriculture, chercheurs en agronomie, aménagement et géographie). Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Conclusion et perspectives FarmSIG est en cours de développement mais permet déjà une exploration temporelle et thématique des territoires étudiés, proposant des éléments de réponse aux questions posées sur les liens entre PPE et caractéristiques agricoles et foncières de ces secteurs. Une restitution 3D est explorée afin de mieux appréhender certaines informations comme les liens entre morcellement ou dispersion des exploitations (perçu via la couleur) et intensité d’adoption des MAEC (perçue via la hauteur). Par ailleurs FarmSIG a vocation à être alimenté avec d’autres résultats d’analyses qui restent à mener comme celles portant sur les dynamiques d’évolutions des EA. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] sans engagement MAEC qui ont conservé une exonération des terres agricoles, et 3) les parcelles forestières qui ont connu un changement lié à leurs propriétaires. Puis une analyse synchronique a été menée en confrontant, pour 2015 et 2018, – l’intensité de l’adoption des MAEC par EA, calculée en cumulant, après les avoir normalisés, les nombres, longueurs et surfaces totales de MAEC ponctuelles, linéaires et surfaciques par EA, – et l’organisation spatiale des parcellaires agricoles des EA, appréhendée par les indicateurs de dispersion et de morcellement (Latruffe et Piet, 2014 ; Renard, 1972). – et l’organisation spatiale des parcellaires agricoles des EA, appréhendée par les indicateurs de dispersion et de morcellement (Latruffe et Piet, 2014 ; Renard, 1972). Ces informations ont été rapportées aux centroïdes des exploitations (obtenus par le barycentre géométrique des îlots de même exploitation). Elles ont ensuite été anonymisées par la méthode de carroyage (agrégation dans des grilles de carreaux) pour rendre impossible l’identification de données relatives à une EA (Loonis, 2018). L’application FarmSIG (http://farmaine.fr/index.php/farmsig/), bâtie sur les bibliothèques javascript Leaflet et Chart.JS, propose une restitution des analyses réalisées avec des fonctionnalités i) de navigation thématique et temporelle dans les données cartographiques aux deux échelles du bassin versant et des territoires d’étude ; et ii) d’exploration des données via un panneau latéral proposant des éléments synthétiques caractérisant les secteurs sélectionnés. Loonis V. (dir), Bellefon M.P. (coord) (2018), Manuel d’analyse spatiale. Théorie et mise en oeuvre pratique avec R, Insee Méthodes n° 131, 406 p. Données, méthodes et résultats Les principales données mobilisées sont le Registre Parcellaire Graphique (RPG) niveau 2 qui comporte, pour chaque îlot de parcelles agricoles, l’identifiant de son EA (numéro PACAGE), les Mesures Agro-Environnementales et Climatiques (MAEC) qui associent à chaque parcelle les dispositifs MAEC adoptés par son EA, et les Fichiers fonciers contenant des informations sur les propriétaires des parcelles. Dans une première analyse diachronique, les changements ont été abordés en croisant les Fichiers fonciers avec le RPG. Cela a abouti à la construction d’une typologie des évolutions foncières des parcelles agricoles dans les Basses Vallées Angevines entre 2009 et 2016. Celle-ci a été réalisée par ACM et classification ascendante hiérarchique. Trois catégories ont été mises en évidence : 1) les parcelles agricoles qui ont fait l’objet d’un engagement en MAEC sans changement lié au propriétaire et ont évolué vers une exonération Natura 2000, 2) les parcelles agricoles 263 Posters 263 Bibliographie Renard M. (1972). Recherches méthodologiques sur le degré d'émiettement parcellaire des exploitations agricoles des bocages de l'ouest. Bulletin de l'Association de géographes français, n° 397, p. 83-94. Puech T., Durpoix A., Barataud F., & Mignolet C. (2020). Une méthode pour caractériser l’organisation spatiale des parcellaires d’exploitations agricoles. Cybergeo: Revue européenne de géographie/European journal of geography, Document 938. Latruffe L., Piet L. (2014), "Does land fragmentation affect farm performance? A case study from Brittany, France", Agricultural Systems, vol. 129, p. 68-80. Loonis V. (dir), Bellefon M.P. (coord) (2018), Manuel d’analyse spatiale. Théorie et mise en oeuvre pratique avec R, Insee Méthodes n° 131, 406 p. 264 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Dorothée JAMES1, Antoine COLLIN1, 2, Mathilde LETARD1 Dorothée JAMES1, Antoine COLLIN1, 2, Mathilde LETARD1 1. EPHE, PSL Université Paris, CNRS UMR 6554 LETG 15 Boulevard de la mer, 35800 Dinard, France dorothee.james@ephe.psl.eu 1. EPHE, PSL Université Paris, CNRS UMR 6554 LETG 15 Boulevard de la mer, 35800 Dinard, France dorothee.james@ephe.psl.eu 2. LabEx CORAIL, Moorea, French Polynesia 2. LabEx CORAIL, Moorea, French Polynesia 2. LabEx CORAIL, Moorea, French Polynesia MOTS-CLES : MOSAIQUE, METASHAPE, EVOLUTION, PRE-SALE, CLASSIFICATION, SVM 265 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Face aux changements globaux, l’étude des milieux et des paysages est essentielle pour comprendre leur fonctionnement. L’environnement littoral est sujet à une modification intense aussi bien par l’activité anthropique que par les processus météo-marins. Afin de comprendre les trajectoires socio-écologiques, l’étude rétrospective du milieu littoral permet d’appréhender les mécanismes pour mener de manière éclairée une gestion prospective. Les données historiques, qui nous renseignent sur l’état du littoral à un moment précis, sont issues de diverses sources telles que les photographies aériennes ou les cartes anciennes. Ces données sont donc définies par des caractéristiques spatiales, spectrales, radiométriques et temporelles spécifiques à chaque campagne aérienne, ce qui pose problème pour établir une série temporelle cohérente. Cette étude vise, dans un premier temps, à mettre en place une méthodologie innovante pour l’utilisation et la standardisation de données anciennes massives (à Très Haute Résolution, THR, et à une échelle régionale) issues de sources diverses (Figure 1). Dans un second temps, grâce la classification supervisée, les cartes créées permettent d’obtenir une comparaison de l’évolution de l’occupation sol à partir d’une classe d’intérêt (pré-salé) ayant une large emprise et à THR. Notre site d’étude, composé d’un vaste pré-salé évoluant le long de la digue Saint-Anne, se localise dans le golfe normano-breton, entre la chapelle Sainte-Anne (48.60°N ; 1.67°W) et le Mont-Saint-Michel (48.63°N ; 1.51°W). Les spartines, l’obione, la salicorne sont les principales communautés végétales qui composent ce pré-salé. Au-devant de celui-ci, s’étend une vaste vasière composée de sédiment vaso-sableux. Les données historiques sont au cœur de cette étude. Fastidieux, le téléchargement des images aériennes est la première étape de ce processus. Ainsi, 9 images ont été retenues pour 1966, 19 pour 1979, 6 pour 1995 et 4 pour 2014. Les images ont ensuite été découpées automatiquement via XnConvert, puis mosaïquées, pour chaque millésime, par l’intermédiaire du logiciel de photogrammétrie Metashape (Rault et al., 2020). Dorothée JAMES1, Antoine COLLIN1, 2, Mathilde LETARD1 À partir de cette compilation d’images, le géoréférencement en RGF93/Lambert93 a été réalisé à l’aide du logiciel SIG ArcGIS. L’Ortholittorale V2 (2014) a servi de référence. La création de filtres texturaux de co-occurrence (moyenne, contraste, homogénéité, variance, dissimilarité, second moment, corrélation et entropie) a été implémentée sur chacune des reconstructions photogrammétriques panchromatiques (logiciel ENVI). In fine, un algorithme de classification supervisée, Support Vector Machine (SVM) a été appliqué sur les 8 bandes de textures. Deux classes ont été identifiées : le pré-salé et la vasière, avec 15000 pixels de validation et 15000 pixels de calibration, extraits aléatoirement. 266 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Étude spatio-temporelle d’un écosystème littoral FIGURE 1. Processus méthodologique pour l’étude rétrospective du littoral. FIGURE 1. Processus méthodologique pour l’étude rétrospective du littoral. Les résultats de classification permettent d’analyser la superficie de l’occupation du sol pour les années 1966, 1979, 1995 et 2014. Ainsi, respectivement, l’enveloppe du pré-salé était de 9,92 km², de 12,69 km², de 9,69 km² et 15,19 km². La saisonnalité, la composition floristique du pré-salé (Valéry et Radureau, 2014) et le développement des activités anthropiques ont contribué à la vitesse de progradation du pré-salé d’environ 1,1% par an en 48 ans. L’utilisation d’algorithme de deep learning (Letard et al., 2020) permettrait d’affiner la détection de la composition du pré-salé. Valéry, A., Radureau, A., (2014). Evolution de la végétation des marais salés de la baie du Mont-Saint-Michel-Analyse cartographique 1984-2013. Rapport Convention CREBS/Université de Rennes1/Conservatoire du Littoral/Inter-SAGE Baie du Mont- Saint-Michel/Agence de l’Eau Seine-Normandie/Agence de l’eau Loire-Bretagne ; Rennes, France, 15p. Bibliographie Letard, M., Collin, A., et Pastol, Y., (2020). Improving mangrove and salt marsh mapping using superpixel-based deep classification of VHR superspectral satellite imagery and topobathymetric LiDAR data, Sea Tech Week 2020, 12-15 Oct. 2020, Brest. Rault, C., Dewez, T., et Aunay, B., (2020). Structure-from-motion processing of aerial archive photographs: sensitivity analyses pave the way for quantifying geomorphological changes since 1978 in la réunion island. ISPRS Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2, 773-780. Valéry, A., Radureau, A., (2014). Evolution de la végétation des marais salés de la baie du Mont-Saint-Michel-Analyse cartographique 1984-2013. Rapport Convention CREBS/Université de Rennes1/Conservatoire du Littoral/Inter-SAGE Baie du Mont- Saint-Michel/Agence de l’Eau Seine-Normandie/Agence de l’eau Loire-Bretagne ; Rennes, France, 15p. 267 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Adaptation de domaine par transport optimal pour la cartographie à large échelle des peupleraies Yousra Hamrouni1, Olivia Bernardoff1, Nicolas Karasiak1, David Sheeren1 1. Université de Toulouse, INRAE, UMR DYNAFOR, Castanet-Tolosan, France yousra.hamrouni@inrae.fr david.sheeren@ensat.fr MOTS-CLES: TELEDETECTION, TRANSPORT OPTIMAL, ADAPTATION DE DOMAINE, TRANSFERT SPATIAL, PEUPLERAIES, CLASSIFICATION, LARGE ÉCHELLE, SENTINEL-2 268 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 La cartographie des peuplements forestiers et leur composition en essences est une donnée incontournable pour les gestionnaires réalisant des opérations sylvicoles. Bien qu’il existe des bases de données forestières à l’échelle nationale, les connaissances sur ces essences sont encore lacunaires en raison, soit d’une caractérisation peu exhaustive des peuplements, soit d’une marge d’erreur importante ou d’une mise à jour trop longue. En particulier, les estimations de surfaces des peupleraies françaises s'avèrent peu précises en raison de leur court cycle de rotation (15 ans en moyenne) . La disponibilité d'images à haute résolution spatiale, spectrale et temporelle provenant des nouveaux satellites Sentinel-2 a offert de nouvelles opportunités pour le suivi des plantations de peupliers sur de grandes étendues. Plusieurs travaux ont démontré le potentiel de la télédétection pour la cartographie des plantations telles que le caoutchouc et le palmier à huile mais seulement quelques-uns se sont intéressés au peuplier. Malgré les résultats prometteurs rapportés, la plupart des études se limitent à des zones d’études relativement restreintes (petites forêts connues) et les performances des modèles de classification dépendent largement des données utilisées pour l'entraînement et la validation des algorithmes. En effet, lorsque ces modèles, calibrés sur une zone, sont appliqués sur de vastes étendues, possiblement très éloignées de la zone initiale, les performances prédictives s’effondrent. Les modèles locaux sont souvent inadaptés au transfert d’une région à l’autre (transfert spatial) car ils ne tiennent pas compte de toute la diversité des spectres de réflectance des classes et de leur caractère non stationnaire dans l’espace (problème du “data shift” en apprentissage automatique). La variabilité spatiale est souvent due à des variations dans les conditions d'acquisition des images, dans la phénologie de la végétation ou dans les conditions stationnelles. Les techniques d’adaptation de domaine et en particulier le transport optimal (TO) sont proposées dans cette étude pour répondre au problème de transfert évoqué. Elles permettent d’adapter les distributions des données de départ sur lesquelles des connaissances sont disponibles (domaine source) à des données d’arrivée pour lesquelles il y a peu ou pas d’informations (domaine cible). [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Transport optimal sur les peupleraies Transport optimal sur les peupleraies classification avec TO ont été comparés à un modèle entraîné sur la source non adaptée. Random Forest a été utilisé comme algorithme de classification et trois classes ont été considérées : peupleraies, robiniers et chênes avec, 3850 échantillons étiquetés par classe pour la source et 1250 pour la cible. Les paramètres optimaux du TO (coefficients de régularisation par l’entropie et par classe) et du classifieur (nombre d’arbre, profondeur maximale) ont été fixés à partir d’une grille de recherche. Les résultats obtenus montrent que l’adaptation de la distribution du domaine source par TO permet d’améliorer la précision globale de la classification de 3 à 35% et la précision producteur (mesure de rappel) des peupleraies de 8 à 27% (TABLE 1). L'amélioration est particulièrement marquée lorsque des étiquettes de la source et de la cible ont été incorporées. Il est important de noter que dans la dernière stratégie nous avons utilisé toutes les étiquettes disponibles du domaine cible et qu’il serait intéressant d’étudier à termes l’impact du nombre d’étiquettes rajoutées sur les performances de la classification. Une des contraintes de l’approche envisagée dans ce travail est la nécessité d’avoir les mêmes classes entre les domaines source et cible. Ainsi, seules les classes communes ont pu être conservées. Par ailleurs, afin de garantir de bonnes performances, les proportions des classes entre les deux domaines doivent être équivalentes (±15%). D’autres approches d’adaptation de domaine qui se focalisent non pas sur l’adaptation des données mais plutôt du classifieur peuvent être envisagées pour contourner ces limites. C’est notamment le cas de l’apprentissage actif qui permet de garder les deux distributions inchangées et de modifier le classifieur en lui incorporant un nombre limité d’échantillons du domaine cible (Hamrouni et al., 2021). TABLE 1: Résultats des différentes stratégies de transport optimal Avant transport optimal TO non supervisé TO non supervisé régularisé TO semi- supervisé + étiquettes de la source TO semi- supervisé + étiquettes (source + cible) Précision globale 61% 64% 66% 69% 96% Précision producteur (peupleraies) 70% 78% 78% 88% 97% Courty N., Flamary R. , Tuia D. and Rakotomamonjy A. (2017). Optimal Transport for Domain Adaptation, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, n° 9, p. 1853-1865. Hamrouni Y., Paillassa, E., Chéret V., Monteil C., Sheeren D. (2021). Posters L’hypothèse qui sous-tend le transport optimal est que si les données sources subissent une phase d’adaptation pour ressembler davantage aux données cibles, le data shift est réduit, et un classifieur entraîné sur les données sources adaptées, donne de meilleures performances qu’un classifieur s’appuyant uniquement sur les données du domaine source ou encore, sur un nombre restreint d’échantillons disponibles dans le domaine cible (Courty et al., 2017). Le transport optimal a été appliqué entre deux tuiles Sentinel-2 de 100 km² chacune. Elles sont situées dans deux bassins populicoles contrastés dans le Nord-Est (T31UEQ) et le Sud-Ouest (T31TCJ) de la France. Les séries temporelles Sentinel-2 datent de 2017 et comportent 34 dates ré-échantillonnées au même pas temporel et dans lesquelles les pixels nuageux ont été corrigés. Le transport optimal a été appliqué depuis la tuile Sud-Ouest (source) vers la tuile Nord-Est (cible) selon quatre stratégies : TO non supervisé (utilise les distributions de la source et de la cible), TO non supervisé régularisé (avec la contrainte de l’entropie), TO non supervisé avec les étiquettes de la source et TO semi-supervisé (utilise les distributions et les étiquettes disponibles de la source et de la cible). Pour chacune des stratégies, les résultats de Posters 269 SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] From local to global: A transfer learning-based approach for mapping poplar plantations at national scale using Sentinel-2. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 171, p. 76-100. TABLE 1: Résultats des différentes stratégies de transport optimal Courty N., Flamary R. , Tuia D. and Rakotomamonjy A. (2017). Optimal Transport for Domain Adaptation, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, n° 9, p. 1853-1865. Courty N., Flamary R. , Tuia D. and Rakotomamonjy A. (2017). Optimal Transport for Domain Adaptation, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, n° 9, p. 1853-1865. Hamrouni Y., Paillassa, E., Chéret V., Monteil C., Sheeren D. (2021). From local to global: A transfer learning-based approach for mapping poplar plantations at national scale using Sentinel-2. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 171, p. 76-100. Hamrouni Y., Paillassa, E., Chéret V., Monteil C., Sheeren D. (2021). From local to global: A transfer learning-based approach for mapping poplar plantations at national scale using Sentinel-2. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 171, p. 76-100. 270 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Jérémy Kalsron1, Jean-Marie Favreau1, Guillaume Touya2 1. LIMOS, UMR CNRS 6158, Université Clermont Auvergne 1 rue de la Chebarde, 63178 Aubière, France jeremy.kalsron@uca.fr, j-marie.favreau@uca.fr 1. LIMOS, UMR CNRS 6158, Université Clermont Auvergne 1 rue de la Chebarde, 63178 Aubière, France jeremy.kalsron@uca.fr, j-marie.favreau@uca.fr 2. UMR LASTIG, IGN 73 avenue de Paris, 94160 Saint-Mandé, France g.touya@ign.fr MOTS-CLES : déficience visuelle, carrefour, modèle de données, génération automatique de texte MOTS-CLES : déficience visuelle, carrefour, modèle de données, génération automatique de texte 271 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Pour les personnes en situation de déficience visuelle (PSDV), le déplacement en milieu urbain est un défi. Celles-ci doivent en effet comprendre l'agencement des infrastructures nécessaires à une traversée en sécurité, une information absente des outils de guidage mobiles actuels. Les cartes tactiles, notamment mobilisées par les instructeurs de locomotion pour enseigner les techniques de déplacement en autonomie, ne permettent pas de représenter une haute densité d’informations ou des détails fins à petite échelle (Touya et al., 2019). Elles nécessitent pour cela d’être complétées d’une information supplémentaire, textuelle ou sonore. , outils de guidage mobiles actuels. Les cartes tactiles, notamment mobilisées par les instructeurs de locomotion pour enseigner les techniques de déplacement en autonomie, ne permettent pas de représenter une haute densité d’informations ou des détails fins à petite échelle (Touya et al., 2019). Elles nécessitent pour cela d’être complétées d’une information supplémentaire, textuelle ou sonore. Parmi les travaux ayant étudié la représentation sonore de l’information géographiques à destination des PSDV, (Boularouk et al., 2017) proposent d’audio- décrire des lieux spécifiques en exploitant la sémantique d’une base de données comme OpenStreetMap. Plus proches de notre problématique, (Guth et al., 2019) réalisent un tableau de données permettant de décrire l’enchaînement des entités qui composent une traversée. Ces travaux présentent l’information sous la forme d’un enchaînement de données brut, et ne proposent pas de modèle de données structuré permettant de formaliser un carrefour et les éléments qui le composent. Nous développons ces aspects dans le travail que nous présentons ici, en proposant un processus de génération automatique de description de carrefour. Ce processus s’appuie sur un modèle de donnée original dédié (Figure 1). [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Le carrefour dont vous êtes le héros à trous complétés procéduralement, faciles à mettre en œuvre, et permettant d’implémenter les phrases telles que nous les avons conçues. Nous proposons une implémentation sous licence libre GPLv31, accessible en démonstration en une preuve de concept interactive par hyperliens depuis un navigateur web2 (Figure 2). FIGURE 2. Les carrefours décrits au sein de notre preuve de concept. Il nécessitent respectivement 72, 64, 31, et 82 pages de description. FIGURE 2. Les carrefours décrits au sein de notre preuve de concept. Il nécessitent respectivement 72, 64, 31, et 82 pages de description. Nous avons conçu ces description en conduisant une série d’entretiens avec une dizaine de PSDV et d’instructeurs de locomotion, et les formulations employées sont issues de leurs propositions. Ceux-ci, familiers avec le territoire que nous avons exploré, ont exprimé l’intérêt de pouvoir visiter un carrefour à distance par le biais d’une description textuelle. Le modèle étant aujourd’hui complété manuellement, nous envisageons pour la prochaine étape de le générer depuis des données ouvertes telles qu’OpenStreetMap. Ce travail est financé dans le cadre du projet ACTIVmap (ANR-19-CE19-0005). 1. https://gitlab.limos.fr/jeremyk6/carrefourV2/-/tree/sageo2021 2. https://carrefours.activmap.limos.fr/publications/sageo2021 Jérémy Kalsron1, Jean-Marie Favreau1, Guillaume Touya2 Le modèle est pensé pour permettre la génération d'un ensemble de textes que l'utilisateur pourra consulter suivant deux modes : une description générale du carrefour, autosuffisante et statique, et une série de descriptions cohérentes permettant une visite interactive du carrefour. FIGURE 1. Le modèle de données proposé représente un graphe piéton du carrefour. FIGURE 1. Le modèle de données proposé représente un graphe piéton du carrefour. La description générale repose sur une description par branche, mobilisée par la nouvelle génération d’instructeurs de locomotion (Ratelle et Couturier, 2018). Une branche correspond à un segment du carrefour, et comprend un nombre de voies et un mode de régulation du trafic. La description interactive place l’utilisateur au sein du carrefour et permet de détailler chaque traversée en proposant de choisir une direction lorsqu’une bifurcation est possible. Ces directions sont matérialisées par un point d’intérêt distant dont le choix, aujourd’hui manuel, pourrait être automatisé par des techniques de fouille de données, selon sa prégnance au sein de la ville. Le texte de la description est généré depuis le modèle présenté à l’aide de patrons, des textes Posters 272 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Bibliographie Boularouk et al., (2017). Ontology for a voice transcription of OpenStreetMap data: the case of space apprehension by visually impaired persons, World Academy of Science, Engineering and Technology 2017, London, United Kingdom. Guth et al., (2019). An Intersection Database Facilitates Access to Complex Signalized Intersections for Pedestrians with Vision Disabilities, Transportation Research Record, vol. 2673, p. 698-709. Ratelle A., Couturier J.-A., (2018). Manuel d’intervention en orientation et mobilité, Les Presses de l’Université de Montréal. Touya et al., (2019). Automatic derivation of on-demand tactile maps for visually impaired people: first experiments and research agenda, International Journal of Cartography , vol. 5, p. 67‑91. 1. https://gitlab.limos.fr/jeremyk6/carrefourV2/-/tree/sageo2021 2. https://carrefours.activmap.limos.fr/publications/sageo2021 273 273 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Intégrer des données géospatiales dans un modèle multi-agents : cas de LittoSIM-GEN Cécilia PIGNON-MUSSAUD1, Ahmed LAATABI1, Marion AMALRIC2, Brice ANSELME3, Élise BECK4, Nicolas BECU1, Nicolas MARILLEAU5 1. 5. UMR 7266 LIENSs, La Rochelle Université 2 rue Olympe de Gouges, 17000 La Rochelle, France cecilia.pignon-mussaud@univ-lr.fr 2. UMR 7324 CITERES, Université de Tours 33, Allée Ferdinand de Lesseps, 37204 TOURS CEDEX 03, France 3. UMR PRODIG, Université Paris 1 2, rue Valette, 75005 Paris, France 4. UMR Pacte, CNRS, Université Grenoble Alpes 1030 avenue Centrale 38100 GRENOBLE, France 5. Institut de Recherche pour le Développement, 32 avenue Henri Varagnat, 93143 Bondy, France MOTS-CLES : SYSTEME MULTI-AGENTS, DONNEES GEOSPATIALES, LITTOSIM-GEN, GENERICITE. 274 1 https://littosim.hypotheses.org/ ; https://github.com/LittoSim/LittoSim_model LittoSIM has been financed by Foundation of France. Intégrer des données géospatiales dans un modèle multi-agents Intégrer des données géospatiales dans un modèle multi-agents En termes de données géospatiales, les liens de mapping représentent l’ensemble des opérations sur les composantes spatiales (création, intersection, fusion, reprojection, contrôle topologique...) et textuelles/attributaires (création, conversion, agrégation, remplacement…). Ce mapping des données permet alors d’extraire, de transformer et d’homogénéiser les données d’entrée (Laatabi et al., 2020). Le protocole ODD+2D permet ainsi de faciliter la réutilisation du modèle, tout en contribuant fortement à améliorer la généricité du dispositif LittoSIM-Gen. Bibliographie Becu N., Amalric M., Anselme B., Beck E., Bertin X., Delaye E., Long N., Marilleau N., Pignon-Mussaud C., Rousseaux F. (2017). Participatory simulation to foster social learning on coastal flooding prevention. Environmental Modelling & Software. vol 98. p.1-11. Laatabi A., Marilleau N., Nguyen-Huu T., Hbid H., Ait Babram M. (2018). ODD+ 2D: an ODD based protocol for mapping data to empirical ABMs. Journal of Artificial Societies and Social Simulation, 21 (2) 9. p.1-24. Laatabi A., Becu N., Marilleau N., Pignon-Mussaud C., Amalric M. (2020). Mapping and Describing Geospatial Data to Generalize Complex Models: The Case of LittoSIM-GEN. International Journal of Geospatial and Environmental Research. Vol 7. p.1-21. 276 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Posters Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Aujourd’hui, un des défis majeurs de la modélisation multi-agents spatialisée porte sur l’exploitation et l’assimilation grandissante de données hétérogènes géolocalisées. A travers le dispositif LittoSIM1 (Becu et al., 2017), nous présenterons, comment l’automatisation de l’intégration des données dans le système multi-agent (SMA) a été géré. Développé suite à la tempête Xynthia de 2010, le dispositif LittoSIM a pour objectif de sensibiliser les élus et les agents communaux, aux risques de submersion marine et accompagner la concertation sur les stratégies de prévention du risque et d’aménagement du territoire. LittoSIM est une plateforme de simulation participative qui combine, à la fois, un jeu multi-acteurs (SMA distribué sous la plateforme GAMA), un modèle hydrodynamique de submersion (LISFLOOD-FP) et une mise en situation pour les acteurs locaux. Initialement développé pour une application sur l’île d’Oléron, le dispositif est désormais déployé sur d’autres territoires littoraux pour évaluer sa généricité et son efficacité en matière d’apprentissage (projet LittoSIM-GEN). Cette ouverture vers la généricité du dispositif s’accompagne d’une re-formalisation et d’une automatisation des traitements des données d’entrée du modèle. LittoSIM nécessite en entrée une grande diversité de données géospatiales issues de sources nombreuses et variées. Par exemple, les données relatives aux ouvrages de défense des côtes proviennent des services de l’État, de collectivités territoriales, de bureaux d’études ou encore de laboratoires de recherche. Elles sont souvent livrées dans des formats et des structures différentes. Ainsi, la transformation, l’agrégation, l’homogénéisation de ces données sont des opérations incontournables et chronophages. Nous commencerons par présenter rapidement l’utilisation du protocole ODD (Overview, Design concepts, and Details) (Laatabi et al., 2018), qui décrit la structure du modèle LittoSIM. Ce protocole ODD constitue un fil d’Ariane pour la compréhension et la réutilisation des modèles multi-agents à travers la description détaillée de l’ensemble de leurs composantes (variables d’état, comportements, ordonnancement, ...). Puis, une large part de la présentation portera sur les données et les traitements nécessaires pour le modèle. Nous présenterons alors l’extension ODD+2D qui permet d’enrichir la description du modèle, notamment pour les données géospatiales, à travers la présentation des sources de données et l’ajout de liens de correspondance/mapping. 275 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Louisa LE ROUZIC1, Bruno GERVAIS2, Jean-François BREILH3 et Axel CREACH4 L’établissement de ces scénarios sont facilités par une connaissance des événements historiques ayant impactés le département (Breilh et al., 2014). Dans un second temps, le projet cherche à modéliser directement la submersion potentielle associée à des conditions météo-marines prévues afin d’anticiper les zones pouvant être impactées. Ces deux objectifs permettront de prédire avec un temps d’avance les zones pouvant être inondées lorsqu’une tempête est annoncée. Pour les services de secours, l’objectif est de traduire une alerte météo « risque submersion » de niveau rouge ou orange, en zone d’impact pour prépositionner des moyens d’intervention ou anticiper une évacuation. Une réflexion a été engagée pour inclure les données de l’indice VIE dans le projet SURVEY 17 afin de pouvoir indiquer les constructions présentant le niveau de danger le plus élevé pour chaque scénario de tempête. Cela permettrait d’ajuster finement et localement les dispositifs d’intervention et d’évacuation et ainsi optimiser le déploiement des services de secours. Cette réflexion s’est néanmoins heurtée à des questions opérationnelles auxquelles ce poster apporte de premières réponses : questions opérationnelles auxquelles ce poster apporte de premières réponses : - Alors que l’indice VIE n’a été appliqué qu’à 7 communes tout en nécessitant un temps de collecte des données important, comment l’appliquer aux 80 000 constructions potentiellement concernées en Charente-Maritime ? - Par ailleurs, pour conserver la valeur de l’outil, celui-ci doit être mis à jour régulièrement pour être fiable et utile pour les services de secours. Comment faciliter cette mise à jour en proposant une méthode facilement reproductible et peu chronophage ? Ces réflexions aboutissent à la proposition d’un indice VIE alternatif permettant une collecte plus rapide des données (mise à jour plus aisée) et la possibilité de couvrir un territoire plus important. Cet indice VIE alternatif est tout d’abord simplifié dans sa structure. Ne sont conservés que deux critères sur les quatre : la hauteur d’eau potentielle et le type architectural. Le premier critère est aisé à extraire et à mettre à jour via un processus de jointure spatiale. Le second nécessite une phase de collecte de terrain qui peut être lourde. Louisa LE ROUZIC1, Bruno GERVAIS2, Jean-François BREILH3 et Axel CREACH4 Louisa LE ROUZIC1, Bruno GERVAIS2, Jean-François BREILH3 et Axel CREACH4 1. Université de Bretagne Occidentale, stagiaire au SDIS 17 de mai à août 2020, louisa.lerouzic@gmail.com 2. Service Départemental d’Incendie et de Secours (SDIS 17), Pôle moyens généraux, Bruno.Gervais@sdis17.fr 3. Union des Marais de la Charente-Maritime (UNIMA), Cellule prévention des Inondation, jf.breilh@unima.fr 4. Sorbonne Université, Laboratoire de Géographie Physique (UMR 8591-CNRS), axel creach@sorbonne universite fr 4. Sorbonne Université, Laboratoire de Géographie Physique (UMR 8591-CNRS), axel.creach@sorbonne-universite.fr Ce poster présente un outil de mesure de la vulnérabilité des maisons pour leurs occupants face au risque de submersion marine en Charente-Maritime. Ce projet a comme point de départ la tempête Xynthia, à l’origine de 41 décès par noyade en 2010. En plus du niveau marin exceptionnel engendré par la conjonction d’une marée haute de vives-eaux et la surcote associée à la tempête, les retours d’expérience ont pointé l’urbanisation des zones basses comme facteurs explicatifs : tous les décès sont survenus à l’intérieur de constructions résidentielles où les occupants ont été piégés, sans possibilité de s’échapper (Vinet et al., 2010). Pl i i iti ti t é é ti i i l é é t ll Plusieurs initiatives ont émergé pour anticiper au mieux les conséquences éventuelles de tels phénomènes dans le futur. A l’échelle des constructions résidentielles, un travail a été réalisé pour pouvoir mesurer la vulnérabilité que représente chacune des constructions pour ses occupants : l’indice VIE (Creach et al., 2015). Cet indice se base sur les caractéristiques de localisation (hauteur d’eau potentielle, proximité aux digues) et la configuration des constructions (type architectural, possibilité de se réfugier). Il permet de définir le niveau de danger qu’elles peuvent représenter pour les occupants. Cet indice permet de dresser des cartes de la vulnérabilité des constructions et ainsi savoir où il serait nécessaire d’évacuer de manière préventive en cas de submersion marine. A l’échelle du département de la Charente-Maritime, le projet SURVEY 17 a comme objectif de mieux anticiper les conséquences de submersions. En premier lieu, le projet s’attache à modéliser les submersions potentielles associées à 96 scénarios de tempêtes aux caractéristiques différentes (vent, houle, marée, etc.). Pour chacun de ces scénarios, une carte des zones inondables et des hauteurs d’eau associées est 277 Posters 277 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] proposée. Louisa LE ROUZIC1, Bruno GERVAIS2, Jean-François BREILH3 et Axel CREACH4 Pour faciliter la mise à jour de ce second critère, une méthodologie alternative à l’indice VIE est proposée : plutôt de que de collecter systématiquement l’information sur le type architectural pour chaque construction, il est proposé de généraliser la donnée selon une maille de 250 m de côté, telle que déjà utilisée par le SDIS. A l’intérieur de ces mailles, est réalisée une estimation de la proportion de chaque type architectural. Cette estimation peut être réalisée via Google Street View et complétée par une vérification de terrain. En croisant ce coefficient avec le niveau de vulnérabilité associé à chaque catégorie architecturale, cela permet d’obtenir une note globale de la vulnérabilité architecturale des bâtiments situés au sein de la maille. Cette note est ensuite croisée avec le niveau d’eau moyen au sein de la maille pour obtenir un nouveau résultat de l’indice VIE présentant le niveau de vulnérabilité à l’échelle de la maille. La figure 1 présente un exemple de résultat de l’indice à cette échelle. Cette méthode peut ainsi être transposée aisément aux 96 scénarios modélisés dans le projet SURVEY 17 et être mise à jour régulièrement. Elle permet ainsi de connaître, à l’échelle d’une maille de 250 m, les secteurs où il faudrait intervenir en priorité en cas de submersion marine. 278 278 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] FIGURE 1. Exemple de résultat de l’indice VIE alternatif pour la commune de Port- des-Barques (17). FIGURE 1. Exemple de résultat de l’indice VIE alternatif pour la commune de Port- des-Barques (17). Finalement, ce travail permet, en associant sapeurs-pompiers et géomaticiens/chercheurs de proposer un outil cartographique opérationnel pour une meilleure appréhension des littoraux urbanisés dans un contexte où le risque de submersion marine devrait se renforcer. Vinet F., Boissier L., Defossez S. (2011). La mortalité comme expression de la vulnérabilité humaine face aux catastrophes naturelles: deux inondations récentes en France (Xynthia, Var, 2010). VertigO- la revue électronique en sciences de l’environnement. Vol. 11, n°2, p. 1-28. Bibliographie Breilh J-F., Bertin X., Chaumillon É., Giloy N., .Sauzeau T. (2014). How frequent is storm- induced flooding in the central part of the Bay of Biscay? Global and Planetary Change. Vol. 122, p. 161‑175. Creach A., Pardo S., Guillotreau P., Mercier D. (2015). The use of a micro-scale index to identify potential death risk areas due to coastal flood surges: lessons from Storm Xynthia on the French Atlantic coast, Natural Hazards, vol 77, no 3, p. 1679-1710. Vinet F., Boissier L., Defossez S. (2011). La mortalité comme expression de la vulnérabilité humaine face aux catastrophes naturelles: deux inondations récentes en France (Xynthia, Var, 2010). VertigO- la revue électronique en sciences de l’environnement. Vol. 11, n°2, p. 1-28. 279 Posters 279 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Construction d’une méta-base de ports à l’échelle mondiale Arnaud Ménard 1, Vincent Lambert de Cursay 2, Cyril Ray 3, Charles Guenois 2, Maryam Maslek Elayam 3, Maxime Dréau 1 Arnaud Ménard 1, Vincent Lambert de Cursay 2, Cyril Ray 3, Charles Guenois 2, Maryam Maslek Elayam 3, Maxime Dréau 1 1. Fabrique Numérique de la Défense arnaud@swail.io,maxime@dreau.fr 1. Fabrique Numérique de la Défense arnaud@swail.io,maxime@dreau.fr 1. Fabrique Numérique de la Défense arnaud@swail.io,maxime@dreau.fr 2. Etat Major de la Marine vcursay@laposte.net,charles.guenois@intradef .gouv.fr 3. Ecole Navale cyril.ray@ecole-navale.fr,maryam.maslek_elayam@ecole-navale.fr 1. Introduction L’étude du trafic maritime requiert d’identifier les escales des navires (Carlini et al., 2020 ; Ducruet, 2017). De nos jours, de nombreuses bases de données listant les localisations des ports du monde existent mais l’identification des zones portuaires et de leur extension géographique reste un problème ouvert (Cotteleer, Koldenhof, 2013). Plusieurs travaux ont abordé ce problème par des mesures de densité extraites des positions de navires (Millefiori et al., 2016). Dans ce travail, nous proposons l’intégration d’une indexation spatiale hexago- nale pour permettre une identification de zones portuaires à différentes échelles et permettre la construction d’une méta base de données des zones portuaires. 2. https://data.humdata.org/dataset/global-ports 3. https://geo.data.gouv.fr/fr/datasets/387e3b19086e5d9ca050a68adc01bd4c7acd1842 1. https://msi.nga.mil/Publications/WPI Source de données Notre méthodologie pour la construction d’une méta-base de zones por- tuaires repose sur les listes (et localisations) des ports du monde issues de 4 bases de données: le World Port Index 1, la World Food Programme 2, les mouillages de la Global Fishing Watch 3 et la base de port de l’Ifremer 4. En as- sociant ces listes, nous pouvons garantir une couverture mondiale sur les ports maritimes ainsi que les fleuves principaux. Méthodologie Afin d e c onstruire l es z ones g éographiques c ontenant d es n avires station- naires et de les rapprocher des points de référence, nous nous appuyons sur une discrétisation de l’espace maritime en utilisant plusieurs résolutions de l’index H3 (Brodsky, 2018). Cet index utilise une grille globale discrète basée sur une hiérarchie d’hexagones dite d’ouverture 7. Une telle approche a montré son efficacité dans plusieurs applications de recherche dans le domaine maritime, (Wang et al., 2019), (Maslek Elayam et al., 2021), (Tsatcha, 2014). Notre mé- thodologie est basée sur quatre grandes étapes : 1. Construction des zones géographiques contenant des navires stationnaires : L’analyse de plusieurs mois de trafic maritime (données AIS) a permis d’identifier les hexagones où des navires ont été observés comme sta- tionnaires. 2. Association des zones dites stationnaires aux ports des listes de référence : Les zones contenant des navires stationnaires sont rapprochées des points de référence afin de les associer ou non à un port de chaque liste de port en entrée, par recherche successive dans les hexagones voisins d’un hexagone étudié jusqu’à une certaine distance définie en fonction de chaque liste de port. 3. Création d’une entité port dans la méta-base : Un port dans la méta-base correspond à un tuple contenant la clé identifiant un port dans cha- cune des listes de référence. 3. Création d’une entité port dans la méta-base : Un port dans la méta-base correspond à un tuple contenant la clé identifiant un port dans cha- cune des listes de référence. 4. Nommage des entités port construites par association : Pour chacun de ces tuples, un nom est construit à partir des noms de port de chaque liste de référence. Ce modèle peut induire un nom composé quand deux listes de référence décrivent plus ou moins précisément un port. 4. Nommage des entités port construites par association : Pour chacun de ces tuples, un nom est construit à partir des noms de port de chaque liste de référence. Ce modèle peut induire un nom composé quand deux listes de référence décrivent plus ou moins précisément un port. Identification d e z ones portuaires L’identification e t l a d élimitation d es e spaces p ortuaires p eut ê tre réalisée à partir des traces historiques des navires (Millefiori et al., 2016). L’objectif de cet article est de proposer une méthode de construction et d’identification des zones portuaires sur la base de plusieurs listes de ports et du trafic maritime constaté (fourni par le système AIS (Bole et al., 2014)) sur plusieurs mois afin d’assurer une grande exhaustivité ainsi qu’un relevé précis des zones portuaires. Posters 280 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 4. https://data.humdata.org/dataset/global-ports [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Méta-base de ports 3. Conclusion La construction d’une méta-base de ports exhaustive à partir de données disparates a été proposée pour l’identification des zones portuaires à l’échelle du globe. Cela facilite l’analyse des routes maritimes empruntées et la com- préhension du trafic maritime mondial (Maslek Elayam et al., 2021). Plusieurs perspectives sont actuellement considérées. Une annotation sémantique prenant en compte le type ou le pavillon des navires ainsi que leur domaine de navi- gation permettraient d’identifier plus précisément le type de port (marchand, pétrolier, pêche, plaisance) ainsi que leur rayonnement (international, national, régional). L’intégration de données de ligne de côte permet d’éviter les associa- tions entre deux entités de port proches mais séparées par une langue de terre. Résultats Notre méthode a été implémentée et testée à l’échelle mondiale. Une méta- base contenant environ 21000 entités a été construite. Cette structuration per- met d’ajouter d’autres listes de port de référence sans altérer les données exis- tantes ou de prendre en compte les évolutions des listes utilisées. 281 Posters [ISBN : 978-2-910545-12-1] Bibliographie Bole A., Wall A., Norris A. (2014). Chapter 5 - automatic identification system (ais). In A. Bole, A. Wall, A. Norris (Eds.), Radar and arpa manual, Third Edition éd., p. 255 - 275. Oxford, Butterworth-Heinemann. Consulté sur http:// www.sciencedirect.com/science/article/pii/B9780080977522000052 Brodsky I. (2018). H3: Uber’s hexagonal hierarchical spatial index. Available from Uber Engineering website: https://eng. uber. com/h3/[22 June 2019]. Carlini E., Lira V. M. de, Soares A., Etemad M., Machado B. B., Matwin S. (2020). Uncovering vessel movement patterns from ais data with graph evolution analy- sis. In Proceedings of the 22nd International Conference on Extending Database Technology (EDBT), March 30-April 2. Cotteleer A., Koldenhof Y. (2013). Ship’s travelling time in approaching the port of rotterdam. In Workshop on moving objects at sea, p. 1–4. Ducruet C. (2017). Advances in shipping data analysis and modeling: Tracking and mapping maritime flows in the age of big data. Routledge. Maslek Elayam M., Ray C., Claramunt C. (2021). Modèle de graphe pour l’analyse des structures de trajectoires maritimes. In 21ème édition de la conférence extraction et gestion des connaissances (egc). Millefiori L. M., Zissis D., Cazzanti L., Arcieri G. (2016). Scalable and distributed sea port operational areas estimation from ais data. In 2016 ieee 16th international conference on data mining workshops (icdmw), p. 374–381. Tsatcha D. (2014). Contribution à l’extraction et à la représentation des connais- sances de l’environnement maritime: Proposition d’une architecture dédiée. aux applications de navigation. Thèse de doctorat non publiée. Wang X., Li J., Zhang T. (2019). A machine-learning model for zonal ship flow prediction using ais data: A case study in the south atlantic states region. Journal of Marine Science and Engineering, vol. 7, no 12, p. 463. 282 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Jessica PIC1, Etienne COSSART1, Mathieu FRESSARD2 1.UMR 5600 – Environnement Ville Société, Université Jean Moulin Lyon 3 18 rue chevreul, 69362 Lyon Cedex, France Correspondance : jessica.pic@univ-lyon3.fr 1.UMR 5600 – Environnement Ville Société, Université Jean Moulin Lyon 3 18 rue chevreul, 69362 Lyon Cedex, France Correspondance : jessica.pic@univ-lyon3.fr 2.UMR 5600 – Environnement Ville Société, CNRS 5 avenue Pierre Mendès-France, 69676 Bron Cedex, France 2.UMR 5600 – Environnement Ville Société, CNRS 5 avenue Pierre Mendès-France, 69676 Bron Cedex, France MOTS-CLES : STRUCTURE PAYSAGERE, MODELISATION SPATIALE, CONNECTIVITE, BEAUJOLAIS Posters 283 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 L’utilisation par l’agriculture de produits phytosanitaires est à l’origine d’une pollution généralisée des sols et des cours d’eau et des enjeux d’ordre environnementaux, économiques et sanitaires (Mateo-Sagasta et al., 2018). Parmi les différents types de cultures, la vigne mobilise d’importantes quantités de produits phytosanitaires. Dans une perspective de limitations des transferts de sédiments et de produits phytosanitaires associés des parcelles agricole au cours d’eau, une prise en compte de la dynamique de transferts à l’échelle du bassin versant – unité hydrologique et de gestion – paraît nécessaire à être mise en œuvre. Toutefois, les transferts de produits phytosanitaires des parcelles agricole au cours d’eau restent difficilement appréhendés en dehors d’une considération stationnaire et bien que diverses études scientifiques (Reichenberger et al., 2007 ; Gascuel-Odoux et al., 2011) aient souligné l’importance de l’interaction entre des objets de différente nature et des processus agissant à des échelles variables. Trois facteurs influencent les transferts de produits phytosanitaires : - les pratiques culturales conditionnent les quantités de produits pouvant être transférées : quantités et des calendriers d’application ; - la dynamique hydrologique : le ruissellement de surface constitue le vecteur préférentiel des transferts par le biais des de l’eau et des sédiments issus de l’érosion des sols ; - la structure paysagère (éléments du paysage et leurs interrelations), intervient dans le schéma d’organisation spatiale et temporelle des transferts (Gascuel- Odoux et al., 2011 ; Reulier, 2015). Jessica PIC1, Etienne COSSART1, Mathieu FRESSARD2 Elle conditionne la connectivité des parcelles, c’est-à-dire leur capacité à transférer les sédiments et produits phytosanitaires au cours d’eau ; Dans une perspective d’étude de la connectivité structurelle, c’est-à-dire de la prise en compte des caractéristiques des éléments du paysage et leur configuration spatiale, nous nous intéressons à l’évaluation de l’influence des éléments du paysage et de leur agencement spatial sur la dynamique des flux de produits phytosanitaires associés aux particules de sol pour deux petits bassins versants (8km²) du Beaujolais viticole. Dans ce cadre, trois types d’éléments du paysage doivent être pris en compte sur le territoire : les modes d’occupation du sol, les infrastructures linéaires de drainage des versants et les bacs de collecte de sédiments. Si la parcelle constitue la maille de référence de la structure paysagère et le lieu de production du ruissellement, ce sont les infrastructures de drainage et leur structuration en réseau qui régissent les potentialités de transferts de sédiments et de produits phytosanitaires d’une parcelle au cours d’eau (fig.1). Nous nous concentrerons ici sur ces éléments et sur la méthode développée pour les envisager. 284 Posters 284 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Implications des infrastructures de drainage des versants dans la dynamique de transferts de sédiments et de produits phytosanitaires, le cas du Beaujolais viticole Un inventaire exhaustif des infrastructures de drainage a été réalisé par travail de terrain, dans le but d’identifier les secteurs de forte connectivité, l’influence de la structuration des sous-réseaux et de construire une méthodologie d’analyse reproductible à différents contextes paysagers. Les réseaux de drainage sont ensuite intégrés et analysés sous SIG sous la forme d’un graphe orienté en fonction de l’écoulement de surface. L’idée étant d’identifier l’ensemble des points de départ et les exutoires des sous-réseaux, ainsi que les séquences empruntées, c’est-à-dire l’enchainement d’infrastructures de drainage entre un point de départ et un exutoire. FIGURE 1. Influence des infrastructures de drainage sur les flux hydro- sédimentaires, secteur de la commune de Villié-Morgon (Beaujolais, France) FIGURE 1. Influence des infrastructures de drainage sur les flux hydro- sédimentaires, secteur de la commune de Villié-Morgon (Beaujolais, France) La densité d’infrastructures linéaires au sein des zones d’étude révèle une forte anthropisation des versants. Reulier, R. (2015). Structure paysagère et dynamiques spatiales des transferts hydro- sédimentaires Approche par simulation multi-agents. Université de Caen Normandie. Jessica PIC1, Etienne COSSART1, Mathieu FRESSARD2 L’analyse des séquences – longueur de réseaux, nombre d’éléments dans la séquence, fréquence d’occurrence de séquence – est très variable en fonction des caractéristiques des zones d’exutoire, appuyant la pertinence de l’analyse de la structure du réseau dans la connectivité des parcelles au cours d’eau. Toutefois, une compréhension approfondie de la connectivité structurelle ne pourra avoir lieu que par l’intégration dans le graphe des bacs de collecte de sédiments et du réseau parcellaire. Mateo-Sagasta, J., Marjani Zadeh, S., Turral, H. (2018). More people, more food, worse water? a global review of water pollution from agriculture. Rome, Italy: FAO. 225p. Gascuel-Odoux, C., Aurousseau, P., Doray, T., Squividant, H., Macary, F., Uny, D., & Grimaldi, C. (2011). Incorporating landscape features to obtain an object-oriented landscape network representing the connectivity of surface flow pathways over rural catchments. Hydrological Processes, 25, 3625–3636 Reichenberger, S., Bach, M., Skitschak, A., & Frede, H. G. (2007). Mitigation strategies to reduce pesticide inputs into ground- and surface water and their effectiveness; A review. Science of the Total Environment, 384(1–3), 1–35. 285 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Marie Ramel-Delobel123, Thomas Coudon 123, Lény Grassot 12, Delphine Praud12, Béatrice Fervers12, Pietro Salizzoni3 1. Département Prévention, Cancer, Environnement Centre Léon Bérard Cheney C, 28 rue Laennec, 69008, Lyon, France Marie.ramel-delobel@lyon.unicancer.fr 1. Département Prévention, Cancer, Environnement Centre Léon Bérard 1. Département Prévention, Cancer, Environnement Centre Léon Bérard 1. Département Prévention, Cancer, Environnement Centre Léon Bérard Cheney C, 28 rue Laennec, 69008, Lyon, France Marie.ramel-delobel@lyon.unicancer.fr Cheney C, 28 rue Laennec, 69008, Lyon, France Marie.ramel-delobel@lyon.unicancer.fr 2. INSERM UA8 Radiations : Défense, Santé, Environnement Centre Léon Bérard, Ministère des Armées, Service de Santé des Armées (SSA) Cheney A, 28 rue Laennec, 69008, Lyon, France 2. INSERM UA8 Radiations : Défense, Santé, Environnement Centre Léon Bérard, Ministère des Armées, Service de Santé des Armées (SSA) Cheney A, 28 rue Laennec, 69008, Lyon, France 3. CNRS UMR 5509 Laboratoire de Mécanique des Fluides et d’Acoustique Ecole Centrale de Lyon, Université Claude Bernard Lyon I 36 avenue Guy de Collongue, 69132 Ecully, France Exposition à la pollution atmosphérique et risque de cancer du sein : vers une prise en compte des mobilités quotidiennes dans les études épidémiologiques Marie Ramel-Delobel123, Thomas Coudon 123, Lény Grassot 12, Delphine Praud12, Béatrice Fervers12, Pietro Salizzoni3 MOTS-CLES : GEOGRAPHIE, POLLUTION ATMOSPHERIQUE, CANCER DU SEIN, MOBILITES QUOTIDIENNES, NO2, PM Contexte : Le cancer du sein est le cancer le plus fréquent chez la femme en France et au niveau mondial. L’augmentation de son incidence ces 30 dernières années est associée au dépistage généralisé, aux traitements hormonaux de la ménopause et aux changements de mode de vie. Cependant, le rôle des facteurs environnementaux est aussi suspecté. Plusieurs études épidémiologiques suggèrent en effet un lien entre l'exposition à long terme à la pollution de l'air ambiant et le risque de cancer du sein (White et al., 2018). De nombreuses études épidémiologiques utilisent l’historique des adresses résidentielles pour évaluer l’exposition aux polluants de l’air, ce qui peut entraîner une imprécision des estimations. En effet, l’exposition « chronique » survient tout au long de la journée, au domicile, mais également sur le lieu du travail et d’autres activités ainsi que pendant les mobilités. Or, bien que les déplacements représentent une faible partie du temps circadien, ils peuvent représenter une part importante de la dose globale qu’une personne inhale dans la journée (de Nazelle et al., 2012). Posters 286 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Ce travail, effectué dans le cadre d’un stage de Master 2 qui se poursuit désormais en thèse s’inscrit dans le projet APoPCo1 - collaboration Département Prévention Cancer Environnement du Centre Léon Bérard et Laboratoire de Mécanique des Fluides et d’Acoustique de l’Ecole Centrale de Lyon (CNRS UMR 5509) ; financé par l’ANSES et la Fondation de France. APoPCo vise à estimer l’association entre le risque de cancer du sein (étude cas-témoins nichée dans la cohorte nationale E3N) et l’exposition à la pollution atmosphérique en prenant en compte les expositions chroniques, à la fois résidentielles, professionnelles, et durant les déplacements domicile-travail (DDT). Objectif : L’objectif est ici de développer une méthode de calcul des expositions pendant les déplacements – et en premiers desquels les DDT – en vue d’évaluer leur contribution dans l’exposition totale de la journée. La part de ce type d’exposition est analysée au regard de l’influence des différents modes de déplacement, des itinéraires et des types de polluants (e.g. NO2 et PM10). 1 Atmospheric Pollution and Physical activity linked with Commute 2 Test de la méthode appliquée à partir d’un couple d’adresses résidentielles issues de E3N et d’adresses professionnelles aléatoires sur la Métropole lyonnaise (partie Master) 3 Application de la méthode sur les adresses professionnelles géocodées (partie Thèse) 2 Test de la méthode appliquée à partir d’un couple d’adresses résidentielles issues de E3N et d’adresses professionnelles aléatoires sur la Métropole lyonnaise (partie Master) 3 Application de la méthode sur les adresses professionnelles géocodées (partie Thèse) 1 Atmospheric Pollution and Physical activity linked with Commute 2 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Mobilités et pollution atmosphérique : risque de cancer du sein Mobilités et pollution atmosphérique : risque de cancer du sein étudié, les DDT représentent entre 3 et 6.6% de la dose en NO2, et 5.7 à 11.2% en PM10 pour une part du temps relativement faible (3% du temps circadien), en comparaison des parts du temps et des expositions estimés en moyenne au domicile (respectivement 60% et 70%) et au travail (30% et 25%). Ensuite, l’exposition attribuable aux DDT varie selon les caractéristiques propres à l’individu. Le premier facteur est le choix du mode (motorisé, vélo, marche) : le vélo est celui qui contribue le plus à augmenter la dose inhalée dans la journée, suivi de la marche, et enfin de la voiture, dont les valeurs sont en moyenne 2 fois inférieures à celles des modes actifs. Cette hiérarchie des modes s’explique par le fait que les modes actifs induisent une activité physique qui accroît le volume respiratoire. Enfin, la dose inhalée est influencée par le choix de l’itinéraire dont ceux exposant l’individu à de plus fortes concentrations empruntent les axes les plus directs et les plus rapides sur lesquels la vitesse autorisée varie entre 80 à 130 km/h. Conclusion : L’exposition aux polluants de l’air pendant les DDT varie selon les caractéristiques propres à l’individu (volume respiratoire), celles de sa mobilité (mode, itinéraire), de même que selon les caractéristiques de l’environnement urbain (structuration du réseau routier, vitesse limite autorisée, trafic). La méthode présentée, toujours en cours de développement, fournit une première estimation de l’exposition pendant les déplacements et elle est amenée à être approfondie pour affiner ces résultats intermédiaires. Ces précisions concerneront notamment la classification des modes les plus ou moins exposés, en intégrant plus de paramètres pour définir l’exposition en automobile. En outre, la pratique d’une activité physique, même dans un environnement pollué, permettrait de réduire les risques de mortalité. La question des bénéfices de l’utilisation des modes actifs sera donc également à appréhender. Enfin, une des perspectives sera de déterminer si la prise en compte des mobilités dans les études épidémiologiques influence les conclusions antérieures sur les effets sanitaires de la pollution. White, A. J., Bradshaw, P. T., & Hamra, G. B. (2018). Air pollution and Breast Cancer : A Review. Current Epidemiology Reports, 5(2), 92‑100. https://doi.org/10.1007/s40471-018- 0143-2 MOTS-CLES : GEOGRAPHIE, POLLUTION ATMOSPHERIQUE, CANCER DU SEIN, MOBILITES QUOTIDIENNES, NO2, PM Matériel et méthode : Ce projet s’appuie sur des données préalablement recueillies et produites par l’équipe dans le cadre du projet XENAIR (2016-2021 ; financement par l’Institut National du Cancer et de la Fondation ARC) : ̶ un questionnaire sur l’histoire résidentielle et professionnelle, et les modes des DDT des sujets d’étude (10 444 femmes, 15 865 adresses résidentielles entre 1990 et 2010, adresses professionnelles en cours de finalisation) ; ̶ un questionnaire sur l’histoire résidentielle et professionnelle, et les modes des DDT des sujets d’étude (10 444 femmes, 15 865 adresses résidentielles entre 1990 et 2010, adresses professionnelles en cours de finalisation) ; ̶ des données de concentrations de NO2 et de PM10 à l’échelle locale (modèle SIRANE, 10x10m, horaire) (Soulhac et al., 2017), et nationale (modèle Land Use Regression (LUR), 50x50m, annuel), sur une période de 21 ans (1990-2010) (Amadou et al., 2020) ; ̶ des données de concentrations de NO2 et de PM10 à l’échelle locale (modèle SIRANE, 10x10m, horaire) (Soulhac et al., 2017), et nationale (modèle Land Use Regression (LUR), 50x50m, annuel), sur une période de 21 ans (1990-2010) (Amadou et al., 2020) ; La méthode de quantification de l’exposition, développée via R, se découpe en 4 étapes qui correspondent à différentes échelles d’analyse, allant de l’échelle globale à celle de l’individu. En premier lieu, les concentrations selon le type de polluant sont estimées à l’échelle de la rue (1). En second lieu, il est possible de déduire les expositions induites par les microenvironnements des modes de déplacement (2). Ensuite, les itinéraires calculés pour chaque individu permettent de définir les concentrations auxquelles il est exposé (3), pour enfin, calculer la dose qu’il inhale selon l’intensité du volume respiratoire (4). En dernier lieu, les résultats sont analysés afin d’identifier la variation de l’exposition qu’induit le choix du mode, de même que la part de l’exposition quotidienne totale attribuable aux déplacements pendulaires. Résultats : Les premiers résultats intermédiaires2 – à confirmer avec toute la population d’étude3 – montrent que la dose de polluant inhalée pendant les DDT participe entre 5 à 10% en moyenne à la dose quotidienne totale. Pour chaque polluant Posters 287 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Samuel ROBERT1, Brieuc CABIOCH1, Eric CARROLL1, Marie- Laure TREMELO1 1. ESPACE, UMR 7300, CNRS, Aix Marseille Université, Avignon Université, Université Côte d’Azur Technopole de l’environnement Arbois-Méditerranée F-13545 Aix-en-Provence Cedex 4, France samuel.robert@univ-amu.fr 1. ESPACE, UMR 7300, CNRS, Aix Marseille Université, Avignon Université, Université Côte d’Azur Technopole de l’environnement Arbois-Méditerranée F-13545 Aix-en-Provence Cedex 4, France samuel.robert@univ-amu.fr 1. ESPACE, UMR 7300, CNRS, Aix Marseille Université, Avignon Université, Université Côte d’Azur Technopole de l’environnement Arbois-Méditerranée F-13545 Aix-en-Provence Cedex 4, France samuel.robert@univ-amu.fr MOTS-CLES : PLAGE, FRÉQUENTATION, ÉVALUATION, MODÈLE, MARSEILLE Contexte Dans les grandes villes littorales, le bord de mer et les plages forment un espace public et récréatif majeur. Les flux de visiteurs qui s’y rendent nécessitent d’être évalués et suivis pour permettre la conception et l’ajustement des mesures de gestion : surveillance de la baignade, nettoiement, travaux d’entretien, services aux usagers, gestion de l’affluence, ou surveillance des faits de délinquance, pour ne citer que quelques exemples. Pourtant, si des expérimentations diverses ont vu le jour (Balouin et al., 2014 ; Guyonnard et al., 2020 ; Guyonnard et Vacher, 2018 ; Guyonnard, 2017), rares sont les villes disposant de dispositifs d’évaluation et de suivi fin de la fréquentation de leurs plages. Références bibliographiques Amadou, A., Praud, D., Coudon, T., Danjou, A. M. N., Faure, E., Leffondré, K., Le Romancer, M., Severi, G., Salizzoni, P., Mancini, F. R., & Fervers, B. (2020). Chronic long-term exposure to cadmium air pollution and breast cancer risk in the French E3N cohort. International Journal of Cancer, 146(2), 341‑351. https://doi.org/10.1002/ijc.32257 de Nazelle, A., Fruin, S., Westerdahl, D., Martinez, D., Ripoll, A., Kubesch, N., & Nieuwenhuijsen, M. (2012). A travel mode comparison of commuters’ exposures to air pollutants in Barcelona. Atmospheric Environment, 59, 151‑159. https://doi.org/10.1016/j.atmosenv.2012.05.013 Soulhac, L., Nguyen, C. V., Volta, P., & Salizzoni, P. (2017). The model SIRANE for atmospheric urban pollutant dispersion. PART III : Validation against NO2 yearly concentration measurements in a large urban agglomeration. Atmospheric Environment, 167, 377‑388. https://doi.org/10.1016/j.atmosenv.2017.08.034 White, A. J., Bradshaw, P. T., & Hamra, G. B. (2018). Air pollution and Breast Cancer : A Review. Current Epidemiology Reports, 5(2), 92‑100. https://doi.org/10.1007/s40471-018- 0143-2 288 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Vers une détermination automatique du nombre d’usagers sur une plage urbaine De l’observation au modèle à base d’intelligence artificielle pour la plage du Prophète à Marseille Samuel ROBERT1, Brieuc CABIOCH1, Eric CARROLL1, Marie- Laure TREMELO1 Objectif Ce poster évoque comment cette problématique est travaillée à Marseille, France. Depuis 2013, plusieurs plages de la ville sont l’objet d’observations des fréquentations et des pratiques au cours de l’été. Ces travaux ont permis de saisir les motivations des usagers, mieux appréhender les plages en termes de qualité environnementale (Labille et al., 2020 ; Tournoud et al., 2019), et révéler les flux de personnes (Robert et Trémélo, 2018 ; Robert et Trémélo, 2020a, 2020b, 2019). Sur 289 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] ce dernier point, de nombreuses données ont été produites pour mettre en évidence une voire des fréquentations-types. Actuellement, les travaux visent à définir un modèle pour l’une des plages les plus étudiées : le Prophète. Résultats Des observations exploratoires répétées sur diverses plages de la ville ont permis de saisir l’hétérogénéité de la fréquentation. Pendant l’été, une grande variabilité a été mise en évidence d’un jour à l’autre sur une même plage, et d’une plage à l’autre lors d’un même jour. Des effets de sites ont également été mis en relief, de même que l’influence d’évènements locaux pouvant affecter une plage à un moment donné. Pour une étude plus approfondie, le Prophète a été choisi comme site expérimental. On y dispose de données variées : des effectifs d’usagers dénombrés en continu aux entrées et sorties de la plage entre 8h et 20h ; des photographies obliques prises d’un point haut à chaque heure (8h, 9h, 10h, etc.), permettant de dénombrer les usagers par photo-interprétation ; et des questionnaires sur les pratiques. L’objectif est de concevoir un protocole de dénombrement automatique des usagers à partir de photos qui seraient prises par un système installé sur site. D’ores et déjà, les photos et les fichiers de semis de points localisant les usagers sur chacune d’elles constituent un jeu de données conséquent, qui permet d’engager une expérimentation en intelligence artificielle (IA), en entrainant un réseau de neurones convolutifs (CNN). Ces algorithmes d’IA en apprentissage profond (deep learning) s’inspirant du cortex humain, sont conçus pour effectuer des tâches descriptives et génératives sur des images numériques. Leurs applications sont nombreuses notamment dans la médecine, la reconnaissance faciale, la transcription manuscrite ou les véhicules autonomes. Leur élaboration nécessite un entrainement sur de nombreuses images pour lesquelles on dispose des données qui font l’objet de l’apprentissage afin de produire un système capable d’inférer les données souhaitées sur de nouvelles images. Les spécificités des photos de plage dont nous disposons nous orientent vers des algorithmes spécifiques du champ du comptage de foule dont l’objectif est de prédire des densités, des effectifs (Zhang et al., 2016, Li et al., 2018), plutôt que ceux communément mobilisés dans le comptage par reconnaissance de forme (Balouin et al., 2014). Les premiers résultats des modèles entraînés sur des photos de foules appliqués sur les photos de la plage du Prophète sont encourageants, mais des biais subsistent. La réussite de l’expérimentation pourrait profondément influer sur le suivi à venir des plages du littoral marseillais. 290 Posters 290 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Vers une détermination automatique du nombre d’usagers sur une plage urbaine Bibliographie non numérotée et références Balouin Y. et al., 2014. Automatic assessment and analysis of beach attendance using video images at the Lido of Sète beach, France. Ocean and Coastal. Management, 102. 10.1016/j.ocecoaman.2014.09.006 Guyonnard V. 2017. Dimensions cachées et attentes spatiales dans un espace de pratique de tourisme et de loisir : une analyse géographique de la plage en Charente-Maritime (France). Th. de géographie, Univ. La Rochelle. Guyonnard V. et al., 2020. REMEPP: a simulation model of beach filling as a tourism and leisure practice. Cybergeo. https://journals.openedition.org/cybergeo/35378 Guyonnard V., Vacher L., 2018. Penser et mesurer les distances de l’interaction sociale dans l’espace de la plage. L’Espace géographique, 47, 2, pp. 159-181. ⟨ 10.3917/eg.472.0159 ⟩ Labille J. et al., 2020. Assessing UV filter inputs into beach waters during recreational activity: A field study of three French Mediterranean beaches from consumer survey to water analysis. Science of the Total Environment, 706. <10.1016/j.scitotenv.2019.136010 > Li Y. et al., 2018. CSRNet : Dilated convolutional neural networks for understanding the highly congested scenes. In : Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, pp.1091-1100. Robert S., Apers-Trémélo M.-L., 2020a. Données de fréquentation de la plage du Prophète (Marseille) le 17 juillet 2019, 26 p. ⟨hal-02461322 ⟩ Robert S., Apers-Trémélo M.-L., 2020b. Données de fréquentation de la plage du Prophète (Marseille) du 10 au 16 juillet 2017, 110 p. ⟨hal-02907086 ⟩ Robert S., Apers-Trémélo M.-L., 2019. Données de fréquentation de la plage du Prophète (Marseille) le 18 juillet 2018, 26 p. ⟨ hal-02438256v2 ⟩ Robert S., Trémélo M.L., 2018. Analyse spatiale des pratiques des usagers de trois plages urbaines de Marseille. merIgéo, Aix-en-Provence, Fr. ⟨hal-01746087 ⟩ Tournoud M.-G. et al., 2019. Microbiological quality of bathing waters versus sanitary regulation and beach users’ practices. A case study in Marseilles. Séminaire annuel du Labex DRIIHM, Oct 2019, Lyon, France. ⟨ hal-02409637 ⟩ Zhang Y. et al., 2016. Single-image crowd counting via multi-column convolutional neural 597. network. In : Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, pp. 589- Zhang Y. et al., 2016. Single-image crowd counting via multi-column convolutional neural 597. network. In : Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, pp. 589- 291 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Apport de l’utilisation du drone et de transects écopaysagers pour l’échantillonnage spatial d’habitats naturels : le cas des digues de la vallée de la Dives (Calvados) transects écopaysagers pour l’échantillonnage spatial d’habitats naturels : le cas des digues de la vallée de la Dives (Calvados) Nicolas ROBINET1, Gaëtan JOLLY2, Guy LEMPERIERE3, Yves PETIT-BERGHEM3 1. Cermosem (Institut d’Urbanisme et de Géographie Alpine – Université Grenoble Alpes / UMR Pacte) 1064 chemin du Pradel 07170 Mirabel, France Nicolas.robinet@univ-grenoble-alpes.fr 2. Diplômé du Master Expertise et Gestion de l’Environnement Littoral, Institut Universitaire Européen de la Mer, Technopôle Brest-Iroise rue Dumont d’Urville - 29280 Plouzané, France jolly.gaetan@gmail.com 3. Larep (Laboratoire de recherche en projet de paysage), École Nationale Supérieure du Paysage 10 avenue du Maréchal Joffre 78000 Versailles, France guy.lemperiere@ird.fr y.petitberghem@ecole-paysage.fr MOTS-CLÉS : drone, écologie du paysage, transects, indices, méthode d’échantillonnage, habitats, digues, milieu littoral, hausse du niveau marin, stratégie de gestion territoriale  KEYWORDS: drone, landscape ecology, transects, metrics, sampling method, habitats, dikes, coastal environment, sea level rise, territorial management strategy Nicolas ROBINET1, Gaëtan JOLLY2, Guy LEMPERIERE3, Yves PETIT-BERGHEM3 1. Cermosem (Institut d’Urbanisme et de Géographie Alpine – Université Grenoble Alpes / UMR Pacte) 1064 chemin du Pradel 07170 Mirabel, France Nicolas.robinet@univ-grenoble-alpes.fr 2. Diplômé du Master Expertise et Gestion de l’Environnement Littoral, Institut Universitaire Européen de la Mer, Technopôle Brest-Iroise rue Dumont d’Urville - 29280 Plouzané, France jolly.gaetan@gmail.com 2. Diplômé du Master Expertise et Gestion de l’Environnement Littoral, Institut Universitaire Européen de la Mer, Technopôle Brest-Iroise rue Dumont d’Urville - 29280 Plouzané, France jolly.gaetan@gmail.com MOTS-CLÉS : drone, écologie du paysage, transects, indices, méthode d’échantillonnage, habitats, digues, milieu littoral, hausse du niveau marin, stratégie de gestion territoriale KEYWORDS: drone, landscape ecology, transects, metrics, sampling method, habitats, dikes, coastal environment, sea level rise, territorial management strategy 292 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 1. Introduction, contexte, problématique et objectif du poster 1. Dans le contexte des changements globaux, la digue est amenée à se transformer et à faire face à un certain nombre de risques (inondation, submersion…) susceptibles de se manifester. Accompagner les acteurs locaux dans ce contexte changeant revient à entrecroiser de multiples enjeux et à réfléchir à des orientations variées de gestion des digues, parfois nouvelles en France, oscillant entre fermeture et ouverture, rehaussement et abaissement, artificialisation et renaturation. L’objectif majeur du programme est l'établissement de propositions nouvelles relatives au devenir des digues et l'évaluation des conditions de faisabilité et d'acceptabilité de ces propositions. Pour y parvenir, l’un des enjeux est de pouvoir connaître précisément l’état des digues et leur capacité à accueillir des habitats naturels dont la spatialisation à échelle fine s’avère indispensable. Dans le cadre du site des digues de la Dives, l’accès aux quatre kilomètres de berges est problématique. De nombreuses portions sont des propriétés privées. De plus, l’entretien des digues a été délaissé depuis une cinquantaine d’années, ce qui a entraîné le recouvrement progressif de ces infrastructures par une végétation haute, rendant la prospection sur site difficile. Cette problématique écarte l’emploi de méthodes terrestres traditionnelles de relevés de végétation et d’habitats, comme par exemple un échantillonnage de type Braun-Blanquet. En effet, cela nécessiterait de défricher des zones d’accès et risquerait d’endommager les habitats pour parvenir au site. Nous avons mobilisé différents moyens pour élaborer une méthode de relevé non destructif, permettant de cartographier les habitats des deux rives. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Drones et transects écopaysagers 3. Résultats et discussion Cet ensemble de données permet d’établir une carte des habitats à une définition scalaire précise (de l’ordre de 0,20 m) et permet de réaliser par la suite des calculs d’indices caractérisant les diversités bêta et gamma. Les indices de Jaccard et Sorensen montrent des paysages aux habitats très différents entre les deux rives avec notamment, en aval du cours d’eau, des enrochements facilitant l’installation d’espèces pionnières, des plantations de peupliers en rive droite, des milieux ouverts en rive gauche essentiellement (schorres). L’indice de Shannon indique une hétérogénéité plus élevée pour la rive droite (H= 1,67) en comparaison de la rive gauche (H=1,46). Cette disparité s’explique en grande partie par l’usage des sols de part et d’autre de la rivière. En partie amont, nous retrouvons des zones de pâturage extensif. Les digues, surmontées de chemins de halage, ont été abandonnées, ce qui a conduit à un enfrichement progressif de ces structures. Des brèches sont disposées tout au long de la rivière, ce qui profite aux bovins pour s’abreuver et accentue la détérioration des digues. En partie, les parcelles attenantes à la Dives s’urbanisent et les berges constituent des espaces publics. Elles sont encore surmontées par des sentiers de halage, ce qui a permis de conserver les digues. Quelques limites ont été observées concernant l'usage du drone. En effet, les survols de certaines zones font l’objet d’autorisations spécifiques, souvent longues à obtenir et à concilier avec les aléas météo. Ce facteur est à prendre en compte, car il peut rallonger, de façon non négligeable, les délais d’étude, sachant qu’un refus n’est pas à exclure. Les espaces boisés représentent eux aussi un frein à la photo- interprétation (étagement de la végétation). Les contraintes techniques peuvent également être un frein : coût, logistique, stockage de la donnée, temps de traitement. L’observation à partir du cours d’eau montre elle aussi quelques limites. Le champ visuel étant restreint par les digues, les habitats écologiques situés de l’autre côté du cours d’eau (face terre) ne peuvent être observés. On ne dispose alors que d’une perception partielle des habitats repérés. 2. Matériel, méthodes et outils Pour analyser dans sa globalité le site d’étude, nous utilisons une approche écopaysagère afin de qualifier le paysage et les habitats des digues aval de la Dives. Pour pallier l’impossibilité d’accéder aux berges, nous avons réalisé deux missions de survol aéroporté qui ont permis de couvrir 7 secteurs pour une surface d’environ 47 hectares. Le traitement numérique des données et la photo-interprétation ont permis la caractérisation des habitats, de la végétation, l’identification des ruptures de pente, l’exposition, la présence de fossés et la réalisation de profils altimétriques. En complément, un parcours aquatique effectué le long de la Dives a permis de réaliser un transect utilisant une méthode dite du quadrat par Point Centré (PCQM) modifiée dans le but d’échantillonner les différents habitats placés le long du cours d’eau sur un parcours aquatique de 4 km et 17 points d’échantillonnage. Cette technique est utilisée pour estimer la surface des différents types d’habitats ou d’objets naturels. Couplée aux relevés aéroportés, elle a permis de déterminer les limites des différents habitats classés selon le Code Corine Biotope, leur taille, leur densité, la fréquence de leur répartition et leur arrangement spatial contagieux ou aléatoire. 293 Posters 293 [ISBN : 978-2-910545-12-1] Le poster présente les éléments suivants : Le poster présente les éléments suivants :  1ère partie : un résumé rapide du contexte de l’étude et de la problématique, accompagné par une carte de localisation du site d’étude.  2e partie : un exposé de la méthode employée, appuyée par une carte de localisation des points d’échantillonnage et les zones de survols par drone. 3e partie : présentation des résultats qui comportera : * une carte des habitats et tableau comparatif des indices, * un bloc-diagramme, * des photographies, * un encart de définition des indices utilisés.  Dernière partie : ouverture sur les limites de la recherche entreprise accompagnée d’une discussion. Posters 294 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] 1. Introduction La pollution plastique est considérée comme un problème environnemental majeur, avec plus de cinq trillions de déchets plastiques répandus dans les océans (Eriksen et al., 2014). Cette accumulation croissante dans les mers et les océans constitue un risque majeur pour les écosystèmes marins, notamment du fait de son introduction dans la chaîne alimentaire. Les déchets plastiques peuvent être classés en fonction de leurs dimensions, et en trois types: macro- plastique (plus de 2,5 cm), méso-plastique (entre 5 mm et 2,5 cm) et micro- plastique (taille inférieure à 5 mm). L’objectif de recherche de cet article est d’introduire une méthodologie gé- nérale pour observer, modéliser, visualiser les déplacements spécifiquement des macro-plastiques en mer afin d’en faciliter la récupération. Processus de modélisation et optimisation pour la collecte de déchets plastiques en mer Loic Salmon 1, Nicolas Maslov 1, Chistophe Claramunt 1 Institut de recherche de l’Ecole Navale, 29240 Brest Cedex 9, France loic.salmon@ecole-navale.fr 2. Méthodologie générale pour le suivi et la collecte de macro-déchets plastiques en mer La chaîne de processus allant de l’observation des macro-déchets plastiques au calcul d’une route otpimale permettant d’aller les collecter est décrite dans la figure 1. La première étape consiste à développer et formaliser (1) un modèle de données pour la récupération et la gestion des données géolocalisées des macro-déchets. La deuxième étape concerne (2) les aspects de fusion et de mise à jour alors que différentes sources d’observation sont disponibles et que les zones de déchets se déplacent constamment en raison des courants et des vents. Une fois que que la base de données est alimentée et que les mécanismes de mise à jour sont effectifs (3) l’extraction d’une carte de densité probabiliste des déchets plastiques permet de visualiser l’emplacement des amas. Enfin la 295 Posters 295 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] dernière étape consiste à développer (4) un algorithme d’aide à la décision pour définir la voie la plus favorable pour collecter autant de déchets plastiques que possible tout en tenant compte des capacités du navire de collecte. Figure 1 Chaîne de traitement pour la collecte de plastiques en mer Figure 1. Chaîne de traitement pour la collecte de plastiques en mer Figure 1. Chaîne de traitement pour la collecte de plastiques en mer (1) Modèle de données générique pour la gestion de données de macro- déchets en mer. Un modèle de données spatio-temporel est nécessaire pour appréhender l’observation et le déplacement d’amas de macro-déchets en mer. Deux difficultés interviennent une concernant la composition de ces amas de déchets pouvant contenir des éléments de différentes sortes. L’autre intervient quant aux différentes sources des données tel que des relevés, des données sa- tellitaires et dont le niveau de confiance et la précision varient. Enfin, des méthodes doivent également être développées pour extraire les positions des amas de déchets à partir d’images. (2) Mécanismes de mise à jour et de fusion des données de macro-déchets. D’abord, les différentes sources d’observations de macro-déchets nécessitent des mécanismes de fusion (Ghamisi et al., 2018). Ensuite, ces amas sont en constant déplacement, soumis aux vents et aux courants marins. Des mécanismes sont donc nécessaires pour mettre à jour la base de données avec les nouvelles posi- tions des amas déplacés par les vents et courants ainsi que les nouveaux reports des données et de les croiser. (3) Extraction de carte de densité probabiliste et visualisation de macro- déchets. [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Méthodologie pour la collecte de plastiques en mer mer a été introduit et permet de générer des cartes de densité de probabilité de répartition des déchets en mer. (4) Optimisation pour trouver la route optimale de collecte de déchets en mer. A partir de la carte de densité de probabilité construite précédemment, il s’agit ensuite de trouver la ou les routes optimales pour le ramassage de déchets plastiques. Une série de paramètres sont à prendre en compte tels que la quantité de plastique collectée, le temps de navigation ou l’énergie consommée. 3. Premiers résultats Des données de positions d’amas plastiques ont été simulées aux abords de la Bretagne, et des données courant fournies par le SHOM ont été utilisées pour estimer la dérive de ces déchets en mer. A gauche (Figure 2), le dégradé de couleur correspond à la probabilité de densité de déchets plastiques pour 1000 simulations de dérive grâce à un modèle probabiliste en fonction des cou- rants sur 12h depuis les reports de positions initiales. A droite (Figure 2), un exemple de trajectoire pour le ramassage de déchets en mer, tenant compte de ces courants. Figure 2. Carte de densité et simulation de ramassage de déchets plastiques Figure 2. Carte de densité et simulation de ramassage de déchets plastiques Eriksen M., Lebreton L. C. M., Carson H. S., Thiel M., Moore C. J., Borerro J. C. et al. (2014, 12). Plastic pollution in the world’s oceans: More than 5 trillion plastic pieces weighing over 250,000 tons afloat at sea. PLOS ONE, vol. 9, no 12, p. 1-15. Ghamisi P., Rasti B., Yokoya N., Wang Q., Hofle B., Bruzzone L. et al. (2018). Multisource and multitemporal data fusion in remote sensing. Shen Y.-T., Lai J.-W., Leu L.-G., Lu Y.-C., Chen J.-M., Shao H.-J. et al. (2019). Applications of ocean currents data from high-frequency radars and current profilers to search and rescue missions around taiwan. Journal of Operational Oceanography, vol. 12, no sup2, p. S126– S136. 2. Méthodologie générale pour le suivi et la collecte de macro-déchets plastiques en mer A partir des données de positions d’amas, pouvoir visualiser ceux-ci et leur déplacement est un élément clef pour pouvoir les collecter. Entre le moment d’observation de ces déchets et le moment de leur collecte, les amas auront été déplacés. Il est donc nécessaire de pouvoir modéliser et prédire le déplacement de ces déchets comme c’est par exemple le cas pour des applications de secours en mer (Shen et al., 2019). Un modèle de déplacement probabiliste permettant de prendre en compte le comportement chaotique et changeant des éléments en Posters 296 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Bibliographie 297 Posters 297 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Magalie Técher1, Hassan Ait Haddou2, Rahim Aguejdad3,4 1. Laboratoire LIFAM, École Nationale Supérieure d’Architecture de Montpellier 179, Rue de l’Esperou, 34093 Montpellier, France magalie.techer.au@gmail.com 1. Laboratoire LIFAM, École Nationale Supérieure d’Architecture de Montpellier 179, Rue de l’Esperou, 34093 Montpellier, France magalie.techer.au@gmail.com 2. Laboratoire LIFAM, École Nationale Supérieure d’Architecture de Montpellier 179, Rue de l’Esperou, 34093 Montpellier, France haithaddou@montpellier.archi.fr 3.USR 3330, UMIFRE 21 CNRS-MAE, IFP, 11 Saint-Louis street, Pondichery – 605 001, Inde 4.UMR TETIS, CNRS, 500 rue Jean-François Breton, 34090 Montpellier, France rahim.aguejdad@cnrs.fr RESUME. L’accroissement de la population dans les villes conduit à une expansion urbaine importante et à une pression sur les milieux urbains. Cette expansion s’accompagne d’une augmentation des températures urbaines plus connu sous le nom d’îlot de chaleur urbain et exacerbée par les effets du réchauffement climatique global. Dès lors, il devient primordial d’apporter des pistes de réflexion pour lutter contre ce phénomène d’ICU et permettre un développement contrôlé de cette expansion urbaine. À À travers une étude menée sur la métropole de Montpellier Méditerranée, ce travail présente une analyse des effets de la planification urbaine, à travers le règlement du plan local d’urbanisme (PLU) français, sur l’îlot de chaleur urbain. Dans un premier temps, grâce à l’intégration des différentes règles du PLU au sein d’une base de données SIG, des cartes de zonage sont générées, et intègrent pour chaque zone, l’ensemble des règles s’y appliquant. Dans un second temps, grâce aux données satellitaires du landsat-8, des cartes de températures du sol (LST) sont générées pour quatre dates en 2019. Enfin, par un outil de géotraitement, nous obtenons par zonage, une température du sol moyenne, ce qui permet de mettre en relation, une règle du PLU et ses effets sur les températures du sol. Les résultats trouvés sont d’une part, conformes à la littérature scientifique et fournissent d’autre part, des leviers d’action pour les décideurs politiques, urbanistes et maîtres d’ouvrage urbain, dans le cadre de l’élaboration de documents d’urbanisme français sur des moyens d’adaptation des territoires aux problématiques d’ICU et de changement climatique. 298 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 ABSTRACT. Population growth in cities leads to significant urban expansion and pressure on urban areas. Along with this expansion, is an increase in urban temperatures better known as the Urban Heat Island (UHI) and exacerbated by the effects of global warming. Magalie Técher1, Hassan Ait Haddou2, Rahim Aguejdad3,4 Therefore, it becomes essential to provide avenues for reflection to fight against this phenomenon of UHI and allow a controlled development of this urban expansion. Through a study carried out on the metropole of Montpellier Méditerranée, this works aims to assess the effects of urban planning policies, through the regulation of French local urban plan (PLU), on urban heat island. First, thanks to the integration of the various PLU rules into a GIS database, zoning maps are generated, and include for each zone all the rules that apply to it. Second, thanks to landsat-8 satellite data, land surface temperature (LST) maps are generated for four dates in 2019. Finally, using a geoprocessing tool, we obtain, by zoning, an average land surface temperature, which shows how PLU rules interact with land surface temperatures. First, thanks to the integration of the various PLU rules into a GIS database, zoning maps are generated, and include for each zone all the rules that apply to it. Second, thanks to landsat-8 satellite data, land surface temperature (LST) maps are generated for four dates in 2019. Finally, using a geoprocessing tool, we obtain, by zoning, an average land surface temperature, which shows how PLU rules interact with land surface temperatures. The results found are, on the one hand, consistent with the scientific literature and, on the other hand, provide levers of action for decision makers and urban planners, within the framework of the development of French urban planning documents on mean of adaptation of territories to problems of UHI and climate change. p f p The results found are, on the one hand, consistent with the scientific literature and, on the other hand, provide levers of action for decision makers and urban planners, within the framework of the development of French urban planning documents on mean of adaptation of territories to problems of UHI and climate change. MOTS-CLES : planification urbaine, forme urbaine, températures de surface de sol, îlot de chaleur urbain, France KEYWORDS: urban planning policies, urban form, land surface temperature, urban heat island, France Bibliographie non numérotée et références Bahi H., Mastouri H, et Radoine H. (2020). Review of methods for retrieving urban heat island. Materials today : proceedings, 2020, 6p. Benmecheta A. (2016) Estimation de la température de surface à partir de l’imagerie satellitale; validation sur une zone côtière d’Algérie. Thèse de doctorat en Géographie, Université Paris-Est Créteil Val de Marne, 118p. Chen H.C., Qi H., et De Vries B. (2020). Modeling the spatial relation between urban morphology, land surface temperature and urban energy demand. Substainable Cities and Society, 35 p. Chen Y., Wu J., Yu K., et Wang D. (2020) Evaluating the impact of the building density and height on the block surface temperature. Building and environment, no 168, 14p. Jusuf, S.K., Ignatius M., Hien W.N., et Akbari H. (2019). Editorial : urban heat island (UHI) and its mitigation through urban planning, design, and landscaping. Architectural science review, vol. 1, no 62, 2 p. 299 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Analyse des effets du règlement d’urbanisme sur les températures de surface du sol Analyse des effets du règlement d’urbanisme sur les températures de surface du sol Melot, R, et Bransiecq M. (2016). Règles d’urbanisme et choix politique : les observations de l’état sur les projets locaux. Revue d’économie régionale et urbaine, Armand Colin, no 4, p 767-98. Melot, R, et Bransiecq M. (2016). Règles d’urbanisme et choix politique : les observations de l’état sur les projets locaux. Revue d’économie régionale et urbaine, Armand Colin, no 4, p 767-98. Mehebub S., Raihan A., et Sajjad H. (2016). Analyzing land surface temperature distribution in response to land use/land cover change using split window algorithm and spectral radiance model in suburban biosphere reserve, India. Modeling Earth Systems and Environment, vol. 2, no 81, 12 p. Oguz H. (2013). LST Calculator : a program for retrieving land surface temperature from landsat TM/ETM + Imagery. Environmental Engineering and Management Journal, vol 12, no 3, p 549-55. Prévost A. (2013). Inférence(s) des documents d’urbanisme sur le territoire: application à la ville de Toulouse : modélisation multicritère et évaluation durable. Thèse de doctorat en Architecture et Aménagement de l’Espace, Université de Technologie de Compiègne, 398p. Salvati A., Palme M., et De la Barrera F. (2018). Urban morphology parametrization for climate modelling in urban planning. 10th International Conference on Urban Climate / 14th Symposium on the urban environment, New York, USA. Bibliographie non numérotée et références Yiannakou A., et Salata K.D.., (2017). Adaptation to climate change through spatial planning in compact urban areas: a case study in the city of Thessaloniki. Sustainability, vol. 9, no 271, 18 p. 300 Posters Posters 300 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 L’état des piscines, construction d’un indicateur de la reprise du tourisme en situation post-catastrophe. L’île de Saint- Martin après le passage du cyclone Irma (sept 2017) Thommeret N.1, Grancher D. 2, Dousseron A.2, Brunstein D.3, Virmoux C.2 Thommeret N.1, Grancher D. 2, Dousseron A.2, Brunstein D.3, Virmoux C.2 2. Laboratoire de Géographie Physique (CNRS – UMR 8591), 1 place Aristide Briand, F-92195 Meudon Cedex, France 3. UMR Lieux, Identités, Espaces et Activités (UMR 6240), Université de Corse Pascal Paoli 3. UMR Lieux, Identités, Espaces et Activités (UMR 6240), Université de Corse Pascal Paoli MOTS-CLES : CYCLONE, TELEDETECTION, PISCINES, TOURISME, NDWI Posters 301 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Les images à très haute résolution spatiale telles que celles des satellites Pléiades permettent de suivre l’évolution du territoire à des échelles spatiales et temporelles très fines. Dans le cadre de la gestion des catastrophes naturelles, la chartre internationale des risques majeurs permet d’obtenir des images pendant l’événement ou juste après l’événement pour évaluer les dégâts et diriger les secours et aides humanitaires. Pour le suivi à plus long terme, on peut avoir accès des images régulières pour le suivi de la reconstruction du territoire ou de la reprise de l’activité économique et sociale. Ces images sont notamment utiles pour caractériser la résilience du territoire et proposer une analyse spatialisée permettant de différencier les quartiers. En 2017, suite au passage du cyclone Irma (5-6 septembre) sur l’ile de Saint Martin, les études d’endommagements et de destructions ont estimé que 95% des bâtiments et infrastructures ont été touchés (Desarthe et al., 2020). Dans le cadre de l’ANR TIREX, plusieurs indicateurs sont explorés pour le suivi de la reconstruction territoriale (Duvat et al., 2021). Parmi les indicateurs spatialisés qui peuvent être considérés, nous nous intéressons ici au suivi de l’état de l’eau des piscines des propriétés sur la partie française de l’île. En effet, sur un territoire tel que Saint Martin dont l’économie repose essentiellement sur le tourisme, une grande partie des logements (individuels ou collectifs – hôtels, résidences) disposent de piscine. En étudiant la réponse spectrale de la surface d’un même bassin sur des images prises à différentes dates successives, on peut identifier la date de remise en service de la piscine. Cette dernière caractérise au minimum une intervention dans la propriété pour redémarrer le moteur ou une intervention plus importante de nettoyage du bassin et de restauration de l’équipement, correspondant à un usage de la propriété. On part de l’hypothèse que la piscine est remise en état lorsque la maison elle-même est de nouveau habitable. Thommeret N.1, Grancher D. 2, Dousseron A.2, Brunstein D.3, Virmoux C.2 Par contre dans le cas d’une piscine mutualisée entre plusieurs logements il est possible qu’il reste des logements non habitables ou non habités et suivant le coût de la réparation par exemple certaines piscines peuvent être laissés en l’état (i.e. non fonctionnelles) alors que la maison est remise en location ou ré- occupée. De manière générale, on peut considérer que la chronologie correspondant à la remise en eau propre des piscines est un bon indicateur pour le suivi de la reconstruction et le retour de l’activité touristique après l'événement majeur qu’est le cyclone Irma. La méthode proposée pour établir cette chronologie repose sur : 1) la localisation des piscines pour un état de référence avant cyclone ; 2) l’identification des piscines fonctionnelles à plusieurs dates après cyclone à partir des images Pléiades. Pour localiser les piscines avant cyclone, nous avons complété par photo- interprétation les données OpenStreetMap (OSM), elles-mêmes issues du cadastre français (Direction Générale des Impôts - Cadastre, 2015). Toutefois, du fait d’un décalage en planimétrie entre les images, la position des piscines varie jusqu’à plusieurs mètres localement, empêchant la superposition des polygones. Nous avons ainsi fait le choix de travailler à l’intérieur d’une zone tampon autour des centroides des polygones OSM. Le rayon de la zone tampon a été fixé à partir des distances minimales inter-piscines permettant de travailler sur un voisinage avec une superposition minimale entre les zones des différentes piscines. Posters 302 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Pour évaluer l’état de l’eau de chacune des piscines, nous avons utilisé l’indice NDWI (Normalized Difference Water Index ; McFeeters, 2013 ; Kim et al., 2011) qui repose sur la différence de réflectance des surfaces dans les longueurs d’onde du vert et du proche-infrarouge (équation 1). 𝑁𝑁𝑁𝑁𝑉𝑉𝑉𝑉𝑃𝑃𝑃𝑉𝑉𝑉𝑉𝑃𝑃𝑃 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁= (𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 − 𝑃𝑃𝑃𝑃𝑃𝑃) (𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 + 𝑃𝑃𝑃𝑃𝑃𝑃) (1) (1) Cet indice nous permet de repérer les piscines fonctionnelles (i.e. en eau claire avec un fond réfléchissant ; figure 1). L’absence de pixel « eau claire » dans la zone tampon est associée à une piscine en eau sale ou vide. Une analyse par arbre de classification (CART) a permis d’une part de montrer que le NDWI était pertinent pour identifier les zones d’eau claire et d’autre part de déterminer un seuil objectif et valable sur les images aux différentes dates. McFeeters, S. K. (2013). Using the normalized difference water index (NDWI) within a geographic information system to detect swimming pools for mosquito abatement: A practical approach, Remote Sensing, 5, 3544-3561. Kim M., Holt J., Eisen R., Padgett K., Reisen W., and Croft J. (2011), Detection of swimming pools by geographic object-based image analysis to support west nile virus control efforts, Photogrammetric Engineering and Remote Sensing, 77, 1169–1179. Thommeret N.1, Grancher D. 2, Dousseron A.2, Brunstein D.3, Virmoux C.2 Les principales raisons de mauvaise classification sont liées d’une part aux images elles-mêmes (présence de nuages ou de zones d’ombre) et d’autre part à un effet de taille (moins bonne détection des piscines < 25m²). FIGURE 1. Piscines ensablées après le cyclone Irma sur images Pléiades et NDWI FIGURE 1. Piscines ensablées après le cyclone Irma sur images Pléiades et NDWI L’approche par le NDWI nous a ainsi permis d’obtenir une première cartographie des piscines fonctionnelles pour neuf dates de 2017 à 2020 et d’établir une chronologie par quartier de remise en fonction des piscines sur la partie française de Saint Martin. Les résultats vont maintenant être confrontés à d’autres indicateurs de la dynamique touristique permettant une analyse plus globale de la reprise sur ce territoire depuis trois ans. Desarthe J., Bauduceau N., Quantin A. (2020). Les défis soulevés par la reconstruction post- Irma, Annales des Mines - Responsabilité et environnement, 36-40. Duvat V., Volto N., Stahl L., Moatty A., Defossez S., Desarthe J., Grancher D., Pillet V. (2021), Understanding interlinkages between long-term trajectory of exposure and vulnerability, path dependency and cascading impacts of disasters in Saint-Martin (Caribbean), Global Environmental Change, 67, 102236. Kim M., Holt J., Eisen R., Padgett K., Reisen W., and Croft J. (2011), Detection of swimming pools by geographic object-based image analysis to support west nile virus control efforts, Photogrammetric Engineering and Remote Sensing, 77, 1169–1179. McFeeters, S. K. (2013). Using the normalized difference water index (NDWI) within a geographic information system to detect swimming pools for mosquito abatement: A practical approach, Remote Sensing, 5, 3544-3561. 303 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Sensibilisation des gestionnaires à la prévention des risques liés à l’eau : modélisation, simulation et participation dans le Delta du Mékong Thi Huyen Tran1, Marion Amalric1, Frédéric Rousseaux2, Benoit Gaudou3 Thi Huyen Tran1, Marion Amalric1, Frédéric Rousseaux2, Benoit Gaudou3 1. UMR CITERES CNRS, Université de Tours, 35 Allée Ferdinand de Lesseps, 37200 Tours, France huyenrosietran@gmail.com, marion.amalric@univ-tours.fr 2. UMR LIENSs CNRS, Université de La Rochelle, 2, rue Olympe de Gouges, 17000 La Rochelle – France frederic.rousseaux@univ-lr.fr 2. UMR LIENSs CNRS, Université de La Rochelle, 2, rue Olympe de Gouges, 17000 La Rochelle – France frederic.rousseaux@univ-lr.fr 3. IRIT, Université de Toulouse, Cours Rose Dieng-Kuntz, 31400 Toulouse – France benoit.gaudou@gmail.com MOTS-CLES : SIMULATION PARTICIPATIVE, GESTION DES RISQUES, APPRENTISSAGE SOCIAL, DELTA DU MEKONG. 304 Solinska-Nowak, A. (2018). An overview of serious games for disaster risk management – Prospects and limitations for informing actions to arrest increasing risk. International Journal of Disaster Risk Reduction, 17. Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Notre étude s’inscrit dans le thème de la simulation participative. Elle présente un jeu sérieux, qui a été réalisé en France « LittoSIM », et qui est en cours d’expérimentation et d’adaptation dans le Delta du Mékong au Vietnam. Cette simulation intitulée « LittoKONG » se base sur une plateforme participative et un système multi-agents. L’objectif principal de notre étude est de sensibiliser les autorités locales à l’adaptation au changement climatique. Plus précisément, il s’agit de favoriser l'apprentissage social des mesures de prévention des risques liés à l’eau avec les autorités et les gestionnaires locaux. En raison du changement climatique, il est probable que l’intensité des submersions marines augmentent sensiblement, tel que l’élévation du niveau de la mer et les événements météo-marins intenses, voire extrêmes (IPCC, 2014). Il est donc essentiel de questionner et adapter la gouvernance des littoraux urbanisés afin de limiter les conséquences liées à la vulnérabilité des sociétés concernées. La démarche collaborative entre les disciplines et les technologies diverses (environnement, société, économie, urbanisation, gouvernance, etc.) est donc indispensable pour la gestion durable du littoral (Lacroix et al., 2019). Dans cette perspective, le jeu sérieux de simulation participative aide les acteurs concernés (communautés affectées, décideurs, parties prenantes) de poursuivre une démarche d’apprentissage et de sensibilisation dans la gestion des risques (Marion Amalric et al., 2017; Solinska-Nowak, 2018). « LittoKONG » est constitué de différents scénarios face à la submersion dans le contexte mondial du changement climatique. A travers les ateliers du jeu, ce dispositif permet aux acteurs locaux de coopérer, « tester » et envisager les solutions de prévention et de gestion des risques littoraux à l’aide d’un ensemble d’actions d’aménagement et une représentation de leur territoire. L'étude présentée dans ce poster a permis de répondre à la question de faisabilité de LittoKONG vis-à-vis du système juridique au Vietnam. L’adaptation du modèle conceptuel initial du jeu sérieux dans un nouveau contexte et les différents défis de traitement des données spatiales sont également abordés. Ce travail de recherche a reçu le soutien financier de l’Ambassade de France au Vietnam, au titre d’une bourse d’excellence de thèse et par du projet CNRS-IRD RISQUES 2019 pour le financement de la mission. 305 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Modélisation, simulation et participation à la prévention des risques liés à l’eau Références bibliographiques IPCC. (2014). Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Intergovernmental Panel on Climate Change. Lacroix, D., Mora, O., De Menthiere, N., & Bethinger, A. (2019). La montée du niveau de la mer: Conséquences et anticipations d’ici 2100, l’éclairage de la prospective. Rapport d'étude, Alliance nationale de recherche pour l'Environnement, 172. Marion Amalric, Anselme, B., Bécu, N., Etienne Delay, Marilleau, N., Pignon, C., & Rousseaux, F. (2017). Sensibiliser au risque de submersion marine par le jeu ou faut-il qu’un jeu soit spatialement réaliste pour être efficace ? Sciences du jeu, 8. Marion Amalric, Anselme, B., Bécu, N., Etienne Delay, Marilleau, N., Pignon, C., & Rousseaux, F. (2017). Sensibiliser au risque de submersion marine par le jeu ou faut-il qu’un jeu soit spatialement réaliste pour être efficace ? Sciences du jeu, 8. Solinska-Nowak, A. (2018). An overview of serious games for disaster risk management – Prospects and limitations for informing actions to arrest increasing risk. International Journal of Disaster Risk Reduction, 17. 306 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Quy Thy Truong1, Anne Ruas2, Serge Lhomme3 Quy Thy Truong1, Anne Ruas2, Serge Lhomme3 Quy Thy Truong1, Anne Ruas2, Serge Lhomme3 1. Université Gustave Eiffel, COSYS/LISIS truong.quythy@gmail.com 2. Université Gustave Eiffel, COSYS/LISIS anne.ruas@univ-eiffel .fr 3. Université Paris Est Créteil, Lab’URBA serge.lhomme@u-pec.fr MOTS-CLES : COLLECTE DE DÉCHETS POST-OURAGANS, SIMULATION, MÉTHODE DE POSITIONNEMENT 2. Université Gustave Eiffel, COSYS/LISIS anne.ruas@univ-eiffel .fr 2. Université Gustave Eiffel, COSYS/LISIS anne.ruas@univ-eiffel .fr 3. Université Paris Est Créteil, Lab’URBA serge.lhomme@u-pec.fr 3. Université Paris Est Créteil, Lab’URBA serge.lhomme@u-pec.fr 3. Université Paris Est Créteil, Lab’URBA serge.lhomme@u-pec.fr MOTS-CLES : COLLECTE DE DÉCHETS POST-OURAGANS, SIMULATION, MÉTHODE DE POSITIONNEMENT 307 Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Les ouragans génèrent d’importants volumes de déchets qu’il faut évacuer d’urgence pour permettre au territoire sinistré de se reconstruire. Pour mieux anticiper les futures catastrophes naturelles de ce genre, le projet ANR DéPOs vise à étudier le processus de collecte des déchets post-ouragans à partir d’événements passés et à proposer des pistes d’amélioration à partir de simulations de collecte de déchets post-ouragans en se concentrant sur les Antilles Françaises. Le travail présenté dans ce poster est centré sur l’algorithme de collecte et est appliqué à Saint- Martin à partir de données réelles après le passage de l’ouragan Irma en septembre 2017. Des campagnes de télédétection post-ouragans sur Saint-Martin réalisées par le SERTIT permettent en effet de disposer de l’inventaire des bâtiments détruits et endommagés très utile pour la mise au point des méthodes proposées. Une fois finalisé, l’algorithme peut être appliqué sur un autre territoire et pour un ouragan simulé. L’objectif de la recherche est d’estimer au mieux les temps de collecte selon les scénarios, les aménagements et les moyens de collecte (acteurs, nombre et types de camions) afin d’aider à minimiser ce temps en situation réelle grâce à une planification éclairée. Le processus de collecte des déchets post-ouragans consiste en un transfert successif des déchets de leur source vers différentes zones de dépôt temporaire, puis de ces zones de dépôts temporaire vers les zones de traitement ou d’enfouissement des déchets. D’après les témoignages recueillis après Irma, les habitants de Saint- Martin rassemblent les déchets post-ouragans devant ou à proximité des bâtiments détruits ou endommagés sur des lieux appelés des aires de dépose (Beraud et al., 2019). Quy Thy Truong1, Anne Ruas2, Serge Lhomme3 Ces déchets sont alors ramassés par des entrepreneurs vers des installations temporaires de transit, avant d’être à nouveau déplacés vers des centres de stockage ou de traitement de déchets. Le processus de collecte est donc composé de deux étapes. Une modélisation a été proposée pour décrire ce processus (Ruas et Lhomme, 2019). Cette modélisation permet de distinguer différents types de déchets mais nous ne traitons ici qu’un type de déchet, ce qui était le cas pour Saint Martin. Nous présentons dans ce poster de la première étape qui s’implémente comme un algorithme de déplacement de stock de déchets entre des origines (proche des habitations endommagées) et des destinations (les installations temporaires de transit). Pour réaliser ce processus, nous proposons des méthodes d’identification des origines, ce qui est une vraie difficulté car ces lieux sont éphémères et nous ne disposons pas d’information pour les localiser puisque ces aires sont informelles donc non planifiées et disparaissent lorsque les déchets ont été collectés. Dans notre recherche, nous proposons plusieurs méthodes de localisation de ces aires de dépose et comparons l’impact de ces choix sur les temps de collecte calculés. L’algorithme proposé est un algorithme de plus court chemin classique calculé sur le réseau routier entre des origines à proximité des bâtiments détruits et endommagés et des zones de transit temporaires, puis des zones de transit aux exutoires, ici la décharge de Saint Martin. Les bâtiments détruits et endommagés ont été identifiés grâce à la campagne de télédétection réalisée par le SERTIT. Le calcul du volume a été estimé grâce à des photos sur place et des connaissances de volumes total en fin de chaîne à l’unique décharge de l’île (Lamour, 2019). Le temps de Posters 308 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] Modélisation spatiale et positionnement de zones de dépôt... Modélisation spatiale et positionnement de zones de dépôt... collecte est la somme du chargement, du déplacement et du déchargement des déchets avec une flotte de camions ayant une certaine capacité de chargement, jusqu’à livraison des déchets dans la zone de transit. Dans notre recherche quatre modes de calcul des origines sont proposées : à la porte des bâtiments, à l’intersection la plus proche, projeté sur la rue la plus proche avec regroupement de déchets en fonction des volumes et aux grandes intersections. Quy Thy Truong1, Anne Ruas2, Serge Lhomme3 La méthode basée sur les regroupements part du constat que les habitants regroupent les déchets sur une même aire, lorsque le volume de déchet n’est pas trop important et que la distance n’est pas trop grande afin de minimiser le nombre de tas de déchets dans les rues. Nous calculons des zones tampons autour des aires de dépose potentielles, avec un rayon variable en fonction du volume de déchets présents. Au niveau expérimental nous présentons ci-dessous les résultats des temps sur le quartier de Grand Case, avec 3 camions de 32m3 et avec des journées de travail de 8 heures (table 1). TABLE 1. Durées de collecte selon quatre scénarios Méthode de positionnement Nombre d’aller-retour Durée de la collecte en jour Au niveau des gisements de déchets 157 71 Intersections proches des gisements 70 65 Projection et regroupement des gisements 46 63.5 Intersections des routes principales 32 62 TABLE 1. Durées de collecte selon quatre scénarios L’analyse des données montre de façon non surprenante que la méthode de projection et regroupement des gisements proches permet de simuler une collecte plus rapide que celle du ‘porte à porte’. On sait aussi que cette méthode est plus réaliste que la méthode de positionnement aux intersections proches. Elle est cohérente avec les témoignages rapportés. Ce qui est intéressant c’est que le temps en jours varie relativement peu d’un scénario à l’autre : c’est en effet la durée de chargement et de déchargement qui est la plus pénalisante dans le processus : dans le scénario de collecte en porte à porte, la durée de collecte résulte à 97% du temps de chargement/déchargement et 3% du temps de transport des déchets. C’est donc bien sur ce point qu’il faut chercher les améliorations pour réduire le temps de collecte. Ruas A., Lhomme S. (2019) L’analyse et l’amélioration de la gestion des déchets après les ouragans. Le cas des Antilles françaises. Cartes et Géomatique. N° 241-242, p 59-64. Bibliographie Beraud H., Nithart C., Durand M. (2019). Le difficile suivi des déchets postcatastrophe : le cas de l’Ouragan Irma à Saint-Martin. Risques urbains, vol. 3, no 1. Consulté sur https://www.openscience.fr/Le-difficile-suivi-des-dechets-post-catastrophe-le-cas-de-l- Ouragan-Irma-a Lamour M., (2019) Système d’information pour la gestion des déchets post-ouragan, Rapport de Master Télédétection et Géomatique appliqué à l’environnement, Université Paris & Gustave Eiffel, 2019 Ruas A., Lhomme S. (2019) L’analyse et l’amélioration de la gestion des déchets après les ouragans. Le cas des Antilles françaises. Cartes et Géomatique. N° 241-242, p 59-64. 309 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Véronique Venturini1, Caroline Tafani2 Véronique Venturini1, Caroline Tafani2 1. Doctorante en Géographie, Urbaniste UMR CNRS 6240 LISA Università di Corsica Pasquale Paoli Campus Mariani, Bâtiment Edmond Simeoni, Avenue Jean Nicoli, 20250 CORTE veronique.venturini@outlook.com 2. Maître de Conférence en Géographie UMR CNRS 6240 LISA Università di Corsica Pasquale Paoli Campus Mariani, Bâtiment Edmond Simeoni, Avenue Jean Nicoli, 20250 CORTE Tafani_c@univ-corse.fr MOTS-CLES : PLANIFICATION, LITTORAL, URBANISME, CONFIGURATIONS SPATIALES DU FONCIER, CORSE MOTS-CLES : PLANIFICATION, LITTORAL, URBANISME, CONFIGURATIONS SPATIALES DU FONCIER, CORSE 310 [ISBN : 978-2-910545-12-1] Planifier en Corse : une problématique incontournable ? La planification littorale spatialisée, en tant que traduction règlementaire des évolutions du territoire (Demazière, Desjardins 2016), apparait comme un sujet complexe en Corse. Sa conception, qui se définit entre la précision et l’adaptation des normes nationales, semble rencontrer des problématiques de mise en œuvre à l’échelle locale. En effet, la gouvernance territoriale, qui s’était heurtée à plusieurs reprises aux difficultés d’élaborer le Plan d’Aménagement et de Développement Durable de la Corse (PADDUC), peine aujourd’hui à le rendre opérationnel. Dans ce contexte, la Balagne ne fait pas figure d’exception. Cette microrégion située au Nord-Ouest de l’île, apparait comme un territoire attractif concentrant les enjeux en matière de développement territorial (Tafani 2012). Cependant, on ne retrouve pas de véritable dynamique de mise en œuvre de la planification stratégique spatialisée. Sur 36 communes, on ne recense que deux documents compatibles avec le PADDUC. Face à ce constat, nous soulevons l’idée que les méthodes d’analyse des territoires, et notamment celles liées à l’identification des espaces non bâtis, ne semblent pas tenir compte de la situation foncière (DDTM 2017 ; Orsini 2014). L’hypothèse soulevée soumet l’idée qu’une analyse foncière plus précise, intégrée à la démarche planificatrice, pourrait permettre d’améliorer son opérationnalité à travers une évaluation plus approfondie des capacités foncières disponibles dans les espaces non bâtis. Approfondir les méthodes d’analyse des espaces bâtis et appréhender la thématique foncière dans la démarche planificatrice En dressant un état des lieux du terrain au regard des méthodes d’analyse des territoires inhérentes à la démarche d’élaboration des documents d’urbanisme nous avons pu constater que la pratique ne tient pas compte de la situation foncière des interstices non bâtis et notamment des configurations spatiales, au sens « d’agencement et de structuration » (Pumain 2004) du foncier. Identifier les configurations spatiales du foncier permet de caractériser ce dernier au regard de sa situation géographique, de son agencement, de son accessibilité, sa structure et enfin de ses usages et qualités (Marraccini et al. 2017). L’ensemble de ces éléments apparaissent comme des indicateurs spatialisés qui vont permettre d’identifier un stock de foncier libre réellement exploitable en traitant sa dimension relationnelle (droits et usages), son accessibilité et les fonctions attribuées (urbanisation, coupure urbaine…). Cette approche, appliquée à l’échelle des interstices non bâtis et intégrée aux méthodes d’analyse, doit permettre de faciliter la démarche planificatrice. Posters 310 Posters [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Révéler les potentialités de développement des territoires littoraux L’analyse des configurations spatiales du foncier, intégrée aux méthodes d’analyse des territoires apparait comme un indicateur pertinent pouvant améliorer l’opérationnalité de la planification. Ce concept, présenté sous la forme d’indicateurs visant à identifier le foncier réellement valorisable, peut permettre ainsi d’éclairer les choix urbanistiques à réaliser. Par conséquent cette proposition méthodologique fondée sur le caractère multidimensionnel du foncier et son niveau de mutabilité se place également comme une première étape d’action à mener dans le cadre de politiques foncières. [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] La planification littorale spatialisée de Balagne à l’épreuve de son opérationnalité La planification littorale spatialisée de Balagne à l’épreuve de son opérationnalité plusieurs effets pour la pratique. D’une part, elle doit permettre de résoudre certaines approximations liées aux ratios démographie/logements pouvant proposer des objectifs de développement parfois inatteignables. D’autre part, cette méthode tient compte des phénomènes de tensions foncières pouvant se traduire par une forme de bicéphalisme entre les noyaux villageois originels situés dans l’arrière-pays avec les formes d’urbanisation littorales récentes. Dès lors, identifier le potentiel foncier au regard de ses configurations spatiales permettrait de garantir une meilleure prise en compte de l’organisation des formes urbaines. Dans le cas du terrain étudié, l’intégration des configurations spatiales du foncier peut exprimer un différentiel de l’ordre de 3 à 4 hectares selon les types de formes urbaines et leur localisation. Les configurations spatiales du foncier, un indicateur pertinent Cette méthode d’analyse des configurations spatiales du foncier doit avoir 311 Posters 311 SAGEO’2021 – La Rochelle, 5-7 mai 2021 Bibliographie non numérotée et références Demazière C., et Desjardins X. (2016). La planification territoriale stratégique : une illusion nécessaire ? Revue internationale d’urbanisme. N° 2. Direction Départementale des Territoires et de la Mer. (2017). La pression foncière en Corse. Rapport d’étude. Prefecture de Corse. Direction Départementale des Territoires et de la Mer. (2017). La pression foncière en Corse. Rapport d’étude. Prefecture de Corse. Marraccini E., Lardon S., et Sabbatini T. (2017). Les configurations spatiales agri-urbaines - Pour intégrer l’agriculture dans l’aménagement des territoires urbains. Spatial Analysis and GEOmatics. Novembre 2017, Rouen. Orsini L. (2014). Les aspects civils et fiscaux de l’indivision en Corse. Droit et Ville, Vol. 77, n° 1, pp. 73‑99. Pumain D. (2004). Les configurations. Hypergeo. 2004. Tafani C. (2012). Dynamiques du foncier agricole et développement territorial en Balagne (Haute-Corse). Conférence inaugurale de l’OHM Littoral méditerranée. Septembre 2012, Marseille. 312 Posters [ISBN : 978-2-910545-12-1] SAGEO’2021 – La Rochelle, 5-7 mai 2021 Narimane ZAABAR1, 2, Simona NICULESCU2, Mustapha Kamel MIHOUBI1, Soumia BENGOUFA1, 2, 3 Après des prétraitements des images S1, une fusion avec les bandes de S2 et des couches d’indices spectraux a été effectuée. Cinq classes d’occupation du sol ont été définies : (1) eau, (2) sols nus, (3) terres agricoles, (4) bâti, (5) végétation. Le choix de ces classes a pris en considération les enjeux situés dans la zone exposée au risque d’inondation. Pour l’approche orientée objet, l’algorithme de Multirésolution (MRS) a été appliqué pour la segmentation de l’image fusionnée. Pour l’approche pixel par pixel, les échantillons créés ont permis d'entraîner directement le classificateur. Tandis que, pour la deuxième approche, des objets ont été extraits à partir de la segmentation, et utilisés en tant qu’échantillons d'entraînement. Afin d’obtenir de meilleures classifications, un processus d’optimisation a été effectué pour choisir les meilleures valeurs des paramètres de segmentation et des algorithmes RF et SVM. La performance des résultats a été évaluée par le calcul d’une matrice de confusion, un indice Kappa, et un indice de précision globale. acquises durant l’année 2019 a été utilisée. Le choix de l’année est justifié par les évènements des inondations que la région d’étude a connu. Après des prétraitements des images S1, une fusion avec les bandes de S2 et des couches d’indices spectraux a été effectuée. Cinq classes d’occupation du sol ont été définies : (1) eau, (2) sols nus, (3) terres agricoles, (4) bâti, (5) végétation. Le choix de ces classes a pris en considération les enjeux situés dans la zone exposée au risque d’inondation. Pour l’approche orientée objet, l’algorithme de Multirésolution (MRS) a été appliqué pour la segmentation de l’image fusionnée. Pour l’approche pixel par pixel, les échantillons créés ont permis d'entraîner directement le classificateur. Tandis que, pour la deuxième approche, des objets ont été extraits à partir de la segmentation, et utilisés en tant qu’échantillons d'entraînement. Afin d’obtenir de meilleures classifications, un processus d’optimisation a été effectué pour choisir les meilleures valeurs des paramètres de segmentation et des algorithmes RF et SVM. La performance des résultats a été évaluée par le calcul d’une matrice de confusion, un indice Kappa, et un indice de précision globale. Les deux approches de classification ont permis de détecter 5 classes de type d’occupation du sol ce qui correspond au niveau 2 d’une nomenclature adaptée au contexte d’inondation selon Maurel et al (2003) (Figure 1). Narimane ZAABAR1, 2, Simona NICULESCU2, Mustapha Kamel MIHOUBI1, Soumia BENGOUFA1, 2, 3 1. Ecole Nationale Supérieure de l’Hydraulique (ENSH), Laboratoire de Mobilisation et de Valorisation des Ressources en Eau (MVRE), 29 route de Soumaa, BLIDA, ALGERIE, n.zaabar@ensh.dz 2. UBO, Université de Bretagne Occidentale, Laboratoire LETG Brest UMR 6554 CNRS, Brest, France 2. UBO, Université de Bretagne Occidentale, Laboratoire LETG Brest UMR 6554 CNRS, Brest, France 2. UBO, Université de Bretagne Occidentale, Laboratoire LETG Brest UMR 6554 CNRS, Brest, France 3. Ecole Nationale Supérieure des Sciences de la Mer et de l’Aménagement du Littoral (ENSSMAL), Dely Ibrahim Bois des Cars, Alger, Algérie 3. Ecole Nationale Supérieure des Sciences de la Mer et de l’Aménagement du Littoral (ENSSMAL), Dely Ibrahim Bois des Cars, Alger, Algérie MOTS-CLES : Inondation, Occupation Du Sol, Ain Témouchent, Classification, Sentinel-1(S1), Sentinel-2(S2), Random Forest (RF), Support Vecteur Machine (SVM) Dans le contexte de l’étude du risques d’inondation, l’analyse de l’occupation du sol permet d’identifier les enjeux socio-économiques et quantifier les dommages provoqués (Pottier, 2005). En effet, l’évolution de la technologie spatiale en matière de capteurs satellitaires, ouvre des possibilités inédites pour l’identification, l’analyse et la surveillance de l’occupation du sol. Dans cette optique, les produits de l’imagerie satellitaire Sentinel-2 (S2) de haute résolution spatiale (HRS) offrent une analyse de haute précision. De même, les images radar Sentinel-1 (S1) à synthèse d’ouverture (SAR), permettent de s’affranchir du problème de la couverture nuageuse. Par conséquent, la fusion des images S1 et S2 offre une meilleure discrimination des éléments de l’occupation du sol, ce qui a été prouvé dans de nombreuses études (Niculescu et al., 2020 ; Mustak et al.,2019), notamment par le biais de approches de Machine Learning et de Deep Learning. En effet, l’objectif de cette étude est d’identifier des méthodes de machine Learning qui permettent de détecter le type d’occupation du sol dans le contexte de la caractérisation du risque d’inondation de la ville de Ain Témouchent (nord-ouest de l’Algérie). Deux approches de classification ont été testées : pixel par pixel et orientée objet, par le biais de deux algorithmes de Machine Learning Random Forest (RF) et Support Vector Machine (SVM). Pour cela, une série temporelle de 12 images S1 et S2, 313 Posters 313 SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] [ISBN : 978-2-910545-12-1] acquises durant l’année 2019 a été utilisée. Le choix de l’année est justifié par les évènements des inondations que la région d’étude a connu. Narimane ZAABAR1, 2, Simona NICULESCU2, Mustapha Kamel MIHOUBI1, Soumia BENGOUFA1, 2, 3 Les résultats obtenus par les deux algorithmes sont satisfaisants, néanmoins la performance de l’algorithme RF a surpassé celle de l’algorithme SVM. Ceci rejoint les constats d’autres études (ex. Htitiou, et al., 2019 ; Niculescu, et al., 2020 ; Inglada, et al., 2016). Selon les résultats, l’approche orientée objet révèle plus de pertinence, en atteignant les valeurs de 96% pour l’indice de précision globale (OA) et 0.95 pour l’indice Kappa (Table 1). Cette approche ainsi que l’avantage de libre droit des produits de HRS, S1 et S2, offrent la possibilité d’analyser des enjeux socio-économiques, dans le cadre de l’évaluation de la vulnérabilité face aux risques d'inondations. Comme perspective de cette étude, il est recommandé de tester ces méthodes sur d’autres types de données afin de détecter les classes du niveau supérieur (niveau 3), ce qui est requis pour l'élaboration des Plans de Prévention des Risques d’Inondation. TABLE 1 : Indices kappa et de précision globale (OA) de RF et SVM appliqués pour les deux approches ; pixel par pixel et orientée objet (OOB) TABLE 1 : Indices kappa et de précision globale (OA) de RF et SVM appliqués pour les deux approches ; pixel par pixel et orientée objet (OOB) OA(%) Kappa RF-pixel 92.3 0.90 SVM-pixel 91.0 0.89 RF-OOB 96.8 0.95 SVM-OOB 92.0 0.90 314 Posters SAGEO’2021 – La Rochelle, 5-7 mai 2021 [ISBN : 978-2-910545-12-1] Méthodes de Machine Learning pour la modélisation de l’occupation du sol Méthodes de Machine Learning pour la modélisation de l’occupation du sol FIGURE 1 : Résultats de la classification; A) RF-pixel B) SVM-pixel C) RF-OOB D)SVM-OOB Légende A B C D A A FIGURE 1 : Résultats de la classification; A) RF-pixel B) SVM-pixel C) RF-OOB D)SVM-OOB Pottier N., Penning-Rowsell E., Tunstall s., ET Hubert G. (2005). Land Use and Flood Protection: Contrasting Approaches and Outcomes in France and in England and Wales. Applied Geography 25 (1) : 1-27. 10.1016/j.apgeog.2004.11.003. Responsables de la publication : Paule-Annick Davoine, Didier Josselin, François Pinet Éditeur : UMR 7300 ESPACE / Avignon, Case 41, 74 rue L. Pasteur, 84029 Avignon Cedex, France Collection : Actes Avignon - ISBN : 978-2-910545-12-1, Mai 2021 Design et Mise en page : N. Brachet (UMR ESPACE – CNRS)t Bibliographie Maurel, P, Raclot D, Puech C (2003). Apports de la télédétection à très haute résolution spatiale à l’étude du risque d’inondation.https://hal.inrae.fr/hal-02582784 Maurel, P, Raclot D, Puech C (2003). Apports de la télédétection à très haute résolution spatiale à l’étude du risque d’inondation.https://hal.inrae.fr/hal-02582784 Maurel, P, Raclot D, Puech C (2003). Apports de la télédétection à très haute résolution spatiale à l’étude du risque d’inondation.https://hal.inrae.fr/hal-02582784 Niculescu, S.; Boissonnat, J.-B.; Lardeux, C.; Roberts, D.; Hanganu, J.; Billey, A.; Constantinescu, A.; Doroftei, M., Synergy of High-Resolution Radar and Optical Images Satellite for Identification and Mapping of Wetland Macrophytes on the Danube Delta. Remote Sensing 2020, 12, 2188, doi.org/10.3390/rs12142188. Pottier N., Penning-Rowsell E., Tunstall s., ET Hubert G. (2005). Land Use and Flood Protection: Contrasting Approaches and Outcomes in France and in England and Wales. Applied Geography 25 (1) : 1-27. 10.1016/j.apgeog.2004.11.003. 315 Posters Responsables de la publication :
https://openalex.org/W2792883393
https://dora.dmu.ac.uk/bitstream/2086/17333/1/Matrix_Factorization_Techniques_for_Context-Aware_.pdf
English
null
Matrix Factorization Techniques for Context-Aware Collaborative Filtering Recommender Systems: A Survey
Computer and information science
2,018
cc-by
6,326
Abstract Collaborative Filtering Recommender Systems predict user preferences for online information, products or services by learning from past user-item relationships. A predominant approach to Collaborative Filtering is Neighborhood-based, where a user-item preference rating is computed from ratings of similar items and/or users. This approach encounters data sparsity and scalability limitations as the volume of accessible information and the active users continue to grow leading to performance degradation, poor quality recommendations and inaccurate predictions. Despite these drawbacks, the problem of information overload has led to great interests in personalization techniques. The incorporation of context information and Matrix and Tensor Factorization techniques have proved to be a promising solution to some of these challenges. We conducted a focused review of literature in the areas of Context-aware Recommender Systems utilizing Matrix Factorization approaches. This survey paper presents a detailed literature review of Context-aware Recommender Systems and approaches to improving performance for large scale datasets and the impact of incorporating contextual information on the quality and accuracy of the recommendation. The results of this survey can be used as a basic reference for improving and optimizing existing Context-aware Collaborative Filtering based Recommender Systems. The main contribution of this paper is a survey of Matrix Factorization techniques for Context-aware Collaborative Filtering Recommender Systems. Keywords: collaborative filtering, context-aware, matrix factorization, performance improvement, accuracy of predictions, quality of recommendations Matrix Factorization Techniques for Context-Aware Collaborative Filtering Recommender Systems: A Survey Mohamed Hussein Abdi1, George Onyango Okeyo2 & Ronald Waweru Mwangi2 1 Department of Information & Communication Technology, Kenya Medical Research Institute, Kenya 2 Department of Computing, School of Computing & Information Technology, Jomo Kenyatta University of Agriculture & Technology, Kenya Correspondence: Mohamed Hussein Abdi Kenya Medical Research Institute PO Box 231-00610 Nairobi Correspondence: Mohamed Hussein Abdi, Kenya Medical Research Institute P.O. Box 231-00610 Nairobi, Kenya. Tel: 254-722-977-728 E-mail: mhabdi@gmail.com Received: January 15, 2018 doi:10.5539/cis.v11n2p1 Received: January 15, 2018 doi:10.5539/cis.v11n2p1 URL: http://dx.doi.org/10.5539/cis.v11n2p1 Computer and Information Science; Vol. 11, No. 2; 2018 ISSN 1913-8989 E-ISSN 1913-8997 Published by Canadian Center of Science and Education Computer and Information Science; Vol. 11, No. 2; 2018 ISSN 1913-8989 E-ISSN 1913-8997 Published by Canadian Center of Science and Education Computer and Information Science; Vol. 11, No. 2; 2018 ISSN 1913-8989 E-ISSN 1913-8997 Published by Canadian Center of Science and Education 1. Introduction Recommender Systems are a class of web applications that assist users to tame the problem of information overload by providing personalized recommendations on various types of information, products and services. The existing Recommender Systems applications can be classified into specific domains like entertainment, e-commerce, e-learning and service recommendations in travel, expert consultation and match making (Rajabpour, Bardsiri, Mohammadighavam, & Molaei, 2014). The various techniques used in generating recommendations suffers sparsity and scalability problem as the volume of accessible information and users continue to grow leading to accuracy and quality reduction. However, the popularity of the approach has drawn a great deal of research towards improving the prediction accuracy and the quality the recommendations. The interest in this area still remains high due to growing demand on practical applications, which are able to provide personalized recommendations and deal with information overload effectively (L. Sharma & Gera, 2013). This is due to the importance of personalization techniques which does not only aim to provide tailored information to customers based on their preferences, restrictions or tastes but also increase profits of commercial systems (Gao, Liu, & Wu, 2010). These growing demands pose some key challenges to Recommender Systems and to deal with these problems many advanced techniques like Content-boosted Collaborative Filtering (Melville, Mooney, & Nagarajan, 2002), Content-boosted Matrix Factorization (Nguyen & Zhu, 2013), and the incorporation of contextual information (Adomavicius & Mobasher, 2011; Adomavicius, Sankaranarayanan, Sen, & Tuzhilin, 2005) have been geared towards improving the accuracy of recommended predictions. Context-aware Recommender Systems (CARS) 1 1 Computer and Information Science Vol. 11, No. 2; 2018 cis.ccsenet.org approaches have been shown to provide more accurate predictions and relevant recommendations (Baltrunas, 2008; Karatzoglou, Amatriain, Baltrunas, & Oliver, 2010). Two major approaches are used in Collaborative Filtering: the Neighborhood-based (Breese, Heckerman, & Kadie, 1998) and the Matrix Factorization (Koren, Bell, & Volinsky, 2009). The Neighborhood approach relies on the preferences of the user’s neighbor, that is, other users with similar preferences to estimate the user preferences. Matrix Factorization in its basic form, characterizes both items and users by vectors of factors. These factors are inferred from patterns of item ratings (Koren & Bell, 2015). Matrix Factorization represents the relation between users and items through a set of latent factors also referred to as features. It forms two low rank matrices, each representing the relation between users (or items) and the set latent features. 1.2 Related Work This section briefly outlines the surveys concerning Recommender Systems. We classify the existing surveys into four categories. The first category deals with general introduction to Recommender Systems research. This includes the works of (Adomavicius & Tuzhilin, 2005; Bouraga, Jureta, Faulkner, & Herssens, 2014; Park, Choi, Kim, & Kim, 2011). The second category of the surveys provide methods; Context-aware systems (Baldauf, Dustdar, & Rosenberg, 2007), approaches and limitations (M. Sharma, 2013), Collaborative Filtering based on social networks (Yang, Guo, Liu, & Steck, 2013); Basic approaches in Recommender Systems (Felfernig et al., 2014). The third category is dedicated to the various applications of Recommender Systems. This include travel package recommendations (Patil & Kolhe, 2014), tourist guide (Umanets, Ferreira, & Leite, 2014), radio station hosting (Ignatov, Nikolenko, Abaev, & Konstantinova, 2014) and the fourth category cover the evaluation of the various Recommender Systems techniques (Hornung et al., 2013; Krohn-Grimberghe, Nanopoulos, & Schmidt-Thieme, 2010; Shinde & Potey, 2015). Most of the work on Collaborative Filtering has mainly focused on the traditional two dimensional user/item problem. Research on Recommender Systems algorithms gained prominence in 2006 with the launch of the Netflix $1Million prize contest to improve the state of movie recommendation. This competition led to much interest among many researchers. Koren et al. have demonstrated that Matrix Factorization models are superior to the classic nearest neighbor techniques (Koren et al., 2009). The extension of standard Matrix Factorization has incorporated context information such as Time-Aware Matrix Factorization (Liu, Cao, Zhao, & Yang, 2010), Context-aware Matrix Factorization (Baltrunas, Ludwig, & Ricci, 2011) and Contextual SLIM (Zheng, Mobasher, & Burke, 2014). The application of Tensor Factorization (Karatzoglou et al., 2010) and (Ricci, De Gemmis, & Semeraro, 2012). A detailed survey by (Frolov & Oseledets, 2016) provides Tensor Factorization methods. This survey will complement and extend the information on previous Context-aware Collaborative Filtering Recommender Systems surveys. In particular, build on the works of (Bokde, Girase, & Mukhopadhyay, 2015; Karydi & Margaritis, 2014; Ricci et al., 2012; Shi, Larson, & Hanjalic, 2014; Su & Khoshgoftaar, 2009). The survey by Su and Khoshgoftaar studied Collaborative Filtering techniques and provided a comprehensive coverage of the memory, model and hybrid based approaches outlining the merits, demerits and challenges. Karydi and Margaritis survey presents the use of parallel and distributed systems in the field of Recommender Systems. The survey provides a detailed background to Recommender Systems that employ parallel and/or distributed techniques. 1. Introduction The multiplication of these two matrices allows estimating users’ future preferences (Chertov, Brun, Boyer, & Aleksandrova, 2015). In this paper we explore the various Matrix Factorization approaches that have been studied in research aimed at improving performance for large scale datasets and the impact of incorporating contextual information. 1.1 Evaluation Metrics There is no common evaluation framework that can be applied to all Recommender Systems. However, a variety of measures exists to evaluate their properties. Three types of metrics to measure the quality of an algorithm are commonly in use: prediction, classification and rank accuracy (Herlocker, Konstan, Terveen, & Riedl, 2004). Prediction accuracy measures the difference between the rating the system predicts and the real rating. The most popular of this kind of metric is the Mean Absolute Error (MAE). Additional related metrics in use are the Mean Squared Error (MSE), Root Mean Squared Error (RMSE) or normalized Mean Absolute Error. The classification accuracy measures how well the system differentiates good items from bad ones. Examples of well-known metrics of this type are Precision, Recall and Receiver Operator Characteristic (ROC) and the area under the ROC curve (Cacheda, Carneiro, Fernández, & Formoso, 2011). The Rank accuracy measures the ability of the system to sort the recommended items like the user would have done. 3.1 Context-aware Collaborative Filtering The concept of Context-awareness in software applications systems has been conversed by research communities in many different application domains. Thus various definitions of context exists in all the disciplines in which it has been studied, including ubiquitous and pervasive computing, mobile computing, e-commerce, information retrieval, marketing and management as well as in several engineering disciplines. Therefore, the precise definition of what is context varies depending on the Recommender Systems application domain (Bazire & Brézillon, 2005). We adopt a widely accepted definition of context as any information that can be used to characterize the situation of an entity (Dey & Abowd, 2000). An entity is a person, place, or object that is considered relevant to the interaction between a user and an application, including the user and applications themselves. Context usually plays the role of additional information besides users, items and the ratings which may be relevant at the current time to make a recommendation. The aim of incorporating contextual information is to generate more relevant suggestions by adapting them to the user’s contextual situation. Context-aware systems sense and adapt their behavior based on the changing contexts (Coutaz, Crowley, Dobson, & Garlan, 2005), which generally consists of four basic components: Context Acquisition, Context Discovery, Context Model, and Context Processing (Baldauf et al., 2007). The traditional two dimensional Recommender Systems deal with two types of entities, users and items and try to estimate unknown ratings in the Users × Items matrix. In the past decade, a number of Recommender Systems algorithms and applications have been developed that incorporate contextual information into the recommendation algorithms. Sarkaleh et al. (Sarkaleh, Mahdavi, & Baniardalan, 2012) proposed a model which is able to recommend new locations to visitors in a museum while the visitor is given essential information about the certain features of the suggested site that puts into consideration the special needs of students, tourists and the ordinary people. The feature based personalization recommendation system categorized under knowledge level, instruments, ethnic language and tools considered the visitors use to acquire the contents of artistic works. In TripAdvisor (Wang, Chan, & Ngai, 2012), an application focusing on the attractions on offer, incorporated demographic data to investigate the applicability in recommender algorithms integrated with machine learning methods for the prediction of ratings of tourist attractions. Bokde et al. studied Matrix Factorization models and how they are used in Collaborative Filtering. The rest of the paper consists of four sections. The following section describes the methodology used in this research while section 3 outlines Context-aware Matrix Factorization techniques. The conclusion and some suggestions for future directions are given in the last section. 3.1 Context-aware Collaborative Filtering The role of emotions as context variable (Zheng, Mobasher, & Burke, 2013) evaluated two types of popular Context-aware recommendation algorithms – Context-aware splitting approaches and differential context modelling. The results showed that emotion-linked context makes an important contribution to Context-aware recommendations. The use of additional contextual information data such as weather, time, social media sentiment and user preferences can provide a more accurate model of the user’s current context resulting in improved recommendations (Meehan, Lunney, Curran, & McCaughey, 2013). 2. Methodology We conducted a focused literature review to identify research papers on Context-aware Matrix Factorization techniques focusing on the limitations, merits and challenges of existing Context-aware Collaborative Filtering Recommender Systems from 2007 to 2016. The following databases were researched: ACM Digital Library, IEEE Xplore digital library, Google Scholar and online Open Access Journals. The search terms for the Boolean search techniques were Recommender Systems, Collaborative Filtering, Matrix Factorization and their combinations with Context-aware. The duplicates were removed from the papers identified in the initial search. Titles and abstracts of the remaining papers were reviewed for relevance to the survey and those that did not meet our requirements were removed. Finally, a full text review was conducted for the remaining papers on CARS based on Matrix Factorization techniques. 1.2 Related Work Ricci et al. present in their works a survey of Matrix and Tensor Factorization while 2 Vol. 11, No. 2; 2018 Computer and Information Science cis.ccsenet.org Bokde et al. studied Matrix Factorization models and how they are used in Collaborative Filtering. 1.3 Organization 3.3 Context-aware Matrix Factorization 3.3 Context-aware Matrix Factorization There are three basic approaches to develop Context-aware recommendation algorithms as described in (Adomavicius & Mobasher, 2011): pre-filtering, post-filtering, and contextual modeling. In pre-filtering approach, contextual information is used to filter out irrelevant ratings before they are used for computing recommendations. Three approaches exists in Context-aware splitting approaches (CASA): User, Item and User-Item splitting. In Item splitting (Baltrunas & Ricci, 2014); (Baltrunas & Ricci, 2009), multiple copies of an item are created to hold ratings generated in different context based on the contexts in which the rating has been rated. When a recommendation is sought, only those items matching the current context are considered. In the case of User splitting (Baltrunas & Amatriain, 2009); (Said, De Luca, & Albayrak, 2011), which splits Users instead of Items. The third CASA is the User-Item splitting, a combination of item splitting and user splitting. The User-Item splitting approach has been shown to be superior to the separate Item or User splitting approaches (Zheng, Mobasher, et al., 2013). In post-filtering approach, contextual information is used after the classical two dimensional recommendation methods are applied to the non-contextual recommendation data (Baltrunas & Ricci, 2014); (Zheng, Burke, & Mobasher, 2014);(Zheng, Mobasher, & Burke, 2015). Two approaches are common in post-filtering: model-based and heuristic. In model-based, items are filtered out from the recommended result list by building models to predict the probability that the recommended item is relevant to the user in a given contextual situation. Items that have a lower probability than a set threshold are filtered from the list and re-ranking is done based on the probability –weighted rating. Contextual modeling consists of using the context within the recommendations models (Zheng, Burke, & Mobasher, 2012); (Zheng, Burke, & Mobasher, 2013). Contextual modeling incorporates contextual information directly into its recommendation process. Predictive models like Context-aware matrix factorizations, regression and decision trees are examples of contextual modeling techniques that incorporate context into their approach. The social network approach to provide recommendations for a user based on the ratings of the users that have direct or indirect social relations with the given user. Trust-based Context-aware Matrix Factorization (TCMF) incorporate trust information into both user bias and user-item-context interaction (Li, Sun, & Lv, 2014). The factorization method builds on the works of (Jamali & Ester, 2010) on Trust-Aware Collaborative Filtering. 3.2 Matrix Factorization Techniques Collaborative Filtering suffers from data sparsity and scalability problem as the number of users and items grow. Matrix decomposition has emerged as a powerful tool to expose hidden structure behind the data. Some of the commonly used Matrix Factorization models are Singular Value Decomposition (Sarwar, Karypis, Konstan, & Riedl, 2000), Principal Component Analysis (PCA) (Goldberg & Roeder, 2014), Probabilistic Matrix 3 Computer and Information Science Vol. 11, No. 2; 2018 cis.ccsenet.org Factorization (PMF) (Salakhutdinov & Mnih, 2008) and Non-Negative Matrix Factorization (Cai, He, Han, & Huang, 2011). The matrix decomposition methods are based on Matrix Factorization (Koren et al., 2009; Rendle et al., 2011; Sarwar et al., 2000; Takács et al., 2008). Their results showed the SVD based prediction algorithms can overcome the sparsity problem by utilizing the latent relationships. SVD could be further improved by adding biases to users and items. Koren et al. proposed SVD++ to make use of implicit feedback, reported high accuracy at the expense of high computational cost. Tensor Factorization (TF) extends the traditional two-dimensional Matrix Factorization problem into an n-dimensional version of the same problem by incorporating contextual information (Gautam, Chaudhary, Sindhwani, & Bedi, 2016). The resulting multi-dimensional matrix is factored into lower-dimensional representation, where the user, the item and each contextual dimension are represented with a lower dimensional feature vector (Baltrunas et al., 2011; Hidasi & Tikk, 2013). Karatzoglou et al. (Karatzoglou et al., 2010) proposed a multiverse recommendation model by employing CF method based on Tensor Factorization. Some studies focus on interpretation of latent features that result from factorization of the user-item matrix. The features extracted from the Non-negative Matrix Factorization are used to establish relationships between users and items (Zhang, Wang, Ford, & Makedon, 2006). In these works, features are viewed as groups of users, groups or items or as attributes of items. These interpretations require human intervention. In (Brun, Aleksandrova, & Boyer, 2014) latent features are interpreted as users. 3.3 Context-aware Matrix Factorization In (Li, Yang, & Jiang, 2016), the static trust model is extended through the use of the Social Network Analysis into new Dynamic Trust-based Context-aware Matrix Factorization (DTCMF) to fully capture the dynamics of trust. The comparative results shows superior performance by the dynamic trust Context-aware Matrix Factorization techniques. Comparative study of Boolean Matrix Factorization (BMF) with SVD is presented by (Akhmatnurov & Ignatov, 2015). The experiments with the Boolean matrix product of binary matrices with and without incorporating contextual information with SVD reported a higher precision for BMF where the number of neighbors is not high. A Tensor Factorization model based on fuzzy mapping between context factors and latent factors is proposed by 4 4 Computer and Information Science Vol. 11, No. 2; 2018 cis.ccsenet.org (Fang & Guo, 2013). In this work movie tags and release time we used as the contextual variables. The combination of tags and release time was modelled as the multidimensional context. The TimeTag (TTSVD) was reported to have achieved better RMSE and HLU while reducing iteration number by 25%. The Sparse Linear method (SLIM) was designed for Top-N recommendation in the traditional Recommender Systems (Ning & Karypis, 2011) improves upon the item-based nearest neighbor Collaborative Filtering by learning directly from the data, a sparse matrix of aggregation coefficients that are similar to the traditional item-item similarities. Contextual SLIM (Zheng, Mobasher, et al., 2014), which is derived from the SLIM incorporates contextual information is a Matrix Factorization approach for Top-N recommendations. Table 1 provides a summary of the various works in Context-aware Matrix Factorization techniques, evaluation metrics and the results of the comparative studies. Table 1. Summary of Context-aware Matrix Factorization techniques 5 Table 1. Summary of Context-aware Matrix Factorization techniques Reference MF Technique Description Metrics Results (Li et al., 2016) Dynamic Trust-based Context-aware Matrix Factorization for Collaborative Filtering (DTCMF). Uses social network analysis to create dynamic trust model with context- aware matrix factorization. DTCM-I (both user bias and item bias are static) while DTCMF-II (user bias is con-text-trust-changing and item bias is context-changing.) MAE The DTCMF-I & II both outperforms TACF, TCMF-I&II (Li et al., 2014) Incorporating context into the Trust-Aware Collaborative Filtering (TACF). Trust-based Context-aware Matrix Factorization for Collaborative Filtering (TCMF-I & TCMF-II). Two approaches to incorporate contextual information into TACF. Incorporates trust information into both user bias and user-item-context interaction. 3.3 Context-aware Matrix Factorization TCMF-I (both user bias and item bias are static while) while in TCMF-II (user bias is con- text-trust-sensitive and item bias is context-sensitive). MAE The TCMF-I & II outperforms ICAMF-I & II models. (Akhmatnurov & Ignatov, 2015) Boolean Matrix Factorisation (BMF) Comparative study BMF with SVD. Experiments on decomposition of the original matrix into a Boolean matrix product of binary matrices with/without incorporating contextual information comparing with SVD. MAE, Precision and Recall, F-measure. Higher precision reported for BMF. (Fang & Guo, 2013) TagSVD (T-SVD) and TimeTagSVD (TT-SVD). TF model based on fuzzy mapping between context factors and latent factors. Improved the performance of SVD, bias SVD and BPMF using SGD and MCMC. MAE, RMSE Half Life Utility (HLU) TT-SVD outperforms SVD, NMF & Slope One. (Karatzoglou et al., 2010) Multiverse Tensor Factorization recommendation model based on CF. Tensor Factorization allowed integration of contextual information resulting in a multidimensional matrix of User-Item Context as a tensor, where context can be taken in any number. MAE TF outperforms both contextual pre-filtering methods and the context free MF (Zheng et al., 2015) Context Correlation-based Context-aware Matrix Factorization (Correlation-based CAMF). Used the correlations between contextual deviations and incorporated into the Matrix Factorization to formulate the correlation-based CAMF algorithm. Precision, Mean Average Precision (MAP) Correlation-based CAMF outperforms deviation-based CAMF and the TF algorithm. (Zheng, Mobasher, et al., 2014) Contextual Sparse Linear Method (CSLIM). MF approach for Top-N recommendations that incorporates contextual information. Precision and MAP CSLIM outperforms TF, CAMF and CASA. (Baltrunas & Ricci, 2014) Context dependent CF using item splitting. Used both gradient descent Matrix Factorization and nearest neighbor CF algorithms. MAE, Precision and Recall. Item splitting with Matrix Factorization outperforms kNN. (Chen, Zheng, Wang, Hong, & Lin, 2014) Context-based Collaborative Topic Regression Social Use spectral clustering for user-item subgroupings, so as to group users and items in similar context. Incorporate topic modelling. MAE, Recall and Root Mean Square Error C-CTR-SMF2 outperforms PMF, SoCo, fLDA, CTR and 5 Computer and Information Science Vol. 11, No. 2; 2018 Vol. 11, No. 2; 2018 cis.ccsenet.org Reference MF Technique Description Metrics Results Matrix Factorization (C-CTR-SMF2). Use hierarchical Bayesian model to make predictions for each user-item subgroup. (RMSE). CTR-SMF. (Li, Feng, & Lv, 2013) Improved CAMF: ICAMF-I Both user bias and item bias are static and ICAMF-II (the timely change of user bias and item bias over different context.) MAE Both ICAMF-I&II outperforms CAMF-C, CAMF-CC and CAMF-CI models. 4. Conclusion and Future Work This work provided a broad overview of available approaches to incorporating contextual information into Collaborative Filtering based Recommender Systems utilizing Matrix Factorization techniques. Although the field of Recommender Systems has developed considerably, the aspects of quality of recommendations, sparsity, scalability, cold start and privacy concerns have remained unresolved since the beginning of Recommender Systems research. Recommender Systems whatever their approach in providing recommendations has proven to be useful in overcoming the problem of information overload by delivering relevant personalized information. CF is the most successful and widely used recommendation technique. However, just like the other approaches, the technique suffers from limitations of data sparsity and scalability problems that hamper the performance, quality and the accuracy of the predictions. The main reason behind data sparsity in Collaborative Filtering is that most users do not rate most of the items making the available ratings sparse. CF suffers from this problem because it is dependent on the rating matrix. The incorporation of contextual information has been suggested as a remedy to overcome the traditional Recommender Systems problems. Context plays the role of additional information besides users, items and the ratings which may be relevant at the current time to make a recommendation. Matrix decomposition is a powerful technique to find the hidden structure behind the data. SVD, Non-negative Matrix Factorization and Probabilistic Matrix Factorization are popular decomposition models. SVD is able to handle large dataset, sparseness of rating matrix and scalability problem of CF algorithm efficiently. NMF is widely used to reduce dimensions and extract latent factors. PMF models place Gaussian priors on user and movie features, and turn the recommendation task into a probabilistic problem. Matrix/Tensor Factorization models could be optimized by using Stochastic Gradient Descent (SGD), Alternating Least Squares (ALS), or Markov Chain Monte Carlo Inference (MCMC). Tensor Factorization models are not popular in practical application for its high computational cost. The various available evaluation metrics can also be applied to evaluate performance of Recommender Systems by measuring their coverage and accuracy, but the current metrics are insufficient for evaluating quality and usefulness of Recommender Systems. Recommender Systems evaluation metrics are trapped in the recommendation accuracy. Future research directions of the CARS should be directed towards a unified definition of context and what constitutes contextual information. Another important aspect to consider is to address the privacy concerns that can be generated by the incorporation of contextual information. 3.3 Context-aware Matrix Factorization (Baltrunas et al., 2011) Dealt with varying granularities of the interaction of the contextual information with ratings. CAMF-C, CAMF-CI and CAMF- CC In CAMF-C every contextual factor had a global influence on the ratings independently from the item. CAMF-CI introduced one parameter per contextual factor and item pair. One parameter was introduced for each item category and contextual factor in CAMF-CC. MAE CAMF-CC outperforms CAMF, CAMF-C and CAMF-CI. (Cai et al., 2011) Graph regularized Non-Negative Matrix Factorization (GNMF) for data representation. Construct an affinity graph to encode the geometrical information of the data space by constructing a nearest neighbor graph. Accuracy (AC) and Normalized Mutual Information (NMI) GNMF outperforms NMF. 4. Conclusion and Future Work The conduct of more research in user studies is critical to obtain the necessary feedback and to get out of the accuracy cage. Although many advanced techniques have been developed to address data sparsity, it still remains a growing problem that needs to be addressed. The need for fast and scalable computations is critical and therefore great effort must be expended to develop efficient and scalable algorithms. Adomavicius, G., & Mobasher, B. (2011). Context-aware Recommender Systems. Artificial Intelligence References Adomavicius, G., & Mobasher, B. (2011). Context-aware Recommender Systems. Artificial Intelligence 6 Vol. 11, No. 2; 2018 Computer and Information Science cis.ccsenet.org Magazine, 32(3), 67-80. Magazine, 32(3), 67-80. Adomavicius, G., & Tuzhilin, A. (2005). Toward the Next Generation of Recommender Systems: A survey of the State-of-the-Art and Possible Extensions. IEEE Transactions on Knowledge and Data Engineering, 17(6), 734-749. https://doi.org/10.1109/TKDE.2005.99 Adomavicius, G., Sankaranarayanan, R., Sen, S., & Tuzhilin, A. (2005). Incorporating Contextual Information in Recommender Systems Using a Multidimensional Approach. ACM Transactions on Information Systems, 23(1), 103-145. https://doi.org/10.1145/1055709.1055714 Akhmatnurov, M., & Ignatov, D. I. (2015). Context-aware recommender system based on Boolean matrix factorisation. In CEUR Workshop Proceedings (Vol. 1466, pp. 99-110). Baldauf, M., Dustdar, S., & Rosenberg, F. (2007). A survey on Context-aware systems. International Journal Ad Hoc and Ubiquitous Computing, 2(4), 263-277. https://doi.org/10.1504/IJAHUC.2007.014070 Baltrunas, L. (2008). Exploiting contextual information in recommender systems. In Proceedings of the 2008 ACM conference on Recommender systems - RecSys ’08 (p. 295). https://doi.org/10.1145/1454008.1454056 Baltrunas, L., & Amatriain, X. (2009). Towards time-dependant recommendation based on implicit feedback. Workshop on Context-Aware Recommender Systems (CARS’09), 25-30. https://doi.org/10.1109/ICDM.2013.139 Baltrunas, L., & Ricci, F. (2009). Context-based splitting of item ratings in Collaborative Filtering. Proceedings of the Third ACM Conference on Recommender Systems - RecSys ’09, 245. https://doi.org/10.1145/1639714.1639759 Baltrunas, L., & Ricci, F. (2014). Experimental evaluation of context-dependent Collaborative Filtering using item splitting. User Modeling and User-Adapted Interaction, 24(1-2), 7-34. https://doi.org/10.1007/s11257-012-9137-9 Baltrunas, L., Ludwig, B., & Ricci, F. (2011). Matrix Factorization Techniques for Context aware recommendation. In Acm Rs (pp. 301-304). https://doi.org/10.1145/2043932.2043988 Bazire, M., & Brézillon, P. (2005). Understanding context before using it. In 5th International and Interdisciplinary Conference (Vol. 3554, pp. 29–40). https://doi.org/10.1007/11508373 Bokde, D., Girase, S., & Mukhopadhyay, D. (2015). Matrix Factorization model in Collaborative Filtering algorithms: A survey. Procedia Computer Science, 49(1), 136-146. https://doi.org/10.1016/j.procs.2015.04.237 Bouraga, S., Jureta, I., Faulkner, S., & Herssens, C. (2014). Knowledge-Based Recommendation Systems: International Journal of Intelligent Information Technologies, 10(2), 1-19. https://doi.org/10.4018/ijiit.2014040101 Breese, J. S., Heckerman, D., & Kadie, C. (1998). Empirical analysis of predictive algorithms for Collaborative Filtering. In UAI’98 Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence (pp. 43-52). https://doi.org/10.1111/j.1553-2712.2011.01172.x Brun, A., Aleksandrova, M., & Boyer, A. (2014). Can latent features be interpreted as users in matrix factorization-based recommender systems? In Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014 (Vol. 2, pp. 257-297). Technologies, 3(4(75)), 4. https://doi.org/10.15587/1729-4061.2015.43074 Technologies, 3(4(75)), 4. https://doi.org/10.15587/1729-4061.2015.43074 Technologies, 3(4(75)), 4. https://doi.org/10.15587/1729-4061.2015.43074 Coutaz, J., Crowley, J. L., Dobson, S., & Garlan, D. (2005). Context is key. Communications of the ACM, 48(3), 49. https://doi.org/10.1145/1047671.1047703 Dey, A. K., & Abowd, G. D. (2000). Towards a better understanding of context and Context-awareness. CHI 2000 Workshop on the What Who Where When and How of Context-awareness, 4(What, Who, Where, When and How of Context-Awareness), 1-6. https://doi.org/10.1007/3-540-48157-5 Fang, Y., & Guo, Y. (2013). A Context-aware Matrix Factorization recommender algorithm. In Proceedings of the IEEE International Conference on Software Engineering and Service Sciences, ICSESS (pp. 914-918). https://doi.org/10.1109/ICSESS.2013.6615454 Felfernig, A., Jeran, M., Ninaus, G., Reinfrank, F., Reiterer, S., & Stettinger, M. (2014). Basic approaches in recommendation systems. Recommendation Systems in Software Engineering, 15-37. https://doi.org/10.1007/978-3-642-45135-5_2 Frolov, E., & Oseledets, I. (2016). Tensor methods and recommender systems. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 7(3), 1-41. https://doi.org/10.1002/widm.1201 Gao, M., Liu, K., & Wu, Z. (2010). Personalisation in web computing and informatics: Theories, techniques, applications, and future research. Information Systems Frontiers, 12(5), 607-629. https://doi.org/10.1007/s10796-009-9199-3 Gautam, A., Chaudhary, P., Sindhwani, K., & Bedi, P. (2016). CBCARS: Content boosted Context-aware recommendations using tensor factorization. 2016 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2016, 75-81. https://doi.org/10.1109/ICACCI.2016.7732028 Goldberg, K., & Roeder, T. M. (2014). Eigentaste: A Constant Time Collaborative Filtering Algorithm. CEUR Workshop Proceedings, 1225(July), 41-42. https://doi.org/10.1023/A Herlocker, J. L., Konstan, J. A., Terveen, L. G., & Riedl, J. T. (2004). Evaluating Collaborative Filtering recommender systems. ACM Transactions on Information Systems, 22(1), 5-53. https://doi.org/10.1145/963770.963772 Hidasi, B., & Tikk, D. (2013). Context-aware item-to-item recommendation within the factorization framework. In Proceedings of the 3rd Workshop on Context-awareness in Retrieval and Recommendation - CaRR ’13 (pp. 19–25). https://doi.org/10.1145/2442670.2442675 Hornung, T., Ziegler, C. N., Franz, S., Przyjaciel-Zablocki, M., Schätzle, A., & Lausen, G. (2013). Evaluating hybrid music recommender systems. Proceedings - 2013 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2013, 1, 57-64. https://doi.org/10.1109/WI-IAT.2013.9 Ignatov, D. I., Nikolenko, S., Abaev, T., & Konstantinova, N. (2014). Online Recommender System for Radio Station Hosting: Experimental Results Revisited. In 2014 IEEE/WIC/ACM International Joint Conferences on Web Intelligence (WI) and Intelligent Agent Technologies (IAT) (pp. 229-236). https://doi.org/10.1109/WI-IAT.2014.38 Jamali, M., & Ester, M. (2010). A Matrix Factorization technique with trust propagation for recommendation in social networks. In Proceedings of the fourth ACM conference on Recommender systems - RecSys ’10 (p. 135). https://doi.org/10.1145/1864708.1864736 Karatzoglou, A., Amatriain, X., Baltrunas, L., & Oliver, N. (2010). References https://doi.org/10.1109/WI-IAT.2014.102 Cacheda, F., Carneiro, V., Fernández, D., & Formoso, V. (2011). Comparison of Collaborative Filtering algorithms: Limitations of Current Techniques and Proposals for Scalable, High-Performance Recommender Systems. ACM Transactions on the Web, 5(1), 1-33. https://doi.org/10.1145/1921591.1921593 Cai, D., He, X., Han, J., & Huang, T. S. (2011). Graph regularized Non-negative Matrix Factorization for data representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33(8), 1548-1560. https://doi.org/10.1007/978-3-642-34528-9_1 Chen, C., Zheng, X., Wang, Y., Hong, F., & Lin, Z. (2014). Context-aware Collaborative Topic Regression with Social Matrix Factorization for Recommender Systems. In Twenty-Eighth AAAI Conference (pp. 9-15). Chertov, O., Brun, A., Boyer, A., & Aleksandrova, M. (2015). Comparative analysis of neighborhood-based approache and Matrix Factorization in recommender systems. Eastern-European Journal of Enterprise 7 7 Vol. 11, No. 2; 2018 Computer and Information Science cis.ccsenet.org Technologies, 3(4(75)), 4. https://doi.org/10.15587/1729-4061.2015.43074 Multiverse recommendation: n-dimensional Tensor Factorization for Context-aware Collaborative Filtering. Proceedings of the Fourth ACM Conference on Recommender Systems - RecSys ’10, 79. https://doi.org/10.1145/1864708.1864727 Karydi, E., & Margaritis, K. (2014). Parallel and Distributed Collaborative Filtering: A Survey. arXiv Preprint arXiv:1409.2762, 49(2). https://doi.org/10.1145/2951952 Koren, Y., & Bell, R. (2015). Advances in Collaborative Filtering. Recommender Systems Handbook, Second Edition, 77–118. https://doi.org/10.1007/978-1-4899-7637-6_3 Koren, Y., Bell, R., & Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, 42(8), 30–37. https://doi.org/10.1109/MC.2009.263 Krohn-Grimberghe, A., Nanopoulos, A., & Schmidt-Thieme, L. (2010). A novel multidimensional framework for evaluating recommender systems. CEUR Workshop Proceedings, 612, 34-41. Li, J., Feng, P., & Lv, J. (2013). ICAMF: Improved Context-aware Matrix Factorization for Collaborative 8 Computer and Information Science Vol. 11, No. 2; 2018 cis.ccsenet.org Filtering. Proceedings - International Conference on Tools with Artificial Intelligence, ICTAI, 63-70. https://doi.org/10.1109/ICTAI.2013.20 Filtering. Proceedings - International Conference on Tools with Artificial Intelligence, ICTAI, 63-70. https://doi.org/10.1109/ICTAI.2013.20 Li, J., Sun, C., & Lv, J. (2014). TCMF:Trust-Based Context-Aware Matrix Factorization for Collaborative Filtering. In 2014 IEEE 26th International Conference on Tools with Artificial Intelligence (pp. 815-821). https://doi.org/10.1109/ICTAI.2014.126 Li, J., Yang, R., & Jiang, L. (2016). DTCMF: Dynamic trust-based Context-aware Matrix Factorization for Collaborative Filtering. In Proceedings of 2016 IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2016 (pp. 914–919). https://doi.org/10.1109/ITNEC.2016.7560495 Liu, N. N., Cao, B., Zhao, M., & Yang, Q. (2010). Adapting neighborhood and Matrix Factorization models for context aware recommendation. In CAMRa2010 (pp. 7–13). https://doi.org/10.1145/1869652.1869653 Meehan, K., Lunney, T., Curran, K., & McCaughey, A. (2013). Context-aware intelligent recommendation system for tourism. 2013 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2013, (March), 328–331. https://doi.org/10.1109/PerComW.2013.6529508 Melville, P., Mooney, R. J., & Nagarajan, R. (2002). Content-boosted Collaborative Filtering for improved recommendations. “Proceedings of the 18th National Conference on Artificial Intelligence (AAAI),” (July), 187-192. https://doi.org/10.1.1.16.4936 Nguyen, J., & Zhu, M. (2013). Content-boosted Matrix Factorization techniques for recommender systems. Statistical Analysis and Data Mining, 6(4), 286-301. https://doi.org/10.1002/sam.11184 Ning, X., & Karypis, G. (2011). SLIM: Sparse LInear Methods for top-N recommender systems. In Proceedings - IEEE International Conference on Data Mining, ICDM (pp. 497-506). https://doi.org/10.1109/ICDM.2011.134 Park, D. H., Choi, I. Y., Kim, H. K., & Kim, J. K. (2011). A Review and Classification of Recommender Systems Research. 2011 International Conference on Social Science and Humanity, 5(1), 290-294. https://doi.org/10.7763/IPEDR Patil, P., & Kolhe, V. L. (2014). Survey of Travel Package Recommendation System. Technologies, 3(4(75)), 4. https://doi.org/10.15587/1729-4061.2015.43074 International Journal of Science and Research, 3(12), 1557-1561. Rajabpour, N., Bardsiri, V. K., Mohammadighavam, A., & Molaei, E. (2014). Application domain of recommender system: a survey. International Journal of Advanced Studies in Computer Science and Engineering, 3(2), 8-14. Rendle, S., Gantner, Z., Freudenthaler, C., & Schmidt-Thieme, L. (2011). Fast Context-aware recommendations with factorization machines. Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information - SIGIR ’11, 635. https://doi.org/10.1145/2009916.2010002 Ricci, G., De Gemmis, M., & Semeraro, G. (2012). Matrix and Tensor Factorization Techniques applied to Recommender Systems: a Survey. Matrix and Tensor Factorization Techniques Applied to Recommender Systems: A Survey, 1(1), 94-98. Said, A., De Luca, E. W., & Albayrak, S. (2011). Inferring contextual user profiles improving recommender performance. CEUR Workshop Proceedings, 791. Salakhutdinov, R., & Mnih, A. (2008). Bayesian probabilistic Matrix Factorization using Markov chain Monte Carlo. In Proceedings of the 25th international conference on Machine learning - ICML ’08 (Vol. 25, pp. 880-887). https://doi.org/10.1145/1390156.1390267 Sarkaleh, M. K., Mahdavi, M., & Baniardalan, M. (2012). Designing a Tourism Recommender System Based on Location, Mobile Device and User Features in Museum. International Journal of Managing Information Technology, 4(2), 13-21. https://doi.org/10.5121/ijmit.2012.4202 Sarwar, B. M., Karypis, G., Konstan, J. a, & Riedl, J. T. (2000). Application of Dimensionality Reduction in Recommender System - A Case Study. Architecture, 1625, 264-8. https://doi.org/10.1.1.38.744 Sharma, L., & Gera, A. (2013). A Survey of Recommendation System: Research Challenges. International Journal of Engineering Trends and Technology, 4(5), 1989-1992. Sharma, M. (2013). A Survey of Recommender Systems: Approaches and Limitations. International Journal of Innovations in Engineering and Technology, 1-9. Shi, Y., Larson, M., & Hanjalic, A. (2014). Collaborative Filtering beyond the User-Item Matrix: A Survey of the 9 9 Vol. 11, No. 2; 2018 Computer and Information Science cis.ccsenet.org State of the Art and Future Challenges. ACM Computing Surveys (CSUR), 47(1), 1-45. https://doi.org/10.1145/2556270 State of the Art and Future Challenges. ACM Computing Surveys (CSUR), 47(1), 1-45. https://doi.org/10.1145/2556270 Shinde, S., & Potey, M. A. (2015). Survey on Evaluation of Recommender Systems. International Journal of Engineering And Computer Science ISSN:2319-7242, 4(2), 1-5. Su, X., & Khoshgoftaar, T. M. (2009). A Survey of Collaborative Filtering Techniques. Advances in Artificial Intelligence, 2009(Section 3), 1–19. https://doi.org/10.1155/2009/421425 Takács, G., Pilászy, I., Németh, B., Tikk, D., Pilaszy, I., N’emeth, B., & Tikk, D. (2008). Investigation of Various Matrix Factorization Methods for Large Recommender Systems. Technologies, 3(4(75)), 4. https://doi.org/10.15587/1729-4061.2015.43074 Proceedings of the 2nd KDD Workshop on Large Scale Recommender Systems and the Netflix Prize Competition, 1, 553-562. https://doi.org/10.1109/ICDMW.2008.86 Umanets, A., Ferreira, A., & Leite, N. (2014). GuideMe – A Tourist Guide with a Recommender System and Social Interaction. Procedia Technology, 17, 407–414. https://doi.org/10.1016/j.protcy.2014.10.248 Wang, Y., Chan, S. C. F., & Ngai, G. (2012). Applicability of demographic recommender system to tourist attractions: A case study on TripAdvisor. Proceedings of the 2012 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology Workshops, WI-IAT 2012, 97-101. https://doi.org/10.1109/WI-IAT.2012.133 Yang, X., Guo, Y., Liu, Y., & Steck, H. (2013). A survey of Collaborative Filtering based social recommender systems. Computer Communications, (0). https://doi.org/http://dx.doi.org/10.1016/j.comcom.2013.06.009 Zhang, S., Wang, W., Ford, J., & Makedon, F. (2006). Learning from Incomplete Ratings Using Non-negative Matrix Factorization. In Proceedings of the 2006 SIAM International Conference on Data Mining (pp. 549– 553). https://doi.org/10.1137/1.9781611972764.58 Zheng, Y., Burke, R., & Mobasher, B. (2012). Differential context relaxation for Context-aware travel recommendation. Lecture Notes in Business Information Processing, 123 LNBIP, 88-99. https://doi.org/10.1007/978-3-642-32273-0_8 Zheng, Y., Burke, R., & Mobasher, B. (2013). Recommendation with differential context weighting. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 7899 LNCS, 152–164. https://doi.org/10.1007/978-3-642-38844-6_13 Zheng, Y., Burke, R., & Mobasher, B. (2014). Splitting approaches for Context-aware recommendation: An Empirical Study. In Proceedings of the 29th Annual ACM Symposium on Applied Computing - SAC ’14 (pp. 274–279). https://doi.org/10.1145/2554850.2554989 Zheng, Y., Mobasher, B., & Burke, R. (2014). CSLIM:Contextual SLIM Recommendation Algorithms. In RecSys 2014 - Proceedings of the 8th ACM Conference on Recommender Systems (Vol. 0, pp. 301-304). https://doi.org/10.1145/2645710.2645756 Zheng, Y., Mobasher, B., & Burke, R. (2015). Incorporating context correlation into Context-aware matrix factorization. In CEUR Workshop Proceedings (Vol. 1440). Zheng, Y., Mobasher, B., & Burke, R. D. (2013). The Role of Emotions in Context-aware Recommendation. Decisions@ RecSys, 2013, 21-28. for this article is retained by the author(s), with first publication rights granted to the journal. Copyrights Copyright for this article is retained by the author(s), with first publication rights granted to the journal. Copyright for this article is retained by the author(s), with first publication rights granted to the journal. Copyright for this article is retained by the author(s), with first publication rights granted to the journal. This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/). This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/). 10
https://openalex.org/W1826733058
https://jlsrf.org/index.php/lsf/article/download/43/pdf
English
null
MEDAPP: Fission neutron beam for science, medicine, and industry
Journal of large-scale research facilities
2,015
cc-by
568
MEDAPP: Fission neutron beam for science, medicine, and industry Heinz Maier-Leibnitz Zentrum Technische Universität München Instrument Scientists: Instrument Scientists: - Christoph Genreith, Heinz Maier-Leibnitz Zentrum (MLZ), Technische Universität München, Garching, Germany, phone: +49(0) 89 289 14756, email: christoph.genreith@frm2.tum.de Abstract: The instrument MEDAPP (Medical Applications), operated by the Technische Universität München, and the respective irradiation position are located at the world-wide unique fast neutron beam tube SR10 to which a uranium converter is attached. Thus, the instrument is operated with unmoderated fssion neutrons and can be used for a broad variety of applications. For selected tasks, an alternative use with thermal neutrons is possible. http://dx.doi.org/10.17815/jlsrf-1-43 Journal of large-scale research facilities, 1, A18 (2015) 1 Introduction MEDAPP is an instrument primarily built for the medical treatment of malignant tumours; but the irra- diation room (see Figure 1 and 2) can also be used for general purposes, e.g. for biological research and technical irradiations. Due to their energy spectrum, fast reactor neutrons have the highest biological efectiveness of clinical neutron beams used in cancer treatment, comparable only to the efectiveness of heavy ions. This advantage comes at the expense of penetration depth in tissue, which - due to the relatively low energy of 2 MeV - limits the application of fast reactor neutrons to near-surface tumours, typically recurrent breast tumours and melanomas. The particularly large beam cross-section of SR10 allows the irradiation of rather large objects, such as groups of cell culture fasks or complete elec- tronic devices. In addition, the FaNGas (Fast Neutron Gamma Spectrometry) instrument, consisting of a movable shielded HPGe detector system, can be installed within the MEDAPP irradiation chamber to directly measure gamma radiation emitted, e.g., in (n,n’), (n,2n), (n,p), and (n,α) reactions and for non-destructive qualitative elemental analysis. 1 1 Journal of large-scale research facilities, 1, A18 (2015) http://dx.doi.org/10.17815/jlsrf-1-43 http://dx.doi.org/10.17815/jlsrf-1-43 Figure 1: The irradiation room of MEDAPP (Copyright by W. Schürmann, TUM). Figure 1: The irradiation room of MEDAPP (Copyright by W. Schürmann, TUM). 3.2 Neutron spectrum • Fission spectrum: Mean energy: 1.9 MeV Flux: up to 7 · 108 n cm-2 s-1 (depending on flter used) • Thermal spectrum of the D2O moderator (without converter): Mean energy: 28 meV Flux: ca. 2 · 109 n cm-2 s-1 2 Typical Applications • Neutron medical treatment of malign tumours • Neutron medical treatment of malign tumours • Biological dosimetry, e.g., irradiations of cell cultures • Irradiations of electronic components (also on-line tests) • Fundamental physics • In beam gamma spectrometry 3.3 Collimation • Multi-leaf collimator, individually adjustable up to 27 cm x 19 cm. • Multi-leaf collimator, individually adjustable up to 27 cm x 19 cm. 2 Journal of large-scale research facilities, 1, A18 (2015) http://dx.doi.org/10.17815/jlsrf-1-43 Figure 2: Schematic drawing of the facilities MEDAPP and NECTAR (Heinz Maier-Leibnitz Zentrum, 2015) at beam tube SR10. Figure 2: Schematic drawing of the facilities MEDAPP and NECTAR (Heinz Maier-Leibnitz Zentrum, 2015) at beam tube SR10. 3.5 Detection systems • Ionisation chambers for dosimetry in custom-made phantoms • Ionisation chambers for dosimetry in custom-made phantoms • 50 %-HPGe detection system shielded with PE, B4C, and Pb y p • 50 %-HPGe detection system shielded with PE, B4C, and Pb • Custom systems can temporarily be installed by users 3.4 Sample space • Max. 40 cm x 30 cm References Heinz Maier-Leibnitz Zentrum. (2015). Nectar: radiography and tomography station using fssion neutrons. Journal of large-scale research facilities, 1, A19. http://dx.doi.org/10.17815/jlsrf-1-45 3 3
https://openalex.org/W4312534659
http://semestreeconomico.unap.edu.pe/index.php/revista/article/download/32/118
es
Medición de la eficiencia técnica de las Escuelas Profesionales de la Universidad Nacional del Altiplano: una aplicación del DEA
Semestre económico
2,020
cc-by
0
W2958067602.txt
https://www.autopsyandcasereports.org/article/10.4322/acr.2019.099/pdf/autopsy-9-3-e2019099.pdf
en
Pediatric Granular Cell Tumor of the Breast: An uncommon neoplasm in an uncommon site and age group
Autopsy and Case Reports
2,019
cc-by
2,660
Article / Clinical Case Report Pediatric Granular Cell Tumor of the Breast: An uncommon neoplasm in an uncommon site and age group Pakesh Baishyaa , Jonali Dasa , Darilin Shangplianga, Biswajit Deya , Evarisalin Marbanianga , Donboklang Lynserb , Yookarin Khonglaha , Vandana Raphaela How to cite: Baishya P, Das J, Shangpliang D, et al. Pediatric granular cell tumor of the breast: An uncommon neoplasm in an uncommon site and age group. Autops Case Rep [Internet]. 2019 Jul-Sep;9(3):e2019099. https://doi.org/10.4322/acr.2019.099 ABSTRACT Granular cell tumor (GCT) is a rare soft tissue neoplasm of Schwann cell origin. Most cases occur in adults; however, the precise incidence is unknown in children. GCT is usually a slow-growing, painless tumor involving the skin and soft tissues that is mostly located in the head and neck region, especially the tongue. The breast is one of the least common sites involved by GCT. This paper presents a 3-year-old girl who presented with a soft to firm, ill-defined swelling on the right breast with painful ulceration of the overlying skin. Fine needle aspiration rendered an initial diagnosis of fibrocystic change accompanied by apocrine metaplasia. Histologic evaluation of the excised breast mass revealed a benign granular cell tumor. Although rare, GCT of the breast should be included in the differential diagnosis for breast masses in pediatric patients. Proper diagnosis and timely management of this tumor are essential because of its malignant potential (<2% of cases) and high rate of local recurrence if not properly excised. Keywords: Breast; Schwann Cells; S100 Proteins. INTRODUCTION Granular Cell Tumor (GCT) is a rare soft tissue neoplasm of probable Schwann cell origin. head and neck, and 40% of all cases arise in the In 1926 tongue.4,5 The breast is involved exceptionally with an this tumor was first described by Abrikossof who overall occurrence of 5-6% of all cases.5 Taking into named it as granular cell myoblastoma thinking the account of all GCTs in all the sites, females are affected tumor originated from myofibroblasts.1,2 In 1962, Fisher 2-3 times more commonly than men.3 Most lesions and Wechles confirmed the neural origin of the tumor.2 are painless and solitary. Notwithstanding 5-25% of GCT most commonly affects adults between 30 and the cases may be multiple.2 Malignant cases account 60 years of age, and it has an exceptional occurrence for 1-2% of all cases and show a poor prognosis.2 among children.3 GCT occurs most commonly in the Incidence is common in people of African descent.4 a b 1,2 North Eastern Indira Gandhi Regional Institute of Health and Medical Sciences (NEIGRIHMS), Department of Pathology. Mawdiangdiang, Shillong, India. North Eastern Indira Gandhi Regional Institute of Health and Medical Sciences (NEIGRIHMS), Department of Radiodiagnosis. Mawdiangdiang, Shillong, India. Autopsy and Case Reports. ISSN 2236-1960. Copyright © 2019. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium provided the article is properly cited. Pediatric Granular Cell Tumor of the Breast: An uncommon neoplasm in an uncommon site and age group CASE REPORT A 3-year-old girl presented with ulcerated painful swelling over the right breast for one and a half months (Figure 1A). On examination, a soft to firm, ill-defined, ulcerated swelling measuring 3x2.2x1.5 cm at the upper outer quadrant of the right breast was noted. Doppler ultrasonography of the breast revealed a well circumscribed, hypoechoic mass with scant internal vascularity (Figure 1B). Fine needle aspiration cytology (FNAC) was performed, and the smears revealed cellular clusters and singly lying cells with round centrally placed nuclei, bland chromatin, and abundant granular cytoplasm with indistinct cytoplasmic borders. The background showed occasionally scattered macrophages (Figure 2A and 2B). The diagnosis of fibrocystic lesion of the breast with apocrine metaplasia was made. The skin nodule was excised. Grossly, the surgical specimen measured Figure 1. A – external examination of the right hemithorax showing the presence of an ulcerated swelling over right breast. B – Doppler ultrasonography of the breast revealing a well circumscribed hypoechoic mass with sparse internal vascularity. Figure 2. Photomicrographs of fine needle aspiration cytology showing clusters of cells with round, centrally placed nuclei, bland chromatin, and abundant granular cytoplasm with indistinct cytoplasmic borders (A – May-Grünwald Giemsa; B – Papanicolaou stain, 400x). 2-6 Autops Case Rep (São Paulo). 2019 Jul-Sep;9(3):e2019099 Baishya P, Das J, Shangpliang D et al. 4.5x2.5x1.4 cm. The overlying skin showed a central ulceration and measured 1x1 cm (Figure 3), and the cut section showed a solid, light grey, well-circumscribed subcutaneous nodule involving the overlying skin measuring 3x2.2x1.5 cm. Figure 3. Surgical specimen showing a solid, light grey, well circumscribed nodule. The histopathological examination showed sheets of polygonal cells exhibiting abundant eosinophilic and granular cytoplasm and vesicular nuclei (Figure 4A). Mitoses were sparse with fewer than 1 per 10 high power field of 200x magnification. There was no necrosis. The tumor cells were Periodic Acid Schiff (PAS) positive and diastase resistant Figure 4. Photomicrographs of the tumor. A – sheets of polygonal cells exhibiting abundant eosinophilic, granular cytoplasm, and vesicular nuclei (H&E, 400X); B – The tumor cells were PAS positive and diastase resistant (PAS, 400X); C – The tumors cells were positive for S100 (200X); D – Ki67 index less than 1% (400x). Autops Case Rep (São Paulo). 2019 Jul-Sep;9(3):e2019099 3-6 Pediatric Granular Cell Tumor of the Breast: An uncommon neoplasm in an uncommon site and age group (Figure 4B). Immunohistochemistry showed S100 and neuron‑specific enolase (NSE) positivity in the tumor cells. S100 showed both nuclear and cytoplasmic positivity in the tumor cells (Figure 4C). Ki67 index was less than 1% (Figure 4D) and p53 expression was negative. Thus, a final diagnosis of benign GCT of the right breast was made. All the surgical margins were free of tumor. After six months of follow-up, the patient is doing well without evidence of tumor recurrence. DISCUSSION Granular cell tumor is a rare soft tissue neoplasm of probable Schwann cell origin.1,2 Clinically, cutaneous GCT has a nonspecific appearance rendering it a challenging diagnosis to make. GCT presents as an asymptomatic, solitary, hard nodule, occasionally sensitive and itchy, with variable coloration including native skin to brown-red color.2 Most cases of GCT are benign but about 1-3% are malignant.2 Although GCT may occur anywhere in the body, the head and neck region is most frequently involved, accounting for 45-65% of all cases.4 About 70% of the tumors in the head and neck occur in the oral cavity, most commonly involving the tongue.4 Others sites include respiratory (10%) and gastrointestinal systems (4%-6%).6 Breast is an uncommon location for GCT being involved in up to 6% of published cases.4,5 GCT of the breast most commonly occurs in women in the fourth and fifth decades with a frequency approximately 1 in 1000 breast neoplasms.5 It is extremely rare in children.3 To the best of our knowledge, only 7 cases of breast GCT in children have been reported in the English literature (Table1).7-13 Our case is the youngest of the 7 reported cases. While most of the reported cases presented with a unilateral painless breast mass, our patient presented with a unilateral painful breast mass, with the pain likely attributed to ulceration of the skin, which is usually not seen in breast GCT (Table 1). The present case is rare due to its uncommon site, age at occurrence, and clinical presentation. Cytologically, breast GCT has a varied differential diagnosis including benign lesions like fibrocystic lesion with apocrine metaplasia, fat necrosis, and malignant tumors like ductal carcinoma with apocrine metaplasia. 14 In the present case, the cytological diagnosis was a fibrocystic lesion of the breast with apocrine metaplasia as the cells showed abundant cytoplasm that was interpreted as apocrine metaplasia of the ductal epithelial cells.14 Cytologically apocrine Table 1. Cases of breast GCT in children reported in the English literature Authors Age (Y) Gender Clinical presentation Tumor size (cm) Side and location Cytological diagnosis Histological diagnosis Apisarnthanarax7 15 F Painless mass; No skin changes 1.5 Left breast Not known GCT (Excision) Aderiran8 17 F Mass 0.9 Left breast Not known GCT (Excision) Yang 9 17 F Palpable painless mass 2.7 Right subareolar Not done GCT (core and excision) De Simone10 14 F Palpable intermittently painful mass; No skin retraction 2.6 Left upper inner quadrant Not done GCT (core and excision) Marshall11 15 F Two palpable painful masses; No skin changes 1.1 Both masses in right breast Not done GCT and fibroadenoma (Excision) Heinzerling12 15 F Palpable painless mass; No skin changes 3 Right upper outer quadrant Not done GCT (Excision) Omar13 9.1 F Palpable painful mass; No skin changes 1.1 Right upper outer quadrant Not done GCT (Excision) 3 F Palpable mass; Skin ulceration 4.5 Right upper outer quadrant Done GCT (Excision) Present case cm = centimeter; F = female; GCT = granular cell tumor; Y = years. 4-6 Autops Case Rep (São Paulo). 2019 Jul-Sep;9(3):e2019099 Baishya P, Das J, Shangpliang D et al. cells have a well-defined cytoplasmic border in contrast to the undefined cytoplasmic borders characteristic of GCT cells.14 Moreover, the cytoplasmic granularity is more prominent in GCT. 14 Benign breast lesions have two-dimensional sheets comprised of ductal and myoepithelial cells, in contrast to syncytial or three‑dimensional clusters of cells in GCT.14 Ultrastructural analyses and immunohistochemical studies based on the S100 positivity demonstrate that GCT originates from the Schwann cells. 15 Histologically, the tumor consists of a well-demarcated dermal proliferation of round or polygonal cells with a centrally located nucleus and granular eosinophilic cytoplasm. 16 Ultrastructural studies elucidated that the granularity seen on light microscopy is due to the numerous lysosomes at the ultrastructural level.17 Cytoplasm contains pustulo-ovoid bodies which are Periodic Acid Schiff (PAS) positive and diastase resistant.14,16 Usually, epidermis overlying the tumor shows pseudoepitheliomatous hyperplasia which introduces squamous cell carcinoma into the potential histologic differential diagnosis of GCT in the skin.16 Apart from S100 and NSE, GCT shows a varying degree of positivity for a wide panel of markers including Inhibin A, Calretinin, Vimentin, Protein gene product (PGP 9,5), Melanoma directed monoclonal antibody NKI/C3, and CD68.16,17 Six diagnostic criteria were described for malignant GCT,17 which include (i) Necrosis, (ii) Spindle formation in the cells, (iii) A vesicular nucleus with a large nucleoli, (iv) Increased mitotic activity (>2 mitosis/ 10 high‑power fields at 200x magnification), (v) High nuclear to cytoplasmic ratio, (vi) Nuclear pleomorphism. 17 The presence of 3 or more of these criteria supports a diagnosis of malignant GCT. 17 Our patient’s GCT diagnosis was of benign nature as none of the above criteria were present. The expression of p53 and a high Ki67 index are associated with an aggressive clinical course.18 In benign tumors, the p53 is negative, and Ki67 is under 1%, while the malignant cases express p53 and Ki67 in 10% and 30% of cells, respectively. 18 According to the current literature, local recurrence is related to insufficient surgical excision.19 The recurrence rate after wide local excision with negative and positive surgical margins is 2-8% and >20%, respectively.20 Therefore, assessment and reporting of surgical margin is crucial to lessen recurrences.20 In our case, the lack Autops Case Rep (São Paulo). 2019 Jul-Sep;9(3):e2019099 of expression of p53, the Ki67 index less than 1%, and the negative surgical margins contributed to the lack of tumor relapse to date. CONCLUSION Although rare, granular cell tumor of the breast should be included in the differential diagnosis in pediatric patients presenting with a breast mass. It should be remembered that the tumor may arise in different locations. Cytopathologists should be aware of the pitfalls in diagnosing GCT on FNAC. Proper diagnosis and timely management of this tumor are essential due to its malignant potential (<2% of cases) and its high local recurrence with suboptimal excision. Total excision with wide margins is vital for treatment success. REFERENCE 1. Bean SM, Eloubeidi MA, Eltoum IA, Cerfolio RJ, Jhala DN, Preoperative diagnosis of a mediastinal granular cell tumour by EUS-FNA: a case report and review of the literature. Cytojournal. 2005;2(8):1-8. 2. Diniz G, Vuruskaner H, Celebiler H, Ortac R, Aktas S, Ozener V. On üç yaşında bir kız çocukta, uyluk yumuşak dokusunda lokalize granüler hücreli tümör olgusu. Türkiye Ekopatoloji Dergisi. 2005;11(1):33-6. 3. Nasier H, Danfort RD Jr, Sunbuli M, Dimitrijevic O. Malignant granular cell tumour: case report with a novel karyotype and review of the literature. Ann Diagn Pathol. 2010;14(4):273-8. http://dx.doi.org/10.1016/j. anndiagpath.2009.08.004. PMid:20637434. 4. R a v i c h A , S t o u t A P , R a v i c h R A . M a l i g n a n t granular cell myoblastoma involving the urinary bladder. Ann Surg. 1945;121(3):361-72. http:// dx.doi.org/10.1097/00000658-194503000-00010. PMid:17858577. 5. Rexeena B, Paul A, Nitish RA, Kurian C, Anila RK. Granular cell tumour of breast: a case report and review of literature. Indian J Surg Oncol. 2015;6(4):446-8. 6. Becelli R, Perugini M, Gasparini G, Cassoni A, Fabiani F. Abrikissoff’s tumour. J Craniofac Surg. 2001;12(1):78-81. http://dx.doi.org/10.1097/00001665-200101000-00013. PMid:11314193. 7. Apisarnthanarax P. Granular cell tumor. An analysis of 16 cases and review of the literature. J Am Acad Dermatol. 1981;5(2):171-82. http://dx.doi.org/10.1016/S01909622(81)70085-9. PMid:6267110. 5-6 Pediatric Granular Cell Tumor of the Breast: An uncommon neoplasm in an uncommon site and age group 8. Adeniran A, Al-Ahmadie H, Mahoney MC, RobinsonSmith TM. Granular cell tumor of the breast: a series of 17 cases and review of the literature. Breast J. 2004;10(6):528-31. http://dx.doi.org/10.1111/j.1075122X.2004.21525.x. PMid:15569210. 9. Yang WT, Edeiken-Monroe B, Sneige N, Fornage BD. Sonographic and mammographic appearances of granular cell tumors of the breast with pathological correlation. J Clin Ultrasound. 2006;34(4):153-60. http:// dx.doi.org/10.1002/jcu.20227. PMid:16615051. 10. De Simone N, Aggon A, Christy C. Granular cell tumor of the breast: clinical and pathologic characteristics of a rare case in a 14-year-old girl. J Clin Oncol. 2011;29(22):e6567. http://dx.doi.org/10.1200/JCO.2011.35.9448. PMid:21646617. 11. Marshall AP, Spottswood SE, Grau AM, Jackson JP. Juvenile fibroadenoma and granular cell tumor of the breast in an adolescent. J Pediatr Surg. 2012;47(10):19303. http://dx.doi.org/10.1016/j.jpedsurg.2012.07.050. PMid:23084210. 12. Heinzerling NP, Koehler SM, Szabo S, Wagner AJ. Pediatric granular cell tumor of the breast: a case report and review of the literature. Case Rep Surg. 2015;2015:568940. http://dx.doi. org/10.1155/2015/568940. PMid:26491597. 13. Omar L, Pfeifer CM, Kulkarni S, Sharma P, Sengupta A, Kwon JK. Granular cell tumor in a premenstrual female breast. Clin Imaging. 2018;52:334-6. http://dx.doi. org/10.1016/j.clinimag.2018.09.003. PMid:30241035. 14. Chung S, Noh WC. Fine needle aspiration cytology of granular cell tumor in breast- a case report. Korean J Cytopathol. 2007;8(2):157-60. 15. Stefansson K, Wollmann RL. S-100 protein in granular cell tumours (granular cell myoblastomas). Cancer. 1982;49(9):1834-8. http://dx.doi. org/10.1002/1097-0142(19820501)49:9<1834::AIDCNCR2820490916>3.0.CO;2-G. PMid:6280846. 16. Bellezza G, Colella R, Sidoni A, et al. Immunohistochemical expression og Galectin-3 and HMBE-1 in granular cell tumours: a new finding. Histol Histopathol. 2008;23(9):1127-30. PMid:18581283. 17. Fanburg-Smith JC, Meis-Kindblom JM, Fante R, Kindblom LG. Malignant granular cell tumour of soft tissue; diagnostic criteria and clinicopathologic correlation. Am J Surg Pathol. 1998;22(7):779-94. http:// dx.doi.org/10.1097/00000478-199807000-00001. PMid:9669341. 18. Le BH, Boyer PJ, Lewis JE, Kapadia SB. Granular cell tumour. Arch Pathol Lab Med. 2004;128(7):771-5. PMid:15214825. 19. Abraham T, Jackson B, Davis I, Yu J, Peterson C. Mohs surgical treatment of a granular cell tumour on the toe of a child. Pediatr Dermatol. 2007;24(3):235-7. http://dx.doi. org/10.1111/j.1525-1470.2007.00392.x. PMid:17542870. 20. Lack EE, Worsham GF, Callihan MD, et al. Granular cell tumour: a clinicopathologic study of 110 patients. J Surg Oncol. 1980;13(4):301-16. http://dx.doi.org/10.1002/ jso.2930130405. PMid:6246310. Author contributions: Baishya P, Das J and Dey B wrote the manuscript. Darilin S and Khonglah Y have reported the cytopathology. Marbaniang E and Raphael V reported the histopathology. Lynser D reported the radiology. All authors collectively proofread the manuscript and approved the final version for publication. The authors retain the informed assent signed by the parents of the patient authorizing the data publication. The manuscript publication is by the Institutional Ethics committee. Conflict of interest: None Financial support: None Submitted on: January 31st, 2019 Accepted on: April 3rd, 2019 Correspondence Evarisalin Marbaniang Department of Pathology - North Eastern Indira Gandhi Regional Institute of Health and Medical Sciences (NEIGRIHMS) Mawdiangdiang, Shillong, Meghalaya, India 793018 Phone: +91 883740-9267 evarisalinmarbaniang@yahoo.com 6-6 Autops Case Rep (São Paulo). 2019 Jul-Sep;9(3):e2019099
https://openalex.org/W2165717238
https://europepmc.org/articles/pmc4059088?pdf=render
English
null
A task-oriented circuit training in multiple sclerosis: a feasibility study
BMC neurology
2,014
cc-by
8,412
© 2014 Straudi et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: The aim of this study was to evaluate the safety, feasibility and preliminary effects of a high-intensity rehabilitative task-oriented circuit training (TOCT) in a sample of multiple sclerosis (MS) subjects on walking competency, mobility, fatigue and health-related quality of life (HRQoL). Methods: 24 MS subjects (EDSS 4.89 ± 0.54, 17 female and 7 male, 52.58 ± 11.21 years, MS duration 15.21 ± 8.68 years) have been enrolled and randomly assigned to 2 treatment groups: (i) experimental group received 10 TOCT sessions over 2 weeks (2 hours/each session) followed by a 3 months home exercise program, whereas control group did not receive any specific rehabilitation intervention. A feasibility patient-reported questionnaire was administered after TOCT. Functional outcome measures were: walking endurance (Six Minute Walk Test), gait speed (10 Meter Walk Test), mobility (Timed Up and Go test) and balance (Dynamic Gait Index). Furthermore, self-reported questionnaire of motor fatigue (Fatigue Severity Scale), walking ability (Multiple Sclerosis Walking Scale – 12) and health-related quality of life (Multiple Sclerosis Impact Scale – 29) were included. Subjects’ assessments were delivered at baseline (T0), after TOCT (T1) and 3 months of home-based exercise program (T2). Results: After TOCT subjects reported a positive global rating on the received treatment. At 3 months, we found a 58.33% of adherence to the home-exercise program. After TOCT, walking ability and health-related quality of life were improved (p < 0.05) with minor retention after 3 months. The control group showed no significant changes in any variables. Conclusions: This two weeks high-intensity task-oriented circuit class training followed by a three months home-based exercise program seems feasible and safe in MS people with moderate mobility impairments; moreover it might improve walking abilities. Trial registration: NCT01464749 Keywords: Task oriented, Rehabilitation, Mobility, Multiple sclerosis A task-oriented circuit training in multiple sclerosis: a feasibility study Sofia Straudi1,2*†, Carlotta Martinuzzi1†, Claudia Pavarelli1, Amira Sabbagh Charabati1†, Maria Grazia Benedetti3, Calogero Foti2, Michela Bonato1†, Eleonora Zancato1† and Nino Basaglia1 Straudi et al. BMC Neurology 2014, 14:124 Straudi et al. BMC Neurology 2014, 14:124 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Open Access * Correspondence: s.straudi@ospfe.it †Equal contributors 1Neuroscience and Rehabilitation Department, Ferrara University Hospital, Ferrara, Italy 2Doctoral Program in Advanced Sciences and Technologies in Rehabilitation Medicine and Sports, Tor Vergata University, Rome, Italy Full list of author information is available at the end of the article Background participation but also to quality of life [3,4]. There are several studies showing a benefit of physical therapy on walking function, mobility [5-9] and fatigue [10-12]; for this reason recent reviews and international guidelines recommend exercise for MS population [13,14]. None- theless, the task-oriented circuit training (TOCT) ap- proach hasn’t been studied enough in people with MS (PwMS) and gait and mobility disorders. TOCT is based on workstations that reproduce physical activities that the subject usually performs during daily living (i.e. walking, climbing stairs, maintain balance) with the aim of pro- moting motor learning and task retention. In addition, a major characteristic of this rehabilitative intervention Multiple sclerosis (MS) is a chronic and progressive neurologic disease that commonly affects young adults worldwide and it represents a major cause of disability over time [1,2]. In this population gait and mobility impairments could have a negative impact on personal activities not only restricted to motor domains and * Correspondence: s.straudi@ospfe.it †Equal contributors 1Neuroscience and Rehabilitation Department, Ferrara University Hospital, Ferrara, Italy 2Doctoral Program in Advanced Sciences and Technologies in Rehabilitation Medicine and Sports, Tor Vergata University, Rome, Italy Full list of author information is available at the end of the article Page 2 of 9 Page 2 of 9 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 TOCT (experimental group) or usual care (control group) through a block randomization approach. Eligible subjects were randomly assigned into a 1:1 ratio to receive experi- mental treatment (task-oriented circuit training), or no spe- cific treatment (usual care). The randomization scheme was generated by using the web site Randomization.com (http://www.randomization.com). The experimental group received 10 task-oriented training sessions (Monday-Friday) over 2 weeks; each session lasted 2 hours. Control group did not receive any intervention. is the exercise intensity that, compared to a conven- tional physiotherapy session, is nearer to the number of repetitions needed to achieve and maintain motor learning of these movements [15]. Moreover, TOCT has been designed for small groups of subjects requir- ing only one physiotherapist compared with conven- tional rehabilitation session in which the ratio is 1:1, giving also the advantage of reducing costs of rehabili- tation. Previous studies [16,17] have shown that TOCT is a good method to improve locomotor function and mobility in stroke survivors. Methods This is a single-blind randomized controlled pilot trial. Subjects were recruited at the outpatient clinic of the Physical Medicine and Rehabilitation Department (Fer- rara University Hospital). Informed written consent was obtained from eligible subjects. The research study has been reviewed and approved by the Ferrara University Hospital Ethics Committee and registered in ClinicalTrial. gov database (NCT01464749). The inclusion criteria were: males and females, age 18 to 75, diagnosis of MS (primary or secondary progressive, relapsing-remitting), without re- lapses in the preceding 3 months, mild to moderate gait impairments referred to Expanded Disability Status Scale (EDSS) score between 4 and 5.5. Subjects were able to walk for at least 100 meters with no constant assistance (cane, crutch or brace) required. The exclusion criteria were: other conditions that may affect motor function, impaired cognitive functioning (Mini Mental Status Exam- ination score less than 24). Subjects were randomized to Home-based training program After the supervised 2 weeks, a home- exercise illus- trated brochure was given to subjects so that they could independently train for the following 3 months. It in- cluded similar exercises that subjects learned during the 2 weeks, gait training (overground or treadmill), stretch- ing and strengthening exercises. We suggested an inde- pendent home training 3 times/week (60 minutes/each session). Our principal aim was to retain positive effects on walking competency and function. Subjects were asked to record in a diary the intensity and duration of exercise; they were allowed to call hospital to have fur- ther information and feedbacks. Usual care (UC) The control group (UC) did not receive any specific re- habilitation treatment for gait performance and mobility improvement. During the entire study, both groups were authorized, at will, to exercise in non-rehabilitative contexts. Task-oriented circuit training OC l d d d ff Task oriented circuit training TOCT included six different workstations in which sub- jects exercised for 5 minutes in each one (3 minutes of exercises and 2 minutes of rest). During each session, subjects underwent 2 laps that took about 60 minutes (6 workstation × 5 minutes × 2 laps), with 10 minutes of rest after each lap. In addition, walking endurance was trained by 30 minutes walking on the treadmill including rests if necessary. For further details on training protocol see Table 1. This was a progressive circuit and subjects while exercising received feedbacks (visual and auditory) by the physiotherapist. Rests were used to discuss about difficulties and to provide further feedbacks. One ses- sion included up to 3 patients and lasted 120 minutes, 5 days/week for 2 weeks. Background The rationale of our study is based on growing evi- dences [18] that demonstrate how human adult brain is capable of significant adaptations providing that the quantity (duration and frequency) and quality (task- specificity) of interventions are appropriate to facilitate enhanced neural reorganization and upper limb motor recovery in chronic stroke survivors [19]. Also, recent neuroimaging studies highlighted how adaptive brain plasticity [20,21] and motor learning [22] are preserved even in progressive MS subjects and rehabilitative strat- egies that harness cortical reorganization of motor func- tion might be an option even in this chronically disabled subjects. Task oriented circuit training (TOCT) is an intensive task-specific intervention, for this reason we designed a pilot randomized-controlled trial: (i) to test the safety, feasibility and acceptability of a two weeks high-intensity task oriented circuit training in multiple sclerosis subjects with moderate gait impairments; (ii) to assess its preliminary effects on walking, mobility, fa- tigue and quality of life; (iii) to evaluate the adherence to a three months home based training program that followed TOCT; (iv) to determine TOCT effect sizes and the sample needed for a future RCT study. 10 Meter Walk Test (10MWT) It assesses walking speed over 10 meters. It has been tested on different kind of subjects (both with neurolo- gical and orthopaedics conditions), including MS [23,24]. The subject walks in total for 14 meters: 2 m for acceler- ation and 2 m for deceleration, gait speed is calculated for the 10 m distance between them. The test is performed three times and best trial is used for analysis. Fatigue Severity Scale (FSS) to treatment (T0), the week after the end of TOCT (T1) and the 3 months home-based training program (T2). Fatigue was monitored through a questionnaire, Fa- tigue Severity Scale (FSS) which evaluates fatigue in MS and other conditions [29]. Essentially, the FSS consists of answering a short questionnaire that re- quires the subjects to rate their level of fatigue. Sub- jects were asked to read each statement and circle a number from 1 to 7, depending on how appropriate they felt the statement applied to them over the pre- ceding week. Multiple Sclerosis Impact Scale – 29 (MSIS-29) h h l h d l f l f This is an health-rated quality of life questionnaire that assesses the impact of MS on physical and psycho- logical functions. It is formed by 29 items on ADL I and II: 20 about physical activity and 9 of psychological status of the person. Each item can be scored with a value from 0 to 5; total score is given by the sum of all the items and then is transformed in a range from 0 to 100. A higher value correspond to a worse perception of subject’s HRQoL [31]. Outcome measures Outcome measures were both functional tests (gait speed, walking endurance, mobility, balance) and self-reported questionnaires (fatigue, health-related quality of life and walking ability). All measures were assessed the week prior Page 3 of 9 Page 3 of 9 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Table 1 Task-oriented circuit training description Patients per group 3 N. of therapists 1 experienced physiotherapist Intensity (I) I: 2 weeks, 5 days per week, 120 minutes Progression (P) P: increasing the number of repetitions completed in 3 minutes at a workstation and increasing treadmill speed Exercises Workstations: 1: Step (the patient goes up and down a 20 cm step both with left and right foot) 2: Slalom (the patient kicks a ball walking through a slalom exercise, formed by 4 cones that are 1 m distant from each other) 3: Tandem exercise (the patient walks in tandem using a line as a guide. If he/she is not able to put one foot in front of the other, it is allowed to do a wider and longer step as long as it is challenging) 4: Goals (the patient must touch with the tip of the foot a goal that is positioned on a mirror in front of him. If necessary, he can use a lateral support) 5: Obstacles (the patient must pass 5 obstacles, 3 of which are 5 cm high) 6: Long step (the patient must walk performing long steps (at least 40-50 cm long) using some signs on the floor as a guide) Subjects exercise for 3 minutes at each workstation followed by a 2 minutes break 7: Treadmill (the patient should walk up to 30 minutes at speed between 0.9 and 2.9 km/h. The speed is self selected by the subject, who can also take a break if she/he needs to) : Treadmill (the patient should walk up to 30 minutes at speed between 0.9 and 2.9 km/h. The speed is self selected by the ubject, who can also take a break if she/he needs to) Multiple Sclerosis Walking Scale – 12 (MSWS-12) Th h f S p g This questionnaire assesses the impact of MS on walking ability. It is formed by 12 items, asking the patient about his perception on gait speed, running, confidence as- cending/discending stairs, balance and fatigue. As in the MSIS 29, the total score is obtained by the sum of the score of each item (0-5) and then transformed into a value from 0 to 100. Again, a higher score represents a major walking disability [30]. Walking endurance was measured with SMWT, a feas- ible, reproducible and reliable measure in MS subjects [25]. Subjects were instructed to walk up and down a 20m walkway as far as possible in 6 minute. Timed Up and Go test (TUG) Subjects were given verbal instruction to stand up from a chair, walk 3 meters, cross a line marked on the floor, turn around, walk back, and sit down. Subjects per- formed 3 trials and their best performance was used for analysis [26]. Biostatistical analysis Descriptive statistic (mean, median, standard deviation, percentiles and confidence interval) was used to describe sample at T0, T1, T2. Baseline characteristics (sex, age, MS type, MS duration, EDSS score) and clinical tests (SMWT, 10 mWT, TUG, DGI, FSS, MSIS-29 and MSWS- 12) were compared among groups to assess the quality of randomization. Graphics frequency and Shapiro-Wilk Test were performed to assess variance in order to fulfill the essential assumptions for parametric statistical analysis for continuous variables. To try reducing Type I error, since variables were normally distributed, we choose ANOVA repeated measures for continuous variables (SMWT, 10 mWTand TUG) and Friedman test [32] for ordinal vari- ables (DGI, MSIS-29 and MSWS-12, FSS) to test within- group differences. Post-hoc analysis have been performed when p < 0.05. Differences in mean changes from baseline (T0) after TOCT (T1) and 3 months home-based exercise (T2) were tested with un-paired t-test or Mann Whitney Test among groups. In order to calculate a sample size for a larger RCT, we used d (Cohen) [33]. An intention-to-treat analysis was carried out, i.e. in case of missing data (lost to follow-up) the last available value was considered (last observation carried forward approach). Statistical analysis All subjects completed the 2 weeks training, while we lost 3 subjects at 3 months follow-up in UC group. The clinical study flow-chart has been reported in Figure 1. Feasibility patient-reported questionnaire At the end of the 2 weeks, subjects who underwent TOCT received a questionnaire formed by twelve questions on their opinions and acceptability referred to the treatment received (for further details see Additional file 1). Results Twenty-four PwMS have been assessed for eligibility in an outpatient rehabilitation clinic (Ferrara University Hospital). After the preliminary visit, all subjects (17 female, 7 male, age 52.58 ± 11.21, 15.21 ± 8.68 years from MS onset) were included in the study. They were randomized into groups: twelve were allocated to TOCT and twelve to UC. Demographic and clinical characteristics are summa- rized in Table 2. 37.5% of subjects received disease- modifying therapy (DMT) during the entire study (8.33% immunosuppressive treatments, 16.66% inter- feron beta, 4.16% glatiramer acetate and 8.33% fingolimod); 20.83% of them received therapies for MS symptoms man- agement (i.e. spasticity, depression, pain). We didn’t report significant differences in DMT or symptomatic therapies among groups at baseline or any changes in pharmaco- logical treatments during the study. Dynamic Gait Index (DGI) It was developed as a clinical tool to assess gait, balance and risk of fall. It doesn’t evaluate only usual steady- state walking, but also walking during more challenging task (i.e. cross obstacles, slalom). Eight functional walk- ing tests are performed by the subject and scored out of three (maximum total score is 24). Scores of 19 or less have been related to an increased risk of falls [27,28]. Page 4 of 9 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 was performed with STATA software, statistical soft- ware v12.1 college station, Texas. Significance level was set at 0.05. nted circuit training. UC = usual care. MS = multiple sclerosis. RR = relapsing remitting. PP = primary progressive. SP = secondary progres d Disability Status Scale. Functional tests (walking endurance, speed, balance, mobility) A repeated-measures analysis of variance for SMWT, 10MWT and TUG and Friedman test for DGI revealed that TOCT and UC didn’t produce a significant Table 2 Sample characteristics n All subjects TOCT UC p value* Total 24 12 12 Gender (male/female) 7/17 5/7 2/10 0.18 MS type 0.56 RR 6 4 2 PP 10 5 5 SP 8 3 5 Mean (SD) median All subjects TOCT UC p value** Age (years) 52.58 (11.21) 49.92 (7.51) 55.25 (13.82) 0.16 53.50 51.5 58 MS duration (years) 15.21 (8.68) 12.16 (6.91) 18.25 (9.46) 0.09 15 10 18 EDSS score 4.89 (0.54) 4.95 (0.61) 4.83 (0.49) 0.56 5 5 4.75 TOCT = task-oriented circuit training. UC = usual care. MS = multiple sclerosis. RR = relapsing remitting. PP = primary progressive. SP = secondary progressive. EDSS = Expanded Disability Status Scale. *Pearson Chi-square. **Mann-Whitney test. Table 2 Sample characteristics Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Straudi et al. BMC Neurology 2014, 14:124 Page 5 of 9 y http://www.biomedcentral.com/1471-2377/14/124 Assessed for eligibility (n=24) Excluded (n=0) Allocated to Usual Care group (n=12) Received allocated intervention (n=12) Allocated to TOCT group (n=12) Received allocated intervention (n=12) Lost to follow-up (n=0) Lost to follow-up (denial) (n=3) Analyzed (n=12) Analyzed (n=12) Randomized (n=24) Allocation 1:1 Figure 1 Study flow diagram. Assessed for eligibility (n=24) Allocated to TOCT group (n=12) Received allocated intervention (n=12) Allocated to Usual Care group (n=12) Received allocated intervention (n=12) Figure 1 Study flow diagram. tandem exercise workstation (25%) as much more en- joyable, while goals workstation as much more boring (33.33%). improvement over time. Between-group differences were found in SMWT after TOCT (p < 0.05). Descrip- tive statistics are shown in Table 3. Patient-reported questionnaires (fatigue, walking ability, health-related quality of life) At 3 months, 7/12 subjects (58.33%) completed their diary with home-exercise program and returned them to investigators. Considering the SMWT, we didn’t find fur- ther gains in subjects who completed the 3 months exer- cise program (good adherence sub-group) compared with subjects who didn’t report any information on their home exercise training (poor adherence sub-group). Specifically, at three months the good adherence sub- group (n = 7) decreased by -13.28 ± 6.72 m compared to post-TOCT, while the poor adherence sub-group (n = 5) of - 4.64 ± 32.37 m (n.s.). MSIS-12 scores were improved in TOCT group (F(2,22) = 3.81, p < 0.05). MSIS-29 physical (F(2,22) = 3.66, p < 0.05) and psychological (F(2,22) = 3.55, p < 0.05) score improved in TOCT group and decreased in UC group (F(2,22) = 3.66, p < 0.05; F(2,22) = 4.28, p < 0.05) over time. Multiple compar- isons were significant T0-T1 and T0-T2 in both groups. Between-group differences were found in MSWS-12 and MSIS-29 physical and psychological scores after TOCT (p < 0.01) and in MSIS-29 psychological score even after 3 months of home-based exercise (p < 0.05). Descriptive statistics are shown in Table 4. Feasibility patient-reported questionnaire Our preliminary results allowed us to define a sample size for a following RCT study. Considering walking en- durance as primary outcome (SMWT), we found a TOCT effect size (d) of 0.85 in multiple sclerosis sub- jects (EDSS < 6). This value is based on a walking endur- ance improvement in TOCT group of 20.05 ± 25.06 m, compared to -9.95 ± 42.84 m in control group. Given equal allocation (1:1) between treatment and control arms, ad using 80% power and alpha of 5%, we would need 46 subjects (23 TOCT + 23 usual care) to complete the study. Conservatively, we expect a 10% rate of drop- out, thus the sample size will be increased by 10% to 51 subjects. y p p q All subjects (n = 12) completed the questionnaire. On a Visual Analog Scale (VAS) 0-10 they reported a positive global rating on treatment received (9.1), considered the time duration adequate (8.5), the goals were achievable (7.6), the feedbacks provided during sessions were easy to understand (9.9), they enjoyed the training (8.4) and were satisfied by their own results (8.7). At the end of each session they felt physically tired (7), however they didn’t report the need of extra rests (0.25). The majority identified the treadmill (50%) as the exercise where they had best results whereas goals workstation (41.6%) where they had their worst; they considered slalom and Straudi et al. Feasibility patient-reported questionnaire BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Page 6 of 9 Page 6 of 9 Table 3 Clinical results Table 3 Clinical results TOCT (n=12) UC (n=12) Mean (SD) Median (25th-75th Percentile) 95% CI Mean (SD) Median (25th-75th Percentile) 95% CI SMWT (m) Pre- TOCT 290.67 (72.92) 283.40 (215.70 – 355.6) 244.33 – 337 285.72 (79.15) 298.40 (230.00 – 329.80) 235.42 – 336.01 Post- TOCT 310.72 (73.73) 316.20 (237.50 – 371.60) 263.87 – 357.57 275.77 (70.65) 293.40 (227.80 – 333) 230.88 – 320.66 3 months follow-up 301.03 (72.58) 308.50 (230.80 – 359.80) 254.92 – 347.15 258.73 (100.80) 274.30 (167.8 – 312.8) 194.69 – 322.78 TUG (s) Pre- TOCT 10.64 (2.48) 9.89 (9.06 – 11.85) 9.06 – 12.21 12.10 (4.00) 11.19 (9.24 – 15) 9.56 – 14.63 Post- TOCT 10.05 (1.89) 9.55 (8.56 – 10.93) 8.85 – 11.25 12.23 (3.78) 11.61 (9.27 – 14.07) 9.82 – 14.63 3 months follow-up 10.50 (2.03) 9.56 (9.08 – 11.98) 9.20 – 11.78 12.43 (3.92) 12.25 (9.89 – 14.22) 9.94 – 14.93 10MWT (m/s) Pre- TOCT 1.08 (0.23) 1.11 (0.94 – 1.20) 0.94 – 1.23 1.03 (0.30) 0.99 (0.77– 1.19) 0.84 – 1.22 Post- TOCT 1.13 (0.24) 1.10 (1.00 – 1.29) 0.98 – 1.28 1.02 (0.23) 1.03 (0.81 – 1.15) 0.87 – 1.16 3 months follow-up 1.15 (0.19) 1.19 (1.01 – 1.30) 1.02 – 1.27 0.99 (0.30) 0.87 (0.82 – 1.14) 0.80 – 1.19 DGI Pre- TOCT 16.83 (3.56) 17 (15.5 – 19.5) 14.57 – 19.10 14.75 (4.92) 14 (12.5 – 18) 11.62 – 17.88 Post - TOCT 17.67 (3.11) 18 (16 – 20) 15.69 – 19.65 14.75 (4.14) 16 (12 – 18) 12.12 – 17.38 3 months follow-up 17.75 (3.93) 18 (14.5 – 10.5) 15.25 – 20.25 13.75 (3.86) 14 (11.5 – 17) 11.29 – 16.21 TOCT = task-oriented circuit training; UC = usual care; SMWT = Six Minute Walk Test; TUG = Timed Up and Go test; 10MWT = 10 Meter Walk Test; DGI = Dynamic Gait Index. includes several principles of experience-dependent neu- roplasticity such as specificity, repetition and intensity [18]; for this reason we hypothesized that a sample of PwMS with moderate mobility impairments might bene- fit from such intervention. Previous studies on TOCT nted circuit training; UC = usual care; MSIS-29 Psy and Phys = Multiple Sclerosis Impact Scale -29 (psychological and physical functions); ple Sclerosis Walking Scale – 12; FSS = Fatigue Severity Scale. CT = task-oriented circuit training; UC = usual care; MSIS-29 Psy and Phys = Multiple Sclerosis Impact Scale -29 (psychological and physi WS-12 = Multiple Sclerosis Walking Scale – 12; FSS = Fatigue Severity Scale. Discussion We didn’t re- port any injuries as a consequence of falls during TOCT sessions or marked fatigability which required the interrup- tion of the exercise; all twelve subjects concluded the 10 sessions with no drop-outs. Furthermore, looking at the feasibility patient-reported questionnaire results, we might conclude that this is a suitable intervention in this popula- tion. To test TOCT effects on walking function we did both short-distance (10MWT) than long-distance (SMWT) walking tests in addition to a patient-reported measure of walking ability (MSWS-12). Long-distance walking test and patient-reported walking ability questionnaires were consid- ered as the best measures to detect changes after rehabi- litation in MS subjects [37]. Furthermore, long-distance walking tests, such as SMWT better reflect walking per- formance and functional capacity in MS population [38]. A good predictor for independency and mobility in PwMS is walking endurance measured by SMWT: it has been shown that subjects have 1% increased probability of being inde- pendent for every 1 m longer they walk during SMWT [23]. In our sample we didn’t find significant effects on gait speed measured with 10MWT probably because short- distance more than long-distance walking test are due to variability. In fact, a recent study reported a within-day vari- ability of >20% for the 10MWT when baseline gait velocity was <1.2 m/s [39]. Another reason would be that we didn’t use the 25 foot walking test, which seems to be more reli- able than 10MWT in detecting changes in PwMS [40]. Our findings in SMWT gains were close to the minimally im- portant change from a patient prospective stated as 22 m [37]. In the same way MSWS-12 score improved by 11 points, showing beneficial effects on walking ability; this pa- tient–reported outcome measure investigates limitation on lki biliti d t MS [30] d t b li bl i In our sample we found a lower adherence to home- exercise program (58%) compared to other studies in MS subjects [43-45]. We can assume that giving a diary with a home-based exercise program wasn’t sufficient to motivate subjects to continue with task-oriented exer- cises at home. We can also speculate that task-oriented training is less suitable at home compared to other intervention tested in the aforementioned studies (mostly resistance exercises). Moreover, depression or cognitive symptoms, that are common in MS, might have reduced subjects’ compliance to rehabilitation, as clearly highlighted for MS medications [46]. Discussion This is a proof-of-concept study with the aim of opti- mizing components of a novel rehabilitative intervention for PwMS with moderate mobility impairments. We proposed a high-intensity task-specific training which Table 4 Questionnaire results TOCT (n = 12) UC (n = 12) Mean (SD) Median (25th-75th percentile) 95% CI Mean (SD) Median (25th-75th Percentile) 95% CI MSIS-29 Psy (0-100) Pre- TOCT 51.66 (20.61) 53.33 (32.22 – 64.44) 38.57 – 64.76 49.44 (20.12) 38.88 (36.11 – 72.22) 36.65 – 62.23 Post- TOCT 41.85 (15.73) 42.22 (30.00 – 46.66) 31.85 – 51.85 53.14 (19.24) 47.77 (40.00 – 72.77) 40.92 – 65.37 3 months follow-up 42.96 (16.20) 43.33 (29.44 – 56.11) 32.66 – 53.25 53.70 (16.43) 50.00 (42.77 – 65.00) 43.26 – 64.14 MSIS-29 Phys (0-100) Pre- TOCT 50.83 (15.60) 53.00 (45.50 – 63.50) 40.92 – 60.74 51.58 (18.67) 50.00 (36.25 – 68.50) 39.71 – 63.44 Post- TOCT 44.41 (11.32) 43.00 (39.25 – 50.50) 37.22 – 51.61 53.75 (18.10) 52.00 (38.50 – 70.25) 42.24 – 65.25 3 months follow-up 49.16 (11.00) 48.00 (41.25 – 58.25) 42.17 – 56.15 53.00 (22.28) 49.00 (37.00 – 71.50) 38.83 – 67.16 MSWS-12 (0-100) Pre- TOCT 63.06 (14.00) 66.67 (57.50 – 71.67) 54.16 – 71.95 61.94 (18.28) 60 (47.5 – 76.67) 50.33 – 73.56 Post- TOCT 52.36 (14.06) 53.33 (42.5 – 62.5) 43.43 – 61.29 70.83 (17.08) 75 (55 – 82.5) 59.98 – 81.68 3 months follow-up 65.42 (16.04) 63.33 (55 – 76.67) 55.23 – 75.61 71.11 (20.34) 74.17 (55.83 – 88.33) 58.19 – 84.04 FSS Pre- TOCT 5.43 (1.23) 5.22 (4.94 – 6.50) 4.65 – 6.22 5.79 (0.98) 6.22 (4.77 –6.55) 5.17 – 6.42 Post- TOCT 5.05 (1.52) 5.44 (4.59 – 6.11) 4.09 – 6.02 6.11 (0.68) 6.11 (5.44 – 6.78) 5.68 – 6.54 3 months follow-up 5.63 (0.78) 5.78 (5.09 – 6.17) 5.13 – 6.13 6.01 (0.91) 6.28 (5.22 – 7) 5.51 – 6.66 TOCT = task-oriented circuit training; UC = usual care; MSIS-29 Psy and Phys = Multiple Sclerosis Impact Scale -29 (psychological and physical functions); MSWS-12 = Multiple Sclerosis Walking Scale – 12; FSS = Fatigue Severity Scale. Table 4 Questionnaire results Page 7 of 9 Page 7 of 9 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 highlighted positive effects on restoring mobility in stroke survivors including a wide range of exercise doses and intensities [16]. Discussion To our knowledge (except for tread- mill, body-weight support treadmill and robot-assisted gait training [34]), the only attempts to apply a task- oriented training for mobility impairments in PwMS were done by Mark et al. [35] who proposed a Constraint- Induced Movement Therapy (CIMT) for the lower extrem- ities and Salbach et al. [36] who tested a community exer- cise program for people with stroke, acquired brain injury and MS. CIMT was administered to a small sample (n = 4) showing positive effects on real-world lower extremity use up to four years. Compared to our TOCT, CIMT protocol was much more intense (3.5 h/d for 3 wk), trained 15 dif- ferent tasks (compared to 7) and a “transfer package” to de- liver gains into the real world was administered [35]. The community exercise program was delivered to 2 MS sub- jects twice/week for 12 weeks with no effects on walking endurance [36]. Our two weeks high-intensity task oriented circuit training was feasible and safe in a small group of PwMS with moderate mobility impairments. We didn’t re- port any injuries as a consequence of falls during TOCT sessions or marked fatigability which required the interrup- tion of the exercise; all twelve subjects concluded the 10 sessions with no drop-outs. Furthermore, looking at the feasibility patient-reported questionnaire results, we might conclude that this is a suitable intervention in this popula- tion. To test TOCT effects on walking function we did both short-distance (10MWT) than long-distance (SMWT) walking tests in addition to a patient-reported measure of walking ability (MSWS-12). Long-distance walking test and patient-reported walking ability questionnaires were consid- ered as the best measures to detect changes after rehabi- litation in MS subjects [37]. Furthermore, long-distance walking tests, such as SMWT better reflect walking per- formance and functional capacity in MS population [38]. A good predictor for independency and mobility in PwMS is walking endurance measured by SMWT: it has been shown that subjects have 1% increased probability of being inde- pendent for every 1 m longer they walk during SMWT [23]. In our sample we didn’t find significant effects on gait speed measured with 10MWT probably because short- distance more than long-distance walking test are due to variability. In fact, a recent study reported a within-day vari- ability of >20% for the 10MWT when baseline gait velocity was <1.2 m/s [39]. Discussion However, we didn’t find any differences in walking endurance deterioration after TOCT among good- and poor-adherence subgroups. Finally, as long walking tests are more appropriate in detecting clinically meaningful improvement after phys- ical rehabilitation in MS population [37,38], we calcu- lated a sample size for a future RCT study based on the SMWT gains (51 subjects). Discussion Another reason would be that we didn’t use the 25 foot walking test, which seems to be more reli- able than 10MWT in detecting changes in PwMS [40]. Our findings in SMWT gains were close to the minimally im- portant change from a patient prospective stated as 22 m detecting changes after rehabilitation with a minimally im- portant change of 10.39 points [37]. Considering how health-related quality of life is strongly reduced in PwMS [4], we found positive effects of TOCT on both physical and psychological functions in MSIS-29; however we have to consider how these results are within the standard error of the measurement which is defined as 5.3-9.5 points for physical domain and 7.6-13 points for psychological do- main [41,42]. Examining fatigue which is a common MS symptom with a negative impact on mobility, activities and quality of life, in our sample we didn’t find any detrimental effects of the task-oriented circuit training (2h/d over 2 weeks). Accordingly with previous studies [10-12] we highlighted a slight fatigue reduction after TOCT. Never- theless, these preliminary results should be taken cau- tiously, considering that this is a pilot study with the aim to test the safety and feasibility of a task-oriented circuit train- ing in PwMS. highlighted positive effects on restoring mobility in stroke survivors including a wide range of exercise doses and intensities [16]. To our knowledge (except for tread- mill, body-weight support treadmill and robot-assisted gait training [34]), the only attempts to apply a task- oriented training for mobility impairments in PwMS were done by Mark et al. [35] who proposed a Constraint- Induced Movement Therapy (CIMT) for the lower extrem- ities and Salbach et al. [36] who tested a community exer- cise program for people with stroke, acquired brain injury and MS. CIMT was administered to a small sample (n = 4) showing positive effects on real-world lower extremity use up to four years. Compared to our TOCT, CIMT protocol was much more intense (3.5 h/d for 3 wk), trained 15 dif- ferent tasks (compared to 7) and a “transfer package” to de- liver gains into the real world was administered [35]. The community exercise program was delivered to 2 MS sub- jects twice/week for 12 weeks with no effects on walking endurance [36]. Our two weeks high-intensity task oriented circuit training was feasible and safe in a small group of PwMS with moderate mobility impairments. Additional file y 13. Latimer-Cheung AE, Martin Ginis KA, Hicks AL, Motl RW, Pilutti LA, Duggan M, Wheeler G, Persad R, Smith KM: Development of evidence-informed physical activity guidelines for adults with multiple sclerosis. Arch Phys Med Rehabil 2013, 94:1829–1836. e7. 13. Latimer-Cheung AE, Martin Ginis KA, Hicks AL, Motl RW, Pilutti LA, Duggan M, Wheeler G, Persad R, Smith KM: Development of evidence-informed physical activity guidelines for adults with multiple sclerosis. Arch Phys Med Rehabil 2013, 94:1829–1836. e7. Conclusions h l Van den Berg M, Dawes H, Wade DT, Newman M, Burridge J, Izadi H, Sackley CM: Treadmill training for individuals with multiple sclerosis: a pilot randomised trial. J Neurol Neurosurg Psychiatry 2006, 77:531–533. 8. Newman MA, Dawes H, van den Berg M, Wade DT, Burridge J, Izadi H: Can aerobic treadmill training reduce the effort of walking and fatigue in people with multiple sclerosis: a pilot study. Mult Scler 2007, 13:113–119. 9. Rietberg MB, Brooks D, Uitdehaag BMJ, Kwakkel G: Exercise therapy for multiple sclerosis. Cochrane Database Syst Rev 2005, CD003980. 10 Andreasen AK Stenager E Dalgas U The effect of exercise therapy on a pilot randomised trial. J Neurol Neurosurg Psychiatry 2006, 77:531 533. 8. Newman MA, Dawes H, van den Berg M, Wade DT, Burridge J, Izadi H: Can aerobic treadmill training reduce the effort of walking and fatigue in people with multiple sclerosis: a pilot study. Mult Scler 2007, 13:113–119. 9. Rietberg MB, Brooks D, Uitdehaag BMJ, Kwakkel G: Exercise therapy for multiple sclerosis. Cochrane Database Syst Rev 2005, CD003980. 10. Andreasen AK, Stenager E, Dalgas U: The effect of exercise therapy on fatigue in multiple sclerosis. Mult Scler 2011, 17:1041–1054. 10. Andreasen AK, Stenager E, Dalgas U: The effect of exercise therapy on fatigue in multiple sclerosis. Mult Scler 2011, 17:1041–1054. g 11. Gervasoni E, Cattaneo D, Jonsdottir J: Effect of treadmill training on fatigue in multiple sclerosis: a pilot study. Int J Rehabil Res 2014, 37:54–60. 11. Gervasoni E, Cattaneo D, Jonsdottir J: Effect of treadmill training on fatigue in multiple sclerosis: a pilot study. Int J Rehabil Res 2014, 37:54–60. g p p y 12. Pilutti LA, Greenlee TA, Motl RW, Nickrent MS, Petruzzello SJ: Effects of exercise training on fatigue in multiple sclerosis: a meta-analysis. Psychosom Med 2013, 75:575–580. 12. Pilutti LA, Greenlee TA, Motl RW, Nickrent MS, Petruzzello SJ: Effects of exercise training on fatigue in multiple sclerosis: a meta-analysis. Psychosom Med 2013, 75:575–580. Additional file 1: Feasibility patient-reported questionnaire. 14. Sá MJ: Exercise therapy and multiple sclerosis: a systematic review. J Neurol 2013. Epub ahead of print 22 Nov 2013. doi:10.1007/s00415-013- 7183-9. 14. Sá MJ: Exercise therapy and multiple sclerosis: a systematic review. J Neurol 2013. Epub ahead of print 22 Nov 2013. doi:10.1007/s00415-013- 7183-9. Acknowledgments CM was supported by the Multiple Sclerosis Italian Society (grant 2010/R/6). CP and ASC were supported by Emilia Romagna region (grant 1786/2012). Thanks to Simonetta Masnada (PT) who delivered the task-oriented circuit training. 21. Rocca MA, Colombo B, Falini A, Ghezzi A, Martinelli V, Scotti G, Comi G, Filippi M: Cortical adaptation in patients with MS: a cross-sectional functional MRI study of disease phenotypes. Lancet Neurol 2005, 4:618–626. 22. Tomassini V, Johansen-Berg H, Leonardi L, Paixão L, Jbabdi S, Palace J, Pozzilli C, Matthews PM: Preservation of motor skill learning in patients with multiple sclerosis. Mult Scler 2011, 17:103–115. Abbreviations OC k TOCT: task-oriented circuit training; MS: multiple sclerosis; HRQoL: Health- related quality of life; EDSS: Expanded Disability Status Scale; PwMS: People with multiple sclerosis; UC: Usual care; 10MWT: 10 Meter Walk Test; SMWT: Six Minute Walk Test; TUG: Timed Up and Go Test; DGI: Dynamic Gait Index; FSS: Fatigue Severity Scale; MSWS-12: Multiple Sclerosis Walking Scale – 12; MSIS-29: Multiple Sclerosis Impact Scale; DMT: Disease-modifying therapy; VAS: Visual Analog Scale; CIMT: Constraint – Induced Movement Therapy. 15. Lang CE, Macdonald JR, Reisman DS, Boyd L, Jacobson Kimberley T, Schindler-Ivens SM, Hornby TG, Ross SA, Scheets PL: Observation of amounts of movement practice provided during stroke rehabilitation. Arch Phys Med Rehabil 2009, 90:1692–1698. 16. English C, Hillier SL: Circuit class therapy for improving mobility after stroke. Cochrane Database Syst Rev 2010, CD007513. 17. Wevers L, van de Port I, Vermue M, Mead G, Kwakkel G: Effects of task-oriented circuit class training on walking competency after stroke: a systematic review. Stroke 2009, 40:2450–2459. Limits and future directions Our pilot study has several limitations. Firstly, our small and heterogeneous sample didn’t allow us to draw definite inferences from this analysis; moreover the inability to blind subjects’ on the treatments received (TOCT or UC) might have reduced the internal validity of the study; secondly, we didn’t use neither a valid method to address home-exercise adherence, nor psychiatric and neuropsychological evalu- ation to identify possible factors related to discontinued therapies; finally, we didn’t have real-world measures (i.e. accelerometer or step counters) to test transfer- ability in everyday life. Future directions: (i) to conduct a large RCT trial to con- firm this preliminary results with a longer follow-up; (ii) to better explain MS mechanisms of functional recovery it Page 8 of 9 Page 8 of 9 Page 8 of 9 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 References 1. Confavreux C, Vukusic S, Moreau T, Adeleine P: Relapses and progression of disability in multiple sclerosis. N Engl J Med 2000, 343:1430–1438. 1. Confavreux C, Vukusic S, Moreau T, Adeleine P: Relapses and progression of disability in multiple sclerosis. N Engl J Med 2000, 343:1430–1438. 2. Sutliff MH: Contribution of impaired mobility to patient burden in multiple sclerosis. Curr Med Res Opin 2010, 26:109–119. 2. Sutliff MH: Contribution of impaired mobility to patient burden in multiple sclerosis Curr Med Res Opin 2010 26:109–119 2. Sutliff MH: Contribution of impaired mobility to patient burden in multiple sclerosis. Curr Med Res Opin 2010, 26:109–119. multiple sclerosis. Curr Med Res Opin 2010, 26:109–119. 3. Pfaffenberger N, Pfeiffer K-P, Deibl M, Höfer S, Günther V, Ulmer H: Association of factors influencing health-related quality of life in MS. Acta Neurol Scand 2006, 114:102–108. 4. Mitchell AJ, Benito-León J, González J-MM, Rivera-Navarro J: Quality of life and its assessment in multiple sclerosis: integrating physical and psychological components of wellbeing. Lancet Neurol 2005, 4:556–566. 4. Mitchell AJ, Benito-León J, González J-MM, Rivera-Navarro J: Quality of life and its assessment in multiple sclerosis: integrating physical and psychological components of wellbeing. Lancet Neurol 2005, 4:556–566. Authors’ contributions SS, MB, EZ, ASC and NB participated in the design of the study (including TOCT protocol); EZ, ASC, MB and CM carried out clinical tests; SS, CM, ASC, CP, MGB, NB and CF drafted the manuscript; SS and CP performed the statistical analysis. All authors read and approved the final manuscript. 19. Jang SH, Kim Y-H, Cho S-H, Lee J-H, Park J-W, Kwon Y-H: Cortical reorganization induced by task-oriented training in chronic hemiplegic stroke patients. Neuroreport 2003, 14:137–141. 20. Tomassini V, Matthews PM, Thompson AJ, Fuglø D, Geurts JJ, Johansen-Berg H, Jones DK, Rocca MA, Wise RG, Barkhof F, Palace J: Neuroplasticity and functional recovery in multiple sclerosis. Nat Rev Neurol 2012, 8:635–646. Received: 27 March 2014 Accepted: 30 May 2014 Published: 7 June 2014 Received: 27 March 2014 Accepted: 30 May 2014 Published: 7 June 2014 would be necessary to add neuroimaging parameters (i.e. functional near-infrared spectroscopy); (iii) to quantify the dose of exercise delivered; (iv) to add real-world based measures (i.e. steps counters and specific patient-reported questionnaires); (v) to test TOCT in a wide range of mobil- ity impairments (EDSS ≥6); (vi) to increase home-exercise program adherence with different methods (i.e. customized pamphlets, phone calls, physiotherapy visits). Competing interests The authors declare that they have no competing interests. 18. Kleim JA, Jones TA: Principles of experience-dependent neural plasticity: implications for rehabilitation after brain damage. J Speech Lang Hear Res 2008, 51:S225–S239. Conclusions h l In this pilot study, we assessed the safety, feasibility and preliminary effects on selected variables (gait speed, gait endurance, mobility, balance and health-related quality of life) of a task-oriented circuit training in a sample of multiple sclerosis subjects. Our main findings revealed that a 2 weeks task-oriented circuit training followed by a three months home exercise program is safe and well tolerated in multiple sclerosis subjects with moderate gait and mobility impairments. TOCT may allow a more intensive (more repetitions) and specific (task oriented) exercise in an enriched rehabilitative setting. No detri- mental effects on subjective fatigue have been observed in this sample or major adverse effects (i.e. fall-related injury). Even though some positive effects on walking ability has been underlined, this trial cannot draw defini- tive conclusions on the effects of such intervention. 5. Benedetti MG, Gasparroni V, Stecchi S, Zilioli R, Straudi S, Piperno R: Treadmill exercise in early mutiple sclerosis: a case series study. Eur J Phys Rehabil Med 2009, 45:53–59. 5. Benedetti MG, Gasparroni V, Stecchi S, Zilioli R, Straudi S, Piperno R: Treadmill exercise in early mutiple sclerosis: a case series study. Eur J Phys Rehabil Med 2009, 45:53–59. y 6. Wiles CM, Newcombe RG, Fuller KJ, Shaw S, Furnival-Doran J, Pickersgill TP, Morgan A: Controlled randomised crossover trial of the effects of physiotherapy on mobility in chronic multiple sclerosis. J Neurol Neurosurg Psychiatry 2001, 70:174–179. 6. Wiles CM, Newcombe RG, Fuller KJ, Shaw S, Furnival-Doran J, Pickersgill TP, Morgan A: Controlled randomised crossover trial of the effects of physiotherapy on mobility in chronic multiple sclerosis. J Neurol Neurosurg Psychiatry 2001, 70:174–179. g y y 7. Van den Berg M, Dawes H, Wade DT, Newman M, Burridge J, Izadi H, Sackley CM: Treadmill training for individuals with multiple sclerosis: a pilot randomised trial. J Neurol Neurosurg Psychiatry 2006, 77:531–533. 8. Newman MA, Dawes H, van den Berg M, Wade DT, Burridge J, Izadi H: Can aerobic treadmill training reduce the effort of walking and fatigue in people with multiple sclerosis: a pilot study. Mult Scler 2007, 13:113–119. 9. Rietberg MB, Brooks D, Uitdehaag BMJ, Kwakkel G: Exercise therapy for multiple sclerosis. Cochrane Database Syst Rev 2005, CD003980. 10 A d AK S E D l U Th ff t f i th y y 7. Author details 1 Forsberg A, Andreasson M, Nilsagård YE: Validity of the dynamic gait index in people with multiple sclerosis. Phys Ther 2013, 93:1369–1376. 29. Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD: The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 1989, 46:1121–1123. 30. Hobart JC, Riazi A, Lamping DL, Fitzpatrick R, Thompson AJ: Measuring the impact of MS on walking ability: the 12-Item MS Walking Scale (MSWS-12). Neurology 2003, 60:31–36. 31. Hobart J, Lamping D, Fitzpatrick R, Riazi A, Thompson A: The Multiple Sclerosis Impact Scale (MSIS-29): a new patient-based outcome measure. Brain J Neurol 2001, 124(Pt 5):962–973. 32. Conover WJ: Practical Nonparametric Statistics. 3rd edition. New York: John Wiley & Sons; 1999. y 33. Cohen J: Statistical Power Analysis for the Behavioral Sciences. Hillsdale: L. Erlbaum Associates; 1988. L. Erlbaum Associates; 1988. 34. Swinnen E, Beckwée D, Pinte D, Meeusen R, Baeyens J-P, Kerckhofs E: Treadmill training in multiple sclerosis: can body weight support or robot assistance provide added value? A systematic review. Mult Scler Int 2012. Article ID 240274, 15 pages, 2012. doi:10.1155/2012/240274. 35. Mark VW, Taub E, Uswatte G, Bashir K, Cutter GR, Bryson CC, Bishop-McKay S, Bowman MH: Constraint-induced movement therapy for the lower extremities in multiple sclerosis: case series with 4-year follow-up. Arch Phys Med Rehabil 2013, 94:753–760. 36. Salbach NM, Howe J-A, Brunton K, Salisbury K, Bodiam L: Partnering to Increase Access to Community Exercise Programs for People With Stroke, Acquired Brain Injury and Multiple Sclerosis. J Phys Act Health. Epub ahead of print 13 May 2013. 37. Baert I, Freeman J, Smedal T, Dalgas U, Romberg A, Kalron A, Conyers H, Elorriaga I, Gebara B, Gumse J, Heric A, Jensen E, Jones K, Knuts K, Maertens de Noordhout B, Martic A, Normann B, Eijnde BO, Rasova K, Santoyo Medina C, Truyens V, Wens I, Feys P: Responsiveness and Clinically Meaningful Improvement, According to Disability Level, of Five Walking Measures After Rehabilitation in Multiple Sclerosis: A European Multicenter Study. Neurorehabil Neural Repair. Epub ahead of print 6 Feb 2014. doi:10.1177/1545968314521010. 38. Gijbels D, Alders G, Van Hoof E, Charlier C, Roelants M, Broekmans T, Eijnde BO ‘t, Feys P: Predicting habitual walking performance in multiple sclerosis: relevance of capacity and self-report measures. Mult Scler 2010, 16:618–626. 39. Author details 1 Feys P, Bibby B, Romberg A, Santoyo C, Gebara B, de Noordhout BM, Knuts K, Bethoux F, Skjerbæk A, Jensen E, Baert I, Vaney C, de Groot V, Dalgas U: Within-day variability on short and long walking tests in persons with multiple sclerosis. J Neurol Sci 2014, 338:183–187. p 40. Kieseier BC, Pozzilli C: Assessing walking disability in multiple sclerosis. Mult Scler 2012, 18:914–924. 40. Kieseier BC, Pozzilli C: Assessing walking disability in multiple sclerosis. Mult Scler 2012, 18:914–924. 41. Riazi A, Hobart JC, Lamping DL, Fitzpatrick R, Thompson AJ: Multiple Sclerosis Impact Scale (MSIS-29): reliability and validity in hospital based samples. J Neurol Neurosurg Psychiatry 2002, 73:701–704. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 42. Learmonth YC, Hubbard EA, McAuley E, Motl RW: Psychometric properties of quality of life and health-related quality of life assessments in people with multiple sclerosis. Qual Life Res. Epub ahead of print 14 Feb 2014. doi:10.1007/s11136-014-0639-2. • Convenient online submission 43. Romberg A, Virtanen A, Ruutiainen J, Aunola S, Karppi S-L, Vaara M, Surakka J, Pohjolainen T, Seppänen A: Effects of a 6-month exercise program on patients with multiple sclerosis: a randomized study. Neurology 2004, 63:2034–2038. 44. Author details 1 1Neuroscience and Rehabilitation Department, Ferrara University Hospital, Ferrara, Italy. 2Doctoral Program in Advanced Sciences and Technologies in Rehabilitation Medicine and Sports, Tor Vergata University, Rome, Italy. 3Physical Medicine and Rehabilitation Department, Istituto Ortopedico Rizzoli, Bologna, Italy. 23. Paltamaa J, Sarasoja T, Leskinen E, Wikström J, Mälkiä E: Measures of physical functioning predict self-reported performance in self-care, mobility, and domestic life in ambulatory persons with multiple sclerosis. Arch Phys Med Rehabil 2007, 88:1649–1657. Page 9 of 9 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 Straudi et al. BMC Neurology 2014, 14:124 http://www.biomedcentral.com/1471-2377/14/124 45. DeBolt LS, McCubbin JA: The effects of home-based resistance exercise on balance, power, and mobility in adults with multiple sclerosis. Arch Phys Med Rehabil 2004, 85:290–297. 24. Dalgas U, Severinsen K, Overgaard K: Relations between 6 minute walking distance and 10 meter walking speed in patients with multiple sclerosis and stroke. Arch Phys Med Rehabil 2012, 93:1167–1172. 46. Bruce JM, Hancock LM, Arnett P, Lynch S: Treatment adherence in multiple sclerosis: association with emotional status, personality, and cognition. J Behav Med 2010, 33:219–227. 46. Bruce JM, Hancock LM, Arnett P, Lynch S: Treatment adherence in multiple sclerosis: association with emotional status, personality, and cognition. J Behav Med 2010, 33:219–227. 25. Goldman MD, Marrie RA, Cohen JA: Evaluation of the six-minute walk in multiple sclerosis subjects and healthy controls. Mult Scler 2008, 14:383–390. 25. Goldman MD, Marrie RA, Cohen JA: Evaluation of the six-minute walk in multiple sclerosis subjects and healthy controls. Mult Scler 2008, 14:383–390. 26. Podsiadlo D, Richardson S: The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc 1991, 39:142–148. 27. Cattaneo D, Jonsdottir J, Repetti S: Reliability of four scales on balance disorders in persons with multiple sclerosis. Disabil Rehabil 2007, 29:1920–1925. 28. Forsberg A, Andreasson M, Nilsagård YE: Validity of the dynamic gait index in people with multiple sclerosis. Phys Ther 2013, 93:1369–1376 26. Podsiadlo D, Richardson S: The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc 1991, 39:142–148. doi:10.1186/1471-2377-14-124 Cite this article as: Straudi et al.: A task-oriented circuit training in multiple sclerosis: a feasibility study. BMC Neurology 2014 14:124. 27. Cattaneo D, Jonsdottir J, Repetti S: Reliability of four scales on balance disorders in persons with multiple sclerosis. Disabil Rehabil 2007, 29:1920–1925. 28. Author details 1 Plow M, Bethoux F, McDaniel C, McGlynn M, Marcus B: Randomized controlled pilot study of customized pamphlets to promote physical activity and symptom self-management in women with multiple sclerosis. Clin Rehabil 2014, 28:139–148.
https://openalex.org/W2118929706
https://academicjournals.org/journal/AJB/article-full-text-pdf/9694B166750.pdf
English
null
Antibacterial potential of extracts of leaves of Parrotia persica
African journal of biotechnology
2,007
cc-by
1,945
Solvent extraction Apparently healthy leaves of P. persica were collected from the mountainous region of Mazandaran province of I.R. Iran during November 2005. The voucher specimen is deposited in the herba- Sample (20 g) dried powder was filled in the thimble and extracted successively with chloroform, ethanol, methanol and petroleum ether in Soxhlet extractor until the extract was clear. Short Communication Ethanol, methanol, chloroform, petroleum ether and aqueous extracts of leaves of Parrotia persica were evaluated for antibacterial activity. The zone of inhibition varied from 13 to 22 mm. The highest inhibit- tion was obtained with methanol and ethanol. Chloroform and petroleum ether extracts did not show any activity. The minimal inhibitory concentration (MIC) value of the methanol extract for the test bacteria ranged between 3.12 and 6.25 mg/ml and that of ethanol extract ranged between 6.25 and 12.5 mg/ml. The results scientifically validate the use of this plant in the traditional medicine of Iran. Key words: Parrotia persica, antibacterial activity, methanol extract, traditional medicine. INTRODUCTION Parrotia persica (DC) CA. Mey (Persian Ironwood) is a deciduous tree of the family Hamamelidaceae, the sole species in the genus Parrotia. It is native to northern Iran, and endemic to the Alborz Mountains. Many herbalists use this plant in the treatment various fevers and respiratory infections. It is also used for food coloring and food flavoring (Rechinger, 1999). There are no scientific reports on the antibacterial properties of the plant. Such lack of scientific knowledge has been major constraint in the use of traditional herbal remedies (Patwardhan et al., 2004). Hence, the aqueous and solvent extracts of leaves of P. persica were quantitatively screened for antibacte- rial activity against five important human pathogenic bacteria, obtained from Microbial Type Culture Collection and Gene bank (MTCC) India. rium of Department of Studies in Botany and Microbiology, Univer- sity of Mysore, Mysore, India. Aqueous extraction Samples (20 g) of thoroughly dried leaves were macerated with 40 ml sterile water in a waring blender for 10 min. The macerate was first filtered through double-layered muslin cloth, and then centri- fuged at 4000 rpm for 10 min. The supernatant was filtered through Whatman No.1 filter paper. The extract was preserved aseptically in a brown bottle at 5ºC until further use. Extraction of plant materials The leaves of P. persica were washed thoroughly three times with running water and once with distilled water. The materials were air dried under shade and powdered. The shade dried leaves and powdered samples were hermetically sealed in separate polythene bags until the time of extraction. *Corresponding author. E-mail: karaveesha@gmail.com. Tel: +91-821-2412418(0). African Journal of Biotechnology Vol. 6 (22), pp. 2526-2528, 19 November, 2007 Available online at http://www.academicjournals.org/AJB ISSN 1684–5315 © 2007 Academic Journals African Journal of Biotechnology Vol. 6 (22), pp. 2526-2528, 19 November, 2007 Available online at http://www.academicjournals.org/AJB ISSN 1684–5315 © 2007 Academic Journals Phytochemical screening The methanol extract, which was antibacterially active, was selec- Ahanjan et al. 252 2527 Ahanjan et al. Table 1. Antibacterial activity of leaves of Parrotia persica extracts at 100 l. Zone of Inhibition (mm) Bacteria Aqueous extract Chloroform extract Ethanol extract Methanol extract Petroleum ether extract Gentamycin 10 (µg) E. coli 13.0±04 0.0 14±0.5 15±0.4 0.0 34±0. K. pneumoniae 16±0.3 0.0 15±0.4 22±0.2 0.0 30±0.4 P. aeruginosa 15±0.4 0.0 18±0.3 17±053 0.0 31±0.5 S. typhi 15±0.6 0.0 22±0.2 21±0.6 0.0 30±0.3 S. aureus 16±0.5 0.0 16±0.5 20±0.5 0.0 33±0.5 Values presented are means of six replicates ± standard error. When subjected to analysis of variance (ANOVA), p < 0.05. Table 2. Minimal inhibitory concentration (MIC) of methanol extract of P. persica against the test bacteria. Zone of Inhibition in mm MIC of methanol extract (mg/ml) Bacteria 50 25 12.5 6.25 3.12 1.56 0.78 Gentamycin (10 µg) E. coli 9±0.3 5±0.4 3±0.3 2±0.2 0.0 0.0 0.0 34±0.7 K. neumoniae 17±0.5 14±0.4 11±0.2 8±0.4 4±0.4 0.0 0.0 30±0.9 P. aeruginosa 13±0.3 8±0.6 5±0.3 3±0.3 2±0.5 0.0 0.0 31±0.3 S. typhi 15±0.3 11±0.7 6±0.3 4±0.4 2±0.3 0.0 0.0 30±0.3 S. aureus 9±0.2 6±0.2 4±0.7 3±0.2 2±0.4 0.0 0.0 33±0.2 Values presented are means of six replicates ± Standard error. When subjected to analysis of variance (ANOVA), p < 0.05. able 2. Minimal inhibitory concentration (MIC) of methanol extract of P. persica against the test bacteria. alues presented are means of six replicates ± Standard error. When subjected to analysis of variance (ANOVA), < 0.05. Determination of minimal inhibitory concentration (MIC) ted for phytochemical analysis. The dried extract was first recon- stituted in methanol and subjected to standard phytochemical analysis for the presence of flavonoids, tannins, saponins, carbohy- drates, terpenoids and protein (Harborne, 1988). The MIC was determined only for the methanol extract by agar well diffusion method as it is showed a wider spectrum of antibacterial activity compared to others. A two-fold serial dilution of the extract was prepared by first reconstituting in methanol then diluting in sterile distilled water to achieve a decreasing concentration range of 50 to 0.781 mg/ml. A 100 µl volume of each dilution was introduced in triplicate wells into MHA plates already seeded with the standardized inoculums (5 × 105 cfu/ml) of the test bacterial cells. All test plates were incubated at 37ºC for 24 h. The least concentration of each extract showing a clear zone of inhibition was taken as the MIC. Test bacterial strains Standard type cultures of Escherichia coli (MTCC 443), Klebseilla pneumoniae (MTCC 109), Pseudomonas aeruginosa (MTCC 1688), Salmonella typhi (MTCC 733) and Staphylococcus aureus (MTCC 737) were obtained from MTCC Chandigrah, India. All the test strains were subcultured on Muller Hinton agar (MHA). Antibacterial activity assay The results of the antibacterial activity of the extracts are recorded in Table 1. The aqueous extract, ethanol extract and methanol extract of leaves of P. persica showed good activity against all the test bacteria. The chloroform and petroleum ether extracts did not show any activity against the test bacterial strains. The methanol extracts of the leaves recorded activity to all standard cultures with highest mean zone of inhibition. The MIC value of methanol extract was 3.12 mg/ml for K. pneumoniae, P. aeruginosa, S. typhi and S. aureus and 6.25 mg/ml for E. coli (Table 2). The results of the antibacterial activity of the extracts are recorded in Table 1. The aqueous extract, ethanol extract and methanol extract of leaves of P. persica showed good activity against all the test bacteria. The chloroform and petroleum ether extracts did not show any activity against the test bacterial strains. The methanol extracts of the leaves recorded activity to all standard cultures with highest mean zone of inhibition. The MIC value of methanol extract was 3.12 mg/ml for K. pneumoniae, P. aeruginosa, S. typhi and S. aureus and 6.25 mg/ml for E. coli (Table 2). The extracts were spot checked for antibacterial activity employing the agar well diffusion technique (Okeke et al., 2001). Standardized inoculum (5× 105 cfu/ml) of each test bacterium was spread on to sterile MHA plates so as to achieve a confluent growth. The plates were allowed to dry and a sterile cork borer of diameter 6.0 mm was used to bore wells in the agar plates. The dried extracts were reconstituted in methanol to a concentration of 50 mg/ml. Subse- quently, a 100 µl volume of the extracts were introduced in triplicate wells into the MHA cultures. The plates were allowed to stand for 1 h. for diffusion to take place and then incubated at 37ºC for 24 h. The zone of inhibition was recorded to the nearest mm. Only the extracts exhibiting apparent zone of inhibition were chosen for further evaluation. For comparative evaluation the antibiotic Genta- mycin (10 µg) was used. The MIC values of ethanol extract for K. pneumoniae 2528 Afr. J. Biotechnol. Table 3. MIC of ethanol extract of P. persica against the test bacteria. Table 3. MIC of ethanol extract of P. persica against the test bacteria. Table 3. MIC of ethanol extract of P. persica against the test bacteria. REFERENCES Ahmet A, Medine G, Meryam S, Hatice O, Fikrettin S, Isa K (2005). Antimicrobial Effects of Ocimum basillicum (Labiatae) Extract. Turk J. biol. 29: 155-160 was 6.25 and 12.5 mg/ml for E. coli and S. aureus (Table 3). The methanol extract of leaves of P. persica was anti- bacterially active to K. pneuminiae with highest mean zone of inhibition (22 mm). The phytochemical analysis showed that the methanol extract of P. persica contained tannin, flavonoid, protein, saponin and carbohydrate (Table 4). Harborne JB (1998). Phytochemical methods: A guide to modern techniques of plant analysis. 3rd edition. Chapman and Hall. London. Khan MR, Kihara M, Omoloso AD (2002). Antimicrobial activity of Calophyllum soulattri. Fitoprapia, 73: 741-743. Newman DJ, Cragg GM, Snader KM (2000). The influence of natural products upon drug discovery. Nat. Prod. Res. 17: 215-234. Nkere CK, Iroegbu CU (2005). Antibacterial screening of the root, seed and stembark extracts of Picralima nitida. Afr. J. Biotechnol. 4: 522-526. Okeke MI, Iroegbu CU, Eze EN, Okoli AS and Esimone CO (2001). Evaluation of extracts of the root of Landolphia owerrience for antibacterial activity. J. Ethnopharmacol. 78: 119-127 Antibacterial activity assay Zone of Inhibition in mm MIC of ethanol extract (mg/ml) Bacteria 50 25 12.5 6.25 3.12 1.56 0.78 Gentamycin (10 µg) E. coli 8.0±0.3 4.5±0.4 3±0.3 0.0 0.0 0.0 0.0 33±0.7 K. neumoniae 11±0.5 8±0.4 4±0.2 2.0±03 0.0 0.0 0.0 29±0.9 P. aeruginosa 12±0.3 7±0.6 4±0.3 2.0±02 0.0 0.0 0.0 30±0.3 S. typhi 13±0.3 8±0.7 4±0.3 2.0±03 0.0 0.0 0.0 30±0.3 S. aureus 7±0.2 4±0.2 2±02 0.0 0.0 0.0 0.0 30±0.2 Values presented are means of six replicates ± Standard error. When subjected to analysis of variance (ANOVA), p < 0.05. Zone of Inhibition in mm Table 4. Phytochemical analysis of methanol Table 4. Phytochemical analysis of methanol extract of P. persica. extraction of the antibacterial principle (Nkere and Iroegbu, 2005). Thus methanol is recommended for the large-scale extraction of the active principle (Taous et al., 2005). Table 4. Phytochemical analysis of methanol extract of P. persica. Constituent Methanol extract Alkaloids - Carbohydrates + Flavonoids + Gum + Phenolics + Proteins + Saponins + Tannins + Terpenoids + Waxes - + Present, - absent. ) Saponins and tannins are reported to possess antibac- terial activity (Newman et al., 2000). Preliminary phyto- chemical analysis of the methanol extract of leaves of P. persica revealed the presence of saponin, tannin and flavonoid, which could be the active principle. Further investigations are in progress to isolate and characterize the active principle. This study scientifically validates the use of this plant for treatment of fever and other respira- tory infection in traditional medicine of Iran. DISCUSSION Okoli AS, Okeke MI, Iroegbu CU, PU (2002). Antibacterial activity of Harungana madagascariensis leaf extracts. Phytother. Res. 16: 174-179. Review of literature reveals lack of information on the antibacterial potential of P. persica. Aqueous and differ- rent solvent extracts of leaves of P. persica evaluated for antibacterial potential, revealed highly significant activity of methanol extract against all the test bacteria followed by ethanol extract. None of the earlier reports have demonstrated the antibacterial potential of this plant (Okoli et al., 2002; Khan et al., 2002; Ahmet et al., 2005). In the present investigation the antibacterial activity of this plant has been demonstrated for the first time. Patwardhan B, Vaidya ADB, Chorghade M (2004). Ayurveda and natural products drug discovery. Curr. Sci. 86: 789-799. Rechinger KH (1999). Flora Iranica. Akademische Druk-u.Verlagsanstalt. Graz, Austria. Taous k, Mansoor A, Hamayun K, Mir AK (2005). Biological activities of aerial parts of Paeonia emodi wall. Afr. J. Biotechnol. 4: 1313-1316. It is also evident from the results that methanol extract has significantly high degree of antibacterial activity, sug- gesting that methanol is the appropriate solvent for
W2778660023.txt
https://content.sciendo.com/downloadpdf/journals/quageo/36/4/article-p103.pdf
en
Innovations and networking fostering tourist destination development in Slovakia
Quaestiones Geographicae
2,017
cc-by
6,517
QUAESTIONES GEOGRAPHICAE 36(4) • 2017 INNOVATIONS AND NETWORKING FOSTERING TOURIST DESTINATION DEVELOPMENT IN SLOVAKIA Tomáš Gajdošík1, Zuzana Gajdošíková1, Vanda Maráková1, Kamila Borseková2 Department of Tourism and Hospitality, Faculty of Economics, Matej Bel University in Banská Bystrica, Tajovského 10, 975 90 Banská Bystrica, Slovakia 2 Research and Innovation Centre, Faculty of Economics, Matej Bel University in Banská Bystrica, Cesta na amfiteáter 1, 974 01 Banská Bystrica, Slovakia 1 Manuscript Received: April 26, 2017 Revised version: July 27, 2017 Gajdošík T., Gajdošíková Z., Maráková V., Borseková K., 2017. Innovations and networking fostering tourist destination development in Slovakia. Quaestiones Geographicae 36(4), Bogucki Wydawnictwo Naukowe, Poznań, pp. 103–115, 8 tables, 5 figs. Abstract: The paper focuses on the implementation of innovations and networking in the sector of tourism in two Slovak mountain destinations of international significance. The main objective of the paper is to identify and evaluate how innovations and networking contribute to tourist destination development in Slovakia. The implementation of institutional innovation resulted in the establishment of formal and informal networks. The developed networks consist of representatives of all sectors co-ordinating all relevant stakeholders. Formal and informal networks and the collaboration among stakeholders have launched other types of innovations in the tourism sector. The interactions and intensity of relations among stakeholders are analysed by network analysis. Destinations are compared with the network of the same size and density through quantitative network characteristics. Based on empirical research we investigate the impact of networks and innovations on tourist destination development. Due to the synergy effect of networking and implementation of multiple innovations, tourist destination development is observed. Key words: destination management organisations, innovation, networking, spatial development, tourism, Slovakia Corresponding author: Tomáš Gajdošík, Department of Tourism and Hospitality, Faculty of Economics, Matej Bel University in Banská Bystrica, Tajovského 10, 975 90 Banská Bystrica, Slovakia; e-mail: Introduction Throughout history, tourism has been a phenomenon characterised by immense innovativeness. Over the past couple of years, the term innovation has increasingly been used to describe the trajectory of development of tourism enterprises, destinations and the tourism sector as a whole (Hall, Williams 2008; Hjalager 2002, 2010; Hjalager, Flagestad 2012). The role of the tourism sector in the regional, national and especially global economy has significantly increased in recent years, and tourism has become an enhancer of economic development (Romao, Guerreiroa, Rodrigues 2013) and innovative behaviour. Moreover, support for tourist activities is an important element of regional development policies in the long term (Navrátil et al. 2013). Tourist organisations and destinations in current dynamic environment must reinvent their cultures and strategies to survive and create a sustainable competitive advantage (Racherla, Hu, Hyun 2008). If tourist destinations want to be competitive, sustain their competitive advantage, and attract new © 2017 Author(s) This is an open access article distributed under the Creative Commons Attribution-NonCommercial-NoDerivs license doi: 10.1515/­quageo-2017-0039 ISSN 0137-477X Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková 104 Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková or keep old customers, innovative behaviour is crucial (Borseková, Vaňová, Vitálišová 2017). The paper focuses on two Slovak destinations of international significance – the High Tatras and Liptov. These two destinations are, in terms of Butler’s (1980) destination life cycle, in the maturity stage. In the initial stages of the destination life cycle, the level of cooperation between destination stakeholders is weak. When the stage of maturity is reached, the destinations will experience a synergic effect of cooperation between the stakeholders, which creates the conditions for the establishment of a destination management organisation. The relationships between the stakeholders and the destination management organisation should result in trust, transparency and leadership. These two selected tourist destinations belong to the most attractive ones with the highest number of accommodations in Slovakia. With their infrastructural capacity and the number of ski slopes, they belong to the largest winter sport areas in Central Europe. Both destinations show good internal potential and the implemented innovations should orchestrate the development of tourist destinations in the future. The scientific goal of the paper is to identify and evaluate how innovations and networking contribute to tourist destination development in Slovakia. The paper focuses on the established networks pushed by institutional innovation and how the created networks generate and introduce other types of innovations in the tourism sector and their impact on the overall development of tourist destinations in Slovakia. Theoretical background There are several types of innovations in the tourism sector. Jafari et al. (2003) distinguish product, process, managerial, marketing and logistic innovations. OECD (2004) adds market and ad hoc innovations, while Mayer (2009) focuses also on organisational, institutional and customer innovations. Another point of view is presented by Castellacci (2008) who deals with demand-driven innovations and those resulting from technologies and state interventions. Moreover, Camisón and Monfort-Mir (2012) focus on innovations deriving from science and research and market-driven innovations. However, the most significant classification of innovations in tourism is the one presented by Hjalager (2010) who creates the typology of innovations focused on product, process, managerial, marketing and institutional innovations. Product or service innovations refer to changes directly observed by the customer, which are regarded as new. It can be a completely new product or service, or new only for a particular enterprise or destination. Process innovations are aimed at escalating efficiency, increasing productivity and flow (Hjalager 2010). Managerial innovations deal with new approaches to internal collaboration, motivating and empowering staff, building careers and compensating workers with extra pay and benefits (Ottenbacher, Gnoth 2005). Marketing innovations focus on new marketing approaches targeting changes in communication between service providers and customers, and building their positive relationship (Hankinton 2004). Institutional innovation is a new comprehensive collaborative or organisational structure, or a legal framework that efficiently enhances business in certain fields of tourism. Networks and alliances in tourism are considered as an essence of fostering innovations in the tourism sector (Lynch, Morrison 2007). The development of tourist destinations is influenced by several stakeholders. A stakeholder theory defines a stakeholder as “any group or individual who can affect or is affected by the achievement of organisation objectives” (Freeman 1984). When applying the stakeholders’ theory on the tourist destination, several types of stakeholders can be found. They can be divided into several groups, e.g. as community-based stakeholders, service providers, employees, market-based stakeholders, owner-based stakeholders, financial stakeholders and others (Flagestad 2001). Sheehan and Ritchie (2005) include local, regional and national tourism associations, accommodation facilities, hospitality facilities, tourist attractions, congress centres, residents and universities. Bieger and Beritelli (2012) look at stakeholders from a wider perspective. They include destination management organisations, hotels and other accommodation facilities, residents, visitors, environmental associations, tradesmen, mountain transport operators, employees, shareholders, political parties, local Innovations and networking fostering tourist destination development in Slovakia Table 1. The most important stakeholders in a destination. Sector Stakeholder Public-private Destination management organisations Tourism associations Public Municipalities Cultural facilities Educational institutions Regional development agencies Private Accommodation facilities Catering facilities Tour operators Congress centres Sport and recreational facilities Transport associations Source: own elaboration. authorities and suppliers. Taking into account these classifications and the importance of stakeholders in the networking process, we propose the following classification of stakeholders in tourist destinations (Table 1). In order to compete in the international market, the cooperation of all relevant stakeholders is very important (Wang, Krakover 2008). Although these stakeholders have various interests and in a particular situation can act as competitors, it is crucial for them to collaborate in order to reach new target groups, new markets and keep their customers. As a result of a collaborative approach among stakeholders within tourist destinations, formal and informal networks are created. Formal networks are relationships institutionalised among existing stakeholders. These networks are based on the joint decision-making process by involving key stakeholders in resolving their conflicts, and on the advantage of sharing common vision (Gray 1989; Jamal, Getz 1995; Hall 1999). It can be done by establishing destination management organisations (DMO) or other legally binding forms of collaboration (Fernández-Cavia et al. 2014). As Pechlaner et al. (2012) state, the DMO acts as a coordinator of stakeholders in tourist destinations. Its role is to coordinate activities in a destination under one coherent strategy (UNWTO 2007). The DMO does not control the actions of its stakeholders, but brings together financial resources and undertake managerial and marketing activities. Informal networks are based on good and productive relations among the network members without creating a formal organisation. 105 The issue of tourism networks was researched by many experts, e.g. Beni (2003), Ferreira (2003), Capone (2004), Baggio (2008), Baggio, Scott and Cooper (2010), Beritelli, Strobl and Peters (2013) and others. Tourism networks can be classified as business networks, where networking is the tool to enhance firm performance, policy networks that provide the access to resources, and co-operating networks reaching the balance between competition and collaboration (Van der Zee, Vanneste 2015). Data and methods This paper is based on empirical research by the exploitation of a solid set of scientific methods. The stakeholders in each destination were identified, on the basis of a database in the destination management system, an official web page and the booking system used in each region. As these sources were designed to inform the visitor about all attractions and products in the destination, therefore it can be assumed that the number of identified stakeholders comes close to the actual number. The population consists of 405 stakeholders (195 in the High Tatras and 210 in Liptov) (Table 2). The networking of stakeholders in tourist destinations was identified on the basis of relations during the creation of tourist destination product and the integrated marketing communication. These two criteria may be perceived as the most crucial activities of destination management, Table 2. The structure of tourism stakeholders in the destination regions of the High Tatras and Liptov. Category of stakeholder DMOs Tourism associations (and clusters) Hotels Pensions Catering facilities Sport and recreational facilities Cultural facilities Transport associations Municipalities Tour operators Σ High Tatras 1 2 68 58 43 5 2 4 3 9 196 Liptov 1 8 43 56 44 10 12 1 25 10 210 Source: based on destination management systems, web pages and a booking system of each destination. 106 Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková Table 3. Selected quantitative characteristics of network analysis. Category Quantitative characteristics Main characteristics of the network Size of network (m) Number of cooperative relations Network density (δ) The ratio between m and the maximum possible number of relations that a network can have Average degree (deg) Average number of relations, that the subject has with neighbours Intensity of cooperation Power of the subject and network Formula Description Clustering of coefficient (C) Tendency of subjects to create homogenous groups with relatively high density Weighted clustering coefficient (C) Probability that a subject cooperates with another subject in the network Degree centrality (CD) Number of relations of the subject Source: processed according to Hanneman, Riddle 2005; Scott, Baggio, Cooper 2008a; Baggio, Scott, Cooper 2010. Note: a = value of a relation (0–1), n = no. of subjects, E = no. of relations, v = no. of graph vertices, σst = no. of the shortest paths from subject s to subject t, t = no. of relations between neighbours, k = degree of a node, A = adjacency matrix, d = distance between subject, Cx(pi) = centrality measure of point i, Cx(p*) = largest centrality measure in the network. therefore the power and resources were not taken into account. For determining the relations, secondary data were used. These include destination brochures, internal materials of destination management organisations, web sites of tourism stakeholders and the destination management system. The relations were identified using a binary system, without taking into account the intensity of cooperation and proceeded by social network analysis. The comparable methodology can be found in the research of Baggio et al. (2010) or Grama and Baggio (2014). Network analysis provides tools to describe the structure of relations between the various entities and allows the measurement of networking. This is an application of the matrix and graph theory (Hanneman, Riddle 2005). It uses primarily the graphic display that consists of nodes representing actors (stakeholders), and lines representing relations (cooperation). Using this concept, network analysis identifies relevant stakeholders, focuses on the connection between them and seeks to form the nature, pattern and implications of the connection between the nodes (Scott, Carrington 2011). Network analysis in tourism is a relatively new way of analysing the interactions between stakeholders, further explored in the works of Scott, Baggio and Cooper (2008), Baggio et al. (2010), Presenza and Cipollina (2010), Beritelli, Strobl and Peters (2013), Del Chiappa and Presenza (2013), Grama and Baggio (2013), O´Sullivan (2014). Apart from the most used graphical interpretation, social network analysis can describe the relationships by quantitative characteristics (Table 3). The data set is analysed by using the UCINET 6.560 program (Borgatti, Everett, Freeman 2002) and NetDraw, which make it possible to calculate all selected quantitative characteristics and visualise cooperative behaviour. The primary data set was examined via several surveys carried out among destination Table 4. The structure of stakeholders that answered the questionnaire. Stakeholder DMOs/ Tourism associations Hotels Pensions Catering facilities Sport and recreational facilities Cultural facilities Transport associations Municipalities Tour operators Σ Source: own study. High Tatras 1 16 10 3 3 1 0 2 3 39 Liptov 3 13 13 4 0 2 0 1 0 36 Innovations and networking fostering tourist destination development in Slovakia 107 Table 5. Networking in tourism in Slovakia since 1989. Time period Form of Partnership Focus 1989–2007 Regional tourism associations, regional associations of cities and municipalities, Euro-regional associations, local action groups Coordinated tourism development Strong competition Organising events 2008–2011 Tourism clusters Open partnerships Coordination of marketing activities Beginning in creation of a single product 2012 till now Destination management organisations following the Law no. 91/2010 Coll. Branding Reservation systems Creation of product and its marketing communication Source: own study. stakeholders. At first, all major stakeholders in the destinations were asked to fill in a questionnaire that focused on the strength of partnerships with other stakeholders. The research sample consists of those who answered the questionnaire. In total, 75 stakeholders responded, leading to a response rate of 18.52%. The survey was not only sent to the members of destination management organisations, but to all businesses operating within the tourism industry in selected destinations as well (Table 4). The implemented innovations in tourist destinations were identified based on in-depth interviews with the members of DMOs and operators of mountain transport facilities. Moreover, the annual reports of these organisations were examined to compare the findings from the interviews and to enrich the analysis. Research results and the discussion Evolution of networking in tourism in Slovakia The networking in the tourism sector in Slovakia started after 1989, when the organisations based on the public-private partnership principle were created. Firstly, they had the character of tourism associations as non-profit organisations. Their main mission was to create the conditions to meet the needs of tourist visitors, to develop the entrepreneurial activities and to coordinate the interests of tourism stakeholders. The legislation in that period did not create sufficient space to finance the activities of tourism associations. These associations, however, represented only a relatively small number of stakeholders, and they were often “dormant” or only active for certain projects (notably the development of the tourism infrastructure). The associations usually did not have permanent professional staff, but their members provided the tasks besides their main jobs. Since 2008, tourism clusters have been created, with the potential for the successful implementation of spatial and tourist development. An important motivation to create them was an effort to coordinate the marketing of a destination, tourist destination development and to promote collaboration. Due to the lack of focus on research and innovations the tourism clusters were far away from the real cluster conception (Table 5). Since 2012, for the first time in Slovakia, the systematic development of organisational structures in tourism has been observed, supported financially by the government. The Tourism Support Act no. 91/2010 Coll. was introduced, and established an organisational structure at the national, regional and local levels, and defined the way of financial subsidies for tourism organisations at the local (NUTS4 and NUTS5/ LAU1 and LAU2) and regional (NUTS3) level. According to this law, a key role regarding the management of strategic business units – destinations, should be fulfilled within newly created destination management organisations (DMOs). Current institutional and legislative conditions supporting networking in the tourism sector in Slovakia Nowadays, the collaboration and stimulus for establishing formal networks in the tourism sector in Slovakia is legislatively governed by the Tourism Support Act no. 91/2010 Coll. So far it has been the biggest institutional innovation in 108 Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková Fig. 1. Map of destination management organisations in Slovakia. Source: own study based on Kidová (2017). the tourism sector in this country. As the Act financially supports the creation of DMOs, it can be regarded as a “top-down” approach, where the organisations are not created from the spontaneous will of stakeholders (as was the case of tourism associations and clusters), but highly stimulated by financial incentives. This institutional innovation created better conditions for networking within tourism stakeholders. Currently, there are 36 DMOs operating at the local level and five at the regional one. Fig. 1 illustrates the location of DMOs in Slovakia. We must critically admit that the number of DMOs operating at the local level is much too large, since some have only one employee and a limited budget. According to the Tourism Support Act, the financial support is especially dedicated to those destinations that have the largest number of overnight stays and the developed tourism infrastructure, implying deepening regional disparities (Kučerová 2015). In particular, it can be observed that the three biggest DMOs in 2016 accounted for 62.71% of total subsidies (Fig. 1). As the collaboration within a newly established DMO is financially supported by state subsidies, newly created DMOs dominate in Slovakia as a consequence of the prevailing institutional scheme. Due to the possibility to receive subsidies from the state budget, many other tourism associations, clusters and other subjects cooperate with DMOs. At present, DMOs are supported from public resources, while other tourism organisations benefit from EU funds, or are financially dependent on membership fees. Networking enhancing further innovations in tourist destinations According to Soteriades (2012), the networking among relevant stakeholders can be one of the best available tools in fostering tourist development. Therefore, it is important to examine how the creation of networks and their operating influence the development of tourist destinations. In order to express how the networking enhances innovations in tourism, the authors focus on a quantitative and qualitative analysis of the two most important mountain destinations in Slovakia. The High Tatras and Liptov are traditional tourist destinations characterised by a fragmentation on the supply side. In both regions, several tourism stakeholders create the destination’s product. These stakeholders differ according to their size, interests, organisational form and character of services. The networking among stakeholders in tourist destinations is analysed using binary network analysis on the basis of the creation of a common regional product and integrated marketing communication. The beginning of collaboration among tourism stakeholders in the High Tatras started in Innovations and networking fostering tourist destination development in Slovakia 109 Fig. 2. Graph of network analysis in the High Tatras. Source: own study. 1994 with the creation of the Tourism Association of the High Tatras. After the implementation of the Tourism Support Act, the High Tatras DMO Region was established in 2012. This DMO expresses the common interest of three municipalities, the Tourism Association of the High Tatras, the operators of mountain transport facilities and an aqua park. The role of public-private partnerships is crucial in order to support the process of development and marketing of a regional tourism product. Fig. 2 indicates that the collaboration of stakeholders in the High Tatras, particularly accommodation facilities, is concentrated in the Tourism Association and the DMO Region High Tatras. The stakeholders cooperate in the development of the destination visitor’s card Tatras Card Winter and Tatras Card Summer, through which visitors can enjoy discounts in sports and recreational facilities, cultural and catering facilities, and also free transport by an aqua ski bus. This cooperation can also be seen in the Go Pass card, which is a loyalty program of a private stakeholder (TMR, Inc.), but due to the inclusion of the major tourist attractions in the destination, this card acts as a destination visitor’s card. The collaboration of stakeholders in Liptov is concentrated on the destination management organisation the Liptov DMO Region and the Liptov Cluster. The Liptov Cluster was established in 2008 through a bottom-up approach. From the membership point of view, the cluster combines the operators of mountain transport facilities and aqua parks, supporting businesses and 19 municipalities (three cities and 16 villages). In total, the cluster has 27 members. The DMO Region Liptov was established in 2012, following the Tourism Support Act. 91/2010 Coll. A top-down approach was driven mainly by the opportunity to benefit from state subsidies. The Liptov Cluster and the Liptov DMO Region do not compete with each other, they built the brand “Liptov” together. The DMO is financed by a state subsidy and membership fees and the cluster is financed by European funds. Fig. 3 indicates collaboration and networking in Liptov, where the cooperative behaviour of accommodation facilities, the Cluster and the DMO can be observed. Stakeholders created the product Liptov Card Winter and Liptov Card Summer. The visitor can benefit from discounts on major attractions in the region. An interesting product of the summer season is the Seven Treasures of Liptov, which uses the elements of geocaching and motivates visitors to get to know the entire region of Liptov. The research is enriched by the quantitative network characteristics (Table 6), where the actual network characteristics of destinations are 110 Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková Fig. 3. Graph of network analysis in Liptov. Source: own study. compared with the network of the same size and density, but with the random distribution of stakeholders and relations. The average in both destinations shows that the average number of relations that the subject has with its neighbours is almost four. As far as the intensity of networking is concerned, the most important characteristics are a clustering coefficient and its weighted value, which measure the tendency of the subject to create homogenous groups where collaboration prevails. Based on the values of these characteristics, which are larger than zero, as well as larger than in networks with a random distribution, the creation of groups where collaboration is higher than in other places in the network in both destinations can be observed. The reason for the creation of such clusters is collaboration and networking stimulated by the established DMOs. The second analysis is also supported by the results based on a questionnaire that focuses on the strength of collaboration from the stakeholders’ point of view (Fig. 4). The stakeholders were asked to evaluate the strength of collaboration with another group of stakeholders on the scale 1-5, where 1 stands for very weak collaboration and 5 stands for very good collaboration. In all, 39 stakeholders from the High Tatras and 36 from the Liptov region took part in the survey. The strongest collaboration can be found with destination management organisations, tourism associations and municipalities. In terms of creating a competitive tourism product, the cooperation with accommodation and catering facilities should be stronger. Both DMOs now play a crucial role in networking based on product development and marketing communication. Their annual budgets exceed €1.5m and their goals in destination development are presented in Table 7. Due to the networking of the major stakeholders, several product and marketing innovations Table 6. Quantitative characteristics of networking in the High Tatras and Liptov. Quantitative characteristics Size of the network Network density Average degree Clustering of coefficient Weighted clustering coefficient Source: own study. High Tatras 696 0.018 3.569 0.543 0.037 Liptov 794 0.017 3.710 0.413 0.035 High Tatras – random 696 0.018 3.528 0.019 0.020 Liptov – random 794 0.017 3.682 0.019 0.018 Innovations and networking fostering tourist destination development in Slovakia DMO Sport and Recreational Facilities Municipalities Tourism Associations Tour Operators Cultural Facilities Hotels Pensions Catering Facilities Transport Associations Liptov 3.542857 2.97058823529412 2.73529411764706 2.60606060606061 2.55882352941176 2.42857142857143 2.36363636363636 2.3030303030303 2.24242424242424 1.97142857142857 111 High Tatras 3.07894736842105 2.64864864864865 2.86486486486486 3.23684210526316 3.13888888888889 2.43243243243243 2.72222222222222 2.25 1.75 1.84210526315789 Fig. 4. Strength of networking in the examined destinations. Source: own study. have been introduced in the High Tatras and in Liptov in the last years (Table 8). Although a DMO would be expected to be the initiator of these innovations in both destinations, several innovations were initiated by the company TMR, Inc., or the Liptov Cluster. These innovations were focused on the product, price, promotion, and distribution. The creation of accommodation packages had an impact on the growth of the average length of stay and increased occupancy rates in hotels. Both DMOs and TMR, Inc. started to create accommodation packages. Destination cards primarily aimed at increasing the sales of a multi-day ski pass and the average length of visitors’ stay. A suitable combination of accommodations with skiing and other discounts on additional services stimulated by collaboration and networking affected tourist destination development. Process innovations were focused on efficiency, productivity and flow. They were connected mainly with implementing information technologies. In both destinations, new destination management systems were introduced. This Table 7. Goals of destination management organisations. High Tatras –– to group the destination’s stakeholders –– to group the financial resources for tourism development –– to build the common infrastructure, product and promotion of the destination region –– to support effective spatial organisation in the region –– to create a good image of the region Source: annual reports of DMOs, 2016. Liptov –– to create a recognisable brand of the destination regions in the domestic and international market –– to create attractive tourism products –– to create a complex system of destination management –– to create and build effective marketing promotion 112 Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková Table 8. Product and marketing innovations implemented in examined destinations. Innovation Accommodation and ski packages Amusement paths for children (Wilderness in Tatras, Marmot country) Destination card (Liptov Card ) Destination card (Tatry Card ) Loyalty programme Gopass Ski pass discounts in accommodation facilities Building an appealing brand in both destination regions Brand Liptov coherent with brand Liptov used for local products Extensive marketing campaign in Poland Utilisation of innovative communication tools, including digital marketing (web side, social networks) Life-style journal about High Tatras Active cooperation with tour operators from Russian speaking countries Online selling of Gopass and ski passes Stakeholder TMR, DMO DMO DMO, Cluster DMO TMR TMR TMR, Cluster DMO, Cluster TMR TMR TMR DMO TMR Source: own study. implementation changed the internal processes in the destination management organisation. The following processes were innovated: –– destination product development, –– destination marketing, –– booking of services (accommodation, cultural events), –– management of visitors’ cards, –– customer relationship management (CRM), –– management of stakeholders’ destination. Conclusions and policy implications Both examined destinations are mountain areas with the rich cultural and historical heritage, and unique water resources suitable for building aqua parks, wellness and spa resorts. Both destinations have good potential for tourist development and they stand a chance of being recognisable on the international market. Fig. 5 shows the impact of innovation processes. In the High Tatras and Liptov, institutional innovations were introduced in the form of formal networks in the tourism sector via the introduction of destination management organisations. Although tourism associations and clusters were also present before, the creation of DMOs offered better possibilities for networking and led to stronger collaboration. The collaboration of organisations within the tourism sector as well as relevant stakeholders enhances the development and implementation of the product, marketing, process and managerial innovations and contributes to the creation of new formal and informal networks. Innovations, formal and informal networks contribute to the development of the tourism sector that orchestrate the overall development of tourism destinations in the High Tatras and Liptov. The Liptov region was visited in 2016 by almost half a million of tourists. The number of visitors increased from the last year by almost 13% and the summer 2016 was the best ever for the region. Networking and innovations lead to the higher number of visitors and thus positively influence the stakeholders. Moreover, these two destinations seized the opportunity to organise major events such as the 27th Winter Universiade and FIS Ladies World Cup 2016. Slovak tourism destinations thanks to hosting important sport events are appearing on the map of significant tourism destinations of international relevance. Since innovations contribute to the higher competitiveness of tourist destinations, via the synergy effect, it brings also economic benefits to the region (higher direct and indirect tourism income, direct and indirect job creation etc.). We believe that further research needs to be conducted within existing informal networks as there is a lack of evidence in the contribution of social capital to higher innovativeness in tourism with particular attention paid to tourist destinations. In order to assess the social capital, the outcomes of the research conducted by the commission of the European Science Foundation within the European Social Survey should be taken into account. The link between social skills, networking and collaboration within a tourist destination as a stimulus of innovation processes leading Innovations and networking fostering tourist destination development in Slovakia 113 Fig. 5. The impact of innovations and networking on tourism destination development. Source: own study. to higher competitiveness in tourism should be investigated. The access to social capital means higher competitiveness and social solidarity, while the lack of social capital is related to the lack of economic success and consequently leads to social exclusion. Moreover, social capital strengthens innovations and thus supports economic growth and development. Therefore, further research should be focused on informal networks, leadership and co-operative relations beyond the tourism sector framework aligned with the destination governance concept. According to the prognosis of the World Travel and Tourism Council (WTTC), the tourism sector and its impact on overall development will be growing in Slovakia. In the ten-year WTTC prognosis, a direct contribution of travel and tourism to GDP is expected to grow by 3.8% p.a. to 2.7% of GDP by 2026 in Slovakia (WTTC 2016). Despite the positive prognosis of the WTTC there is a large space for improvement. The Slovak tourism sector is still underdeveloped and not enough attention is paid to its smart and sustainable development. In the long-term growth forecast for 2014–2024, Slovakia will rank 148th out of 184 countries. This situation cannot be perceived as satisfactory. Policy and decision makers at the local, regional and national level in Slovakia should create efficient support tools for tourist development. The Slovak tourism sector, despite its potential, lacks systematic approach integrating all spatial levels and all relevant stakeholders. The collaboration among stakeholders and efficient networks (formal as well as informal) can act as building blocks for successful tourist development in Slovakia. Innovations are crucial for the development of the tourism sector and for ensuring the competitiveness in tourist destinations. As discussed in the paper, the implementation of institutional innovations in the form of establishing destination management organisations, supported the creation of formal and informal networks and together with tourism associations and clusters generated additional innovations. Due to the synergy effect of innovations, collaboration and networking, the development of the tourism sector is orchestrating the overall development of tourist destinations by an increase in employment, local or regional economics and well-being. The reasons for implementing innovations can vary in each region. Every municipality, region or destination has its own history, tradition, policy and leadership style. They are influenced by the external environment, state and European policy. It is therefore quite challenging to assess which innovations are in general the most suitable for orchestrating tourist destination development with important economic benefits. The aim of implemented innovations should be to increase the competitiveness of the destination on the tourism market with the respect to the principles of sustainable and smart tourist development. Thus, they should meet the economic, environmental and social needs of a tourist destination and entire regions to ensure their smart and sustainable development. Although the institutional innovation in the form of national law results in the collaboration of tourism stakeholders in destinations, which we believe is a positive feature, this top-down approach stimulated the creation of weak destination management organisations as well. We are of the opinion that without an impulse in the form of the state subsidy, several of these organisations would never be established and their 114 Tomáš Gajdošík, Zuzana Gajdošíková, Vanda Maráková, Kamila Borseková perspective for sustainability and a long-term efficient operation would be low, which is not conducive to smart and sustainable tourist development. However, we can expect merges of several small DMOs covering an area with limited potential in tourism in the near future. Acknowledgements The research was financially supported by the research project VEGA 1/0809/17 Reengineering of destination management organisations and good destination governance conformed to principles of sustainable development. References Baggio R., 2008. Network analysis of a tourism destination. Dissertation Thesis. The University of Queensland, Queensland. Baggio R., Scott N., Cooper C., 2010. Network science. A review focused on tourism. Annals of Tourism Research 37(3): 802–827. Beni M., 2003. Globalizaçâo do Turismo; Megaicndências do Sector e a Realidade Brasileira. Editora Aleph, Sao Paulo. Beritelli P., Strobl A., Peters M., 2013. Interlocking directorships against community closure: A trade-off for development in tourist destinations. Tourism Review 68(1): 21–33. Bieger T., Beritelli P., 2012. Management von Destinationen. Oldenbourg, München. Borgatti S.P., Everett M.G., Freeman L.C., 2002. Ucinet for Windows: Software for social network analysis. Analytic Technologies, Harvard, MA. Borseková K., Vaňová A., Vitálišová K., 2017. Smart specialization for smart spatial development: Innovative strategies for building competitive advantages in tourism in Slovakia. Socio-Economic Planning Sciences 58(2): 39–50. Britton S., 1991. Tourism, capital, and place: Towards a critical geography of tourism. Environ Plan D Environment and Planning D: Society and Space 9(4): 452–475. Butler R.W., 1980. The concept of a tourism area cycle of evolution: Implication for management of resources. The Canadian Geographer 24(1): 5–12. Camisón C., Monformt-Mir V.M., 2012. Measuring innovation in tourism from the Schumpeterian and the dynamic-capabilities perspectives. Tourism Management 33(4): 776–789. Capone F., 2004. Regional competitiveness in tourism local systems. 44th European congress of the ERSA. Regions and fiscal federalism, Universidad de Porto. Castellaci F., 2008. Innovation and the competitiveness of industries: Comparing the mainstream and the evolutionary approaches. Journal of Technological Forecasting & Social Change 75(7): 984–1006. Del Chiappa G., Presenza A., 2013. The use of network analysis to assess relationships among stakeholders within a tourism destination. An empirical investigation on costa Smeralda-Gallura (Italy). Tourism Analysis 18(1): 1–13. Fernández-Cavia J., Rovira C., Díaz-Luque P., Cavaller V., 2014. Web Quality Index (WQI) for official tourist destination websites. Proposal for an assessment system. Tourism Management Perspectives 9(1): 5–13. Flagestad A., Hope C.A., 2001. Strategic success in winter sports destinations: A sustainable value creation perspective. Tourism Management 22(5): 445–461. Freeman R.E., 1984. Strategic management: A stakeholder approach. Pitman, Boston, MA. Grama C., Baggio R., 2014. A network analysis of Sibiu County, Romania. Annals of Tourism Research 47(2): 77–95. Gray B., 1989. Collaborating: Finding common ground for multiparty problems. Jossey-Bass, San Francisco. Hall C.M., 1999. Rethinking collaboration and partnership: A public policy perspective. Journal of Sustainable Tourism 7(3–4): 274–289. Hall C.M., Williams A.M., 2008. Tourism and innovation. Routledge, London. Hankinton G., 2004. Relational network brands: Towards a conceptual model of place brands. Journal of Vacation Marketing 10(2): 109–121. Hanneman A., Riddle M., 2005. Introduction to social network methods. University of California, Riverside, CA. Hjalager A., 2002. Repairing innovation defectiveness in tourism. Tourism Management 23(5): 465–474. Hjalager A., 2010. A review of innovation research in tourism. Tourism Management 31(1): 1–12. Hjalager A., Flagestad A., 2012. Innovations in well-being tourism in the Nordic countries. Current Issues in Tourism 15(8): 725–740. Jafari J. et al., 2003. Encyclopedia of tourism. Routledge, London. Jamal T.B., Getz D., 1995. Collaboration theory and community tourism planning. Annals of Tourism Research 22(1): 186–204. Kidová A., 2017. Map of destination management organisations in Slovakia. Geographical institute of Slovak academy of Sciences, Bratislava. Kučerová J., 2015. Plánovanie a politika v cieľových miestach cestovného ruchu (Planning and policy in tourist destinations). Belianum, Banská Bystrica. Lynch P., Morrison A., 2007. The role of networks. In: Michael E. (ed.), Micro clusters and networks: The growth of tourism. Elsevier, Oxford: 43–62. Mayer M., 2009. Innovation as a success factor in tourism: Empirical evidence from western Austrian cable-car companies. Erdkunde 63(2): 123–139. Navrátil J., Pícha K., Martinát S., Knotek J., Kučera T., Balounová Z., Gilliam V., Švec R., Rajchard J., 2013. A model for the identification of areas favourable for the development of tourism: A case study of the Šumava mts. and South Bohemia tourist regions. Moravian Geographical Reports 21(1): 25–39. Novelli M., Schmitz, B., Spencer, T., 2006. Networks, clusters and innovation in tourism: A UK experience. Tourism Management 27(6): 1141–1152. O’Sullivan D., 2014. Spatial network analysis. In: Handbook of Regional Science. Springer, London: 1253–1273. OECD 2004. A conceptual and dynamic approach to innovation in tourism – OECD. Retrieved February 18, 2016 from http://www.oecd.org/cfe/ tourism/34267921.pdf. Oslo manual: Guidelines for collecting and interpreting innovation data 2005. Organisation for Economic Co-operation and Development, Paris. Innovations and networking fostering tourist destination development in Slovakia Ottenbacher M., Gnoth J., 2005. How to develop successful hospitality innovation. Cornell Hotels and Restaurant Administration Quarterly 46(2): 205–222. Pechlaner et al., 2012. Destination management organizations as interface between destination governance and corporate governance. Anatolia 23(2): 151–168. Porter M.E., 1998. Clusters and the new economics of competition. Harvard Business Review 76(6): 77–90. Presenza A., Cipollina M., 2010. Analysing tourism stakeholders’ networks. Tourism Review 65(4): 17–30. Racherla P., Hu C., Hyun M.Y., 2008. Exploring the role of innovative technologies in building a knowledge-based destination. Current Issues in Tourism 11(5): 407–428. Rodrigues A.B., 2001. Turismo rural. Contexto, São Paulo. Romão J., Guerreiro J., Rodrigues P., 2013. Regional tourism development: Culture, nature, life cycle and attractiveness. Current Issues in Tourism 16(6): 517–534. Scott J., Carrington P.J., 2011. The SAGE handbook of social network analysis. Sage, London. Scott N., Baggio R., Cooper Ch., 2008. Network analysis and tourism: From theory to practice. Cromwell Press Ltd., Clevedon. 115 Sheehan L.R., Ritchie J.B., 2005. Destination stakeholders exploring identity and salience. Annals of Tourism Research 32(3): 711–734. Soteriades M., 2012. Tourism destination marketing: Approaches improving effectiveness and efficiency. Journal of Hospitality and Tourism Technology 3(2): 107–120. UNWTO 2007. A practical guide to tourism destination management. World Tourism Organization, Madrid. UNWTO 2010. Survey on destination governance: Evaluation report. World Tourism Organization, Madrid. van der Zee E., Vanneste D., 2015. Tourism networks unravelled; a review of literature on networks in tourism management studies. Tourism Management Perspectives 15(2): 46–56. Wang Y., Krakover S., 2008. Destination marketing: Competition, cooperation or coopetition? International Journal of Contemporary Hospitality Management 20(2): 126–141. Wasserman S., Faust K., 1994. Social network analysis: Methods and applications. Cambridge University Press, Cambridge. World Travel and Tourism Council, 2016. Travel & Tourism. Economic Impact 2016. Slovakia.
https://openalex.org/W3014070799
https://europepmc.org/articles/pmc7177949?pdf=render
English
null
The Role of Chondrocyte Hypertrophy and Senescence in Osteoarthritis Initiation and Progression
International journal of molecular sciences
2,020
cc-by
9,476
Received: 10 March 2020; Accepted: 27 March 2020; Published: 29 March 2020 Abstract: Osteoarthritis (OA) is the most common joint disease that causes pain and disability in the adult population. OA is primarily caused by trauma induced by an external force or by age-related cartilage damage. Chondrocyte hypertrophy or chondrocyte senescence is thought to play a role in the initiation and progression of OA. Although chondrocyte hypertrophy and cell death are both crucial steps during the natural process of endochondral bone formation, the abnormal activation of these two processes after injury or during aging seems to accelerate the progression of OA. However, the exact mechanisms of OA progression and these two processes remain poorly understood. Chondrocyte senescence and hypertrophy during OA share various markers and processes. In this study, we reviewed the changes that occur during chondrocyte hypertrophy or senescence in OA and the attempts that were made to regulate them. Regulation of hypertrophic or senescent chondrocytes might be a potential therapeutic target to slow down or stop OA progression; thus, a better understanding of the processes is required for management. Keywords: osteoarthritis; articular cartilage; chondrocyte; hypertrophy; senescence The Role of Chondrocyte Hypertrophy and Senescence in Osteoarthritis Initiation and Progression Yeri Alice Rim 1 , Yoojun Nam 1 and Ji Hyeon Ju 1,2,* 1 Catholic iPSC Research Center, College of Medicine, The Catholic University of Korea, Seoul 06591, Korea; llyerill0114@gmail.com (Y.A.R.); givingtreemax@gmail.com (Y.N.) 2 Division of Rheumatology, Department of Internal Medicine, Seoul St. Mary’s Hospital, Institute of Medical Science, College of Medicine, The Catholic University of Korea, Seoul 06591, Korea * Correspondence: juji@catholic.ac.kr; Tel.: +82-2-2258-6895 1 Catholic iPSC Research Center, College of Medicine, The Catholic University of Korea, Seoul 06591, Korea; llyerill0114@gmail.com (Y.A.R.); givingtreemax@gmail.com (Y.N.) 2 Division of Rheumatology, Department of Internal Medicine, Seoul St. Mary’s Hospital, Institute of Medical Science, College of Medicine, The Catholic University of Korea, Seoul 06591, Korea * Correspondence: juji@catholic.ac.kr; Tel.: +82-2-2258-6895 Int. J. Mol. Sci. 2020, 21, 2358; doi:10.3390/ijms21072358 International Journal of Molecular Sciences International Journal of Molecular Sciences 2. Nature of Articular Cartilage Articular cartilage mainly refers to hyaline cartilage. Hyaline cartilage is a smooth, lubricated surface of the cartilage that is 2 to 4 mm thick [6]. It is composed of a dense layer of ECM that also contains chondrocytes and water. Chondrocytes are quiescent cells, yet they actively synthesize a large volume of ECM proteins such as collagen, hyaluronans or glycoproteins, and proteoglycans [7–9]. Because of the dense layer of ECM between each lacuna, which is the small space containing each chondrocyte in the cartilage, chondrocyte only occupies 1%–5% of the total volume of the cartilage tissue. Cartilage is an avascular and alymphatic tissue, and therefore chondrocytes rely on the diffusion of nutrients from the articular surface [10]. Given these environmental characteristics, the inner layers of cartilage maintain a low oxygen environment with a low turnover rate of chondrocytes. SRY-box 9 (SOX9) is the major transcription factor that regulates chondrogenesis. Chondrocytes express SOX9, which activates ECM genes such as collagen type II, alpha 1 (COL2A1), and aggrecan (ACAN). Other ECM genes, including collagen type IX and XI, were also shown to be regulated by SOX9 [11]. Collagens and other ECM molecules, especially hyaluronan, are water-retentive, and are responsible for the high water content of cartilage. Up to 80% of the wet weight of cartilage consists of water. Collagens form about 60% of the dry weight of the cartilage, which makes them the most abundant type of protein found in ECM [9]. Although the superficial layers of cartilage mostly consist of collagen type II, the terminally differentiated hypertrophic chondrocytes in the deep zone actively synthesize collagen type X. 1. Introduction The simple yet complicated architecture of the articular cartilage compromises its natural capacity for self-repair [1]. The cartilage, with an avascular and alymphatic nature, is composed of dense layers of extracellular matrix (ECM) containing embedded distributed chondrocytes [2]. Cartilage lines the surface of the bone and provides a low-friction surface by absorbing external pressure or stimuli, which enables painless joint movement [3]. Osteoarthritis (OA) is the most common joint disorder throughout the whole human population. The incidence of OA is expected to increase due to the aging population throughout the world. OA accompanies progressive degradation of the articular cartilage, which leads to a loss of joint mobility and function and eventually to a low quality of life in patients due to both pain and restricted lifestyles. OA is characterized by pain, joint stiffness, and joint swelling. Severe stages of OA also include symptoms such as osteophyte and synovitis formation, which are outgrowths of bone or synovial membrane followed by inflammation. ECM degradation caused by matrix-degrading enzymes, including aggrecanases (a disintegrin and metalloprotease with thrombospondin motifs (ADAMTSs)) and matrix metalloproteinases (MMPs), is a well-known characteristic of OA. Healthy chondrocytes usually display moderate metabolic activity and proliferation under normal conditions; however, some articular chondrocytes lose their differentiated phenotype under diseased conditions and enter an endochondral ossification (EO)-like Int. J. Mol. Sci. 2020, 21, 2358; doi:10.3390/ijms21072358 www.mdpi.com/journal/ijms 2 of 15 Int. J. Mol. Sci. 2020, 21, 2358 state of proliferation along with abnormal hypertrophic differentiation [4,5]. Cellular senescence can also occur alongside hypertrophy due to similar stimuli. Cellular senescence and hypertrophy share various markers and processes, and both events are reported to play a role in the development of OA. This review focuses on the hypertrophic changes and cellular senescence in chondrocytes under OA conditions. 3. OA and Chondrocyte Hypertrophy Chondrocyte hypertrophy and cell death are natural phenomena that usually occur during a developmental process called EO. Hypertrophic chondrocytes appear and play a crucial role in EO. Hyaline cartilage can be divided into two groups, (1) temporary and (2) permanent cartilage. Healthy cartilage is usually called permanent cartilage or resting chondrocytes, which are present in the articulating joint. Usually, permanent cartilage has a low proliferation rate and does not undergo terminal differentiation and EO under normal conditions [12]. Temporary cartilage is initially formed as cartilage, but the final product is bone. Unrestricted differentiation of precursor cells into the chondrocyte lineage does not lead to permanent cartilage but instead leads to bone [12]. Chondrocytes undergo active proliferation and generate a cascade of cells; whereas some of them undergo enlargement, others undergo hypertrophical changes and become hypertrophic chondrocytes. These cells increase their volume dramatically and the surroundings become mineralized to develop bone tissue [13]. The elastic nature of cartilage begins to change and harden through calcification. This makes it more difficult for the chondrocytes to receive nutrients, as most of the cells undergo apoptosis and leave small cavities within the tissue, which leaves enough room in the hardened bone for blood vessel invasion. Through this process, the cartilage turns into trabecular bone. However, the major highlight events of EO, such as chondrocyte proliferation, hypertrophic differentiation of chondrocytes, cell death, calcification or mineralization, blood vessel invasion, and chondrocyte apoptosis, occur equally in OA (Figure 1). Int. J. Mol. Sci. 2020, 21, 2358 decreases instead. This i like process in OA 3 of 15 he EO 3 of 15 he EO Figure 1. Schematic image of (a) endochondral ossification in the embryonic cartilage and (b) progression of osteoarthritis in the articular cartilage. Figure 1. Schematic image of (a) endochondral ossification in the embryonic cartilage and (b) progression of osteoarthritis in the articular cartilage. Figure 1. Schematic image of (a) endochondral ossification in the embryonic cartilage and (b) progression of osteoarthritis in the articular cartilage. Figure 1. Schematic image of (a) endochondral ossification in the embryonic cartilage and (b) progression of osteoarthritis in the articular cartilage. 4. OA and Chondrocyte Senescence OA and cellular senescence are characterized by various epigenetic changes due to altered cellular phenotype and disease progression. Therefore, cellular senescence is also a factor that is thought to play a significant role in the pathology of OA [23]. 3. OA and Chondrocyte Hypertrophy Cellular senescence can be caused by various factors, including external stimuli or stress, oncogene activation, radiation, oxidative stress, l l h d d b l DNA l OA h d h b f h Cell hypertrophy generally refers to an increase in cell size and volume. Hypertrophic differentiation of chondrocytes can also be characterized by the high expression of collagen type X, runt-related transcription factor 2 (RUNX2), and MMP13. Hyaline cartilage markers, such as aggrecan, collagen type II, and SOX9, are decreased in the hypertrophic cells. Although hypertrophical changes in chondrocytes are required in bone growth and development, the mechanism acts as a double-edged sword in diseased environments [14]. telomere length reduction, and abnormal DNA replication. OA chondrocytes exhibit features that are similar to senescent cells, such as telomere shortening and increased senescence-associated β- galactosidase (SA-βgal) activity. The term ‘cellular senescence’ was first mentioned in the 1960s by Leonard Hayflick and Paul Moorhead, after finding that in vitro cultured primary human fibroblasts have a restricted lifespan, which is approximately 50 cell divisions [24]. Senescent cells undergo Hypertrophic differentiation is also an important pathological change in OA cartilage [15]. Cartilage tissue fractures at sites that receive high mechanical stress, which is a process called ‘fibrillation’ [3]. During fibrillation in the cartilage, the tissue loses several ECM proteins, including proteoglycans and collagen fibers. Abnormal chondrocyte hypertrophy in articular cartilage is seen in cartilage destruction, which is a potential therapeutic target for OA treatment [16,17]. Hypertrophic chondrocytes start to express genes related to osteogenic differentiation and start to produce mineralized ECM proteins [18,19]. It is unclear if the abnormally differentiated hypertrophic chondrocytes give rise to fully differentiated bone cells; however, apoptosis of hypertrophic chondrocytes has been suggested in previous studies. Lacunar emptying within OA cartilage has been observed, which led to the loss of the articular cartilage [20,21]. After these changes, the subchondral bone generates osteophytes. Chondrocyte calcification by calcium deposition is the final stage of chondrocyte hypertrophy [12]. Previous studies report that the layer bordering the calcified cartilage and the non-calcified cartilage increases after the age of 60 years [22]. Although this layer of calcified cartilage tends to increase throughout the aging process, in some cases, the calcified region decreases instead. This indicates that the lower calcified cartilage is replaced by bone due to the EO-like process in OA. 4. OA and Chondrocyte Senescence OA and cellular senescence are characterized by various epigenetic changes due to altered cellular phenotype and disease progression. Therefore, cellular senescence is also a factor that is thought to play a significant role in the pathology of OA [23]. Cellular senescence can be caused by various 4 of 15 Int. J. Mol. Sci. 2020, 21, 2358 factors, including external stimuli or stress, oncogene activation, radiation, oxidative stress, telomere length reduction, and abnormal DNA replication. OA chondrocytes exhibit features that are similar to senescent cells, such as telomere shortening and increased senescence-associated β-galactosidase (SA-βgal) activity. The term ‘cellular senescence’ was first mentioned in the 1960s by Leonard Hayflick and Paul Moorhead, after finding that in vitro cultured primary human fibroblasts have a restricted lifespan, which is approximately 50 cell divisions [24]. Senescent cells undergo morphological changes such as flattening due to increased cell size, vacuolization, and accumulation of stress granules. Growth arrest, resistance to cell death, and altered gene expression that leads to the production of bioactive senescence-associated secretory phenotype (SASP) are induced due to these morphological changes [25,26]. SASPs are various inflammatory cytokines and growth factors secreted by senescent cells. Interestingly, SASPs that are reported to be produced by senescent cells (i.e., GM-CSF, IL-1α and β, IL-6, IL-7, IL-8, MCP-1, MCP-2, MMP1, MMP3, MMP10, MMP13, TIMP-2, and more) have been found in OA tissues or in synovial fluids [27]. SASPs are produced by senescent cells that secrete various inflammatory cytokines and growth factors [6,28]. The growth arrest of senescent cells is caused by the expression of cell cycle inhibitors that can be induced by various stimuli such as DNA damage, telomere shortening, and oncogene activation [25]. These genetic changes can be induced by other stimuli such as reactive oxygen species (ROS), mitotic stress, and inflammation. When chondrocytes are isolated from native culture and maintained in an in vitro culture as a monolayer, they easily lose their characteristics and express various senescence and dedifferentiation-related genes [29]. Senescence and dedifferentiation of chondrocytes during culture is accompanied by decreased expression of chondrocyte-specific proteins such as collagen type II, and other glycoproteins [30]. Although various cell types are involved in OA pathology, chondrocytes are primarily thought to play a major role in OA induction by cellular senescence [23]. Two different mechanisms of senescence are suggested in chondrocytes: (1) replicative senescence and (2) stress-induced premature senescence (SIPS) [31]. 5. Hypertrophy and Senescence-Related Markers in OA Several markers are known as chondrocyte hypertrophy and senescence markers. Both hypertrophy and senescence are not usually characterized by a specific set of markers, but are rather associated with a set of cellular phenotypes that often co-exist in a stressed environment (Figure 2) [23]. Although both chondrocyte hypertrophy and senescence are not clearly understood, markers such as IL-1β, collagen type X, RUNX2, vascular endothelial growth factor (VEGF), osteopontin, osteocalcin, and Indian hedgehog (IHH) are thought to be related [31,40]. The accumulation of MMPs induced by pro-inflammatory cytokines is also a positive senescence marker in OA chondrocytes. Int. J. Mol. Sci. 2020, 21, x 6 of 15 Figure 2. A theoretical model for the relationships of chondrocyte hypertrophy and chondrocyte senescence and the initiation or progression of osteoarthritis (OA). When trauma is induced by mechanical injury or aging, biomechanical stress and oxidative stress increase in the cartilage environment. Increased chondrocyte hypertrophy and senescence induce decreased anabolic signaling and increased catabolic signaling. This results in suppressed glycosaminoglycan (GAG) production by normal chondrocytes and increases the expression of extracellular matrix (ECM)- degrading enzymes and inflammatory cytokines. Figure 2. A theoretical model for the relationships of chondrocyte hypertrophy and chondrocyte senescence and the initiation or progression of osteoarthritis (OA). When trauma is induced by mechanical injury or aging, biomechanical stress and oxidative stress increase in the cartilage environment. Increased chondrocyte hypertrophy and senescence induce decreased anabolic signaling and increased catabolic signaling. This results in suppressed glycosaminoglycan (GAG) production by normal chondrocytes and increases the expression of extracellular matrix (ECM)-degrading enzymes and inflammatory cytokines. Figure 2. A theoretical model for the relationships of chondrocyte hypertrophy and chondrocyte senescence and the initiation or progression of osteoarthritis (OA). When trauma is induced by mechanical injury or aging, biomechanical stress and oxidative stress increase in the cartilage environment. Increased chondrocyte hypertrophy and senescence induce decreased anabolic signaling and increased catabolic signaling. This results in suppressed glycosaminoglycan (GAG) production by normal chondrocytes and increases the expression of extracellular matrix (ECM)- degrading enzymes and inflammatory cytokines. Figure 2. A theoretical model for the relationships of chondrocyte hypertrophy and chondrocyte senescence and the initiation or progression of osteoarthritis (OA). When trauma is induced by mechanical injury or aging, biomechanical stress and oxidative stress increase in the cartilage environment. Increased chondrocyte hypertrophy and senescence induce decreased anabolic signaling and increased catabolic signaling. 4. OA and Chondrocyte Senescence There are two main aging pathways, where p53/p21 is thought to act on replicative senescence as well as chondrocyte apoptosis, and p38/p16 induces SIPS [31,32]. Replicative senescence generally occurs after 30–40 passaging and shows features of senescent phenotype including enlarged flattened cells in culture and the expression of SA-βgal. Chondrocytes have also been shown to have telomere shortening with age [33]. Because of this, chondrocyte senescence is thought to be closely related to OA, which also occurs with age. Given the chondrocyte characteristics, it is unlikely that we will find evidence of cell division in normal chondrocytes. However, chondrocyte proliferation is a feature observed in OA as well as telomere shortening and SA-βgal expression [34]. Although replicative senescence is caused by telomere shortening, SIPS is caused by oxidative stress and DNA damage without the change in telomere length [35]. Cellular senescence can be caused by chronic stress in cells; however, there is still the possibility that post-traumatic OA can be characterized or triggered by the senescent cells within the damaged region [23]. In the tissue of one patient, however, senescent cells were found to be absent in normal cartilage and instead were observed near the osteoarthritic lesions [36]. When senescent cells were transplanted into the knee joint of wild type mice, an OA-like state was induced, which included pain, impaired mobility, and morphological and histological changes [37]. SASPs secreted by senescent cells can alter the tissue microenvironment and impair tissue regeneration induced by stem cells or progenitor cells, which can eventually lead to the senescence of the neighboring cells [38]. Beside chondrocytes, synovial fibroblasts are also thought to initiate or progress OA through senescence. Nuclear expression of p16 was detected in higher amounts in OA synovial tissue samples when compared to that of normal synovial tissues, which indicates senescence in OA synovial fibroblasts [39]. Senescent synovial fibroblasts induced by H2O2 or tumor necrosis factor-α (TNFα) expressed increased CDKN1A and CDGN2A and other pro-inflammatory SASP-associated factors such as IL-6, CXCL8, CCL2, and MMP3. Int. J. Mol. Sci. 2020, 21, 2358 5 of 15 5. Hypertrophy and Senescence-Related Markers in OA This results in suppressed glycosaminoglycan (GAG) production by normal chondrocytes and increases the expression of extracellular matrix (ECM)-degrading enzymes and inflammatory cytokines. RUNX2 is the main transcription factor that is involved in hypertrophic chondrocyte differentiation and early osteogenesis [48,59]. One of the hallmarks of OA is the upregulation of RUNX2. Therefore, RUNX2 is assumed to be a major transcriptional factor that directly regulates the expression of matrix degradation enzymes in the damaged articular cartilage [60]. When the destabilization of the medial meniscus (DMM) osteoarthritis model was induced in RUNX2 knockout mice, the gene expression of matrix degradation enzymes (i.e., MMP9, MMP13, ADAMTS4, ADAMTS5, ADAMTS7, and ADAMTS12) was significantly reduced compared with DMM-induced Cre-negative control. The deletion of RUNX2 in DMM-induced mice decreased MMP13 protein levels in the articular cartilage. Cells expressing ectopic RUNX2 showed a senescent-like phenotype that was characterized by an enlarged and flattened morphology and β-galactosidase staining; p53 Collagen type X is a major marker used to detect chondrocyte hypertrophy. This type of collagen is usually not expressed in healthy articular cartilage, though it is thought to play a role in the early stage of endochondral bone formation since it has been detected at sites of hypertrophic chondrocyte regions and calcification. Protein and mRNA levels of collagen type X are also detected in human OA cartilage but their expression shows significant local variation [12,16,41–43]. Despite the overall expression of collagen type X, alpha 1 (COL10A1) was increased in human OA cartilage, and the expression was higher in the relatively less degenerated area [44]. Isolated chondrocytes cultured in vitro lose their normal morphology and start synthesizing collagen type X after several passages [29,45]. For example, whereas passage 2 chondrocytes show high levels of collagen type II and low levels of collagen type X, passage 6 chondrocytes show the opposite characteristics with increased collagen type X [31]. The exact Int. J. Mol. Sci. 2020, 21, 2358 6 of 15 mechanism of senescence and collagen type X is not fully understood, but its expression is thought to be the main reason for chondrocyte dedifferentiation [46]. Elevated levels of MMP13 have been reported in OA cartilage [47,48]. The gene level of MMP13 was highly expressed in OA cartilage when compared to normal levels. Elevated expression of MMP13 and collagen type X was reported after inducing knee joint stability by the transection of knee ligaments and meniscus removal in mouse OA models [49]. 5. Hypertrophy and Senescence-Related Markers in OA MMP13 is also thought to play a central role in the irreversible degradation of collagen type II in OA [50,51]. Knockdown of activin-like kinase 5 (ALK5) was reported to increase MMP13 mRNA levels [52]. TGFβ is important for the maintenance and protection of healthy articular cartilage; therefore, it is commonly used in the in vitro culture of chondrocytes or chondrogenesis. The binding of TGFβ to its receptor recruits the type I receptor ALK5. In turn, this molecule complex phosphorylates the smad2 and 3 in the smad2/3 pathway, which is known to repress chondrocyte terminal differentiation [53,54]. Another alternative receptor ALK1 activates the smad1/5/8 signaling pathway, which is known to induce chondrocyte terminal differentiation [55,56]. The change in the ALK1/ALK5 ratio during aging or OA development may result in increased dominance of the smad1/5/8 signaling in chondrocytes and induce a hypertrophic-like state by high levels of MMP13. MMP13 is also a central regulator of chondrocyte senescence [57]. Chondrocytes undergoing senescence produce ECM-degrading enzymes such as MMP13, which promotes the degradation of the main protein of the articular cartilage, aggrecan, and collagen type II. MMP13 produced by cultured chondrocytes isolated from older donors appeared to be increased [58]. Changes in chondrocyte function (e.g., decreased anabolic activity and increased catabolic activity) are thought to be related to MMP13 production. RUNX2 is the main transcription factor that is involved in hypertrophic chondrocyte differentiation and early osteogenesis [48,59]. One of the hallmarks of OA is the upregulation of RUNX2. Therefore, RUNX2 is assumed to be a major transcriptional factor that directly regulates the expression of matrix degradation enzymes in the damaged articular cartilage [60]. When the destabilization of the medial meniscus (DMM) osteoarthritis model was induced in RUNX2 knockout mice, the gene expression of matrix degradation enzymes (i.e., MMP9, MMP13, ADAMTS4, ADAMTS5, ADAMTS7, and ADAMTS12) was significantly reduced compared with DMM-induced Cre-negative control. The deletion of RUNX2 in DMM-induced mice decreased MMP13 protein levels in the articular cartilage. Cells expressing ectopic RUNX2 showed a senescent-like phenotype that was characterized by an enlarged and flattened morphology and β-galactosidase staining; p53 signaling was required for this process [61]. A characteristic feature of hypertrophy and OA cartilage is the increased production of VEGF. VEGF induces the migration of endothelial cells by chemotactic actions and induces angiogenesis in vivo. 5. Hypertrophy and Senescence-Related Markers in OA Chondrocytes expressing p53 showed similar morphology to OA chondrocytes and underwent apoptosis, whereas downregulation of p53 expression prevented chondrocytes from undergoing senescence or apoptosis [71]. Chondrocytes expressing p16 lose their normal phenotype and express increased inflammatory cytokines and MMPs such as IL-1β, IL-6, and MMP13 [46]. In chondrocytes isolated from cadaveric donors without evidence of OA between 17 to 72 years old, age was found to be responsible for 27% of the variation in p16 levels. However, loss of p16 was not critical enough to prevent the development of age-induced or injury-induced traumatic OA [72]. Collagen type II expression is negatively correlated with p21 expression [73]. Oxidative stress caused by ROS is a major regulator in cell senescence. ROS induce multiple genes that spark senescence or dedifferentiation in chondrocytes. The main ROS reported to induce cartilage damage in chondrocytes are hydrogen peroxide (H2O2) and peroxynitrite [74,75]. Degenerated cartilage regions in OA patients had significantly lower anti-oxidative potential than the intact cartilage region, confirming increased oxidative damage in degenerated cartilage [76]. Chondrocytes treated with H2O2 showed shorter telomere length in vitro. When OA cartilage tissue was cultured with H2O2 addition, the number of glycosaminoglycans (GAGs) gradually decreased in a time-dependent manner. However, the addition of an anti-oxidative agent reversed these effects, inhibited GAG loss, and maintained telomere length. H2O2 addition significantly increased p21 expression in both early and late passage chondrocytes [77]. The expression started to increase one hour after treatment and reached its highest level after three hours, and the change in early passage chondrocytes was nine-fold higher than the initial level. Growth factors such as BMPs and other cytokines are reported to play a role in chondrocyte hypertrophy. Several BMPs enhance chondrogenesis; however, dysregulated levels of BMPs may induce calcification of cartilage and accelerate cartilage degradation. During OA development, chondrocytes reportedly express BMPs [9,78]. The BMP canonical smad1/5/8 pathway is thought of as an inducer of chondrocyte hypertrophy and EO [79]. Cytokines such as IL-6 and IL-8 are known to induce growth arrest of senescent cells when secreted into the cell microenvironment. IL-1β is associated with chondrocyte dedifferentiation, and induces repression of SOX9 and the synthesis of nitric oxide (NO) [31,80]. The induction of NO by IL-1β activates c-Jun N-terminal kinases (JNK) signaling, which is a well-known inducer of senescence [81]. IL-7 is a cytokine that is upregulated age-dependently [27]. 5. Hypertrophy and Senescence-Related Markers in OA VEGF also promotes angiogenesis in the cartilage tissue, which is related to the calcification of chondrocytes that can lead to dysregulated osteogenesis of the normal cartilage. Neoangiogenesis in the cartilage growth plate plays an important role in EO; therefore, VEGF is thought of as a critical mediator during EO. Carlevaro et al. investigated the expression of VEGF in mammalian and avian embryo long bone growth plates [62]. Although VEGF was observed in fully mature hypertrophic chondrocytes, it was completely absent in proliferating and quiescent cells in both chicken and mice. VEGF mRNA generates five different isoforms with a different number of amino acid residues by alternative splicing, labeled VEGF121, VEGF145, VEGF165, VEGF189, and VEGF206 [63]. Although only three types (VEGF121, VEGF165, and VEGF189) were detectable in both OA and normal cartilage, their receptors were only found in OA cartilage. OA cartilage secreted a significantly higher amount of VEGF than normal cartilage when cultured in vitro. VEGF expression is induced by mechanical forces on articular cartilage [64]. Therefore, it is thought that VEGF may regulate chondrocyte differentiation in response to mechanical forces [65]. After inducing DMM in mice, immunostaining showed increased levels of VEGF in articular cartilage, subchondral bone, synovium, and meniscus. IHH is also suggested as a marker for chondrocyte hypertrophy. IHH is a key signaling molecule that is expressed in pre-hypertrophic chondrocytes during growth plate development and that regulates 7 of 15 Int. J. Mol. Sci. 2020, 21, 2358 chondrocyte hypertrophy during EO [66]. The upregulation of IHH in postnatal cartilage showed an increased chondrocyte hypertrophy and degradation [67]. Activation of IHH downstream signaling pathways results in a reduction in articular cartilage thickness and proteoglycan content levels, while inhibition of IHH signaling reversed these effects [68,69]. IHH is reported to regulate the onset of chondrocyte hypertrophy by a negative feedback loop with parathyroid hormone-related protein (PTHrP) [70]. IHH activates the expression of PTHrP in particular chondrocytes and PTHrP signals through the receptor PTHR1 to inhibit chondrocyte hypertrophy and maintain the proliferating state. Mak et al. reported that IHH signaling can also directly regulate chondrocyte hypertrophy in the absence of PTHrP. IHH also promoted chondrocyte hypertrophy through Wnt/β-catenin activation and bone morphogenetic proteins (BMPs) signaling. Expressions of p16, p21, and p53 increase in senescent cells, which result in cell cycle arrest. Articular chondrocytes with increased expression of p16 and p21 exhibit increased senescence. 6. Cartilage Treatment and Regeneration Strategies Targeting Hypertrophy or Senescence Chondrocyte hypertrophy has been targeted in cartilage regeneration rather than treatment Several strategies have been attempted to inhibit chondrogenic hypertrophy. g p g yp p y The addition of TGFβ into an in vitro culture was reported to enhance chondrogenesis and reduce chondrocyte proliferation and hypertrophy [91–93]. TGFβ1 and TGFβ3 were mostly used in in vitro chondrogenesis in various studies [94]. Some studies report that there is no significant difference between TGFβ subtypes in suppressing hypertrophy during chondrogenesis; several other studies show that TGFβ1 is superior in suppressing hypertrophy [95–97]. Smad3 disrupted in mutant mice showed a progressive loss of articular cartilage and symptoms that resemble OA; however, the addition of TGFβ1 reversed these symptoms, suggesting that TGFβ/smad3 signals inhibit terminal hypertrophic differentiation of chondrocytes [54]. TGFβ1 also enhanced hyaline-like properties using BMP2 in bovine synovial explants, and suppressed chondrocyte hypertrophy [98]. PTHrP is a parathyroid hormone with various isoforms, and it is involved in the development of multiple tissues including skeletal development [99]. PTHrP increased GAG contents and expression of SOX9 and COL2A1 in pellets generated from bone marrow stromal cells and adipose-derived stromal cells, and downregulated the expression of COL10A1 and RUNX2 in a dose-dependent manner [100]. Between other isoforms, PTHrP(1-34) was reported to induce chondrogenesis and reduce hypertrophy in mesenchymal stromal cells [101]. Previous work done by Lee and colleagues showed that of four PTHrP isoforms (1-34, 1-86, 7-34, and 107-139), PTHrP(1-34) showed the most significant enhancement in chondrogenesis and suppression in hypertrophy in human bone marrow stromal cells [102]. Other studies also showed that PTHrP(1-34) increased the gene and protein expression of collagen type II. Collagen type X gene expression was less affected; however, PTHrP(1-34) downregulated the expression of IHH and alkaline phosphatase (ALP) [101]. Various groups have attempted to alter the levels of MMPs despite the negative musculoskeletal side effects in humans [103]. A broad spectrum of MMP inhibitors has been proposed and tested for OA [104]. Bertram et al. confirmed the activity of proteases and MMPs in chondrogenesis and how they influenced the early and late stages of chondrogenesis [105]. The addition of pan-MMP inhibitors suppresses chondrogenesis in a dose-dependent manner, which was confirmed by decreased levels of proteoglycan disposition, collagen type II and X staining levels, ALP activity, and SOX9 and COL2A1 expressions. The MMP13-specific inhibitor GG86/2 did not affect early chondrogenesis and did not alter the collagen type II accumulation. 5. Hypertrophy and Senescence-Related Markers in OA The expression of IL-7 is higher in cultured articular chondrocytes from older donors compared to that of the younger donors. Furthermore, increased levels of MMP13 are induced in chondrocytes after IL-7 treatment [82]. Advanced glycation end products (AGEs) are produced from reducing sugars and free amino groups of proteins, lipids, or nucleic acids by a non-enzymatic reaction [83,84]. AGE accumulation suppresses collagen type II production and also stimulates the expression of MMPs and ADAMTSs [85–87]. Collagen cross-linking is increased by AGEs, which enhances tissue stiffness [88,89]. Increased AGEs are pathogenic and induce oxidative stress and inflammation. AGEs also induce the secretion of TNF-α, and NO [86,90]. Int. J. Mol. Sci. 2020, 21, 2358 8 of 15 6. Cartilage Treatment and Regeneration Strategies Targeting Hypertrophy or Senescence GG86/2 partially reduced ALP activity. There are several drugs developed to target MMPs in OA [103]. Novartis described an active hydroxamate MMP inhibitor, CGS-27023A, which can be taken orally. This drug inhibited MMP1, MMP2, MMP3, MMP9, MMP12, and MMP13, and exhibited a chondroprotective effect in OA animal models. However, it was halted due to musculoskeletal side effects related to treatment duration during a phase I clinical trial [106]. Other companies attempted to develop MMP-13 inhibitors such as CP-54439 (Pfizer) and WAY-170523 (Wyeth Research), and were taken to phase II clinical trials. Although the trials of pan-MMP inhibitors had rather disappointing results, selective MMP13 inhibitors are under development with both improved delivery and reduced side effects [103]. Senescent cells secreting pro-inflammatory cytokines and matrix-degrading enzymes were targeted using senolytic (a combination of the words ‘senescence’ and ‘-lytic’) compounds that have emerged as a potential therapeutic [107]. Baker et al. attempted to eliminate senescent cells with a tailored drug through the expression of a transgene that is only expressed in P16(Ink4a)-positive cells [108]. The direct effect of a senolytic agent in OA was tested in an anterior cruciate ligament transection (ACLT) model induced in transgenic p16-3MR mice, which was designed to eliminate p16-positive cells with ganciclovir [28]. The senolytic compound UBX0101 was repeatedly delivered by intra-articular injections after two weeks of ACLT induction. Although the half-life of the compound is only 1.5 h, UBX0101 successfully reduced the senescent chondrocytes and the production of SASP factors and showed suppression of the progression of OA. Int. J. Mol. Sci. 2020, 21, 2358 9 of 15 7. Conclusions The molecular mechanisms of OA initiation and progression require considerable further study, despite significant progress in recent years. OA is mainly caused by trauma induced by an external force or cartilage damage accumulated during aging. During these processes, chondrocyte hypertrophy and senescence are thought to play a critical role in OA initiation or progression. However, the remaining question is: which came first, the chicken or the egg? There is still little understanding of whether these two independent processes (i.e., chondrocyte hypertrophy and senescence) are dependent on penetration in the other. Further study on which event is the cause or the effect should be conducted to better understand these processes. Investigating the effect of chondrocyte hypertrophy and/or senescence in OA would provide a promising research field in developing a potential therapeutic agent for OA treatment. Author Contributions: Y.A.R. proposed the concept of this review article and constructed the draft. Y.N. and Y.A.R. provided the necessary information. Y.A.R. and J.H.J. reviewed the draft. All authors have read and agreed to the published version of the manuscript. Funding: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2019R1I1A1A01060753 and 2019R1I1A1A01062060). This work was also supported by a grant of the Korea Healthcare Technology R&D project, Ministry for Health, Welfare and Family Affairs, Republic of Korea (HI16C2177 and HI18C1178). Conflicts of Interest: The authors declare no conflict of interest. References 1. Nam, Y.; Rim, Y.A.; Jung, S.M.; Ju, J.H. Cord blood cell-derived iPSCs as a new candidate for chondrogenic differentiation and cartilage regeneration. Stem Cell Res. Ther. 2017, 8, 16. [CrossRef] [PubMed] 2. Rim, Y.A.; Nam, Y.; Ju, J.H. Application of Cord Blood and Cord Blood-Derived Induced Pluripotent Stem Cells for Cartilage Regeneration. Cell Transpl. 2019, 28, 529–537. [CrossRef] [PubMed] 3. Jeon, O.H.; David, N.; Campisi, J.; Elisseeff, J.H. Senescent cells and osteoarthritis: A painful connection. J Clin. Invest. 2018, 128, 1229–1237. [CrossRef] [PubMed] 4. Dreier, R. Hypertrophic differentiation of chondrocytes in osteoarthritis: The developmental as degenerative joint disorders. Arthritis Res. Ther. 2010, 12, 216. [CrossRef] 5. Riedl, M.; Witzmann, C.; Koch, M.; Lang, S.; Kerschbaum, M.; Baumann, F.; Krutsch, W.; Docheva, D.; Alt, V.; Pfeifer, C. Attenuation of Hypertrophy in Human MSCs via Treatment with a Retinoic Acid Receptor Inverse Agonist. Int. J. Mol. Sci. 2020, 21, 1444. [CrossRef] 6. Abula, K.; Muneta, T.; Miyatake, K.; Yamada, J.; Matsukura, Y.; Inoue, M.; Sekiya, I.; Graf, D.; Economides, A.N.; Rosen, V.; et al. Elimination of BMP7 from the developing limb mesenchyme leads to articular cartilage degeneration and synovial inflammation with increased age. FEBS Lett. 2015, 589, 1240–1248. [CrossRef] 7. Goldring, M.B. Chondrogenesis, chondrocyte differentiation, and articular cartilage metabolism in health and osteoarthritis. Ther. Adv. Musculoskelet Dis. 2012, 4, 269–285. [CrossRef] 8. Archer, C.W.; Francis-West, P. The chondrocyte. Int. J. Biochem. Cell Biol. 2003, 35, 401–404. [CrossR 9. Akkiraju, H.; Nohe, A. Role of Chondrocytes in Cartilage Formation, Progression of Osteoarthritis and Cartilage Regeneration. J. Dev. Biol. 2015, 3, 177–192. [CrossRef] 10. Poole, C.A. Articular cartilage chondrons: Form, function and failure. J. Anat. 1997, 191 (Pt 1), 1–13. [CrossRef] 11. Oh, C.D.; Lu, Y.; Liang, S.; Mori-Akiyama, Y.; Chen, D.; de Crombrugghe, B.; Yasuda, H. SOX9 regulates multiple genes in chondrocytes, including genes encoding ECM proteins, ECM modification enzymes, receptors, and transporters. PLoS ONE 2014, 9, e107577. [CrossRef] [PubMed] 12. van der Kraan, P.M.; van den Berg, W.B. Chondrocyte hypertrophy and osteoarthritis: Role in initiation and progression of cartilage degeneration? Osteoarthr. Cartil. 2012, 20, 223–232. [CrossRef] [PubMed] 13. Mackie, E.J.; Ahmed, Y.A.; Tatarczuch, L.; Chen, K.S.; Mirams, M. Endochondral ossification: How cartilage is converted into bone in the developing skeleton. Int. J. Biochem. Cell Biol. 2008, 40, 46–62. [CrossRef] [PubMed] 14. van Donkelaar, C.C.; Wilson, W. Mechanics of chondrocyte hypertrophy. Biomech. Model. Mechanobiol. 2012, 11, 655–664. [CrossRef] [PubMed] 15. Abbreviations Abbreviations ECM Extracellular matrix OA Osteoarthritis ADAMTS A disintegrin and metalloprotease with thrombospondin motifs MMP Matrix metallopeptidase EO Endochondral ossification SOX9 SRY-box 9 COL2A1 Collagen type II, alpha 1 ACAN Aggrecan RUNX2 Runt-related transcription factor 2 SA-βgal Senescence-associated beta-galactosidase SASP Senescence-associated secretory phenotype ROS Reactive oxygen specie SIPS Stress-induced premature senescence TNF-α Tumor necrosis factor-α VEGF IHH Vascular endothelial growth factor Indian hedgehog COL10A1 Collagen type X, alpha 1 ALK5 Activin-like kinase 5 DMM Destabilization of the medial meniscus PTHrP BMP Parathyroid hormone-related protein Bone morphogenetic protein H2O2 GAG Hydrogen peroxide Glycosaminoglycan NO JNK AGE ALP Nitric oxide c-Jun N-terminal kinases Advanced glycation end product Alkaline phosphatase ACLT Anterior cruciate ligament transection Abbreviations ECM Extracellular matrix OA Osteoarthritis ADAMTS A disintegrin and metalloprotease with thro MMP Matrix metallopeptidase EO Endochondral ossification SOX9 SRY-box 9 COL2A1 Collagen type II, alpha 1 ACAN Aggrecan RUNX2 Runt-related transcription factor 2 SA-βgal Senescence-associated beta-galactosidase SASP Senescence-associated secretory phenotype ROS Reactive oxygen specie SIPS Stress-induced premature senescence TNF-α Tumor necrosis factor-α VEGF IHH Vascular endothelial growth factor Indian hedgehog COL10A1 Collagen type X, alpha 1 ALK5 Activin-like kinase 5 DMM Destabilization of the medial meniscus PTHrP BMP Parathyroid hormone-related protein Bone morphogenetic protein H2O2 GAG Hydrogen peroxide Glycosaminoglycan NO JNK AGE ALP Nitric oxide c-Jun N-terminal kinases Advanced glycation end product Alkaline phosphatase ACLT Anterior cruciate ligament transection 10 of 15 10 of 15 Int. J. Mol. Sci. 2020, 21, 2358 References Local clearance of senescent cells attenuates the development of post-traumatic osteoarthritis and creates a pro-regenerative environment. Nat. Med. 2017, 23, 775–781. [CrossRef] 29. Stokes, D.G.; Liu, G.; Dharmavaram, R.; Hawkins, D.; Piera-Velazquez, S.; Jimenez, S.A. Regulation of type-II collagen gene expression during human chondrocyte de-differentiation and recovery of chondrocyte-specific phenotype in culture involves Sry-type high-mobility-group box (SOX) transcription factors. Biochem. J. 2001, 360, 461–470. [CrossRef] 30. Benya, P.D.; Padilla, S.R.; Nimni, M.E. Independent regulation of collagen types by chondrocytes during the loss of differentiated function in culture. Cell 1978, 15, 1313–1321. [CrossRef] 31. Ashraf, S.; Cha, B.H.; Kim, J.S.; Ahn, J.; Han, I.; Park, H.; Lee, S.H. Regulation of senescence associated signaling mechanisms in chondrocytes for cartilage tissue regeneration. Osteoarthr. Cartil. 2016, 24, 196–205. [CrossRef] [PubMed] 32. Zhao, X.; Huang, P.; Li, G.; Feng, Y.; Zhendong, L.; Zhou, C.; Hu, G.; Xu, Q. Overexpression of Pitx1 attenuates the senescence of chondrocytes from osteoarthritis degeneration cartilage-A self-controlled model for studying the etiology and treatment of osteoarthritis. Bone 2020, 131, 115177. [CrossRef] [PubMed] 33. Martin, J.A.; Buckwalter, J.A. Telomere erosion and senescence in human articular cartilage chondr Gerontol. A Biol. Sci. Med. Sci. 2001, 56, B172–B179. [CrossRef] [PubMed] 34. Price, J.S.; Waters, J.G.; Darrah, C.; Pennington, C.; Edwards, D.R.; Donell, S.T.; Clark, I.M. The role of chondrocyte senescence in osteoarthritis. Aging Cell 2002, 1, 57–65. [CrossRef] 35. Muller, M. Cellular senescence: Molecular mechanisms, in vivo significance, and redox considerations. Antioxid. Redox Signal. 2009, 11, 59–98. [CrossRef] 36. Erusalimsky, J.D.; Kurz, D.J. Cellular senescence in vivo: Its relevance in ageing and cardiovascular disease. Exp. Gerontol. 2005, 40, 634–642. [CrossRef] 37. Xu, M.; Bradley, E.W.; Weivoda, M.M.; Hwang, S.M.; Pirtskhalava, T.; Decklever, T.; Curran, G.L.; Ogrodnik, M.; Jurk, D.; Johnson, K.O.; et al. Transplanted Senescent Cells Induce an Osteoarthritis-Like Condition in Mice. J. Gerontol. A Biol. Sci. Med. Sci. 2017, 72, 780–785. [CrossRef] 38. Acosta, J.C.; Banito, A.; Wuestefeld, T.; Georgilis, A.; Janich, P.; Morton, J.P.; Athineos, D.; Kang, T.W.; Lasitschka, F.; Andrulis, M.; et al. A complex secretory program orchestrated by the inflammasome controls paracrine senescence. Nat. Cell. Biol. 2013, 15, 978–990. [CrossRef] 39. Del Rey, M.J.; Valin, A.; Usategui, A.; Ergueta, S.; Martin, E.; Municio, C.; Canete, J.D.; Blanco, F.J.; Criado, G.; Pablos, J.L. Senescent synovial fibroblasts accumulate prematurely in rheumatoid arthritis tissues and display an enhanced inflammatory phenotype. Immun. Ageing 2019, 16, 29. [CrossRef] 40. References Lian, C.; Wang, X.; Qiu, X.; Wu, Z.; Gao, B.; Liu, L.; Liang, G.; Zhou, H.; Yang, X.; Peng, Y.; et al. Collagen type II suppresses articular chondrocyte hypertrophy and osteoarthritis progression by promoting integrin beta1-SMAD1 interaction. Bone Res. 2019, 7, 8. [CrossRef] [PubMed] 16. Kirsch, T.; Swoboda, B.; Nah, H. Activation of annexin II and V expression, terminal differentiation, mineralization and apoptosis in human osteoarthritic cartilage. Osteoarthr. Cartil. 2000, 8, 294–302. [CrossRef] 17. Tchetina, E.V.; Kobayashi, M.; Yasuda, T.; Meijers, T.; Pidoux, I.; Poole, A.R. Chondrocyte hypertrophy can be induced by a cryptic sequence of type II collagen and is accompanied by the induction of MMP-13 and collagenase activity: Implications for development and arthritis. Matrix Biol. 2007, 26, 247–258. [CrossRef] 18. Findlay, D.M.; Atkins, G.J. Osteoblast-chondrocyte interactions in osteoarthritis. Curr. Osteoporos Rep. 2014, 12, 127–134. [CrossRef] 19. Li, J.; Dong, S. The Signaling Pathways Involved in Chondrocyte Differentiation and Hypertrophic Differentiation. Stem Cells Int. 2016, 2016, 2470351. [CrossRef] 20. Sharif, M.; Whitehouse, A.; Sharman, P.; Perry, M.; Adams, M. Increased apoptosis in human osteoarthritic cartilage corresponds to reduced cell density and expression of caspase-3. Arthritis Rheum. 2004, 50, 507–515. [CrossRef] 21. Charlier, E.; Relic, B.; Deroyer, C.; Malaise, O.; Neuville, S.; Collee, J.; Malaise, M.G.; De Seny, D. Insights on Molecular Mechanisms of Chondrocytes Death in Osteoarthritis. Int. J. Mol. Sci. 2016, 17, 2146. [CrossRef] [PubMed] 11 of 15 Int. J. Mol. Sci. 2020, 21, 2358 22. Lane, L.B.; Bullough, P.G. Age-related changes in the thickness of the calcified zone and the number of tidemarks in adult human articular cartilage. J. Bone Joint Surg. Br. 1980, 62, 372–375. [CrossRef] [PubMed] 23. McCulloch, K.; Litherland, G.J.; Rai, T.S. Cellular senescence in osteoarthritis pathology. Aging Cell 2017, 16, 210–218. [CrossRef] [PubMed] 24. Hayflick, L.; Moorhead, P.S. The serial cultivation of human diploid cell strains. Exp. Cell Res. 1961, 25, 585–621. [CrossRef] 25. Vinatier, C.; Dominguez, E.; Guicheux, J.; Carames, B. Role of the Inflammation-Autophagy-Senescence Integrative Network in Osteoarthritis. Front. Physiol. 2018, 9, 706. [CrossRef] [PubMed] 26. Kuilman, T.; Michaloglou, C.; Mooi, W.J.; Peeper, D.S. The essence of senescence. Genes Dev. 2010, 24, 2463–2479. [CrossRef] [PubMed] 27. Greene, M.A.; Loeser, R.F. Aging-related inflammation in osteoarthritis. Osteoarthr. Cartil. 2015, 23, 1966–1971. [CrossRef] 28. Jeon, O.H.; Kim, C.; Laberge, R.M.; Demaria, M.; Rathod, S.; Vasserot, A.P.; Chung, J.W.; Kim, D.H.; Poon, Y.; David, N.; et al. References Johnson, K.A.; van Etten, D.; Nanda, N.; Graham, R.M.; Terkeltaub, R.A. Distinct transglutaminase 2-independent and transglutaminase 2-dependent pathways mediate articular chondrocyte hypertrophy. J. Biol. Chem. 2003, 278, 18824–18832. [CrossRef] 41. Hennig, T.; Lorenz, H.; Thiel, A.; Goetzke, K.; Dickhut, A.; Geiger, F.; Richter, W. Reduced chondrogenic potential of adipose tissue derived stromal cells correlates with an altered TGFbeta receptor and BMP profile and is overcome by BMP-6. J. Cell Physiol. 2007, 211, 682–691. [CrossRef] [PubMed] 42. Pullig, O.; Weseloh, G.; Ronneberger, D.; Kakonen, S.; Swoboda, B. Chondrocyte differentiation in human osteoarthritis: Expression of osteocalcin in normal and osteoarthritic cartilage and bone. Calcif. Tissue Int. 2000, 67, 230–240. [CrossRef] [PubMed] Int. J. Mol. Sci. 2020, 21, 2358 12 of 15 43. Aigner, T.; Fundel, K.; Saas, J.; Gebhard, P.M.; Haag, J.; Weiss, T.; Zien, A.; Obermayr, F.; Zimmer, R.; Bartnik, E. Large-scale gene expression profiling reveals major pathogenetic pathways of cartilage degeneration in osteoarthritis. Arthritis Rheum. 2006, 54, 3533–3544. [CrossRef] [PubMed] 44. Fukui, N.; Ikeda, Y.; Ohnuki, T.; Tanaka, N.; Hikita, A.; Mitomi, H.; Mori, T.; Juji, T.; Katsuragawa, Y.; Yamamoto, S.; et al. Regional differences in chondrocyte metabolism in osteoarthritis: A detailed analysis by laser capture microdissection. Arthritis Rheum. 2008, 58, 154–163. [CrossRef] 45. Cha, B.H.; Lee, J.S.; Kim, S.W.; Cha, H.J.; Lee, S.H. The modulation of the oxidative stress response in chondrocytes by Wip1 and its effect on senescence and dedifferentiation during in vitro expansion. Biomaterials 2013, 34, 2380–2388. [CrossRef] 46. Philipot, D.; Guerit, D.; Platano, D.; Chuchana, P.; Olivotto, E.; Espinoza, F.; Dorandeu, A.; Pers, Y.M.; Piette, J.; Borzi, R.M.; et al. p16INK4a and its regulator miR-24 link senescence and chondrocyte terminal differentiation-associated matrix remodeling in osteoarthritis. Arthritis Res. Ther. 2014, 16, R58. [CrossRef] 47. Shlopov, B.V.; Gumanovskaya, M.L.; Hasty, K.A. Autocrine regulation of collagenase 3 (matrix metalloproteinase 13) during osteoarthritis. Arthritis Rheum. 2000, 43, 195–205. [CrossRef] 48. Wang, X.; Manner, P.A.; Horner, A.; Shum, L.; Tuan, R.S.; Nuckolls, G.H. Regulation of MMP-13 exp by RUNX2 and FGF2 in osteoarthritic cartilage. Osteoarthr. Cartil. 2004, 12, 963–973. [CrossRef] 49. Kamekura, S.; Hoshi, K.; Shimoaka, T.; Chung, U.; Chikuda, H.; Yamada, T.; Uchida, M.; Ogata, N.; Seichi, A.; Nakamura, K.; et al. Osteoarthritis development in novel experimental mouse models induced by knee joint instability. Osteoarthr. Cartil. 2005, 13, 632–641. [CrossRef] 50. References Neuhold, L.A.; Killar, L.; Zhao, W.; Sung, M.L.; Warner, L.; Kulik, J.; Turner, J.; Wu, W.; Billinghurst, C.; Meijers, T.; et al. Postnatal expression in hyaline cartilage of constitutively active human collagenase-3 (MMP-13) induces osteoarthritis in mice. J. Clin. Invest. 2001, 107, 35–44. [CrossRef] 51. Little, C.B.; Barai, A.; Burkhardt, D.; Smith, S.M.; Fosang, A.J.; Werb, Z.; Shah, M.; Thompson, E.W. Matrix metalloproteinase 13-deficient mice are resistant to osteoarthritic cartilage erosion but not chondrocyte hypertrophy or osteophyte development. Arthritis Rheum. 2009, 60, 3723–3733. [CrossRef] [PubMed] 52. Blaney Davidson, E.N.; Remst, D.F.; Vitters, E.L.; van Beuningen, H.M.; Blom, A.B.; Goumans, M.J.; van den Berg, W.B.; van der Kraan, P.M. Increase in ALK1/ALK5 ratio as a cause for elevated MMP-13 expression in osteoarthritis in humans and mice. J. Immunol. 2009, 182, 7937–7945. [CrossRef] [PubMed] 53. Ferguson, C.M.; Schwarz, E.M.; Reynolds, P.R.; Puzas, J.E.; Rosier, R.N.; O’Keefe, R.J. Smad2 and 3 mediate transforming growth factor-beta1-induced inhibition of chondrocyte maturation. Endocrinology 2000, 141, 4728–4735. [CrossRef] [PubMed] 54. Yang, X.; Chen, L.; Xu, X.; Li, C.; Huang, C.; Deng, C.X. TGF-beta/Smad3 signals repress chondrocyte hypertrophic differentiation and are required for maintaining articular cartilage. J. Cell Biol. 2001, 153, 35–46. [CrossRef] [PubMed] 55. Goumans, M.J.; Mummery, C. Functional analysis of the TGFbeta receptor/Smad pathway through gene ablation in mice. Int. J. Dev. Biol. 2000, 44, 253–265. [PubMed] 56. Ito, H.; Akiyama, H.; Shigeno, C.; Nakamura, T. Noggin and bone morphogenetic protein-4 coordinately regulate the progression of chondrogenic differentiation in mouse clonal EC cells, ATDC5. Biochem. Biophys. Res. Commun. 1999, 260, 240–244. [CrossRef] [PubMed] 57. Zhang, W.; Zhang, C.; Luo, C.; Zhan, Y.; Zhong, B. Design, cyclization, and optimization of MMP13-TIMP1 interaction-derived self-inhibitory peptides against chondrocyte senescence in osteoarthritis. Int. J. Biol. Macromol. 2019, 121, 921–929. [CrossRef] 58. Forsyth, C.B.; Cole, A.; Murphy, G.; Bienias, J.L.; Im, H.J.; Loeser, R.F., Jr. Increased matrix metalloproteinase-13 production with aging by human articular chondrocytes in response to catabolic stimuli. J. Gerontol. A Biol. Sci. Med. Sci. 2005, 60, 1118–1124. [CrossRef] 59. Hecht, J.; Seitz, V.; Urban, M.; Wagner, F.; Robinson, P.N.; Stiege, A.; Dieterich, C.; Kornak, U.; Wilkening, U.; Brieske, N.; et al. Detection of novel skeletogenesis target genes by comprehensive analysis of a Runx2(-/-) mouse model. Gene Expr. Patterns 2007, 7, 102–112. [CrossRef] 60. Chen, D.; Shen, J.; Zhao, W.; Wang, T.; Han, L.; Hamilton, J.L.; Im, H.J. Osteoarthritis: Toward a comprehensive understanding of pathological mechanism. Bone Res. References 2017, 5, 16044. [CrossRef] 61. Kilbey, A.; Blyth, K.; Wotton, S.; Terry, A.; Jenkins, A.; Bell, M.; Hanlon, L.; Cameron, E.R.; Neil, J.C. Runx2 disruption promotes immortalization and confers resistance to oncogene-induced senescence in primary murine fibroblasts. Cancer Res. 2007, 67, 11263–11271. [CrossRef] [PubMed] 13 of 15 13 of 15 Int. J. Mol. Sci. 2020, 21, 2358 62. Carlevaro, M.F.; Cermelli, S.; Cancedda, R.; Descalzi Cancedda, F. Vascular endothelial growth factor (VEGF) in cartilage neovascularization and chondrocyte differentiation: Auto-paracrine role during endochondral bone formation. J. Cell Sci. 2000, 113 (Pt 1), 59–69. 63. Enomoto, H.; Inoki, I.; Komiya, K.; Shiomi, T.; Ikeda, E.; Obata, K.; Matsumoto, H.; Toyama, Y.; Okada, Y. Vascular endothelial growth factor isoforms and their receptors are expressed in human osteoarthritic cartilage. Am. J. Pathol. 2003, 162, 171–181. [CrossRef] 64. Beckmann, R.; Houben, A.; Tohidnezhad, M.; Kweider, N.; Fragoulis, A.; Wruck, C.J.; Brandenburg, L.O.; Hermanns-Sachweh, B.; Goldring, M.B.; Pufe, T.; et al. Mechanical forces induce changes in VEGF and VEGFR-1/sFlt-1 expression in human chondrocytes. Int. J. Mol. Sci. 2014, 15, 15456–15474. [CrossRef] [PubMed] 65. Nagao, M.; Hamilton, J.L.; Kc, R.; Berendsen, A.D.; Duan, X.; Cheong, C.W.; Li, X.; Im, H.J.; Olsen, B.R. Vascular Endothelial Growth Factor in Cartilage Development and Osteoarthritis. Sci. Rep. 2017, 7, 13027. [CrossRef] [PubMed] 66. Wei, F.; Zhou, J.; Wei, X.; Zhang, J.; Fleming, B.C.; Terek, R.; Pei, M.; Chen, Q.; Liu, T.; Wei, L. Activation of Indian hedgehog promotes chondrocyte hypertrophy and upregulation of MMP-13 in human osteoarthritic cartilage. Osteoarthr. Cartil. 2012, 20, 755–763. [CrossRef] [PubMed] 67. Lin, A.C.; Seeto, B.L.; Bartoszko, J.M.; Khoury, M.A.; Whetstone, H.; Ho, L.; Hsu, C.; Ali, S.A.; Alman, B.A. Modulating hedgehog signaling can attenuate the severity of osteoarthritis. Nat. Med. 2009, 15, 1421–1425. [CrossRef] [PubMed] 68. Beaupre, G.S.; Stevens, S.S.; Carter, D.R. Mechanobiology in the development, maintenance, and degeneration of articular cartilage. J. Rehabil. Res. Dev. 2000, 37, 145–151. 69. Mak, K.K.; Kronenberg, H.M.; Chuang, P.T.; Mackem, S.; Yang, Y. Indian hedgehog signals independently of PTHrP to promote chondrocyte hypertrophy. Development 2008, 135, 1947–1956. [CrossRef] 70. St-Jacques, B.; Hammerschmidt, M.; McMahon, A.P. Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev. 1999, 13, 2072–2086. [CrossRef] 71. Hashimoto, S.; Nishiyama, T.; Hayashi, S.; Fujishiro, T.; Takebe, K.; Kanzaki, N.; Kuroda, R.; Kurosaka, M. Role of p53 in human chondrocyte apoptosis in response to shear strain. Arthritis Rheum. References 2009, 60, 2340–2349. [CrossRef] [PubMed] 72. Diekman, B.O.; Sessions, G.A.; Collins, J.A.; Knecht, A.K.; Strum, S.L.; Mitin, N.K.; Carlson, C.S.; Loeser, R.F.; Sharpless, N.E. Expression of p16(INK) (4a) is a biomarker of chondrocyte aging but does not cause osteoarthritis. Aging Cell 2018, 17, e12771. [CrossRef] [PubMed] 73. Kim, H.J.; Park, S.R.; Park, H.J.; Choi, B.H.; Min, B.H. Potential predictive markers for proliferative capacity of cultured human articular chondrocytes: PCNA and p21. Artif. Organs 2005, 29, 393–398. [CrossRef] [PubMed] 74. De la Fuente, M.; Miquel, J. An update of the oxidation-inflammation theory of aging: The involvement of the immune system in oxi-inflamm-aging. Curr. Pharm. Des. 2009, 15, 3003–3026. [CrossRef] 75. Minguzzi, M.; Cetrullo, S.; D’Adamo, S.; Silvestri, Y.; Flamigni, F.; Borzi, R.M. Emerging Players at the Intersection of Chondrocyte Loss of Maturational Arrest, Oxidative Stress, Senescence and Low-Grade Inflammation in Osteoarthritis. Oxid. Med. Cell Longev. 2018, 2018, 3075293. [CrossRef] 76. Yudoh, K.; Nguyen v, T.; Nakamura, H.; Hongo-Masuko, K.; Kato, T.; Nishioka, K. Potential involvement of oxidative stress in cartilage senescence and development of osteoarthritis: Oxidative stress induces chondrocyte telomere instability and downregulation of chondrocyte function. Arthritis Res. Ther. 2005, 7, R380–R391. [CrossRef] 77. Brandl, A.; Hartmann, A.; Bechmann, V.; Graf, B.; Nerlich, M.; Angele, P. Oxidative stress induces senescence in chondrocytes. J. Orthop. Res. 2011, 29, 1114–1120. [CrossRef] 78. Nakase, T.; Miyaji, T.; Tomita, T.; Kaneko, M.; Kuriyama, K.; Myoui, A.; Sugamoto, K.; Ochi, T.; Yoshikawa, H. Localization of bone morphogenetic protein-2 in human osteoarthritic cartilage and osteophyte. Osteoarthr. Cartil. 2003, 11, 278–284. [CrossRef] 79. Retting, K.N.; Song, B.; Yoon, B.S.; Lyons, K.M. BMP canonical Smad signaling through Smad1 and Smad5 is required for endochondral bone formation. Development 2009, 136, 1093–1104. [CrossRef] 80. McNulty, A.L.; Estes, B.T.; Wilusz, R.E.; Weinberg, J.B.; Guilak, F. Dynamic loading enhances integrative meniscal repair in the presence of interleukin-1. Osteoarthr. Cartil. 2010, 18, 830–838. [CrossRef] Int. J. Mol. Sci. 2020, 21, 2358 14 of 15 14 of 15 81. Clancy, R.; Rediske, J.; Koehne, C.; Stoyanovsky, D.; Amin, A.; Attur, M.; Iyama, K.; Abramson, S.B. Activation of stress-activated protein kinase in osteoarthritic cartilage: Evidence for nitric oxide dependence. Osteoarthr. Cartil. 2001, 9, 294–299. [CrossRef] [PubMed] 82. Long, D.; Blake, S.; Song, X.Y.; Lark, M.; Loeser, R.F. Human articular chondrocytes produce IL-7 and respond to IL-7 with increased production of matrix metalloproteinase-13. Arthritis Res. Ther. 2008, 10, R23. [CrossRef] [PubMed] 83. References Zhang, W.; Randell, E.W.; Sun, G.; Likhodii, S.; Liu, M.; Furey, A.; Zhai, G. Hyperglycemia-related advanced glycation end-products is associated with the altered phosphatidylcholine metabolism in osteoarthritis patients with diabetes. PLoS ONE 2017, 12, e0184105. [CrossRef] [PubMed] 84. Uribarri, J.; Woodruff, S.; Goodman, S.; Cai, W.; Chen, X.; Pyzik, R.; Yong, A.; Striker, G.E.; Vlassara, H. Advanced glycation end products in foods and a practical guide to their reduction in the diet. J. Am. Diet. Assoc. 2010, 110, 911–916 e912. [CrossRef] 85. Huang, C.Y.; Lai, K.Y.; Hung, L.F.; Wu, W.L.; Liu, F.C.; Ho, L.J. Advanced glycation end products cause collagen II reduction by activating Janus kinase/signal transducer and activator of transcription 3 pathway in porcine chondrocytes. Rheumatology (Oxford) 2011, 50, 1379–1389. [CrossRef] 86. Nah, S.S.; Choi, I.Y.; Lee, C.K.; Oh, J.S.; Kim, Y.G.; Moon, H.B.; Yoo, B. Effects of advanced glycation end products on the expression of COX-2, PGE2 and NO in human osteoarthritic chondrocytes. Rheumatology (Oxford) 2008, 47, 425–431. [CrossRef] 87. Nah, S.S.; Choi, I.Y.; Yoo, B.; Kim, Y.G.; Moon, H.B.; Lee, C.K. Advanced glycation end products increases matrix metalloproteinase-1, -3, and -13, and TNF-alpha in human osteoarthritic chondrocytes. FEBS Lett. 2007, 581, 1928–1932. [CrossRef] 88. Chen, A.C.; Temple, M.M.; Ng, D.M.; Verzijl, N.; DeGroot, J.; TeKoppele, J.M.; Sah, R.L. Induction of advanced glycation end products and alterations of the tensile properties of articular cartilage. Arthritis Rheum. 2002, 46, 3212–3217. [CrossRef] 89. Bank, R.A.; Bayliss, M.T.; Lafeber, F.P.; Maroudas, A.; Tekoppele, J.M. Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. Biochem. J. 1998, 330 (Pt 1), 345–351. [CrossRef] 90. Liu, F.C.; Hung, L.F.; Wu, W.L.; Chang, D.M.; Huang, C.Y.; Lai, J.H.; Ho, L.J. Chondroprotective effects and mechanisms of resveratrol in advanced glycation end products-stimulated chondrocytes. Arthritis Res. Ther. 2010, 12, R167. [CrossRef] 91. Ballock, R.T.; Heydemann, A.; Wakefield, L.M.; Flanders, K.C.; Roberts, A.B.; Sporn, M.B. TGF-beta 1 prevents hypertrophy of epiphyseal chondrocytes: Regulation of gene expression for cartilage matrix proteins and metalloproteases. Dev. Biol. 1993, 158, 414–429. [CrossRef] 92. Tschan, T.; Bohme, K.; Conscience-Egli, M.; Zenke, G.; Winterhalter, K.H.; Bruckner, P. Autocrine or paracrine transforming growth factor-beta modulates the phenotype of chick embryo sternal chondrocytes in serum-free agarose culture. J. Biol. Chem. 1993, 268, 5156–5161. [PubMed] 93. Serra, R.; Johnson, M.; Filvaroff, E.H.; LaBorde, J.; Sheehan, D.M.; Derynck, R.; Moses, H.L. References Expression of a truncated, kinase-defective TGF-beta type II receptor in mouse skeletal tissue promotes terminal chondrocyte differentiation and osteoarthritis. J. Cell. Biol. 1997, 139, 541–552. [CrossRef] [PubMed] Rim, Y.A.; Nam, Y.; Park, N.; Jung, H.; Lee, K.; Lee, J.; Ju, J.H. Chondrogenic Differentiation from Induced Pluripotent Stem Cells Using Non-Viral Minicircle Vectors. Cells 2020, 9, 582. [CrossRef] [PubMed] 95. Mueller, M.B.; Fischer, M.; Zellner, J.; Berner, A.; Dienstknecht, T.; Prantl, L.; Kujat, R.; Nerlich, M.; Tuan, R.S.; Angele, P. Hypertrophy in mesenchymal stem cell chondrogenesis: Effect of TGF-beta isoforms and chondrogenic conditioning. Cells Tissues Organs 2010, 192, 158–166. [CrossRef] [PubMed] 96. Pei, M.; He, F.; Li, J.; Tidwell, J.E.; Jones, A.C.; McDonough, E.B. Repair of large animal partial-thickness cartilage defects through intraarticular injection of matrix-rejuvenated synovium-derived stem cells. Tissue Eng. Part A 2013, 19, 1144–1154. [CrossRef] [PubMed] 97. Cals, F.L.; Hellingman, C.A.; Koevoet, W.; Baatenburg de Jong, R.J.; van Osch, G.J. Effects of transforming growth factor-beta subtypes on in vitro cartilage production and mineralization of human bone marrow stromal-derived mesenchymal stem cells. J. Tissue Eng. Regen. Med. 2012, 6, 68–76. [CrossRef] 15 of 15 Int. J. Mol. Sci. 2020, 21, 2358 98. Shintani, N.; Siebenrock, K.A.; Hunziker, E.B. TGF-ss1 enhances the BMP-2-induced chondrogenesis of bovine synovial explants and arrests downstream differentiation at an early stage of hypertrophy. PLoS ONE 2013, 8, e53086. [CrossRef] 9. Frieling, J.S.; Lynch, C.C. Proteolytic Regulation of Parathyroid Hormone-Related Protein: Functio Implications for Skeletal Malignancy. Int. J. Mol. Sci. 2019, 20, 2814. [CrossRef] 100. Kim, Y.J.; Kim, H.J.; Im, G.I. PTHrP promotes chondrogenesis and suppresses hypertrophy from both bone marrow-derived and adipose tissue-derived MSCs. Biochem. Biophys. Res. Commun. 2008, 373, 104–108. [CrossRef] 101. Fischer, J.; Aulmann, A.; Dexheimer, V.; Grossner, T.; Richter, W. Intermittent PTHrP(1-34) exposure augments chondrogenesis and reduces hypertrophy of mesenchymal stromal cells. Stem Cells Dev. 2014, 23, 2513–2523. [CrossRef] 102. Lee, J.M.; Im, G.I. PTHrP isoforms have differing effect on chondrogenic differentiation and hypertrophy of mesenchymal stem cells. Biochem. Biophys. Res. Commun. 2012, 421, 819–824. [CrossRef] [PubMed] 103. Li, N.G.; Shi, Z.H.; Tang, Y.P.; Wang, Z.J.; Song, S.L.; Qian, L.H.; Qian, D.W.; Duan, J.A. New hope for the treatment of osteoarthritis through selective inhibition of MMP-13. Curr. Med. Chem. 2011, 18, 977–1001. [CrossRef] [ ] 104. Peterson, J.T. The importance of estimating the therapeutic index in the development of matrix metalloproteinase inhibitors. Cardiovasc. Res. 2006, 69, 677–687. [CrossRef] [PubMed] 105. References Bertram, H.; Boeuf, S.; Wachters, J.; Boehmer, S.; Heisel, C.; Hofmann, M.W.; Piecha, D.; Richter, W. Matrix metalloprotease inhibitors suppress initiation and progression of chondrogenic differentiation of mesenchymal stromal cells in vitro. Stem Cells Dev. 2009, 18, 881–892. [CrossRef] [PubMed] 106. Levitt, N.C.; Eskens, F.A.; O’Byrne, K.J.; Propper, D.J.; Denis, L.J.; Owen, S.J.; Choi, L.; Foekens, J.A.; Wilner, S.; Wood, J.M.; et al. Phase I and pharmacological study of the oral matrix metalloproteinase inhibitor, MMI270 (CGS27023A), in patients with advanced solid cancer. Clin. Cancer Res. 2001, 7, 1912–1922. 107. Collins, J.A.; Diekman, B.O.; Loeser, R.F. Targeting aging for disease modification in osteoarthritis. Curr. Opin. Rheumatol. 2018, 30, 101–107. [CrossRef] 108. Baker, D.J.; Childs, B.G.; Durik, M.; Wijers, M.E.; Sieben, C.J.; Zhong, J.; Saltness, R.A.; Jeganathan, K.B.; Verzosa, G.C.; Pezeshki, A.; et al. Naturally occurring p16(Ink4a)-positive cells shorten healthy lifespan. Nature 2016, 530, 184–189. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2896492379
https://documentserver.uhasselt.be//bitstream/1942/27547/1/Rahmouni%20et%20al%202018%20PeerJ.pdf
English
null
Seven new species of<i>Cichlidogyrus</i>Paperna, 1960 (Monogenea: Dactylogyridae) parasitizing the gills of Congolese cichlids from northern Lake Tanganyika
PeerJ
2,018
cc-by
20,988
Subjects Biodiversity, Parasitology, Taxonomy, Freshwater Biology Keywords Congo, Cichlidae, Interochromis, Petrochromis, Callochromis, Platyhelminthes, Tylochromis, C. koblmuelleri sp. nov., C. habluetzeli sp. nov., C. masilyai sp. nov., C. antoineparisellei sp. nov., Cardiopharynx, C. sergemorandi sp. nov., Cyphotilapia, C. adkoningsi sp. nov., Xenotilapia, C. salzburgeri sp. nov. How to cite this article Rahmouni et al. (2018), Seven new species of Cichlidogyrus Paperna, 1960 (Monogenea: Dactylogyridae) parasitizing the gills of Congolese cichlids from northern Lake Tanganyika. PeerJ 6:e5604; DOI 10.7717/peerj.5604 Seven new species of Cichlidogyrus Paperna, 1960 (Monogenea: Dactylogyridae) parasitizing the gills of Congolese cichlids from northern Lake Tanganyika Chahrazed Rahmouni1, Maarten P.M. Vanhove1,2,3,4 and Andrea Šimková1 1 Department of Botany and Zoology, Faculty of Science, Masaryk University, Brno, Czech Republic 2 Zoology Unit, Finnish Museum of Natural History, University of Helsinki, Helsinki, Finland 3 Laboratory of Biodiversity and Evolutionary Genomics, Department of Biology, University of Leuven, Leuven, Belgium 2 Zoology Unit, Finnish Museum of Natural History, University of Helsinki, Helsinki, Finland 3 Laboratory of Biodiversity and Evolutionary Genomics, Department of Biology, University of Leuven, Leuven, Belgium 4 Centre for Environmental Sciences, Research Group Zoology: Biodiversity and Toxicology, Universiteit Hasselt, Diepenbeek, Belgium 4 Centre for Environmental Sciences, Research Group Zoology: Biodiversity and Toxicology, Universiteit Hasselt, Diepenbeek, Belgium Submitted 2 March 2018 Accepted 16 August 2018 Published 23 October 2018 Corresponding author Chahrazed Rahmouni, rahmouni.chahrazed@gmail.com Academic editor Jean-Lou Justine Additional Information and Declarations can be found on page 31 DOI 10.7717/peerj.5604 Copyright 2018 Rahmouni et al. Distributed under Creative Commons CC-BY 4.0 INTRODUCTION With an estimated 3,000 species distributed from Central and South America, across Africa to Madagascar, and to the Middle East and the Indian subcontinent (Chakrabarty, 2004), cichlid fishes represent one of the most species-rich families of vertebrates, accounting for about 10% of total teleost diversity (Takahashi & Koblmüller, 2011; Wanek & Sturmbauer, 2015). The Great African Rift Lakes Malawi, Victoria, and Tanganyika harbour cichlid flocks exhibiting high morphological, ecological, and genetic diversity (Takahashi & Sota, 2016). The exact number of species inhabiting these three lakes is still unknown, but approximately 2,000 species have been described (Koblmüller, Sefc & Sturmbauer, 2008). Lake Tanganyika (LT), located in the Great Rift Valley in central East Africa, is the deepest and oldest lake in Africa (Cohen et al., 1997) and the second deepest and oldest lake in the world (Salzburger et al., 2005). It holds the most diverse cichlid assemblages, comprised of several lineages of mostly endemic species classified into more than 50 genera and 12–14 tribes (Snoeks, 2000; Koblmüller, Sefc & Sturmbauer, 2008; Takahashi & Sota, 2016). Over 250 cichlid species are known to inhabit this lake (Takahashi & Koblmüller, 2011). Cichlids represent a textbook model in evolutionary biology (Kocher, 2004). Their mechanisms of speciation by rapid radiation make them crucial to the study of biological diversification, dynamics, and functions (Barluenga & Meyer, 2010; Takahashi & Koblmüller, 2011). Cichlid monogeneans are a promising tool for elucidating the speciation of both fish and parasites (Vanhove et al., 2015, 2016). Among the 14 monogenean parasite genera known to infect cichlids, six (Urogyrus Bilong Bilong, Birgi & Euzet, 1994; Enterogyrus Paperna, 1963; Onchobdella Paperna, 1968; Scutogyrus Pariselle & Euzet, 1995; Cichlidogyrus (Dactylogyridae Bychowski, 1933), and Gyrodactylus von Nordmann, 1832 (Gyrodactylidae Van Beneden & Hesse, 1863)) were reported from African cichlids (Pariselle & Euzet, 2009; Pariselle et al., 2011; Mendoza-Palmero et al., 2017). More than 100 African and Levantine cichlid species have been investigated for the presence of monogenean parasites (Pariselle & Euzet, 2009; Vanhove et al., 2016). Cichlidogyrus Paperna, 1960 is the most species-rich genus and is mostly restricted to African and Levantine hosts (a few species were isolated from non-cichlid hosts, see for instance Birgi & Lambert (1986)) (Pariselle & Euzet, 2009). To date, 111 valid species of Cichlidogyrus have been recognized in African cichlids (see the overview of Tanganyikan and non-Tanganyikan species of Cichlidogyrus published recently by Rahmouni et al. (2017)). ABSTRACT Seven new species of Cichlidogyrus Paperna, 1960 (Monogenea: Dactylogyridae) isolated from the gills of six cichlid host species belonging to four tribes and sampled from the Congolese coastline of Lake Tanganyika (LT) are described: Cichlidogyrus adkoningsi sp. nov. from Cyphotilapia frontosa (tribe Cyphotilapiini); C. koblmuelleri sp. nov. from Cardiopharynx schoutedeni (Ectodini); C. habluetzeli sp. nov. from C. schoutedeni and C. frontosa; C. antoineparisellei sp. nov. from Interochromis loocki (Tropheini); C. masilyai sp. nov. from Petrochromis orthognathus (Tropheini); C. salzburgeri sp. nov. from P. trewavasae, and C. sergemorandi sp. nov. from Tylochromis polylepis (Tylochromini). This study represents the first parasitological examination of cyphotilapiine cichlid hosts. Representatives of the Tanganyikan ectodine, tropheine, and tylochromine cichlids previously sampled from various localities in the lake yielded nine, twelve, and two described species of Cichlidogyrus, respectively. The study further includes a morphological characterization of the male copulatory organ of six undescribed species of Cichlidogyrus found on the gills of the tropheines I. loocki and P. orthognathus, and on those of Callochromis melanostigma and Xenotilapia flavipinnis (both Ectodini). Geographical variation in the monogenean fauna of I. loocki was observed. The most closely related cichlid species investigated in this study harboured Cichlidogyrus spp. exhibiting some similarities in their sclerotized structures. Thus, our paper provides additional evidence of the high species richness of Cichlidogyrus and the link with their hosts’s phylogenetic affinities in LT. Submitted 2 March 2018 Accepted 16 August 2018 Published 23 October 2018 Corresponding author Chahrazed Rahmouni, rahmouni.chahrazed@gmail.com Academic editor Jean-Lou Justine Additional Information and Declarations can be found on page 31 DOI 10.7717/peerj.5604 Copyright 2018 Rahmouni et al. Distributed under Creative Commons CC-BY 4.0 Submitted 2 March 2018 Accepted 16 August 2018 Published 23 October 2018 Corresponding author Chahrazed Rahmouni, rahmouni.chahrazed@gmail.com Academic editor Jean-Lou Justine Additional Information and Declarations can be found on page 31 DOI 10.7717/peerj.5604 Copyright 2018 Rahmouni et al. Distributed under Creative Commons CC-BY 4 0 Submitted 2 March 2018 Accepted 16 August 2018 Published 23 October 2018 Distributed under Creative Commons CC-BY 4.0 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 INTRODUCTION Some Tanganyikan cichlid tribes remain to be investigated for their gill flatworms. No parasitological data are available on the Cyphotilapiini Salzburger et al., 2002 with its three endemic representatives Cyphotilapia frontosa (Boulenger, 1906), Cyphotilapia gibberosa (Takahashi & Nakaya, 2003), and Trematochromis benthicola (Matthes, 1962) (Muschick, Indermaur & Salzburger, 2012; Meyer, Matschiner & Salzburger, 2015; Takahashi & Sota, 2016). Among the 34 valid cichlid species that belong to the endemic Tanganyikan tribe Ectodini Poll, 1986, which includes 10 genera, only four species were studied for the presence of parasites and nine Cichlidogyrus spp. were described. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 2/37 2/37 Vanhove, Volckaert & Pariselle (2011) described four Cichlidogyrus spp. (C. centesimus, C. makasai, and C. vandekerkhovei Vanhove, Volckaert & Pariselle, 2011 on Congolese and Zambian Ophthalmotilapia ventralis (Boulenger, 1898) and Tanzanian O. boops (Boulenger, 1901) and O. nasuta (Poll & Matthes, 1962), and C. sturmbaueri Vanhove, Volckaert & Pariselle, 2011 on Zambian O. ventralis and Tanzanian O. nasuta). Later, Rahmouni et al. (2017) investigated Aulonocranus dewindti (Boulenger, 1899) and O. nasuta from the Burundese part of LT and described two Cichlidogyrus spp. on A. dewindti (C. discophonum and C. pseudoaspiralis Rahmouni, Vanhove & Šimková, 2017), and three species on O. nasuta (C. aspiralis, C. glacicremoratus and C. rectangulus Rahmouni, Vanhove & Šimková, 2017). Cardiopharynx Poll, 1942 is monotypic and represented by C. schoutedeni Poll, 1942 (Konings, 2015). Still in the Ectodini, Callochromis Regan, 1920 consists of three nominal species: Callochromis macrops (Boulenger, 1898), C. melanostigma (Boulenger, 1906), and C. pleurospilus (Boulenger, 1906) (Konings, 2015), whereas Xenotilapia Boulenger, 1898 includes 13–17 species (Kidd et al., 2012). Xenotilapia flavipinnis Poll, 1985 has a lake-wide distribution (Konings, 2015). In contrast to some other members of the Ectodini, there are no data available on the parasite fauna hosted by representatives of Callochromis and Xenotilapia. p The Tropheini is one of the most species-rich cichlid tribes endemic to LT with nine genera including approximately 24 species (Takahashi & Koblmüller, 2014). At least eight species of Petrochromis Boulenger, 1898, a representative of the Tropheini, have been described (Sturmbauer et al., 2003; Takahashi & Sota, 2016). Petrochromis orthognathus Matthes, 1959 is restricted to the northern two-thirds of the lake. Petrochromis trewavasae Poll, 1948 is found in the southern part of the lake, usually in sympatry with Petrochromis ephippium Brichard, 1989, a morphologically similar species considered conspecific to P. trewavasae (Konings, 2015). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 INTRODUCTION Interochromis Yamaoka, Hori & Kuwamura, 1998 is a monotypic genus erected because of the morphological and ecological similarities between I. loocki and species of Petrochromis, and the differences between I. loocki and species of the tropheine Simochromis Boulenger, 1898 (see overview in Pariselle et al., 2015b). Several studies have been carried out on the parasitic flatworms of these cichlids. Gillardin et al. (2012) described three Cichlidogyrus spp. (C. steenbergei and C. irenae Gillardin et al., 2012 from Zambian and Congolese Limnotilapia dardennii (Boulenger, 1899) and ‘Gnathochromis’ pfefferi (Boulenger, 1898), respectively, and C. gistelincki Gillardin et al., 2012 from Congolese, Tanzanian, and Zambian ‘Ctenochromis’ horei (Günther, 1894)). Then, Pariselle et al. (2015b) examined Zambian I. loocki and described three Cichlidogyrus spp. (C. buescheri, C. schreyenbrichardorum, and C. vealli Pariselle & Vanhove, 2015). In the same study, they compared the haptoral structures of representatives of Cichlidogyrus infecting I. loocki with those observed in some undescribed Cichlidogyrus spp. isolated from representatives of Petrochromis. The same team described six species of Cichlidogyrus infecting Congolese, Tanzanian and Zambian tropheine cichlids (C. banyankimbonai, C. muterezii, and C. raeymaekersi Pariselle & Vanhove, 2015 on Simochromis schreyenbrichardorum, and C. vealli Pariselle & Vanhove, 2015). In the same study, they compared the haptoral structures of representatives of Cichlidogyrus infecting I. loocki with those observed in some undescribed Cichlidogyrus spp. isolated from representatives of Petrochromis. The same team described six species of Cichlidogyrus infecting Congolese, Tanzanian and Zambian tropheine cichlids (C. banyankimbonai, C. muterezii, and C. raeymaekersi Pariselle & Vanhove, 2015 on Simochromis diagramma (Günther, 1894); C. georgesmertensi Pariselle & Vanhove, 2015 on schreyenbrichardorum, and C. vealli Pariselle & Vanhove, 2015). In the same study, they compared the haptoral structures of representatives of Cichlidogyrus infecting I. loocki with those observed in some undescribed Cichlidogyrus spp. isolated from representatives of Petrochromis. The same team described six species of Cichlidogyrus infecting Congolese, Tanzanian and Zambian tropheine cichlids (C. banyankimbonai, C. muterezii, and C. raeymaekersi Pariselle & Vanhove, 2015 on Simochromis diagramma (Günther, 1894); C. georgesmertensi Pariselle & Vanhove, 2015 on Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 3/37 Pseudosimochromis babaulti (Pellegrin, 1927); C. franswittei Pariselle & Vanhove, 2015 on P. marginatus (Poll, 1956) and P. curvifrons (Poll, 1942); and finally C. frankwillemsi Pariselle & Vanhove, 2015 on P. curvifrons (Van Steenberge et al., 2015)). Lake Tanganyika harbours representatives of a few non-endemic tribes resulting from colonisation from the lacustrine environment. INTRODUCTION This is the case of Tylochromini Poll, 1986 with Tylochromis polylepis (Boulenger, 1900) as its sole representative in the lake. Members of Tylochromis Regan, 1920 inhabit rivers, lakes, and coastal lagoons throughout central and western Africa. Muterezi Bukinga et al. (2012) studied the gill monogeneans of T. polylepis and described two species of Cichlidogyrus from Congolese host specimens, that is, C. mulimbwai and C. muzumanii Muterezi Bukinga et al., 2012. In addition, they presented drawings of the hard parts (haptor and reproductive organs) of an undescribed species referred to as Cichlidogyrus sp. ‘T. polylepis 3’. The aim of this paper is to study gill monogenean diversity in cichlids belonging to four tribes from the Congolese lakeshore of northern LT. We describe seven new species of Cichlidogyrus and provide the morphological characterization of six undescribed congeners. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 MATERIAL AND METHODS Parasite identification was conducted using original descriptions, the systematic revision of dactylogyridean parasites of African cichlids by Pariselle & Euzet (2009), and the recent overview focusing on the genitals of African Cichlidogyrus spp. by Rahmouni et al. (2017). Measurements and photographs were taken using an Olympus BX51 phase-contrast microscope and Olympus Stream Image Analysis v. 1.9.3 software. All measurements are included in the species descriptions. They are in micrometres, and are given as the mean followed by the range and the number of measurements (n) in parentheses (measurements of some undescribed species are given as the length of the structure in question followed by the number of measurements in parentheses). Drawings of the haptoral sclerotized parts and copulatory organs were made on flattened specimens using an Olympus BX51 microscope equipped with a drawing tube and edited with a graphic tablet compatible with Adobe Illustrator CS6 v. 16.0.0 and Adobe Photoshop v. 13.0. The terminology of haptoral sclerotized parts (anchors and hooks; also termed gripi and uncinuli, respectively) follows Gussev (1983). The numbering of hook pairs (Roman letters I–VII) is that recommended by Mizelle (1936). This method is preferred in adult specimens because it takes into consideration both antero-posterior and dorso–ventral positions of hooks (Kritsky, Thatcher & Boeger, 1986; Řehulková, Mendlová & Šimková, 2013). The lengths of hook pairs (short or long) was assigned following Pariselle & Euzet (2009). The classification of haptoral groups follows Vignon, Pariselle & Vanhove (2011). The metrics used for the hard structures are shown in Fig. 2. The type material was deposited in the Invertebrate collection of the MRAC, Tervuren, Belgium; the Finnish Museum of Natural History (MZH), Helsinki, Finland; and the Muséum National d’Histoire Naturelle (MNHN), Paris, France. Host nomenclature follows FishBase (Froese & Pauly, 2017). The list of museum specimens used for comparison with the new species is presented in Table 1. Sampling was carried out under mission statements 022/ MINEURS/CRH-U/2013 and 031/MINRST/CRH-U/2016 from the Centre de Recherche en Hydrobiologie-Uvira. In the absence of relevant animal welfare regulations in the D.R. Congo, the same strict codes of practice enforced within the European Union were applied. This study was approved by the Animal Care and Use Committee of the Faculty of Science, Masaryk University, Brno (Czech Republic), approval nuber CZ01308. MATERIAL AND METHODS Cichlid specimens were obtained in September 2013 and August 2016 from LT along the shoreline of the Democratic Republic of the Congo (DRC) (Fig. 1). In total, 26 fish specimens belonging to eight cichlid species of four tribes were purchased from fish markets or captured using gill nets during snorkelling or diving. Fish were placed in a cool box containing ice, transported to the laboratory, and dissected immediately. Cichlid hosts were identified in situ on the basis of morphological characters by Walter Salzburger (Zoological Institute, University of Basel, Switzerland), Donatien Muzumani Risasi (Centre de Recherche en Hydrobiologie, Uvira, DRC), and Maarten Van Steenberge (the Royal Museum for Central Africa (MRAC), Tervuren and the Royal Belgian Institute of Natural Sciences, Brussels, Belgium). We performed molecular analysis on the samples, and obtained sequence data from the partial cytochrome b (cyt-b) mitochondrial gene to confirm the identity of the investigated cichlids. Fin clips from the cichlid specimens were preserved in 96% ethanol. Cichlid DNA was extracted using the DNeasy Tissue kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. The partial cyt-b gene was amplified following Mendlová et al. (2012). The PCR products were loaded onto a 1% agarose gel and subsequently purified, and sequencing was performed following Rahmouni et al. (2017). Nucleotide sequences were edited using Sequencher software v. 5.0 (Gene Codes, Ann Arbor, MI, USA). The identification of cichlid species based on the sequence similarity approach was carried out using the Basic Local Alignment Search Tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi: blastn, default settings), available through the website of the National Centre for Biotechnology Information (NCBI Resource Coordinators, 2017). The newly generated sequences were deposited in GenBank under the accession numbers MH297985–MH298008. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 4/37 4/37 Figure 1 Map of sampling localities in LT. Full-size  DOI: 10.7717/peerj.5604/fig-1 Figure 1 Map of sampling localities in LT. Full-size  DOI: 10.7717/peerj.5604/fig-1 Figure 1 Map of sampling localities in LT. Figure 1 Map of sampling localities in LT. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Gill arches were separated via dorsal and ventral section using standard parasitological procedures and transferred into a Petri dish containing water. Monogeneans were detached from the gills and isolated according to Musilová, Řehulková & Gelnar (2009) using an MST130 stereoscopic microscope and mounted on slides with glycerine ammonium picrate mixture (GAP) (Malmberg, 1957). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 MATERIAL AND METHODS The electronic version of this article in portable document format will constitute a The electronic version of this article in portable document format will constitute a published work according to the International Commission on Zoological Nomenclature (ICZN), and hence the new names contained in the electronic version are effectively published under that Code from the electronic edition alone. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved, and the associated information viewed through any standard web browser by appending the Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 6/37 Figure 2 Measurements used in the descriptions of the new species of Cichlidogyrus. (A) anchor. (DA) dorsal anchor. (VA) ventral anchor. (a) total length. (b) blade length. (c) shaft length. (d) guard length. (e) point length. (DB) dorsal bar: (h) auricle length. (w) maximum straight width. (x) total length. (y) distance between auricles. (VB) ventral bar: (x) length of one ventral bar branch. (w) maximum width. (H) hook length. (MCO) male copulatory organ straight length. (Ct) copulatory tube curved length. (He) heel straight length. (Ap) accessory piece straight length. (Vg) vagina: (V) vagina total length. (v) vagina width. Full-size  DOI: 10.7717/peerj.5604/fig-2 Figure 2 Measurements used in the descriptions of the new species of Cichlidogyrus. (A) anchor. (DA) dorsal anchor. (VA) ventral anchor. (a) total length. (b) blade length. (c) shaft length. (d) guard length. (e) point length. (DB) dorsal bar: (h) auricle length. (w) maximum straight width. (x) total length. (y) distance between auricles. (VB) ventral bar: (x) length of one ventral bar branch. (w) maximum width. (H) hook length. (MCO) male copulatory organ straight length. (Ct) copulatory tube curved length. (He) heel straight length. (Ap) accessory piece straight length. (Vg) vagina: (V) vagina total length. (v) vagina width. Full-size  DOI: 10.7717/peerj.5604/fig-2 Table 1 Comparative museum material examined in the present study. Cichlidogyrus spp. Host species Locality Accession number Cichlidogyrus mulimbwai Muterezi Bukinga et al., 2012 T. polylepis (Boulenger, 1900) Mulembwe & Moba (DRC) MRAC 37701 Cichlidogyrus muzumanii Muterezi Bukinga et al., 2012 T. polylepis (Boulenger, 1900) Mulembwe & Moba (DRC) MRAC 37699 Cichlidogyrus buescheri Pariselle & Vanhove, 2015 I. loocki (Poll, 1949) Kalambo Lodge (Zambia) MRAC 37744 Cichlidogyrus schreyenbrichardorum Pariselle & Vanhove, 2015 I. SPECIES DESCRIPTIONS SPECIES DESCRIPTIONS Dactylogyridae Bychowski, 1933 Cichlidogyrus Paperna, 1960 Cichlidogyrus adkoningsi sp. nov. Fig. 3 urn:lsid:zoobank.org:act:526C2A74-E3D6-4357-B1E8-03B08B95CE38. Dactylogyridae Bychowski, 1933 Cichlidogyrus Paperna, 1960 Cichlidogyrus adkoningsi sp. nov. Fig. 3 urn:lsid:zoobank.org:act:526C2A74-E3D6-4357-B1E8-03B08B95CE38. Description. Based on three specimens fixed in GAP. Body 516 (482–550; n = 3) long, 93 (80–107; n = 3) wide at mid-body. Dorsal anchors with short shaft and more pronounced guard and curved blade with arched point: a = 31 (30–32; n = 3); b = 25 (24–26; n = 3); c = 1–2 (n = 3); d = 9 (7–10; n = 3); e = 10 (9–12; n = 3). Dorsal bar curved with constant width and relatively long auricles: h = 38 (35–44; n = 3); w = 3–4 (n = 3); x = 28 (26–30; n = 3); y = 2 (1–4; n = 3). Ventral anchors with shorter shaft than guard, curved blade with arched point: a = 28 (26–29; n = 3); b = 24 (23–25; n = 3); c = 1–2 (n = 3), d = 7 (6–8; n = 3); e = 11 (10–13; n = 3). V-shaped ventral bar: w = 4 (3–5; n = 3); x = 32 (29–36; n = 3). Haptor with seven short hook pairs, hooks V retain their larval size (sensu Pariselle & Euzet, 2003, 2009), each hook with erect thumb and shank comprised of two subunits: pair I = 11 (10–12; n = 3) long, pair II = 14–15 (n = 3) long, pair III = 15–16 (n = 3) long, pair IV = 16–17 (n = 3) long, pair V = 11–12 (n = 3) long, pair VI = 17–18 (n = 3) long, and pair VII = 17 (16–18, n = 3) long. Male copulatory organ with relatively short copulatory duct, slightly curved halfway and tapered distally: MCO = 45 (42–47; n = 3); Ct = 31 (29–33; n = 3). Heel irregularly shaped, He = 10 (9–11; n = 3). Accessory piece linked to the basal bulb, C-shaped, thick in the middle part, ending in hook, Ap = 26 (24–28; n = 13). Vagina non-sclerotized. Diagnosis. Cichlidogyrus adkoningsi sp. nov. belongs to the group of species which exhibit short hook pairs I–IV, VI, and VII (sensu Vignon, Pariselle & Vanhove, 2011), a copulatory duct without a swollen proximal portion, and a non-sclerotized vagina (see Pariselle & Euzet, 2003), just like C. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Molecular identification of cichlid hosts The mitochondrial cytochrome b (cyt-b) gene fragment of 25 cichlid specimens was successfully amplified. The length of each consensus sequence was 419 bp. The Blast search processed on the NCBI website confirmed the species identification of the cichlid species investigated for the presence of gill parasitic flatworms. MATERIAL AND METHODS loocki (Poll, 1949) Kalambo Lodge (Zambia) MRAC, MRAC 37741 Cichlidogyrus vealli Pariselle & Vanhove, 2015 I. loocki (Poll, 1949) Kalambo Lodge (Zambia) MRAC, MRAC 37743 Cichlidogyrus discophonum Rahmouni, Vanhove & Šimková, 2017 A. dewindti (Boulenger, 1899) Nyaruhongoka (Burundi) MRAC, MRAC 37956 Cichlidogyrus pseudoaspiralis Rahmouni, Vanhove & Šimková, 2017 A. dewindti (Boulenger, 1899) Nyaruhongoka (Burundi) MRAC 37955 Cichlidogyrus aspiralis Rahmouni, Vanhove & Šimková, 2017 O. nasuta (Poll & Matthes, 1962) Magara (Burundi) MRAC 37954 Table 1 Comparative museum material examined in the present study. Table 1 Comparative museum material examined in the present study. Cichlidogyrus mulimbwai Muterezi Bukinga et al., 2012 Cichlidogyrus buescheri Pariselle & Vanhove, 2015 Cichlidogyrus schreyenbrichardorum Pariselle & Vanhove, 2015 Cichlidogyrus vealli Pariselle & Vanhove, 2015 Cichlidogyrus discophonum Rahmouni, Vanhove & Šimková, 2017 Cichlidogyrus discophonum Rahmouni, Vanhove & Šimková, 2017 Cichlidogyrus pseudoaspiralis Rahmouni, Vanhove & Šimková, 2017 Cichlidogyrus pseudoaspiralis Rahmouni, Vanhove & Šimková, 2017 Cichlidogyrus aspiralis Rahmouni, Vanhove & Šimková, 2017 Cichlidogyrus aspiralis Rahmouni, Vanhove & Šimková, 2017 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 LSID to the prefix http://zoobank.org/. The LSID for this publication is: urn:lsid:zoobank. org:pub:7076794A-B9EB-4FFC-AC49-66C304EC5BFB. The online version of this work is archived and available from the following digital repositories: PeerJ, PubMed Central, and CLOCKSS. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 SPECIES DESCRIPTIONS attenboroughi Kmentová, Gelnar, Koblmüller et al., 2016; C. banyankimbonai; C. berminensis Pariselle, Bitja Nyom & Bilong Bilong, 2013; C. bifurcatus Paperna, 1960; C. brunnensis Kmentová, Gelnar, Koblmüller et al., 2016; C. buescheri; C. consobrini Jorissen, Pariselle & Vanhove, 2018; Koblmüller et al., 2016; C. banyankimbonai; C. berminensis Pariselle, Bitja Nyom & Bilong Bilong, 2013; C. bifurcatus Paperna, 1960; C. brunnensis Kmentová, Gelnar, Koblmüller et al., 2016; C. buescheri; C. consobrini Jorissen, Pariselle & Vanhove, 2018; Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Figure 3 Sclerotized structures of Cichlidogyrus adkoningsi sp. nov. ex Cyphotilapia frontosa. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copu- latory organ; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-3 Figure 3 Sclerotized structures of Cichlidogyrus adkoningsi sp. nov. ex Cyphotilapia frontosa. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copu- latory organ; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-3 C. discophonum; C. evikae Rahmouni, Vanhove & Šimková, 2017; C. fontanai Pariselle & Euzet, 1997; C. frankwillemsi; C. franswittei; C. georgesmertensi; C. gillardinae; C. gistelincki; C. glacicremoratus; C. haplochromii Paperna & Thurston, 1969; C. irenae; C. jeanloujustinei Rahmouni, Vanhove & Šimková, 2017; C. longipenis Paperna & Thurston, 1969; C. makasai; C. milangelnari Rahmouni, Vanhove & Šimková, 2017; C. mulimbwai; C. muterezii; C. nageus Řehulková, Mendlová & Šimková, 2013; C. discophonum; C. evikae Rahmouni, Vanhove & Šimková, 2017; C. fontanai Pariselle & Euzet, 1997; C. frankwillemsi; C. franswittei; C. georgesmertensi; C. gillardinae; C. gistelincki; C. glacicremoratus; C. haplochromii Paperna & Thurston, 1969; C. irenae; C. jeanloujustinei Rahmouni, Vanhove & Šimková, 2017; C. longipenis Paperna & Thurston, 1969; C. makasai; C. milangelnari Rahmouni, Vanhove & Šimková, 2017; C. mulimbwai; C. muterezii; C. nageus Řehulková, Mendlová & Šimková, 2013; C. raeymaekersi; C. rognoni Pariselle, Bilong Bilong & Euzet, 2003; C. schreyenbrichardorum; C. sanjeani Pariselle & Euzet, 1997; C. sigmocirrus Pariselle, Bitja Nyom & Bilong Bilong, 2014; C. steenbergei; C. tilapiae Paperna, 1960; C. vandekerkhovei; and C. vealli. Dorsal and ventral bars as well as the accessory piece in C. adkoningsi sp. nov. and C. sturmbaueri described from O. ventralis and O. nasuta are of similar size. Additionally, C. discophonum; C. evikae Rahmouni, Vanhove & Šimková, 2017; C. fontanai Pariselle & Euzet, 1997; C. frankwillemsi; C. franswittei; C. georgesmertensi; C. gillardinae; C. gistelincki; C. glacicremoratus; C. SPECIES DESCRIPTIONS haplochromii Paperna & Thurston, 1969; C. irenae; C. jeanloujustinei Rahmouni, Vanhove & Šimková, 2017; C. longipenis Paperna & Thurston, 1969; C. makasai; C. milangelnari Rahmouni, Vanhove & Šimková, 2017; C. mulimbwai; C. muterezii; C. nageus Řehulková, Mendlová & Šimková, 2013; C. raeymaekersi; C. rognoni Pariselle, Bilong Bilong & Euzet, 2003; C. schreyenbrichardorum; C. sanjeani Pariselle & Euzet, 1997; C. sigmocirrus Pariselle, Bitja Nyom & Bilong Bilong, 2014; C. steenbergei; C. tilapiae Paperna, 1960; C. vandekerkhovei; and C. vealli. Dorsal and ventral bars as well as the accessory piece in C. adkoningsi sp. nov. and C. sturmbaueri described from O. ventralis and O. nasuta are of similar size. Additionally, the copulatory duct in both species is similarly shaped (see Vanhove, Volckaert & Pariselle, 2011). However, the new species is distinguishable from C. sturmbaueri by (i) the longer dorsal anchors (30–32 mm in C. adkoningsi sp. nov. vs 19–21 mm in C. sturmbaueri), (ii) the longer dorsal bar auricles (35–44 mm in C. adkoningsi sp. nov. vs 12–15 mm in C. sturmbaueri), (iii) the different hook pairs (C. adkoningsi sp. nov. exhibits short hook pairs I–IV, VI, and VII while C. sturmbaueri displays short hook pair I and long pairs II–IV, VI, and VII (see Pariselle & Euzet, 2009; Rahmouni, Vanhove & Šimková, 2017 for the importance of hook length in the systematics of Cichlidogyrus), and (iv) the accessory piece (C-shaped, thick in the middle part ending in hook in C. adkoningsi sp. nov. vs H-shaped in C. sturmbaueri). The long auricles of C. adkoningsi sp. nov. are reminiscent of C. vandekerkhovei from O. ventralis, O. boops and O. nasuta, C. glacicremoratus from O. nasuta, and C. discophonum from A. dewindti, which are Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 all parasites of ectodine cichlids (Vanhove, Volckaert & Pariselle, 2011; Rahmouni, Vanhove & Šimková, 2017). Further, C. adkoningsi sp. nov. shares with C. makasai similarly sized dorsal and ventral bars, hook pairs, and accessory pieces. However, in C. adkoningsi sp. nov. (i), the dorsal anchors are longer (30–32 mm in C. adkoningsi sp. nov. vs 19–23 mm in C. makasai), (ii) the dorsal bar auricles are much longer (35–44 in C. adkoningsi sp. nov. vs 17–23 mm in C. makasai), (iii) the heel is longer (9–11 in C. adkoningsi sp. nov. vs 2–4 mm in C. Paratype: MRAC M.T.38433. Etymology: the specific epithet of the new species, ‘adkoningsi’, honours the Dutch biologist Dr. Adrianus Johannes Franciscus Marinus Maria Konings, known as Ad Konings, who has published extensively on cichlids. His books on the cichlids of LT have been crucial to our research on the parasite fauna of these fishes. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Site of infection: Gills. Site of infection: Gills. Site of infection: Gills. Type-locality: Makabola village (332′S, 299′E; purchase from fisherman), DRC, LT. Prevalence & intensity of infection: one fish specimen infected/two fish specimens examined, three parasite specimens on the infected fish. Holotype: MRAC M.T.38432. Cichlidogyrus koblmuelleri sp. nov. Fig. 4 urn:lsid:zoobank.org:act:473DB764-6798-43EE-8DD8-3B10D1AC1BBF. SPECIES DESCRIPTIONS makasai), (iv) the copulatory duct is much shorter and differently shaped (relatively short copulatory duct, slightly curved halfway, with a length of 33–29 mm in C. adkoningsi sp. nov. vs thin curved duct which tapers distally, 69–79 mm long in C. makasai), and (v) the accessory piece is differently shaped (thick, C-shaped, ending in a hook in C. adkoningsi sp. nov. vs simple, slightly bent at distal third, resembling a spanner in C. makasai). Type-host: Cyphotilapia frontosa (Boulenger, 1906) (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Cyphotilapiini Salzburger et al., 2002). Host accession numbers: MH297995–96 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 rn:lsid:zoobank.org:act:473DB764-6798-43EE-8DD8-3B10D1AC1BBF. Description. Based on seven specimens fixed in GAP. Body 462 (402–564; n = 4) long, 85 (68–121; n = 4) wide at mid-body. Dorsal anchors long with short shaft and more pronounced guard, curved blade and arched point: a = 27 (26–28; n = 4); b = 22 (21–23; n = 4); c = 1–2 (n = 4); d = 9 (8–10; n = 4); e = 9–10 (n = 4). Dorsal bar slightly curved with blunt endings and long auricles: h = 26 (25–28; n = 4); w = 4–5 (n = 4); x = 26 (25–27; n = 4); y = 4 (2–5; n = 4). Ventral anchors similar to dorsal ones: a = 24–25 (n = 4); b = 21–22 (n = 4); c = 1–2 (n = 4), d = 6 (5–7; n = 4); e = 9–10 (n = 4). V-shaped ventral bar: w = 2–3 (n = 4); x = 29 (28–30; n = 10). Haptor with seven short hook pairs, hooks V retain their larval size (see above), each hook with erect thumb and shank comprised of two subunits: pair I = 11 (10–12; n = 4) long, pair II = 13–14 (n = 4) Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 10/37 l i III 14 15 ( 4) l i IV 15 16 ( 4) l i V 10 11 Figure 4 Sclerotized structures of Cichlidogyrus koblmuelleri sp. nov. ex Cardiopharynx schoutedeni. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copu- latory organ; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-4 Figure 4 Sclerotized structures of Cichlidogyrus koblmuelleri sp. nov. ex Cardiopharynx schoutedeni. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copu- latory organ; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-4 long, pair III = 14–15 (n = 4) long, pair IV = 15–16 (n = 4) long, pair V = 10–11 (n = 4) long, pair VI = 16 (15–17; n = 4) long, and pair VII = 13–14 (n = 4) long. Male copulatory organ with a C-shaped copulatory duct: MCO = 46 (36–49; n = 7); Ct = 63 (59–65; n = 7). No heel. rn:lsid:zoobank.org:act:473DB764-6798-43EE-8DD8-3B10D1AC1BBF. Accessory piece curved with two superimposed parts, one of them thicker and longer than the other, ending in moderately curved hook, Ap = 38 (37–40; n = 7). Vagina non-sclerotized. long, pair III = 14–15 (n = 4) long, pair IV = 15–16 (n = 4) long, pair V = 10–11 (n = 4) long, pair VI = 16 (15–17; n = 4) long, and pair VII = 13–14 (n = 4) long. Male copulatory organ with a C-shaped copulatory duct: MCO = 46 (36–49; n = 7); Ct = 63 (59–65; n = 7). No heel. Accessory piece curved with two superimposed parts, one of them thicker and longer than the other, ending in moderately curved hook, Ap = 38 (37–40; n = 7). Vagina non-sclerotized. Diagnosis. Cichlidogyrus koblmuelleri sp. nov. belongs to the same group as C. adkoningsi sp. nov. as it shares the small size of all hook pairs. The new species most closely resembles C. discophonum described from A. dewindti (see Rahmouni, Vanhove & Šimková, 2017) regarding the morphology of the dorsal and ventral anchors and the absence of a heel. However, it differs from the latter by (i) the longer copulatory duct (59–65 mm in C. koblmuelleri sp. nov. vs 41–47 mm in C. discophonum), and (ii) the longer and differently shaped accessory piece (curved with two superimposed parts, one of them thicker and longer than the other ending in a hook, 37–40 mm in C. koblmuelleri sp. nov. vs short with two thick distinct parts, twisted distally, ending in hook, 15–22 mm in C. discophonum). Like C. adkoningsi sp. nov. and the other Cichlidogyrus spp. of the ectodine cichlids listed in the previous diagnosis, C. koblmuelleri sp. nov. exhibits long dorsal bar auricles. However, they are longer in the new species than those of C. glacicremoratus (25–28 mm in C. koblmuelleri sp. nov. vs 14–18 mm in C. glacicremoratus). Moreover, C. koblmuelleri sp. nov. is mainly distinguishable from C. glacicremoratus by (i) the longer and differently shaped copulatory duct (C-shaped copulatory duct, 59–65 mm in C. koblmuelleri sp. nov. vs wavy copulatory duct, with thick Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Figure 5 Sclerotized structures of Cichlidogyrus habluetzeli sp. nov. ex Cardiopharynx schoutedeni and Cyphotilapia frontosa. rn:lsid:zoobank.org:act:473DB764-6798-43EE-8DD8-3B10D1AC1BBF. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece; Vg, vagina. Full-size  DOI: 10.7717/peerj.5604/fig-5 Figure 5 Sclerotized structures of Cichlidogyrus habluetzeli sp. nov. ex Cardiopharynx schoutedeni and Cyphotilapia frontosa. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece; Vg, vagina. Full-size  DOI: 10.7717/peerj.5604/fig-5 wall, constricted and curved approximately at proximal third, with wide terminal opening, 42–47 mm in C. glacicremoratus), and (ii) the heel (no heel in C. koblmuelleri sp. nov. vs small, irregular sclerotized flange-like structure in C. glacicremoratus). wall, constricted and curved approximately at proximal third, with wide terminal opening, 42–47 mm in C. glacicremoratus), and (ii) the heel (no heel in C. koblmuelleri sp. nov. vs small, irregular sclerotized flange-like structure in C. glacicremoratus). Type-host: Cardiopharynx schoutedeni Poll, 1942 (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Ectodini Poll, 1986). Type-host: Cardiopharynx schoutedeni Poll, 1942 (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Ectodini Poll, 1986). Host accession numbers: MH297989–94. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Host accession numbers: MH297989–94. Type-locality: Mulongwe fish market (322′S, 296′E), Uvira, DRC, LT. Prevalence & intensity of infection: four fish specimens infected/six fish specimens examined, one to three parasite specimens per infected host. Holotype: MRAC M.T.38434. Holotype: MRAC M.T.38434. Paratypes: MRAC M.T.38435, MRAC M.T.38439. Paratypes: MRAC M.T.38435, MRAC M.T.38439. Paratypes: MRAC M.T.38435, MRAC M.T.38439. Etymology: the specific epithet of the new species, ‘koblmuelleri’ honours the biologist Dr. Stephan Koblmüller (Austria), an all-round specialist in LT’s ichthyofauna, in recognition of his crucial contribution to parasitological work. Cichlidogyrus habluetzeli sp. nov. Fig. 5 urn:lsid:zoobank.org:act:EC79CE69-2D88-4D96-A255-AC0E31B76EAE. Cichlidogyrus habluetzeli sp. nov. Fig. 5 urn:lsid:zoobank.org:act:EC79CE69-2D88-4D96-A255-AC0E31B76EAE. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Description. Based on 28 specimens fixed in GAP. Body 507 (286–790; n = 19) long, 99 (62–152; n = 19) wide at mid-body. Dorsal anchors long with short shaft and more pronounced guard, curved blade with slightly arched point: a = 60 (53–66; n = 20); b = 43 (38–46; n = 20); c = 6 (3–9; n = 21); d = 22 (16–26; n = 20); e = 11 (9–14; n = 20). Dorsal bar relatively curved, thick in the middle part with blunt endings and straight auricles: h = 14 (10–18; n = 19); w = 11 (9–14; n = 18); x = 59 (54–63; n = 18); y = 25 (22–29; n = 19). Ventral anchors shorter than dorsal ones with shorter shaft than guard, curved blade with arched point: a = 47 (44–52; n = 12); b = 44 (40–48; n = 20); c = 6 (5–8; n = 18), d = 12 (8–15; n = 18); e = 11 (10–13; n = 21). Long and thick V-shaped ventral bar: w = 7 (6–10; n = 20); x = 53 (47–60; n = 20). Haptor with seven long hook pairs, pair I large in comparison with remaining pairs, hooks V retain their larval size (sensu Pariselle & Euzet, 2003, 2009), and each hook with erect thumb and shank comprised of two subunits: pair I = 38 (33–42; n = 20) long, pair II = 27 (23–31; n = 18) long, pair III = 27 (22–31; n = 19) long, pair IV = 24 (21–28; n = 19) long, pair V = 12 (11–13; n = 19) long, pair VI = 28 (24–31; n = 20) long, and pair VII = 28 (25–32; n = 20) long. Male copulatory organ with a relatively short straight copulatory duct: MCO = 44 (39–53; n = 23); Ct = 21 (19–23; n = 24). Well-developed straight heel, He = 23 (18–29; n = 24). Thin accessory piece distally with leaf-shaped ending, Ap = 15 (13–21; n = 21). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Paratypes: MRAC M.T.38435, MRAC M.T.38439. Sclerotized vagina: V = 23 (19–32; n = 6); v = 11 (10–14; n = 7). Description. Based on 28 specimens fixed in GAP. Body 507 (286–790; n = 19) long, 99 (62–152; n = 19) wide at mid-body. Dorsal anchors long with short shaft and more pronounced guard, curved blade with slightly arched point: a = 60 (53–66; n = 20); b = 43 (38–46; n = 20); c = 6 (3–9; n = 21); d = 22 (16–26; n = 20); e = 11 (9–14; n = 20). Dorsal bar relatively curved, thick in the middle part with blunt endings and straight auricles: h = 14 (10–18; n = 19); w = 11 (9–14; n = 18); x = 59 (54–63; n = 18); y = 25 (22–29; n = 19). Ventral anchors shorter than dorsal ones with shorter shaft than guard, curved blade with arched point: a = 47 (44–52; n = 12); b = 44 (40–48; n = 20); c = 6 (5–8; n = 18), d = 12 (8–15; n = 18); e = 11 (10–13; n = 21). Long and thick V-shaped ventral bar: w = 7 (6–10; n = 20); x = 53 (47–60; n = 20). Haptor with seven long hook pairs, pair I large in comparison with remaining pairs, hooks V retain their larval size (sensu Pariselle & Euzet, 2003, 2009), and each hook with erect thumb and shank comprised of two subunits: pair I = 38 (33–42; n = 20) long, pair II = 27 (23–31; n = 18) long, pair III = 27 (22–31; n = 19) long, pair IV = 24 (21–28; n = 19) long, pair V = 12 (11–13; n = 19) long, pair VI = 28 (24–31; n = 20) long, and pair VII = 28 (25–32; n = 20) long. Male copulatory organ with a relatively short straight copulatory duct: MCO = 44 (39–53; n = 23); Ct = 21 (19–23; n = 24). Well-developed straight heel, He = 23 (18–29; n = 24). Thin accessory piece distally with leaf-shaped ending, Ap = 15 (13–21; n = 21). Sclerotized vagina: V = 23 (19–32; n = 6); v = 11 (10–14; n = 7). Diagnosis. Cichlidogyrus habluetzeli sp. nov. Paratypes: MRAC M.T.38435, MRAC M.T.38439. belongs to the group of species with long hook pairs I–IV, VI and VII, large hook pair I, a copulatory duct without a swollen proximal portion, and a sclerotized vagina (see above). This group is restricted to a single species, the Tanganyikan C. centesimus (see Vanhove, Volckaert & Pariselle, 2011). Diagnosis. Cichlidogyrus habluetzeli sp. nov. belongs to the group of species with long hook pairs I–IV, VI and VII, large hook pair I, a copulatory duct without a swollen proximal portion, and a sclerotized vagina (see above). This group is restricted to a single species, the Tanganyikan C. centesimus (see Vanhove, Volckaert & Pariselle, 2011). species, the Tanganyikan C. centesimus (see Vanhove, Volckaert & Pariselle, 2011). Therefore, C. habluetzeli sp. nov. is only the second record with this configuration of sclerotized structures throughout the entire genus. The dorsal bar auricles in C. habluetzeli sp. nov. are small hollow outgrowths on the anterior face, a feature observed in congeners infecting representatives of Tylochromini and Ectodini. With this morphology, we find Tanganyikan C. mulimbwai and C. muzumanii, both from T. polylepis (Muterezi Bukinga et al., 2012), C. pseudoaspiralis from A. dewindti, and C. aspiralis from O. nasuta (Rahmouni, Vanhove & Šimková, 2017) to be just like non-Tanganyikan C. chrysopiformis, C. djietoi, and C. sigmocirrus Pariselle, Bitja Nyom & Bilong Bilong, 2014, all parasites of T. sudanensis Daget, 1954 (Pariselle, Bitja Nyom & Bilong Bilong, 2014), C. kothiasi Pariselle & Euzet, 1994 from T. jentinki (Steindachner, 1862) (Pariselle & Euzet, 1994), and C. dageti, C. euzeti, and C. falcifer Dossou & Birgi, 1984 from Hemichromis fasciatus (Dossou & Birgi, 1984; Pariselle & Euzet, 2009). In addition to its haptoral features, C. habluetzeli sp. nov. exhibits a similar morphotype of the reproductive organs as C. aspiralis, C. pseudoaspiralis (Rahmouni, Vanhove & Šimková, 2017), C. casuarinus Pariselle, Muterezi Bukinga & Vanhove, 2015 isolated from a range of cichlid representatives of the tribe Bathybatini Poll, 1986 (Pariselle et al., 2015a), C. centesimus (Vanhove, Volckaert & Pariselle, 2011), and C. nshomboi Muterezi Bukinga et al., 2012 from Boulengerochromis microlepis (Boulenger, 1899) Boulengerochromini Takahashi, 2003 (all Tanganyikan species) (Muterezi Bukinga et al., 2012). C. habluetzeli sp. nov. can Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 be compared to C. aspiralis regarding the similarly shaped and sized copulatory duct and accessory piece, and the presence of a sclerotized vagina. However, the new species differs from C. Paratypes: MRAC M.T.38435, MRAC M.T.38439. aspiralis by (i) the longer dorsal and ventral anchors (respectively, 60–66; 46–52 mm in C. habluetzeli sp. nov. vs 39–43; 33–34 mm in C. aspiralis), (ii) the longer dorsal and ventral bars (respectively, 54–63; 52–60 mm in C. habluetzeli sp. nov. vs 45–47, 37–40 mm in C. aspiralis), (iii) and the longer dorsal bar auricles (14–18 mm in C. habluetzeli sp. nov. vs 6–10 mm in C. aspiralis). On the other hand, the dorsal and ventral anchors in C. habluetzeli sp. nov. are similar to those of C. casuarinus and C. nshomboi, while in C. centesimus and C. pseudoaspiralis, they are shorter (respectively, 53–66; 44–52 mm in C. habluetzeli sp. nov. vs 41–55; 34–44 mm in C. centesimus and 37–42; 37–40 mm in C. pseudoaspiralis). Also, the dorsal bar in the new species is similar to that of C. nshomboi, shorter than that of C. casuarinus (54–63 mm in C. habluetzeli sp. nov. vs 64–85 mm in C. casuarinus), and longer than those of C. centesimus and C. pseudoaspiralis (54–63 mm in C. habluetzeli sp. nov. vs 37–52; 31–32 mm in C. centesimus, and C. pseudoaspiralis, respectively). Like C. aspiralis and C. pseudoaspiralis, C. habluetzeli sp. nov. lacks a spirally coiled thickening in the distal part of its copulatory duct, a feature present in C. casuarinus, C. centesimus and C. nshomboi. Further, the copulatory duct in C. habluetzeli sp. nov. is similar in size to that in C. centesimus and C. nshomboi, while it is longer in C. casuarinus and C. pseudoaspiralis (19–23 mm in C. habluetzeli sp. nov. vs 33–44; 40–43 mm in C. casuarinus and C. pseudoaspiralis, respectively). Unlike C. pseudoaspiralis, C. habluetzeli sp. nov. exhibits a sclerotized vagina similarly to C. casuarinus, C. centesimus, and C. nshomboi (see Vanhove, Volckaert & Pariselle, 2011; Muterezi Bukinga et al., 2012; Pariselle et al., 2015a; Rahmouni, Vanhove & Šimková, 2017). Type-host: Cardiopharynx schoutedeni Poll, 1942 (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Ectodini Poll, 1986). Site of infection: Gills. Type-locality: Mulongwe fish market (322′S, 296′E), Uvira, DRC, LT. Type-locality: Mulongwe fish market (322′S, 296′E), Uvira, DRC, LT. Prevalence & intensity of infection in the type host: three fish specimens infected/six fish specimens examined, one to six parasite specimens per infected host. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. Description. Based on 12 specimens fixed in GAP. Body 436 (330–509; n = 5) long, 82 (76–86; n = 5) wide at mid-body. Dorsal anchors with poorly marked shaft and more pronounced guard (approximately four times the length of shaft), bent blade with arched point: a = 25 (24–26; n = 7); b = 22–23 (n = 7); c = 1–2 (n = 7); d = 6 (4–7; n = 7); e = 9 (8–10; n = 7). Dorsal bar curved, thick in the middle part with blunt endings and well-developed auricles: h = 11–12 (n = 7); w = 5–6 (n = 7); x = 27 (26–29; n = 7); y = 9 (8–11; n = 7). Ventral anchors similar in shape and size to dorsal ones: a = 26 (25–27; n = 7); b = 24 (23–25; n = 7); c = 2 (1–3; n = 7); d = 6 (3–7; n = 7); e = 10 (9–12; n = 7). V-shaped ventral bar: w = 4–5 (n = 7); x = 26 (24–27; n = 7). Haptor with seven short hook pairs, hooks V retain their larval size (see above), each hook with erect thumb and shank comprised of two subunits: pair I = 11–12 (n = 6) long, pair II = 13 (12–14; n = 7) long, pair III = 15 (13–16; n = 7) long, pair IV = 17 (15–18; n = 7) long, pair V = 11 (9–12; n = 6) long, pair VI = 16 (14–17; n = 7) long, and pair VII = 13–14 (n = 7) long. Male copulatory organ composed of long, curved copulatory duct that tapers distally: MCO = 42 (39–46; n = 10); Ct = 59 (54–62; n = 11). Heel relatively short, He = 6 (5–7; n = 11). Accessory piece is linked to basal bulb by thin filament, C-shaped in the middle part, composed of two distinct parts, one extremity ending in hook while the other connected to the latter with swollen base, Ap = 29 (22–33; n = 10). Long tubular vagina with two characteristic extremities, a broadened part and one extremity covered by a triangular-like structure: V = 40 (35–47; n = 11); v = 9 (7–13; n = 11). Description. Based on 12 specimens fixed in GAP. Body 436 (330–509; n = 5) long, 82 (76–86; n = 5) wide at mid-body. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Prevalence & intensity of infection in the type host: three fish specimens infected/six fish specimens examined, one to six parasite specimens per infected host. Prevalence & intensity of infection in the type host: three fish specimens infected/six fish specimens examined, one to six parasite specimens per infected host. Additional host: found on the gills of C. frontosa, one fish specimen infected/two fish specimens examined, nine parasite specimens on the infected host. Holotype: MRAC M.T.38437. Paratypes: MRAC M.T.38436, MRAC M.T.38438; MNHN HEL748–49, HEL752; MZH KN10058–59. Paratypes: MRAC M.T.38436, MRAC M.T.38438; MNHN HEL748–49, HEL752; MZH KN10058–59. Etymology: the specific epithet of the new species, ‘habluetzeli’ honours the biologist Dr. Pascal István Hablützel (Switzerland/Belgium) in honour of his pioneering work in the eco-immunology of Tanganyikan cichlids. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Figure 6 Sclerotized structures of Cichlidogyrus antoineparisellei sp. nov. ex Interochromis loocki. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece; Vg, vagina. Full-size  DOI: 10.7717/peerj.5604/fig-6 Figure 6 Sclerotized structures of Cichlidogyrus antoineparisellei sp. nov. ex Interochromis loocki. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece; Vg, vagina. Full-size  DOI: 10.7717/peerj.5604/fig-6 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. (2018), PeerJ, DOI 10.7717/peerj.5604 airs I–IV, VI, and VII, and a copulatory duct without a swollen proximal portion, pairs I–IV, VI, and VII, and a copulatory duct without a swollen proximal portion, and exhibits a sclerotized vagina (see above). This group includes C. acerbus Dossou, 1982; C. amieti Birgi & Euzet, 1983; C. amphoratus Pariselle & Euzet, 1996; C. berrebii Pariselle & Euzet, 1994; C. cirratus Paperna, 1964; C. cubitus Dossou, 1982; C. djietoi; C. giostrai Pariselle, Bilong Bilong & Euzet, 2003; C. karibae Douëllou, 1993; C. kothiasi Pariselle & Euzet, 1994; C. lagoonaris Paperna, 1969; C. levequei Pariselle & Euzet, 1996; C. louipaysani Pariselle & Euzet, 1995; C. mvogoi Pariselle, Bitja Nyom & Bilong Bilong, 2014; C. njinei Pariselle, Bilong Bilong & Euzet, 2003; C. ornatus Pariselle & Euzet, 1996; C. pouyaudi Pariselle & Euzet, 1994; C. sclerosus Paperna & Thurston, 1969; C. slembroucki Pariselle & Euzet, 1998; and C. zambezensis Douëllou, 1993. C. mbirizei Muterezi Bukinga et al., 2012, a parasite described from a representative of the tribe Oreochromini Dunz & Schliewen, 2013, Oreochromis tanganicae (Günther, 1894), is the only species hitherto known to have short hook pairs I–IV, VI, and VII and a sclerotized vagina in LT (see Muterezi Bukinga et al., 2012), and therefore C. antoineparisellei sp. nov. is the second known representative with the combination of these sclerotized structures in the lake. However, C. antoineparisellei sp. nov. is the first representative of Cichlidogyrus recognized on Tanganyikan tropheines which has a sclerotized vagina. Further, because of the systematic and phylogenetic affinities among the Tropheini and the Haplochromini Poll, 1986 (see Salzburger et al., 2005), C. antoineparisellei sp. nov. is considered to be the second representative species exhibiting this feature within the parasites infecting representatives of the Haplochromini lineage, like its congener C. zambezensis reported from the non-Tanganyikan haplochromines Serranochromis macrocephalus (Boulenger, 1899), S. robustus jallae (Günther, 1864) (Vanhove et al., 2013), S. mellandi (Boulenger, 1905), S. stappersi Trewavas, 1964, S. thumbergi (Castelnau, 1861), and S. angusticeps (Boulenger, 1905) (C. zambezensis was also found on O. mortimeri Trewavas, 1966; see Douëllou, 1993; Jorissen et al., 2018). Morphologically, C. antoineparisellei sp. nov. exhibits similar haptoral features to those of Tanganyikan C. brunnensis described from Trematocara unimaculatum Boulenger, 1901 Trematocarini Poll, 1986 (Kmentová et al., pairs I–IV, VI, and VII, and a copulatory duct without a swollen proximal portion, and exhibits a sclerotized vagina (see above). Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. This group includes C. acerbus Dossou, 1982; C. amieti Birgi & Euzet, 1983; C. amphoratus Pariselle & Euzet, 1996; C. berrebii Pariselle & Euzet, 1994; C. cirratus Paperna, 1964; C. cubitus Dossou, 1982; C. djietoi; C. giostrai Pariselle, Bilong Bilong & Euzet, 2003; C. karibae Douëllou, 1993; C. kothiasi Pariselle & Euzet, 1994; C. lagoonaris Paperna, 1969; C. levequei Pariselle & Euzet, 1996; C. louipaysani Pariselle & Euzet, 1995; C. mvogoi Pariselle, Bitja Nyom & Bilong Bilong, 2014; C. njinei Pariselle, Bilong Bilong & Euzet, 2003; C. ornatus Pariselle & Euzet, 1996; C. pouyaudi Pariselle & Euzet, 1994; C. sclerosus Paperna & Thurston, 1969; C. slembroucki Pariselle & Euzet, 1998; and C. zambezensis Douëllou, 1993. C. mbirizei Muterezi Bukinga et al., 2012, a parasite described from a representative of the tribe Oreochromini Dunz & Schliewen, 2013, Oreochromis tanganicae (Günther, 1894), is the only species hitherto known to have short hook pairs I–IV, VI, and VII and a sclerotized vagina in LT (see Muterezi Bukinga et al., 2012), and therefore C. antoineparisellei sp. nov. is the second known representative with the combination of these sclerotized structures in the lake. However, C. antoineparisellei sp. nov. is the first representative of Cichlidogyrus recognized on Tanganyikan tropheines which has a sclerotized vagina. Further, because of the systematic and phylogenetic affinities among the Tropheini and the Haplochromini Poll, 1986 (see Salzburger et al., 2005), C. antoineparisellei sp. nov. is considered to be the second representative species exhibiting this feature within the parasites infecting representatives of the Haplochromini lineage, like its congener C. zambezensis reported from the non-Tanganyikan haplochromines Serranochromis macrocephalus (Boulenger, 1899), S. robustus jallae (Günther, 1864) (Vanhove et al., 2013), S. mellandi (Boulenger, 1905), S. stappersi Trewavas, 1964, S. thumbergi (Castelnau, 1861), and S. angusticeps (Boulenger, 1905) (C. zambezensis was also found on O. mortimeri Trewavas, 1966; see Douëllou, 1993; Jorissen et al., 2018). Morphologically, C. antoineparisellei sp. nov. exhibits similar haptoral features to those of Tanganyikan C. brunnensis described from Trematocara unimaculatum Boulenger 1901 Trematocarini Poll 1986 (Kmentová et al pairs I–IV, VI, and VII, and a copulatory duct without a swollen proximal portion, and exhibits a sclerotized vagina (see above). This group includes C. acerbus Dossou, 1982; C. amieti Birgi & Euzet, 1983; C. amphoratus Pariselle & Euzet, 1996; C. berrebii Pariselle & Euzet, 1994; C. cirratus Paperna, 1964; C. cubitus Dossou, 1982; C. djietoi; C. Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. giostrai Pariselle, Bilong Bilong & Euzet, 2003; C. karibae Douëllou, 1993; C. kothiasi Pariselle & Euzet, 1994; C. lagoonaris Paperna, 1969; C. levequei Pariselle & Euzet, 1996; C. louipaysani Pariselle & Euzet, 1995; C. mvogoi Pariselle, Bitja Nyom & Bilong Bilong, 2014; C. njinei Pariselle, Bilong Bilong & Euzet, 2003; C. ornatus Pariselle & Euzet, 1996; C. pouyaudi Pariselle & Euzet, 1994; C. sclerosus Paperna & Thurston, 1969; C. slembroucki Pariselle & Euzet, 1998; and C. zambezensis Douëllou, 1993. C. mbirizei Muterezi Bukinga et al., 2012, a parasite described from a representative of the tribe Oreochromini Dunz & Schliewen, 2013, Oreochromis tanganicae (Günther, 1894), is the only species hitherto known to have short hook pairs I–IV, VI, and VII and a sclerotized vagina in LT (see Muterezi Bukinga et al., 2012), and therefore C. antoineparisellei sp. nov. is the second known representative with the combination of these sclerotized structures in the lake. However, C. antoineparisellei sp. nov. is the first representative of Cichlidogyrus recognized on Tanganyikan tropheines which has a sclerotized vagina. Further, because of the systematic and phylogenetic affinities among the Tropheini and the Haplochromini Poll, 1986 (see Salzburger et al., 2005), C. antoineparisellei sp. nov. is considered to be the second representative species exhibiting this feature within the parasites infecting representatives of the Haplochromini lineage, like its congener C. zambezensis reported from the non-Tanganyikan haplochromines Serranochromis macrocephalus (Boulenger, 1899), S. robustus jallae (Günther, 1864) (Vanhove et al., 2013), S. mellandi (Boulenger, 1905), S. stappersi Trewavas, 1964, S. thumbergi (Castelnau, 1861), and S. angusticeps (Boulenger, 1905) (C. zambezensis was also found on O. mortimeri Trewavas, 1966; see Douëllou, 1993; Jorissen et al., 2018). Morphologically, C. antoineparisellei sp. nov. exhibits similar haptoral features to those of Tanganyikan C. brunnensis described from Trematocara unimaculatum Boulenger, 1901 Trematocarini Poll, 1986 (Kmentová et al., 2016a), and a variety of West African parasite species such as C. amphoratus from Tilapia louka Thys van den Audenaerde, 1969 (Pariselle & Euzet, 1996), C. sclerosus Paperna & Thurston, 1969 from O. mossambicus (Peters, 1852), O. niloticus (Linnaeus, 1758), Haplochromis sp., O. leucosticus (Trewavas, 1933), C. zillii (Gervais, 1848), O. spilurus (Günther, 1894), O. aureus (Steindachner, 1864) (Paperna & Thurston, 1969; Pariselle & Euzet, 2009), O. mortimeri, S. microcephalus (Douëllou, 1993), and O. mweruensis Trewavas, 1983 (Jorissen et al., 2018), and C. giostrai from Sarotherodon caudomarginatus (Boulenger, 1916) (Pariselle, Bilong Bilong & Euzet, 2003). Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. These include the characteristic broad base and almost non-incised roots of the anchors. C. antoineparisellei sp. nov. shares its host species I. loocki with C. buescheri in addition to the morphology of its hook pairs and dorsal and ventral anchors (see Pariselle et al., 2015b). However, it is mainly distinguishable from the latter by (i) the shorter dorsal bar (26–29 mm in C. antoineparisellei sp. nov. vs 31–47 mm in C. buescheri), and (ii) the sclerotized vagina and exhibits a sclerotized vagina (see above). This group includes C. acerbus Dossou, 1982; C. amieti Birgi & Euzet, 1983; C. amphoratus Pariselle & Euzet, 1996; C. berrebii Pariselle & Euzet, 1994; C. cirratus Paperna, 1964; C. cubitus Dossou, 1982; C. djietoi; C. giostrai Pariselle, Bilong Bilong & Euzet, 2003; C. karibae Douëllou, 1993; C. kothiasi Pariselle & Euzet, 1994; C. lagoonaris Paperna, 1969; C. levequei Pariselle & Euzet, 1996; C. louipaysani Pariselle & Euzet, 1995; C. mvogoi Pariselle, Bitja Nyom & Bilong Bilong, 2014; C. njinei Pariselle, Bilong Bilong & Euzet, 2003; C. ornatus Pariselle & Euzet, 1996; C. pouyaudi Pariselle & Euzet, 1994; C. sclerosus Paperna & Thurston, 1969; C. slembroucki Pariselle & Euzet, 1998; and C. zambezensis Douëllou, 1993. C. mbirizei Muterezi Bukinga et al., 2012, a parasite described from a representative of the tribe Oreochromini Dunz & Schliewen, 2013, Oreochromis tanganicae (Günther, 1894), is the only species hitherto known to have short hook pairs I–IV, VI, and VII and a sclerotized vagina in LT (see Muterezi Bukinga et al., 2012), and therefore C. antoineparisellei sp. nov. is the second known representative with the combination of these sclerotized structures in the lake. However, C. antoineparisellei sp. nov. is the first representative of Cichlidogyrus recognized on Tanganyikan tropheines which has a sclerotized vagina. Further, because of the systematic and phylogenetic affinities among the Tropheini and the Haplochromini Poll, 1986 (see Salzburger et al., 2005), C. antoineparisellei sp. nov. is considered to be the second representative species exhibiting this feature within the parasites infecting representatives of the Haplochromini lineage, like its congener C. zambezensis reported from the non-Tanganyikan haplochromines Serranochromis macrocephalus (Boulenger, 1899), S. robustus jallae (Günther, 1864) (Vanhove et al., 2013), S. mellandi (Boulenger, 1905), S. stappersi Trewavas, 1964, S. thumbergi (Castelnau, 1861), and S. angusticeps (Boulenger, 1905) (C. zambezensis was also found on O. mortimeri Trewavas, 1966; see Douëllou, 1993; Jorissen et al., 2018). Morphologically, C. antoineparisellei sp. nov. Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. Dorsal anchors with poorly marked shaft and more pronounced guard (approximately four times the length of shaft), bent blade with arched point: a = 25 (24–26; n = 7); b = 22–23 (n = 7); c = 1–2 (n = 7); d = 6 (4–7; n = 7); e = 9 (8–10; n = 7). Dorsal bar curved, thick in the middle part with blunt endings and well-developed auricles: h = 11–12 (n = 7); w = 5–6 (n = 7); x = 27 (26–29; n = 7); y = 9 (8–11; n = 7). Ventral anchors similar in shape and size to dorsal ones: a = 26 (25–27; n = 7); b = 24 (23–25; n = 7); c = 2 (1–3; n = 7); d = 6 (3–7; n = 7); e = 10 (9–12; n = 7). V-shaped ventral bar: w = 4–5 (n = 7); x = 26 (24–27; n = 7). Haptor with seven short hook pairs, hooks V retain their larval size (see above), each hook with erect thumb and shank comprised of two subunits: pair I = 11–12 (n = 6) long, pair II = 13 (12–14; n = 7) long, pair III = 15 (13–16; n = 7) long, pair IV = 17 (15–18; n = 7) long, pair V = 11 (9–12; n = 6) long, pair VI = 16 (14–17; n = 7) long, and pair VII = 13–14 (n = 7) long. Male copulatory organ composed of long, curved copulatory duct that tapers distally: MCO = 42 (39–46; n = 10); Ct = 59 (54–62; n = 11). Heel relatively short, He = 6 (5–7; n = 11). Accessory piece is linked to basal bulb by thin filament, C-shaped in the middle part, composed of two distinct parts, one extremity ending in hook while the other connected to the latter with swollen base, Ap = 29 (22–33; n = 10). Long tubular vagina with two characteristic extremities, a broadened part and one extremity covered by a triangular-like structure: V = 40 (35–47; n = 11); v = 9 (7–13; n = 11). Diagnosis. According to the sclerites of the haptoral and reproductive organs, C. antoineparisellei sp. nov. belongs to the group which includes species with short hook C. antoineparisellei sp. nov. belongs to the group which includes species with short hook Rahmouni et al. Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. exhibits similar haptoral features to those of Tanganyikan C. brunnensis described from Trematocara unimaculatum Boulenger, 1901 Trematocarini Poll, 1986 (Kmentová et al., 2016a), and a variety of West African parasite species such as C. amphoratus from Tilapia louka Thys van den Audenaerde, 1969 (Pariselle & Euzet, 1996), C. sclerosus Paperna & Thurston, 1969 from O. mossambicus (Peters, 1852), O. niloticus (Linnaeus, 1758), Haplochromis sp., O. leucosticus (Trewavas, 1933), C. zillii (Gervais, 1848), O. spilurus (Günther, 1894), O. aureus (Steindachner, 1864) (Paperna & Thurston, 1969; Pariselle & Euzet, 2009), O. mortimeri, S. microcephalus (Douëllou, 1993), and O. mweruensis Trewavas, 1983 (Jorissen et al., 2018), and C. giostrai from Sarotherodon caudomarginatus (Boulenger, 1916) (Pariselle, Bilong Bilong & Euzet, 2003). These include the characteristic broad base and almost non-incised roots of the anchors. C. antoineparisellei sp. nov. shares its host species I. loocki with C. buescheri in addition to the morphology of its hook pairs and dorsal and ventral anchors (see Pariselle et al., 2015b). However, it is mainly distinguishable from the latter by (i) the shorter dorsal bar (26–29 mm in C. antoineparisellei sp. nov. vs 31–47 mm in C. buescheri), and (ii) the sclerotized vagina (no sclerotized vagina in C. buescheri). p g y Trematocara unimaculatum Boulenger, 1901 Trematocarini Poll, 1986 (Kmentová et al., 2016a), and a variety of West African parasite species such as C. amphoratus from Tilapia louka Thys van den Audenaerde, 1969 (Pariselle & Euzet, 1996), C. sclerosus Paperna & Thurston, 1969 from O. mossambicus (Peters, 1852), O. niloticus (Linnaeus, 1758), Haplochromis sp., O. leucosticus (Trewavas, 1933), C. zillii (Gervais, 1848), O. spilurus (Günther, 1894), O. aureus (Steindachner, 1864) (Paperna & Thurston, 1969; Pariselle & Euzet, 2009), O. mortimeri, S. microcephalus (Douëllou, 1993), and O. mweruensis Trewavas, 1983 (Jorissen et al., 2018), and C. giostrai from Sarotherodon caudomarginatus (Boulenger, 1916) (Pariselle, Bilong Bilong & Euzet, 2003). These include the characteristic broad base and almost non-incised roots of the anchors. C. antoineparisellei sp. nov. shares its host species I. loocki with C. buescheri in addition to the morphology of its hook pairs and dorsal and ventral anchors (see Pariselle et al., 2015b). However, it is mainly distinguishable from the latter by (i) the shorter dorsal bar (26–29 mm in C. antoineparisellei sp. nov. vs 31–47 mm in C. buescheri), and (ii) the sclerotized vagina (no sclerotized vagina in C. buescheri). Rahmouni et al. Paratypes: MRAC M.T.38441–43; MNHN HEL750–51. Etymology: the species epithet ‘antoineparisellei’ honours the French parasitologist Dr. Antoine Pariselle, researcher at the Institut de Recherche pour le Développement, who extensively studied the monogeneans of African fish, trained countless people in fish parasitology, and described more than fifty species of Cichlidogyrus, of which three parasitize I. loocki. Cichlidogyrus antoineparisellei sp. nov. Fig. 6 urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. (2018), PeerJ, DOI 10.7717/peerj.5604 Type-host: Interochromis loocki (Poll, 1949) (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840 T h i i P ll 1986) Figure 7 Sclerotized structures of Cichlidogyrus masilyai sp. nov. ex Petrochromis orthognathus. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-7 Figure 7 Sclerotized structures of Cichlidogyrus masilyai sp. nov. ex Petrochromis orthognathus. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-7 Type-host: Interochromis loocki (Poll, 1949) (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Tropheini Poll, 1986). Type-host: Interochromis loocki (Poll, 1949) (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Tropheini Poll, 1986). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Holotype: MRAC M.T.38440. Holotype: MRAC M.T.38440. Paratypes: MRAC M.T.38441–43; MNHN HEL750–51. Cichlidogyrus masilyai sp. nov. Fig. 7 urn:lsid:zoobank.org:act:4D5AF9E3-9CCE-4D3F-95B4-CCDA7B1A5C91. Site of infection: Gills. Type-locality: Pemba (340′S, 2910′E; caught when snorkelling), DRC, LT. Type-locality: Pemba (340′S, 2910′E; caught when snorkelling), DRC, LT. Prevalence & intensity of infection: two fish specimens infected/two fish specimens examined, one to seven parasite specimens per infected host. urn:lsid:zoobank.org:act:4D5AF9E3-9CCE-4D3F-95B4-CCDA7B1A5C91. Description. Based on 12 specimens fixed in GAP. Body 622 (523–721; n = 2) long, 107 (56–158; n = 2) wide at mid-body. Dorsal anchors with short shaft and pronounced Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 guard (approximately two times the length of shaft), bent blade with arched point: a = 28 (27–29; n = 3); b = 21 (20–23; n = 3); c = 5 (4–6; n = 3); d = 10 (9–11; n = 3); e = 7 (5–8; n = 4). Dorsal bar well-arched, C-shaped with relatively short auricles: h = 10 (8–11; n = 3); w = 6 (5–7; n = 3); x = 28 (22–24; n = 3); y = 14 (13–15; n = 3). Ventral anchors with shorter shaft than guard and arched point: a = 28 (27–29; n = 3); b = 24 (23–25; n = 3); c = 4 (2–5; n = 3); d = 9 (6–10; n = 3); e = 9 (8–10; n = 3). W-shaped ventral bar: w = 5 (4–6; n = 3); x = 32 (31–33; n = 3). Haptor with seven short hook pairs, hooks V retain their larval size (see above), each hook with erect thumb and shank comprised of two subunits: pair I = 12–13 (n = 3) long, pair II = 16 (15–17; n = 3) long, pair III = 20 (19–21; n = 3) long, pair IV = 22 (21–23; n = 3) long, pair V = 12 (10–13; n = 3) long, pair VI = 22–23 (n = 3) long, and pair VII = 19 (17–21; n = 3) long. Male copulatory organ beginning as an irregularly shaped bulb with thin copulatory duct, slightly curved proximally, straight halfway, folded back distally: MCO = 33 (31–34; n = 3); Ct = 35 (32–36; n = 3). Poorly developed heel at the side of the basal bulb, He = 1–2 (n = 3). Accessory piece seems to be unattached to basal bulb, is thin, slightly curved in the middle part with pincer-like ending, Ap = 26 instead of 29 (24–29; n = 3). Vagina non-sclerotized. Diagnosis. Cichlidogyrus masilyai sp. nov. belongs to the same group as C. adkoningsi sp. nov. and C. koblmuelleri sp. nov. (see above). The new species shows a similar morphotype to several representatives of Cichlidogyrus described from non-Tanganyikan haplochromine cichlids such as C. Type-host: Petrochromis orthognathus Matthes, 1959 (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Tropheini Poll, 1986). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 urn:lsid:zoobank.org:act:4D5AF9E3-9CCE-4D3F-95B4-CCDA7B1A5C91. bifurcatus from Astatotilapia flaviijosephi (Lortet, 1883) (Paperna, 1960), C. haplochromii from H. guiarti (Pellegrin, 1904) (Paperna & Thurston, 1969) and C. zambezensis, in addition to C. gillardinae from the Tanganyikan haplochromine A. burtoni (Günther, 1894) (Muterezi Bukinga et al., 2012). Further, the exceptional shape of the dorsal bar in the new species, that is, well-arched, C-shaped, is observed for the first time in Cichlidogyrus spp. infecting a tropheine cichlid. The new species exhibits the same hook pairs, similarly sized dorsal and ventral anchors, a non-sclerotized vagina, and an accessory piece ending in a pincer-like structure as in C. buescheri, a parasite of I. loocki. However, the shape of the dorsal and ventral anchors is different, that is, the shaft is more clearly marked in C. masilyai sp. nov. compared to that observed in C. buescheri. Further, it differs from C. buescheri by (i) the shorter and differently shaped dorsal bar (C-shaped, 22–24 mm in C. masilyai sp. nov. vs moderately arched, 31–45 mm in C. buescheri), (ii) the copulatory duct (thin, slightly curved proximally, straight halfway, folded distally, 31–34 mm in C. masilyai sp. nov. vs curved duct with narrow extremity, 49–58 mm in C. buescheri), (iv) the heel (poorly developed, lateral to the bulb, 1–2 mm in C. masilyai sp. nov. vs large heel of irregular shape at the basis of the bulb, 6–11 mm in C. buescheri), and (v) the accessory piece (seems to be unattached to basal bulb and thin, 24–29 mm in C. masilyai sp. nov. vs wide, directly attached to the basal bulb, 31–52 mm in C. buescheri). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Figure 8 Sclerotized structures of Cichlidogyrus salzburgeri sp. nov. ex Petrochromis trawavasae. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-8 Figure 8 Sclerotized structures of Cichlidogyrus salzburgeri sp. nov. ex Petrochromis trawavasae. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-8 Paratype: MRAC M.T.38446. Etymology: the species epithet ‘masilyai’ is given in honour of the Congolese biologist Prof. Dr. Pascal Masilya Mulungula of the Institut Supérieur Pédagogique de Bukavu and general director of the Centre de Recherche en Hydrobiologie-Uvira (DRC), in appreciation of the hospitality and help received during the fieldtrip for the present study. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Site of infection: Gills. Type-locality: Pemba (340′S, 2910′E; caught when snorkelling), DRC, LT. Prevalence & intensity of infection: two fish specimens infected/four fish specimens examined, one to two parasite specimens per infected host. Holotype: MRAC M.T.38447. Holotype: MRAC M.T.38447. Cichlidogyrus salzburgeri sp. nov. Fig. 8 urn:lsid:zoobank.org:act:612A6D2B-09DD-4902-8640-A4877277E7F3. urn:lsid:zoobank.org:act:612A6D2B-09DD-4902-8640-A4877277E7F3. Description. Based on four specimens fixed in GAP. Dorsal anchors with short shaft and more pronounced guard (approximately three times the length of shaft), bent blade with arched point: a = 27 (26–29; n = 3); b = 18 (17–19; n = 3); c = 4–5 (n = 3); d = 12–13 (n = 3); e = 8 (7–9; n = 4). Dorsal bar slightly curved with well-developed auricles: h = 22 (21–24; n = 3); w = 6–7 (n = 3); x = 32 (35–38; n = 3); y = 11–12 (n = 4). Ventral anchors with shorter shaft than guard and arched point: a = 32 (31–33; n = 3); Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 b = 26–27 (n = 3); c = 5–6 (n = 3); d = 11–12 (n = 3); e = 12–13 (n = 3). V-shaped ventral bar: w = 7–8 (n = 3); x = 39 (38–40; n = 3). Haptor with short hook pair I, hooks V retain their larval size (see above), pairs II–IV, VI, and VII long, each hook with erect thumb and shank comprised of two subunits: pair I = 12–13 (n = 3) long, pair II = 23–24 (n = 3) long, pair III = 28–29 (n = 3) long, pair IV = 31–32 (n = 3) long, pair V = 11–12 (n = 3) long, pair VI = 25–26 (n = 3) long, and pair VII = 25–26 (n = 3) long. Male copulatory organ with straight, wide copulatory duct and large distal opening: MCO = 42 (41–44; n = 4); Ct = 40 (39–42; n = 4). Short heel, He = 3 (2–4; n = 4). Accessory piece linked to basal bulb, thin proximally, straight, pincer-like ending with one extremity shorter than the other, Ap = 29 (28–30; n = 4). Vagina non-sclerotized. Diagnosis. Cichlidogyrus salzburgeri sp. nov. belongs to the group of Cichlidogyrus which are characterized by a shorter hook pair I (pair V with larval size) and longer pairs II–IV, VI, and VII, a copulatory duct without a swollen distal portion, and a non-sclerotized vagina. Of all Cichlidogyrus spp., only C. halli Price & Kirk, 1967 and the Tanganyikan C. sturmbaueri (Vanhove, Volckaert & Pariselle, 2011) and C. rectangulus (Rahmouni, Vanhove & Šimková, 2017) are included in this group. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 urn:lsid:zoobank.org:act:612A6D2B-09DD-4902-8640-A4877277E7F3. Therefore, the new species is the third representative described from endemic Tanganyikan cichlids displaying this hook configuration. C. salzburgeri sp. nov. differs from its congeners C. buescheri, C. schreyenbrichardorum, and C. vealli isolated from I. loocki, and from the undescribed species previously identified on some representatives of Petrochromis, by the shape of the anchors, specifically the marked bases in the ventral ones (Pariselle et al., 2015b). The dorsal bar in C. salzburgeri sp. nov. with its well-developed auricles is reminiscent of C. muterezii described from S. diagramma (see Van Steenberge et al., 2015). However, C. salzburgeri sp. nov. differs from the latter by (i) the hook pairs (shorter hook pair I and longer pairs II–IV, VI, and VII in C. salzburgeri sp. nov. vs short hook pairs I–IV, VI, and VII in C. muterezii) and (ii) the pincer-like ending in the accessory piece, a feature missing in C. muterezii. Further, according to the comparative morphology of the reproductive organs, C. salzburgeri sp. nov. differs from C. franswittei found on the gills of P. marginatus and P. curvifrons (see Van Steenberge et al., 2015) by (i) the hook pairs (shorter hook pair I and longer pairs II–IV, VI, and VII in C. salzburgeri sp. nov. vs short hook pairs I–IV, VI, and VII in C. franswittei), (ii) the shorter dorsal anchors (26–29 mm in C. salzburgeri sp. nov. vs 31–40 mm in C. franswittei), (iii) the shorter copulatory duct (39–42 mm in C. salzburgeri sp. nov. vs 47–57 mm in C. franswittei), and (iv) the shorter accessory piece (28–30 mm in C. salzburgeri sp. nov. vs 31–46 mm in C. franswittei). C. salzburgeri sp. nov. is distinguishable from the newly described species C. masilyai sp. nov. by (i) the longer dorsal bar (35–38 mm in C. salzburgeri sp. nov. vs 22–24 mm in C. masilyai sp. nov.), (ii) the length of its auricles (21–24 mm in C. salzburgeri sp. nov. vs 8–11 mm in C. masilyai sp. nov.), (iii) the differently shaped copulatory duct (straight, large copulatory duct and large distal opening in C. salzburgeri sp. nov. vs thin, slightly curved proximally, straight halfway, folded distally in C. masilyai sp. nov.), and (v) the differently shaped accessory piece (linked to basal bulb, thin proximally, straight, ending pincer-like with one extremity shorter than the Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 20/37 Figure 9 Sclerotized structures of Cichlidogyrus sergemorandi sp. nov. urn:lsid:zoobank.org:act:612A6D2B-09DD-4902-8640-A4877277E7F3. ex Tylochromis polylepis. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece; Vg, vagina. Full-size  DOI: 10.7717/peerj.5604/fig-9 Figure 9 Sclerotized structures of Cichlidogyrus sergemorandi sp. nov. ex Tylochromis polylepis. DA, dorsal anchor; DB, dorsal bar; VA, ventral anchor; VB, ventral bar; I–VII, hooks; MCO, male copulatory organ; He, heel; Ct, copulatory tube; Ap, accessory piece; Vg, vagina. Full-size  DOI: 10.7717/peerj.5604/fig-9 other in C. salzburgeri sp. nov. vs thin and slightly curved in the middle part ending in pincer-like structure in C. masilyai sp. nov.). other in C. salzburgeri sp. nov. vs thin and slightly curved in the middle part ending in pincer-like structure in C. masilyai sp. nov.). Type-host: Petrochromis trewavasae Poll, 1948 (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Tropheini Poll, 1986). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Paratype: MRAC M.T.38449. Paratype: MRAC M.T.38449. Etymology: the specific epithet ‘salzburgeri’ honours the biologist Dr. Walter Salzburger, professor at the University of Basel (Switzerland), for his work on cichlid evolution and in appreciation of his assistance during the fieldtrip. Cichlidogyrus sergemorandi sp. nov. Fig. 9 urn:lsid:zoobank.org:act:7A4653B3-2C25-4E3A-9D4A-3FA76165C167. Host accession number: MH298007. Type-locality: Pemba (340′S, 2910′E; caught when snorkelling), DRC, LT. Type-locality: Pemba (340′S, 2910′E; caught when snorkelling), DRC, LT. Prevalence & intensity of infection: one fish specimen infected/two fish specimen examined, four parasite specimens on infected fish. Prevalence & intensity of infection: one fish specimen infected/two fish specimen examined, four parasite specimens on infected fish. Holotype: MRAC M.T.38448. urn:lsid:zoobank.org:act:7A4653B3-2C25-4E3A-9D4A-3FA76165C167. Description. Based on seven specimens fixed in GAP. Body 659 (594–769; n = 4) long, 115 (95–146; n = 4) wide at mid-body. Dorsal anchors with short shaft and more Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 pronounced guard (approximately two times the length of shaft), slightly bent blade with arched point: a = 24 (23–26; n = 5); b = 19 (18–20; n = 5); c = 2–3 (n = 5); d = 7 (6–8; n = 5); e = 6–7 (n = 5). C-shaped dorsal bar with blunt endings and short auricles: h = 4–5 (n = 5); w = 3–4 (n = 5); x = 18 (16–20; n = 5); y = 9 (7–10; n = 5). Ventral anchors with shorter shaft than guard and arched point: a = 23 (22–24; n = 4); b = 20–21 (n = 5); c = 3 (2–4; n = 5), d = 7 (6–8; n = 5); e = 9 (8–10; n = 5). V-shaped ventral bar: w = 4–5 (n = 5); x = 26 (25–27; n = 5). Haptor with seven short hook pairs, hooks V retain their larval size, each hook with erect thumb and shank comprised of two subunits (see above): pair I = 13–14 (n = 5) long, pair II = 18 (17–19; n = 5) long, pair III = 21 (19–22; n = 5) long, pair IV = 22 (21–23; n = 5) long, pair V = 10 (9–11; n = 5) long, pair VI = 22 (21–24; n = 5) long, and pair VII = 22 (20–24; n = 4) long. Male copulatory organ with an elongated basal bulb prolonged into a long, spirally coiled copulatory duct with thick walls, ending distally in a sharp extremity: MCO = 40 (38–44; n = 6); Ct = 45 (42–47; n = 7). Distinct heel, He = 5 (4–6; n = 6). Large accessory piece unattached to the basal bulb and pierced distally by the copulatory duct, Ap = 38 (33–39; n = 7). Vagina pouch-like in shape, variable in size: V = 29 (19–38; n = 6); v = 9 (7–11; n = 6). Diagnosis. Regarding the morphology of the hook pairs and the vagina, C. sergemorandi sp. nov. belongs to the same morphological group as C. antoineparisellei sp. nov. Diagnosis. Regarding the morphology of the hook pairs and the vagina, C. sergemorandi sp. nov. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 urn:lsid:zoobank.org:act:7A4653B3-2C25-4E3A-9D4A-3FA76165C167. belongs to the same morphological group as C. antoineparisellei sp. nov. sp. nov. belongs to the same morphological group as C. antoineparisellei sp. nov. (see above). The newly described species possesses a dorsal bar with short auricles, a spirally shaped copulatory duct, and an accessory piece separated from the basal bulb, features observed in many representatives described on tylochromine cichlids such as C. mulimbwai and C. muzumanii, parasites of the same cichlid host from LT (Muterezi Bukinga et al., 2012), and the non-Tanganyikan C. berrebii, C. kothiasi, and C. pouyaudi from T. jentinki (Steindachner, 1894) (Pariselle & Euzet, 1994), and C. chrysopiformis and C. djietoi from T. sudanensis Daget, 1954 (Pariselle, Bitja Nyom & Bilong Bilong, 2014). C. sergemorandi sp. nov. is most reminiscent of C. mulimbwai. The former differs from C. mulimbwai by (i) the shorter dorsal and ventral anchors (respectively, 23–26; 22–24 mm in C. sergemorandi sp. nov. vs 32–42; 31–40 mm in C. mulimbwai), (ii) the shorter ventral and dorsal bars (respectively, 16–20; 25–27 mm in C. sergemorandi sp. nov. vs 33–44; 38–52 mm in C. mulimbwai), and (iii) the vagina (no sclerotized vagina in C. mulimbwai unlike in C. sergemorandi sp. nov.). C. sergemorandi sp. nov. differs from C. muzumanii by (i) the first hook pair (thin pair I in C. sergemorandi sp. nov. vs large pair in C. muzumanii), (ii) the shorter dorsal and ventral anchors (respectively, 23–26; 22–24 mm in C. sergemorandi sp. nov. vs 44–58; 37–47 mm in C. muzumanii), (iii) the shorter dorsal bar and its auricles (respectively, 16–20; 4–5 mm in C. sergemorandi sp. nov. vs 45–62; 10–19 mm in C. muzumanii), (iv) the shorter ventral bar (25–27 mm in C. sergemorandi sp. nov. vs 47–63 mm in C. muzumanii), (v) the shorter copulatory duct distally differently shaped (sharp extremity, 42–47 mm in C. sergemorandi sp. nov. vs broad extremity (not mentioned in the original description), 57–68 mm in C. muzumanii), (vi) the longer accessory piece (33–39 mm in C. sergemorandi sp. nov. vs 17–20 mm in C. muzumanii), and (vii) the vagina (no sclerotized vagina in C. muzumanii vs sclerotized vagina in C. sergemorandi sp. nov.). Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Table 2 Overview of the undescribed species of Cichlidogyrus characterized in the present study with their type-host and locality of sampling. Cichlidogyrus spp. Host Type-locality C. sp. ‘Callochromis melanostigma’ Fig. 10A C. Paratype: MRAC M.T.38445. Paratype: MRAC M.T.38445. Etymology: the specific epithet of the new species ‘sergemorandi’ honours the French evolutionary biologist and ecologist Dr. Serge Morand from CNRS and CIRAD (France), and CILM (Laos) for his work on ecology and the evolution of parasites. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 urn:lsid:zoobank.org:act:7A4653B3-2C25-4E3A-9D4A-3FA76165C167. melanostigma (Boulenger, 1906) Kilomoni beach (320′S, 2910′E) C. sp. ‘Xenotilapia flavipinnis 1’ Fig. 10B X. flavipinnis Poll, 1985 Pemba (337′S, 299′E) C. sp. ‘Xenotilapia flavipinnis 2’ Fig. 10C C. sp. ‘Interochromis loocki 5’ Fig. 10D I. loocki (Poll, 1949) Pemba (337′S, 299′E) C. sp. ‘Petrochromis orthognathus 2’ Fig. 10E P. orthognathus Matthes, 1959 Pemba (337′S, 299′E) C. sp. ‘Petrochromis orthognathus 3’ Fig. 10F Table 2 Overview of the undescribed species of Cichlidogyrus characterized in the present study with their type-host and locality of sampling. Previous record: Cichlidogyrus sp. ‘T. polylepis 3’ Muterezi Bukinga et al., 2012. Type-host: Tylochromis polylepis (Boulenger, 1900) (Perciformes Bleeker, 1859: Cichlidae Heckel, 1840: Tylochromini Poll, 1986). Cichlidogyrus sp. ‘C. melanostigma’ Fig. 10A Host accession number: MH298008. Site of infection: Gills. Type-locality: Mulongwe fish market (322′S, 296′E), Uvira, DRC, LT Type-locality: Mulongwe fish market (322′S, 296′E), Uvira, DRC, LT. Prevalence & intensity of infection: two fish specimens infected/three fish specimens examined, one to five parasite specimens per infected host. Prevalence & intensity of infection: two fish specimens infected/three fish specimens examined, one to five parasite specimens per infected host. Holotype: MRAC M.T.38444. Holotype: MRAC M.T.38444. Paratype: MRAC M.T.38445. Morphological characterisation of the MCO of the undescribed On the basis of the MCO features, we characterize six new undescribed species, namely C. sp. ‘C. melanostigma’, C. sp. ‘X. flavipinnis 1’, and C. sp. ‘X. flavipinnis 2’ (all from ectodine hosts), and C. sp. ‘I. loocki 5’, C. sp. ‘P. orthognathus 2’, and C. sp. ‘P. orthognathus 3’ (all from tropheine hosts) (Table 2; Fig. 10A–10F). These species could not be formally described because of the lack of material. Further, the haptoral parts of some specimens were not clearly visible for drawings. However, the general shape of the MCO provided sufficient morphological information to distinguish these species that can be considered as new to science. The morphological characterization of the MCO of the undescribed species of Cichlidogyrus recognized on Congolese host specimens did not allow us to confirm the presence/absence of a sclerotized vagina. Cichlidogyrus sp. ‘C. melanostigma’ Fig. 10A Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Figure 10 Sclerotized structures of the undescribed species of Cichlidogyrus characterized in this study. (A) Cichlidogyrus sp. ‘C. melanostigma.’ (B) Cichlidogyrus sp. ‘X. flavipinnis 1.’ (C) Cichlido- gyrus sp. ‘X. flavipinnis 2.’ (D) Cichlidogyrus sp. ‘I. loocki 5.’ (E) Cichlidogyrus sp. ‘P. orthognathus 2.’ (F) Cichlidogyrus sp. ‘P. orthognathus 3.’ He, heel; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-10 Figure 10 Sclerotized structures of the undescribed species of Cichlidogyrus characterized in this study. (A) Cichlidogyrus sp. ‘C. melanostigma.’ (B) Cichlidogyrus sp. ‘X. flavipinnis 1.’ (C) Cichlido- gyrus sp. ‘X. flavipinnis 2.’ (D) Cichlidogyrus sp. ‘I. loocki 5.’ (E) Cichlidogyrus sp. ‘P. orthognathus 2.’ (F) Cichlidogyrus sp. ‘P. orthognathus 3.’ He, heel; Ct, copulatory tube; Ap, accessory piece. Full-size  DOI: 10.7717/peerj.5604/fig-10 In the gills of the ectodine C. melanostigma we found a single monogenean species, namely C. sp. ‘C. melanostigma’. Its MCO lacks a heel and possesses an ovoid basal bulb prolonged into a curved copulatory duct ending in a large opening: MCO = 46 (45–47; n = 2); Ct = 47–48 (n = 2). The accessory piece is curved, L-shaped, attached to the basal bulb and ending in a thick hook, Ap = 36–37 (n = 2). Further, a visible second wall lines the surface of the copulatory duct, which confers a twisted appearance. Cichlidogyrus sp. ‘C. melanostigma’ resembles C. banyankimbonai described from S. diagramma regarding the large distal opening of the copulatory duct (see Van Steenberge et al., 2015). However, the MCO of C. sp. ‘C. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Cichlidogyrus sp. ‘X. flavipinnis 1’ Fig. 10B The MCO of C. sp. ‘X. flavipinnis 1’ exhibits a small heel and a thick, well curved copulatory duct, MCO = 37 (n = 1); He = 2 (n = 1); Ct = 36 (n = 1). The subterminal opening of the copulatory duct seems to be located at the last third. Such a feature has never been reported in Cichlidogyrus species described so far. Thus, we cannot confirm whether C. sp. ‘X. flavipinnis 1’ really exhibits this characteristic in its copulatory duct, or it is only an artefact due to isolation or fixation procedures. The accessory piece attached to the basal bulb is thick, curved both proximally and distally with a finger-like ending, Ap = 29 (n = 1). Cichlidogyrus sp. ‘X. flavipinnis 1’ is most similar to C. sp. ‘C. melanostigma’. Both species occur on closely related ectodine host species, that is, X. flavipinnis and C. melanostigma. However, C. sp. ‘X. flavipinnis 1’ presents a heel, a feature missing in C. sp. ‘C. melanostigma’, in addition to the differently shaped proximal endings of the accessory piece (finger-like in C. sp. ‘X. flavipinnis 1’ vs thick hook in C. sp. ‘C. melanostigma’). Therefore, we consider C. sp. ‘X. flavipinnis 1’ and C. sp. ‘C. melanostigma’ as two different species. Morphological characterisation of the MCO of the undescribed melanostigma’ lacks a heel, unlike that of C. banyankimbonai. In addition, C. sp. ‘C. melanostigma’ exhibits a differently shaped accessory piece ending in a hook, a feature not observed in C. banyankimbonai. According to the comparative morphology of the MCO, C. sp. ‘C. melanostigma’ is reminiscent of C. discophonum from A. dewindti, both parasites of ectodine cichlids (see Rahmouni, Vanhove & Šimková, 2017). The two species lack a heel and possess a curved copulatory duct, and an accessory piece ending in a hook. However, C. sp. ‘C. melanostigma’ exhibits a longer accessory piece (36–37 mm in C. sp. ‘C. melanostigma’ vs 15–22 mm in C. discophonum). Further, the copulatory duct in C. sp. ‘C. melanostigma’ appears double walled and ends in a large opening, while there is a thick proximal part which tapers distally in C. discophonum. Also, the accessory piece in C. sp. ‘C. melanostigma’ is In the gills of the ectodine C. melanostigma we found a single monogenean species, namely C. sp. ‘C. melanostigma’. Its MCO lacks a heel and possesses an ovoid basal bulb prolonged into a curved copulatory duct ending in a large opening: MCO = 46 (45–47; n = 2); Ct = 47–48 (n = 2). The accessory piece is curved, L-shaped, attached to the basal bulb and ending in a thick hook, Ap = 36–37 (n = 2). Further, a visible second wall lines the surface of the copulatory duct, which confers a twisted appearance. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 24/37 L-shaped, while C. discophonum possesses an accessory piece composed of two parts twisted distally. Regarding these differences, we consider C. sp. ‘C. melanostigma’ to be a different species from C. discophonum and new to science. Cichlidogyrus sp. ‘X. flavipinnis 2’ Fig. 10C The second undescribed species isolated from X. flavipinnis has a shorter MCO compared to that of C. sp. ‘X. flavipinnis 2’, and a slightly curved copulatory duct with a narrow extremity, MCO = 22 (n = 1); Ct = 24 (n = 1). Heel reduced to inconspicuous, He = 1 (n = 1). The accessory piece is proximally thick and irregularly shaped, attached to the basal bulb by a small additional part, distally with pincer-like ending, Ap = 16 (n = 1). The heel structure in C. sp. ‘C. melanostigma’, C. sp. ‘X. flavipinnis 1’, and C. sp. ‘X. flavipinnis 2’ is absent to inconspicuous. Further, C. sp. ‘X. flavipinnis 2’ is easily distinguishable from C. sp. ‘C. melanostigma’ and C. sp. ‘X. flavipinnis 1’ (i) by its shorter MCO and a dissimilar copulatory duct with a narrow extremity, and (ii) by the differently sized accessory piece and its characteristic pincer-like shape at its extremity, the distal parts of which are finger and hook-like in C. sp. ‘X. flavipinnis 1’ and C. sp. ‘C. melanostigma’, respectively. On the basis of these differences, we consider C. sp. ‘X. flavipinnis 2’ to be different from C. sp. ‘C. melanostigma’ and C. sp. ‘X. flavipinnis 1’ and to represent a new species to science. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Cichlidogyrus ‘I. loocki 5’ Fig. 10D Specimens of I. loocki harboured a second parasite species and, according to the morphology of the MCO, we consider it as new to science. Cichlidogyrus sp. ‘I. loocki 5’ exhibits a long MCO with a heel attached to the side of an irregularly shaped basal bulb, MCO = 79 (n = 1); He = 6 (n = 1). The copulatory duct of constant width is long, wavy, and well curved proximally, Ct = 104 (n = 1). The accessory piece is separated Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 proximally into two thin parts, curved halfway with a twisted gutter-like appearance, distally with an additional small outgrowth and an irregular blunt ending, Ap = 64 (n = 1). However, we cannot reliably confirm the blunt shape in the distal part of the accessory piece since we cannot reject the possibility that this feature is an artefact of parasite fixation. According to the morphology of the MCO, C. sp. ‘I. loocki 5’ is most similar to C. georgesmertensi from P. babaulti in having a long and wavy copulatory duct, in addition to a curved accessory piece attached to the basal bulb (see Van Steenberge et al., 2015). However, C. sp. ‘I. loocki 5’ exhibits a heel attached to the side of the basal bulb like in C. masilyai sp. nov., while in C. georgesmertensi, it is located at the bottom of the basal bulb as in most species of Cichlidogyrus. Further, no gutter-like appearance or additional outgrowth in the distal part of the accessory piece were reported for C. georgesmertensi. Cichlidogyrus sp. ‘I. loocki 5’ can be compared to C. vealli (see Pariselle et al., 2015b). Indeed, the new undescribed species shares the host I. loocki with C. vealli, but the two parasites were sampled in different localities (northern lakeshore vs southern tip of the lake in the case of C. vealli). Moreover, in C. sp. ‘I. loocki 5’, the extremity of the MCO is narrower than that of C. vealli. Also, the undescribed species shows an additional small outgrowth at the proximal third of its MCO, a feature missing in its congener. Further, the accessory piece in C. sp. ‘I. loocki 5’ presents a gutter-like structure only in the middle part and a blunt extremity. In C. vealli, this feature exists throughout the accessory piece. On the basis of these differences, we consider C. sp. ‘I. Cichlidogyrus ‘I. loocki 5’ Fig. 10D loocki 5’ to be a different species from C. vealli. Cichlidogyrus sp. ‘P. orthognathus 2’ Fig. 10E Cichlidogyrus sp. ‘P. orthognathus 2’ possesses a long MCO exhibiting an elongated irregularly shaped basal bulb with long heel, MCO = 55 (53–57; n = 4); He = 7 (6–8; n = 4). The copulatory duct is well curved, Ct = 49 (47–51; n = 4). The accessory piece is linked to the basal bulb by a thin filament, curved in the middle part, distally with an outgrowth ending in a hook, Ap = 38 (37–40; n = 4). Cichlidogyrus sp. ‘P. orthognathus 2’ presents a morphotype similar to that of its undescribed congener C. sp. ‘I. loocki 5’, especially in having the additional small outgrowth in the distal part of the accessory piece. However, the MCO and the copulatory duct are shorter in C. sp. ‘P. orthognathus 2’ than in C. sp. ‘I. loocki 5’. Further, the heel is located differently (see above). Thus, on the basis of the morphological differences listed above, we separated the species C. sp. ‘P. orthognathus 2’ from its congener C. sp. ‘I. loocki 5’. Cichlidogyrus sp. ‘P. orthognathus 2’ possesses a long MCO exhibiting an elongated irregularly shaped basal bulb with long heel, MCO = 55 (53–57; n = 4); He = 7 (6–8; n = 4). The copulatory duct is well curved, Ct = 49 (47–51; n = 4). The accessory piece is linked to the basal bulb by a thin filament, curved in the middle part, distally with an outgrowth ending in a hook, Ap = 38 (37–40; n = 4). Cichlidogyrus sp. ‘P. orthognathus 2’ presents a morphotype similar to that of its undescribed congener C. sp. ‘I. loocki 5’, especially in having the additional small outgrowth in the distal part of the accessory piece. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 DISCUSSION The cichlid fishes of LT have undergone spectacular diversification, filling a diversity of ecological niches within a short time period, and therefore represent one of the most interesting models of adaptive radiation (Tsuboi et al., 2016). Research on cichlids is nowadays combined with the investigation of their parasite diversity in order to study the speciation of both cichlids and their specific parasites (Vanhove et al., 2015, 2016). Currently, only 32 Cichlidogyrus spp. are known from cichlids inhabiting LT (see the overview by Rahmouni et al. (2017)). The present study, based on morphological characters, increases knowledge of the diversity of host specific monogenean species in the cichlids living the northern sub-basin of the lake. We provide descriptions of seven gill monogenean species parasitizing six Congolese representatives of four cichlid tribes in LT. They are C. adkoningsi sp. nov. on C. frontosa (Cyphotilapiini); C. koblmuelleri sp. nov. on C. schoutedeni (Ectodini); C. habluetzeli sp. nov. on C. frontosa and C. schoutedeni; C. antoineparisellei sp. nov. on I. loocki; C. masilyai sp. nov. on P. orthognathus; C. salzburgeri sp. nov. on P. trewavasae (both Tropheini); and finally C. sergemorandi sp. nov. on T. polylepis (Tylochromini). We characterized also six undescribed parasite species on the basis of the morphology of their MCO. They are C. sp. ‘C. melanostigma’ from C. melanostigma; C. sp. ‘X. flavipinnis 1’ and C. sp. ‘X. flavipinnis 2’ from X. flavipinnis (both Ectodini); C. sp. ‘I. loocki 5’ from I. loocki; and finally C. sp. ‘P. orthognathus 2’ and C. sp. ‘P. orthognathus 3’ from P. orthognathus (all Tropheini). First, we identified the cichlid species using morphology and DNA sequence data from the mitochondrial cyt-b region. Nowadays, DNA barcoding targeting mitochondrial regions such as the COI (cytochrome oxidase I) or cyt-b genes for cichlid fish identification is well established and documented (Kullander et al., 2014; Breman et al., 2016). Molecular analyses confirmed the morphological identification of the cichlids analysed for the presence of monogenean parasites in this study. Then, we morphologically characterized the newly described species of Cichlidogyrus on the basis of the sclerotized parts of their attachment (haptor) and reproductive organs (MCO and sclerotized vagina when visible). The sclerotized structures of dactylogyridean monogeneans have been extensively investigated in various ecological and evolutionary contexts. Cichlidogyrus sp. ‘P. orthognathus 3’ Fig. 10F Cichlidogyrus sp. ‘P. orthognathus 3’ exhibits a short MCO with a reduced heel, MCO = 33 (31–34; n = 2); He = 2–3 (n = 2). The copulatory duct is thick, starting in an ovoid basal bulb, curved halfway with a big distal opening, Ct = 30 (28–31; n = 2). The accessory piece is linked to the basal bulb, thick and spirally coiled ending in a hook, Ap = 21 (20–22; n = 2). The general shape of the MCO of C. sp. ‘P. orthognathus 3’ is Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 reminiscent of C. raeymaekersi, a species described from the tropeine S. diagramma (see Van Steenberge et al., 2015). C. raeymaekersi shows a short and wide copulatory duct with a poorly developed heel, and a thick spirally coiled accessory piece attached to the basal bulb as observed in C. sp. ‘P. orthognathus 3’. However, the latter possesses a copulatory duct with an ovoid basal bulb with a distal wide part, different from the elongated bulb and the bevelled ending in C. raeymaekersi, which makes C. sp. ‘P. orthognathus 3’ distinct from C. raeymaekersi. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 DISCUSSION Several studies have reported the influence of such sclerotized structures on host specificity, parasite specialization, and reproductive isolation among congeners through niche ecology (Šimková & Morand, 2008; Vignon, Pariselle & Vanhove, 2011; Messu Mandeng et al., 2015). Because of the limitations of light microscopy as regards some morphological characters, sclerotized structures are increasingly studied Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 using enzymatic digestion followed by scanning electron microscopy. In the case of Cichlidogyrus, we can cite recent studies of Fannes, Vanhove & Huyse (2017) and Igeh, Dos Santos & Avenant-Oldewage (2017). Using this method, they redescribed three Cichlidogyrus spp. (C. tiberianus, C. dossoui, and C. philander), and provided morphological details on the haptoral and reproductive hard parts, which are not visible with light microscopy. Regarding our study, most sclerites of C. adkoningsi sp. nov. found on C. frontosa are reminiscent of species previously described from ectodine cichlids (see diagnosis). Despite these similarities, C. adkoningsi sp. nov., which represents the first record of a gill ectoparasite on a Tanganyikan cyphotilapiine, is easily distinguishable from its Tanganyikan congeners by the shape and dimensions of its sclerotized structures (such as the dorsal bar and its very characteristic long auricles, and the MCO with its irregularly shaped heel and S-shaped accessory piece ending in a hook). Similarly, C. habluetzeli sp. nov., a common species found on the same host and on the gills of C. schoutedeni, exhibits the same morphotype as species infecting ectodines, as well as other Cichlidogyrus species described from various Tanganyikan tribes. Morphologically, most of the sclerotized structures of C. habluetzeli sp. nov. are bigger than in similar species. Yet, the dorsal bar (with its small hollow outgrowths on the anterior face and its straight auricles), in addition to the straight copulatory duct, heel, and accessory piece, are structures reminiscent of those of C. aspiralis, C. pseudoaspiralis, C. centesimus (Vanhove, Volckaert & Pariselle, 2011; Rahmouni, Vanhove & Šimková, 2017), C. casuarinus (Pariselle et al., 2015a), and finally C. nshomboi (Muterezi Bukinga et al., 2012). However, some features are present in some species and missing in others. This is the case with the spirally-ornamented wall of the MCO observed exclusively in C. casuarinus, C. centesimus, and C. nshomboi, or the sclerotized vagina present in all the species listed above (including C. habluetzeli sp. nov.) except for C. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 DISCUSSION In relation to the unique morphology of the hook pairs observed in C. habluetzeli sp. nov. and its congener C. centesimus, Rahmouni et al. (2017) already suggested the existence of more haptoral groups in Cichlidogyrus than previously reported by Vignon, Pariselle & Vanhove (2011), and proposed a reinvestigation of the structural diversity of the hook pairs in Cichlidogyrus spp. and a redefinition of the ‘boundaries’ between the haptoral groups. Rahmouni, Vanhove & Šimková, 2017). Also in the parasite fauna of tropheine cichlids, morphologically similar Cichlidogyrus species infect phylogenetically related cichlid species (Vanhove, 2012; Pariselle et al., 2015b; Van Steenberge et al., 2015). As reported by Braga, Araújo & Boeger (2014), the phylogenetic relatedness of the host species has a strong influence on the distribution of monogenean parasites. On the basis of molecular data, recent work on Tanganyikan cichlids placed the Ectodini as a sister group to a clade formed by Cyphotilapiini and two other tribes (Eretmodini Poll, 1986 and Haplochromini), all mouthbrooding lineages (Meyer, Matschiner & Salzburger, 2015). This could explain the presence of C. habluetzeli sp. nov. on both C. frontosa and C. schoutedeni. At the same time, Boulengerochromini and Bathybatini are phylogenetically distant from Cyphotilapiini and Ectodini (Meyer, Matschiner & Salzburger, 2015), although they all share the morphotype of Cichlidogyrus exhibiting a spiral thickening in the wall of the copulatory duct. Possibly, living in at least partially overlapping habitat(s) could allow unrelated cichlids to host similar Cichlidogyrus species. In relation to the unique morphology of the hook pairs observed in C. habluetzeli sp. nov. and its congener C. centesimus, Rahmouni et al. (2017) already suggested the existence of more haptoral groups in Cichlidogyrus than previously reported by Vignon, Pariselle & Vanhove (2011), and proposed a reinvestigation of the structural diversity of the hook pairs in Cichlidogyrus spp. and a redefinition of the ‘boundaries’ between the haptoral groups. In addition to hook morphology, Rahmouni et al. (2017) also discussed the sclerotization of the vagina. On the basis of this study, we find that a few Cichlidogyrus spp. described from Tanganyikan hosts and possessing a sclerotized vagina belong to none of the haptoral groups defined by Vignon, Pariselle & Vanhove (2011) (who did not include species of Cichlidogyrus from LT (Pariselle & Euzet, 2009; Rahmouni, Vanhove & Šimková, 2017)). This is the case for C. habluetzeli sp. nov., C. casuarinus, C. centesimus, and C. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 DISCUSSION pseudoaspiralis (Vanhove, Volckaert & Pariselle, 2011; Muterezi Bukinga et al., 2012; Pariselle et al., 2015a; Kmentová et al., 2016b; Rahmouni et al., 2017). Regarding the haptoral configuration, C. habluetzeli sp. nov. with its hook pairs (long pairs I–IV, VI, and VII with large pair I) joins the group that harbours, so far, a single member, C. centesimus. In recent studies on the monogenean diversity of Tanganyikan cichlids, similarities in the haptoral and reproductive organs of Cichlidogyrus spp. parasitizing phylogenetically closely related cichlid hosts are far from unusual. In the present study, for instance, C. habluetzeli sp. nov. and C. koblmuelleri sp. nov. infecting C. schoutedeni are morphologically similar to some congeners of ectodine cichlids. This is also true for the undescribed species C. sp. ‘C. melanostigma’ and C. discophonum from A. dewindti sampled in north-eastern LT. In fact, Rahmouni et al. (2017) reported that two cichlid hosts Eretmodus marksmithi Burgess, 2012 and Tanganicodus irsacae Poll, 1950, both belonging to the Eretmodini, host two morphologically similar monogenean species C. jeanloujustinei and C. evikae, respectively. Moreover, monogenean communities composed of morphologically similar Cichlidogyrus species were reported in Burundese O. nasuta and its congeners from more southern localities. This observation was explained by the large distribution of ectodine hosts in the lake (Vanhove, Volckaert & Pariselle, 2011; Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Rahmouni, Vanhove & Šimková, 2017). Also in the parasite fauna of tropheine cichlids, morphologically similar Cichlidogyrus species infect phylogenetically related cichlid species (Vanhove, 2012; Pariselle et al., 2015b; Van Steenberge et al., 2015). As reported by Braga, Araújo & Boeger (2014), the phylogenetic relatedness of the host species has a strong influence on the distribution of monogenean parasites. On the basis of molecular data, recent work on Tanganyikan cichlids placed the Ectodini as a sister group to a clade formed by Cyphotilapiini and two other tribes (Eretmodini Poll, 1986 and Haplochromini), all mouthbrooding lineages (Meyer, Matschiner & Salzburger, 2015). This could explain the presence of C. habluetzeli sp. nov. on both C. frontosa and C. schoutedeni. At the same time, Boulengerochromini and Bathybatini are phylogenetically distant from Cyphotilapiini and Ectodini (Meyer, Matschiner & Salzburger, 2015), although they all share the morphotype of Cichlidogyrus exhibiting a spiral thickening in the wall of the copulatory duct. Possibly, living in at least partially overlapping habitat(s) could allow unrelated cichlids to host similar Cichlidogyrus species. DISCUSSION nshomboi (Vanhove, Volckaert & Pariselle, 2011; Muterezi Bukinga et al., 2012; Pariselle et al., 2015a). Moreover, only three Tanganyikan species of Cichlidogyrus specifically C. mbirizei from O. tanganicae (Oreochromini) (Muterezi Bukinga et al., 2012) and two of the newly described species (C. antoineparisellei sp. nov. from I. loocki (Tropheini), and C. sergemorandi sp. nov. from T. polylepis (Tylochromini)) exhibit short hook pairs I–IV, VI, and VII and a sclerotized vagina. Members of the Oreochromini and Tylochromini tribes occur in the lake but colonized the lake relatively recently (Meyer, Matschiner & Salzburger, 2015). The tribe Oreochromini was proposed only recently (Dunz & Schliewen, 2013). Therefore, formulating hypotheses on the significance of the hooks and the sclerotized vagina, seen together in Cichlidogyrus spp., is problematic. Rahmouni et al. (2017) has already underlined the necessity to investigate and analyse whether there is a correlation between the reproductive organs (the presence/ absence of a sclerotized vagina) and the haptoral sclerites (the morphology of the hook pairs). Muterezi Bukinga et al. (2012) reported the presence of Cichlidogyrus sp. ‘T. polylepis 3’, in addition to two other Cichlidogyrus species isolated from T. polylepis sampled in the central sub-basin of the lake. Because of the low number of helminth specimens collected and studied, they did not provide a formal description. On the basis of the Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 29/37 original drawing of the hard parts of Cichlidogyrus sp. ‘T. polylepis 3’, we were easily able to identify the species, complete the observations made by Muterezi Bukinga et al. (2012), and formally describe the new species herein as C. sergemorandi sp. nov. The Tropheini is one of the best studied cichlid tribes with respect to ectoparasitic monogeneans. Twelve gill species of Cichlidogyrus were previously recognized on eight cichlid species of this tribe (Gillardin et al., 2012; Pariselle et al., 2015b; Van Steenberge et al., 2015). In this study, I. loocki, P. orthognathus, and P. trewavasae sampled from the northern Congolese lakeshore revealed the presence of C. antoineparisellei sp. nov., C. masilyai sp. nov., and C. salzburgeri sp. nov. The latter two species represent the first formal descriptions of gill monogeneans on representatives of Petrochromis. I. loocki from the Zambian lakeshore was previously studied for monogenean parasites and three species, C. buescheri, C. schreyenbrichardorum, and C. vealli, were described (Pariselle et al., 2015b). DISCUSSION The new species of the present study are morphologically highly differentiated from those described by Pariselle et al. (2015b). Our record is strong new evidence of the geographical variation in parasite diversity in a particular cichlid species throughout LT. Indeed, Vanhove, Volckaert & Pariselle (2011) described four Cichlidogyrus spp. from the ectodine host O. nasuta and its two congeners O. ventralis and O. boops sampled along the more southern coastline of LT in DRC, Zambia, and Tanzania, while Rahmouni et al. (2017) reported three different species of Cichlidogyrus found on O. nasuta inhabiting the northern part of the lake. Moreover, Pariselle et al. (2015b) previously compared the three Cichlidogyrus spp. described on I. loocki to parasite specimens belonging to hitherto undescribed species retrieved from Petrochromis spp. from the collection of the MRAC. They provided micrographs of the haptoral hard parts of potentially six Cichlidogyrus spp. isolated from three cichlids (P. fasciolatus Boulenger, 1914, a species phylogenetically closely related to I. loocki; P. famula Matthes & Trewavas, 1960, an intermediate species between the Petrochromis/Interochromis clade and the ‘large’ Petrochromis clade; and finally, P. trewavasae ephippium Brichard, 1989, a member of the ‘large’ Petrochromis) (Yamaoka, 1997). On the basis of these micrographs, we were able to compare the morphology of the dorsal and ventral anchors and bars and identify differences between the newly described species on representatives of Petrochromis and the undescribed species reported by Pariselle et al. (2015b). However, due to the lack of haptoral parts in our specimens (damaged parts), and the lack of reproductive organs in the micrographs of Cichlidogyrus included in Pariselle et al. (2015b), we were not able to clearly state whether Cichlidogyrus spp. isolated from P. orthognathus represents one of the species collected from P. famula, P. fasciolatus, and P. trawavasae ephippium. In contrast, C. masilyai sp. nov. and C. salzburgeri sp. nov. isolated respectively from P. orthognathus and P. trawavasae were easily distinguishable from their congeners reported by Pariselle et al. (2015b) by their typical C-shaped dorsal bar with straight auricles and a dorsal bar with relatively wide auricles, respectively. Regarding the undescribed species of Cichlidogyrus presented herein, resampling C. melanostigma, I. loocki, P. orthognathus, and X. flavipinnis would allow future Regarding the undescribed species of Cichlidogyrus presented herein, resampling C. melanostigma, I. loocki, P. orthognathus, and X. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 INSTITUTIONAL ABBREVIATIONS INSTITUTIONAL ABBREVIATIONS MNHN Muséum National d’Histoire Naturelle, Paris, France. MZH Finish Museum of Natural History, Helsinki, Finland. MRAC Royal Museum for Central Africa, Tervuren, Belgium. MNHN Muséum National d’Histoire Naturelle, Paris, France. MZH Finish Museum of Natural History, Helsinki, Finland. MRAC Royal Museum for Central Africa, Tervuren, Belgium. DISCUSSION flavipinnis would allow future researchers to collect more parasite specimens, characterize the haptoral parts, and provide formal descriptions. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 30/37 ACKNOWLEDGEMENTS The authors would like to thank the following people and institutions for their invaluable support in the realization of this study: Walter Salzburger (University of Basel, Switzerland), Donatien Muzumani Risasi (CRH-Uvira) and Maarten Van Steenberge (MRAC/RBINS) for fish identification; Ludmila Raisingerová and Nikol Kmentová (Masaryk University, Brno, Czech Republic) and Fidel Muterezi Bukinga (Centre de Recherche en Hydrobiologie, Uvira, DRC) for helping with fish dissection, parasite isolation, and fixation; Pascal Masilya Mulungula, Vercus Lumami Kapepula, Théophile Mulimbwa N’sibula, Eugène Bahane Byaragi, Simon Kambale Mukeranya, and all the staff working in the Centre de Recherche en Hydrobiologique in Uvira for their precious help with the co-organisation of the fieldwork and for their scientific input on Tanganyikan cichlids; Katerina Francová, Evá Řehulková, Ivá Prikrylová, Carlos Mendoza-Palmero, Mária Seifertová and Jirí Vorel (Masaryk University, Brno, Czech Republic) for advice concerning taxonomy and genetic questions; and Kristína Civáňová, Eliška Jirounková, Kristýna Koukalová and Lenka Gettová (Masaryk University, Brno, Czech Republic) for their assistance in the molecular laboratory. We are also grateful to many of the above for providing specimens or hosting the authors during the fieldtrip. Special thanks go to Ondrej Hájek (Masaryk University, Brno, Czech Republic) for making the map with the sampling sites. Finally, the authors would like to thank Matthew Nicholls and Joseph Lennon for English language revision of the manuscript. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 CONCLUSION Lake Tanganyika has an impressive degree of cichlid parasite diversity with numerous species as yet undescribed. Cichlidogyrus includes (to date) 39 species from LT, including the new species described in the present study, and 79 species from other locations. Species of Cichlidogyrus infecting closely related cichlid fishes are often morphologically similar, but taxonomically distinct. Further morphological and molecular studies are necessary to elucidate the origin of the various lineages of Cichlidogyrus in LT and their evolutionary history. INSTITUTIONAL ABBREVIATIONS MNHN Muséum National d’Histoire Naturelle, Paris, France. MZH Finish Museum of Natural History, Helsinki, Finland. MRAC Royal Museum for Central Africa, Tervuren, Belgium. Competing Interests The authors declare that they have no competing interests. Author Contributions  Chahrazed Rahmouni conceived and designed the experiments, performed the experiments, analysed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, approved the final draft.  Maarten P.M. Vanhove conceived and designed the experiments, analysed the data, contributed reagents/materials/analysis tools, authored or reviewed drafts of the paper, approved the final draft, co-organised the field trip and the parasite collection.  Andrea Šimková analysed the data, contributed reagents/materials/analysis tools, authored or reviewed drafts of the paper, approved the final draft. Animal Ethics The following information was supplied relating to ethical approvals (i.e. approving body and any reference numbers): This study was approved by the Animal Care and Use Committee of the Faculty of Science, Masaryk University in Brno (Czech Republic) (approval number CZ01308). DNA Deposition The following information was supplied regarding the deposition of DNA sequences: GenBank: MH297985–MH298008. Field Study Permissions The following information was supplied relating to field study approvals (i.e. approving body and any reference numbers): Sampling was carried out under mission statements 022/MINEURS/CRH-U/2013 and 031/MINRST/CRH-U/2016 from the Centre de Recherche en Hydrobiologie-Uvira. In the absence of relevant animal welfare regulations in the D.R. Congo, the same strict codes of practice enforced within the European Union were applied. Funding This research was funded by the Czech Science Foundation project no. P505/12/G112— European Centre of Ichtyoparasitology (ECIP). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. 31/37 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Grant Disclosures The following grant information was disclosed by the authors: Czech Science Foundation project no. P505/12/G112—European Centre of Ichtyoparasitology (ECIP). Data Availability The following information was supplied regarding data availability: The type-material of the new species described in this study is housed in the invertebrate collection of the Royal Museum for Central Africa (MRAC), Tervuren, Belgium; the Finnish Museum of Natural History (MZH), Helsinki, Finland; and the Muséum National d’Histoire Naturelle (MNHN), Paris, France (see taxonomic summaries for details on repositories and accession numbers). New Species Registration The following information was supplied regarding the registration of a newly described species: The following information was supplied regarding the registration of a newly described species: Publication LSID: urn:lsid:zoobank.org:pub:7076794A-B9EB-4FFC-AC49- 66C304EC5BFB. Cichlidogyrus adkoningsi sp. nov. LSID urn:lsid:zoobank.org:act:526C2A74-E3D6-4357-B1E8-03B08B95CE38. Cichlidogyrus koblmuelleri sp. nov. LSID urn:lsid:zoobank.org:act:526C2A74-E3D6-4357-B1E8-03B08B95CE38. Cichlidogyrus koblmuelleri sp. nov. LSID urn:lsid:zoobank.org:act:473DB764-6798-43EE-8DD8-3B10D1AC1BBF. LSID urn:lsid:zoobank.org:act:473DB764-6798-43EE-8DD8-3B10D1AC1BBF. 32/37 Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 Cichlidogyrus habluetzeli sp. nov. LSID urn:lsid:zoobank.org:act:EC79CE69-2D88-4D96-A255-AC0E31B76EAE. Cichlidogyrus antoineparisellei sp. nov. LSID urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. Cichlidogyrus masilyai sp. nov. LSID urn:lsid:zoobank.org:act:4D5AF9E3-9CCE-4D3F-95B4-CCDA7B1A5C91. Cichlidogyrus salzburgeri sp. nov. LSID urn:lsid:zoobank.org:act:612A6D2B-09DD-4902-8640-A4877277E7F3. Cichlidogyrus sergemorandi sp. nov. urn:lsid:zoobank org:act:7A4653B3-2C25-4E3A-9D4A-3FA76165C167 LSID urn:lsid:zoobank.org:act:EC79CE69-2D88-4D96-A255-AC0E31B76EAE. Cichlidogyrus antoineparisellei sp. nov. LSID urn:lsid:zoobank.org:act:B324AD44-4520-4129-8DBB-8985A981783D. Cichlidogyrus masilyai sp. nov. LSID urn:lsid:zoobank.org:act:4D5AF9E3-9CCE-4D3F-95B4-CCDA7B1A5C91. Cichlidogyrus salzburgeri sp. nov. LSID urn:lsid:zoobank.org:act:612A6D2B-09DD-4902-8640-A4877277E7F3. Cichlidogyrus sergemorandi sp. nov. urn:lsid:zoobank.org:act:7A4653B3-2C25-4E3A-9D4A-3FA76165C167. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 REFERENCES Barluenga M, Meyer A. 2010. Phylogeography, colonization and population history of the Midas cichlid species complex (Amphilophus spp.) in the Nicaraguan crater lakes. BMC Evolutionary Biology 10(1):326 DOI 10.1186/1471-2148-10-326. Barluenga M, Meyer A. 2010. Phylogeography, colonization and population history of the Midas cichlid species complex (Amphilophus spp.) in the Nicaraguan crater lakes. BMC Evolutionary Biology 10(1):326 DOI 10.1186/1471-2148-10-326. Birgi E, Lambert A. 1986. Présence chez un Nandidae (Téléostéen), Polycentropsis abbreviata Boulenger, 1901, du genre Cichlidogyrus (Monogenea, Monopisthocotylea, Ancyrocephalidae). Description de Cichlidogyrus nandidae n. sp. et de Cichlidogyrus euzeti n. sp. Annales de Parasitologie Humaine et Comparée 61(5):521–528 DOI 10.1051/parasite/1986615521. Braga MP, Araújo SBL, Boeger WA. 2014. Patterns of interaction between Neotropical freshwater fishes and their gill Monogenoidea (Platyhelminthes). Parasitology Research 113(2):481–490 DOI 10.1007/s00436-013-3677-8. Breman FC, Loix S, Jordaens K, Snoeks J, Van Steenberge M. 2016. Testing the potential of DNA barcoding in vertebrate radiations: the case of the littoral cichlids (Pisces, Perciformes, Cichlidae) from Lake Tanganyika. Molecular Ecology Resources 16(6):1455–1464 DOI 10.1111/1755-0998.12523. Chakrabarty P. 2004. Cichlid biogeography: comment and review. Fish and Fisheries 5(2):97–119 DOI 10.1111/j.1467-2979.2004.00148.x. Cohen AS, Lezzar KE, Tiercelin AJJ, Soreghan M. 1997. New palaeogeographic and lake-level reconstructions of Lake Tanganyika: implications for tectonic, climatic and biological evolution in a rift lake. Basin Research 9(2):107–132 DOI 10.1046/j.1365-2117.1997.00038.x. Dossou C, Birgi E. 1984. Monogènes parasites d’Hemichromis fasciatus Peters, 1857 (Teleostei, Cichlidae). Annales des Sciences Naturelles. Zoologie et Biologie Animale 6:101–109. Douëllou L. 1993. Monogeneans of the genus Cichlidogyrus Paperna, 1960 (Dactylogyridae: Ancyrocephalinae) from cichlid fishes of Lake Kariba (Zimbabwe) with descriptions of five new species. Systematic Parasitology 25(3):159–186 DOI 10.1007/BF00007007. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 33/37 Dunz AR, Schliewen UK. 2013. Molecular phylogeny and revised classification of the haplotilapiine cichlid fishes formerly referred to as “Tilapia”. Molecular Phylogenetics and Evolution 68(1):64–80 DOI 10.1016/j.ympev.2013.03.015. Fannes W, Vanhove MPM, Huyse T. 2017. Redescription of Cichlidogyrus tiberianus Paperna, 1960 and C. dossoui Douellou, 1993 (Monogenea: Ancyrocephalidae), with special reference to the male copulatory organ. Systematic Parasitology 94(1):133–144 DOI 10.1007/s11230-016-9685-1. Froese R, Pauly D. 2017. Fishbase. World Wide Web electronic publication. Available at http://www.fishbase.org (accessed September 2017). Gillardin C, Vanhove MPM, Pariselle A, Huyse T, Volckaert FAM. 2012. Ancyrocephalidae (Monogenea) of Lake Tanganyika: II: description of the first Cichlidogyrus spp. parasites from Tropheini fish hosts (Teleostei, Cichlidae). Parasitology Research 110(1):305–313 DOI 10.1007/s00436-011-2490-5. Gussev AV. 1983. The method of collecting and processing of fish parasitic monogenean material. Nauka, Leningrad. 48 [in Russian]. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 REFERENCES Igeh PC, Dos Santos QM, Avenant-Oldewage A. 2017. Redescription of Cichlidogyrus philander (Monogenea, Ancyrocephalidae) using scanning electron microscopy (SEM) and molecular analysis. Parasite 24:49 DOI 10.1051/parasite/2017046. Jorissen MWP, Pariselle A, Huyse T, Vreven EJ, Snoeks J, Volckaert FAM, Manda AC, Kasembele GK, Artois T, Vanhove MPM. 2018. Diversity and host specificity of monogenean gill parasites (Platyhelminthes) of cichlid fishes in the Bangweulu-Mweru ecoregion. Journal of Helminthology 92(4):417–437 DOI 10.1017/S0022149X17000712. Kidd MR, Duftner N, Koblmüller S, Sturmbauer C, Hofmann HA. 2012. Repeated parallel evolution of parental care strategies within Xenotilapia, a genus of cichlid fishes from Lake Tanganyika. PLOS ONE 7(2):e31236 DOI 10.1371/journal.pone.0031236. Kmentová N, Gelnar M, Koblmüller S, Vanhove MPM. 2016a. Deep-water parasite diversity in Lake Tanganyika: description of two new monogenean species from benthopelagic cichlid fishes. Parasites & Vectors 9(1):426 DOI 10.1186/s13071-016-1696-x. Kmentová N, Gelnar M, Mendlová M, Van Steenberge M, Koblmüller S, Vanhove MPM. 2016b. Reduced host-specificity in a parasite infecting non-littoral Lake Tanganyika cichlids evidenced by intraspecific morphological and genetic diversity. Scientific Reports 6(1):39605 DOI 10.1038/srep39605. Koblmüller S, Sefc KM, Sturmbauer C. 2008. The Lake Tanganyika cichlid species assemblage: recent advances in molecular phylogenetics. Hydrobiologia 615(1):5–20 DOI 10.1007/s10750-008-9552-4. Kocher TD. 2004. Adaptive evolution and explosive speciation: the cichlid fish model. Nature Reviews Genetics 5(4):288–298 DOI 10.1038/nrg1316. Konings A. 2015. Tanganyika cichlids in their natural habitat. El Paso: Cichlid Press. Kritsky D, Thatcher DE, Boeger W. 1986. Neotropical monogenea. 8. Revision of Urocleidoides (Dactylogyridae, Ancyrocephalinae). Proceedings of the Helminthological Society of Washington 53:1–37. Kullander SO, Karlsson M, Karlsson M, Norén M. 2014. Chalinochromis cyanophleps, a new species of cichlid fish (Teleostei: Cichlidae) from Lake Tanganyika. Zootaxa 3790(3):425–438 DOI 10.11646/zootaxa.3790.3.2. Malmberg G. 1957. On the occurrence of Gyrodactylus on Swedish fishes. Skrifter utgivna av Södra Sveriges Fiskeriföreningen 1956:19–76 [in Swedish]. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 34/37 Mendlová M, Desdevises Y, Civáňová K, Pariselle A, Šimková A. 2012. Monogeneans of west African cichlid fish: evolution and cophylogenetic interactions. PLOS ONE 7(5):e37268 DOI 10.1371/journal.pone.0037268. Mendoza-Palmero CA, Blasco-Costa I, Hernandez-Mena D, Perez-Ponce De Leon G. 2017. Parasciadicleithrum octofasciatum n. gen., n. sp. (Monogenoidea: Dactylogyridae), parasite of Rocio octofasciata (Regan) (Cichlidae: Perciformes) from Mexico characterised by morphological and molecular evidence. Parasitology International 66(2):152–162 DOI 10.1016/j.parint.2017.01.006. Messu Mandeng FD, Bilong Bilong CF, Pariselle A, Vanhove MPM, Bitja Nyom AR, Agnèse JF. 2015. A phylogeny of Cichlidogyrus spp. REFERENCES (Monogenea, Dactylogyridea) clarifies a host-switch between fish families and reveals an adaptive component to attachment organ morphology of this parasite genus. Parasites & Vectors 8(1):582 DOI 10.1186/s13071-015-1181-y. Meyer BS, Matschiner M, Salzburger W. 2015. A tribal level phylogeny of Lake Tanganyika cichlid fishes based on a genomic multi-marker approach. Molecular Phylogenetics and Evolution 83:56–71 DOI 10.1016/j.ympev.2014.10.009. Mizelle J. 1936. New species of trematodes from the gills of Illinois fishes. American Midland Naturalist 17(5):785–806 DOI 10.2307/2420687. Muschick M, Indermaur A, Salzburger W. 2012. Convergent evolution within an adaptive radiation of cichlid fishes. Current Biology 22(24):2362–2368 DOI 10.1016/j.cub.2012.10.048. Musilová N, Řehulková E, Gelnar M. 2009. Dactylogyrids (Platyhelminthes: Monogenea) from the gills of Labeo (Teleostei: Cyprinidae) from West Africa. Zootaxa 2241:47–68 DOI 10.5281/zenodo.190572. Muterezi Bukinga F, Vanhove MPM, Van Steenberge M, Pariselle A. 2012. Ancyrocephalidae (Monogenea) of Lake Tanganyika: III: Cichlidogyrus infecting the world’s biggest cichlid and the non-endemic tribes Haplochromini, Oreochromini and Tylochromini (Teleostei, Cichlidae). Parasitology Research 111(5):2049–2061 DOI 10.1007/s00436-012-3052-1. NCBI Resource Coordinators. 2017. Database resources of the National Center for Biotechnology Information. Nucleic Acids Research 45(D1):D12–D17 DOI 10.1093/nar/gkw1071. Paperna I. 1960. Studies on monogenetic trematodes in Israel. 2 Monogenetic trematodes of cichlids. Bamidgeh 12:20–33. Paperna I, Thurston JP. 1969. Monogenetic Trematodes collected from cichlid fish in Uganda; including the description of five new species of Cichlidogyrus. Revue de Zoologie et de Botanique Africaines 79:15–33. Pariselle A, Bilong Bilong CF, Euzet L. 2003. Four new species of Cichlidogyrus Paperna, 1960 (Monogenea, Ancyrocephalidae), all gill parasites from African mouthbreeder tilapias of the genera Sarotherodon and Oreochromis (Pisces, Cichlidae), with a redescription of C. thurstonae Ergens, 1981. Systematic Parasitology 56(3):201–210 DOI 10.1023/B:SYPA.0000003807.27452.bd. Pariselle A, Boeger WA, Snoeks J, Bilong Bilong CF, Morand S, Vanhove MPM. 2011. The monogenean parasite fauna of cichlids: a potential tool for host biogeography. International Journal of Evolutionary Biology 2011:471–480 DOI 10.4061/2011/471480. Pariselle A, Euzet L. 1994. Three new species of Cichlidogyrus Paperna, 1960 (Monogenea: Ancyrocephalidae) parasitic on Tylochromis jentinki (Steindachner, 1895) (Pisces, Cichlidae) in West Africa. Systematic Parasitology 29(3):229–234 DOI 10.1007/BF00009678. Pariselle A, Euzet L. 1996. Cichlidogyrus Paperna, 1960 (Monogenea, Ancyrocephalidae): gill parasites from West African Cichlidae of the subgenus Coptodon Regan, 1920 (Pisces), Pariselle A, Boeger WA, Snoeks J, Bilong Bilong CF, Morand S, Vanhove MPM. 2011. The monogenean parasite fauna of cichlids: a potential tool for host biogeography. International Journal of Evolutionary Biology 2011:471–480 DOI 10.4061/2011/471480. Pariselle A, Euzet L. 1994. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 REFERENCES Three new species of Cichlidogyrus Paperna, 1960 (Monogenea: Ancyrocephalidae) parasitic on Tylochromis jentinki (Steindachner, 1895) (Pisces, Cichlidae) in West Africa. Systematic Parasitology 29(3):229–234 DOI 10.1007/BF00009678. Pariselle A, Euzet L. 1996. Cichlidogyrus Paperna, 1960 (Monogenea, Ancyrocephalidae): gill parasites from West African Cichlidae of the subgenus Coptodon Regan, 1920 (Pisces), Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 35/37 with descriptions of six new species. Systematic Parasitology 34(2):109–124 DOI 10.1007/BF00009685. Pariselle A, Euzet L. 2003. Four new species of Cichlidogyrus (Monogenea: Ancyrocephalidae), gill parasites of Tilapia cabrae (Teleostei: Cichlidae), with discussion on relative length of haptoral sclerites. Folia Parasitologica 50(3):195–201 DOI 10.14411/fp.2003.035. Pariselle A, Euzet L. 2009. Systematic revision of dactylogyridean parasites (Monogenea) from cichlid fishes in Africa, the Levant and Madagascar. Zoosystema 31(4):849–898 DOI 10.5252/z2009n4a6. Pariselle A, Bitja Nyom AR, Bilong Bilong CF. 2014. Four new species of Cichlidogyrus (Monogenea, Ancyrocephalidae) from Sarotherodon mvogoi and Tylochromis sudanensis (Teleostei, Cichlidae) in Cameroon. Zootaxa 3881(3):258–266 DOI 10.11646/zootaxa.3881.3.4. Pariselle A, Muterezi Bukinga F, Van Steenberge M, Vanhove MPM. 2015a. Ancyrocephalidae (Monogenea) of Lake Tanganyika: IV. Cichlidogyrus parasitizing species of Bathybatini (Teleostei, Cichlidae): reduced host-specificity in the deepwater realm? Hydrobiologia 748(1):99–119 DOI 10.1007/s10750-014-1975-5. Pariselle A, Van Steenberge M, Snoeks J, Volckaert F, Huyse T, Vanhove M. 2015b. Ancyrocephalidae (Monogenea) of Lake Tanganyika: does the Cichlidogyrus parasite fauna of Interochromis loocki (Teleostei, Cichlidae) reflect its host’s phylogenetic affinities? Contributions to Zoology 84:25–38. Rahmouni I, Řehulková E, Pariselle A, Berrada Rkhami O, Šimková A. 2017. Four new species of Dactylogyrus Diesing, 1850 (Monogenea: Dactylogyridae) parasitising the gills of northern Moroccan Luciobarbus Heckel (Cyprinidae): morphological and molecular characterisation. Systematic Parasitology 94(5):575–591 DOI 10.1007/s11230-017-9726-4. Rahmouni C, Vanhove MPM, Šimková A. 2017. Underexplored diversity of gill monogeneans in cichlids from Lake Tanganyika: eight new species of Cichlidogyrus Paperna, 1960 (Monogenea: Dactylogyridae) from the northern basin of the lake, with remarks on the vagina and the heel of the male copulatory organ. Parasites & Vectors 10(1):591 DOI 10.1186/s13071-017-2460-6. Řehulková E, Mendlová M, Šimková A. 2013. Two new species of Cichlidogyrus (Monogenea: Dactylogyridae) parasitizing the gills of African cichlid fishes (Perciformes) from Senegal: morphometric and molecular characterization. Parasitology Research 112(4):1399–1410 DOI 10.1007/s00436-013-3291-9. Salzburger W, Mack T, Verheyen E, Meyer A. 2005. Out of Tanganyika: genesis, explosive speciation, key-innovations and phylogeography of the haplochromine cichlid fishes. BMC Evolutionary Biology 5(1):17 DOI 10.1186/1471-2148-5-17. Šimková A, Morand S. 2008. Co-evolutionary patterns in congeneric monogeneans: a review of Dactylogyrus species and their cyprinid hosts. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 REFERENCES Journal of Fish Biology 73(9):2210–2227 DOI 10.1111/j.1095-8649.2008.02064.x. Snoeks J. 2000. How well known is the ichthyodiversity of the large East African lakes? Advances in Ecological Research 31:17–38 DOI 10.1016/S0065-2504(00)31005-4. Sturmbauer C, Hainz U, Baric S, Verheyen E, Salzburger W. 2003. Evolution of the tribe Tropheini from Lake Tanganyika: synchronized explosive speciation producing multiple evolutionary parallelism. Hydrobiologia 500(1–3):51–64 DOI 10.1023/a:1024680201436. Takahashi T, Koblmüller S. 2011. The adaptive radiation of cichlid fish in Lake Tanganyika: a morphological perspective. International Journal of Evolutionary Biology 2011:620754 DOI 10.4061/2011/620754. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 36/37 Takahashi T, Koblmüller S. 2014. A new species of Petrochromis (Perciformes: Cichlidae) from Lake Tanganyika. Ichthyological Research 61:252–264 DOI 10.1007/s10228-014-0396-9. Takahashi T, Sota T. 2016. A robust phylogeny among major lineages of the East African cichlids. Molecular Phylogenetics and Evolution 100:234–242 DOI 10.1016/j.ympev.2016.04.012. Tsuboi M, Kotrschal A, Hayward A, Buechel SD, Zidar J, Lovlie H, Kolm N. 2016. Evolution of brain-body allometry in Lake Tanganyika cichlids. Evolution: International Journal of Organic Evolution 70(7):1559–1568 DOI 10.1111/evo.12965. Vanhove MPM. 2012. Species flocks and parasite evolution towards a co-phylogenetic analysis of monogenean flatworms of cichlids and gobies. D. Phil. thesis Belgium: Leuven University. Vanhove M, Hablützel PI, Pariselle A, Šimková A, Huyse T, Raeymaekers JAM. 2016. Cichlids: a host of opportunities for evolutionary parasitology. Trends in Parasitology 32(10):820–832 DOI 10.1016/j.pt.2016.07.002. Vanhove MPM, Pariselle A, Van Steenberge M, Raeymaekers JAM, Hablützel PI, Gillardin C, Hellemans B, Breman FC, Koblmüller S, Sturmbauer C, Snoeks J, Volckaert FAM, Huyse T. 2015. Hidden biodiversity in an ancient lake: phylogenetic congruence between Lake Tanganyika tropheine cichlids and their monogenean flatworm parasites. Scientific Reports 5(1):13669 DOI 10.1038/srep13669. Vanhove MPM, Van Steenberge M, Dessein S, Volckaert FAM, Snoeks J, Huyse T, Pariselle A. 2013. Biogeographical implications of Zambezian Cichlidogyrus species (Platyhelminthes: Monogenea: Ancyrocephalidae) parasitizing Congolian cichlids. Zootaxa 3608(5):398–400 DOI 10.11646/zootaxa.3608.5.8. Vanhove MPM, Volckaert FAM, Pariselle A. 2011. Ancyrocephalidae (Monogenea) of Lake Tanganyika: I: four new species of Cichlidogyrus from Ophthalmotilapia ventralis (Teleostei: Cichlidae), the first record of this parasite family in the basin. Zoologia (Curitiba, Impresso) 28(2):253–263 DOI 10.1590/S1984-46702011000200016. Van Steenberge M, Pariselle A, Huyse T, Volckaert F, Snoeks J, Vanhove MPM. 2015. Morphology, molecules, and monogenean parasites: an example of an integrative approach to cichlid biodiversity. PLOS ONE 10(4):e0124474 DOI 10.1371/journal.pone.0124474. Vignon M, Pariselle A, Vanhove MPM. 2011. Modularity in attachment organs of African Cichlidogyrus (Platyhelminthes: Monogenea: Ancyrocephalidae) reflects phylogeny rather than host specificity or geographic distribution. Rahmouni et al. (2018), PeerJ, DOI 10.7717/peerj.5604 REFERENCES Biological Journal of the Linnean Society 102(3):694–706 DOI 10.1111/j.1095-8312.2010.01607.x. Wanek KA, Sturmbauer C. 2015. Form, function and phylogeny: comparative morphometrics of Lake Tanganyika’s cichlid tribe Tropheini. Zoologica Scripta 44(4):362–373 DOI 10.1111/zsc.12110. Yamaoka K. 1997. Trophic ecomorphology of Tanganyikan cichlids. In: Kawanabe H, Hori M, Nagoshi M, eds. Fish communities in Lake Tanganyika. Kyoto: Kyoto University Press, 25–56. 37/37
https://openalex.org/W3039894784
https://www.lenus.ie/bitstream/10147/630437/1/s41598-020-67701-3.pdf
English
null
HAPPENN is a novel tool for hemolytic activity prediction for therapeutic peptides which employs neural networks
Scientific reports
2,020
cc-by
15,474
Find this and similar works at - http://www.lenus.ie/hse HAPPENN is a novel tool for hemolytic activity prediction for therapeutic peptides which employs neural networks. Item Type Article;Other Authors Timmons, Patrick Brendan;Hewage, Chandralal M DOI 10.1038/s41598-020-67701-3 Journal Scientific reports Download date 24/10/2024 05:56:06 Link to Item http://hdl.handle.net/10147/630437 Find this and similar works at - http://www.lenus.ie/hse Find this and similar works at - http://www.lenus.ie/hse www.nature.com/scientificreports www.nature.com/scientificreports HAPPENN is a novel tool for hemolytic activity prediction for therapeutic peptides which employs neural networks Patrick Brendan Timmons & Chandralal M. Hewage* The growing prevalence of resistance to antibiotics motivates the search for new antibacterial agents. Antimicrobial peptides are a diverse class of well-studied membrane-active peptides which function as part of the innate host defence system, and form a promising avenue in antibiotic drug research. Some antimicrobial peptides exhibit toxicity against eukaryotic membranes, typically characterised by hemolytic activity assays, but currently, the understanding of what differentiates hemolytic and non- hemolytic peptides is limited. This study leverages advances in machine learning research to produce a novel artificial neural network classifier for the prediction of hemolytic activity from a peptide’s primary sequence. The classifier achieves best-in-class performance, with cross-validated accuracy of 85.7% and Matthews correlation coefficient of 0.71. This innovative classifier is available as a web server at https​://resea​rch.timmo​ns.eu/happe​nn, allowing the research community to utilise it for in silico screening of peptide drug candidates for high therapeutic efficacies. All living organisms exist in an environment teeming with harmful microbes, which they can become exposed to through contact, ingestion or ­inhalation1. An important part of the host defence mechanisms that protect against these pathogens are antimicrobial peptides (AMPs).h All living organisms exist in an environment teeming with harmful microbes, which they can become exposed to through contact, ingestion or ­inhalation1. An important part of the host defence mechanisms that protect against these pathogens are antimicrobial peptides (AMPs).h g p g p p The serious issue of pathogen resistance to multiple antibiotics is the motivation for the search of novel drugs that can be used without the development of resistance. AMPs are a class of compounds that are promising as novel antibiotics, due to good selectivity and only a limited number of cases of resistance, attributed to their relatively non-specific mechanism of ­action2,3.hhfi y pi Therapeutic peptides possess many advantages over traditional drugs. They are more efficacious, selective and specific than small molecules, and their products of degradation are amino acids, which present a reduced risk of drug-drug interactions. Additionally, their short half-life means a lower propensity for accumulation in ­tissues4. Their immediate response and potent activity against various pathogens, including bacteria, fungi, parasites and viruses, means that these compounds can be utilised both as substitutes and as part of a combination therapy with conventional ­antibiotics5. Furthermore, therapeutic peptides have been identified for other applications, such as cancer, immune disorders, cardiovascular diseases, gastrointestinal dysfunction, hemostasis and ­diabetes6–9. UCD School of Biomolecular and Biomedical Science, UCD Centre for Synthesis and Chemical Biology, UCD Conway Institute, University College Dublin, Dublin 4, Ireland. *email: chandralal.hewage@ucd.ie HAPPENN is a novel tool for hemolytic activity prediction for therapeutic peptides which employs neural networks Although therapeutic peptides were initially isolated from plants or animals that secrete them as part of their host defence ­mechanism10, they can now also be obtained from ­genetic11, ­recombinant12 and ­chemical13 libraries as well, which presents a largely unexplored chemical space, with only a limited number of peptide-based drugs currently available on the market. Among those are Enfuvirtide, Leuprolide, Bacitracin and Boceprevir, which act against ­HIV14, prostate ­cancer15, ­pneumonia16 and hepatitis-C17, respectively. g p p p p y Many peptides never reach clinical trials because of a therapeutic potential that’s hindered by low metabolic stability, poor oral bioavailability, or a poor toxicity profile, which is typically assessed by measuring the activ- ity that the peptide exerts against eukaryotic ­erythrocytes18. Peptide modifications such as substitution with D-amino acids have been proved to improve peptide stability to ­proteolysis19. Toxicity can be divided into three classes: immunotoxicity, cytotoxicity and hemotoxicity. It is the aim of this work to develop a method of predict- ing peptides’ hemotoxicity prior to their chemical synthesis. Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Figure 1. Schematic representation of the interactions between the peptides (red) and the lipid bilayer (grey) during (A) initial approach and binding, (B) carpet model, (C) barrel stave model and (D) toroidal pore model. Figure 1. Schematic representation of the interactions between the peptides (red) and the lipid bilayer (grey) during (A) initial approach and binding, (B) carpet model, (C) barrel stave model and (D) toroidal pore model. AMPs typically, but not exclusively, possess amphipathic structures with a positively charged and a hydro- phobic interface, and primarily exert their activity at the charged surface of the bacterial plasma membrane. A number of mechanisms of action have been characterised to date; the most commonly observed mechanisms of action are the pore-forming barrel-stave and toroidal-pore models, and the non-pore-forming carpet model, which are illustrated in Fig. 120. All three mechanisms share the development of an electrostatic attraction between the negatively charged membrane phospholipid headgroups and the positive charge of ­AMPs21 as their initial step. Once the AMP is in close proximity to the membrane, it establishes hydrophobic interactions between its hydrophobic face and the membrane’s hydrophobic interior. Interestingly, anionic AMPs have been identi- fied despite the general requirement for cationicity; these peptides instead possess an increased hydrophobicity ­profile22. Methods Datasets. Datasets. The HAPPENN dataset consists of 3,738 peptides sequences between 7–35 amino acids in length and their corresponding hemolytic activities. All sequences are composed of exclusively natural amino acids, and the only modifications included are N-terminal acetylation and C-terminal amidation. Secondary structure properties are not considered in the creation of the dataset. The dataset is available as supplementary material. 3,408 of these peptide sequences were extracted from the DBAASP ­database28 and 1,174 from the Hemolytik ­database30. 844 peptide sequences were present in both databases. Of the sequences extracted from DBAASP, Datasets. The HAPPENN dataset consists of 3,738 peptides sequences between 7–35 amino acids in length and their corresponding hemolytic activities. All sequences are composed of exclusively natural amino acids, and the only modifications included are N-terminal acetylation and C-terminal amidation. Secondary structure properties are not considered in the creation of the dataset. The dataset is available as supplementary material. p g y q p y , and the only modifications included are N-terminal acetylation and C-terminal amidation. Secondary structure properties are not considered in the creation of the dataset. The dataset is available as supplementary material. 3,408 of these peptide sequences were extracted from the DBAASP ­database28 and 1,174 from the Hemolytik ­database30. 844 peptide sequences were present in both databases. Of the sequences extracted from DBAASP, 861 were ribosomally synthesised, while 2,547 were chemically synthesised.h p ph pp y 3,408 of these peptide sequences were extracted from the DBAASP ­database28 and 1,174 from the Hemolytik ­database30. 844 peptide sequences were present in both databases. Of the sequences extracted from DBAASP, 861 were ribosomally synthesised, while 2,547 were chemically synthesised.h y y y y The dataset consists of 1,543 experimentally validated hemolytic peptides and 2,195 experimentally validated on-hemolytic peptides, as determined by criteria detailed in Table 1. Redundancy reduced dataset. The HAPPENN dataset was internally redundancy reduced using CD-HIT31–33, removing sequences so that no two sequences were ≥90% similar to each other, which yielded the HAPPENN- RR90 dataset, which consists of 823 experimentally validated hemolytic peptides, and 1,100 experimentally vali- dated non-hemolytic peptides. Dataset for additional benchmarking. Discriminating compositionally similar peptides with different biologi- cal activities is one of the greatest challenges in developing prediction ­methods34,35. HAPPENN is a novel tool for hemolytic activity prediction for therapeutic peptides which employs neural networks Barrel-stave model AMPs form a barrel shape in the cell membrane in which hydrophobic residues are juxtaposed with lipid chains and hydrophobic residues which line the central pore. AMPs which employ the toroidal pore model of action coerce the membrane lipids to bend in a manner such that the AMPs are positioned nearest to the phospholipid head groups which shape the pore diameter. The carpet model, meanwhile, requires for membrane thinning to occur between the anionic membrane phospholipids and the cationic AMPs. Following insertion, AMPs induce a change in the cell membrane polarization, which results in an increase in membrane permeability and a reduction of the transmembrane electrical ­potential23.hf p y p The basis for selectivity of AMPs lies in the different composition of the prokaryotic and eukaryotic cell membranes. Bacterial cell membranes are rich in phospholipids, including phosphatidylglycerol, phosphatidyl- serine, phosphoglycerol, gangliosides, phosphatidylcholine, ­sphingomyelin24, but do not contain cholesterol, like eukaryotic membranes. Classification of peptides as hemolytic or non-hemolytic is complicated as most antimicrobial peptides exert their activity at the plasma membrane. The differentiator between hemolytic and non-hemolytic peptides is whether or not they are active at the zwitterionic eukaryotic membrane as well as the anionic prokaryotic membrane.fi p y In silico methods provide a more efficient avenue for the screening of the large chemical space and accelerate drug design by reducing the number of peptides that have their hemolytic activity screened experimentally. Deep learning has proven itself useful in other areas of bioinformatics, with numerous examples such as DeepPPISP for the prediction of protein-protein interaction ­sites25, SCLpred for protein subcellular localization ­prediction26 and CPPpred for prediction of cell-penetrating ­peptides27. As a number of databases exist that detail the biological activities of peptides, such as ­DBAASP28, ­CAMP29 and ­Hemolytik30, we have exploited the available data to train a deep neural network that classifies peptides as hemolytic or non-hemolytic based on their primary sequence. Herein we present a novel method for the prediction of the hemolytic activity of antimicrobial peptides. Methods Datasets. An additional dataset was created, HAPPENN-hard, wherein positive examples are the experimentally validated hemolytic peptides of HAPPENN-RR90, and the negative examples are experimentally validated non-hemolytic peptides exhibiting the greatest compositional similarity to the positive peptides. A negative sequence was deemed to be the most similar to a positive sequence if it possessed the minimum Euclidean distance to the positive ­sequence36–38. Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Table 1. Criteria for designating a peptide as hemolytic or non-hemolytic. Peptides which satisfied neither or both of these criteria were excluded. Hemolytic peptides Non-hemolytic peptides Hemolytic activity (%) Peptide conc. (μM) Hemolytic activity (%) Peptide conc. (μM) 50 ≤300 45 > 270 55 ≤330 40 > 240 60 ≤360 35 > 210 65 ≤390 30 > 180 70 ≤420 25 > 150 75 ≤450 20 > 120 80 ≤480 15 > 90 85 ≤510 10 > 60 90 ≤540 5 > 30 95 ≤570 0 > 30 100 ≤600 Table 1. Criteria for designating a peptide as hemolytic or non-hemolytic. Peptides which satisfied neither or both of these criteria were excluded. Model validation. It is critical that any classifier model created by machine learning is thoroughly vali- dated. For that reason, tenfold cross-validations and validation by an external test set were employed to evaluate the performance of all models presented herein. The HAPPENN dataset was split into twelve parts, ten of which were used for cross-validation, whereby one of the subsets was selected for use in validation while the other nine were employed for training. The resultant models were ensembled and evaluated with an independent test set, which consists of the remaining two of the twelve parts. To avoid possible bias arising from the choice of a randomly selected test set, the procedure is repeated six times in total, allowing for a rigorous assessment of the model’s overall performance. Validation comparison with HemoPI and HemoPred. The different dataset construction and validation pro- cedure employed by HAPPENN compared to other available tools, namely HemoPI and HemoPred, prevents a direct comparison of their respective validation statistics. To facilitate a more direct comparison with the HemoPI and HemoPred classifiers, a model was trained and tested under equivalent conditions. Neural network architecture.  All input features are scaled to have minimum and maximum values and 1, respectively. Both a randomized grid search and genetic algorithm were employed to identify the optimal neural network architecture and hyperparameters. The optimized neural network applies a Gaussian noise layer with a standard deviation of 0.03 to the input layers, which mitigates overfitting. The first hidden layer has 1024 nodes and the second hidden layer has 64 nodes. Batch ­normalization118 is applied before the ReLU activation function. Each hidden layer is followed by a Dropout layer, with a rate of 0.93, which aids in the prevention of ­overfitting119. y y p y pi g The final output layer consisted of a single node with a sigmoid activation function. A summary of the overall architecture described is shown in Fig. 2. mplementation. The neural network was implemented with Keras, a popular deep learning framework sing a ­Tensorflow120 back-end. The binary cross-entropy loss function was employed, which is defined as: (1) −1 N N  i=1 [yi log( ˆyi) + (1 −yi) log(1 −ˆyi)] (1) (1) whereby yi is the true value of the ith sample, and ˆyi is the predicted value of the ith sample.hi whereby yi is the true value of the ith sample, and ˆyi is the predicted value of the ith sample. Thi l f ti i l d i bi l ifi ti bl A th di t d l This loss function is commonly used in binary classification problems. As the predicted labels of all training data approach their respective true values, the value of the function approaches zero.h pp p pp The optimizer employed is Adaptive Momentum (Adam), which updates the neural network weights accord- ing to the following ­formula121: (2) t+1 = t − η ˆmt √ ˆvt + ǫ (2) whereby the ˆmt and ˆvt are the bias-corrected estimates of the mean and the variance of the gradients, respectively. The neural network was trained for 600 epochs, without stopping criteria. The model with the highest valida- tion accuracy encountered during training was saved for each of the cross-validation splits. whereby the ˆmt and ˆvt are the bias-corrected estimates of the mean and the variance of the gradients, respectively. The neural network was trained for 600 epochs, without stopping criteria. The model with the highest valida- tion accuracy encountered during training was saved for each of the cross-validation splits. www.nature.com/scientificreports/ ­hydrophobicities72–76, hydrophobic ­moments77, partition ­energies78–80 and retention coefficients in ­HPLC81,82. Sim- ilarly, the sequence hydrophilicity was characterised using properties based on the amino acids’ ­hydrophilicity50, ­charges83, ­polarities84,85, free energies of solution in ­water77,86, numbers of hydrogen bond ­donors87 and fractions of site occupied by ­water88. Descriptors relating to the amino acids’ sterics were calculated based on their resi- due ­volume89–92, residue ­flexibility54, steric ­hindrance93, ­bulkiness52, 8Å and 14Å contact ­numbers94,95, average reduced side-chain ­distance96 and accessible molar fractions ­ratio97 properties. As membrane interaction plays an important role in peptides’ hemolysis mechanism of action, features based on membrane-propensities98, membrane-buried preference ­parameters47,99 and the side-chain100,101 and electron-ion ­interactions102,103 were calculated. Descriptors were also calculated based on the number of full non-bonding ­orbitals87, SWIGM ­index51, and the ­IFH104 and ­z1105 scales. Composition descriptors. Amino acid, dipeptide, and tripeptide compositions were calculated for the conven- tional 20-amino acid alphabet, as well as the reduced alphabets of Veltri et al.106, Thomas and ­Dill107, and the conjoint ­alphabet108. To account for the three-dimensional structure of the peptides, g-gap dipeptide and trip- eptide compositions were ­calculated109. Finally, pseudo amino acid ­composition110, conjoint triad, composition, transition and ­distribution111 descriptors were also calculated. Machine learning approaches. Support vector machine (SVM)112, random forest (RF)113, principal com- ponent analysis (PCA)114, t-distributed Stochastic Neighbour Embedding (t-SNE)115 and dense fully connected neural ­networks116 are employed in this study. p y y Both a linear and non-linear (RBF) kernel were employed with SVMs. SVM and RF hyperparameters were tuned using a grid search in conjunction with the previously described cross-validation. Feature selection. Only features which were non-zero for at least 100 samples were retained. Furthermore, features were selected for retention by SVM and random forest. y Features importances were calculated individually for each of the splits during tenfold cross-validation using both support vector machines and random forests. Features which had SVM absolute weights near-zero ( < 0.05 ) were excluded, as practised by Brank et al.117. Features which an ensemble of random forests decided were important (importance > 0.0005 ) were included. Methods Datasets. An altered dataset, HAPPENN-HemoPI3-equiv was created, wherein all the peptide sequences present in the HAPPENN dataset that form part of the HemoPI-3 test dataset were set aside as the test dataset, and the remaining non-test set sequences were used for training and validation as part of a fivefold cross-validation. Amino acid composition analysis. An analysis of the amino acid composition of the hemolytic and non- hemolytic peptides was carried out, completed by an analysis of peptides randomly extracted from proteins in Swiss-Prot39. The analysis comprises the peptides’ full sequences, the 10 N-terminal residues, and the C-terminal 10 residues. Residue position preference analysis. Enrichment depletion logos (EDLogo)40 were created to identify preferences for certain amino acid residues at certain positions in the hemolytic peptides’ sequences. The logo plots were constructed using the experimentally validated non-hemolytic peptide sequences as the baseline. Motif analysis. Motif analysis was carried out on the HAPPENN dataset to identify motifs occurring exclu- sively in hemolytic and non-hemolytic peptides. Motifs with a length between 2–5 amino acids which occurred in at least ten peptides were considered. Features extraction. A large selection of features was extracted from the peptides’ primary sequences, which can be divided into two subcategories, amino acid composition based features and physicochemical descriptors. Physicochemical descriptors. The ­modlAMP41, ­ChemoPy42 and RDKit packages were used for the calculation of global physicochemical descriptors, as well as amino acid scale-based descriptors. g p y p p Global physicochemical descriptors include sequence length, molecular formula, molecular weight, sequence charge, charge density, isoelectric point, instability index, aromaticity ­index43, aliphatic ­index44, Boman ­index45 and the hydrophobic ratio. y Meanwhile, amino acid scale-based descriptors include ­AASI46, ­ABHPRK41, ­hydrophobicity47–51, side- chain ­bulkiness52, amino acid charges, COUGAR​41, ­Ez53, side-chain ­flexibility54, ­polarity52,55, ­ISAECI56, α-helix ­propensity57, ­MSS58, ­MSW59, ­pepArc41, ­PPCALI60, ­refractivity61, ­t_scale62, transmembrane ­propensity63, ­z364 and ­z565. Additionally, physicochemical descriptors were calculated from the amino acid properties in the ­AAindex66. Secondary structure related descriptors were calculated based on the ­turn67, ­helical47, 68, ­coil69 and ­amphiphilic70 propensities. The sequence hydrophobicity was quantified using the amino acids’ ­hydropathies49,71, Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Neural network architecture.  All input features are scaled to have minimum and maximum values and 1, respectively. y g g p During training, the loss function was weighted to adjust for the slightly unequal number of positive and negative samples. Performance evaluation. The robustness of the predictor is evaluated by a number of standard param- eters, namely accuracy (Acc), sensitivity (Sn), specificity (Sp), the Matthews correlation coefficient (MCC), and by the receiver operating characteristic (ROC) curve.hii y p g The first four of these are defined by the following equations: y p g The first four of these are defined by the following equations: Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Figure 2. Summary of the model development architecture. Peptide sequences and their corresponding activities were extracted from databases, peptides outside the experiment’s scope were removed, and descriptors were calculated. The peptides’ descriptors are used as training input to a neural network with two hidden layers, which then predicts whether or not the peptide possesses hemolytic activity. Figure 2. Summary of the model development a activities were extracted from databases, peptides were calculated. The peptides’ descriptors are used which then predicts whether or not the peptide p Figure 2. Summary of the model development architecture. Peptide sequences and their corresponding activities were extracted from databases, peptides outside the experiment’s scope were removed, and descriptors were calculated. The peptides’ descriptors are used as training input to a neural network with two hidden layers, which then predicts whether or not the peptide possesses hemolytic activity. Figure 2. Summary of the model development architecture. Peptide sequences and their corresponding activities were extracted from databases, peptides outside the experiment’s scope were removed, and descriptors were calculated. The peptides’ descriptors are used as training input to a neural network with two hidden layers, which then predicts whether or not the peptide possesses hemolytic activity. Figure 2. Summary of the model development architecture. Peptide sequences and their corresponding activities were extracted from databases, peptides outside the experiment’s scope were removed, and descriptors were calculated. The peptides’ descriptors are used as training input to a neural network with two hidden layers, which then predicts whether or not the peptide possesses hemolytic activity. Resultsh The HAPPENN dataset was constructed from peptide sequences whose hemolytic activity, or lack thereof, has previously been evaluated. Peptides were separated into a positive (hemolytic) class and a negative (non- hemolytic) class based on criteria outlined in Table 1. The peptide sequences were subjected to an amino acid composition analysis, residue position preference analysis and motif analysis. Peptides were then represented by feature vectors composed of physicochemical and compositional descriptors of the peptides. The feature vectors are visualised using principal component analysis (PCA) and t-stochastic neighbour embedding (t-SNE) plots, both of which show an incomplete separation of the positive and negative classes. Finally, the feature vectors are used to train support vector machine, random forest and neural network hemolytic activity classifiers, the prediction results of which are evaluated. Amino acid composition analysis. To determine whether a preference exists for certain residues in hemolytic peptides compared to non-hemolytic peptides, an amino acid residue composition analysis was per- formed, the results of which are shown in Fig. 3. It is apparent that hemolytic peptides are most enriched in the hydrophobic leucine and isoleucine residues, and to a lesser extent phenylalanine, tryptophan and glycine. Meanwhile, non-hemolytic peptides are enriched in the positively charged lysine and arginine residues. Inter- estingly, both hemolytic and non-hemolytic peptides are depleted in the negatively charged aspartic and glu- tamic acid residues compared to the sequences randomly extracted from Swiss-Prot, with the hemolytic peptides exhibiting greater depletion. Residue position preference analysis. To ascertain whether or not there exists a preference for certain residues at certain positions in the peptide sequence, an enrichment-depletion logo plot was created (Fig. 4) for the hemolytic peptide class, with the non-hemolytic peptide class serving as the baseline for the plot. Enriched residues, therefore, are those which are more common at that position in hemolytic peptides, relative to the non- hemolytic class, and depleted residues are those which are less common.hi The first inspection of the EDlogo plot reveals information that is consistent with the amino acid composition analysis: hemolytic peptides are enriched in hydrophobic residues, and predominantly depleted in negatively charged residues. On further inspection, position-specific enrichments become apparent. Hemolytic peptides are enriched in the negatively charged aspartic acid residue at position 4, and at the last, third- and fourth- and eleventh-last position, despite being depleted in this residue for the remainder of the sequence. Resultsh Hemolytic pep- tides are depleted in the positively charged arginine residue throughout the sequence, but enriched in lysine at positions 7, 8, 11, 12 and 15. A preference exists at the positions 2 and 3 for tryptophan, position 3 for proline, and positions 3 and 4 for serine. A notable preference exists at position 14 for proline, which is indeed a com- mon feature in the brevinin-1 family of peptides, which do possess hemolytic ­activity122. Hemolytic peptides are also enriched in glutamine exclusively at their C-terminus, while being depleted in glutamine throughout the remainder of the sequence. Motif analysis. A motif analysis was undertaken on the HAPPENN dataset to identify any motifs present exclusively in hemolytic or non-hemolytic peptides. The top twenty motifs occurring exclusively in hemolytic peptides are ’LKHI’, ’KIIKV’, ’TLLKK’, ’VNWK’, ’GAIA’, ’VNWKK’, ’KKILG’, ’VLKAA’, ’LWKT’, ’ALWKT’, ’MAL’, ’KITK’, ’PKIF’, ’GKEV’, ’KIAS’, ’CKITK’, ’KHILK’, ’IKVV’, ’IKVA’, ’IASI’. The top twenty motifs occurring exclu- sively in non-hemolytic peptides are ’PRP’, ’RPRP’, ’AAAA’, ’PRPR’, ’PRPRP’, ’RPRPR’, ’AAAAA’, ’AFA’, ’AAFA’, ’AFAA’, ’LKYG’, ’WKI’, ’KYGK’, ’ILKYG’, ’LKYGK’, ’PRL’, ’RRKK’, ’AAFAA’, ’KPS’, ’RPG’. Data visualisation. Principal component analysis (PCA). Principal component analysis (PCA) was un- dertaken for the full computed dataset, the dataset with only the physicochemical features and the dataset with only composition descriptors (Fig. 5). Inspection of the results of all three indicates that while a separation exists between the hemolytic and non-hemolytic classes, the separation is not clear-cut and a significant overlap exists between the classes. The overlap between classes is most significant for the set consisting of only the physico- chemical descriptors, while a greater separation between classes exists in the composition descriptor plot. T‑distributed stochastic neighbour embedding (t‑SNE). Similarly to the aforementioned PCA analysis, a t-dis- tributed Stochastic Neighbour Embedding (t-SNE) analysis was undertaken for the full computed dataset, the dataset with only the physiochemical features and the dataset with only composition descriptors (Fig. 6). As is the case with the PCA results, there exists an incomplete separation between the hemolytic and non-hemolytic classes in all three datasets. In many cases, positive and negatives peptides are near-coincident in the plots, and appear, therefore, to be physicochemically and compositionally similar. Hemolytic activity prediction. This novel study employed a number of popular machine learning classi- fiers for predicting peptides’ hemolytic activity on the basis of features calculated from their primary sequence. Neural network architecture.  All input features are scaled to have minimum and maximum values and 1, respectively. whereby (3) Acc = TP + TN TP + TN + FP + FN × 100 (4) Sn = TP TP + FN × 100 (5) Sp = TN TN + FP × 100 (6) MCC = TP × TN −FP × FN √(TP + FP)(TP + FN)(TN + FP)(TN + FN) (3) Acc = TP + TN TP + TN + FP + FN × 100 (4) Sn = TP TP + FN × 100 (5) Sp = TN TN + FP × 100 (3) (6) MCC = TP × TN −FP × FN √(TP + FP)(TP + FN)(TN + FP)(TN + FN) (6) whereby whereby whereby https://doi.org/10.1038/s41598-020-67701-3 https://doi.org/10.1038/s41598-020-67701-3 Scientific Reports | (2020) 10:10869 | www.nature.com/scientificreports/ • TP = True positives: the number of correctly predicted positive (hemolytic) peptides. • TP = True positives: the number of correctly predicted positive (hemolytic) peptides. • FP = False positives: the number of non-hemolytic peptides incorrectly predicted as being hemo • TN = True negatives: the number of correctly predicted negative (non hemolytic) peptides p y p p ( y ) p p • FP = False positives: the number of non-hemolytic peptides incorrectly predicted as bein TN T i h b f l di d i ( h l i ) id • FP = False positives: the number of non-hemolytic peptides incorrectly predicted as being y y g y • TN = True negatives: the number of correctly predicted negative (non-hemolytic) peptides. l h b f h l d l d d b h l y y • TN = True negatives: the number of correctly predicted negative (non-hemolytic) pe FN = False negatives: the number of hemolytic peptides incorrectly predicted as being non-hemolytic. Resultsh The predictive power was evaluated using tenfold cross-validation, and the final ensemble of ten neural networks was evaluated by means of external validation. Accuracy, sensitivity, specificity, Matthews correlation coeffi- Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ entificreports/ Figure 3. Percentage average amino acid residue composition of the (A) full sequences, (B) N-terminal 10 residues, and (C) C-terminal 10 residues of experimentally validated hemolytic peptides (orange), experimentally validated non-hemolytic peptides (blue) and peptide sequences randomly extracted from Swiss- Prot proteins (green). Figure 3. Percentage average amino acid residue composition of the (A) full sequences, (B) N-terminal 10 residues, and (C) C-terminal 10 residues of experimentally validated hemolytic peptides (orange), experimentally validated non-hemolytic peptides (blue) and peptide sequences randomly extracted from Swiss- Prot proteins (green). cient statistical parameters are reported with their confidence intervals. A receiver operating characteristic curve (ROC) with a calculated area under the curve (AUC) is also reported. To the authors’ knowledge, three machine-learning based classifiers for the prediction of hemolytic activity peptides are described in the literature, namely ­HemoPI123, ­HemoPred124, and ­HemoPImod125. The former two predict the hemolytic activity of natural amino-acid-based peptides, while the latter specializes in predicting the hemolytic potency of chemically modified peptides. The results of the present study are compared to those of the former two classifiers, HemoPI and HemoPred. As HemoPImod specifically addresses chemically modified peptides, and therefore differs in its aims to HAPPENN, it is excluded from the comparisons. cient statistical parameters are reported with their confidence intervals. A receiver operating characteristic curve (ROC) with a calculated area under the curve (AUC) is also reported.i To the authors’ knowledge, three machine-learning based classifiers for the prediction of hemolytic activity peptides are described in the literature, namely ­HemoPI123, ­HemoPred124, and ­HemoPImod125. The former two predict the hemolytic activity of natural amino-acid-based peptides, while the latter specializes in predicting the hemolytic potency of chemically modified peptides. The results of the present study are compared to those of the former two classifiers, HemoPI and HemoPred. As HemoPImod specifically addresses chemically modified peptides, and therefore differs in its aims to HAPPENN, it is excluded from the comparisons. Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Comparison of classifier methods. The prediction statistics achieved by support vector machine (SVM) dom forest (RF) and neural network (NN) models are presented in Table 2 Figure 4. Resultsh Enrichment-depletion logo plot of (A) N-terminal 15 residues and (B) C-terminal 15 residues of experimentally validated hemolytic peptides of the HAPPENN dataset. Data is scaled to account for the background probability of each amino acid, based on the experimentally validated non-hemolytic peptides Figure 4. Enrichment-depletion logo plot of (A) N-terminal 15 residues and (B) C-terminal 15 residues of experimentally validated hemolytic peptides of the HAPPENN dataset. Data is scaled to account for the background probability of each amino acid, based on the experimentally validated non-hemolytic peptides. Comparison of classifier methods. The prediction statistics achieved by support vector machine (SVM), ran- dom forest (RF) and neural network (NN) models are presented in Table 2.hh The SVM hyperparameters were optimised using a grid search. The linear kernel SVM achieved its highest performance with the regularization parameter C = 0.1 . The non-linear RBF kernel SVM achieved its highest performance with the regularization parameter C = 10 and the kernel coefficient γ = 2 × 10−4 . Both the RBF and linear kernel SVM approaches achieve the worst level of performance of the three methods studied, with a validation accuracies of 78% and 81%, and MCCs of 0.54 and 0.61, respectively.hh p y The RF hyperparameters were also optimised using a grid search. The highest performance, with an accuracy and MCC of 83% and 0.65 was achieved with the number of estimators set to be 1024, with unrestricted tree depth. The optimal value for max_features was found to be 70.h ph p _ The neural network approach, meanwhile, achieves the highest accuracy and MCC score, with scores of 86% and 0.71, respectively, marking it as the most capable predictor. Furthermore, the neural network approach Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Figure 5. Principal component analysis of (A) all the computed descriptors, (B) only the physicochemical descriptors and (C) composition descriptors. Hemolytic peptides (positives) are coloured red, non-hemolytic peptides (negatives) are coloured blue, false-positives are coloured black, false-negatives are coloured orange. Figure 5. Principal component analysis of (A) all the computed descriptors, (B) only the physicochemical descriptors and (C) composition descriptors. Hemolytic peptides (positives) are coloured red, non-hemolyti peptides (negatives) are coloured blue, false-positives are coloured black, false-negatives are coloured orange Figure 6. t-SNE visualisation of (A) all the computed descriptors, (B) only the physicochemical descriptors and (C) composition descriptors. Resultsh Hemolytic peptides (positives) are coloured red, non-hemolytic peptides (negatives) are coloured blue, false-positives are coloured black, false-negatives are coloured orange. Figure 6. t-SNE visualisation of (A) all the computed descriptors, (B) only the physicochemical descriptors and (C) composition descriptors. Hemolytic peptides (positives) are coloured red, non-hemolytic peptides (negatives) are coloured blue, false-positives are coloured black, false-negatives are coloured orange. Table 2. Validation and test results for the SVM, RF and NN models trained on the HAPPENN dataset. Method Acc Sn Sp MCC Cross-validation SVM (linear) 81.22 ± 2.58 76.87 ± 4.06 84.26 ± 3.13 0.61 ± 0.05 SVM (RBF) 77.51 ± 2.91 71.26 ± 4.77 81.94 ± 3.44 0.54 ± 0.06 RF 83.30 ± 2.35 77.68 ± 4.10 87.26 ± 2.39 0.65 ± 0.05 NN 85.66 ± 1.93 84.96 ± 3.37 86.09 ± 3.43 0.71 ± 0.04 External validation SVM (linear) 81.49 ± 1.80 76.77 ± 2.46 84.85 ± 2.13 0.62 ± 0.04 SVM (RBF) 77.79 ± 2.11 71.33 ± 2.78 82.37 ± 2.97 0.54 ± 0.04 RF 84.06 ± 1.38 78.56 ± 2.69 87.96 ± 1.75 0.67 ± 0.03 NN 84.00 ± 1.67 82.85 ± 2.31 84.86 ± 2.23 0.67 ± 0.03 Table 2. Validation and test results for the SVM, RF and NN models trained on the HAPPENN dataset. Method Acc Sn Sp MCC Cross-validation SVM (linear) 81.22 ± 2.58 76.87 ± 4.06 84.26 ± 3.13 0.61 ± 0.05 SVM (RBF) 77.51 ± 2.91 71.26 ± 4.77 81.94 ± 3.44 0.54 ± 0.06 RF 83.30 ± 2.35 77.68 ± 4.10 87.26 ± 2.39 0.65 ± 0.05 NN 85.66 ± 1.93 84.96 ± 3.37 86.09 ± 3.43 0.71 ± 0.04 External validation SVM (linear) 81.49 ± 1.80 76.77 ± 2.46 84.85 ± 2.13 0.62 ± 0.04 SVM (RBF) 77.79 ± 2.11 71.33 ± 2.78 82.37 ± 2.97 0.54 ± 0.04 RF 84.06 ± 1.38 78.56 ± 2.69 87.96 ± 1.75 0.67 ± 0.03 NN 84.00 ± 1.67 82.85 ± 2.31 84.86 ± 2.23 0.67 ± 0.03 Table 2. Validation and test results for the SVM, RF and NN models trained on the HAPPENN dataset. achieves the best balance between sensitivity and specificity. As it was the most capable, the neural network approach was selected as the classifier of choice for the prediction of hemolytic activity. Resultsh The predictive power of HAPPENN was further evaluated by means of the receiver operating characteristic (ROC) curve, and its associ- ated area under the curve (AUC) (Fig. 7), which is equivalent to the probability that the predictor will rank a randomly selected positive instance higher than a negative one. We note that the performance is nearly excellent on both the validation and test sets, with both yielding an AUC of 0.90. Comparison to HemoPI and HemoPred. HemoPI and HemoPred are in silico peptide hemolytic activity predic- tion models previously reported in the literature, to which the HAPPENN model is compared.h The former approach employs a support vector machine (SVM) trained on a combination of single residue-, dipeptide- and property-based features, while the latter employs a random forest (RF) trained on a combination of amino acid and dipeptide composition features. Both models achieve similar cross-validated accuracies and MCC scores not exceeding 78% and 0.56 when trained on the HemoPI-2 and HemoPI-3 datasets. Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Figure 7. Receiver operating characteristic plot of HAPPENN performance on (A) the tenfold cross-validation sets and (B) the external validation. Figure 7. Receiver operating characteristic plot of HAPPENN performance on (A) the tenfold cross-validation sets and (B) the external validation. Table 3. Validation and test results for HemoPI, HemoPred and HAPPENN. HemoPI, HemoPred and HAPPENN-HemoPI3-Equiv datasets are subjected to fivefold cross-validation, while HAPPENN employs tenfold cross-validation for the HAPPENN, HAPPENN-RR90 and HAPPENN-hard datasets. Resultsh Dataset Classifier Acc (%) Sn (%) Sp (%) MCC Cross-validation HemoPI-2 HemoPI 78.0 78.3 77.6 0.56 HemoPred 76.18 ± 0.40 76.57 ± 0.34 75.66 ± 0.53 0.52 ± 0.01 HemoPI-3 HemoPI 77.98 79.24 76.48 0.56 HemoPred 77.60 ± 0.70 77.91 ± 0.94 77.18 ± 0.58 0.55 ± 0.01 HAPPENN HAPPENN 85.66 ± 1.93 84.96 ± 3.37 86.09 ± 3.43 0.71 ± 0.04 HAPPENN-RR90 HAPPENN 82.73 ± 2.73 83.38 ± 4.82 82.19 ± 4.22 0.65 ± 0.05 HAPPENN-hard HAPPENN 77.54 ± 3.31 82.12 ± 6.41 72.45 ± 7.74 0.55 ± 0.06 HAPPENN-HemoPI3-equiv HAPPENN 85.44 ± 1.23 83.45 ± 3.16 86.79 ± 1.94 0.70 ± 0.02 External validation HemoPI-2 HemoPI 75.7 78.2 78.3 0.51 HemoPred 76.82 ± 3.40 78.91 ± 3.82 74.29 ± 6.62 0.53 ± 0.07 HemoPI-3 HemoPI 77.16 81.92 71.43 0.54 HemoPred 79.91 ± 0.68 85.20 ± 2.09 73.33 ± 1.76 0.59 ± 0.01 HAPPENN HAPPENN 84.00 ± 1.67 82.85 ± 2.31 84.86 ± 2.23 0.67 ± 0.03 HAPPENN-RR90 HAPPENN 80.65 ± 2.41 81.75 ± 4.30 79.84 ± 1.96 0.61 ± 0.05 HAPPENN-hard HAPPENN 73.94 ± 2.74 78.26 ± 3.56 69.49 ± 2.46 0.48 ± 0.06 HAPPENN-HemoPI3-Equiv HAPPENN 84.96 ± 0.53 84.67 ± 1.19 85.27 ± 0.63 0.70 ± 0.01 Table 3. Validation and test results for HemoPI, HemoPred and HAPPENN. HemoPI, HemoPred and HAPPENN-HemoPI3-Equiv datasets are subjected to fivefold cross-validation, while HAPPENN employs tenfold cross-validation for the HAPPENN, HAPPENN-RR90 and HAPPENN-hard datasets. The HAPPENN model achieves good validation statistics, with a tenfold cross-validated accuracy of 85.66% and MCC value of 0.71. The HAPPENN approach achieves prediction performance that significantly exceeds that of both HemoPI and HemoPred (Table 3), although the cross-validation scheme and test set used differ. In order to facilitate a more direct comparison, an altered dataset was created, termed HAPPENN-HemoPI3- equiv, wherein the test dataset consists exclusively of HAPPENN dataset peptides which also form part of the HemoPI-3 test set. The remaining non-test set peptides were used for training and validation. Using this altered dataset, a neural network sharing the architecture and hyperparameters of the main HAPPENN neural network was trained under fivefold cross-validation, achieving a test set accuracy of 84.96% and an MCC of 0.70. Resultsh While these results again exceed those of the available classifiers, it should be noted that the test set in this case is not truly independent as its constituent peptides had been previously used in optimising the hyperparameters of the main HAPPENN neural network. The HAPPENN model achieves good validation statistics, with a tenfold cross-validated accuracy of 85.66% and MCC value of 0.71. The HAPPENN approach achieves prediction performance that significantly exceeds that of both HemoPI and HemoPred (Table 3), although the cross-validation scheme and test set used differ. In order to facilitate a more direct comparison, an altered dataset was created, termed HAPPENN-HemoPI3- equiv, wherein the test dataset consists exclusively of HAPPENN dataset peptides which also form part of the HemoPI-3 test set. The remaining non-test set peptides were used for training and validation. Using this altered dataset, a neural network sharing the architecture and hyperparameters of the main HAPPENN neural network was trained under fivefold cross validation achieving a test set accuracy of 84 96% and an MCC of 0 70 While The HAPPENN model achieves good validation statistics, with a tenfold cross-validated accuracy of 85.66% and MCC value of 0.71. The HAPPENN approach achieves prediction performance that significantly exceeds that of both HemoPI and HemoPred (Table 3), although the cross-validation scheme and test set used differ. d f l d l d d d d f In order to facilitate a more direct comparison, an altered dataset was created, termed HAPPENN-HemoPI3- equiv, wherein the test dataset consists exclusively of HAPPENN dataset peptides which also form part of the HemoPI-3 test set. The remaining non-test set peptides were used for training and validation. Using this altered dataset, a neural network sharing the architecture and hyperparameters of the main HAPPENN neural network was trained under fivefold cross-validation, achieving a test set accuracy of 84.96% and an MCC of 0.70. While these results again exceed those of the available classifiers, it should be noted that the test set in this case is not truly independent as its constituent peptides had been previously used in optimising the hyperparameters of the main HAPPENN neural network. Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Figure 8. Plot of the neural network’s output values against x, the peptides’ multiple of the threshold concentration. Where x < 4 , the values are presented to-scale. Where x > 4 , the values are presented not-to- scale. Resultsh Peptide’s which the literature states possess no hemolytic activity are presented separately, also not-to- scale. Correctly predicted peptides are coloured blue, incorrectly predicted peptides are coloured red. Figure 8. Plot of the neural network’s output values against x, the peptides’ multiple of the threshold concentration. Where x < 4 , the values are presented to-scale. Where x > 4 , the values are presented not-to- scale. Peptide’s which the literature states possess no hemolytic activity are presented separately, also not-to- scale. Correctly predicted peptides are coloured blue, incorrectly predicted peptides are coloured red. Figure 8. Plot of the neural network’s output values against x, the peptides’ multiple of the threshold concentration. Where x < 4 , the values are presented to-scale. Where x > 4 , the values are presented not-to- scale. Peptide’s which the literature states possess no hemolytic activity are presented separately, also not-to- scale. Correctly predicted peptides are coloured blue, incorrectly predicted peptides are coloured red. Figure 8. Plot of the neural network’s output values against x, the peptides’ multiple of the threshold concentration. Where x < 4 , the values are presented to-scale. Where x > 4 , the values are presented not-to- scale. Peptide’s which the literature states possess no hemolytic activity are presented separately, also not-to- scale. Correctly predicted peptides are coloured blue, incorrectly predicted peptides are coloured red. Relationship between prediction and hemolytic activity values. Peptides were classified as hemolytic or non- hemolytic based on criteria given in Table 1, which relates hemolytic activity (H) to concentration (c). A pep- tide’s concentration can be expressed as a multiple (x) of the threshold concentration c. For instance, a peptide which exhibits 65% hemolytic activity at 195 μM can be said to have x = 0.5 , as 0.5 × 390 μM = 195 μM. As x is less than 1, the concentration is lower than the threshold concentration (390 μM  for 65% hemolytic activity), and the peptide is considered hemolytic.hi p p y The neural network’s output is obtained from the sigmoid activation function of its final layer. Resultsh As the sigmoid function produces values ranging between 0 and 1, these output values can be interpreted as the probability of a peptide being non-hemolytic (0) and hemolytic (1).h p p g y ( ) y ( ) The relationship between the neural network’s output values and x, the multiple of the threshold concentra- tion, are shown in Fig. 8. The upper left quadrant shows the true positives, the lower right quadrant shows the true negatives, and the upper right and lower left quadrant show the false negatives and false positives, respectively. g pp gt q g p p y It can be seen from Fig. 8, that not many peptides are found at the y = 0.5 hemolytic-non-hemolytic predic- tion boundary. While there are many peptides at the x = 1.0 activity boundary, most peptides are seen to be correctly predicted. Descriptor‑set specific results. Several approaches were trialled for constructing the input feature space (Table 4), namely dipeptide and tripeptide composition, the corresponding g-gap compositions, N- and C-ter- minus composition and physicochemical features. Dipeptide and tripeptide composition. Dipeptide composition is defined as the proportion of a given dipeptide in the sequence, while similarly the tripeptide composition is defined as the proportion of a given tripeptide in the sequence. These composition features capture both the chemical nature of the peptide composition while retaining information about the local sequence order. The models achieved respectable accuracies of 82.56% and 82.62%, respectively, and MCC values of 0.65 and 0.64, respectively. g‑gap composition. g-gap dipeptide composition is described as the proportion of a given pair of amino acids separated by 1, 2 or 3 residues, which corresponds to residues which are adjacent in three-dimensional space, especially in regular secondary structures where such non-adjoining residues may be connected by hydrogen bonds. Interestingly, models trained on these features perform better than those trained on the more conven- tional dipeptide and tripeptide compositions. This can be attributed to these features better capturing the chemi- cal environment that the peptide exposes to the membrane upon contact. For instance, the g-gap feature can represent the spatial adjacency separated by one turn of the α-helix, which has a turn of 3.6 residues. Termini. A feature set was created to capture the information on the residues specific location in the sequence. This feature set consists of a binary profile for the first and last 15 residues of each peptide. Resultsh fore represented by a number of vectors, the first of which represents the conventional 20 amino acid alphabet is of length 30 × 20 , and the remaining vectors represent the conjoint alphabet and the reduced alphabets of Veltri, Thomas and Dill. The model trained on this feature set achieved an accuracy of 82.11% and an MCC of 0.64. fore represented by a number of vectors, the first of which represents the conventional 20 amino acid alphabet is of length 30 × 20 , and the remaining vectors represent the conjoint alphabet and the reduced alphabets of Veltri, Thomas and Dill. The model trained on this feature set achieved an accuracy of 82.11% and an MCC of 0.64. Physicochemical. Interestingly, the worst-performing model is the model trained on the physicochemical fea- tures, achieving an accuracy of only 80.63% and an MCC of 0.61 on the non-redundancy reduced dataset.h Of the single feature-set approaches trialled, none outperform the model trained on the full feature set. The compositional descriptor-based models are seen to benefit from sequence similarity to an extent, and exhibit somewhat reduced performance on the redundancy-reduced dataset. The physicochemical descriptor-based model, conversely, maintains a comparable performance even on the redundancy-reduced dataset. Feature importance analysis. Random forest feature importance. Random forests have the advan- tage of being easily interpretable and provide an easy method of ranking the importance of input features. The most useful features as determined by cross-validated random forests are the Eisenberg direction of the hydrophobic moment (EISD860103)77, the Eisenberg normalized hydrophobicity scale (EISD840101)73, hy- dropathy (NADH010102, NADH010103, NADH010104)71, the hydrophobic parameter pi (FAUJ830101)72, the Boman ­index45, apparent partition energies (GUYH850105)80, membrane-propensity (PUNT030102)98, trans- membrane ­propensity63, side-chain hydrophobicity values (BLAS910101)76 and Hopp-Woods hydrophobicity (HOPT810101)50.f Effectively all of these features directly or indirectly quantify hydrophobicity, which points to it being impor- ant for hemolytic activity. Analysis of neural network weights using Garson’s method. In order to understand the basis for the neural net- work’s predictive power, we analysed the importance assigned to various input features by Garson’s ­method126, iteratively reducing the feature input space by approximately halving the number of input features, until  300 composition features were retained in each split. 24 features were identified as having large weights in each of the 10 splits.h p The occurrences of the FS, LH, KIK, VAK, VLK dipeptides and tripeptides in the peptide sequence were found to be important. Resultsh Each peptide is there- Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ Table 4. Validation statistics achieved by neural networks trained with just single sets of descriptor Dataset Features Acc (%) Sn (%) Sp (%) MCC Cross-validation HAPPENN Composition, dipeptide 82.56 ± 2.42 82.59 ± 3.52 82.50 ± 3.12 0.65 ± 0.05 Composition, tripeptide 82.62 ± 1.74 80.46 ± 4.45 84.06 ± 2.27 0.64 ± 0.04 g-gap composition, dipeptide 84.02 ± 2.38 83.87 ± 3.14 84.13 ± 3.20 0.68 ± 0.05 g-gap composition, tripeptide 83.55 ± 1.85 82.07 ± 3.06 84.54 ± 2.86 0.66 ± 0.04 Termini 82.11 ± 2.00 80.92 ± 3.75 82.95 ± 2.58 0.63 ± 0.04 Physicochemical 80.63 ± 2.60 82.02 ± 5.13 79.64 ± 4.27 0.61 ± 0.05 HAPPENN-RR90 Composition, dipeptide 78.05 ± 2.57 81.25 ± 4.90 75.67 ± 3.86 0.56 ± 0.05 Composition, tripeptide 77.47 ± 2.62 72.79 ± 8.53 80.92 ± 5.63 0.54 ± 0.06 g-gap composition, dipeptide 79.52 ± 2.79 81.50 ± 6.21 78.04 ± 4.86 0.59 ± 0.06 g-gap composition, tripeptide 78.61 ± 2.75 78.49 ± 6.16 78.68 ± 4.90 0.57 ± 0.06 Termini 79.38 ± 3.01 80.39 ± 6.52 78.63 ± 3.02 0.59 ± 0.06 Physicochemical 79.02 ± 4.42 81.70 ± 6.77 77.01 ± 5.75 0.58 ± 0.09 External validation HAPPENN Composition, dipeptide 81.42 ± 1.61 81.25 ± 1.51 81.58 ± 2.49 0.62 ± 0.03 Composition, tripeptide 81.42 ± 1.01 79.91 ± 1.73 82.49 ± 1.11 0.62 ± 0.02 g-gap composition, dipeptide 82.98 ± 1.76 82.76 ± 2.47 83.17 ± 1.64 0.65 ± 0.04 g-gap composition, tripeptide 81.80 ± 1.17 80.53 ± 1.59 82.72 ± 1.19 0.63 ± 0.02 Termini 81.34 ± 1.15 80.02 ± 3.18 82.33 ± 1.26 0.62 ± 0.02 Physicochemical 79.30 ± 1.82 80.93 ± 3.89 78.21 ± 2.27 0.58 ± 0.04 HAPPENN-RR90 Composition, dipeptide 76.92 ± 1.43 80.03 ± 2.43 74.62 ± 1.69 0.54 ± 0.03 Composition, tripeptide 76.44 ± 0.96 74.63 ± 4.15 77.79 ± 2.57 0.52 ± 0.02 g-gap composition, dipeptide 78.14 ± 1.82 79.00 ± 2.75 77.47 ± 3.37 0.56 ± 0.03 g-gap composition, tripeptide 77.16 ± 1.80 77.06 ± 3.52 77.28 ± 1.17 0.54 ± 0.04 Termini 78.07 ± 2.54 78.30 ± 3.88 77.93 ± 2.11 0.56 ± 0.05 Physicochemical 77.11 ± 3.58 79.33 ± 5.38 75.47 ± 3.46 0.54 ± 0.07 Table 4. Validation statistics achieved by neural networks trained with just single sets of descriptors. Discussion A decreasing number of drug approvals and a rising research and development cost base has contributed to a resurgence of interest in peptide therapeutics. An ideal peptide drug should possess a high therapeutic index, specifically high activity against the biological target and limited toxicity. The therapeutic potential of peptides, however, is highly dependent on it possessing little to no hemolytic activity. Minimizing hemolytic activity is important for improving the therapeutic index of a peptide. g Many research groups have studied the structures of natural peptides as well as engineered peptide ana- logues in order to characterise how their structure determines their biological ­activities127–129. A comprehensive understanding of the relationship between structure and function, however, remains elusive. A computational method that can provide information about a peptide’s biological activity from its primary structure prior to chemical synthesis, however, would allow for rapid and efficient exploration of the chemical space and present a significant cost and-time saving. gi g To accelerate the lead molecule design and optimization pipeline, this study aimed to create an in silico method for classifying therapeutic peptides as hemolytic or non-hemolytic based on their primary sequence. The prediction task is challenging, however, as it requires distinguishing between desirable activity at the peptide’s target, the prokaryotic plasma membrane in the case of most antimicrobial peptides, and activity at the membrane of eukaryotic erythrocytes. The task is further complicated by the varying extent to which many peptides display hemolytic activity, which makes arbitrarily classifying them as hemolytic or non-hemolytic challenging. Indeed, there is limited consensus on the most appropriate metric to quantify hemolysis, with many articles reporting only one metric recorded at a single concentration, which consequently precludes a regression approach instead of a classification approach. The topic is complicated further by a lack of consensus on the definition of a key metric, MHC. Most studies define it as the minimum hemolytic concentration, but differ on the specific criteria, with different studies defining it as the concentration at which 5%, 10%, 50% or even 100% hemolysis occurs. Some even define it as the maximum concentration that does not cause any ­hemolysis130. Ideally, studies would present the analysis of hemolytic activity as a series of measurements undertaken at several concentrations, which would allow for a fuller understanding of the toxicity-concentration profile. Discussion Until such a time, however, using the MHC values for training classifiers requires investigating its actual meaning on a case by case basis. In the course of verifying the activity of the sequences in our dataset, and comparing our dataset to the HemoPI dataset, we identified a number of instances of misclassified sequences, sequences whose hemolytic activities were not clear, and sequences whose presence in the literature we were unable to independently verify. These sequences were not included in the HAPPENN dataset.hii q The success and validity of a machine learning classifier are predicated on the correct definition of the problem at hand, which in this case encompasses the definition of positive and negative datasets. Both the positive and negative datasets consist of experimentally validated peptide sequences which exhibit antimicrobial or other bio- logical activities. Unlike HemoPI and HemoPred, we chose not to conduct a machine learning experiment where the negative dataset consists of peptides randomly extracted from proteins in Swiss-Prot, as a machine learning classifier is most dependable when only one property of interest is varied. Using randomly extracted sequences as the negative dataset, and hemolytic antimicrobial peptides as the positive set, likely results in the classifier learning to predict general membrane activity, rather than specifically activity against eukaryotic erythrocytes. The authors believe that the HAPPENN dataset represents a major improvement on the HemoPI datasets, both in terms of size and reliability, as it contains 3738 peptides with confirmed biological activities, compared to the 904 and 1623 sequences present in the HemoPI-2 and HemoPI-3 datasets, and therefore has been made available for download both as supplementary information and on the server’s website. Once a reliable dataset was constructed, the peptide sequences were translated into vectors of physicochemi- cal and composition features, and a number of different machine learning approaches, namely support vector machines, random forests and neural networks, were trialled for relating the peptides’ features to their hemolytic activities. The neural network approach proved most promising, and was therefore retained, further optimised, and had its predictive power thoroughly evaluated by means of tenfold cross-validation and external validation on an independent test set.hii The final neural network model achieved a tenfold cross-validated accuracy, sensitivity and specificity of 85.66%, an MCC of 0.71, and an AUC of 0.90. Resultsh Additionally, the LLL Veltri reduced alphabet tripeptide and the RGV and VCR Thomas and Dill (length 3) reduced alphabet tripeptides were found to contribute strongly to the final classification. Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 12 www.nature.com/scientificreports/ Additionally, the occurrence of g-gap i,i+3 residue pairs FG, FL, LK, WV, the occurrence of g-gap i, i+4 residue pair FK and the occurrence of g-gap i, i+2 residue pairs AR, FL, FS, LF were found to be meaningful. Furthermore, the occurrence of the g-gap i,i+3 Thomas and Dill (10) reduced alphabet residue pairs PS and WW and g-gap i,i+3 Veltri reduced alphabet residue pair QQ were also important. g g p p p p Finally, the network weights associated with the EstateVSA3 and EstateVSA4 (MOE-type descriptors using Estate indices and surface area contributions), Geary autocorrelation-lag8 weight by atomic polarizabilities, and the acetylation of the N-terminus inputs were also large. y p g Interestingly, the predictive power of the reduced feature space neural network is nearly as strong as the main network. www.nature.com/scientificreports/ the neural network approach are possible and indeed expected, as the number of peptides in the literature with characterised hemolytic activity increases.h The main HAPPENN dataset was not redundancy-reduced, as even a single amino acid substitution can affect a peptide’s bioactivity. Nonetheless, to determine to what extent sequence similarity contributes to the model’s performance, the experiments were repeated with a redundancy-reduced dataset, which achieved a cross- validated accuracy of 82.73% and MCC of 0.65. While these values are lower than the non-redundancy-reduced dataset, they illustrate that the majority of the predictive power of the model is not derived from sequence simi- larity, especially considering that neural networks perform best when trained with larger datasets, and redun- dancy reduction significantly reduces the amount of data available for training. Finally, to ascertain the model’s power in distinguishing between similar peptides with different hemolytic activities, a model was trained on the HAPPENN-hard dataset, which contains the positive examples from the HAPPENN-RR90 dataset, and for each positive example, the most compositionally similar non-hemolytic peptide, as measured by the Euclidean distance between their amino acid composition vectors. Despite the similarity between the positive and negative peptides, the model achieves a respectable accuracy of 77.54% and MCC of 0.55. p p p y Interestingly, the results of training a neural network with reduced feature spaces are in close agreement with the unsupervised principal component analysis (PCA) (Fig. 5) and t-distributed Stochastic Neighbour Embedding (t-SNE) (Fig. 6) analysis. The neural network trained on just the physicochemical features had the least predictive power among the networks trained, which coincides with the physicochemical features’ PCA plot having the least separation between the hemolytic and non-hemolytic classes. The relatively lower predic- tive power of the physicochemical descriptors highlights the need for the development of novel, peptide-specific descriptors that account for their capacity to adopt complex three-dimensional structures. When trained on the redundancy-reduced dataset, the difference in predictive power between the compositional and physicochemi- cal descriptors is reduced.i p To ascertain the source of misclassification of the wrongly predicted peptides, the main model’s false positives and false negatives were highlighted on the PCA and t-SNE plots. It is apparent that many of the misclassifications occur due to the peptides’ compositional and/or physicochemical similarity to peptides with differing hemolytic activity. www.nature.com/scientificreports/ To gain further insight into the source of misclassification, the peptide with the most similar percentage amino acid composition but opposite hemolytic activity was identified for each wrongly predicted peptide. For 16% of misclassified peptides, a compositionally identical peptide with opposite hemolytic activity was identified, compared with just 5% for correctly classified peptides. Overall, misclassified peptides had a smaller Euclidean distance to their most compositionally similar opposite-activity peptide than correctly classified peptides did. y y yi To gain insight into which features were most important for hemolytic activity, the importance assigned to features by random forests was investigated. Hydrophobicity, as quantified by a selection of different metrics, appears to be critical for hemolytic activity, with more hydrophobic sequences generally being found to be more hemolytic than less hydrophobic sequences. These findings are not surprising, and are consistent with the available ­literature131,132. A number of compositional descriptors were also found to be indicative of hemolytic propensity, with FS, LH, KIK, VAK and VLK being ranked as important. p p y g p HAPPENN’s power is demonstrated by an alanine scan applied to maximin 3, a non-hemolytic ­peptide127,133. Interestingly, the classifier predicts maximin 3 and all of its alanine scan mutants to be non-hemolytic, with the single exception of [E20A]maximin 3, which is consistent with the literature, which acknowledges the relation- ship between a peptide’s net charge and its hemolytic ­activity134.hii This study presents a significant improvement in the area of in silico hemolytic activity classification, with its results forming the new state-of-the-art. The novel application of a neural network combined with the HAPPENN dataset’s superior data quality and quantity has facilitated a 35% decrease in classification error, compared to the results achieved by the best currently available tools. y y To conclude, accurate prediction of hemolytic activity of antimicrobial peptides can facilitate in silico design of novel peptide-based therapeutics, thereby accelerating the design phase and reducing its cost. HAPPENN distinguishes itself from existing methods through its focus on antimicrobial peptides, more accurate prediction and incorporation of novel features. Although HAPPENN displays advantages compared to competing methods, it is limited by the lower inter- pretability of the neural network’s hidden layers. Prediction of hemolytic activity from primary sequence remains a challenging problem, as it is characterised by a complex interplay between numerous features, which also contribute to the desirable antimicrobial activities. www.nature.com/scientificreports/ Nonetheless, HAPPENN possesses an error rate 35% lower than the most accurate existing classifiers, and we believe that this work will aid future studies focused on the identification and design of novel peptide therapeutics. Discussion The validation statistics demonstrated that the model is capable of discriminating between hemolytic and non-hemolytic peptides, and that it exhibits minimal bias towards one class or another. The model performs very well compared to the existing methods, with a 35.3% decrease in cross-validated error relative to HemoPI and HemoPred. The model’s residual prediction error rate can likely be attributed to a limited sample size for neural networks to accurately learn from, as well as the fine boundary between the definition of hemolytic and non-hemolytic peptides combined with the margin of error associated with the experimental determination of hemolytic activity. Further improvements to the predictive power of Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ References J., Kei, W. K. & Cho, C. H. Cathelicidin a potential therapeutic peptide for gastrointestinal inflammation and cancer. World J. Gastroenterol. 19, 2731–2735 (2013). 9. Bercier, J. G., Al-Hashimi, I., Haghighat, N., Rees, T. D. & Oppenheim, F. G. Salivary histatins in patients with recurrent ora candidiasis. J. Oral Pathol. Med. 28, 26–29 (1999).h J ( ) 10. Lau, J. L. & Dunn, M. K. Therapeutic peptides: historical perspectives, current development trends, and future directions. Bioorg. Med. Chem. 26, 2700–2707 (2018). 11. Sohrabi, C., Foster, A. & Tavassoli, A. Methods for generating and screening libraries of genetically encoded cyclic peptides in drug discovery. Nat. Rev. Chem. 4, 90–101 (2020).i g y 12. Bozovičar, K. & Bratkovič, T. Evolving a peptide: library platforms and diversification strategies. Int. J. Mol. Sci. 21, 215 (2020) k Á b d b l h d f d h f l d 12. Bozovičar, K. & Bratkovič, T. Evolving a peptide: libr Á 12. Bozovičar, K. & Bratkovič, T. Evolving a peptide: library platforms and diversification strategies. Int. J. Mol. Sci. 21, 215 (2020). 13. Furka, Á, Sebestyén, F., Asgedom, M. & Dibó, G. General method for rapid synthesis of multicomponent peptide mixtures. Int. J. Pept. Protein Res. 37, 487–493 (1991). , , g p p y pi g J , ( ) 13. Furka, Á, Sebestyén, F., Asgedom, M. & Dibó, G. General method for rapid synthesis of multicomponent peptide mixtures. Int. J. Pept. Protein Res. 37, 487–493 (1991). p 14. Lalezari, J. P. et al. A phase II clinical study of the long-term safety and antiviral activity of enfuvirtide-based antiretrovira therapy. AIDS 17, 691–698 (2003).fi y 15. Heyns, C., Simonin, M.-P., Grosgurin, P., Schall, R. & Porchet, H. Comparative efficacy of triptorelin pamoate and leuprolide acetate in men with advanced prostate cancer. BJU Int. 92, 226–231 (2003).h p 16. Reisner, E. H., Bailey, F. N. & Appelbaum, E. The treatment of pneumonia with bacitracin. Ann. Intern. Med. 34, 1232–1242 (1951).h ( ) 17. Ascione, A. Boceprevir in chronic hepatitis C infection: a perspective review. Ther. Adv. Chron. Dis. 3(3), 113–121 (2012).h h 18. Bruno, B. J., Miller, G. D. & Lim, C. S. Basics and recent advances in peptide and protein drug delivery. Ther. Deliv. 4(11) 1443–1467 (2013). 19. Hamamoto, K., Kida, Y., Zhang, Y., Shimizu, T. & Kuwano, K. References 1. Hultmark, D. Drosophila immunity: paths and patterns. Curr. Opin. Immunol. 15, 12–19 (2003). 1. Hultmark, D. Drosophila immunity: paths and patterns. Curr. Opin. Immunol. 15, 12–19 (2003). 2. Yeaman, M. R. & Yount, N. Y. Mechanisms of antimicrobial peptide action and resistance. Pharmacol. Rev. 55, 27–55 (200 3. Guilhelmelli, F. et al. Antibiotic development challenges: the various mechanisms of action of antimicrobial peptides an 2. Yeaman, M. R. & Yount, N. Y. Mechanisms of antimicrobial peptide action and resistance. Pharmacol. Rev. 55, 27–55 (2003). 3. Guilhelmelli, F. et al. Antibiotic development challenges: the various mechanisms of action of antimicrobial peptides and o bacterial resistance. Front. Microbiol. 4, 353 (2013). bacterial resistance. Front. Microbiol. 4, 353 (2013 4. Vlieghe, P., Lisowski, V., Martinez, J. & Khrestchatisky, M. Synthetic therapeutic peptides: science and market. Drug Discov Today 15, 40–56 (2010). 5. Gordon, Y. J., Romanowski, E. G. & McDermott, A. M. Mini review: A review of antimicrobial peptides and their therapeutic potential as anti-infective drugs. Curr. Eye Res. 30, 505–515 (2005). p g y 6. Conlon, J. M., Mechkarska, M., Lukic, M. L. & Flatt, P. R. Potential therapeutic applications of multifunctional host-defense peptides from frog skin as anti-cancer, anti-viral, immunomodulatory, and anti-diabetic agents. Peptides 57, 67–77 (2014). 7. Karapetyan, A. V. et al. Bioactive lipids and cationic antimicrobial peptides as new potential regulators for trafficking of bone d i d t ll i ti t ith t di l i f ti St C ll D 22 1645 1656 (2013) 6. Conlon, J. M., Mechkarska, M., Lukic, M. L. & Flatt, P. R. Potential therapeutic applications of multifunctional host-defense peptides from frog skin as anti-cancer, anti-viral, immunomodulatory, and anti-diabetic agents. Peptides 57, 67–77 (2014). 7. Karapetyan, A. V. et al. Bioactive lipids and cationic antimicrobial peptides as new potential regulators for trafficking of bone marrow-derived stem cells in patients with acute myocardial infarction. Stem Cells Dev. 22, 1645–1656 (2013).l pept des o og s as a t ca ce , a t v a , u o odu ato y, a d a t d abet c age ts. eptides 57, 67 77 ( 0 ). 7. Karapetyan, A. V. et al. Bioactive lipids and cationic antimicrobial peptides as new potential regulators for trafficking of bone marrow-derived stem cells in patients with acute myocardial infarction. Stem Cells Dev. 22, 1645–1656 (2013).l 8. Chow, J. Y. C., Li, Z. Data availability Data availability All data generated or analysed during this study are included in this published article’s supplementary data sets a availability ata generated or analysed during this study are included in this published article’s supplementary data sets. y All data generated or analysed during this study are included in this published article’s supplementary data sets Received: 3 March 2020; Accepted: 9 June 2020 References Antimicrobial activity and stability to proteolysis of small linear cationic peptides with D-amino acid substitutions. Microbiol. Immunol. 46, 741–749 (2002). 20. Wimley, W. C. Describing the mechanism of antimicrobial peptide action with the interfacial activity model. ACS Chem. Biol 5, 905–917 (2010). ( ) 21. Hu, Y., Sinha, S. K. & Patel, S. Investigating hydrophilic pores in model lipid bilayers using molecular simulations: correlating bilayer properties with pore-formation thermodynamics. Langmuir 31, 6615–6631 (2015). h b l d f d b h h 22. Lai, R., Liu, H., Hui Lee, W. & Zhang, Y. An anionic antimicrobial peptide from toad Bombina maxima. Biochem. Biophys. Res Commun. 295, 796–799 (2002). 23. Matsuzaki, K., Sugishita, K. I., Fujii, N. & Miyajima, K. Molecular basis for membrane selectivity of an antimicrobial peptide Magainin 2. Biochemistry 34, 3423–3429 (1995). g y 24. Gomes, B. et al. Designing improved active peptides for therapeutic approaches against infectious diseases. Biotechnol. Adv. 36 415–429 (2018). 24. Gomes, B. et al. Designing improved active peptides for therapeutic approaches against infectious diseases. Biotechnol. Adv. 36, 415–429 (2018). 25 Zeng M et al Protein-protein interaction site prediction through combining local and global features with deep neural networks 415–429 (2018). 25. Zeng, M. et al. Protein-protein interaction site prediction through combining local and global features with deep neural networks. Bioinformatics 36 1114 1120 (2020) 25. Zeng, M. et al. Protein-protein interaction site prediction through combining local and global features with deep neural networks Bioinformatics 36, 1114–1120 (2020). f ( ) 26. Oti, M., Ballouz, S. & Wouters, M. a. In silico tools for gene discovery. Methods Mol. Biol. 760, 175–187 (2011). l ll h ld d d f ll d 26. Oti, M., Ballouz, S. & Wouters, M. a. In silico tools for gene discovery. Methods Mol. Biol. 760, 175–187 (2011). 27. Holton, T. A., Pollastri, G., Shields, D. C. & Mooney, C. CPPpred: prediction of cell penetrating peptides. Bioinformatics 29, ( ) 27. Holton, T. A., Pollastri, G., Shields, D. C. & Mooney, C. CPPpred: prediction of cell penetrating peptides. Bioinformatics 29 3094–3096 (2013). 28. Pirtskhalava, M. et al. Erratum: DBAASP vol 2: an enhanced database of structure and antimicrobial/cytotoxic activity of natural and synthetic peptides (Nucleic Acids Research 44 (D1104–D1112) DOI 10.1093/nar/gkv1174). Nucleic Acids Res. 44, 6503 (2016). ( ) 29. Waghu, F. H. et al. CAMP: collection of sequences and structures of antimicrobial peptides. Nucleic Acids Res. 42, D1154-8 (2014). www.nature.com/scientificreports/ lytic, based on the neural network’s prediction. This probability is on a scale of 0–1, where 0 is most probably non-hemolytic and 1 is most probably hemolytic. Mutation analysis. Mutation analysis is available for single sequences, provided in FASTA format. After inputting the sequence, the user should select the mutation analysis option, input the residue number that they wish to mutate, and run the prediction. The server will predict the hemolytic activity of each of the peptide’s mutants attained by substituting the residue at the selected position with each of the other natural 20 amino acids. Residue scan. A residue scan, for instance an alanine-scan, is available for single sequences provided in FASTA format. After inputting the sequence, the user should select the residue scan option, choose the residue they wish to scan with and run the prediction. The server will predict the hemolytic activity of each of the pep- tide’s mutants attained by substituting successive residue positions with the selected residue. Web server implementationi To best serve the scientific community, we have made the classifier algorithm available online at https​://resea​ rch.timmo​ns.eu/happe​nn in the form of an easy to use web-server, which is available for free use by academic researchers. The web server is capable of predicting the hemolytic activity of peptides’ based on their primary sequence, as well as the presence or absence of N-terminal acetylation or C-terminal amidation modifications. Prediction is limited to peptides composed of the 20 natural amino acids; non-natural amino acids are not sup- ported. The web server possesses many features. Neural network models trained on the HAPPENN, HAPPENN- RR90 and HAPPENN-hard datasets are available for prediction. Hemolytic activity prediction. Hemolytic activity prediction is available for both single and multiple sequences. The user should submit the peptide sequence or sequences in FASTA format, select the neural net- work model they wish to use for prediction, and the server will return the probability of the peptide being hemo- Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 www.nature.com/scientificreports/ References Scientific Reports | (2020) 10:10869 | https://doi.org/10.1038/s41598-020-67701-3 15 www.nature.com/scientificreports/ 30. Gautam, A. et al. Hemolytik: a database of experimentally determined hemolytic and non-hemolytic peptides. Nucleic Acids Res. 42, D444-9 (2014). 31. Li, W. & Godzik, A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinfor‑ matics 22, 1658–1659 (2006). 32. Huang, Y., Niu, B., Gao, Y., Fu, L. & Li, W. CD-HIT Suite: a web server for clustering and comparing biological sequences Bioinformatics 26, 680–682 (2010). f 33. Fu, L., Niu, B., Zhu, Z., Wu, S. & Li, W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152 (2012). 28, 3150 3152 (2012). 34. Loose, C., Jensen, K., Rigoutsos, I. & Stephanopoulos, G. A linguistic model for the rational design of antimicrobial peptides. Nature 443, 867–869 (2006). 34. Loose, C., Jensen, K., Rigoutsos, I. & Stephanopoulos, G. A linguistic model for the rational design of antimicrobial peptides Nature 443, 867–869 (2006). 35. Porto, W. F., Pires, A. S. & Franco, O. L. Computational tools for exploring sequence databases as a resource for antimicrobia peptides. Biotechnol. Adv. 35, 337–349 (2017).hi h gi 37. Agrawal, P. et al. In silico approach for prediction of antifungal peptides. Front. Microbiol. 9, 323 (2018). 38. Agrawal, P., Kumar, S., Singh, A., Raghava, G. P. & Singh, I. K. NeuroPIpred: a tool to predict, design and scan insect neuropep- tides. Sci. Rep. 9, 5129 (2019). p 39. Apweiler, R. et al. Update on activities at the Universal Protein Resource (UniProt) in 2013. Nucleic Acids Res. 41, D43-7 (2013) p 39. Apweiler, R. et al. Update on activities at the Universal Protein Resource (UniProt) in 2013. Nucleic Acids Res. 41, D43-7 (2013). 40. Dey, K. K., Xie, D. & Stephens, M. A new sequence logo plot to highlight enrichment and depletion. BMC Bioinform. 19, 473 (2018). 40. Dey, K. K., Xie, D. & Stephens, M. A new sequence logo plot to highlight enrichment and depletion. BMC Bioinform. 19, 473 (2018). 41. Müller, A. T., Gabernet, G., Hiss, J. A. & Schneider, G. modlAMP: Python for antimicrobial peptides. Bioinformatics (Oxford England) 33, 2753–2755 (2017). g 42. Cao, D. S., Xu, Q. S., Hu, Q. N. & Liang, Y. Z. ChemoPy: freely available python package for computational biology and chemo- informatics. Bioinformatics 29, 1092–1094 (2013). f 43. Lobry, J. R. & Gautier, C. References Hydrophobicity, expressivity and aromaticity are the major trends of amino-acid usage in 999 Escheri chia coli chromosome-encoded genes. Nucleic Acids Res. 22, 3174–3180 (1994).h g 44. Ikai, A. Thermostability and aliphatic index of globular proteins. J. Biochem. 88, 1895–8 (1980).i h 45. Boman, H. G., Wade, D., Boman, I. A., Wåhlin, B. & Merrifield, R. B. Antibacterial and antimalarial properties of peptides tha are cecropin–melittin hybrids. FEBS Lett. 259, 103–106 (1989). 46. Juretić, D., Vukičević, D., Ilić, N., Antcheva, N. & Tossi, A. Computational design of highly selective antimicrobial peptides. J Chem. Inf. Model. 49, 2873–2882 (2009). f 47. Argos, P., Rao, J. K. & Hargrave, P. A. Structural prediction of membrane-bound proteins. Eur. J. Biochem. 128, 565–575 (1982) 47. Argos, P., Rao, J. K. & Hargrave, P. A. Structural prediction of membrane-bound proteins. Eur. J. Biochem. 128, 565–575 (1982). 48. Eisenberg, D., Weiss, R. M., Terwilliger, T. C. & Wilcox, W. Hydrophobic moments and protein structure. Faraday Symp. Chem. Soc. 17, 109–120 (1982). 48. Eisenberg, D., Weiss, R. M., Terwilliger, T. C. & Wilcox, W. Hydrophobic moments and protein structure. Faraday Symp. Chem. Soc. 17, 109–120 (1982). 49. Kyte, J. & Doolittle, R. F. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105–132 (1982) 49. Kyte, J. & Doolittle, R. F. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105–132 (1982). 50. Hopp, T. P. & Woods, K. R. Prediction of protein antigenic determinants from amino acid sequences. Proc. Natl. Acad. Sci. USA 78, 3824–3828 (1981). y p p y g y p p 50. Hopp, T. P. & Woods, K. R. Prediction of protein antigenic determinants from amino acid sequences. Proc. Natl. Acad. Sci. USA 78, 3824–3828 (1981). 51. Cornette, J. L. et al. Hydrophobicity scales and computational techniques for detecting amphipathic structures in proteins. J Mol. Biol. 195, 659–685 (1987).h 52. Zimmerman, J. M., Eliezer, N. & Simha, R. The characterization of amino acid sequences in proteins by statistical methods. J Theor. Biol. 21, 170–201 (1968). h 53. Senes, A. et al. Ez, a depth-dependent potential for assessing the energies of insertion of amino acid side-chains into membranes: derivation and applications to determining the orientation of transmembrane and interfacial helices. J. Mol. Biol. 366, 436–448 (2007).l ( ) 54. Bhaskaran, R. & Ponnuswamy, P. K. References Positional flexibilities of amino acid residues in globular proteins. Int. J. Pept. Protein Res 32, 241–255 (1988).f 55. Grantham, R. Amino acid difference formula to help explain protein evolution. Science 185, 862–864 (1974). 56. Collantes, E. R. & Dunn, W. J. Amino acid side chain descriptors for quantitative structure–activity relationship studies of peptid analogues. J. Med. Chem. 38, 2705–2713 (1995). 57. Levitt, M. & Levitt, M. Conformational preferences of amino acids in globular proteins. Biochemistry 17, 4277–4285 (1978). h dh l l h d f d d fi f l ll l f 58. Raychaudhury, C., Banerjee, A., Bag, P. & Roy, S. Topological shape and size of peptides: identification of potential allele specif helper T cell antigenic sites. J. Chem. Inf. Comput. Sci. 39, 248–254 (1999). 59. Zaliani, A. & Gancia, E. MS-WHIM scores for amino acids: a new 3D-description for peptide QSAR and QSPR studies. J. Chem Inf. Comput. Sci. 39, 525–533 (1999). f p 60. Koch, C. P. et al. Scrutinizing MHC-I binding peptides and their limits of variation. PLoS Comput. Biol. 9, e1003088 (2013). k l k f d f l d d f 60. Koch, C. P. et al. Scrutinizing MHC-I binding peptides and their limits of variation. PLoS Comput. Biol. 9, e1003088 (2013). 61. McMeekin, T. L., Wilensky, M. & Groves, M. L. Refractive indices of proteins in relation to amino acid composition and specific volume Biochem Biophys Res Commun 7 151–156 (1962) 61. McMeekin, T. L., Wilensky, M. & Groves, M. L. Refractive indices of proteins in relation to amino acid composition and specifi volume. Biochem. Biophys. Res. Commun. 7, 151–156 (1962). p y 62. Cocchi, M. & Johansson, E. Amino acids characterization by GRID and multivariate data analysis. Quant. Struct. Act. Relatsh 12, 1–8 (1993). 63. Zhao, G. & London, E. An amino acid transmembrane tendency scale that approaches the theoretical limit to accuracy for prediction of transmembrane helices: relationship to biological hydrophobicity. Protein Sci. 15, 1987–2001 (2006). llb k b ld d l h l 64. Hellberg, S., Sjöström, M., Skagerberg, B. & Wold, S. Peptide quantitative structure–activity relationships, a multivariate approach. J. Med. Chem. 30, 1126–1135 (1987). 65. Sandberg, M., Eriksson, L., Jonsson, J., Sjöström, M. & Wold, S. New chemical descriptors relevant for the design of biologically active peptides. A multivariate characterization of 87 amino acids. J. Med. Chem. 41, 2481–2491 (1998). p p ( ) 66. Kawashima, S. References & Jernigan, R. L. Estimation of effective interresidue contact energies from protein crystal structures: quasi- chemical approximation. Macromolecules 18, 534–552 (1985). pp ( ) 80. Guy, H. R. Amino acid side-chain partition energies and distribution of residues in soluble proteins. Biophys. J. 47, 61–70 (1 80. Guy, H. R. Amino acid side-chain partition energies and distribution of residues in soluble proteins. Biophys. J. 47, 61–70 (1985). 81 Meek J L Prediction of peptide retention times in high pressure liquid chromatography on the basis of amino acid composi 80. Guy, H. R. Amino acid side-chain partition energies and distribution of residues in soluble proteins. Biophys. J. 47, 61–70 (1985). 81. Meek, J. L. Prediction of peptide retention times in high-pressure liquid chromatography on the basis of amino acid composi- 80. Guy, H. R. Amino acid side-chain partition energies and distribution of residues in soluble proteins. Biophys. J. 47, 61–70 (1985). 81. Meek, J. L. Prediction of peptide retention times in high-pressure liquid chromatography on the basis of amino acid composi- tion Proc Natl Acad Sci USA 77 1632 1636 (1980) 81. Meek, J. L. Prediction of peptide retention times in high-pressure liquid chromatography on the basis of amino acid composi- tion. Proc. Natl. Acad. Sci. USA 77, 1632–1636 (1980). 82. Parker, J. M., Guo, D. & Hodges, R. S. New hydrophilicity scale derived from high-performance liquid chromatography peptide retention data: correlation of predicted surface residues with antigenicity and X-ray-derived accessible sites. Biochemistry 25, 5425–5432 (1986). ( ) 83. Klein, P., Kanehisa, M. & DeLisi, C. Prediction of protein function from sequence properties. Discriminant analysis of a data base. Biochim. Biophys. Acta (BBA) Protein Struct. Mol. 787, 221–226 (1984). , C. R. Evolution of the genetic code. Die Nat. 60, 447–459 (1973) g 85. Radzicka, A. & Wolfenden, R. Comparing the polarities of the amino acids: side-chain distribution coefficients between the vapor phase, cyclohexane, 1-octanol, and neutral aqueous solution. Biochemistry 27, 1664–1670 (1988). h h Th l d d f d h d h b Th l 86. Charton, M. & Charton, B. I. The structural dependence of amino acid hydrophobicity parameters. J. Theor. Biol. 99, 629–644 (1982). 87. Fauchère, J. L., Charton, M., Kier, L. B., Verloop, A. & Pliska, V. Amino acid side chain parameters for correlation studies in biology and pharmacology. Int. J. Pept. Protein Res. 32, 269–278 (1988). gy gy p 88. Krigbaum, W. R. & Komoriya, A. References et al. AAindex: amino acid index database, progress report 2008. Nucleic Acids Res. 36, D202-5 (2008). , , p g p , ( ) 67. Monné, M., Hermansson, M. & Von Heijne, G. A turn propensity scale for transmembrane helices. J. Mol. Biol. 288, 141–145 (1999). 68. Aurora, R. & Rose, G. D. Helix capping. Protein Sci. 7, 21–38 (1998). 68. Aurora, R. & Rose, G. D. Helix capping. Protein Sci. 7, 21–38 (1998). 69. Qian, N. & Sejnowski, T. J. Predicting the secondary structure of globular proteins using neural network models. J. Mol. Biol 202, 865–884 (1988). 70. Mitaku, S., Hirokawa, T. & Tsuji, T. Amphiphilicity index index of polar amino acids as an aid in the characterization of am acid preference at membrane water interfaces Bioinformatics 18 608 616 (2002) 70. Mitaku, S., Hirokawa, T. & Tsuji, T. Amphiphilicity index index of polar amino acids as an aid in the characterization of amino acid preference at membrane-water interfaces. Bioinformatics 18, 608–616 (2002). 71. Naderi-Manesh, H., Sadeghi, M., Arab, S. & Moosavi Movahedi, A. A. Prediction of protein surface accessibility with informa tion theory. Proteins: Struct. Funct. Genet. 42, 452–459 (2001). 72. Fauchere, J.-L. & Pliska, V. Hydrophobic parameters pi of amino-acid side chains from the partitioning of N-acetyl-amino-acid amides. Eur. J. Med. Chem 18, 369–375 (1983).h ( ) 73. Eisenberg, D. Three-dimensional structure of membrane and surface proteins. Annu. Rev. Biochem. 53, 595–623 (1984). 74. Ponnuswamy, P. K., Prabhakaran, M. & Manavalan, P. Hydrophobic packing and spatial arrangement of amino acid residues in globular proteins. BBA Protein Struct. 623, 301–316 (1980). g p ( ) 75. Wilce, M. C., Aguilar, M. I. & Hearn, M. T. Physicochemical basis of amino acid hydrophobicity scales: evaluation of four new scales of amino acid hydrophobicity coefficients derived from RP-HPLC of peptides. Anal. Chem. 67, 1210–1219 (1995). https://doi.org/10.1038/s41598-020-67701-3 Scientific Reports | (2020) 10:10869 | www.nature.com/scientificreports/ 76. Black, S. D. & Mould, D. R. Development of hydrophobicity parameters to analyze proteins which bear post- or cotranslationa modifications. Anal. Biochem. 193, 72–82 (1991). i ( ) 77. Eisenberg, D. & Mclachlan, A. D. Solvation energy in protein folding and binding. Nature 319, 199–203 (1986).fi 78. Pliška, V., Schmidt, M. & Fauchère, J. L. Partition coefficients of amino acids and hydrophobic parameters π of their side-chains as measured by thin-layer chromatography. J. Chromatogr. A 216, 79–92 (1981).f y y g y g 79. Miyazawa, S. References Is it possible to analyze DNA and protein sequences by the methods o digital signal processing?. IEEE Trans. Biomed. Eng. 32, 337–341 (1985).h 102. Veljkovic, V., Cosic, I., Dimitrijevic, B. & Lalovic, D. Is it possible to analyze DNA a digital signal processing?. IEEE Trans. Biomed. Eng. 32, 337–341 (1985). 103. Cosic, I. Macromolecular bioactivity: is it resonant interaction between macromolecules? Theory and applications. IEEE Trans Biomed. Eng. 41, 1101–1114 (1994).h g ( ) 104. Jacobs, R. E. & White, S. H. The nature of the hydrophobic binding of small peptides at the bilayer interface: implications for the insertion of transbilayer helices. Biochemistry 28, 3421–3437 (1989). 105. Wold, S. et al. Principal property values for six non-natural amino acids and their application to a structure–activity relationship for oxytocin peptide analogues. Can. J. Chem. 65, 1814–1820 (1987). 105. Wold, S. et al. Principal property values for six non-natural amino acids and their appl for oxytocin peptide analogues. Can. J. Chem. 65, 1814–1820 (1987). for oxytocin peptide analogues. Can. J. Chem. 65, 1814–1820 (1987). 106. Veltri, D., Kamath, U. & Shehu, A. Deep learning improves antimicrobial peptide recognition. Bioinformatics 34, 2740–2747 (2018) 106. Veltri, D., Kamath, U. & Shehu, A. Deep learning improves antimicrobial peptide recognition. Bioinformatics 34, 2740–2747 (2018).h 107. Thomas, P. D. & Dill, K. A. An iterative method for extracting energy-like quantities from protein structures. Proc. Natl. Acad Sci. USA 93, 11628–11633 (1996). 108. Shen, J. et al. Predicting protein-protein interactions based only on sequences information. Proc. Natl. Acad. Sci. USA 104 4337–4341 (2007).i 109. Ding, H., Feng, P. M., Chen, W. & Lin, H. Identification of bacteriophage virion proteins by the ANOVA feature selection and analysis. Mol. BioSyst. 10, 2229–2235 (2014). 110. Cao, D. S. et al. PyDPI: freely available python package for chemoinformatics, bioinformatics, and chemogenomics studies. J Chem. Inf. Model. 53, 3086–3096 (2013). f 111. Dong, J. et al. PyBioMed: a python library for various molecular representations of chemicals, proteins and DNAs and their interactions. J. Cheminformatics 10, 16 (2018). interactions. J. Cheminformatics 10, 16 (2018). 112. Cortes, C. Support-Vector Networks (Tech, Rep, 1995). J f , ( ) 112. Cortes, C. Support-Vector Networks (Tech, Rep, 1995). 112. Cortes, C. Support-Vector Networks (Tech, Rep, 1995). pp p 113. Ho, T. K. Random decision forests. In Proceedings of the International Conference on Document Analysis and Recognition, ICD vol. References Local interactions as a structure determinant for protein molecules: II. BBA Protein Struct 576, 204–228 (1979). 89. Goldsack, D. E. & Chalifoux, R. C. Contribution of the free energy of mixing of hydrophobic side chains to the stability of the tertiary structure of proteins. J. Theor. Biol. 39, 645–651 (1973).h y ph 90. Tsai, J., Taylor, R., Chothia, C. & Gerstein, M. The packing density in proteins: standard radii and volumes. J. Mol. Biol. 290 253–266 (1999). 91. Pontius, J., Richelle, J. & Wodak, S. J. Deviations from standard atomic volumes as a quality measure for protein crystal structures J. Mol. Biol. 264, 121–136 (1996). 92. Harpaz, Y. ., Gerstein, M. . & Chothia, C. . Volume changes on protein folding. Structure 2, 641–649 (1994).h 93. Charton, M. Protein folding and the genetic code: an alternative quantitative model. J. Theor. Biol. 91, 115–123 (1981). h k d f h f d f l b l b l h d 94. Nishikawa, K. & Ooi, T. Prediction of the surface-interior diagram of globular proteins by an empirical method. Int. J. Pept Protein Res. 16, 19–32 (1980). 95. Nishikawa, K. & Ooi, T. Radial locations of amino acid residues in a globular protein: correlation with the sequence. J. Biochem 100, 1043–1047 (1986).f 96. Meirovitch, H., Rackovsky, S. & Scheraga, H. A. Empirical studies of hydrophobicity. 1. Effect of protein size on the hydrophobic behavior of amino acids. Macromolecules 13, 1398–1405 (1980). 97. Janin, J. Surface and inside volumes in globular proteins [20]. Nature 277, 491–492 (1979). Janin, J. Surface and inside volumes in globular proteins [20]. N 97. Janin, J. Surface and inside volumes in globular proteins [20]. Nature 277, 491–492 (1979). 98. Punta, M. & Maritan, A. A knowledge-based scale for amino acid membrane propensity. Proteins: Struct. Funct. Genet. 50 114–121 (2003).f 99. Zhou, H. & Zhou, Y. Quantifying the effect of burial of amino acid residues on protein stability. Proteins: Struct. Funct. Genet 54, 315–322 (2004). 100. Oobatake, M., Kubota, Y. & Ooi, T. Optimization of Amino Acid Parameters for Correspondence of Sequence to Tertiary Structures of Proteins. Tech. Rep. 2 (1985). 101. Warme, P. K. & Morgan, R. S. A survey of amino acid side-chain interactions in 21 proteins. J. Mol. Biol. 118, 289–304 (1978) g y 102. Veljkovic, V., Cosic, I., Dimitrijevic, B. & Lalovic, D. Scientific Reports | (2020) 10:10869 | References A comparison of neural network and expert systems algorithms with common multivariate procedures for analysis of social science data. Soc. Sci. Comput. Rev. 9, 399–434 (1991). i 26. Garson, G. D. A comparison of neural network and expert syst of social science data. Soc. Sci. Comput. Rev. 9, 399–434 (1991 p 27. Benetti, S., Timmons, P. B. & Hewage, C. M. NMR model structure of the antimicrobial peptide maximin 3. Eur. Biophys. J. 48 203–212 (2019). ( ) 28. Timmons, P. B., O’Flynn, D., Conlon, J. M. & Hewage, C. M. Structural and positional studies of the antimicrobial peptide brevinin-1BYa in membrane-mimetic environments. J. Pept. Sci. 25, e3208 (2019). 128. Timmons, P. B., O’Flynn, D., Conlon, J. M. & Hewage, C. M. Structural and positional st brevinin-1BYa in membrane-mimetic environments. J. Pept. Sci. 25, e3208 (2019). p 129. Timmons, P. B., O’Flynn, D., Conlon, J. M. & Hewage, C. M. Insights into conformation and membrane interactions of the acyclic and dicarba-bridged brevinin-1BYa antimicrobial peptides. Eur. Biophys. J. 48, 701–710 (2019). 129. Timmons, P. B., O’Flynn, D., Conlon, J. M. & Hewage, C. M. Insights into conformation and membrane inter acyclic and dicarba bridged brevinin 1BYa antimicrobial peptides Eur Biophys J 48 701 710 (2019) y g p y 130. Dawson, R. M. & Liu, C. Q. Properties and applications of antimicrobial peptides in biodefense against biological warfare threat agents. Crit. Rev. Microbiol. 34, 89–107 (2008). g 31. Chen, Y. et al. Role of peptide hydrophobicity in the mechanism of action of α-helical antimicrobial peptides. Antimicrob. Agents Chemother. 51, 1398–1406 (2007). 132. Hollmann, A. et al. Role of amphipathicity and hydrophobicity in the balance between hemolysis and peptide-membrane interactions of three related antimicrobial peptides. Colloids Surf. B Biointerfaces 141, 528–536 (2016). p p f f 133. Lai, R. et al. Antimicrobial peptides from skin secretions of Chinese red belly toad Bombina maxima. Peptides 23, 427–435 (2002).f 34. Jiang, Z. et al. Effects of net charge and the number of positively charged residues on the biological activity of amphipathic alpha-helical cationic antimicrobial peptides. Adv. Exp. Med. Biol. 611, 561–562 (2009). Acknowledgementsh g The authors would like to thank Prof. Denis Shields, Dr. Gianluca Pollastri and Dr. Manaz Kaleel for enlighten- ing conversations. The authors would also like to thank University College Dublin for the Research Scholarship granted to P.B.T. Finally, this work is dedicated to the memory of Janina Jarońska, who inspired a love of learning and a thirst for knowledge. References 1 of ICDAR ’95, 278–282 (IEEE Computer Society, Washington, DC, USA, 1995).i 114. Pearson, K. L. I. I. I. On lines and planes of closest fit to systems of points in space. Lond. Edinb. Dublin Philos. Mag. J. S 559–572 (1901). 115. Van Der Maaten, L. & Hinton, G. Visualizing data using t-SNE. J. Mach. Learn. Res. 9, 2579–2625 (2008).h 116. White, B. W. & Rosenblatt, F. Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms Vol. 76 (Spartan Books, New York, 1963). 117. Brank, J., Grobelnik, M., Milić-Frayling, N. & Mladenić, D. Feature selection using support vector machines. Tech. Rep. MSR- TR-2002-63 (2002).ft 118. Ioffe, S. & Szegedy, C. Batch normalization: accelerating deep network training by reducing internal covariate shift. 32nd Inter‑ national Conference on Machine Learning, ICML 2015 1, 448–456. arXiv​:1502.03167​ (2015) 119. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014). i 120. Abadi, M. et al. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems (2016). arXiv​:1603.04467​. https://doi.org/10.1038/s41598-020-67701-3 Scientific Reports | (2020) 10:10869 | www.nature.com/scientificreports/ 121. Kingma, D. P. & Ba, J. L. Adam: A method for stochastic optimization. 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings (2015). arXiv​:1412.6980. f g 122. Conlon, J. M. et al. Isolation of peptides of the brevinin-1 family with potent candidacidal activity from the skin secretio the frog Rana boylii. J. Pept. Res. 62, 207–213 (2003). g y J p ( ) 123. Chaudhary, K. et al. A web server and mobile app for computing hemolytic potency of peptides. Sci. Rep. 6, 22843 (2016) 123. Chaudhary, K. et al. A web server and mobile app for com 124. Win, T. S. et al. HemoPred: a web server for predicting the hemolytic activity of peptides. Future Med. Chem. 9, 275–291 (2 p g y y p p 25. Kumar, V., Kumar, R., Agrawal, P., Patiyal, S. & Raghava, G. P. A method for predicting hemolytic potency of chemically modi fied peptides from its structure. Front. Pharmacol. 11, 54 (2020). 125. Kumar, V., Kumar, R., Agrawal, P., Patiyal, S. & Raghava, G. P. A m fied peptides from its structure. Front. Pharmacol. 11, 54 (2020). i p p 26. Garson, G. D. Competing interests h p g The authors declare no competing interests. Author contributions P.B.T. assembled the dataset, built the feature description and machine learning programmes, trained the machine learning algorithms, created all figures and tables and created the web interface. P.B.T. and C.M.H. determined the scope of the experiments and wrote the manuscript. Both authors contributed to and have approved the final manuscript. Additional information Supplementary information is available for this paper at https​://doi.org/10.1038/s4159​8-020-67701​-3. Supplementary information is available for this paper at https​://doi.org/10.1 Correspondence and requests for materials should be addressed to C.M.H. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and nstitutional affiliations. Open Access  This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. © The Author(s) 2020 https://doi.org/10.1038/s41598-020-67701-3 Scientific Reports | (2020) 10:10869 |
https://openalex.org/W2247151075
https://mech.novtex.ru/jour/article/download/215/156
Russian
null
Specific Features of Designing of the Robotic Systems for the Restorative Medicine
Mehatronika, avtomatizaciâ, upravlenie
2,015
cc-by
6,139
1 Pабота выпоëнена пpи поääеpжке ãpанта Пpезиäента PФ МК-2511.2014.8. Особенности пpоектиpования pобототехнических систем для восстановительной медицины1 Pассматpиваются необходимость пpименения pобототехнических систем для восстановительной медицины и особенности пpоектиpования этих систем. Отмечается пpиоpитет отечественных pазpаботок в области аппаpатных сpедств восста- новительной медицины и pост интеpеса к ним в миpе. Подpобно описываются технические тpебования к паpаметpам и узлам pобототехнической системы для восстановительной медицины: pабочей зоне, pазвиваемым усилиям, скоpостям, точности, по- гpешности, системе упpавления, интеpфейсу упpавления, инстpументам. Отмечается, что пpи пpоектиpовании необходимо учитывать вопpосы безопасности, экономической эффективности, взаимодействия pобота и человека-пациента и психоло- гического воспpиятия pобота человеком-пациентом. Пpиводится пpимеp технических тpебований к pазpаботке pобототех- нической системы для восстановительной медицины. В заключении суммиpуются особенности пpоектиpования pобототехни- ческих систем для восстановительной медицины. Ключевые слова: pобототехническая система, восстановительная медицина, биомехатpонный модуль, технические тpе- бования, жизненный цикл, эpгономика Мехатроника, автоматизация, управление, Том 16, № 10, 2015 УДК 616-085:681.5 DOI: 10.17587/mau.16.664-671 В. Ф. Головин, канд. техн. наук, зав. лаб. "Pобототехника", medicalrobot@mail.ru, М. В. Аpхипов, канд. техн. наук, доц., Московский госудаpственный индустpиальный Унивеpситет, Москва, Pоссия, В. Е. Павловский, д-р физ.-мат. наук, пpоф., vlpavl@mail.ru, Институт пpикладной математики им. М. В. Келдыша PАН, Москва, Pоссия Особенности пpоектиpования pобототехнических систем для восстановительной медицины1 Введение ВМ, на пеpвый пëан пpи этоì выхоäят вопpосы соз- äания и испоëüзования спеöиаëизиpованных pоботов. ВМ, на пеpвый пëан пpи этоì выхоäят вопpосы соз- äания и испоëüзования спеöиаëизиpованных pоботов. В настоящее вpеìя появëяþтся спеöиаëизиpован- ные аппаpатные сpеäства (вкëþ÷ая ìассажные кpесëа), котоpые, оäнако, зна÷итеëüно уступаþт возìожностяì ÷еëове÷еской pуки [5]. Известны не- котоpые иссëеäоватеëüские pазpаботки на Запаäе [3]. Как вывоä ìожно отìетитü, ÷то необхоäиìы PТС äëя ВМ, спpоектиpованные с у÷етоì потpебностей зäpавоохpанения и возìожностей совpеìенной pо- бототехники. Госуäаpственноãо заказа на пpоекти- pование PТС äëя ВМ пока нет, но пpовоäятся по- исковые pаботы и созäаþтся необхоäиìые заäеëы. Pис. 1. Взаимодействие между компонентами эpгатической сис- темы "вpач — pобот — пациент" Пpиоpитет иäеоëоãии pазpаботки унивеpсаëü- ных ìеäиöинских pоботов äëя ВМ пpинаäëежит Pоссии. Существует патент на способ и устpойство äëя выпоëнения ìассажа с пpиìенениеì ìанипу- ëяöионноãо pобота [8]. жения коне÷ностей ÷еëовека (паöиента) в суставах пpи взаиìоäействии их с pоботоì [5]. В основноì äаëее буäет pассìатpиватüся пpоектиpование PТС, выпоëняþщих ìассажные возäействия. Отìетиì, ÷то ìассаж, как оäин из основных эëеìентов ВМ, явëяется эффективныì и øиpокоäоступныì сpеä- ствоì восстановëения pаботоспособности пpи фи- зи÷ескоì утоìëении и наäежныì сpеäствоì пpофи- ëактики пpофессионаëüных забоëеваний. Массаж утоìëенных ìыøö не тоëüко восстанавëивает пеpво- на÷аëüнуþ ìыøе÷нуþ pаботоспособностü, но äаже увеëи÷ивает ее. Повыøение pаботоспособности утоì- ëенных ìыøö набëþäается и пpи ìассаже ìыøö, не пpиниìавøих у÷астия в физи÷еской pаботе. жения коне÷ностей ÷еëовека (паöиента) в суставах пpи взаиìоäействии их с pоботоì [5]. В основноì äаëее буäет pассìатpиватüся пpоектиpование PТС, выпоëняþщих ìассажные возäействия. Отìетиì, ÷то ìассаж, как оäин из основных эëеìентов ВМ, явëяется эффективныì и øиpокоäоступныì сpеä- ствоì восстановëения pаботоспособности пpи фи- зи÷ескоì утоìëении и наäежныì сpеäствоì пpофи- ëактики пpофессионаëüных забоëеваний. Массаж утоìëенных ìыøö не тоëüко восстанавëивает пеpво- на÷аëüнуþ ìыøе÷нуþ pаботоспособностü, но äаже увеëи÷ивает ее. Повыøение pаботоспособности утоì- ëенных ìыøö набëþäается и пpи ìассаже ìыøö, не пpиниìавøих у÷астия в физи÷еской pаботе. Пpивеäеì нескоëüко пpиìеpов. Наибоëüøуþ заинтеpесованностü к pаботаì по pазpаботке pобо- тов äëя ìассажа пpоявëяþт у÷еные из Китая, ãäе ìетоäы ìассажной физиотеpапии pазвиты зна÷и- теëüно боëüøе, ÷еì на Запаäе. У÷итывая скоpостü освоения новой техники и pазвития пpоìыøëен- ноãо потенöиаëа, ìожно пpеäпоëожитü, ÷то Китай ìожет бытü пеpвыì пpоизвоäитеëеì pоботов äëя ìассажа [13]. Такой pобот pазpаботан в Бейäжин- скоì техноëоãи÷ескоì институте (pис. 2, сì. вто- pуþ стоpону обëожки). Пpеäпоëаãается, ÷то PТС не тоëüко выпоëняþт ìехани÷еские возäействия на паöиента, но оöени- ваþт взаиìоäействие с биоëоãи÷ескиìи МТ как объектаìи жизнеäеятеëüности ÷еëовека, äиаãности- pуя изìенения психофизиоëоãи÷ескоãо состояния паöиента и контpоëиpуя пpоöесс [5]. Упpавëение ìехани÷ескиìи взаиìоäействияìи буäет pассìатpи- ватüся как pеаëизаöия позиöионно-сиëовоãо упpав- ëения (ПСУ), а упpавëение психофизиоëоãи÷ескиì состояниеì паöиента — как биотехни÷еское упpав- ëение (БТУ). ВМ, на пеpвый пëан пpи этоì выхоäят вопpосы соз- äания и испоëüзования спеöиаëизиpованных pоботов. Чтобы поä÷еpкнутü актуаëüностü и эконоìи÷е- ские пpеиìущества испоëüзования pобототехники äëя восстановитеëüной ìеäиöины (ВМ), пpежäе всеãо необхоäиìо отìетитü, ÷то ìассаж и активные и пассивные äвижения коне÷ностей в суставах яв- ëяþтся пpовеpенныìи тыся÷еëетияìи озäоpови- теëüныìи неìеäикаìентозныìи сpеäстваìи и ÷то пока ëþäи поëüзуþтся этиìи сpеäстваìи в зна÷и- теëüно ìенüøей степени, ÷еì это иì необхоäиìо. Мноãие наpоäы ìиpа векаìи созäаваëи куëüтуpы озäоpовëения, ãäе веäущуþ pоëü и зна÷ение иìеëи äвижение и соответствуþщие спеöиаëизиpованные ìехани÷еские возäействия. К пpиìеpу, на Востоке äëя физи÷еской куëüтуpы хаpактеpно и иìеет осо- бое зна÷ение ÷еpеäование наãpужений и pазãpузок, конöентpаöий и pассëабëений, активноãо и пассив- ноãо отäыха. Дëя совpеìенноãо ÷еëовека особенно актуаëüна такая физи÷еская куëüтуpа, äëя котоpой хаpактеpны не тоëüко активные воëевые äвижения, но и пассивные äвижения, вкëþ÷ая pазнообpазный ìассаж, постизоìетpи÷ескуþ pеëаксаöиþ и äpуãие пpиеìы pеëаксаöии и ìобиëизаöии. Эффективныìи ìоãут бытü новые техноëоãии совìестноãо пpиìене- ния ìассажа, напpиìеp, с ìеëотеpапией, теpìоте- pапией, аpоìатеpапией, активныìи äвиженияìи [5]. Вообще пpоектиpованиþ техни÷еских систеì по- священ pяä спеöиаëüных pабот, существуþт ГОСТы на пpоектиpование техни÷еских систеì, известны конкpетные пpиìеpы пpоектиpования систеì [1]. Поäобная инфоpìаöия известна äëя ìеäиöинской техники [2] и pобототехники [3, 4]. Основой пpо- ектиpования как пpоöесса явëяется инноваöион- ная äеятеëüностü, коãäа кpоìе pазpаботки пpоекта pассìатpивается äовеäение изäеëия äо конкуpен- тоспособноãо виäа [1]. Пpоектиpование pобототех- ни÷еских систеì (PТС) äëя ВМ в основноì сëеäует пpинöипаì и ìетоäаì пpоектиpования техни÷еских систеì, но иìеет опpеäеëенные особенности. Особенности пpоектиpования PТС äëя ВМ свя- заны, в пеpвуþ о÷еpеäü, с объектоì упpавëения — живыìи биоëоãи÷ескиìи ìяãкиìи тканяìи ÷еëо- века, их изìеняþщиìися вязко-упpуãиìи свойст- ваìи, неинвазивныì хаpактеpоì ìехани÷ескоãо возäействия, а сëеäоватеëüно, с виäаìи инстpуìен- тов. Эти особенности пpоявëяþтся в кинеìатике ìанипуëятоpов и постpоении их пpивоäов. Опpеäеëиì кëасс pассìатpиваеìых äаëее PТС äëя ВМ. Это систеìы упpавëяеìоãо ìехани÷ескоãо неинвазивноãо взаиìоäействия pоботов с биоëоãи- ÷ескиìи ìяãкиìи тканяìи (МТ) ÷еëовека. Взаиìо- äействия ìоãут бытü как непосpеäственныìи ìа- нипуëяöияìи pобота на МТ в виäе pазнообpазноãо ìассажа [5] иëи пеpеìещений äат÷иков вäоëü теëа паöиента во вpеìя уëüтpазвуковой äиаãностики [6], так и косвенныìи возäействияìи, такиìи как äви- Сказанное пpивоäит к вывоäу о необхоäиìости созäания спеöифи÷еских техни÷еских сpеäств äëя Сказанное пpивоäит к вывоäу о необхоäиìости созäания спеöифи÷еских техни÷еских сpеäств äëя 1 Pабота выпоëнена пpи поääеpжке ãpанта Пpезиäента PФ МК-2511.2014.8. 664 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Pис. 1. Взаимодействие между компонентами эpгатической сис- темы "вpач — pобот — пациент" еìкой pаботы ìассажистов и ìануаëüных теpапев- тов пpи ìассовоì обсëуживании паöиентов. Вìесте с теì, pусский вpа÷ Н. Б. Забëуäовский [7] с÷итаë необхоäиìыì пpиìенение аппаpатных сpеäств ìассажа äëя äозиpования ìассажных возäействий. ВМ, на пеpвый пëан пpи этоì выхоäят вопpосы соз- äания и испоëüзования спеöиаëизиpованных pоботов. В öеëоì упpавëение PТС äëя ВМ яв- ëяется эpãати÷ескиì в тpехкоìпонентной систеìе "вpа÷ — pобот — паöиент", пpи÷еì паöиент, в от- ëи÷ие от техни÷ескоãо объекта, наä котоpыì пpо- воäит ìанипуëяöии pобот, пpоявëяет свои психо- физиоëоãи÷еские свойства (pис. 1) [7]. Дëя совìестной pаботы pобота и ÷еëовека, нахо- äящеãося в pабо÷ей зоне pобота, неìеöкой коìпа- нией Kuka быë pазpаботан pобот LWR (light weight robot), котоpый по сpавненииþ с äpуãиìи, иìеþ- щиìи такуþ же pабо÷уþ зону, pоботаìи иìеет бо- ëее ëеãкуþ pуку. Пpи сеìи степенях поäвижности и ãpузопоäъеìности, pавной 7 кã, pобот весит 22 кã. Pобот ìожет pаботатü без оãpажäения, и пpи пpи- косновении к неìу pуки ÷еëовека он ìяãко отво- äит своþ pуку в стоpону [11] (pис. 3, а, сì. втоpуþ стоpону обëожки). Такой же способностüþ безопасной pаботы без оãpажäения обëаäаþт pоботы äатской коìпании Universal Robots UR5 (pис. 3, б, сì. втоpуþ стоpону обëожки) и UR10 [12]. Заìетиì, ÷то в ка÷естве конкуpента ìанипуëяöи- онноìу pоботу ìожно pассìатpиватü уже упоìи- навøееся ìассажное кpесëо — безpукий pобот, вы- поëняþщий пpиеìы пpиäавëиваþщеãо ìассажа [9]. Массажные кpесëа иìеþт сëеäуþщий pяä пpеиìу- ществ пеpеä ìанипуëяöионныìи pоботаìи: Хаpактеpныìи äëя пpоектиpования техни÷еских систеì явëяþтся äва этапа НИP — pазpаботка техни- ÷ескоãо заäания и пpеäваpитеëüное пpоектиpова- ние, äва этапа ОКP — эскизное и техни÷еское пpо- ектиpование, затеì пpоизвоäство и экспëуатаöия. То же необхоäиìо и в пpоектиpовании PТС äëя ВМ. z боëüøая конöентpаöия необхоäиìых äвижений в ìаëоì объеìе pабо÷ей зоны; z оãpани÷енностü пpиìенения инстpуìентов äëя ìассажа. Существует зна÷итеëüное pазнообpазие выпус- каеìых сеpийно ìассажных кpесеë, оäнако функ- öионаëüно уступаþщих ìанипуëяöионныì pобо- таì. Пеpе÷исëиì тепеpü тpебования к такиì pоботаì. Пpиводы pобота. Чтобы обеспе÷итü необхоäиìые скоpости и усиëия, а также пëавностü пеpеìещений, пpеäпо÷титеëüно пpоектиpоватü pобот с эëектpопpи- воäаìи. Мощностü эëектpопpивоäов суставов äоëж- на бытü в пpеäеëах 50...100 Вт. Чтобы обеспе÷итü ìа- ëуþ инеpöионностü ìанипуëяöионной pуки, сëеäует pаспоëаãатü эëектpоäвиãатеëи пpивоäов не в звенüях ìанипуëятоpа, а вìесте с pеäуктоpаìи на поäвижноì основании. Связü пpивоäов со звенüяìи в этоì сëу÷ае осуществëяется ëеãкиìи тpосовыìи пеpеäа- ÷аìи. Констpукöии звенüев ìоãут бытü тpуб÷атыìи. Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Основания для pазpаботки PТС для ВМ z посаäка паöиента в кpесëо ãаpантиpует еãо оpи- ентаöиþ; Напpавëение pазpаботки PТС äëя ВМ возникëо как естественное pазвитие систеì äëя заìены тpуäо- z боëüøие усиëия за с÷ет ìаëых pы÷аãов в кpесëе. 665 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Кинеìатика pобота äëя выпоëнения äвижений коне÷ностей в суставах ìожет бытü pеаëизована на основе экзоскеëетона, коãäа звенüя pобота накëа- äываþтся на коне÷ности паöиента и оси суставов pобота совпаäаþт с осяìи суставов паöиента. В äpу- ãих кинеìати÷еских коìпоновках звенüя pобота ìоãут pаспоëаãатüся как pуки вpа÷а, нахоäящеãося напpотив паöиента [5]. Оäнако неäостатки ìассажных кpесеë в сpавне- нии с pукой pобота существенные: z ìенüøие ìобиëüностü и pазнообpазие äвижений; z невозìожностü ìассажа пеpеäней повеpхности теëа паöиента, в тоì ÷исëе ëиöа; z невозìожностü ìассажа пеpеäней повеpхности теëа паöиента, в тоì ÷исëе ëиöа; z невозìожностü сìещения то÷ек пpи то÷е÷ноì ìассаже; z невозìожностü захватных и öаpапатеëüных пpиеìов; Pабо÷ая зона pобота зависит не тоëüко от pазìе- pов у÷астка, на котоpоì выпоëняþтся пpоöеäуpы. Дëя паöиента ìоãут бытü необpеìенитеëüныìи не- котоpые установо÷ные пеpеìещения äаже пpи ис- поëüзовании унивеpсаëüноãо pобота. Pабо÷ая зона pобота ìожет бытü увеëи÷ена за с÷ет пpиìенения поäвижноãо стоëа, на котоpоì нахоäится паöиент, в ка÷естве оäной из степеней поäвижности pобота. z невозìожностü выпоëнения äвижений коне÷- ностей в суставах; z оãpани÷енностü пpиìенения инстpуìентов äëя ìассажа. Функциональные тpебования. Кинематика, пpиводы и инстpументы манипулятоpов для ВМ Также боëüøая повеpхностü теëа поäвеpãается äав- ëениþ пpи ãиäpоìассаже, напpиìеp в äжакузи. Оäнако äëя pяäа техник необхоäиìа конöентpаöия вниìания в небоëüøой обëасти теëа, а поэтоìу äостато÷но оäной pуки и оäноãо инстpуìента [5]. ìанипуëятоpа, способы упpавëения ìанипуëятоpоì и систеìой упpавëения PТС в öеëоì, способы обу÷ения. Спеöифи÷ескиìи тpебованияìи к PТС äëя ВМ явëяþтся: pазвиваеìые ìанипуëятоpоì уси- ëия, ÷исëо упpавëяеìых сиëовых коìпонент, по- ãpеøностü воспpоизвеäения сиëовых коìпонент, инстpуìенты äëя взаиìоäействия с паöиентоì, виä äиаãностики психофизиоëоãи÷ескоãо состояния паöиента, усиëенные тpебования безопасности. ìанипуëятоpа, способы упpавëения ìанипуëятоpоì и систеìой упpавëения PТС в öеëоì, способы обу÷ения. Спеöифи÷ескиìи тpебованияìи к PТС äëя ВМ явëяþтся: pазвиваеìые ìанипуëятоpоì уси- ëия, ÷исëо упpавëяеìых сиëовых коìпонент, по- ãpеøностü воспpоизвеäения сиëовых коìпонент, инстpуìенты äëя взаиìоäействия с паöиентоì, виä äиаãностики психофизиоëоãи÷ескоãо состояния паöиента, усиëенные тpебования безопасности. Пpиìеpно поëовину ìассажных пpиеìов пpеä- ставëяþт нажиìание на теëо, а äpуãуþ поëовину — äействия с захватоì и оттяãиваниеì ìяãких тканей. Нажиìные инстpуìенты pобота ìоãут иìитиpоватü pазëи÷ные ÷асти pук и ноã ÷еëовека: поäуøе÷ки боëüøоãо паëüöа, pебpа ëаäони, пëоскости ëаäони. Даëее pассìотpиì техни÷еские хаpактеpистики, свойственные PТС äëя ВМ. Задаваемые усилия. Зна÷ения заäаваеìых усиëий зависят от виäа пpоöеäуpы и ìассоãеоìетpи÷еских паpаìетpов паöиента, еãо психофизиëоãи÷ескоãо состояния. Пpи выжиìании и pазìинании ìыøö, напpиìеp споpтсìенов, пpикëаäывается усиëие поpяäка 200...500 Н, пpи поãëаживании усиëие не äоëжно пpевыøатü 1 Н. Поãpеøностü воспpоизве- äения заäанных усиëий пpи выпоëнении пpиеìов общеукpепëяþщеãо ãиãиени÷ескоãо ìассажа ìожет составëятü 15... 30 % от заäаваеìых усиëий. В пpиеìах то÷е÷ноãо ìассажа боëüøие зна÷ения уäеëüных усиëий — äавëений — äостиãаþтся за с÷ет ìаëой контактной пëощаäи инстpуìента. Такиì инстpуìентоì поëüзуется вpа÷, такой же инстpу- ìент закpепëяется на сиëовоì ìоäуëе pобота. Пpиеìы pастиpания, тpакöии с боëüøиì тpениеì ìожет выпоëнятü инстpуìент с øеpøавой повеpх- ностüþ, пpиеìы с ìиниìаëüныì тpениеì ìожно выпоëнятü pоëикаìи иëи со сìазкой, как в аpоìа- теpапии. То÷ностü воспpоизвеäения заäанной тpаектоpии пpи заäанноì усиëии ìожет обеспе÷иватüся сущест- вуþщиìи сеpийно выпускаеìыìи pоботаìи, иìеþ- щиìи поãpеøностü позиöиониpования ±0,1 ìì. В этоì сëу÷ае äëя оpãанизаöии ПСУ äостато÷но ìоäеpнизаöии пpоãpаììноãо обеспе÷ения. В äpу- ãоì боëее äеøевоì ваpианте позиöионный pобот ìожет иìетü пониженнуþ то÷ностü поpяäка ±1,0 ìì. Контактнуþ повеpхностü инстpуìента сëеäует покpыватü ìатеpиаëаìи со свойстваìи, бëизкиìи к ÷еëове÷еской коже. Опpеäеëенная коìфоpтностü в контакте с ÷еëове÷ескиì теëоì ìожет бытü äостиã- нута инстpуìентоì, поäоãpеваеìыì токоì, напpиìеp, с поìощüþ pезистоpа, вìонтиpованноãо в инстpуìент. Частü пpиеìов ìассажа и äвижений коне÷но- стей связана с захватываниеì иëи у÷астков ìяãких тканей, иëи коне÷ностей. Сpеäи ìассажных пpиеìов это оpäинаpное pазìинание, äвойной ãpиф, äвой- ное коëüöевое защипывание. Функциональные тpебования. Кинематика, пpиводы и инстpументы манипулятоpов для ВМ Инстpуìент pобота äëя воспpоизвеäения этих пpиеìов ìожет пpеä- ставëятü äвухпаëüöевый захват с контpоëеì усиëия пpижатия паëüöев, напpиìеp, ÷еpез пpоãpаììно устанавëиваеìое äавëение в пневìоöиëинäpе схвата. Pазвиваемые скоpости. Движения pобота пpи вы- поëнении пpоöеäуp ВМ äоëжны бытü пëавныìи, ÷то тpебует контуpноãо упpавëения. Максиìаëüные скоpости äвижений pуки пpи ìеханотеpапии не пpевыøаþт 0,5 ì/с. Пpи выпоëнении pеëаксаöи- онных пpиеìов, как пpавиëо, скоpости ìенüøе, ÷еì пpи выпоëнении возбужäаþщих пpиеìов. Система упpавления. Дëя обеспе÷ения наибоëü- øей эффективности выпоëнения пpоöеäуp ВМ не- обхоäиìа эpãати÷еская систеìа упpавëения, у÷иты- ваþщая психофизиоëоãи÷еские свойства паöиен- та-÷еëовека. Чтобы обеспе÷итü заäанные усиëия, пеpеìещения и скоpости, необхоäиìо ПСУ с у÷а- стиеì сиëовоãо обу÷ения, напpиìеp, с испоëüзо- ваниеì сиëовоãо ìоäуëя. Конкуpентоспособностü ìеäиöинскоãо pобота в сpавнении с известныìи аппаpатныìи сpеäстваìи ìожет бытü äостиãнута за с÷ет унивеpсаëüности, ìноãофункöионаëüности, в тоì ÷исëе, за с÷ет наве- øивания на pобот в ка÷естве инстpуìента сеpийно выпускаеìых фиpìенных насаäок. Это всевозìож- ные вибpатоpы, pоëики, иãоëü÷атые инстpуìенты, опти÷еские изëу÷атеëи, äиаãности÷еские пpибоpы, напpиìеp, по систеìе Фоëëя, изìеpяþщие кожное сопpотивëение, и äp. Все пеpе÷исëенные насаäки äоëжны пpижиìатüся сиëовыì ìоäуëеì pобота к теëу паöиента с заäанныì усиëиеì. Дëя pеаëизаöии БТУ необхоäиìы изìеpитеëи паpаìетpов психофизиоëоãи÷ескоãо состояния — эëектpокожноãо сопpотивëения и ìыøе÷ноãо то- нуса, в котоpых в наибоëüøей степени пpоявëяется äействие ìеханотеpапии. Интеpфейс опеpатоpа. Необхоäиìые усиëия и скоpости в пеpвуþ о÷еpеäü пpоãpаììно устанавëи- вает и пpи необхоäиìости изìеняет вpа÷, поëüзуясü опытоì пpеäыäущих сеансов с äанныì паöиентоì. Вpа÷ ìожет поëüзоватüся "советаìи" äиаãности÷е- ской систеìы pобота. На некотоpых этапах отëа- женных пpоöеäуp вpа÷ ìожет поëностüþ äовеpятü pоботу, пеpекëþ÷ая еãо в автоìати÷еский pежиì pаботы. В автоìати÷ескоì pежиìе pобот поääеp- живает те усиëия и скоpости, котоpые pекоìенäует äиаãности÷еский бëок, анаëизиpуþщий психофи- Функциональные тpебования. Кинематика, пpиводы и инстpументы манипулятоpов для ВМ Кинематические модели манипулятоpов. Поскоëüку pассìатpивается кëасс PТС äëя ВМ, то техни÷еские тpебования к кинеìатике ìанипуëятоpа опpеäеëя- þтся пpибëижениеì ìанипуëятоpа к ÷еëове÷еской pуке ìассажиста иëи ìануаëüноãо теpапевта. Из сообpажений биони÷ескоãо копиpования ÷еëове- ÷еской pуки уìестны ìанипуëятоpы с анãуëяpной кинеìатикой. Дëя обеспе÷ения пëавности поäхоäа инстpуìента pобота к паöиенту и ìяãкоãо взаиìоäействия с МТ пpивоä сиëовоãо ìоäуëя ìожет бытü пневìати- ÷ескиì. Так как повыøенной жесткости и то÷ности pо- ботов äëя ВМ не тpебуþтся, то кинеìатика pуки ìанипуëятоpа ìожет бытü посëеäоватеëüной, и нет необхоäиìости пpибеãатü к ìанипуëятоpаì с па- pаëëеëüной стpуктуpой. Инстpументы манипулятоpов. Так же как äëя ëþ- бой pобототехни÷еской систеìы, испоëнитеëüные функöии PТС äëя ВМ опpеäеëяþтся инстpуìентаìи, сìенныìи иëи закpепëяеìыìи на коне÷ноì звене ìанипуëятоpа. Оäниì из отëи÷итеëüных тpебова- ний, пpеäъявëяеìых к инстpуìентаì ìеäиöинских pоботов, явëяется стеpиëüностü. Инстpуìенты äоëж- ны бытü ëеãкосъеìныìи иëи автоìати÷ески заìе- няеìыìи с возìожностüþ их стеpиëизаöии. Осо- бенное вниìание этоìу уäеëяется в pоботах äëя пpовеäения хиpуpãи÷еских и ìаëоинвазивных опе- pаöий, напpиìеp в pоботе Da Vinci [3]. Хотя PТС äëя ВМ выпоëняþт неинвазивные пpоöеäуpы, тpе- бования к стеpиëüности сохpаняþтся. Дëя pоботов, пpеäназна÷енных äëя выпоëнения äëинных пpоäоëüных äвижений вäоëü теëа паöи- ента (pобот äëя øëейф-ìассажа [5]), уìестной бу- äет äекаpтовая кинеìатика. Гаììа PТС äëя ВМ ìожет охватыватü ìанипуëя- тоpы с ÷исëоì степеней поäвижности от оäноãо äо сеìи. К оäностепенныì ìанипуëятоpаì относятся пpостейøие спеöиаëизиpованные, напpиìеp, pо- бот äëя капиëëяpноãо ìассажа ãоëовы, выпоëняþ- щий поступатеëüные äвижения в напpавëении оси инстpуìента [5]. Двухстепенной ìанипуëятоp äëя øëейф-ìассажа выпоëняет öикëовыì способоì упpавëяеìые äви- жения вäоëü теëа паöиента, пpижиìаясü пpужиной к теëу паöиента. Основной особенностüþ ìеäиöинскоãо pобота äëя ìануаëüной теpапии и ìассажа явëяется ìеха- ни÷еский контакт pобота с паöиентоì. Этот контакт осуществëяется ÷еpез инстpуìент. Поэтоìу в техни- ках, воспpоизвоäящих pуки ÷еëовека, инстpуìент äоëжен иìитиpоватü контактные свойства ÷еëове- ÷еской pуки: упpуãостü, тепëоту, вëажностü, фpикöи- онные свойства (øеpоховатостü, ãëаäкостü, скоëüз- костü), кооpäинаöионные возìожности (ìноãо- паëüöевостü, способностü захватыватü). Шестистепенные ìанипуëятоpы pоботов, наи- боëее pаспpостpаненных äëя техноëоãи÷еских заäа÷, позвоëяþт в своей pабо÷ей зоне выпоëнятü пpо- стpанственное позиöиониpование и оpиентаöиþ инстpуìента. Шестистепенной ìанипуëятоp испоëü- зуется как основа пpототипа [9]. Оäнако обеспе÷е- ние необхоäиìой оpиентаöии в ëþбой то÷ке pабо- ÷ей зоны тpебует избыто÷ности ìанипуëятоpа. Сеìистепенной ìанипуëятоp испоëüзован в опыт- ноì обpазöе pобота äëя пеpеìещения äат÷ика пpи уëüтpазвуковой äиаãностике [6]. Оäновpеìенный контакт на боëüøой повеpхно- сти теëа ìожет обеспе÷итü ìноãоpукий pобот иëи pобот, иìеþщий äве кисти иëи нескоëüко паëüöев. Возìожно пpиìенение øиpокоãо эëасти÷ноãо, саìо- устанавëиваþщеãося инстpуìента. Это, в некотоpой 666 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 степени, иìитаöия pаботы нескоëüких ìассажистов. Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Безопасность pобототехники для восстановительной медицины В ìеäиöинской pобототехнике тpебования к безопасности явëяþтся наибоëее стpоãиìи, так как в боëüøой степени относятся не к опеpатоpу, пpо- фессионаëüно взаиìоäействуþщеìу с pоботоì, а к паöиенту, котоpый äоëжен бытü увеpен, ÷то о еãо безопасности позаботятся без еãо у÷астия [5]. Наибоëее опасной явëяется инвазивная ìеäи- öинская pобототехника, напpиìеp, хиpуpãи÷еская pобототехника, ãäе испоëнитеëüный эëеìент pобота выпоëняет пункöии, pезекöии по пpоãpаììныì тpа- ектоpияì [3]. Зäесü фатаëüной ìожет бытü оøибка на уpовне составëения пpоãpаììы иëи оøибка опеpа- тоpа пpи äистанöионноì пpовеäении опеpаöии. В этоì сëу÷ае безопасностü в боëüøей степени зави- сит не от pобота, а от ìастеpства, обу÷енности опе- pатоpа, на котоpоãо возëаãается ответственностü. В ìеäиöинской pобототехнике тpебования к безопасности явëяþтся наибоëее стpоãиìи, так как в боëüøой степени относятся не к опеpатоpу, пpо- фессионаëüно взаиìоäействуþщеìу с pоботоì, а к паöиенту, котоpый äоëжен бытü увеpен, ÷то о еãо безопасности позаботятся без еãо у÷астия [5]. Условия эксплуатации зиоëоãи÷еское состояние паöиента. Дëя уäобства настpойки и экспëуатаöии pобота äоëжен бытü pазpаботан эpãоноìи÷ный оконный интеpфейс по анаëоãии с окнаìи пpоãpаìì в ОС MS Windows. Pазpаботка интуитивно понятноãо интеpфейса явëя- ется заäа÷ей эpãоноìики. Интеpфейс äоëжен пpеäос- тавëятü возìожностü поëüзоватеëþ в ãpафи÷ескоì pежиìе настpаиватü сëеäуþщие паpаìетpы пpо- öеäуp: скоpостü выпоëнения, усиëия взаиìоäейст- вия, ÷исëо повтоpений, пеpекëþ÷ение в автоìати÷е- ский pежиì pаботы и äp. Необхоäиìо пpеäусìотpетü вывоä на ìонитоp текущих и пpеäыäущих паpаìет- pов психофизиоëоãи÷ескоãо состояния. Пуëüт обу÷е- ния и упpавëения äоëжен соäеpжатü эëеìенты пози- öионноãо и сиëовоãо обу÷ения и упpавëения pоботоì. PТС иìеет явные пpеиìущества пеpеä pу÷ной ÷еëове÷еской pаботой как неутоìиìая систеìа. За- ìеняя ìассажиста иëи ìануаëüноãо теpапевта, PТС äëя ВМ äоëжна обеспе÷иватü непpеpывнуþ безот- казнуþ pаботу, такуþ же, как пpоìыøëенные pо- боты, — по кpайней ìеpе 20 ÷ в сутки. Pесуpс pа- боты такой же, как у пpоìыøëенных pоботов. Кëиìати÷еские усëовия экспëуатаöии опpеäе- ëяþтся как äоìаøние усëовия иëи усëовия стаöио- наpов кëиник. Технические тpебования к PТС для ВМ Как пpавиëо, äëя техни÷еских систеì тpебования охватываþт ãабаpитные pазìеpы, ìассу, то÷ностü, наäежностü, энеpãопотpебëение [1]. Дëя PТС к этиì тpебованияì äобавëяþтся сëеäуþщие: кинеìати÷е- ская ìоäеëü ìанипуëятоpа, pабо÷ая зона ìанипуëя- тоpа, ÷исëо степеней поäвижности ìанипуëятоpа, типы пpивоäов ìанипуëятоpа, pазвиваеìые ìанипу- ëятоpоì усиëия, скоpости пеpеìещения инстpуìента 667 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 зиоëоãи÷еское состояние паöиента. Дëя уäобства настpойки и экспëуатаöии pобота äоëжен бытü pазpаботан эpãоноìи÷ный оконный интеpфейс по анаëоãии с окнаìи пpоãpаìì в ОС MS Windows. Pазpаботка интуитивно понятноãо интеpфейса явëя- ется заäа÷ей эpãоноìики. Интеpфейс äоëжен пpеäос- тавëятü возìожностü поëüзоватеëþ в ãpафи÷ескоì pежиìе настpаиватü сëеäуþщие паpаìетpы пpо- öеäуp: скоpостü выпоëнения, усиëия взаиìоäейст- вия, ÷исëо повтоpений, пеpекëþ÷ение в автоìати÷е- ский pежиì pаботы и äp. Необхоäиìо пpеäусìотpетü вывоä на ìонитоp текущих и пpеäыäущих паpаìет- pов психофизиоëоãи÷ескоãо состояния. Пуëüт обу÷е- ния и упpавëения äоëжен соäеpжатü эëеìенты пози- öионноãо и сиëовоãо обу÷ения и упpавëения pоботоì. Пpимеp технических тpебований пpи постpоении PТС для ВМ Назна÷ение: PТС пpеäназна÷ена äëя автоìати÷е- скоãо выпоëнения пpиеìов озäоpовитеëüноãо ìас- сажа äëя насеëения в кëиниках, фитнес-öентpах, веëнес-öентpах, в äоìаøних усëовиях. Наибоëее опасной явëяется инвазивная ìеäи- öинская pобототехника, напpиìеp, хиpуpãи÷еская pобототехника, ãäе испоëнитеëüный эëеìент pобота выпоëняет пункöии, pезекöии по пpоãpаììныì тpа- ектоpияì [3]. Зäесü фатаëüной ìожет бытü оøибка на уpовне составëения пpоãpаììы иëи оøибка опеpа- тоpа пpи äистанöионноì пpовеäении опеpаöии. В этоì сëу÷ае безопасностü в боëüøей степени зави- сит не от pобота, а от ìастеpства, обу÷енности опе- pатоpа, на котоpоãо возëаãается ответственностü. Выпоëняеìые пpиеìы озäоpовитеëüноãо ìассажа: поãëаживание, выжиìание, pазìинание, pастиpание, защипывание, вибpаöии с испоëüзованиеì сìен- ных насаäок-инстpуìентов. Состав коìпëекса: ìанипуëятоp, куøетка, набоp сìенных инстpуìентов, пуëüт упpавëения и обу÷е- ния, эëектpонный бëок пpивоäов, пеpсонаëüный коìпüþтеp с теpìинаëоì, бëок биотехни÷ескоãо упpавëения, бëок пневìопитания. Стpуктуpа pобота соäеpжит тpи контуpа с обpат- ныìи связяìи по поëожениþ, усиëиþ, биоìеäиöин- скиì пеpеìенныì. Поэтоìу исто÷никаìи опасных сбоев систеìы упpавëения этоãо pобота буäут, в пеpвуþ о÷еpеäü, pазpывы заìыканий указанных контуpов, отказы äат÷иков обpатных связей. Дpуãиì исто÷никоì опасных ситуаöий ìожет бытü непpа- виëüная оöенка заäания паpаìетpов пpоöеäуpы. Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Пpимеpы pазpаботки биомехатpонных модулей PТС для ВМ и специализиpованных медицинских pоботов для ВМ 7) испоëüзование систеìы техни÷ескоãо зpения äëя сpавнения текущеãо изобpажения тpаектоpии инстpуìента с обpазöовыì; В ка÷естве возìожных ваpиантов ниже pассìат- pиваþтся пpиìеpы pазpаботки биоìехатpонных си- ëовых ìоäуëей PТС äëя ВМ и pазpаботка спеöиаëи- зиpованных ìеäиöинских pоботов äëя ВМ. Эти иссëеäования и pазpаботки выпоëнены в Москов- скоì ãоpоäскоì инäустpиаëüноì унивеpситете. 8) виpтуаëüное пëаниpование пpоöеäуpы; 9) пpиìенение упpуãих суставов. Необхоäиìо вывоäитü на ìонитоp текущие зна- ÷ения ìехани÷еских и биоìеäиöинских пеpеìен- ных и выäеëятü обëастü откëонений, пpеäупpеж- äаþщуþ звуковыìи и световыìи сиãнаëаìи о поä- хоäе к зоне опасности. Чтобы äопоëнятü сеpийно выпускаеìые pоботы неìеäиöинскоãо назна÷ения äо pоботов, выпоë- няþщих пpоöеäуpы ВМ, пpеäëожен сиëовой био- ìехатpонный ìоäуëü (pис. 4, сì. втоpуþ стоpону обëожки), соäеpжащий äат÷ик усиëия, коìпенса- тоp и сìенный инстpуìент [5]. Некотоpые связи ìежäу коìпонентаìи систеìы обpазуþт заìкнутые петëи (коëüöа). Поэтоìу не- обхоäиìо обеспе÷итü устой÷ивостü этих контуpов. Pассìотpенные тpебования обеспе÷ения безо- пасности сëеäует у÷итыватü пpи pазpаботке аппа- pатноãо интеpфейса ìежäу вpа÷оì, паöиентоì и pоботоì, а также пpи pазpаботке языка. p pу [ ] Коìпенсатоp необхоäиì äëя äеìпфиpования äинаìи÷еских пеpеãpузок пpи взаиìоäействии pо- бота с МТ паöиента. Дат÷ик усиëия испоëüзуется äëя позиöионно-сиëовоãо упpавëения и как изìе- pитеëü ìыøе÷ноãо тонуса äëя äиаãностиpования психофизиоëоãи÷ескоãо состояния паöиента и БТУ. В констpукöии пассивноãо сиëовоãо ìоäуëя ìе- хани÷еская пpужина выпоëняëа pоëü коìпенсатоpа с постоянной поäатëивостüþ. Дëя изìенения поäат- ëивости сиëовоãо ìоäуëя и, сëеäоватеëüно, поäат- ëивости ìанипуëятоpа быë пpеäëожен активный сиëовой биоìехатpонный ìоäуëü. Коìпенсатоp необхоäиì äëя äеìпфиpования äинаìи÷еских пеpеãpузок пpи взаиìоäействии pо- бота с МТ паöиента. Дат÷ик усиëия испоëüзуется äëя позиöионно-сиëовоãо упpавëения и как изìе- pитеëü ìыøе÷ноãо тонуса äëя äиаãностиpования психофизиоëоãи÷ескоãо состояния паöиента и БТУ. Есëи сëеäоватü биони÷ескоìу поäхоäу в пpоекти- pовании pоботов äëя ВМ, то, в пеpвуþ о÷еpеäü, сëе- äует стаpатüся воспpоизвести тактику вpа÷а и äвиже- ния еãо pук пpи выпоëнении ìеханотеpапевти÷еских пpоöеäуp. Вìесте с теì, pобот ìожет выпоëнятü pяä новых пpиеìов, несвойственных ÷еëовеку. В констpукöии пассивноãо сиëовоãо ìоäуëя ìе- хани÷еская пpужина выпоëняëа pоëü коìпенсатоpа с постоянной поäатëивостüþ. Дëя изìенения поäат- ëивости сиëовоãо ìоäуëя и, сëеäоватеëüно, поäат- ëивости ìанипуëятоpа быë пpеäëожен активный сиëовой биоìехатpонный ìоäуëü. Пpи pазpаботке пеpвоãо унивеpсаëüноãо pобота, выпоëняþщеãо коìпëексы äвижений ìассажа и ìануаëüной теpапии, уìестно в ка÷естве основы воспоëüзоватüся сеpийно выпускаеìыìи pоботаìи неìеäиöинскоãо назна÷ения. Pасøиpение сеpвиса, необхоäиìоãо пpи обу÷ении pобота, а также повы- øение безопасности взаиìоäействия pобота с вpа÷оì и особенно с паöиентоì на пеpвых этапах pазpаботки возìожны с пpиìенениеì в ка÷естве базовоãо пpоìыøëенных pоботов типа UR-10 (стоиìостü pобота €25.000) иëи Kuka LWR (стои- ìостü pобота €100.000). Технические тpебования Технические тpебования Pазвиваеìые ìанипуëятоpоì усиëия . . . . . . . . . . . . . . . . . . 70 Н Pабо÷ая зона ìанипуëятоpа . . Сфеpа pаäиусоì 1000—2000 ìì Чисëо степеней поäвижности ìанипуëятоpа . . . . . . . . . . . . 6 Поãpеøностü позиöиониpова- ния ìанипуëятоpа . . . . . . . . . 0,1 ìì Макс. скоpостü инстpуìента . 0,5 ì/с Масса pуки ìанипуëятоpа (без пpивоäа 1-ãо сустава) . . . 3 кã Пpивоäы ìанипуëятоpа . . . . . С тpосовыìи пеpеäа÷аìи, пpи- воäы 2...6 суставов установëе- ны на поäвижноì основании, ìощностü эëектpоäвиãатеëей: 1...3 суставов — 40 Вт, 4...6 суставов — 20 Вт Дат÷ик усиëия . . . . . . . . . . . . Тpехкоìпонентный Инстpуìенты. . . . . . . . . . . . . Набоp сìенных стеpиëüных инстpуìентов äëя кажäоãо пpиеìа ìассажа Упpавëение pобота . . . . . . . . Позиöионно-сиëовое и био- техни÷еское Обу÷ение pобота . . . . . . . . . . С поìощüþ сиëовых то÷ек Вpеìя пpовеäения ìассажа . . 15...60 ìин Диаãностика . . . . . . . . . . . . . По эëектpо-кожноìу сопpо- тивëениþ Эëектpопитание. . . . . . . . . . . 220 В, 50 Гö Тpебования безопасности . . . Повыøенные Пpоãpаììное обеспе÷ение. . . Спеöиаëизиpованное на базе совpеìенных опеpаöионных систеì Эpãоноìика. . . . . . . . . . . . . . Работа с у÷етоì взаиìоäейст- вия тpех коìпонент эpãати÷е- ской систеìы Возìожныìи сpеäстваìи пpеäотвpащения опас- ности ìоãут бытü сëеäуþщие: 1) откëþ÷ение пpивоäов иëи отвеäение pуки pо- бота по сиãнаëу äат÷ика ìаксиìаëüноãо усиëия, котоpый ìожет поäкëþ÷атüся ÷еpез позиöионно- сиëовуþ систеìу упpавëения иëи непосpеäственно к питаниþ пpивоäов; 2) откëþ÷ение пpивоäов иëи отвеäение pуки pо- бота по биоìеäиöинскиì сиãнаëаì; 3) пpовеpка пpивоäов в на÷аëе пpоöеäуp, а также сиëовоãо äат÷ика в пpоãpаììных контpоëüных то÷ках; 4) выпоëнение пеpвых äвижений с ìиниìаëü- ныìи усиëияìи; 5) pазìещение äвиãатеëей пpивоäов не в pуке pобота, а на пëатфоpìе, ÷то обеспе÷ит ìаëуþ инеp- öионностü и ëеãкостü pуки; 6) ìяãкое неэëектpопpовоäное покpытие pуки и pаспоëожение на кисти и pуке pобота тактиëüных äат÷иков касания; 668 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Заключение URL: http://www.kuka-labs.com/en/medical_ robotics/lightweight_robotics/ (äата обpащения: 22.01.2015). 12. Pоботы UR. URL: http://fam-robotics.ru/product/ur5 (äата обpащения: 22.01.2015). 12. Pоботы UR. URL: http://fam-robotics.ru/product/ur5 (äата обpащения: 22.01.2015). Пpеäëоженные в статüе тpебования к пpоекти- pованиþ pоботов äëя ВМ ìоãут бытü основой äëя составëения техни÷еских заäаний на pазpаботку. 13. Pобот äëя ìассажа Бейäжинãскоãо техноëоãи÷ескоãо ин- ститута. URL: http://www.youtube.com/watch?v=8LBFUaylbnc# t=40 (äата обpащения: 22.01.2015). 13. Pобот äëя ìассажа Бейäжинãскоãо техноëоãи÷ескоãо ин- ститута. URL: http://www.youtube.com/watch?v=8LBFUaylbnc# t=40 (äата обpащения: 22.01.2015). Дëя уто÷нения техни÷ескоãо заäания необхо- äиìо совìестное у÷астие в НИP техни÷еских и ìе- äиöинских спеöиаëистов. 14. Колобов А. А., Кочетов В. В., Омельченко И. Н. и äp. Эко- ноìика инноваöионной äеятеëüности наукоеìких пpеäпpи- ятий. М.: Изä-во МГТУ иì. Бауìана, 2007. 383 с. Пpимеpы pазpаботки биомехатpонных модулей PТС для ВМ и специализиpованных медицинских pоботов для ВМ Констpукöия активноãо сиëовоãо ìоäуëя вкëþ- ÷ает ëинейный пpивоä, изìенение коэффиöиента усиëения котоpоãо вызывает изìенение еãо поäат- ëивости. Тензоäат÷ик испоëüзуется как ÷увстви- теëüный эëеìент систеì позиöионно-сиëовоãо и биотехни÷ескоãо упpавëения. Мноãосуставный pобот ìожет выпоëнятü ìассаж на нескоëüких у÷астках теëа и пеpеìещатü коне÷но- сти вокpуã нескоëüких осей. Зна÷итеëüно äеøевëе pоботы с ìаëыì ÷исëоì суставов, особенно оäно- пpивоäные pоботы, оäнако они обëаäаþт ìенüøиìи возìожностяìи, ìоãут обсëуживатü ëиøü отäеëü- ные у÷астки и тоëüко опpеäеëенные коне÷ности вокpуã оäной оси. Остаëüные установо÷ные äвиже- ния выпоëняет паöиент, пеpеìещая новый у÷асток теëа в pабо÷уþ зону pобота. Пpи оöенке стоиìости pабот на НИОКP по пpо- екту "Pобототехника äëя восстановитеëüной ìеäиöи- ны" ìожно выäеëитü сëеäуþщие особенности в по- pяäке их зна÷иìости по этапаì пpоектиpования [14]: z основаниеì äëя бþäжетноãо финансиpования pазpаботки pобота äëя ВМ явëяется соöиаëüно зна÷иìая пpобëеìа общества — поääеpжание зäоpовüя насеëения; Несìотpя на своþ оãpани÷енностü, как автоìа- ти÷еские пpибоpы они буäут боëее эффективны, ÷еì pаспpостpаненные аппаpатные сpеäства ìеханиза- öии. Эффективностü таких pоботов ìожно повыситü пpи испоëüзовании в них äиаãности÷еских сpеäств, ìехатpонноãо поäхоäа, ПСУ, БТУ, нестанäаpтных äат÷иков и äвиãатеëей. z на÷аëо финансиpования pабот зависит от поëо- житеëüности экспеpтных оöенок pазpаботки pо- бота в виäе патентов, "ноу-хау"; z пpототип pобота с äеìонстpаöионныìи функ- öияìи ìожет бытü pеаëизован на базе унивеp- саëüноãо pобота неìеäиöинскоãо назна÷ения и закупаеìых äат÷иков (ìноãокоìпонентноãо си- ëовоãо, биоìеäиöинских паpаìетpов); Возìожные констpукöии таких pоботов описаны в ìоноãpафии [5], пpиìеpаìи явëяþтся pобот äëя косìети÷ескоãо то÷е÷ноãо ìассажа ëиöа, pобот с упpуãиì пpовоëо÷ныì инстpуìентоì äëя капиëëяp- ноãо ìассажа повеpхности ãоëовы (pис. 5, сì. вто- pуþ стоpону обëожки), pобот äëя øëейф-ìассажа повеpхности спины, pобот-ìаятник äëя выжиìа- ния äëинных ìыøö спины, pобот äëя выпоëнения пассивных со стоpоны паöиента äвижений pуки в öеëях pазpаботки контpактуp. z äаëüнейøие вëожения в совеpøенствование то- ваpноãо виäа pобота äëя ВМ пpевpащаþт еãо в коììеp÷еский пpоäукт, котоpый ìожет бытü пpеäìетоì äëя пеpехоäа к пеpеäа÷е техноëоãии на стаäиþ пpоизвоäства. 669 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Список литеpатуpы Заключение 1. Юpевич Е. И. Основы пpоектиpования техники: у÷еб. по- собие. СПб.: Изä. СПбГТУ, 2001. PТС äëя ВМ в зависиìости от заäа÷ пpеäстав- ëяþт øиpокий кëасс, в котоpоì унивеpсаëüные и спеöиаëизиpованные pоботы вхоäят как некото- pые отäеëüные ãpуппы. 2. Лощилов В. И., Шукин С. И. Пpинöипы анаëиза и син- теза биотехни÷еских систеì. М.: МВТУ иì. Н. Э. Бауìана, 1987. 67 с. Pазвитие pобототехники äëя ВМ в ìиpе в настоя- щее вpеìя поäтвеpжäается сеpийныìи выпускаìи спеöиаëизиpованных pоботов äëя ВМ. 3. Tavakoli M., Patel R. V., Moallem M., Aziminejad A. Haptics for teleoperated surgical robotic systems. Monograph series in the World scientific publishing under the title "New frontiers in robotics". Singapore: World Scientific Publishing Company, 2003, 290 p. Основаниеì äëя бþäжетноãо финансиpования pазpаботки PТС äëя ВМ явëяется соöиаëüно зна÷и- ìая пpобëеìа общества — поääеpжание зäоpовüя насеëения. 4. Савpасов Г. В. Меäиöинская pобототехника: состояние, пpобëеìы и общие пpинöипы пpоектиpования // Вестник МГТУ иì. Бауìана Н. Э. Спеöвыпуск "Биоìеäиöинская техни- ка и техноëоãия. Сеp. "Пpибоpостpоение". 1998. С. 35—50. 5. Golovin V., Zhuravlev V., Arkhipov M. Robotics in Restorative Medicine. LAP LAMBERT Academic Publishing, GmbH & Co. KG, 2012. 270 с Дëя PТС äëя ВМ тpебования безопасности в зна- ÷итеëüной степени pеøаþтся постpоениеì упpавëе- ния, позвоëяþщеãо pоботу pаботатü без оãpажäения. 6. Golovin V., Arhipov M., Juravlev V. Force Training for Po- sition/Force Control of Message Robots. Mechanisms and Machine Science. New Trends in Medical and Service Robots. Springer In- ternational Publishing. Switzerland. 2014. Vol. 20. P. 95—109. На пеpвоì этапе pазpаботки PТС äëя ВМ в ка- ÷естве базовых ìоãут бытü испоëüзованы иìпоpт- ные pоботы неìеäиöинскоãо назна÷ения Kuka LWR иëи UR-10. 7. Заблудовский В. И. Матеpиаëы к вопpосу о äействии ìас- сажа на зäоpовых ëþäей": äис. ... äокт. ìеä. наук. СПб.: 1982. С. 33—37. Есëи в основу пpоектиpования PТС äëя ВМ пpеäëаãается в ка÷естве базовоãо сеpийно выпускае- ìый pобот, то основныì аппаpатныì äопоëнениеì буäет биоìехатpонный сиëовой ìоäуëü, необхоäи- ìый и äëя ПСУ, и äëя БТУ. 8. Головин В. Ф., Самоpуков А. Е. Способ ìассажа и устpой- ство äëя еãо осуществëения. Патент PФ № 2145833, 08.05.1998. 9. Массажные кpесëа. URL: http://www.discountleatherchair. com/osaki-massage-recliners.html (äата обpащения: 22.01.2015). 9. Массажные кpесëа. URL: http://www.discountleatherchair. com/osaki-massage-recliners.html (äата обpащения: 22.01.2015). 10. Массажеpы äëя ãоëовы и теëа. URL: http://www.mas- sagerell.ru (äата обpащения: 22.01.2015). Оäниì из äопоëнитеëüных тpебований пеp- спективноãо техни÷ескоãо заäания явëяется pазpа- ботка спеöиаëüных автоìати÷ески заìеняеìых и стеpиëüных инстpуìентов. 11. Pоботы Kuka. URL: http://www.kuka-labs.com/en/medical_ robotics/lightweight_robotics/ (äата обpащения: 22.01.2015). 11. Pоботы Kuka. Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Кинематика манипулятоpа с упpавляемым изгибом на базе твеpдых элементов со сфеpической повеpхностью Pассматpивается новый класс манипулятоpов, постpоенных на базе звеньев с упpавляемым изгибом, фоpмиpуемых из твеp- дых элементов со сфеpической повеpхностью. Описана методика постpоения полной кинематической модели звена подобного манипулятоpа. Пpиводятся pезультаты компьютеpного моделиpования движения манипулятоpа (пpямая задача кинематики), постpоенного на базе одного, двух, тpех и чеpыpех звеньев, котоpые подтвеpждают адекватность постpоенных математи- ческих моделей с учетом пpинятых допущений. Получено pешение обpатной задачи кинематики для pабочей точки звена с уче- том его положения в pабочем пpостpанстве. манипулятоp с упpавляемым изгибом, кинематика, кинематический анализ, пpямая и обpатная задачи Ключевые слова: манипулятоp с упpавляемым изгибом, кинематика, кинематический анализ, пpямая и обpатная задачи DOI: 10.17587/mau.16.664-671 DOI: 10.17587/mau.16.664-671 8. Golovin V. F., Samorukov A. E. Sposob massazha i ustrojstvo dlja ego osushhestvlenija (Massage method and device for its implemen- tation), Ros.patent № 2145833 ot 08.05.1998 (in Russian). V. F. Golovin, medicalrobot@mail.ru , M. V. Arkhipov, Moscow State Industrial University, Moscow, 115280, Russian Federation, V. E. Pavlovskiy, vlpavl@mail.ru, Institute of Applied Mathematics, Moscow, 125047, Russian Federation V. F. Golovin, medicalrobot@mail.ru , M. V. Arkhipov, Moscow State Industrial University, Moscow, 115280, Russian Federation, V. E. Pavlovskiy, vlpavl@mail.ru, Institute of Applied Mathematics, Moscow, 125047, Russian Federation V. F. Golovin, medicalrobot@mail.ru , M. V. Arkhipov, Moscow State Industrial University, Moscow, 115280, Russian Federation, V. E. Pavlovskiy, vlpavl@mail.ru, Institute of Applied Mathematics, Moscow, 125047, Russian Federation Corresponding author: Golovin Vadim F., Chief of Robotics Laboratory, Moscow State Industrial University, Moscow, 115280, Russian Federation, e-mail: medicalrobot@mail.ru Corresponding author: Golovin Vadim F., Chief of Robotics Laboratory, Moscow State Industrial University, Moscow, 115280, Russian Federation, e-mail: medicalrobot@mail.ru Received on April 30, 2015 Accepted on May 15, 2015 Received on April 30, 2015 Accepted on May 15, 2015 The topic of this article is the role of the robotic systems in the restorative medicine and specific features of designing of such systems. The authors note the priority of the domestic developments in the area of the restorative medicine and the growing interest to them in the world. The specific features of the design of robotic systems for the restorative medicine are connected, first of all, with the live biological human soft tissues, their variable viscous-elastic properties, non-invasive nature of the me- chanical influence, and various tools. Presently the robotic systems appear, which encourages development of the service sys- tems of the public health care. The theoretical value of the work consists in presentation of the robotic systems for the restorative medicine with three components of the ergotic system, of which two components — physician and patient — are not technical objects but live intelligent subsystems. In this work the technical requirements for the parameters and units of the robotic systems for the restorative medicine are described, including the working area, developed efforts, speeds, accuracy, control systems, interface and tools. It is necessary to take into account the problems of safety, economic efficiency, interaction of robots and human-patients. The authors provide examples of technical requirements to the development of the robotic systems for the re- storative medicine. In conclusion, they summarize the specific features of the robotic systems for the restorative medicine. For citation: Golovin V., Arkhipov M., Pavlovskiy V. Specific Features of Desig- ning of the Robotic Systems for the Restorative Medicine, Mekhatro- nika, Avtomatizatsiya, Upravlenie, 2015, vol. 16, no. 10, pp. 664—671. 7. Zabludovskij V. I. Materialy к voprosu о dejstvii massazha na zdorovyh ljudej (Materials for the issue of the effect of massage for healthy people), Phd theses, SPb., 1882 (in Russian). DOI: 10.17587/mau.16.664-671 Д. P. Богданов, науч. сотp., dr_bog@mail.ru, О. В. Даpинцев, д-p техн. наук, зав. лаб., ovd@uimech.org Институт механики им. P. P. Мавлютова Уфимского научного центpа PАН, г. Уфа Кинематика манипулятоpа с упpавляемым изгибом на базе твеpдых элементов со сфеpической повеpхностью Д. P. Богданов, науч. сотp., dr_bog@mail.ru, О. В. Даpинцев, д-p техн. наук, зав. лаб., ovd@uimech.org, Институт механики им. P. P. Мавлютова Уфимского научного центpа PАН, г. Уфа V. F. Golovin, medicalrobot@mail.ru , M. V. Arkhipov, Moscow State Industrial University, Moscow, 115280, Russian Federation, V. E. Pavlovskiy, vlpavl@mail.ru, Institute of Applied Mathematics, Moscow, 125047, Russian Federation Keywords: robotic system, restorative medicine, bio-mechatronic modules, technical requirements, life cycle, ergonomics 670 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Мехатроника, автоматизация, управление, Том 16, № 10, 2015 Acknowledgements: This work was supported by the grant of the President of the Russian Federation MK-2511.2014.8 Acknowledgements: This work was supported by the grant of the President of the Russian Federation MK-2511.2014.8 6. Golovin V. F., Arhipov M. V., Zhuravlev V. E. Force Training for Position/Force Control of Message Robots. Mechanisms and Machine Science, New Trends in Medical and Service Robots, Springer International Publishing, Switzerland, 2014, vol. 20, pp. 95—109. References 9. Massazhnye kresla (Massage Chairs), available at: http:// www.discountleatherchair.com/osaki-massage-recliners.html (data of access: 22.01.2015). 1. Jurevich E. I. Osnovy proektirovanija tehniki (Basics of tech- nical systems designing), SPb., Publishing house of SPbGTU, 2001 (in Russian). 10. Massazhery dlja golovy i tela (Massage for the head and the body), available at: http://www.massagerell.ru (data of access: 22.01.2015). ( ) 2. Loshhilov V. I., Shukin S. I. Principy analiza i sinteza bioteh- nicheskih system (The principles of analysis and synthesis of biotech- nical systems), Mocsow, Publishing house of MVTU im. N. Je. Bau- mana, 1987, 67 p. (in Russian). 11. Roboty Kuka (Robots Kuka), available at: http://www.kuka- labs.com/en/medical_robotics/lightweight_robotics/ (data of access: 22.01.2015). 3. Tavakoli M., Patel R. V., Moallem M., Aziminejad A. Haptics for teleoperated surgical robotic systems, Monograph series in the World scientific publishing under the title "New frontiers in robotics", Singapore, World Scientific Publishing Company, 2003, 290 p. 12. Robots UR, available at: http://fam-robotics.ru/product/ur5 (data of access: 22.01.2015). 13. Robot dlja massazha Bejdzhingskogo tehnologicheskogo institute (Robot massage Beydzhing Institute of Technology), available at: http:// www.youtube.com/watch?v=8LBFUaylbnc#t=40 (data of access: 22.01.2015). 4. Savrasov G. V. Medicinskaja robototehnika: sostojanie, proble- my i obshhie principy proektirovanija (Medical robotics: state, problems and general design principles), Vestnik MGTU im. Baumana N. Je., Specvypusk "Biomedicinskaja Tehnika i Tehnologija, Ser. "Priboro- stroenie", 1998 (in Russian). 14. Kolobov A. A., Kochetov V. V., Omel’chenko I. N. Jekonomika innovacionnoj dejatel’nosti naukoemkih predprijatij (The economy of innovation high-tech enter), Moscow, Publishing house of MGTU im. Baumana, 2007 (in Russian). 5. Golovin V. F., Zhuravlev V. E., Arhipov M. V. Robotics in Re- storative Medicine, LAP LAMBERT Academic Publishing, GmbH & Co. KG, 2012, 270 с. УДК 621.865.8 DOI: 10.17587/mau.16.671-678 Введение енные на базе звенüев с упpавëяеìыì изãибоì [1]. Пpиìеp оäноãо из посëеäних на сеãоäняøний äенü pеøений по созäаниþ такоãо ìанипуëятоpа ìожно увиäетü в pаботе заpубежных у÷еных [2]. Вопpосы упpавëения поäобныìи ìанипуëятоpаìи освещены в оте÷ественной и заpубежной ëитеpатуpе не в поëной ìере, так как синтез систеì упpавëения äостато÷но сëожен и обусëовëен ìноãообpазиеì констpуктоpских pеøений таких ìанипуëятоpов, спеöифи÷ностüþ базовых пеpеìещений и сpеäств их pеаëизаöии. Особый интеpес пpеäставëяет ис- поëüзование ãибких ìанипуëятоpов в ìикpоpобото- технике, но такой ваpиант испоëüзования еще боëее усëожняет пpобëеìу синтеза, поскоëüку зна÷итеëüно увеëи÷ивается возìущаþщее возäействие сpеäы на В посëеäнее вpеìя зна÷итеëüно pасøиpяþтся обëасти пpиìенения pоботов. Так, ускоpиëисü pа- боты в сеpвисной и соöиаëüной pобототехнике, отìе÷ается увеëи÷ение ÷исëа теì по pоботаì äвой- ноãо назна÷ения, не остаëасü без вниìания иссëе- äоватеëей также теìа спеöтехники. Общей хаpактеp- ной ÷еpтой выøепеpе÷исëенных напpавëений явëяется необхоäиìостü pазpаботки новых ìанипу- ëяöионных схеì, ëиøенных основноãо неäостатка пpоìыøëенных ìанипуëятоpов — жесткости кон- стpукöии. Оäниì из пpеäставитеëей сеìейства не- типи÷ных констpукöий ìанипуëятоpов явëяþтся пëоские и пpостpанственные ìеханизìы, постpо- Мехатроника, автоматизация, управление, Том 16, № 10, 2015 671 Мехатроника, автоматизация, управление, Том 16, № 10, 2015
https://openalex.org/W1999101300
https://revistas.ucr.ac.cr/index.php/rbt/article/download/11136/10497
Spanish; Castilian
null
Especies arborescentes del género &lt;i&gt;Blechnum&lt;/i&gt; (Blechnaceae: Pteridophyta)
Revista de Biología Tropical
2,013
cc-by
19,734
Recibido 20-I-2012. Abstract: Arborescent species of the genus Blechnum (Blechnaceae: Pteridophyta). Although the genus Blechnum has been largely studied, problems of species characterization, nomenclature, classification and taxonomy remain unsolved. For these reasons, in this study we intend to deepen in the characterization of the arborescent group of Blechnum with a review of eight species: B. brasiliense, B. columbiense, B. cycadifolium, B. magellanicum, B. moritzianum, B. schomburgkii, B. tabulare and B. werckleanum.Type material and several collections from diverse herbaria were studied. For morphological studies we used both herbarium and fresh material collected by the authors in fieldwork between 2005 and 2010. Rhizomes, stipes, vascularization of cos- tae, as well as several characters of the lamina, such as outline, division, venation, margin structure, epidermal patterns, stomata, presence and types of hypodermis, and mesophyll of pinnae were studied. Important aspects of the morphology of the sporophyte have been analyzed, and external characters have been correlated with internal diagnostic ones to use the most reliable external traits in keys and descriptions. Taxonomy has been updated, and corrections have been introduced to the species nomenclature and comments were included to summarize the current knowledge of the geographical distribution and ecology. The presence of B. tabulare in America is con- firmed, and the taxonomy of the species is updated, since external and internal morphology indicated that other related, American and African species, such as B. spannagelii and B. madagascariense must be included under that name. Similar studies carried out on B. schomburgkii indicated that the name is a synonym of B. auratum subsp. auratum, while B. auratum subsp. columbiense is treated here as a well characterized species. Detailed study of numerous specimens allowed to characterize B. moritzianum, being the correct name to apply to the recently described B. yungense. A refreshed description of the species is presented together with a list of new, selected materials to illustrate current known geographical distribution. Expanded data on ecology, geographical distribution and affinities are given, and a key for species identification is presented. Rev. Biol. Trop. 61 (1): 377-408. Epub 2013 March 01. Key words: review, arborescent species, morphology, nomenclature, diagnostic characters. La bibliografía sobre el género Blechnum L. es muy abundante tanto en lo que respecta a las especies neotropicales como paleotropi- cales. Especies arborescentes del género Blechnum (Blechnaceae: Pteridophyta) istina H. Rolleri1, Carmen Prada2*, José María Gabriel & Galán3 & Lilian M. Passarelli4 Cristina H. Rolleri , Carmen Prada , José María Gabriel & Galán & Lilian M. Passarelli 1. Laboratorio de Estudios de Anatomía Vegetal Evolutiva y Sistemática (LEAVES), Facultad de Ciencias Naturales y Museo, 64 entre 120 y diagonal 113, B1904 DZB, Universidad Nacional de La Plata, Argentina. (2) Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET); tinar@speedy.com.ar 1. Laboratorio de Estudios de Anatomía Vegetal Evolutiva y Sistemática (LEAVES), Facultad de Ciencias Naturales y Museo, 64 entre 120 y diagonal 113, B1904 DZB, Universidad Nacional de La Plata, Argentina. (2) Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET); tinar@speedy.com.ar g y ( ) @ p y 2. Departamento de Biología Vegetal I, Facultad de Ciencias Biológicas, Universidad Complutense, Ciudad Universitaria, 28040 Madrid, España; cpm@bio.ucm.es 2. Departamento de Biología Vegetal I, Facultad de Ciencias Biológicas, Universidad Complutense, Ciudad Universitaria, 28040 Madrid, España; cpm@bio.ucm.es p p @ 3. Departamento de Biología Vegetal I, Facultad de Ciencias Biológicas, Universidad Complutense, Ciudad Universitaria, 28040 Madrid, España; jmgabriel@bio.ucm.es 3. Departamento de Biología Vegetal I, Facultad de Ciencias Biológicas, Universidad Complutense, Ciudad Universitaria, 28040 Madrid, España; jmgabriel@bio.ucm.es p j g @ 4. Laboratorio de Estudios de Anatomía Vegetal Evolutiva y Sistemática (LEAVES), Facultad de Ciencias Naturales y Museo, 64 entre 120 y diagonal 113, B1904 DZB, Universidad Nacional de La Plata, Argentina; lmpassarelli@yahoo com ar 4. Laboratorio de Estudios de Anatomía Vegetal Evolutiva y Sistemática (LEAVES), Facultad de Cienci Museo, 64 entre 120 y diagonal 113, B1904 DZB, Universidad Nacional de La Plata, Argentina; lmpassarelli@yahoo com ar 4. Laboratorio de Estudios de Anatomía Vegetal Evolutiva y Sistemática (LEAVES), Facultad de Ciencias Naturales y Museo, 64 entre 120 y diagonal 113, B1904 DZB, Universidad Nacional de La Plata, Argentina; lmpassarelli@yahoo.com.ar p @y * Correspondencia. * Correspondencia. Recibido 20-I-2012. Sin embargo, aún persisten problemas en la nomenclatura y tipificación de algunas especies, así como la falta de un estudio ana- tómico completo y una caracterización actua- lizada de su morfología externa e interna, taxonomía, citología y afinidades, algo que aplica especialmente para varias especies de Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 377 variación, sus afinidades y su área geográfica preferentemente andina. hábito arborescente con escamas rizomáticas aciculares. Por esta razón, se han selecciona- do para una caracterización y revisión, ocho especies del género: B. brasiliense Desv., B. columbiense Hieron., B. cycadifolium (Colla) Sturm, B. magellanicum (Kaulf.) Mett., B. moritzianum (Klotzsch) Moore, B. schombur- gkii (Klotzsch) C. Chr., B. tabulare (Thunb.) Kuhn y B. werckleanum (H. Christ) C. Chr., algunas poco conocidas. hábito arborescente con escamas rizomáticas aciculares. Por esta razón, se han selecciona- do para una caracterización y revisión, ocho especies del género: B. brasiliense Desv., B. columbiense Hieron., B. cycadifolium (Colla) Sturm, B. magellanicum (Kaulf.) Mett., B. moritzianum (Klotzsch) Moore, B. schombur- gkii (Klotzsch) C. Chr., B. tabulare (Thunb.) Kuhn y B. werckleanum (H. Christ) C. Chr., algunas poco conocidas. Blechnum tabulare se revisa en relación con la confirmación de su presencia en Sud- américa, ya citada previamente (Rolleri et al. 2008, Passarelli et al. 2010) y de este análisis se desprende que es coespecífica de B. mada- gascariense y B. spannagelii, dos especies poco estudiadas en obras recientes. También se analizan su variación, sus afinidades y su área geográfica, que en este caso incluye localidades selváticas, bordes de selva, áreas de bañados y pajonales inundados en áreas de selvas lluvio- sas de Brasil, Paraguay, Bolivia y en las selvas subtropicales del NO y NE de la Argentina. Blechnum auratum subsp. auratum fue el nombre con el que Tryon & Stolze (1993) recombinaron Lomaria aurata Fée, incluyen- do en su sinonimia a B. buchtienii Rosenst. En las claves propuestas por esos autores las diferencias de este taxón con B. schomburgkii no estaban establecidas con suficiente claridad. Blechnum columbiense, descrita por Hieron- ymus (1908) fue también recombinada como una subespecie de B. auratum (Fée) Tryon y Stolze (Tryon & Stolze, 1993), pero por el conjunto de sus caracteres se la trata aquí con rango de especie. Recibido 20-I-2012. Los caracteres analizados fueron rizomas, estípites y costas, en éstas especialmente la estructura interna y vascularización, caracteres de las láminas como contorno, división, presen- cia o ausencia de pinnas reducidas, aflebias o aguijones esclerosados procedentes de la atrofia temprana de aquéllas, estructura de los márge- nes, venación, modelos epidérmicos, presencia y tipos de hipodermis, mesofilo e indusios. Blechnum cycadifolium es una planta del Archipiélago de Juan Fernández, que por su distribución se considera endémica y no hay estudios recientes sobre la especie; convive en ambiente insular con B. magellanicum, que tam- bién crece en la patagonia argentino-chilena. Los caracteres de la hipodermis, una zona dermatoide adaxial, de espesor variable, que se distingue de la epidermis y del mesofilo en todas las especies estudiadas se describen e ilustran por primera vez. Los análisis de la hipodermis son novedosos para estas especies ya que los previos se refieren en general a gim- nospermas y angiospermas (Metcalfe & Chalk 1950, Esau 1953). Payne & Peterson (1973) son los únicos autores que integraron, en un estudio comparativo, la descripción de las hipodermis presentes en helechos leptosporan- giados, con una actualización de las propuestas de autores previos sobre su origen, como Foster (1936) y Bower (1923). La especie Blechnum moritzianum, basada en Lomaria moritziana Klotzsch, es el nombre que debe aplicarse al material descrito como B. yungense Ramos Giacosa (Ramos Giacosa, 2010). Aquí se la estudia en detalle, se analizan sus afinidades y diferencias con B. wercklea- num (H. Christ) C. Chr., descrita sobre material de Costa Rica y con B. schomburgkii. Las tres especies tienen una distribución amplia en los Andes sudamericanos y tanto B. moritzianum como B. werckleanum presentan un área geo- gráfica extensa no citada previamente. En trabajos anteriores sobre el género Blechnum (Rolleri & Prada 2006a, Prada et al. 2008, Rolleri et al. 2008, Gabriel y Galán et al. 2008) se han analizado varios de los rasgos citados para diferentes especies y se ha comprobado su valor diagnóstico en la carac- terización de los taxones y en la evaluación de sus afinidades. El alcance del valor diagnóstico de ciertos rasgos externos se puede certificar a Blechnum schomburgkii y B. tabulare son exhaustivamente analizados aquí en materiales de toda su área. Blechnum schomburgkii se considera coespecífica de B. auratum subsp. auratum y de B. buchtienii, y el nombre que tiene prioridad es el basado en Loma- ria schomburgkii Klotzsch. Recibido 20-I-2012. Se analizan su Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 378 arborescentes, de las que sólo trató B. schom- burgkii y B. spannagelii. través del conocimiento de su correlación con algunos caracteres anatómicos de fácil acceso. g Los estudios palinológicos amplios tam- bién son recientes en el género y escasos para estas especies. Tryon & Lugardon (1990) ilus- traron 39 especies de todo el género, incluyendo en esa obra las de varias arborescentes; Ramos Giacosa et al. (2009) analizaron con microsco- pía electrónica de transmisión, entre otras espe- cies del Noroeste de Argentina, las esporas de B. tabulare. Passarelli et al. (2010) efectuaron un análisis amplio utilizando microscopía elec- trónica de barrido, de 64 especies neotropicales y paleotropicales de Blechnum, incluyendo las de las especies arborescentes estudiadas aquí, proponiendo tipos morfológicos esporales y mostrando la importancia que tiene la orna- mentación del perisporio en la caracterización de especies y grupos de afinidad intragenéricos y con otros géneros de Blechnaceae. En este trabajo se proporcionan nuevos caracteres diagnósticos del esporófito y se revisan otros utilizados previamente con el fin de ponerlos a prueba en la caracterización de los taxones. Los nombres de algunas especies arborescentes se confunden frecuentemente en el material de herbario y esa es otra razón para revisar rasgos externos e internos, fácilmente ponderables en laboratorio, que conduzcan a su mejor determinación. Se elaboran descrip- ciones detalladas, con una actualización de la nomenclatura, tipificación y sinonimia, nuevos datos sobre ecología, distribución geográfica, aspectos adaptativos, afinidades de las especies y se propone una clave para la determinación de las especies. Materiales y Métodos rizomas adultos es de 20cm o más; carecen de estolones y son escamosos. Para estudios morfológicos se utilizó material de herbario y recolectado por los auto- res en trabajos de campo efectuados entre 2005 y 2010. El material de herbario se restauró con una mezcla de butil cellosolve y agua 1:4. Los fragmentos de pinnas se aclararon con hidróxi- do de sodio al 6% y se colorearon con azul de toluidina (TBO) al 1% acuoso. La venación y los modelos epidérmicos se analizaron en pin- nas basales, medias y apicales, pero se ilustran sólo la de las pinnas medias. Las dimensiones y densidad de los estomas se tomaron en pinnas medias de todos los ejemplares estudiados; las dimensiones se expresan en largo x ancho, en µm y la densidad en número de estomas por mm2. Los valores dados representan el prome- dio de 25 medidas por muestra. Estos datos se incluyen en el Cuadro 1. Toda la terminología relacionada con el esporófito se utiliza según Lellinger (2002). Se priorizan los términos pteridológicos sobre los angiospérmicos, tal como se explica en esa obra (Lellinger 2002), y se prefiere por lo tanto, rizoma, estípite y fronda en lugar de tallo, pecíolo y hoja. El término arborescente se utiliza tal como se indica en Font Quer (1982) y en Lellinger (2002), y no implica la existencia de un hábito arbóreo, derivado de la actividad cambial, sino que hace referencia al rizoma erecto y robusto que generalmente presentan estas especies. Los vocablos relacionados con la epidermis han sido utilizados por los autores en trabajos pte- ridológicos previos, como Prada et al. (2008) y Prada & Rolleri (2005) y la relacionada con los estomas se ha tomado de Prabhakar (2003). Las escamas del rizoma son lineares, cas- tañas, más o menos rígidas cuando secas y en las partes viejas del rizoma, concoloras o dis- coloras, con una zona central pluriestratificada que se torna de color castaño más oscuro que el resto y puede esclerosarse y con márgenes hialinos angostos o anchos, más claros, unies- tratificados, a veces algo irregulares o repan- dos, hasta 6cm de largo según las especies; se disponen cubriendo densamente el rizoma y la base de los estípites, tienden a curvarse todas en la misma dirección en plantas adultas de algunas especies como B. columbiense, B. moritzianum, B. schomburgkii o bien son más rectas y erguidas, algo retorcidas, como en B. Materiales y Métodos La mayoría de los taxones seleccionados fueron tratados en floras y catálogos, con rango taxonómico de especies o subespecies, por Hieronymus (1908), Schelpe (1952), Stolze (1981), Tryon & Tryon (1982), Tryon & Stolze (1993), Moran (1995), Rodríguez Ríos (1995), Dittrich (2005) y Rolleri & Prada (2006b). Se estudiaron colecciones del Herbario Nacional Colombiano, del Instituto de Cien- cias Naturales de la Universidad Nacional de Colombia (COL), del Museo de La Plata (LP), Instituto-Fundación Miguel Lillo (LIL), Real Jardín Botánico de Madrid (MA), University Herbarium, University of California, Berke- ley (UC), United States National Herbarium, Smithsonian Institution (US), Universidad Nacional Mayor de San Marcos (USM) y los tipos relacionados con especies arborescentes existentes en otros herbarios, tales como el her- bario de Berlín (Botanic Garden and Botanical Museum Berlin-Dahlem, Freie Universität Ber- lin, B) y el Herbario de París, Museo Nacional de Historia Natural (P). Tryon & Tryon (1982) propusieron sub- dividir el género en siete grupos informales basados en los caracteres de los ejes y las fron- das, aporte que suministró un primer paso en el ordenamiento de las especies, pero los grupos establecidos resultaron insuficientes o incom- pletos (Rolleri & Prada 2006a). De acuerdo con esos autores, algunas especies arborescentes y subarborescentes con escamas lineares fueron reunidas en el grupo B. buchtienii, que incluyó B. brasiliense, B. imperiale (Fée & Glaziou) H. Christ, B. insularum C. V. Morton & Lellinger, B. schomburgkii, B. spannagelii, B. underwoo- dianum (Broadh.) C. Chr., B. tabulare y B. werckleanum. Blechnum cycadifolium y B. magellanicum no se mencionaron entonces, aunque su fenotipo sugería su pertenencia al grupo. Dittrich (2005) renombró algunos grupos según la especie más antigua y reem- plazó la denominación de grupo B. buchtie- nii por grupo B. tabulare para las especies Las especies tratadas se citan con los basónimos y sinónimos correspondientes, se hacen comentarios en forma de observaciones, se resumen los datos actualizados de la distri- bución geográfica, aspectos ecológicos y se da una lista actualizada del material de respaldo seleccionado. En Blechnum magellanicum y B. tabulare, los sinónimos están en Rolleri et al. (2008) y sólo se incluyen sinónimos nue- vos derivados de este estudio. Los materiales de referencia de ambas especies son nuevos Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 379 y deben adicionarse a los previamente citados (Rolleri et al. 2008). Materiales y Métodos brasiliense y B. magellanicum. Las frondas son monomorfas en B. brasi- liense y B. columbiense, y dimorfas en las espe- cies restantes. Los estípites nacen más o menos aglomerados, abundantes o escasos, erguidos a divergentes, con un surco adaxial profundo flanqueado por dos zonas elevadas que corres- ponden a prominencias longitudinales más o menos afiladas, apenas salientes en B. colum- biense y extremadamente sobresalientes y cur- vas en otras especies, como B. cycadifolium, B. magellanicum y B. schomburgkii, mientras que en la porción abaxial son curvos. En general son rígidos y duros, debido a una zona fibrosa de varias capas de espesor, presente por debajo de la epidermis, que es papilosa; internamente son dictiostélicos, con 2-4 haces adaxiales más grandes que los abaxiales, éstos en un número variable de meristelas desde 5-6 hasta 10. Los raquis son lisos y glabros o escamo- sos, pero en general, esta condición es irregular porque en frondas adultas las escamas suelen ser caducas. Resultados Las costas son marcadamente carenadas en la cara abaxial en B. brasiliense, B. colum- biense, B. cycadifolium y B. schomburgkii, moderadamente curvas en B. magellanicum, B. moritzianum, B. tabulare y B. werckleanum y con un surco adaxial que es más profundo en las especies con costas carenadas, con excepción de B. brasiliense (Fig. 1A-B). Son En las especies estudiadas las plantas tienen gran porte y alcanzan una altura que va desde 180 hasta 300cm. Los rizomas son ergui- dos, desde 50-70 hasta 120cm o más en plantas adultas. Las plantas jóvenes suelen tener hábito arrosetado, pero aún así desarrollan una corona de frondas de hasta 180cm. El diámetro de los Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 380 CUADRO 1 Caracteres de las pinnas en especies arborescentes del género Blechnum TABLE 1 Characters of the pinnae of arborescent species of genus Blechnum Especie Nº de venas/cm (medido en el margen) Dimensiones estomas (largo x ancho, en µm) Densidad (estomas M por mm2) Haz costal Hipodermis, capas Hipodermis, pared celular Hipodermis, rel. l/a celular B. brasiliense 15-20 (39) 45 (50) x (23) 25 (28) (48) 64 (80) inmerso en parénquima 1 continua pared no engrosada ni lignificada, paredes delgadas, espacios intercelulares 1:1 B. columbiense 13-14 (52) 57 (63) x (34) 35 (38) (12) 20 (32) traba completa, esclerosada 2 continua pared engrosada, lignificada 6-9:1 B. cycadifolium 10-12 (55) 61 (66) x (30) 35 (39) (20) 28 (32) traba completa, esclerosada 2 continua pared engrosada, lignificada 1-1.5:1 B. schomburgkii 16-21 (57) 61 (65) x (35) 37 (40) (16) 24 (32) traba completa, esclerosada 2 continua pared engrosada, lignificada 6-9:1 B. magellanicum 10-14 (38) 44 (52) x (34) 36 (40) (36) 40 (58) traba completa, esclerosada 1 continua pared engrosada, no lignificada 2-4:1 B. moritzianum 16-20 (62) 68 (78) x (46) 50 (56) (12) 16 (20) traba completa, esclerosada 1 continua pared engrosada, no lignificada 2-4:1 B. tabulare 20-23 (47) 58 (64) x (31) 35 (39) (16) 24 (32) traba completa, esclerosada 1 continua pared engrosada, no lignificada, continua, pero con espacios intercelulares pequeños 2-3:1 B. Resultados werkleanum 15-16 (48) 56 (66) x (35) 38 (42) (40) 51 (60) traba completa, esclerosada 1 continua pared engrosada, no lignificada 2-4:1 CUADRO 1 Caracteres de las pinnas en especies arborescentes del género Blechnum TABLE 1 Characters of the pinnae of arborescent species of genus Blechnum Especie Nº de venas/cm (medido en el margen) Dimensiones estomas (largo x ancho, en µm) Densidad (estomas M por mm2) Haz costal Hipodermis, capas Hipodermis, pared celular Hipodermis, rel. l/a celular B. brasiliense 15-20 (39) 45 (50) x (23) 25 (28) (48) 64 (80) inmerso en parénquima 1 continua pared no engrosada ni lignificada, paredes delgadas, espacios intercelulares 1:1 B. columbiense 13-14 (52) 57 (63) x (34) 35 (38) (12) 20 (32) traba completa, esclerosada 2 continua pared engrosada, lignificada 6-9:1 B. cycadifolium 10-12 (55) 61 (66) x (30) 35 (39) (20) 28 (32) traba completa, esclerosada 2 continua pared engrosada, lignificada 1-1.5:1 B. schomburgkii 16-21 (57) 61 (65) x (35) 37 (40) (16) 24 (32) traba completa, esclerosada 2 continua pared engrosada, lignificada 6-9:1 B. magellanicum 10-14 (38) 44 (52) x (34) 36 (40) (36) 40 (58) traba completa, esclerosada 1 continua pared engrosada, no lignificada 2-4:1 B. moritzianum 16-20 (62) 68 (78) x (46) 50 (56) (12) 16 (20) traba completa, esclerosada 1 continua pared engrosada, no lignificada 2-4:1 B. tabulare 20-23 (47) 58 (64) x (31) 35 (39) (16) 24 (32) traba completa, esclerosada 1 continua pared engrosada, no lignificada, continua, pero con espacios intercelulares pequeños 2-3:1 B. werkleanum 15-16 (48) 56 (66) x (35) 38 (42) (40) 51 (60) traba completa, esclerosada 1 continua pared engrosada, no lignificada 2-4:1 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 381 Fig. 1. Anatomía de las pinnas en especies arborescentes de Blechnum. A: sección transversal de la costa en B. brasilie B: sección transversal de la costa en B. schomburgkii. C-E: márgenes de las pinnas. C: B. brasiliense. D-E: B. tabul F: glándula de mucílago en el raquis de B. tabulare. G: protuberancias intercelulares pécticas en parénquima costal d brasiliense. Escala: 200µm en A-B, 0.5mm en C, 50µm en D-G. Fig. 1. Anatomy of the pinnae in arborescent species of Blechnum. A: Cross section of the costa in B. brasiliense. B: C section of the costa in B. schomburgkii. Resultados C-E: Margins of the pinnae. C: B. brasiliense. D-E: B. tabulare. F: Mucilage gl on the rachis of B. tabulare. G: Intercellular pectic protuberances in costal parenchyma of B. brasiliense. Scale: 200µm A-B, 0.5mm in C, 50µm in D-G. Fig. 1. Anatomía de las pinnas en especies arborescentes de Blechnum. A: sección transversal de la costa en B. brasiliense. B: sección transversal de la costa en B. schomburgkii. C-E: márgenes de las pinnas. C: B. brasiliense. D-E: B. tabulare. F: glándula de mucílago en el raquis de B. tabulare. G: protuberancias intercelulares pécticas en parénquima costal de B. brasiliense. Escala: 200µm en A-B, 0.5mm en C, 50µm en D-G. Fig. 1. Anatomy of the pinnae in arborescent species of Blechnum. A: Cross section of the costa in B. brasiliense. B: Cross section of the costa in B. schomburgkii. C-E: Margins of the pinnae. C: B. brasiliense. D-E: B. tabulare. F: Mucilage gland on the rachis of B. tabulare. G: Intercellular pectic protuberances in costal parenchyma of B. brasiliense. Scale: 200µm in A-B, 0.5mm in C, 50µm in D-G. Fig. 1. Anatomía de las pinnas en especies arborescentes de Blechnum. A: sección transversal de la costa en B. brasiliense. B: sección transversal de la costa en B. schomburgkii. C-E: márgenes de las pinnas. C: B. brasiliense. D-E: B. tabulare. F: glándula de mucílago en el raquis de B. tabulare. G: protuberancias intercelulares pécticas en parénquima costal de B. brasiliense. Escala: 200µm en A-B, 0.5mm en C, 50µm en D-G. Fig. 1. Anatomy of the pinnae in arborescent species of Blechnum. A: Cross section of the costa in B. brasiliense. B: Cross section of the costa in B. schomburgkii. C-E: Margins of the pinnae. C: B. brasiliense. D-E: B. tabulare. F: Mucilage gland on the rachis of B. tabulare. G: Intercellular pectic protuberances in costal parenchyma of B. brasiliense. Scale: 200µm in A-B, 0.5mm in C, 50µm in D-G. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 382 glabrescentes y abaxialmente pueden ser esca- mosas, pero la densidad del indumento varía ya que las escamas son caducas. En B. mage- llanicum y B. tabulare se presentan pequeñas glándulas de mucílago 1-celulares, largamen- te persistentes aún en material herborizado (Fig. 1F). Resultados En sección transversal presentan un solo haz, con xilema en forma de V de brazos divergentes, rodeado por floema, parénquima y una endodermis con bandas de Caspary o engro- samientos en forma de U. Los haces en V están trabados por tejido esclerosado que se desarro- lla hacia ambas epidermis, abaxial y adaxial, en todas las especies (Fig. 1B) con excepción de B. brasiliense, en la que están inmersos en parénquima laxo, sin tejido esclerosado aso- ciado (Fig. 1A). En esta última el parénquima del área costal está constituido por grandes células en cuyas paredes primarias celulósicas se encuentran protuberancias intercelulares péc- ticas en forma de verrugas claramente visibles (Fig. 1G). Ocasionalmente en pinnas jóvenes de B. werckleanum se ha observado que los haces presentan una traba esclerosada parcial hacia el hipofilo y un área en forma de casquete esclero- sado hacia el epifilo (Cuadro 1). ascendentes, angostamente oblongas y con ápices caudados, curvados acroscópicamente y algo aserrados en B. werckleanum (Fig. 3H). Los estípites representan 1/3 o más de la longi- tud de la lámina. En B. columbiense (Fig. 2D), B. cycadi- folium (Fig. 2F), B. magellanicum (Fig. 2E) y B. schomburgkii (Fig. 2G) las láminas tienen contorno oblanceolado elíptico. En las tres pri- meras las bases son reducidas más bien abrup- tamente; en B. cycadifolium y B. magellanicum pueden verse 2-3 pares de pinnas basales algo más cortas pero raramente aflebias y son fre- cuentes los aguijones resultantes de la esclero- sis temprana de pinnas atrofiadas. Los estípites pueden alcanzar 1/3 de la longitud de la lámina en B. columbiense y son más cortos en B. cyca- difolium y B. magellanicum. En B. columbiense y B. cycadifolium la fronda es pinnada: en B. columbiense con pinnas lineares divergentes o ascendentes, cuyas láminas están marcadamen- te enrolladas hacia la cara abaxial, tienen ápices romos enrollados y son sub-pecioluladas (Fig. 3A), mientras que en B. cycadifolium las pinnas son lanceoladas con láminas planas pero con los márgenes revolutos, ápices agudos rígidos y bases sésiles (Fig. 3E). En B. magellanicum la lámina es pinnatífida a pinnada, con segmentos adnatos en el tercio apical o hasta la mitad de la lámina, luego pinnada y con pinnas sésiles planas ascendentes y ápices agudos atenuados. La forma y división de las láminas es variable en las especies. En B. brasiliense el contorno es oblanceolado a espatulado (Fig. Resultados 2A), con ápice obtuso y base atenuada, muy gradualmente debido a la presencia de pinnas cuya lámina se reduce más y más basípeta- mente, hasta formar pequeñas aurículas planas, herbáceas o papiráceas, según la edad de la fronda. La lámina es pinnatífida, con pinnas adnatas planas cuyas bases surcurrentes y decurrentes se unen. El estípite es muy corto a casi inexistente. El contorno es también oblanceolado a elíptico en B. schomburgkii, con bases redon- deadas, muy raramente atenuadas y estípites cortos o largos, sin aflebias; el desarrollo de los estípites varía en la misma planta, pero en general en frondas adultas son más bien lar- gos, alcanzando hasta 1/3 de la longitud de la lámina, que es pinnada; aunque ocasionalmente las pinnas apicales son adnatas, la condición pinnada es la más frecuente, con pinnas ascen- dentes, lanceolado-lineares a lanceoladas, ate- nuadas en el ápice y la base, que es sésil, y con los márgenes revolutos prácticamente en toda su extensión (Fig. 3B-D). En B. moritzianum la lámina es lanceola- do elíptica (Fig. 2C) y en B. werckleanum es oblonga (Fig. 2B), en ambos casos con ápices agudos y bases abruptamente reducidas; en general sólo uno o dos pares basales de pin- nas son algo más cortos que los restantes. Las láminas son pinnadas en ambas especies, con pinnas planas distanciadas, que son horizonta- les, lanceoladas y con ápices obtusos a obtuso- mucronados en B. moritzianum (Fig. 3G) y En B. tabulare el contorno es elíptico (Fig. 2H-I), la base se atenúa en pares de pinnas gradualmente reducidas hasta auriculiformes, Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 383 Fig. 2. Contorno de las frondas en especies arborescentes de Blechnum. A: B. brasiliense. B: B. werckleanum. C: B moritzianum. D: B. columbiense. E: B. magellanicum. F: B. cycadifolium. G: B. schomburgkii. H-I: B. tabulare. Escala 50cm. Fig. 2. Outline of the laminae in fronds of arborescent species of Blechnum. A: B. brasiliense. B: B. werckleanum. C: B moritzianum. D: B. columbiense. E: B. magellanicum. F: B. cycadifolium. G: B. schomburgkii. H-I: B. tabulare. Scale: 50cm Fig. 2. Contorno de las frondas en especies arborescentes de Blechnum. A: B. brasiliense. B: B. werckleanum. C: B. moritzianum. D: B. columbiense. E: B. magellanicum. F: B. cycadifolium. G: B. schomburgkii. H-I: B. tabulare. Escala: 50cm. Fig. 2. Outline of the laminae in fronds of arborescent species of Blechnum. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 Fig. 3. External morphology of the pinnae in arborescent species of Blechnum. A: B. columbiense. B-D: B schomburgkii. E: B. cycadifolium. F: B. tabulare. G: B. moritzianum. H: B. werckleanum. Scale: 0.5cm in A-B, D, 1cm in C, E, F 2cm, 3cm in G, H. Resultados 7F; 8F). planas y no esclerosadas ni transformadas en aguijones; los estípites tienen longitud muy variable, desde tres o cinco cm hasta 1/3 de la lámina; ésta es pinnatífida a pinnada, la adnación de las pinnas puede incluir porciones apicales y basales de la lámina; las pinnas son planas, oblongas angostas, horizontales a ascen- dentes, con ápices obtusos, a veces mucronados y márgenes también planos (Fig. 3F). ( g ) y ( g ) Las epidermis tienen células curvas en sección transversal. Los estomas son algo hundidos, diacíticos, anomocíticos, actinocíti- cos, y menos frecuentemente, estaurocíticos y tetracíticos. Presentan engrosamientos de pared epidérmica y pared de contacto de oclusivas, de pared poral y piezas polares en T. La pared epi- dérmica y la pared de contacto de oclusivas, así como la pared poral está muy engrosada en B. columbiense, B. cycadifolium, B. magellanicum y B. schomburgkii. Las piezas polares en T se observaron especialmente en B. magellanicum y B. schomburgkii. En las pinnas con láminas más anchas, los estomas se encuentran en 4-6 hileras entre las venas. Las dimensiones y den- sidades estomáticas se incluyen en el Cuadro 1. Los márgenes son regularmente aserrados en B. brasiliense (Fig. 1C), repandos planos en B. magellanicum, B. moritzianum, B. tabulare (Fig. 1D-E) y B. werckleanum, o repandos revolutos en B. columbiense, B. cycadifolium y B. schomburgkii. Las venas son predominantemente sim- ples, con algunas bifurcadas a distintas dis- tancias de la costa en B. brasiliense (Fig. 4E), simples y bifurcadas una vez, casi en origen (geminadas) o cerca de la costa en B. colum- biense (Fig. 4A), B. cycadifolium (Fig. 4B), B. magellanicum (Fig. 4C) y B. schomburgkii (Fig. 4D); o bien como las anteriores pero con algunas venas bifurcadas dos veces, la segunda división más próxima al margen, en B. mori- tzianum (Fig. 4F) , B. tabulare (Fig. 4H) y B. werckleanum (Fig. 4G). La distancia entre las venas es similar en todas las especies excepto en B. tabulare, que presenta venas muy próxi- mas con dicotomías de ramas casi paralelas (Cuadro 1). y La hipodermis está presente en todas las especies estudiadas, en posición adaxial y es continua. Está constituida por 1-2 capas de células alargadas en el sentido transversal, perpendiculares a la vena y de longitudes variables, equivalentes a 1-2 y hasta 5-6 células epidérmicas suprayacentes. Resultados A: B. brasiliense. B: B. werckleanum. C: B. moritzianum. D: B. columbiense. E: B. magellanicum. F: B. cycadifolium. G: B. schomburgkii. H-I: B. tabulare. Scale: 50cm. Fig. 2. Contorno de las frondas en especies arborescentes de Blechnum. A: B. brasiliense. B: B. werckleanum. C: B. moritzianum. D: B. columbiense. E: B. magellanicum. F: B. cycadifolium. G: B. schomburgkii. H-I: B. tabulare. Escala: 50cm. Fig. 2. Outline of the laminae in fronds of arborescent species of Blechnum. A: B. brasiliense. B: B. werckleanum. C: B. moritzianum. D: B. columbiense. E: B. magellanicum. F: B. cycadifolium. G: B. schomburgkii. H-I: B. tabulare. Scale: 50cm. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 384 g. 3. Morfología externa de las pinnas en especies arborescentes de Blechnum. A: B. columbiense. B-D: B. schomburg B. cycadifolium. F: B. tabulare. G: B. moritzianum. H: B. werkleanum. Escala: 0.5cm en A- B, D, 1 cm en C, E, 2cm 3cm en G, H. g. 3. External morphology of the pinnae in arborescent species of Blechnum. A: B. columbiense. B-D: B schomburg B. cycadifolium. F: B. tabulare. G: B. moritzianum. H: B. werckleanum. Scale: 0.5cm in A-B, D, 1cm in C, E, F 2 m in G H Fig. 3. Morfología externa de las pinnas en especies arborescentes de Blechnum. A: B. columbiense. B-D: B. schomburgkii. E: B. cycadifolium. F: B. tabulare. G: B. moritzianum. H: B. werkleanum. Escala: 0.5cm en A- B, D, 1 cm en C, E, 2cm en F, 3cm en G, H. Fig. 3. Morfología externa de las pinnas en especies arborescentes de Blechnum. A: B. columbiense. B-D: B. schomburgkii. E: B. cycadifolium. F: B. tabulare. G: B. moritzianum. H: B. werkleanum. Escala: 0.5cm en A- B, D, 1 cm en C, E, 2cm en F, 3cm en G, H. Fig. 3. External morphology of the pinnae in arborescent species of Blechnum. A: B. columbiense. B-D: B schomburgkii. E: B. cycadifolium. F: B. tabulare. G: B. moritzianum. H: B. werckleanum. Scale: 0.5cm in A-B, D, 1cm in C, E, F 2cm, 3cm in G, H. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 385 magellanicum (Fig. 6F), B. schomburgkii (Fig. 5F), B. tabulare (Figs. 6I; 7C; 8E) y B. werc- kleanum (Fig. 6C), y compuestas, con engro- samientos en curvas y senos en B. brasiliense (Fig. 7I) y B. moritzianum (Figs. Fig. 4. Patrones de venación en especies arborescentes de Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. tabulare. Escala: 1cm en A-D y H, 0.75cm en E-G. Fig. 4. Venation patterns in arborescent species of Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. tabulare. Scale: 1cm in A-D and H, 0.75cm in E-G. Resultados Las relaciones lon- gitud/anchura (l/a) de estas células hipodérmi- cas van desde 1:1, es decir, casi isodiamétricas, en B. brasiliense (Fig. 7H) y B. cycadifolium (Fig. 5B), 2-4:1 en B. magellanicum (Fig. 6E), B. moritzianum (Fig. 7E), B. tabulare (Figs. 6H; 7B) y B. werckleanum (Fig. 6B), hasta las células muy largas, con relación l/a 6-9:1 de B. columbiense (Fig. 5H; 9A) y B. schomburgkii (Fig. 5E). En estas tres últimas especies la hipodermis suele formar dos y hasta tres capas de células con paredes marcadamente engrosa- das, de contorno rectangular con punteaduras y paredes esclerosadas que reaccionan dando color azul verdoso intenso con TBO (Fig. 9A; C-D). En las especies restantes la hipodermis suele presentar una sola capa, con células cuyas paredes no están lignificadas pero sí variable- mente engrosadas (Fig. 9B, E-F), excepto en B. brasiliense, que las presenta con paredes delgadas y casi isodiamétricas, con espacios Los modelos epidérmicos difieren en epi- filos e hipofilos. Los epifilos tienen células de paredes casi rectas a levemente sinuosas en B. cycadifolium (Figs. 5A; 8A) y son diver- samente sinuosos en las especies restantes. Las ondulaciones son simples, someras, poco frecuentes en B. schomburgkii (Fig. 5D) y B. werckleanum (Figs. 6A; 8B) y más frecuentes en B. columbiense (Fig. 5G) y B. magellanicum (Fig. 6D). Las ondas son también simples, pero profundas en B. tabulare (Figs. 6G; 7A; C). Las ondulaciones son compuestas y curvas en B. brasiliense (Fig. 7G) y compuestas, curvas y angulosas, en B. moritzianum (Figs. 7D; 8D). Los hipofilos son menos variables, con ondas simples, generalmente profundas, en B. colum- biense (Fig. 5I), B. cycadifolium (Fig. 5C), B. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 386 ig. 4. Patrones de venación en especies arborescentes de Blechnum. A: B. columbiense. B: B. cycadifolium. C: agellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. tabulare. Escala: 1 n A-D y H, 0.75cm en E-G. ig. 4. Venation patterns in arborescent species of Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicu : B schomburgkii E: B brasiliense F: B moritzianum G: B werckleanum H: B tabulare Scale: 1cm in A-D and Fig. 4. Patrones de venación en especies arborescentes de Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. Fig. 6. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. werckleanum. D-F: B. magellanicum. G-I: B. tabulare, de África. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala 100µm. Fig. 6. Epidermal patterns in arborescent species of Blechnum. A-C: B. werckleanum. D-F: B. magellanicum. G-I: B. tabulare, from Africa. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Resultados tabulare. Escala: 1cm en A-D y H, 0.75cm en E-G. Fig. 4. Venation patterns in arborescent species of Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. tabulare. Scale: 1cm in A-D and H, 0.75cm in E-G. Fig. 4. Patrones de venación en especies arborescentes de Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. tabulare. Escala: 1cm en A-D y H, 0.75cm en E-G. Fig. 4. Venation patterns in arborescent species of Blechnum. A: B. columbiense. B: B. cycadifolium. C: B. magellanicum. D: B. schomburgkii. E: B. brasiliense. F: B. moritzianum. G: B. werckleanum. H: B. tabulare. Scale: 1cm in A-D and H, 0.75cm in E-G. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 387 g. 5. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. cycadifolium. D-F: B. schomburgkii. G-I: lumbiense. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala: 100µm. g. 5. Epidermal patterns in arborescent species of Blechnum. A-C: B. cycadifolium. D-F: B. schomburgkii. G-I: lumbiense. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Fig. 5. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. cycadifolium. D-F: B. schomburgkii. G-I: B. columbiense. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala: 100µm. Fig. 5. Epidermal patterns in arborescent species of Blechnum. A-C: B. cycadifolium. D-F: B. schomburgkii. G-I: B. columbiense. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 388 3 v. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 g. 6. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. werckleanum. D-F: B. magellanicum. G-I: bulare, de África. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala 100µm. g. 6. Epidermal patterns in arborescent species of Blechnum. A-C: B. werckleanum. D-F: B. magellanicum. G-I: bulare, from Africa. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Fig. 6. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. werckleanum. D-F: B. magellanicum. G-I: B. tabulare, de África. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala 100µm. Fig. 6. Epidermal patterns in arborescent species of Blechnum. A-C: B. werckleanum. D-F: B. magellanicum. G-I: B. tabulare, from Africa. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Rev. Biol. Trop. (Int. J. Trop. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 Fig. 8. Modelos epidérmicos en especies arborescentes de Blechnum. A-D: epifilos. A: B. cycadifolium. B: B. werckleanum. C: B. tabulare. D: B. moritzianum. E-F: hipofilos. E: B. tabulare. F: B. moritzianum. Escala: 35µm en A-B, 40µm en C, 85µm en D, 100µm en E-F. Fig. 8. Epidermal patterns in arborescent species of Blechnum. A-D: Epiphylls. A: B. cycadifolium. B: B. werckleanum. C: B. tabulare. D: B. moritzianum. E-F: Hypophylls. E: B. tabulare. F: B. moritzianum. Scale: 35µm in A-B, 40µm in C, 85µm in D, 100µm in E-F. Resultados Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 389 . 7. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. tabulare, de América. D-F: B. moritzian : B. brasiliense. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala: 100µm. . 7. Epidermal patterns in arborescent species of Blechnum. A-C: B. tabulare, from America. D-F: B. moritzianum. brasiliense. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Fig. 7. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. tabulare, de América. D-F: B. moritzianum. G-I: B. brasiliense. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala: 100µm. Fig. 7. Epidermal patterns in arborescent species of Blechnum. A-C: B. tabulare, from America. D-F: B. moritzianum. G-I: B. brasiliense. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Fig. 7. Modelos epidérmicos en especies arborescentes de Blechnum. A-C: B. tabulare, de América. D-F: B. moritzianum. G-I: B. brasiliense. Izquierda, epifilos; derecha hipofilos; centro, hipodermis. Escala: 100µm. Fig. 7. Epidermal patterns in arborescent species of Blechnum. A-C: B. tabulare, from America. D-F: B. moritzianum. G-I: B. brasiliense. Left, epiphylls, right, hypophylls, center, hypodermis. Scale: 100µm. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 390 g. 8. Modelos epidérmicos en especies arborescentes de Blechnum. A-D: epifilos. A: B. cycadifolium. B: B. wercklean B. tabulare. D: B. moritzianum. E-F: hipofilos. E: B. tabulare. F: B. moritzianum. Escala: 35µm en A-B, 40µm en µm en D, 100µm en E-F. g. 8. Epidermal patterns in arborescent species of Blechnum. A-D: Epiphylls. A: B. cycadifolium. B: B. werckleanum tabulare. D: B. moritzianum. E-F: Hypophylls. E: B. tabulare. F: B. moritzianum. Scale: 35µm in A-B, 40µm in C, 85 D, 100µm in E-F. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 391 9. Hipodermis e indusios en especies arborescentes de Blechnum. A-B: hipodermis en vista superficial. A: B mbiense. B: B. tabulare. C-F: hipodermis en secciones transversales de pinnas. C: B. schomburgkii. D: B. columbiens tabulare. F: B: werckleanum. G-H: márgenes de los indusios. G: B. brasiliense. H. B: moritzianum. Escalas: 40µm e 250µm en C-D; 125µm en E; 400µm en F; 500µm en G: 250µm en H. 9. Hypodermis and indusia in arborescent species of Blechnum. A-B: Hypodermis, superficial view. A: B. columbiens tabulare. CF: Hypodermis, in cross sections of pinnae. C: B. schomburgkii. D: B. columbiense. E: B. tabulare. F: B kleanum. G-H: Margins of the indusia. G: B. brasiliense. H. Resultados B: moritzianum. Scale: 40 µm in A-B, 250 µm in C-D µm in E, 400µm in F, 500µm in G, 250µm in H. Fig. 9. Hipodermis e indusios en especies arborescentes de Blechnum. A-B: hipodermis en vista superficial. A: B. columbiense. B: B. tabulare. C-F: hipodermis en secciones transversales de pinnas. C: B. schomburgkii. D: B. columbiense. E: B. tabulare. F: B: werckleanum. G-H: márgenes de los indusios. G: B. brasiliense. H. B: moritzianum. Escalas: 40µm en A-B; 250µm en C-D; 125µm en E; 400µm en F; 500µm en G: 250µm en H. Fig. 9. Hypodermis and indusia in arborescent species of Blechnum. A-B: Hypodermis, superficial view. A: B. columbiense. B: B. tabulare. CF: Hypodermis, in cross sections of pinnae. C: B. schomburgkii. D: B. columbiense. E: B. tabulare. F: B: werckleanum. G-H: Margins of the indusia. G: B. brasiliense. H. B: moritzianum. Scale: 40 µm in A-B, 250 µm in C-D, 125µm in E, 400µm in F, 500µm in G, 250µm in H. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 392 Los indusios son anchos, alcanzan los 4-5mm, se curvan, tiene márgenes repandos denticulados en B. brasiliense (Fig. 9G) y en el resto de las especies se endurecen marcada- mente en pinnas fértiles adultas y tienen már- genes irregulares (Fig. 9H); están diferenciados en dos porciones: la proximal a su inserción constituida por 1-2 (3) estratos celulares con modelos epidérmicos poligonales o sinuosos, que suele esclerosarse y sus paredes celulares reaccionan dando color azul verdoso o verde intenso con TBO, y una porción distal unis- tratificada, desflecada y con prolongaciones dentadas o piliformes, formada por células poligonales, cuya reacción de color intensa- mente púrpura ante el TBO sugiere que están formadas por celulosa o celulosa y cutina. intercelulares pequeños (Cuadro 1). En sección transversal las células hipodérmicas se distin- guen bien por tener mayor lumen y tamaño que las epidérmicas. El mesofilo es compacto con excepción del de B. brasiliense. Por debajo de la hipodermis se observa parénquima de translocación. En la mayoría de las especies el tejido translocador esponjoso y el clorénquima no muestran una transición morfológica clara entre uno y otro. Hacia la cara adaxial se puede presentar un clo- rénquima laxo, formado por células con forma de horqueta y dispuestas con espacios interce- lulares amplios, como en B. brasiliense y B. Clave para determinar las especies arborescentes de Blechnum 1. Pinnas lineares con láminas completamente enrolladas hacia la cara abaxial; frondas monomorfas de contorno oblan- ceolado-elíptico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. columbiense 1. Pinnas con láminas planas o sólo revolutas en los márgenes, ni lineares ni enrolladas completamente hacia la cara abaxial; frondas monomorfas o dimorfas de contorno variado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Frondas monomorfas oblanceolado-espatuladas, con bases gradualmente reducidas en numerosos pares de pinnas; pinnas adnatas con márgenes planos regularmente aserrados en toda su extensión . . . . . . . . . . . . . . . . . . B. brasiliense 2. Frondas dimorfas de contorno variable, con bases atenuadas o truncadas; pinnas subpecioluladas, sésiles o adnatas; márgenes planos o revolutos, repandos o finamente aserrados sólo en el ápice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3. Láminas con bases truncadas; pinnas distanciadas, subpecioluladas y márgenes repandos o finamente aserrados sólo en el ápice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3. Clave para determinar las especies arborescentes de Blechnum Láminas con bases redondeadas o atenuadas y prolongadas en pinnas reducidas, aflébicas o aguijones esclerosados; pinnas próximas, sésiles o adnatas, con márgenes planos y enteros, revolutos en la base o desde la base hasta el ápice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4. Láminas oblongas, pinnas ascendentes, con ápices curvados acroscópicamente, caudados y finamente aserrados . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. werckleanum 4. Láminas elípticas, pinnas horizontales, con ápice entero, obtuso a agudo . . . . . . . . . . . . . . . . . . . . . . . . B. moritzianum 5. Pinnas con márgenes revolutos desde la base hasta el ápice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5. Pinnas con márgenes planos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6. Resultados moritzianum, o con aspecto homogéneo, con espacios intercelulares más reducidos, como en las especies restantes. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 TRATAMIENTO TAXONÓMICO DE LAS ESPECIES 1897. PROTÓLO- GO: Growing in running water, Colapampa. Collected July 4, 1894, 2314. TIPO: Bolivia. Colapampa, July 1894, Bang 2314 (holótipo no localizado; isótipos B!, GH, NY! US 827874!). Blechnum brasiliense var. angustifolium Sehnem, Pesquisas, Bot. 3: 529. 1959. PRO- TÓLOGO: Tal vez se trate de Bl. nitidum Pr. que segundo o Index Fil. é considerado iden- tical con Bl. brasiliense Desv., que Tweedie teria coleccionado no RS (?). TIPO: Brasil: Rio grande do Sul, Vacaria, Passo do Socorro, Sehnem 5747 (holótipo PACA?). LOGO: Hab. in Brasilia ad Rio Janeiro. TIPO: Brasil: Rio de Janeiro, Pohl? (isótipos B!, GH). Blechnum nigrosquamatum Gilbert, Bull. Torrey Bot. Club 24: 258. 1897. PROTÓLO- GO: Growing in running water, Colapampa. Collected July 4, 1894, 2314. TIPO: Bolivia. Colapampa, July 1894, Bang 2314 (holótipo no localizado; isótipos B!, GH, NY! US 827874!). Blechnum brasiliense var. angustifolium Sehnem, Pesquisas, Bot. 3: 529. 1959. PRO- TÓLOGO: Tal vez se trate de Bl. nitidum Pr. que segundo o Index Fil. é considerado iden- tical con Bl. brasiliense Desv., que Tweedie teria coleccionado no RS (?). TIPO: Brasil: Rio grande do Sul, Vacaria, Passo do Socorro, Sehnem 5747 (holótipo PACA?). Distribución geográfica y ecología: Mesoamérica (Guatemala) y Sudamérica (Colombia, Venezuela, Ecuador, Perú, Bolivia, Brasil, Paraguay, Uruguay y Argentina hasta el delta del Paraná). Moran (1995) cita la espe- cie para Mesoamérica (Guatemala) a partir de una colección única. Crece en selvas o en áreas abiertas, selvas en galería y marginales; es palustre, se encuentra preferentemente en áreas ribereñas de ambiente léntico, en baña- dos y pantanos, desde los 500 hasta los 1 500m de altitud; ocasionalmente se encuentra como terrestre en suelos muy húmedos. Forma colo- nias grandes y abundantes; las láminas varían en textura en relación con el ambiente, y son más coriáceas en los elementos palustres. Blechnum brasiliense var. angustifolium Sehnem, Pesquisas, Bot. 3: 529. 1959. PRO- TÓLOGO: Tal vez se trate de Bl. nitidum Pr. que segundo o Index Fil. é considerado iden- tical con Bl. brasiliense Desv., que Tweedie teria coleccionado no RS (?). TIPO: Brasil: Rio grande do Sul, Vacaria, Passo do Socorro, Sehnem 5747 (holótipo PACA?). Plantas terrestres o palustres, con rizomas erectos robustos, de 60cm o más, no estolonífe- ros, escamosos. Escamas rizomáticas lineares, enteras, erguidas y algo retorcidas, de color cas- taño oscuro hasta negruzco, generalmente con- coloras o con márgenes más claros, de 2-4cm de longitud. Clave para determinar las especies arborescentes de Blechnum Láminas con base redondeada; pinnas sésiles marcadamente ascendentes, linear lanceoladas, con la base atenuada y el ápice agudo u obtuso; estípite con aguijones esclerosados ocasionales . . . . . . . . . . . . . . . . . . . . . . . . B. schomburgkii 6. Láminas con base atenuada y prolongada en pinnas gradualmente reducidas; pinnas sésiles ascendentes, casi tan anchas en la base como en la parte media, con el ápice agudo rígido; estípite siempre con aguijones esclerosados . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. cycadifolium 7. Lámina oblanceolado-elíptica, con ápice agudo y base interrumpida abruptamente o algo atenuada, a lo sumo con 2-4 pares de pinnas basales reducidas; pinnas oblongo-triangulares, próximas, algo incurvadas acroscópicamente, con ápice agudo; estípite con aguijones endurecidos frecuentes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. magellanicum 7. Lámina elíptica, con ápice obtuso y base gradualmente reducida, con hasta 14 pares de pinnas basales y con aflebias laminadas; pinnas oblongas u oblanceolado-oblongas, horizontales, imbricadas en el ápice, luego distanciadas, con ápices obtusos u obtuso-mucronulados; estípite sin aguijones endurecidos . . . . . . . . . . . . . . . . . . . . . . . . . . B. tabulare Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 393 TRATAMIENTO TAXONÓMICO DE LAS ESPECIES cartáceas. Raquis glabro. Costas glabrescentes o sólo escamosas en la porción abaxial basal, adaxialmente sobresalientes y abaxialmente carenadas; haces inmersos en parénquima, no trabados; protuberancias intercelulares pécticas verrucosas presentes en el tejido parenquimá- tico costal. Segmentos laterales 35-50 pares, ascendentes, completamente adnatos con el raquis, surcurrentes-percurrentes, lanceolado triangulares, planos, con el ápice agudo, acumi- nado u obtuso y margen regularmente aserrado. Venas visibles, próximas, simples y una vez bifurcadas a distancias variables de la costa, con ramas escasamente divergentes, termina- das en hidatodos grandes de contorno elíptico. Indusios repandos-denticulados, con la mitad basal formada por células engrosadas de con- torno sinuoso frecuente, la mitad marginal por células engrosadas de contorno poligonal y el margen mismo formado por ondas irre- gulares pluricelulares rematadas en dentículos cónicos unicelulares. 1. Blechnum brasiliense Desv., Mag. Neuesten Entdeck. Gesammten Naturk. Ges. Naturf. Freunde Berlin 5: 330. 1811. Blechnop- sis brasiliensis (Desv.) C. Presl, Epimel. Bot.: 115. 1851. PROTÓLOGO: Habitat in Brasi- lia (n. v.). TIPO: Brasil. Dombey (P-Jussieu 1390!). Blechnum corcovadense Raddi, Opusc. Sci. 3: 294. 1819. PROTÓLOGO: Brasil, Cor- covado, Raddi. TIPO: Brasil: Rio de Janeiro, Corcovado, Raddi (holótipo PI; isótipos B!, FI). Blechnum nitidum C. Presl in J. Presl & C. Presl, Delic. prag. 1: 187. 1822. PROTÓ- LOGO: Hab. in Brasilia ad Rio-Janeiro. TIPO: Brasil: Rio de Janeiro, Pohl? (isótipos B!, GH). Blechnum nigrosquamatum Gilbert, Bull. Torrey Bot. Club 24: 258. 1897. PROTÓLO- GO: Growing in running water, Colapampa. Collected July 4, 1894, 2314. TIPO: Bolivia. Colapampa, July 1894, Bang 2314 (holótipo no localizado; isótipos B!, GH, NY! US 827874!). Blechnum brasiliense var. angustifolium Sehnem, Pesquisas, Bot. 3: 529. 1959. PRO- TÓLOGO: Tal vez se trate de Bl. nitidum Pr. que segundo o Index Fil. é considerado iden- tical con Bl. brasiliense Desv., que Tweedie teria coleccionado no RS (?). TIPO: Brasil: Rio grande do Sul, Vacaria, Passo do Socorro, Sehnem 5747 (holótipo PACA?). Blechnum corcovadense Raddi, Opusc. Sci. 3: 294. 1819. PROTÓLOGO: Brasil, Cor- covado, Raddi. TIPO: Brasil: Rio de Janeiro, Corcovado, Raddi (holótipo PI; isótipos B!, FI). Blechnum nitidum C. Presl in J. Presl & C. Presl, Delic. prag. 1: 187. 1822. PROTÓ- Blechnum nitidum C. Presl in J. Presl & C. Presl, Delic. prag. 1: 187. 1822. PROTÓ- LOGO: Hab. in Brasilia ad Rio-Janeiro. TIPO: LOGO: Hab. in Brasilia ad Rio-Janeiro. TIPO: Brasil: Rio de Janeiro, Pohl? (isótipos B!, GH). Blechnum nigrosquamatum Gilbert, Bull. Torrey Bot. Club 24: 258. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Frondas monomorfas de hasta 150cm de longitud, ascendentes y arqueadas, las juveniles a veces de color rosado a magenta. Estípites rígidos de hasta 16-18cm de longitud, con escamas oscuras en la base, similares a las del rizoma pero en general de color castaño más claro y brillantes. Láminas oblanceolado- espatuladas, pinnatífidas, gradualmente reduci- das hacia la base y el ápice, herbáceas firmes a Ejemplares de referencia seleccionados: ARGENTINA: Corrientes: Ituzaingó, Rincón del Playadito, 5km N de San Alonso, islote de selva con arroyo afluente del Playadito, Tressens et al. 5042 (LP). Estancia Santa Rita, interior selva marginal, Tressens et al. 3951 (LP). Misiones: San Ignacio, camino al río, a 200m de la casa de Horacio Quiroga, Palacios Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 394 et al. 1295 (LP). Iguazú, sendero Jacatía, ex Ruta 12, 2km del CIES, Vanni et al. 3239 (LP). Iguazú: Cataratas, Prada s. n. (MA). BRASIL: Goiás: Serra Dourada, Ca. 17km of Goiás Velho, Ca. 6km OW of Mossamedes, 750m, along stream at base of East side of Sierra, Anderson 9957 (LP). Paraná: Yacarehý, Dusén s. n. (MA 213329). Rio de Janeiro: Gua- nabara, Serra Carioca, Caranta N-638 (LP). Tijuca, Floresta de Tijuca, Castellanos 23070 (LP). Rio Grande do Sul: sin loc., Reinerck & Czermak 164 (BA). Santa Catarina: Pelle- grini, camino de Santo Amaro da Imperatriz a Fazenda Jomar, Krapovickas & Cristóbal 44827 (LP). Municipio Garuvá, Krapovickas et al. 23073 (CTES). GUYANA: Region Potaro- Siparuni, Pakaraima Mts., Upper Ireng. 0.5km section of Tay-Slay-O Creek, 4-8km upstream of Ireng mouth, Henkel 873 (MA 571361). PARAGUAY: Canindeyú: cerca de Curguatý, junto al arroyo Schiní, FC5887 (MA 477474). Central: Capiatá, Schinini 4157 (BA). URU- GUAY: Cerro Largo, Sierra Aceguá, Osten 18588 (BA). 1m de largo. Estípites rígidos, de hasta 25cm de largo, adaxialmente surcados, con escamas basales similares a las del rizoma. Láminas oblanceolado-elípticas, abruptamente reduci- das hacia la base y con ápice más o menos obtuso. Raquis y costas densamente escamosos, con escamas castañas largamente acintadas, caducas. Costas adaxialmente surcadas, con un surco profundo, abaxialmente carenadas, con el haz vascular trabado por tejido parenquimático esclerosado y fibroso hacia las caras adaxial y abaxial. Pinna terminal conforme, discreta, a veces más corta que las laterales. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES columbiense, se distinguen por el menor enro- llamiento de la lámina, la venación y el número de venas por cm y por los modelos epidérmicos de epifilos e hipofilos. Blechnum columbiense tiene una distribución geográfica más restrin- gida, limitada a los páramos de Colombia, Ecuador y Perú. 3. Blechnum cycadifolium (Colla) J.W. Sturm., Abh. naturhist. Ges. Nürnberg 2: 173. 1858. Lomaria cycadifolia Colla, Herb. pedem. 6: 219. 1836. Blechnum magellanicum var. cycadifolium (Colla) C. Chr., Dansk Bot. Ark. 10(2): 9. 1910. Struthiopteris cycadifolia (Colla) Ching, Sunyatsenia 5: 243. 1940. PRO- TÓLOGO: [Lomaria] magellanica an diversa? Berter. In sylvaticis collium et montium ins. Juan-Fernández. TIPO: Masatierra, Yunque. 24.08.1908, Leg.: C. J. F. Skottsberg 407. Notes: Expeditio suecica 1907-1909. (Holóti- po, isótipo B!). 3. Blechnum cycadifolium (Colla) J.W. Sturm., Abh. naturhist. Ges. Nürnberg 2: 173. 1858. Lomaria cycadifolia Colla, Herb. pedem. 6: 219. 1836. Blechnum magellanicum var. cycadifolium (Colla) C. Chr., Dansk Bot. Ark. 10(2): 9. 1910. Struthiopteris cycadifolia (Colla) Ching, Sunyatsenia 5: 243. 1940. PRO- TÓLOGO: [Lomaria] magellanica an diversa? Berter. In sylvaticis collium et montium ins. Juan-Fernández. TIPO: Masatierra, Yunque. 24.08.1908, Leg.: C. J. F. Skottsberg 407. Notes: Expeditio suecica 1907-1909. (Holóti- po, isótipo B!). Ejemplares de referencia selecciona- dos: COLOMBIA: Cundinamarca: Bogotá: Andes, Cordillera Oriental, La Peña, 2 800m, Cuatrecasas 2081 (MA 213330) –2 ejempla- res–. Páramo de Montserrate, El Granizo, 3 250m, Fernández Alonso & Arenas 5134 (MA 517122), planta de la derecha. Páramo de Montserrate, 3 200-3 300m, Cardiel & Rubi 5 (MA 494869). Carretera Zipaquirá-Paho, 3 000m, Hagemann 165 (COL 118540). Boya- cá: Páramo de Guantiva, entre Santa Rosita y Onzaga, en la Quebrada de Sietecueral, 3 410m, Jaramillo Mejía & van der Hammen 4287 (MA 381471). Entre Zipaquirá y Pacho, Páramo de Guerrero, 3 100m, Murillo & Villa- rreal 1370 (MA 228368). Santander: Páramo de Guantiva, lado occidental, hacia Onzaga, 3 460m, Jaramillo Mejía & van der Hammen 4313 (MA 380907). Cerca Páramo de Santur- bán, 3 000m, Barklay & Araque 185176 (COL 495862). Valle del Río Susa, en el camino de Hato Viejo a La Figuera o Jiquera (Jequera en el mapa), arriba del sitio de Santa Bárbara, en la Loma del Reventón, 3 600m, Jaramillo Mejía & van der Hammen 4399 (MA 380819). Santander del Norte: al oriente de Pamplona, Páramo de Fontibón, 2 800m, Murillo & Jara- millo Mejía 1307 (MA 228367). TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Pinnas laterales hasta 40 pares, lineares, coriáceas, marcadamente ascendentes, próximas pero no imbricadas, las apicales sésiles, luego subpe- cioluladas, con ápice obtuso y lámina comple- tamente enrollada hacia la cara abaxial, con escamas caducas acintadas y largas, enroscadas y de aspecto piliforme. Venas inmersas, dis- tantes, simples y bifurcadas una vez con una división en dos ramas claramente divergentes cerca de la costa, algunas geminadas. Láminas fértiles de hasta 50 cm de largo, pinnas fértiles angostas, lineares, con margen recurvado y cenosoros mediales que cubren por completo la cara abaxial. Indusios eroso-fimbriados, con células basales engrosadas. 2. Blechnum columbiense Hieron., Hed- wigia 47: 244. 1908. PROTÓLOGO: Colum- bia, crescit in regione páramo dicta Montis Huila, in páramo de Moras (n. 1268). TIPO: Colombia: Huila: Páramo de Moras, Stübel 1268 (holotipo B!). Distribución geográfica y ecología: Pára- mos de Colombia, Ecuador y Perú, desde 2 800 hasta 3 600m o más. Blechnum columbiense var. bogotense Hieron., Hedwigia 47: 244.1908. PROTÓLO- GO: Columbia: hábitat prope urbem Bogotá (Karsten s. n.), alio loco in regiones urbem Bogotá, alt. 2 800m (Lindig 27). TIPO: Colom- bia: Bogotá: Karsten s. n. (B!); Lindig 27 (sín- tipos B!; isosíntipo 27 BM n. v.). Observaciones: Tryon & Stolze (1993) combinaron Blechnum columbiense como una subespecie de B. auratum, ésta basada en Lomaria aurata Fée. Blechnum columbiense es una especie bien caracterizada cuyo material tipo no genera dudas sobre su determinación. Se distingue por sus pinnas lineares, angostas, completamente revolutas y enrolladas hacia la cara abaxial, llegando hasta la costa y la ausen- cia de dimorfismo, ya que las frondas fértiles son indistinguibles de las estériles y sólo en el caso de la presencia de abundantes esporangios maduros se hace visible el carácter de láminas esporógenas. Es afín con B. schomburgkii. Las formas con pinnas más angostas de esta Blechnum auratum subsp. columbiense (Hieron) Tryon & Stolze. Fieldiana, Botany 32: 67. 1993. Blechnum auratum subsp. columbiense (Hieron) Tryon & Stolze. Fieldiana, Botany 32: 67. 1993. Plantas terrestres, con rizoma erecto de hasta 200cm de alto, escamoso. Escamas rizo- máticas aciculares a lineares, de erguidas a algo curvadas todas en la misma dirección, de color castaño oscuro, rígidas, lustrosas, de 3-4cm de largo. Frondas monomorfas, pinnadas, de hasta Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 395 (UC 1870542). Piura: Huancabamba, Hutchin- son 1619 (UC, USM). última especie, que podrían confundirse con B. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Láminas fértiles erectas, erguidas en el centro de la espiral de frondas; pinnas fér- tiles angostas, lineares, con cenosoros mediales a marginales. Indusios eroso-fimbriados, con células basales engrosadas. Observaciones: Blechnum cycadifolium forma comunidades muy densas en lugares abiertos o despejados, hasta los 1 200m de alti- tud. Es abundante en matorrales y pastizales, forma extensos bosquecillos, aún en suelos ero- sionados. Muestra cierta variación en la textura y grado de desarrollo de las frondas de acuerdo con la cantidad de luz que recibe y se han mencionado formas umbrófilas (Johow 1896). Según Skottsberg (1953), es más abundante en las faldas expuestas al viento, en Masatierra, donde es la planta dominante y crece desde el nivel del mar hasta las cumbres más elevadas. En Masafuera es menos frecuente y asciende por encima del límite arbóreo hasta cerca de la cima de la isla. Es una de las especies arbores- centes más robustas del género, ya que su rizo- ma erecto puede alcanzar los 250cm de altura. Las colecciones son escasas. Distribución geográfica y ecología: Chile, Juan Fernández. En general, se la con- sidera como una especie endémica de las islas Robinson Crusoe y Alejandro Selkirk, donde convive con B. magellanicum. En la isla Robinson Crusoe forma colonias abundantes que se distribuyen desde el Sector de Puerto Francés hasta Cerro Chumacera; en Cordón y Quebrada de Salsipuedes, Cerro Alto, Sector de Puerto Inglés hasta Quebradas Vaquería y Juanango; en isla Alejandro Selkirk se encuen- tran en Quebrada las Chozas, Quebrada de las Casas, Quebrada la Lágrima, Quebradas los Inocentes, El Guatón, La Lobería, Varadero, Óvalo, el Pasto, El Sándalo, Planos de Sán- chez (Johow 1896, Skottsberg 1924, 1953, Barrera 1997). Rodríguez Ríos (1995) y Ricci (1990, 1992) registraron más de 5 000 plantas distribuidas en diferentes sectores de ambas islas, donde la especie es abundante en lugares abiertos y forma agrupaciones muy tupidas. Esos autores hacen referencia a la regeneración natural de las colonias pero también conside- ran que a pesar del gran número de plantas que se detectaron en ambas islas, la especie estaría amenazada por el progreso de Rubus ulmifolius Schott (zarzamora), Aristotelia chi- lensis (Molina) Stuntz (maqui) y Ugni molinae Turcz. (murtilla), así como por los incendios y los deslizamientos de terreno producidos por las frecuentes lluvias. Ricci (1990, 1992) la considera fuera de peligro, mientras que para Rodríguez Ríos (1995), Baeza et al. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Along road between Pamplona and Chorro Colorado, Vía Chiquirá, 2 580m, Croat 56408 (COL 344728). Plantae Expeditionis Botanicae Mutisii Vice- Regni Novae Granatae (1783-1808), 3119 (MA 659308), 3116 (MA 659331), 3123 (MA 659334). ECUADOR: Loja: Cajanuma, Leh- nert 833 (UC 1791891). PERU: Luya: Distrito Camporredondo, Tullanya, cerro Wiscocunga, Campos & Campos 3138 (UC 1733388). Oxa- pampa: Huancabamba, Sector Santa Bárbara- Milpo, 2 750-3 000m Mellado & Ortiz 2745 Lomaria lanuginosa Kunze, Analecta pteridogr.: 19. 1837. Blechnum lanuginosum (Kunze) J.W. Sturm, Abh. naturhist. Ges. Nür- nberg 2: 175. 1858. PROTÓLOGO: Ex insula Juan Fernandez misit Larrain, Decembri 1829 lectam (Lomariae sp. n.? nº 1110). Fertilem dedit Bory de St.Vincent; sterilem unio itinerar. e plant. Berteroanis (Herb. propr.). Lomaria lanuginosa Kunze, Analecta pteridogr.: 19. 1837. Blechnum lanuginosum (Kunze) J.W. Sturm, Abh. naturhist. Ges. Nür- nberg 2: 175. 1858. PROTÓLOGO: Ex insula Juan Fernandez misit Larrain, Decembri 1829 lectam (Lomariae sp. n.? nº 1110). Fertilem dedit Bory de St.Vincent; sterilem unio itinerar. e plant. Berteroanis (Herb. propr.). Plantas terrestres, de hasta 300cm de alto, con rizoma erecto escamoso de hasta 250cm. Escamas rizomáticas lineares, erguidas a algo curvadas todas en la misma dirección, de color castaño oscuro, rígidas, de 3-6cm de largo. Frondas dimorfas, pinnadas, de hasta 120cm de largo, las estériles más largas que las fértiles, éstas erguidas, de hasta 80cm de largo. Estípites rígidos, de hasta 30cm de largo, adaxialmente surcados y con pinnas vestigiales esclerosadas reducidas a aguijones espinosos. Láminas estériles oblanceoladas, reducidas hacia la base y con ápice obtuso a mucronado. Raquis y costas con escamas castañas, caducas. Costas adaxialmente surcadas con un surco profundo y abaxialmente carenadas, con un solo haz vascular trabado por tejido esclero- sado y fibroso que se extiende hacia las caras adaxial y abaxial. Pinna terminal conforme o más corta que las laterales, a veces adnata con 1-2 pares apicales. Pinnas laterales hasta 50 pares, coriáceas, ascendentes, imbricadas en el tercio apical y luego muy próximas, sésiles, de Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 396 contorno elíptico angosto, ápice obtuso rígido en plantas adultas, margen repando revolu- to, abaxialmente escamosas en la costa, con escamas caducas. Venas inmersas, distantes, simples y bifurcadas una vez cerca de la costa en dos ramas claramente divergentes, algunas geminadas. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Lámina estéril oblanceolado-elíptica, ancha, con ápice agudo y base interrumpida abruptamente y sin pinnas reducidas ni aflebias laminares en frondas adultas, en éstas con aguijones esclero- sados resultantes de la atrofia de pinnas y algo atenuada, con 2-4 pares de pinnas reducidas no esclerosadas en frondas jóvenes. Raquis paleáceo. Costas acanaladas en la cara adaxial y casi carenadas en la abaxial, externamente con glándulas unicelulares de mucílago y esca- mas lineares acintadas con aspecto lanuginoso, con el haz vascular trabado por tejido paren- quimático esclerosado, fibroso, hacia las caras adaxial y abaxial. Pinna terminal conforme con las laterales pero en general más angosta, más ancha en la base y adnata con uno o más pares. Pinnas laterales hasta 40 pares, coriáceas, alternas o sub-opuestas, oblongo-triangulares, con ápices agudos atenuados y bases adnatas en el tercio apical de las frondas, luego sésiles y truncadas, asimétricas, con margen repando plano. Láminas fértiles erectas, con pinnas de 0.5cm de ancho, dirigidas acroscópicamente. Indusio con margen irregular. 5. Blechnum moritzianum (Klotzsch) Moore. Ind. fil.: 200. 1857. Lomaria moritzia- na Klotzsch, Linnaea 20: 347. 1847. TIPO: Venezuela: Mérida, in sylv. subalpin. J. W. K. Moritz 301 (B 20 0031858!, B 20 0031859!, B 20 0031860!, B 200031861!l; isótipos K!, BM!). 5. Blechnum moritzianum (Klotzsch) Moore. Ind. fil.: 200. 1857. Lomaria moritzia- na Klotzsch, Linnaea 20: 347. 1847. TIPO: Venezuela: Mérida, in sylv. subalpin. J. W. K. Moritz 301 (B 20 0031858!, B 20 0031859!, B 20 0031860!, B 200031861!l; isótipos K!, BM!). Distribución geográfica y ecología: Argentina y Chile. En Argentina desde Neu- quén hacia el S, en bosques hiperhúmedos cordilleranos y en áreas más expuestas, sobre todo en los bosques subantárticos, en suelos sueltos o algo rocosos, no acidófila; en Chile, desde Valparaíso y el Archipiélago de Juan Fernández hacia el S, en bosques, bordes de bosques, valles y quebradas húmedas, entre rocas y en tierra, con agua cerca pero no en suelos encharcados. En el archipiélago de Juan Fernández en asociaciones con B. cycadifolium y B. schottii (Colla) C. Chr., un elemento de las áreas más húmedas, escandente o hemiepífito. Blechnum yungense Ramos Giacosa, Novon 20(1): 68-72. 2010. TIPO: Argentina: Jujuy: Depto. Ledesma, camino a Valle Grande, 9.IV-1971, Legname & Cuezzo 8219 (Holótipo LP!, Isótipos LIL!, CTES!). Plantas terrestres, con rizomas ascendentes a erguidos, de hasta 70-80cm de alto, escamo- sos. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES (1998) y Danton (2004) sería vulnerable. Ejemplares de referencia selecciona- dos: CHILE: Archipiélago de Juan Fernández: Masafuera, Correspondencia Camp, Skottsberg & Skottsberg 405 (US 1085782). Los Inocen- tes, Solbrig et al. 3760 (US 2421520). Masa- tierra: high ridge above Rabanal, Skottsberg & Skottsberg 580 (US1055780). Masatierra, Aréte del Centinela, Skottsberg & Skottsberg 580 (P). Slopes of Damajuana, Solbrig et al. 3790 (US 2421535). Ex insula Juan Fernández, Bertero s. n. (P 0140440). 4. Blechnum magellanicum (Desv.) Mett. Fil. Lechl. 1. 14. 1856. Lomaria magellanica Desv. Mag. Neuesten Entdeck. Gesammten Naturk. Ges. Naturf. Freunde Berlin 5: 330. 1811. PROTÓLOGO: [¿Chile?] Habita ad fre- tum magellanicum n. v. 4. Blechnum magellanicum (Desv.) Mett. Fil. Lechl. 1. 14. 1856. Lomaria magellanica Desv. Mag. Neuesten Entdeck. Gesammten Naturk. Ges. Naturf. Freunde Berlin 5: 330. 1811. PROTÓLOGO: [¿Chile?] Habita ad fre- tum magellanicum n. v. Plantas terrestres con rizoma erecto de hasta 100cm de alto y 20cm de diámetro, cubierto por las bases persistentes de los estípi- tes caducos, escamosos. Escamas rizomáticas lineares, erguidas a algo curvadas todas en la misma dirección, enteras, largamente ate- nuadas, sésiles, de 3.5-4cm de largo. Frondas dimorfas; las estériles arqueadas, de 50-150cm, las fértiles erectas, iguales o algo más largas. Estípites rígidos, 1/3 del largo total de la Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 397 Isla Victoria. Nahuel Huapi, Diem 1914 (LIL). Río Negro: Laguna Frías, Meyer 7338 (LIL 113135). Santa Cruz: Lago Argentino, Brazo Mayo, Vervoorst 4806 (LIL). Tierra del Fuego: Isla de Los Estados, Puerto Cook, below the multiple peaks of Monte Orejas de Burro, Dud- ley & al. 692 (LP). Tierra del Fuego: Torres1179 (LP). CHILE: Región V: Valparaíso, Archipié- lago de Juan Fernández: Isla Robinson Crusoe (Mas a Tierra), Pérez Moreau 22936 (BA). Plazoleta del Yunque [Plaz. del Yunque, im Hochwalde on volcán, 300m], Bock 33 (US 15994193 & US 1787893). Mountain apur E of Quebrada de la Damajuana, Skottsberg & Skottsberg 74 (US 2751260). Region XII: Provincia Última Esperanza, Puerto Edén, Isla Wellington, Capitanía de Puerto, Pisano 6455 (CONC 104825, MA 700919). Fiordo Peel, Río Martillar, Pisano 5942 (CONC 95870, MA 7000121). Bahía Micaela, Isla Saumarez, canal Grappler, Pisano 6573 (CONC 104778, MA 700330). [Chile ] Pto. Egmont, Née s. n. (MA 230549). fronda, paleáceos, con un surco adaxial, con aguijones o mamelones esclerosados y sección casi circular; internamente huecos, con área central parenquimática reabsorbida. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Escamas rizomáticas lineares, curvadas todas en la misma dirección, rígidas, de color castaño oscuro, concoloras o bicoloras y con un área central castaño-negruzca, sésiles, ente- ras, largamente atenuadas, de 2-3cm de largo. Frondas dimorfas, las estériles de hasta 150cm de largo, con láminas elípticas, anchas, con ápices agudos, bases truncadas sin aflebias ni pinnas reducidas; las fértiles iguales o más Ejemplares de referencia seleccionados: ARGENTINA: Chubut: Lago Menéndez, Aler- zal del Norte, Meyer 9517 (LIL). Islas Malvi- nas: Puerto Stanley, Hicken 31 (LIL). Neuquén: Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 398 largas. Estípites de hasta 30-35cm de largo o más a veces tan largos como la lámina, adaxialmente surcados, con escamas basales similares a las rizomáticas. Raquis escamosos, con escamas largas y lineares, castañas. Costas abaxialmente curvas y adaxialmente surcadas, con un surco poco profundo, con el haz trabado hacia la cara adaxial por tejido parenquimático esclerosado que también forma casquetes fibro- sos hipodérmicos hacia la cara abaxial, que no siempre traban el haz. Láminas estériles pinna- das con la pinna terminal conforme, algo más corta y basalmente más ancha que las laterales. Pinnas laterales hasta 25-30 pares, lanceola- das a oblongo-lanceoladas, horizontales, más o menos distanciadas, con bases asimétricas cuneadas, ápices agudos poco atenuados y márgenes repandos planos. Venas inmersas próximas, simples y una vez bifurcadas cerca o lejos de la costa, geminadas raras o ausentes. Indusios profundamente eroso-desflecados. ese nombre, ya que no se encuentra en obras posteriores y B. moritzianum aparece en los herbarios determinada indistintamente como B. auratum subsp. auratum, B. schomburgkii o B. buchtienii. Blechnum moritzianum es el nombre que corresponde aplicar a los materia- les recientemente descritos por Ramos Giacosa (2010) como B. yungense. El tipo depositado en B es un fragmento de fronda y son más representativos los isótipos depositados en BM y K, en los que las frondas están completas e incluyen estípites con sus escamas basales curvadas. Blechnum moritzianum difiere de B. schomburgkii por el contorno de las láminas, las pinnas horizontales anchas y distanciadas, de láminas bien desarrolladas, con bases asi- métricas subpecioluladas y márgenes enteros planos, no revolutos. La especie más afín es B. werckleanum, con pinnas patentes incurvadas, oblongas angostas, atenuadas, caudadas, con ápices acuminados aserrados dirigidos acroscó- picamente, marcadamente bicolores y abaxial- mente mucho más claras, un rasgo que también puede estar presente en B. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES moritzianum pero es más evidente aún en material seco, en B. werc- kleanum. Los modelos epidérmicos de las tres especies difieren en la ondulación de los epifi- los, más somera en B. schomburgkii y B. werc- kleanum, y de los hipofilos, con ondas simples, subangulosas en B. schomburgkii, curvas, en B. werckleanum y compuestas curvas en B. moritzianum. Las hipodermis también difieren, ya que en B. schomburgkii está formada por células largas traqueiformes, lignificadas mien- tras que en B. moritzianum y B. werckleanum las células son más cortas, menos engrosadas y no se esclerosan. Distribución geográfica y ecología: Sud- américa, desde Venezuela hasta el NO de Argentina. Es una especie abundante en los Andes; crece tanto en la selva montana como en el bosque parcialmente caducifolio más alto, en pastizales y pajonales, sobre barrancas y laderas expuestas. Es una planta terrestre, resis- tente y rígida, de color verde más claro a glauco en los hipofilos. Desde los 1 200-1 330m hasta los 3 500m de altitud. Observaciones: Lomaria moritziana Klotzsch fue transferida a Blechnum por Moore (1857), quien incluyó en su sinonimia Salpi- chlaena volubilis (Kaulf.) Sm. Esta combi- nación es válida de acuerdo con el ICBN, y Moore (1857), además, incluye una breve refe- rencia al basónimo, acompañada del nombre de su autor: “Moritzianum, Kl.”. Hieronymus (1908) aceptó la combinación de Moore (1857) pero excluyó, correctamente, la sinonimia, citó tanto el basónimo como la nueva combinación en sus comentarios sobre las afinidades de B. guascense Hieron. y también citó el tipo, pro- cedente de Venezuela (Moritz 301), depositado en el Herbario B. Es el último autor que utilizó Ejemplares de referencia seleccionados: ARGENTINA: Salta: Parque Nacional Baritú, Los Toldos, camino al Lipeo, Prada s. n. (LP, MA). Santa Victoria, entre El Lipeo y Canto del Monte, 200m, Novara & Adzet 9133 (CORD, CTES, MCNS). Camino Los Toldos al Lipeo, Desvío San José, 1 750m, Cassá et al. 286 y 287 (LP, MCNS). Canto del Monte, camino de Los Toldos, 6km al N antes de llegar a Lipeo, 1 200-1 500m, Novara & Barboza 11556 (CTES, Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 399 MCNS). Santa Victoria, camino al Lipeo, 1 600m, Martínez & de la Sota s.n. (MCNS). TRATAMIENTO TAXONÓMICO DE LAS ESPECIES BOLIVIA: Santa Cruz: Florida, Parque Nacio- nal Amboró, 8km (by air) NE of Mairana, along trail and near ridge tops, 5.5km N of Campa- mento La Yunga Park Station, 2 300m, Sundue et al. 545 (UC 1780949). Caballero, Cerro Curu- cucho, ca. 1km ENE de Liberia, 3 000m, Saldias & Fernández 4564 (CTES 291019). COLOM- BIA: Chocó: Municipio San José del Palmar, Cerro del Torrá, Silverstone-Stopkins et al. 1374 (CTES). Bogotá: montes detrás de Bogotá, 3 200m, Carvajalino 214 (COL 89678). Boyacá: Sierra Nevada de Cocuy, Cordillera Oriental, Grubb & Guymer 66 (COL 89414). Cundina- marca: San Miguel, extremo SW de la Sabana de Bogotá, 2 800m, Jaramillo Mejía, Murillo & van der Hammen 3418 (MA 381510 y MA 380762). Below Alto del Cuchuco, 7km SW of Sibate, Tryon & Tryon 6111 (COL 95378). Zipaquirá, inmedianciones de Pantano Redondo, 2 900m, Huertas & Camargo 580 (COL 106268). Al N de Bogotá, bosques arriba de la Hacienda La Francia, Torquita, 2 800m, Jaramillo Mejía & van der Hammen 4522 (COL 243992). Meta, Cordillera Oriental main north affluent of Rio Grande, south of Cordillera de Las Cruces, south slopes of Páramo de Sumapaz, 3 130m, Fosberg 20866 (COL 244351). Santander del Norte: al oriente de Pamplona, Páramo de Fontibón, 2 800m, Murillo & Jaramillo Mejía 1321 (COL 125892). ECUADOR: Zamora- Chinchipe: Estación Científica San Francisco, Lehnert 849 (UC 1791755). PERÚ: Amazonas: Chachapoyas, Chachapoyas-Cajamarca road, between Limebamba and Calla-Calla, 12-15km from Leimebamba, coll? 4984 (USM 214613). Pasco: Oxapampa, from Antenna past Chacos to the Laguna, van der Werff et al. 19642 (UC 1872875). Piura: Ayabaca, Bosque de Huamba, Cano 1613 (USM). Blechnum schomburgkii (Klotzsch) C.Chr. var. squamulosa Hieron., Plantae Stübelianae, Filices, no. 444. TIPO: Colombia: Bogotá Excurs.; oberhalb Pasca. Alt.: 9 000’. A. Stübel 444 (B 20 0033242!). Lomaria aurata Fée, Mém. foug. 8: 71. 1854-57. Blechnum auratum (Fée) R. M. Tryon & Stolze subps. auratum, Fieldiana, Bot. 32: 67. 1993. PROTÓLOGO: Nouvelle-Grena- de, province d´Ocaña, dans les Paramos, à 2 600-3 300 mètres d´altitude (L. Schlim nº 394; 1846-1852). TIPO: Colombia: Ocaña, Schlim 394 (holotipo P!). Blechnum buchtienii Rosenst., Repert. Spec. Nov. Regni Veg. 5: 231. 1908. Struthiop- teris buchtienii (Rosenst.) Maxon & C. V. Morton, Bull. Torrey Bot. Club 66: 44. 1939. PROTÓLOGO: Yungas borealis; Undua- vi, 3 800m alt., in silvis, 12-II-1907., Dr. O. Buchtien nº 878. TIPO: Bolivia (Nor Yungas), Unduavi, Buchtien 878 (holotipo B!, isótipo US 867062!). TRATAMIENTO TAXONÓMICO DE LAS ESPECIES schomburgkii varían en rigidez y enrollamiento marginal en ejemplares juveniles y adultos, pero las láminas no alcanzan el desarrollo que se observa en B. moritzianum o en B. werckleanum. La mencio- nada variación externa de B. schomburgkii ha derivado en determinaciones diferentes, ya que en los herbarios aparece indistintamente como B. auratum subsp. auratum, B. buchtienii o B. schomburgkii. Del mismo origen puede ser la determinación de ejemplares de B. moritzia- num como B. schomburgkii. Dittrich (2005) lo reconoce para Brasil, señala que tiene afinida- des con B. tabulare, e incluye en la sinonimia a B. exiguum Dutra y B. bradei Markgraf, cuyos tipos proceden de Rio Grande do Sul. El holótipo R de B. exiguum no ha sido visto aquí, aunque sí se ha podido consultar el isótipo ICN. El tipo de B. bradei en B parece muy similar al isótipo ICN de B. exiguum, los nombres podrían ser sinónimos y ambos materiales están dentro de la variación de B. tabulare más que de B. schomburgkii. La distribución geográfica de B. schombugkii parece más bien mesoa- mericana y sudamericana andina y tal vez la especie no se encuentre realmente en Brasil, un área en la que en cambio, está presente B. tabulare en ambientes que no serían los típicos de B. schomburgkii, una planta de pajonales de altura, selvas, bosques montanos y páramos. Distribución geográfica y ecología: Mesoamérica, Sudamérica (Colombia, Vene- zuela, Ecuador, Perú, Bolivia, Brasil, Paraguay). Crece en selvas montanas y pedemontanas, en bordes de selvas, áreas boscosas húmedas, hasta pantanosas y desde los 1 400-1 500 hasta los 3 800m de altitud. Observaciones: Blechnum schomburgkii es una especie insuficientemente caracteriza- da hasta el momento. Tryon & Stolze (1993) la consideran diferente de B. auratum subsp. auratum y tratan sub B. schomburgkii las formas con pinnas patentes, anchas y con márgenes planos que aquí se han agrupado sub B. moritizianum o B. werckleanum, ambas especies abundantes en Perú. Moran (1995) no menciona B. auratum subsp. auratum y trata el material mesoamericano como B. buchtie- nii. Dittrich (2005) señala que es una especie variable, posiblemente afín con B. tabulare, señalando que requiere más estudios. Ramos Giacosa (2010) describe B. yunguense sobre material argentino de B. moritzianum y señala que las diferencias con B. schomburgkii se vin- culan con el contorno de la lámina, la anchura de las pinnas, las escamas y las esporas. El material tipo de B. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Blechnum buchtienii Rosenst., Repert. Spec. Nov. Regni Veg. 5: 231. 1908. Struthiop- teris buchtienii (Rosenst.) Maxon & C. V. Morton, Bull. Torrey Bot. Club 66: 44. 1939. PROTÓLOGO: Yungas borealis; Undua- vi, 3 800m alt., in silvis, 12-II-1907., Dr. O. Buchtien nº 878. TIPO: Bolivia (Nor Yungas), Unduavi, Buchtien 878 (holotipo B!, isótipo US 867062!). Lomaria obtusifolia Presl, Tent. Pterid. 143. 1836. Blechnum obtusifolium (Ettingsh.) C. Chr., Denkschr. Kaiserl. Akad. Wiss. Wien 23: 59. 1864. TIPO: [Según Presl]“Herb. Bras. Reg. Berol. 101”. Según Dittrich (2005), hay dos ejemplares con este nombre: Brasilia, Sellow.1825 (B 20-0031991!) y otro en BM, con la misma indicación. Lonchitis-aspera tabularis var. saligna Farwell, Amer. Midl. Naturalist 12: 277. 1931. TIPO: Bolivia, Mapiri, Rusby 313 (holotipo? isótipo US 834067!). Plantas terrestres, de hasta 3m de altura, con rizomas erectos no estoloníferos, escamo- sos. Escamas rizomáticas aciculares rígidas, curvadas todas en la misma dirección, pluries- tratificadas, con área central oscura y márgenes hialinos, lustrosas, de 2.5-4.5cm de largo. Frondas dimorfas, con láminas oblanceolado- elípticas, pinnadas, coriáceas. Estípites rígidos, con un surco adaxial y escamas basales simila- res a las del rizoma. Raquis y costas con esca- mas triangulares largas, pardo-amarillentas, concoloras o discoloras, caducas a persistentes. Costas abaxialmente carenadas y adaxialmente 6. Blechnum schomburgkii (Klotzsch) C. Chr., Index filic.: 159. 1905. Lomaria schomburgkii Klotzsch, Linnaea 20: 346. 1847. PROTÓLOGO Y TIPO: Guiana angl. Richard Schomburgk n. 1162 (holotipo B!, fragm NY!, isótipo K). Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 400 el herbario B, incluyendo el de B.schomburgkii var. squamulosa, ilustran la variación morfoló- gica de la especie, cuyas pinnas son ascenden- tes, de contorno linear a lanceolado angosto con ápices agudos a obtusos, bases sésiles y márgenes revolutos. El tipo de Lomaria aurata Fée, basónimo de B. auratum subps. auratum, que se encuentra en el herbario P y el de B. butchienii son considerados aquí también como parte de la variación morfológica esperable de la especie. Las pinnas de B. schomburgkii varían en rigidez y enrollamiento marginal en ejemplares juveniles y adultos, pero las láminas no alcanzan el desarrollo que se observa en B. moritzianum o en B. werckleanum. La mencio- nada variación externa de B. schomburgkii ha derivado en determinaciones diferentes, ya que en los herbarios aparece indistintamente como B. auratum subsp. auratum, B. buchtienii o B. schomburgkii. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Del mismo origen puede ser la determinación de ejemplares de B. moritzia- num como B. schomburgkii. Dittrich (2005) lo reconoce para Brasil, señala que tiene afinida- des con B. tabulare, e incluye en la sinonimia a B. exiguum Dutra y B. bradei Markgraf, cuyos tipos proceden de Rio Grande do Sul. El holótipo R de B. exiguum no ha sido visto aquí, aunque sí se ha podido consultar el isótipo ICN. El tipo de B. bradei en B parece muy similar al isótipo ICN de B. exiguum, los nombres podrían ser sinónimos y ambos materiales están dentro de la variación de B. tabulare más que de B. schomburgkii. La distribución geográfica de B. schombugkii parece más bien mesoa- mericana y sudamericana andina y tal vez la especie no se encuentre realmente en Brasil, un área en la que en cambio, está presente B. tabulare en ambientes que no serían los típicos de B. schomburgkii, una planta de pajonales de altura, selvas, bosques montanos y páramos. surcadas, con un surco profundo, con el haz vascular trabado por tejido parenquimático esclerosado y fibroso hacia las caras adaxial y abaxial. Pinna terminal discreta, linear o linear lanceolada, conforme pero más corta y angos- ta que las laterales, con ápice agudo rígido y base uni o bi-auriculada, a veces adnata con el primer par apical. Pinnas laterales hasta 40-50 pares, sésiles, marcadamente ascenden- tes, próximas pero no imbricadas, lanceoladas angostas a anchas, discretas, con ápices agudos u obtusos, márgenes repandos revolutos desde la base hasta el ápice, carenadas, con venas distanciadas inmersas pero visibles, simples y bifurcadas una vez cerca de la costa, algu- nas geminadas, con una sola dicotomía de ramas claramente divergentes. Indusio cartá- ceo, firme, de color castaño claro, submarginal o marginal, lacerado, de hasta 0.5cm de ancho. el herbario B, incluyendo el de B.schomburgkii var. squamulosa, ilustran la variación morfoló- gica de la especie, cuyas pinnas son ascenden- tes, de contorno linear a lanceolado angosto con ápices agudos a obtusos, bases sésiles y márgenes revolutos. El tipo de Lomaria aurata Fée, basónimo de B. auratum subps. auratum, que se encuentra en el herbario P y el de B. butchienii son considerados aquí también como parte de la variación morfológica esperable de la especie. Las pinnas de B. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES San José: Villa Mills, hacia el Cerro de la Muerte, Gabriel y Galán & Sundue s.n. (MA). ECUA- DOR: Azuay: Sevilla de Oro, en el camino de Sevilla de Oro-Méndez, llegando al Páramo del Castillo, Garmendia & Cisneros 1138 & 1106 (MA 630212 & 630213). Carchi, Tulcán, Camino Tufio, Aguas Hediondas en la base del Volcán Chiles, límite con Colombia, 3 650m, Palacios 11823 (UC 1744624). Loja: km 51 on Panamerican Highway, 2 900m, Holm-Nielssen et al. 4677 (UC 1477913). PERÚ: Cuzco: Camino a Buenos Aires, Kcosñipata, Peralta 26 (USM 214611). Mariscal Cáceres, Río Abiseo National Park, 3 300m, s. coll. 2762 (USM 83386). Puno, Sandía, 3 250-3 400m, Metcalf 30519 (LIL 146372). San Martín: Mariscal Cáceres, Chochos, NW corner of Río Abiseo National Park, 3 500m, Young & León 4824 (USM 214612). VENEZUELA: Amazonas: Depto Atabapo, Cerro Marahuaca, Liebner 24709 (UC 1557150). Trujillo: Muni- cipio Boconó, Parque Nacional Guaramacal, Páramo de Vicuyal, 2 730m, Stergios 20148 (UC 1784887). imperiale (Fée & Glaz.) H. Christ, in Schwac- ke, Pl. nov. mineir. 2: 27. 1900. Struthiopteris imperialis (Fée & Glaziou) Ching, Sunyatsenia 5: 243. 1940. TIPO: Brasil: Río de Janeiro, Serra dos Órgãos, Glaziou 2801 (holotipo P!, isótipo B!, foto BM!). Blechnum spannagelii Rosenst., Hedwigia 46: 93. 1907. PROTÓLOGO: [Brasil, Estado Santa Catharina] Lages, an schattigen Orten, C. Spanna­gel n. 86. TIPO: Brasil: Santa Catarina, Lages, Spannagel 86.1 (holótipo B!; isótipos HB, NY!, S, UC!, US!). Blechnum spannagelii f. pectinata Rosenst., Hedwigia 46: 93. 1907. PROTÓLO- GO: [Brasil, Estado San­ta Catharina] Lages, C. Spannagel n. 86.1. TIPO: Brasil: Santa Catarina, Lages, Spannagel 86.1 (holótipo B!; isótipos HB, NY!, S, UC!, US!). Blechnum madagascariense Tardieu, Mém. Inst. Sci. Madagascar, Sér. B, Biol. Vég. 6: 230, f. 3. 1955. PROTÓLOGO Y TIPO: Madagascar: Vallée de la Lokoho (NE), Mont Beondroka au nord de Maroambihy, rochers de la crête, 1 400-1 450m, Humbert H. 23466 (Holótipo?, Isótipos P00518459! P00483203!). Blechnum madagascariense Tardieu, Mém. Inst. Sci. Madagascar, Sér. B, Biol. Vég. 6: 230, f. 3. 1955. PROTÓLOGO Y TIPO: Madagascar: Vallée de la Lokoho (NE), Mont Beondroka au nord de Maroambihy, rochers de la crête, 1 400-1 450m, Humbert H. 23466 (Holótipo?, Isótipos P00518459! P00483203!). Plantas terrestres, con rizoma erecto de hasta 120cm, cubierto por las bases persis- tentes de los estípites, escamoso. Lomaria imperialis Fée & Glaziou in Fée, Cryp. vasc. Brésil 1: 21. 1869. Blechnum TRATAMIENTO TAXONÓMICO DE LAS ESPECIES schomburgkii depositado en Ejemplares de referencia seleccio- nados: BOLIVIA: Cochabamba: Chapare, Caja-Región, Llanta Aduana, Tarija, 3 200m, Steinbach 9551 (LIL 20657). Chapare, Inca- chaca-San Antonioni, 2 600m, Steinbach 9062 (LIL 96885). Santa Cruz, Ruinas Incaicas-El Fuerte, 2 000m, Steinbach 8259 (LIL 20684) Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 401 -planta de de la izquierda, la de la derecha es B. tabulare. COLOMBIA: Cundinamarca: PNN, Chingaza, Sendero Suasie, Acosta 024 (COL 466668). Al N de Bogotá, bosque arriba de la Hacienda La Francia (Torquita), 3 030m, Jara- millo Mejía, Murillo & van der Hammen 4575 (MA 380245). Tolima: Cordillera Oriental, ver- tiente oriental, bajando del Ruiz, carretera entre Letras y balcones, 3 400-3 200m, Cuatrecasas 9365 (COL 35766). COSTA RICA: Limón: Cordillera de Talamanca, Atlantic slope, unna- med cordillera, between Rio Terbi and Rio Siní, Davidse et al. 28996 (UC 1506870). San José: Villa Mills, hacia el Cerro de la Muerte, Gabriel y Galán & Sundue s.n. (MA). ECUA- DOR: Azuay: Sevilla de Oro, en el camino de Sevilla de Oro-Méndez, llegando al Páramo del Castillo, Garmendia & Cisneros 1138 & 1106 (MA 630212 & 630213). Carchi, Tulcán, Camino Tufio, Aguas Hediondas en la base del Volcán Chiles, límite con Colombia, 3 650m, Palacios 11823 (UC 1744624). Loja: km 51 on Panamerican Highway, 2 900m, Holm-Nielssen et al. 4677 (UC 1477913). PERÚ: Cuzco: Camino a Buenos Aires, Kcosñipata, Peralta 26 (USM 214611). Mariscal Cáceres, Río Abiseo National Park, 3 300m, s. coll. 2762 (USM 83386). Puno, Sandía, 3 250-3 400m, Metcalf 30519 (LIL 146372). San Martín: Mariscal Cáceres, Chochos, NW corner of Río Abiseo National Park, 3 500m, Young & León 4824 (USM 214612). VENEZUELA: Amazonas: Depto Atabapo, Cerro Marahuaca, Liebner 24709 (UC 1557150). Trujillo: Muni- cipio Boconó, Parque Nacional Guaramacal, Páramo de Vicuyal, 2 730m, Stergios 20148 (UC 1784887). -planta de de la izquierda, la de la derecha es B. tabulare. COLOMBIA: Cundinamarca: PNN, Chingaza, Sendero Suasie, Acosta 024 (COL 466668). Al N de Bogotá, bosque arriba de la Hacienda La Francia (Torquita), 3 030m, Jara- millo Mejía, Murillo & van der Hammen 4575 (MA 380245). Tolima: Cordillera Oriental, ver- tiente oriental, bajando del Ruiz, carretera entre Letras y balcones, 3 400-3 200m, Cuatrecasas 9365 (COL 35766). COSTA RICA: Limón: Cordillera de Talamanca, Atlantic slope, unna- med cordillera, between Rio Terbi and Rio Siní, Davidse et al. 28996 (UC 1506870). TRATAMIENTO TAXONÓMICO DE LAS ESPECIES La especie presenta cierta variación en el grado de división de la lámina, desde pinnas sésiles hasta adnatas, condición que es común tanto en ejemplares americanos como africanos; el extremo de esta condición corresponde en parte a plantas previamente asignadas a B. madagascariense y B. spanna- gelii, ésta considerada hasta ahora endémica de Brasil, aunque aquí se ha visto material de Paraguay con esa morfología. Blechnum span- nagelii fue considerado afín con B. brasiliense por Dittrich (2005) por la presencia en ambas de láminas pinnatisectas. Se la ha incluido en B. tabulare por los caracteres externos e internos tales como división de las láminas, contorno y márgenes de las pinnas, modelos epidérmicos, estructura de las costas, tipo de hipodermis y mesofilo. En relación con B. madagascariense, no hay menciones ni estudios posteriores a su diagnosis y descripción (Tardieu Blot 1955, 1960), pero esa autora ya sugiere que podría ser una “forma” reducida o pequeña de B. tabulare y esa sugerencia se ajusta a lo observado aquí en la morfología vegetativa y en Passarelli et al. (2010), en las esporas, aunque en ese trabajo se ilustraron separadamente de las de B. tabu- lare, por estar inconcluso el estudio actual. Los caracteres que llevan a incluirla en B. tabulare son los mismos señalados para B. spannagelii. Blechnum tabulare es común en áreas selváti- cas del S-SE de Brasil, Paraguay, NE y NO de la Argentina y Uruguay, y se la ha confundido con B. schomburgkii, que se considera aquí una especie preferentemente andina, de las laderas orientales de los Andes, páramos y áreas mon- tañosas de Venezuela y las Guayanas, menos común en áreas selváticas. El límite de área de ambas coincide en regiones montanas de las Yungas (Bolivia y NO de la Argentina). Distribución geográfica y ecología: Sud- américa: Argentina, Brasil, Bolivia, Paraguay y Uruguay. África: Nigeria, Camerún, Gabón, Zaire, Ruanda, Burundi, Etiopía, Zambia, Malawi, Mozambique, Zimbabue, Tanzania, Uganda, Congo, Angola, Kenia, Sudáfrica, Islas del océano Índico (Madagascar, Reunión, Mauricio) y del océano Atlántico (Marianas, Tristán da Cunha). En América tropical y sub- tropical crece en bosques húmedos, bordes de selvas, depresiones húmedas, cerca de áreas anegadas, en pastizales, pajonales y praderas graminosas, formando poblaciones no muy abundantes; las plantas pueden ser acidófilas y asociarse en áreas encharcadas con Sphagnum. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Escamas rizomáticas concoloras o discoloras y con un área central castaño-negruzca, sésiles, lineares, enteras, largamente atenuadas y terminadas en una sola célula, laxas o rígidas, curvadas todas en la misma dirección, con frecuencia retorcidas helicoidalmente, de hasta 6cm de largo. Frondas agrupadas erectas y rígidas o algo divergentes y laxas, las estériles de hasta 200cm de largo y las fértiles más largas. Estí- pites de 10-25cm de largo (los fértiles más largos y algo más gruesos), estriados o arru- gados, casi glabros excepto en la base que es densamente paleácea, con escamas similares a las del rizoma y con glándulas mucilaginíferas cilíndricas, esparcidas; internamente masivos. Raquis y costas subglabros, con glándulas mucilaginíferas unicelulares dispersas. Costas abaxialmente curvas y adaxialmente surcadas, con un surco poco profundo, internamente con 7. Blechnum tabulare (Thunb.) Kuhn, Fil. Afr.: 94 (1868). Pteris tabularis Thunb., Prodr. pl. cap.: 171 (1800). Lomaria tabularis (Thunb.) Mett. ex Baker, In Mart., Fl. bras. 1(2): 418 (1870). TIPO: [Sudáfrica] e Cap b. Spei, in lateribus (aquo) sis summi Taffel- berg, Thunberg s.n. (Holótipo UPS-THUNB 24965!). Lomaria imperialis Fée & Glaziou in Fée, Cryp. vasc. Brésil 1: 21. 1869. Blechnum Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 402 el haz inmerso en tejido esclerosado. Lámina estéril de hasta 110cm, elíptica, con ápice obtuso y base cuneada, gradualmente reduci- da, con hasta 14 pares de pinnas basales más distanciadas que las apicales, que pueden estar casi imbricadas en el ápice y con aflebias lami- nadas. Pinna terminal adnata con otros pares apicales, conforme, oblonga, más corta que las laterales. Pinnas laterales 40-42 pares, hori- zontales a ascendentes, oblongas, distanciadas, variablemente adnatas con el raquis a sésiles, con láminas planas, ápices obtusos a obtuso- mucronulados y márgenes repandos planos o basalmente revolutos; venas próximas, sim- ples, geminadas y bifurcadas, las bifurcadas divididas una o dos veces a distancias variables de la costa, adaxialmente inmersas y abaxial- mente visibles, sin hidatodos o con éstos poco apreciables. Indusio cartáceo, firme, de color castaño claro, submarginal o marginal, lacera- do, de hasta 5mm de ancho. Observaciones: Aquí se ha estudiado material adicional, no analizado en estudios previos de B. tabulare (Rolleri et al. 2008), de toda su área de distribución, que incluye Amé- rica tropical y África. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Rio Grande do Sul: Bom Jesus, Dutra 278 (SI). s. loc. Jürgens 11 (SI) Sã L ld R h 280 (LIL 147188) d´Itremo, W of Ambositra, van der Werff & McPherson 13587 (UC 1604667). Antsiranana: Montagne d’Ambre PN, near sommet d’Ambre, Andrianantoanina et al. 54 (UC 1872011). R. N. 1, 50km W of Antananarivo, Tapia forest, Barnett et al. 457 (UC 1549512). Antsiranana: Andapa, Begingotra, RS d´Anjanaharibe-Sud, sur le versant Sud.Est, 2km à l´Ouest-Sud Ouest de Befingra, 1 950-2 000m, Raconto- dranairibe & Raharimalala 2514 (P). Toliara: Andohahela RNI, Mont Trafanaomby, Parcelle I d´Andohahela Réserve Naturelle Intégrale, Tolagnaro, 1 000-1 957m, Randriamampio- nona 762 (P, UC). MALAWI: Mt. Mulanje, likhubula Valley, Chapalaka Trail, Chapman 7109 (UC 1587945). SUDÁFRICA: Transvaal: Woodbush Forest, Taylor 675 (UC 956260). Cape Town: Table Mountain, near Skeleton Gorge, Rodin 3211 (UC 802107). TANZA- NIA: Tanganiika: Irungi, Mufindi, N gwazi, Lovett & Thompson 2381 (UC 1585473). ZIM- BABUE: Orange Kop: Meltsetter, Wild 1953 (UC 085447). SUDAMÉRICA. ARGENTI- NA: Corrientes: Ituzaingó; desembocadura del arroyo Garapé en el río Paraná, 45km al E de Ituzaingó, Schinini et al. 11239 (CTES). Misio- nes: Iguazú, Paraje Aguaray, Lote P, bañado en rodal 5 y 6, sector anegado con Eryngium sp., Keller, Colcombet & Ferreira 1774 (MA 710843). BOLIVIA: Santa Cruz: Ruinas Incai- cas-El Fuerte, Steinbach 8259 (LIL 20684, planta de la derecha). BRASIL: Goiás: Serra do Cristais, 2km by Road of Cristalina, ele- vat. 1 200, Anderson 8148 (UC 1444720). Serra do Cristais, 2km by Road of Cristalina, Anderson 6403 (P, UC). Minas Gerais: Sierra Cipó, Palacio Hotel, de la Sota 2616 (LIL 473356). Paraná: Curitiba, Colombo, Dombro- vski 4442 & Kuniyoshi 3620 (CTES, duplica- dos en MBM, PACA, citas de Dittrich 2005). Río de Janeiro: RS-Jaquirana, Fazenda Boa Vista, 900m, Watsum & al. 3995 (MA 422162). Petropolis, Cuatrecasas 6137 (MA 233154, 9 pliegos). San Francisco de Paula, Villa Oliva, Rambo 31197 (LIL 206540 & 182815). Serra do Itatiaia, Dussén 168 (P). Rio Grande do Sul: Bom Jesus, Dutra 278 (SI). s. loc. Jürgens 11 (SI). São Leopoldo, Rohr 280 (LIL 147188). Santa Cruz, Rosenstock Filices Austrobrasi- liensis 189, Juergens s. n. (LIL 474039). Santa Catarina: “Santa Catarina, nos campos”, Reitz H680 (LIL146037). Bom Jardim, em bosques de Araucaria, Eskuche 0497 (CTES). Vienem bei Taquari, Serra do Mar, Ule 227 (P). São Paulo: Municipio Presidente Bernardes, Região do Pontal de Paranapanema, ca. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Toleran la exposición (no son frecuentes en selvas cerradas) pero requieren agua abundante en el suelo y crecen hasta los 2 000m. En Áfri- ca crece en selvas montanas como terrestre o epipétrica, a lo largo de corrientes de agua, no es edáficamente selectiva y produce frondas fértiles solamente durante el verano (Burrows 1990); se encuentra desde los 900 hasta los 1 600m de altitud. Ejemplares de referencia seleccionados: AFRICA. BURUNDI: Burundi: Nwisare, pres- so Burundi, Pichi Sermolli 7035 (UC 429945). MADAGASCAR: Antananarivo: Massif Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 403 d´Itremo, W of Ambositra, van der Werff & McPherson 13587 (UC 1604667). Antsiranana: Montagne d’Ambre PN, near sommet d’Ambre, Andrianantoanina et al. 54 (UC 1872011). R. N. 1, 50km W of Antananarivo, Tapia forest, Barnett et al. 457 (UC 1549512). Antsiranana: Andapa, Begingotra, RS d´Anjanaharibe-Sud, sur le versant Sud.Est, 2km à l´Ouest-Sud Ouest de Befingra, 1 950-2 000m, Raconto- dranairibe & Raharimalala 2514 (P). Toliara: Andohahela RNI, Mont Trafanaomby, Parcelle I d´Andohahela Réserve Naturelle Intégrale, Tolagnaro, 1 000-1 957m, Randriamampio- nona 762 (P, UC). MALAWI: Mt. Mulanje, likhubula Valley, Chapalaka Trail, Chapman 7109 (UC 1587945). SUDÁFRICA: Transvaal: Woodbush Forest, Taylor 675 (UC 956260). Cape Town: Table Mountain, near Skeleton Gorge, Rodin 3211 (UC 802107). TANZA- NIA: Tanganiika: Irungi, Mufindi, N gwazi, Lovett & Thompson 2381 (UC 1585473). ZIM- BABUE: Orange Kop: Meltsetter, Wild 1953 (UC 085447). SUDAMÉRICA. ARGENTI- NA: Corrientes: Ituzaingó; desembocadura del arroyo Garapé en el río Paraná, 45km al E de Ituzaingó, Schinini et al. 11239 (CTES). Misio- nes: Iguazú, Paraje Aguaray, Lote P, bañado en rodal 5 y 6, sector anegado con Eryngium sp., Keller, Colcombet & Ferreira 1774 (MA 710843). BOLIVIA: Santa Cruz: Ruinas Incai- cas-El Fuerte, Steinbach 8259 (LIL 20684, planta de la derecha). BRASIL: Goiás: Serra do Cristais, 2km by Road of Cristalina, ele- vat. 1 200, Anderson 8148 (UC 1444720). Serra do Cristais, 2km by Road of Cristalina, Anderson 6403 (P, UC). Minas Gerais: Sierra Cipó, Palacio Hotel, de la Sota 2616 (LIL 473356). Paraná: Curitiba, Colombo, Dombro- vski 4442 & Kuniyoshi 3620 (CTES, duplica- dos en MBM, PACA, citas de Dittrich 2005). Río de Janeiro: RS-Jaquirana, Fazenda Boa Vista, 900m, Watsum & al. 3995 (MA 422162). Petropolis, Cuatrecasas 6137 (MA 233154, 9 pliegos). San Francisco de Paula, Villa Oliva, Rambo 31197 (LIL 206540 & 182815). Serra do Itatiaia, Dussén 168 (P). TRATAMIENTO TAXONÓMICO DE LAS ESPECIES Ejemplares de referencia seleccionados: COLOMBIA: Cauca: Macizo Colombiano, Páramo de Las Papas entre El Boquerón y La Hoyola, 3 000-3 500m, Hidrobo et al. 3451 (COL 64782). Cundinamarca: San Miguel, extremo SW de la Sabana de Bogotá, 2 800m, Jaramillo Mejía, Murillo & van der Hammen 3422 (MA 3480728). Cordillera Oriental, Pára- mo de Guasca, Laguna, 3 200m, Cuatrecasas 2163 (MA 213337). Municipio de San Fran- cisco, Vereda Sabaneta, Hacienda San Pedro, 2 600-2 800m, Medina & Zamudio 90 (COL 253210). Huila: Comisaría del Caquetá, Cor- dillera oriental, sobre el filo divisorio, El Gabi- nete, 2 300-2 450m, Cuatrecasas 8490 (COL 14924). Nariño: Páramo El Consuelo, 10km al O de la Victoria, 3 100m, Hagemann & Leist (COL 236369). PANAMÁ: Bocas del Toro: 1 000m, Churchill et al. 4510 (UC 1506934). Chiriquí, along trail between North fork of Rio Palo Alto and Cerro Pate Macho, ca. 6km NE of Boquete, Smith et al. 2358 (UC 1524194). PERÚ: Cuzco: Paucartambo, 3 350m, León & Huallapa 2682 (USM). Piura: Huancabamba, above Huancabamba, road to Canchaque, 3 000m, Hutchinson 1620 (UC, USM 214614). VENEZUELA: Amazonas, Cerro de la Nebli- na, Camp 2, Neblina massif, 2.8km NE of Pico Phelps, 2 700m, Stein & Gentry 1537 (UC 1511485). Bolívar, Distrito Pilar, Camarcarba- rai-tepui, summit 4th of 4 tepuis from W to E in Asparaman range, 2 300m, Holst 3636 (UC 1535660). Distribución geográfica y ecología: Costa Rica, Panamá, Venezuela, Colombia, Ecuador, Perú. Selvas de neblina y páramos; 1 000-2 500m de altitud. Observaciones: Moran (1995) señala que B. werckleanum puede ser tanto terrestre como epífito, que carece de aeróforos y que la lámi- na seca es de color pardo oscuro en el haz y amarillo en el envés, algo que es útil en las identificaciones de material de herbario y que se ha comprobado aquí. Este autor sugiere que las plantas mesoamericanas podrían ser coes- pecíficas con B. schomburgkii, pero aquí se la trata como una especie característica, afín con B. moritzianum y con una distribución geográ- fica similar. Las pinnas discretas, ascendentes, curvadas acroscópicamente, distantes, subpe- cioluladas, discolores, con bases asimétricas y ápices caudados aserrados, difieren claramente de las pinnas de B. schomburgkii y son algo similares a las de B. moritzianum, aunque en esta especie son horizontales, más próximas, pecioluladas, con bases subcordiformes y ápi- ces enteros agudos a subobtusos. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES 430m de alt., Rodovia SP-272 Pirapozinho-Mirante do Para- napanema, entre kms 22-26, Silva & Fernandes 383 (CTES). Campos de Jordão, Leite & Leite 3504 (LIL 162347). PARAGUAY: Alto Paraná, Reserva Biológica Limoy, Caballero Marmo- ri 1597 (CTES). Caaguazú, Yhu, Sparre & Vervoorst 2042 (LIL 382985). In regionis flu- minis Alto Paraná, Fiebrig 5714 (LIL 41755). Guairá:Colonia Independencia, Schinini 1107 (CTES). In viam Caaguazú, Hassler 9389 (P). URUGUAY: Canelones, Bañados de Pando, Lombardo s. n. (LIL 206930). Bañados de Pando, Rosengurst B-4328 (LIL 15672). Baña- dos de Guazubirá, en la base de las dunas esta- bles Carretera interbalnearia, Prada s. n (MA). El Bañado de Los Pájaros, cerca de la carretera de Montevideo a Rocha, pr. Guazuvirá, Fuertes & Prada s. n. (MA). 8. Blechnum werckleanum (H. Christ) C. Chr., Index filic.: 161. 1905. Lomaria werc- kleana H. Christ, Bull. Herb. Boissier, ser. 2, 4(11): 1091. 1904. Struthiopteris werckleana (H. Christ) Broadh., Bull. Torrey Bot. Club 39: 382. 1912. PROTÓLOGO: I. Wercklé 169. 1903. TIPO: Costa Rica, Wercklé 169 (US 575242-575243!; isótipo NY!). Plantas terrestres con rizoma erecto de hasta 120cm de alto, escamoso. Escamas rizo- máticas lineares, de color castaño oscuro, rígi- das, brillantes, curvadas todas en la misma dirección, hasta de 3.5cm de largo. Frondas dimorfas, pinnadas, hasta de 180cm de largo. Estípites rígidos adaxialmente surcados y con escamas basales similares a las rizomáticas. Láminas estériles oblongas, abruptamente reducidas en la base con algunas pinnas más cortas pero no vestigiales. Raquis y costas con escamas caducas pardo-amarillentas. Costas Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 404 schomburgkii es frecuente, tanto en el caso de B. werckleanum como de B. moritzianum. abaxialmente curvas y adaxialmente surcadas, con un surco poco profundo, con el haz vas- cular trabado por tejido parenquimático escle- rosado que puede formar casquetes fibrosos independientes del haz. Pinna terminal con- forme, en general más larga que las laterales, no adnata y prolongada en un ápice aserrado. Pinnas laterales 25-45 pares, herbáceas firmes, papiráceas y con márgenes rasgados cuando secas, distanciadas, oblongas angostas, ascen- dentes, sub-pecioluladas a pecioluladas, con ápice acroscópicamente curvado, caudado y aserrado, margen repando plano, abaxialmente más claras y cuando secas, con el envés pardoa- marillento. Venas visibles, próximas, simples y bifurcadas una vez cerca de la costa, algunas geminadas, y siempre con una sola división en dos ramas claramente divergentes. Indusios eroso-fimbriados. TRATAMIENTO TAXONÓMICO DE LAS ESPECIES En ambas las láminas de las pinnas están más desarro- lladas que en otras especies arborescentes y los márgenes son planos. La confusión con B. DISCUSIÓN Se ha realizado una revisión de ocho espe- cies arborescentes de Blechnum: B. brasiliense, B. columbiense, B. cycadifolium, B. magella- nicum, B. moritzianum, B. schomburgkii, B. tabulare y B. werckleanum. Todas estas espe- cies tienen en común el hábito arborescente y las escamas aciculares, dos caracteres que originalmente se utilizaron para su agrupación (Tryon & Tryon 1982), y el xilema de las costas Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 405 en forma de V de brazos divergentes. Con excepción de B. brasiliense, también tienen en común las pinnas coriáceas, los haces costales trabados con esclerénquima, la presencia de 1-2 capas de hipodermis con células de paredes engrosadas, lignificadas o no, y las esporas con perisporio muriforme rugado, espaciadamente a densamente plegado, con muros bajos redon- deados no anastomosantes. diferencias como profundidad y frecuencia de las sinuosidades, ondas simples o compuestas, engrosamiento parietal general o en las curvas y ángulos y la relación l/a celular, son más evi- dentes por la ausencia de los estomas. p Los tipos esporales han sido analizados en detalle por Passarelli et al. (2010). Las esporas de la mayoría de las especies tienen perisporio plegado, con excepción de B. brasiliense que tiene perisporio venuloso. Esta especie también difiere del resto de los taxones arborescentes por las pinnas planas con márgenes aserrados y la presencia de protuberancias intercelulares pécticas en el parénquima que rodea al haz costal, aunque tiene en común el hábito, las escamas del rizoma y base del estípite y el tipo de haz costal. Las esporas de B. tabulare tienen una notable constancia en el área de la especie, con perisporios plegados y pliegues más bajos, menos evidentes que los que se presentan en las especies arborescentes andinas o de áreas templado-frías; las semejanzas con las esporas de B. madagascariense se señalaron previa- mente (Passarelli et al. 2010) al igual que las de B. schomburgkii y B. auratum subsp. auratum, aunque en ese trabajo se ilustraron separada- mente, por estar por entonces inconcluso el estudio actual. DISCUSIÓN Entre los caracteres estudiados relaciona- dos con los ejes es destacable la constancia en la estructura vascular de las costas, diferente de la de otros grupos de especies afines (datos no publicados), constituida por un haz con xilema en V de brazos divergentes, rodeado por una endodermis y asociado con tejido esclerosado que forma una vaina perivascular que alcanza su mayor desarrollo (haces trabados adaxial- abaxialmente) en especies que crecen entre 1 200 y más de 3 000m (páramos, bosques fríos, bosques montanos tropicales) como B. colum- biense, B. cycadifolium, B. magellanicum, B. moritzianum, B. schomburgkii, B. tabulare y B. werckleanum, mientras que falta en B. brasi- liense, una especie que crece a menor altura en selvas lluviosas. El desarrollo de este refuerzo de las pinnas podría estar relacionado con otro carácter no previamente estudiado en este grupo, que es la hipodermis. Esta capa dermatoide adaxial está formada por hasta tres capas de células dispuestas de manera continua. Las paredes celulares están lignificadas y las células son traqueiformes y largas en B. columbiense, B. cycadifolium y B. schomburgkii, mientras que no están lignificadas, son más cortas y tienen paredes menos engrosadas en B. magellanicum, B. moritzianum, B. tabulare y B. werckleanum, pero aún así contituyen un refuerzo mecánico continuo adicional para las pinnas. Las paredes de las células hipodérmicas son más delgadas que las de las epidérmicas en B. brasiliense. En todas las especies, con excepción tam- bién de B. brasiliense, se presenta además un mesofilo compacto. La clave para la determinación de las espe- cies está basada en caracteres que se pueden apreciar en material vivo pero que se mantie- nen sin cambio en material de herbario y cuya apreciación no es difícil si se trabaja con mate- rial herborizado. Esto se ha hecho así porque a menudo este material es el único disponible para el botánico y en el caso particular de las especies arborescentes es escaso. Blechnum schomburgkii y B. tabulare han sido revisados a la luz de nuevos enfoques de su morfología externa e interna, así como de su distribución geográfica, ecología y afinidades. Blechnum schomburgkii se considera aquí una especie preferentemente andina, presente en las laderas orientales de los Andes, los páramos y áreas montañosa de Venezuela y las Guayanas; es menos común en áreas selváticas del S-SE de Brasil, donde la especie se ha confundido con B. tabulare; es afín con B. Resumen y B. werckleanum, dos especies que frecuen- temente se determinan como B. schomburgkii en los herbarios, especialmente la primera. La variación morfológica que se podría atribuir a B. schomburgkii en función de las diferen- tes determinaciones halladas en material de herbario no es importante ni amplia si se la caracteriza como aquí se ha hecho, y se la dife- rencia de B. moritzianum y B. werckleanum. El análisis de los ejemplares de toda el área de distribución de B. schomburgkii ha permitido caracterizar de una forma más precisa las tres especies. En el caso de B. tabulare el análisis de ejemplares de toda su área geográfica ha llevado a reunir bajo ese nombre materiales que representaban parte de la variación mor- fológica de la especie y que se trataban como especies diferentes. La bibliografía sobre el género Blechnum L. es muy abundante tanto en lo que respecta a las especies neotropicales como paleotropicales. Sin embargo, aún faltan estudios anatómicos completos, una caracterización actualizada de su morfología externa e interna, taxonomía, citología y afinidades. En este estudio se revisaron ocho especies de Blechnum de hábito arborescente: B. brasi- liense, B. columbiense, B. cycadifolium, B. magellanicum, B. moritzianum, B. schomburgkii, B. tabulare y B. werc- kleanum. Se analizaron caracteres de rizomas, estípites y costas, división de la lámina, y venación, estructura de márgenes, modelos epidérmicos, presencia y tipos de hipo- dermis y mesofilo de las pinnas. Se estudió por primera vez la hipodermis, un área dermatoide adaxial presente en el mesofilo de todas las especies. Se correlacionaron caracteres externos e internos con el fin de utilizar los externos más confiables en claves y descripciones. Los estudios realizados confirmaron nuevamente la presencia de B. tabulare en América, que incluyen en su sinonimia a B. spannagelii de América tropical y B. madagascariense de África. También se concluyó que el nombre B. schom- burgkii es un sinónimo de B. auratum subsp. auratum, en tanto que B. auratum subsp. columbiense se trata aquí con rango de especie. Se caracterizó a B. moritzianum, el nombre que corresponde aplicar a la recientemente descrita B. yungense. Se presentó una descripción actualizada de las especies, se resolvieron varios problemas nomenclaturales pendientes; se presentaron datos amplios sobre la ecología, distribución geográfica y afinidades entre las especies, así como una clave para determinarlas. DISCUSIÓN moritzianum Los modelos epidérmicos y la combinación de los tipos de epifilos e hipofilos son caracte- res específicos, pero en los epifilos ciertas Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 406 AGRADECIMIENTOS Este trabajo fue realizado con el apoyo de las siguientes Instituciones: Ministerio de Ciencia e Innovación de España (Proyecto CGL 2009-13622, Subprograma BOS), Uni- versidad Complutense de Madrid, España, Consejo Nacional de Investigaciones Científi- cas y Técnicas (CONICET), Buenos Aires y Programa de Incentivos para Docentes Inves- tigadores de la Universidad Nacional de La Plata (Argentina). El Ministerio de Ciencia e Innovación de España apoyó la estadía de C. H. Rolleri en la Universidad Complutense de Madrid durante los meses de febrero a junio de 2010 a través de su Programa de Visitantes Distinguidos Universidad Complutense-Grupo Santander y la estancia de L.M. Passarelli en la Universidad Complutense en noviembre 2010. El Programa de Becas Complutense-del Amo financió una estancia de investigación de J.M. Gabriel y Galán en el Herbario de la Univer- sidad de California en Berkeley. Se agradece la revisión crítica de aspectos nomenclaturales a Alberto Herrero, Real Jardín Botánico de Madrid, España. Palabras clave: revisión, especies arborescentes, morfolo- gía, nomenclatura, caracteres diagnósticos. Palabras clave: revisión, especies arborescentes, morfolo- gía, nomenclatura, caracteres diagnósticos. Palabras clave: revisión, especies arborescentes, morfolo- gía, nomenclatura, caracteres diagnósticos. REFERENCIAS A revision of the African spe- cies of Blechnum. J. Linn. Soc. Bot. 53: 487-510. Moran, R.C. 1995. Blechnaceae, p. 325-333. In G. David- se, M. Souza & S. Knapp (eds.). Flora Mesoame- ricana I, Psilotaceae a Salviniaceae. Universidad Autónoma de México, Missouri Bot. Gard. & The Nat. Hist. Museum, Londres, Inglaterra. Skottsberg, C. 1924. The phanerogams of the Juan Fer- nández Islands, p 95-240. In C. Skottsberg (ed.). The Natural History of Juan Fernández and Easter Island, Vol II Botany. part VI. Almqvist & Wiksells Boktryc- keri AB, Uppsala, Suecia. Moore, T. 1857. Index Filicum. W. Pamplin, Londres, Inglaterra. Skottsberg, C. 1953. The vegetation of the Juan Fernandez Islands, p. 793-959. In C. Skottsberg (ed.). The Natu- ral History of Juan Fernández and Easter Island, Vol II. Botany. Part VI. Almqvist & Wiksells Boktryckeri AB, Uppsala, Suecia. Passarelli, L.M., J.M. Gabriel y Galán, C. Prada & C.H. Rolleri. 2010. Spore morphology and ornamentation in the genus Blechnum (Blechnaceae). Grana 49: 243-262. Stolze, R.G. 1981. Ferns and fern allies of Guatemala. Part 2: Polypodiaceae. Fieldiana Bot. n.s. 6: 1-522. Payne, W.W. & K.M. Peterson. 1973. Observations of the hypodermises of ferns. Am. Fern J. 63: 34-42. Tardieu-Blot, M.L. 1955. Blechnum. In H. Humbert (ed.). Une merveille de la nature à Madagascar. Première exploration botanique du Massif du Marojejy et de ses satellites. Mém. Inst. Sci. Madagascar, Sér. B, Biol. Vég. 6: 230. Prabhakar, M. 2003. Structure, delimitation, nomenclature and classification of stomata. Acta Bot. Sin. 46: 242-252. Prada, C. & C.H. Rolleri. 2005. A new species of Isoetes (Isoetaceae) from Turkey, with a study of microphyll intercellular pectic protuberances and their potential taxonomic value. Bot. J. Linn. Soc. 147: 213-228. Tardieu-Blot, M.L. 1960. Blechnum. In H. Humbert (ed.). Flore de Madagascar et des Comores, 2: 1-19. Polypodiacees s. lat. (Blechnacees a Polypodiacees s.str.). Museum National d’Histoire Naturelle, Paris, Francia. Prada, C., C.H. Rolleri & L.M. Passarelli. 2008. Morfolo- gía, caracterización y distribución geográfica de Ble- chnum cordatum (Blechnaceae-Pteridophyta). Acta Bot. Malac. 33: 29-46. Tryon, R.M. & A.F. Tryon. 1982. Ferns and Allied Plants with Special Reference to Tropical America. Sprin- ger, Nueva York, EE.UU. Ramos Giacosa, J.P. 2010. Blechnum yungense (Pterido- phyta, Blechnaceae), una nueva especie de Argentina y Bolivia. Novon 20: 68-72. Tryon, R.M. & H. Lugardon 1990. Spores of the Pterido- phyta. Springer, Nueva York, EE.UU. Ramos Giacosa, J.P., M.A. Morbelli & G.E. Giúdice. 2009. REFERENCIAS Baeza, M., E. Barrera, J. Flores, C. Ramírez & R. Rodrí- guez. 1998. Categorías de conservación de Pterido- phyta nativas de Chile. Bol. Mus. Nac. Hist. Nat. 47: 23-46. Barrera, E. 1997. Helechos de Juan Fernández. Museo Nacional de Historia Natural, Publicación Ocasional Nº 51. Bower, F.O. 1923. The Ferns. Cambridge University, Cam- bridge, England. Burrows, J.E. 1990. Southern African ferns and fern-allies. Fandsen, Sandton, South Africa. Danton, P.H. 2004. Plantas silvestres de la Isla Robinson Crusoe, Guía de reconocimiento. Orgraf, Chile. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 407 Ricci, M. 1990. Programa de conservación y recuperación de plantas amenazadas de Juan Fernández. Informe Final, 2ª etapa, Proyecto CONAF-WWF-3313, Chile. Dittrich, V.A. 2005. Estudos taxonômicos no gênero Ble- chnum L. (Pteridophyta-Blechnaceae) para as regiões Sudeste e Sul do Brasil. Tesis de Doctorado, Univer- sidad Estadual Paulista Júlio de Mesquita Filho, Rio Claro, São Paulo, Brasil. Ricci, M. 1992. Programa de conservación y recuperación de plantas amenazadas de Juan Fernández. Informe Final, 3ª etapa, Proyecto CONAF-WWF-3313, Chile. Esau, K. 1953. Plant Anatomy. Wiley, Nueva York, EE.UU. Font Quer, P. 1982. Diccionario de Botánica. Labor, Bar- celona, España. Rodríguez Ríos, R. 1995. Pteridophyta, p. 119-309. In C. Marticorena & R. Rodriguez (eds.). Flora de Chile. Vol. 1. Pteridophyta-Gymnospermae. Universidad de Concepción, Chile. Foster, A.S. 1936. Leaf differentiation in Angiosperms. Bot. Rev. 24: 304-316. Rolleri, C.H. & C. Prada. 2006a. Revisión de los grupos de especies del género Blechnum (Blechnaceae, Pterido- phyta): el grupo Blechnum penna-marina. Acta Bot. Malac. 31: 7-50. Gabriel y Galán, J.M., L.M. Passarelli, C. Prada & C.H. Rolleri. 2008. Study of sporophyte morphology and gametophyte development of Blechnum sprucei (Blechnaceae-Pteridophyta). Rev. Biol. Trop. 56: 2027-2040. Rolleri, C.H. & C. Prada 2006b. Catálogo comentado de las especies de Blechnum (Blechnaceae-Pteridophyta) de Mesoamérica y Sudamérica. Anales Jard. Bot. Madrid 63: 67-106. Hieronymus, G. 1908. Plantae Stübelianae. Hedwigia 47: 204-249. Johow, F. 1896. Estudio sobre la Flora de las Islas de Juan Fernández. Cervantes, Santiago de Chile, Chile. Rolleri, C.H., C. Prada & L.M. Passarelli. 2008. Estudios morfológicos y taxonómicos en Blechnum (Blechna- ceae-Pteridophyta): Blechnum tabulare y B. magella- nicum. Anales Jard. Bot. Madrid 65: 169-175. Lellinger, D.B. 2002. A Modern Multilingual Glossary of taxonomic Pteridology. Pteridologia 3: 1-246. Metcalfe, C.R. & L. Chalk. 1950. Anatomy of the Dicotyle- dons. Clarendon, Oxford, Inglaterra. Schelpe, E.A.C.L.E. 1952. REFERENCIAS Spore morphology and wall ultrastructure of Ble- chnum L. species from North West Argentina. Rev. Palaeobot. Palynol. 156: 185-197. Tryon, R.M. & R.G. Stolze. 1993. Pteridophyta of Peru. Part. 5: Aspleniaceae-Polypodiaceae. Fieldiana Bot.n.s. 32: 54-70. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 61 (1): 377-408, March 2013 408
https://openalex.org/W3033809025
https://europepmc.org/articles/pmc7286485?pdf=render
English
null
A Paleolithic bird figurine from the Lingjing site, Henan, China
PloS one
2,020
cc-by
13,806
RESEARCH ARTICLE A Paleolithic bird figurine from the Lingjing site, Henan, China RESEARCH ARTICLE A Paleolithic bird figurine from the Lingjing site, Henan, China Zhanyang Li1, Luc DoyonID1,2☯, Hui Fang1*, Ronan Ledevin2‡, Alain QueffelecID2‡, Emeline Raguin3‡, Francesco d’ErricoID2,4☯¤* Zhanyang Li1, Luc DoyonID1,2☯, Hui Fang1*, Ronan Ledevin2‡, Alain QueffelecID2‡, Emeline Raguin3‡, Francesco d’ErricoID2,4☯¤* 1 Institute of Cultural Heritage, Shandong University, Qingdao, Shandong Province, P.R. of China, 2 Centre National de la Recherche Scientifique CNRS UMR5199 PACEA, Universite´ de Bordeaux, Pessac, Nouvelle- Aquitaine, France, 3 Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel, 4 SFF Centre for Early Sapiens Behaviour (SapienCE), University of Bergen, Bergen, Norway ☯These authors contributed equally to this work. ¤ Current address: CNRS UMR5199 –PACEA, Universite´ de Bordeaux, Alle´e Geoffroy Saint Hilaire, Pessac, France ☯These authors contributed equally to this work. ¤ Current address: CNRS UMR5199 –PACEA, Universite´ de Bordeaux, Alle´e Geoffroy Saint Hilaire, Pessac, France a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 ‡ These authors also contributed equally to this work. ‡ These authors also contributed equally to this work. * francesco.derrico@u-bordeaux.fr (FdE); fangh@sdu.edu.cn (HF) ‡ q y * francesco.derrico@u-bordeaux.fr (FdE); fangh@sdu.edu.cn (HF) * francesco.derrico@u-bordeaux.fr (FdE); fangh@sdu. OPEN ACCESS Editor: Karen Hardy, Institucio Catalana de Recerca i Estudis Avancats, SPAIN Editor: Karen Hardy, Institucio Catalana de Recerca i Estudis Avancats, SPAIN Received: February 10, 2020 Accepted: May 4, 2020 Published: June 10, 2020 Copyright: © 2020 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: The Lingjing bird carving is curated at the Henan Provincial Institute of Cultural Relics and Archaeology, Zhengzhou, Henan Province (Repository ID: 09L5鸟-01). All data is available in the main text or the supplementary materials. Funding: This research was supported by grants given to ZL by the National Natural Science Foundation of China (grant number: 41630102 and 41672020; http://www.nsfc.gov.cn/english/site_1/ index.html), and to HF by the Shandong University 111 Project (grant number: 111-2-09; https://en. Abstract OPEN ACCESS Citation: Li Z, Doyon L, Fang H, Ledevin R, Queffelec A, Raguin E, et al. (2020) A Paleolithic bird figurine from the Lingjing site, Henan, China. PLoS ONE 15(6): e0233370. https://doi.org/ 10.1371/journal.pone.0233370 OPEN ACCESS The recent identification of cave paintings dated to 42–40 ka BP in Borneo and Sulawesi highlights the antiquity of painted representations in this region. However, no instances of three-dimensional portable art, well attested in Europe since at least 40 ka BP, were docu- mented thus far in East Asia prior to the Neolithic. Here, we report the discovery of an excep- tionally well-preserved miniature carving of a standing bird from the site of Lingjing, Henan, China. Microscopic and microtomographic analyses of the figurine and the study of bone fragments from the same context reveal the object was made of bone blackened by heating and carefully carved with four techniques that left diagnostic traces on the entire surface of the object. Critical analysis of the site’s research history and stratigraphy, the cultural remains associated with the figurine and those recovered from the other archeological lay- ers, as well as twenty-eight radiometric ages obtained on associated archeological items, including one provided by a bone fragment worked with the same technique recorded on the object, suggest a Late Paleolithic origin for the carving, with a probable age estimated to 13,500 years old. The carving, which predates previously known comparable instances from this region by 8,500 years, demonstrates that three-dimensional avian representations were part of East Asian Late Pleistocene cultural repertoires and identifies technological and stylistic peculiarities distinguishing this newly discovered art tradition from previous and contemporary examples found in Western Europe and Siberia. Citation: Li Z, Doyon L, Fang H, Ledevin R, Queffelec A, Raguin E, et al. (2020) A Paleolithic bird figurine from the Lingjing site, Henan, China. PLoS ONE 15(6): e0233370. https://doi.org/ 10.1371/journal.pone.0233370 Citation: Li Z, Doyon L, Fang H, Ledevin R, Queffelec A, Raguin E, et al. (2020) A Paleolithic bird figurine from the Lingjing site, Henan, China. PLoS ONE 15(6): e0233370. https://doi.org/ 10.1371/journal.pone.0233370 Editor: Karen Hardy, Institucio Catalana de Recerca i Estudis Avancats, SPAIN Received: February 10, 2020 Accepted: May 4, 2020 Published: June 10, 2020 Copyright: © 2020 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PLOS ONE PLOS ONE Introduction Our knowledge about the origins of symbolically mediated behaviors has substantially increased over the last twenty years. The idea of a symbolic explosion occurring in Europe 40,000 years ago [1–3], associated with the arrival of anatomically modern human populations PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 1 / 24 PLOS ONE A Paleolithic bird figurine from China sdu.edu.cn/). Funding was also provided to LD and FD by the Sino-French PHC Cai Yuanpei program (grant number: 36707NF; https://cn.ambafrance. org/-Le-programme-Cai-Yuanpei-1628-), the Programme Investissements d’Avenir IdEx of the Bordeaux University (https://idex.u-bordeaux.fr/fr/), and the Research Council of Norway through its Centre’s of Excellence funding scheme, SFF Centre for Early Sapiens Behaviour (SapienCE), project number 262618 (https://www.forskningsradet.no/ en/). L.D. was granted financial support from the China/Shandong University International Postdoctoral Exchange Program (http://www. chinapostdoctor.org.cn/) and the Sino-French PHC Xu Guangqi (grant number: 41230RB; https://cn. ambafrance.org/-Le-programme-XU-Guangqi-). PACEA (UMR5199 CNRS) is a Partner team of the Labex LaScArBx-ANR n˚ ANR-10-LABX-52 (https:// lascarbx.labex.u-bordeaux.fr/). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. sdu.edu.cn/). Funding was also provided to LD and FD by the Sino-French PHC Cai Yuanpei program (grant number: 36707NF; https://cn.ambafrance. org/-Le-programme-Cai-Yuanpei-1628-), the Programme Investissements d’Avenir IdEx of the Bordeaux University (https://idex.u-bordeaux.fr/fr/), and the Research Council of Norway through its Centre’s of Excellence funding scheme, SFF Centre for Early Sapiens Behaviour (SapienCE), project number 262618 (https://www.forskningsradet.no/ en/). L.D. was granted financial support from the China/Shandong University International Postdoctoral Exchange Program (http://www. chinapostdoctor.org.cn/) and the Sino-French PHC Xu Guangqi (grant number: 41230RB; https://cn. ambafrance.org/-Le-programme-XU-Guangqi-). PACEA (UMR5199 CNRS) is a Partner team of the Labex LaScArBx-ANR n˚ ANR-10-LABX-52 (https:// lascarbx.labex.u-bordeaux.fr/). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. in the region, has given way to a complex gradualist scenario [4–7]. Multiple evidence now demonstrates that behaviors generally associated with symbolic thought, such as producing abstract drawings and engravings, using pigments, wearing personal ornaments and perform- ing complex mortuary practices, are three to ten times older than what was acknowledged two decades ago. It is also becoming clear that these practices emerged gradually among both Afri- can Middle Stone Age populations and the so-called archaic populations living in Europe and Asia [8–16]. Figurative representations were considered until recently the only symbolic mani- festation for which Europe could claim precedence. Introduction This view was challenged in the last few years by the dating of 42,000-year-old calcite deposits covering animal, human, and hand sten- cil depictions at sites from Southeast Asia [17–19]. Furthermore, the modern human author- ship of all Paleolithic paintings has likewise been questioned with the dating of calcite deposits covering hand stencils and painted signs from three Iberian caves, suggesting that these repre- sentations were made some 63,000 years ago, a time at which only Neanderthal populations were living in Europe [20]. Although repeatedly disputed [21–24], these dates were produced with the same methods applied to establish the age of the earliest Southeast Asian cave paint- ings [25] and were obtained following strict protocols, which apparently duly considered pos- sible sources of error [26–28]. The carving of small figurines is, for the time being, the only artistic practice that may have originated in Europe and that could represent an innovation created by anatomically modern populations colonizing this region. The earliest known carv- ings consist of animal and human figurines sculpted in mammoth ivory, and were found at sites from the Swabian Jura, Germany, in layers containing Early Aurignacian artifacts and dated to circa 40–38 ka [29,30]. For vast regions of the world, however, it remains unclear when the production of three-dimensional representations became a part of the cultural reper- toire of prehistoric groups, and whether this happened independently or through diffusion from a point of origin. Competing interests: The authors have declared that no competing interests exist. Here we report the discovery of a diminutive carving, depicting a standing bird, found at the Paleolithic site of Lingjing, Henan, China. Careful consideration of the site’s research his- tory and stratigraphy, of the cultural remains it yielded in association with the figurine and from other archeological layers, as well as of the numerous radiometric ages obtained from the associated archeological items argues in favor of a Late Paleolithic origin for the carving, with a probable age of 13,500 years. The exceptional state of preservation of the figurine, unmatched by comparable Paleolithic carvings, and the application of state-of-the-art methodology to its study allowed us to reconstruct in detail the technology and evaluate the skills that led to its manufacture. Introduction Our results highlight peculiar features that make this carving the first known instance of an original artistic tradition, i.e., a body of technical, thematic, and stylistic traits applied to the production of symbolic artifacts shared by a society and transmitted to new gen- erations of artists [31]. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Archeological context and dating Lingjing is an open-air site located in Henan, 120 km south of the Yellow River (Fig 1a). This water-lain deposit was discovered in 1965 [32,33] and excavated yearly from 2005 to 2018 under the direction of one of us (LZ). The excavation, extending over a surface of 551 m2, identified eleven layers, numbered from 1 at the top to 11 at the bottom, within a 9 m deep sed- imentary sequence (Fig 1b). Seven layers contain archeological remains. The uppermost layers 1–4 are Holocene in age; they were identified over the entire excavated surface but were associ- ated with archeological material only along the northern edge of the site, i.e., >50 meters north of the area where layer 5 was identified in the stratigraphy. Layers 1–4 yielded a few dozen isolated, fine pottery sherds that could not be refitted with one another. Their outer PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 2 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 1. Localization and stratigraphy of the Lingjing site and 14C dating for the bird carving archeological context. (A) Location of Lingjing (Henan, China). (B) Stratigraphy of the sedimentary sequence with indication of main findings. (C) Calibrated 14C ages obtained on burnt bone, charcoal, and charred residues present on pottery sherds from the context in which the carving was recovered (see Table 1 for details). The red bar identifies the age of a burnt bone sample bearing traces of modification similar to those recorded on the carving. https://doi.org/10.1371/journal.pone.0233370.g001 Fig 1. Localization and stratigraphy of the Lingjing site and 14C dating for the bird carving archeological context. (A) Location of Lingjing (Henan, China). (B) Stratigraphy of the sedimentary sequence with indication of main findings. (C) Calibrated 14C ages obtained on burnt bone, charcoal, and charred residues present on pottery sherds from the context in which the carving was recovered (see Table 1 for details). The red bar identifies the age of a burnt bone sample bearing traces of modification similar to those recorded on the carving. https://doi.org/10.1371/journal.pone.0233370.g001 https://doi.org/10.1371/journal.pone.0233370.g001 https://doi.org/10.1371/journal.pone.0233370.g001 surface bears decors that allow their cultural attribution to periods spanning from the Yang- shao Neolithic (~6.5–5 ka BP) to the Shang-Zhou Bronze Age (~4–2.5 ka BP). Neither stone tools nor faunal remains were recovered from these layers. Layer 5 and sediments originating from it (see below) included artifacts reflecting an occupation spanning from the LGM to the Younger Dryas. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Archeological context and dating Layers 6 to 9 are sterile. Layers 10 and 11, attributed to the early Late Pleisto- cene, were dated by OSL between 99 ka and 118 ka [34]. Both layers yielded lithic artifacts and faunal remains [35]. Two incomplete human skulls were also found in layer 11. They bear a mosaic of morphological features interpreted as indicating both regional continuity and inter- regional population dynamics [33,36]. Analysis of faunal remains from layer 11 identified the earliest known evidence for pressure flaking [37], the first bone retouchers from East Asia [38], and the use of metapodials as organic soft hammer for marrow extraction [39]. Two weathered bone fragments bearing parallel engraved lines and traces of ocher were also found in this layer [15]. When one of us (LZ) excavated the site in 2005, he found that most of layer 5 had been removed by well diggers in 1958 and that only a small portion of this layer was still present in the stratigraphic profile near the southern limit of trench T1. The layers overlaying the rem- nants of layer 5 were completely sterile in this area. The excavation of the intact portion of layer 5 yielded a small amount of quartz tools, high quality black chert microliths, and pottery sherds. During the 2008 and 2013 excavation campaigns, the spoil heap left by the well diggers in 1958 was identified less than 10 m away from the water cistern built over the spring open- ing, which is located near the southern edge of the excavated area. Water sieving of these sedi- ments produced a rich microcore and microblade industry made of high-quality black chert, a raw material only found in layer 5, few pottery sherds, burnt and unburnt faunal remains, charcoal, ostrich egg shell fragments, a perforated ostrich egg shell pendant (Figs 2 and 3), and the bird figurine described in the present study. The pottery sherds found in the intact 3 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China Fig 2. Archeological remains associated with the bird carving. (A) Selection of microblade cores. (B) Retouched tools including scrapers, burins and points. (C) End-scrapers. (D) Ostrich egg shell pendant. (E) High quality black chert flake from layer 5. Scales = 1 cm. https://doi org/10 1371/journal pone 0233370 g002 Fig 2. Archeological remains associated with the bird carving. (A) Selection of microblade cores. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Archeological context and dating (B) Retouched tools including scrapers, burins and points. (C) End-scrapers. (D) Ostrich egg shell pendant. (E) High quality black chert flake from layer 5. Scales = 1 cm https://doi.org/10.1371/journal.pone.0233370.g002 https://doi.org/10.1371/journal.pone.0233370.g002 4 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China Fig 3. Burnt bone fragments associated with the bird carving. (A) Fragment dated by 14C bearing traces of gouging (close up at the right) comparable to those recorded on the carving. (B-T) Other fragments, some of which bearing traces of modification (see Table 2 for details). Scale = 1 cm. Fig 3. Burnt bone fragments associated with the bird carving. (A) Fragment dated by 14C bearing traces of gouging (close up at the right) comparable to those recorded on the carving. (B-T) Other fragments, some of which bearing traces of modification (see Table 2 f d il ) S l 1 Fig 3. Burnt bone fragments associated with the bird carving. (A) Fragment dated by 14C bearing traces of gouging (close up at the right) comparable to those recorded on the carving. (B-T) Other fragments, some of which bearing traces of modification (see Table 2 for details). Scale = 1 cm. https://doi.org/10.1371/journal.pone.0233370.g003 https://doi.org/10.1371/journal.pone.0233370.g003 https://doi.org/10.1371/journal.pone.0233370.g003 5 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China remnants of layer 5 and in the spoil heap differ greatly from those recovered from Holocene contexts i.e., layers 1–4. They are thick, crude, simple in shape, with plain surfaces, very fragile, and fired at low temperatures [40,41]. The lithic assemblage from layer 5 is dominated by pyra- midal type microblade cores, followed by boat-shaped and wedge-shaped cores [41–43]. Retouched tools include in decreasing order short end-scrapers, scrapers, and burins. Analysis of this assemblage identified consistencies in raw material use (Fig 2e), technology and typol- ogy supporting the syndepositional nature of the assemblage and striking similarities with Late Glacial industries from Northern China [40–43]. Aside from the few ostrich eggshell frag- ments, the faunal assemblage found in association with the figurine contains 215 remains con- sisting of 80 unidentifiable, blackened bone fragments, a quarter of which was radiocarbon dated (see below), as well as 135 fragmentary equids and bovids molars. Archeological context and dating Burnt bones, charcoals and charred residues from the pottery sherds recovered in the spoil heap in which the bird carving was found were radiocarbon dated (Fig 1c, Table 1) at three dating laboratories: the Institute of Accelerator Analysis Ltd., Kawasaki City, Kanagawa, Japan (IAAA), the University of Tokyo Radiocarbon Laboratory (TKa), and the Beijing University Radiocarbon Laboratory (MTC), China [40–43]. The thirty-six available ages, three of which obtained in the framework of this study (see below), were calibrated with the OxCal 4.3.2 online software [44] using the IntCal13 atmospheric calibration curve [45]. With the exception of a single old age obtained from a charcoal sample (IAAA-100080: 28,690±120 14C years), not included in Fig 1c, two distinct sets of ages are identified by the dated material. The first set, which includes all ages obtained on burnt bone and charcoal, ranges between ~13.8 and ~13.0 ka cal BP. Considering the age of Northern Chinese sites with comparable microblade assem- blages, this set covers the time span for the occupation of the site by Late Glacial hunter-gath- erers bearing microlithic technologies. The ages of the second set, all obtained from charred residues present on pottery sherds, range from ~12.1 to ~8.6 ka cal BP. This last set probably reflects two successive occupations by ceramics users, centered around 10.3 ka cal BP and 9.2 ka cal BP [40]. In order to estimate the age of the bird figurine, three new burnt bone samples, one of which bearing evidence of gouging, an anthropogenic modification observed on the bird carv- ing (see below), were sent to the Beta Analytic Testing Laboratory, Miami (FL, USA). All of them were analyzed following the ISO/IEC 17025:2005 Testing Accreditation PJLA #59423 protocols. Charred bone pretreatment with alkali allowed for the extraction of collagen. All work was done at Beta facilities in four in-house NEC accelerator mass spectrometers and four Termo IRMSs. The “Conventional Radiocarbon Age”, calculated using the Libby half-life i.e., 5,568 years, was corrected for total isotopic fraction, and was used for calendar calibration. Conventional Radiocarbon Ages and sigma were rounded to the nearest 10 years and are reported as radiocarbon years before present, i.e., before 1950. When counting statistics pro- duced sigma lower than 30 years, a conservative ± 30 BP was cited for the results. δ13C values were obtained on the material itself, not on the AMS δ13C. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Archeological context and dating The three burnt bone samples selected for radiocarbon dating yielded sufficient amount of collagen within the expected value required for passing the quality assurance tests. All three ages fall within the first cluster previously identified and range between 13,4 and 13,1 ka cal BP (Fig 1c, Table 1). The age of the bone bearing a deep notch produced by gouging is 11,520 ± 40 (Beta-515953), which corresponds to an age of 13,448–13,279 cal BP (95.4%). It is noteworthy that twenty-one of the twenty-eight (75%) 14C ages from the first set statistically overlap (95%) the age obtained on the modified bone (Fig 1c). The coherent results obtained for the ages of bone and charcoal samples indicate that both the faunal remains and the fire fuel, i.e., the char- coals, are of the same age, which rules out the possibility that the craftsman would have selected a sub-fossil fragment to burn and to carve the figurine [46]. 6 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China Table 1. Calibrated 14C ages obtained on burnt bone and charcoal from the sediment in which the bird carving was found. PLOS ONE PLOS ONE A Paleolithic bird figurine from China Table 1. (Continued) Lab ID Material 14C age Error Calibrated date (1σ) Calibrated date (2σ) References From To Probability From To Probability 11071 10950 30.7% 10623 10601 1.6% MTC-16892 CM 8630 50 9630 9536 66.5% 9703 9523 95.4% 40 9652 9549 1.7% MTC-16893 CM 9250 210 10746 10196 68.2% 11148 9911 95.4% 40 MTC-16894 CM 7880 50 8764 8595 68.2% 8797 8559 78.1% 40 8976 8881 11.0% 8869 8827 6.3% MTC-16896 CM 8310 90 9445 9240 57.7% 9489 9077 93.5% 40 9175 9140 7.7% 9055 9033 1.9% 9220 9206 2.8% BB: Burnt bone; CH: Charcoal; CM: Carbonized material from pottery sherds  Object bearing traces of manufacture similar to those on the bird carving BB: Burnt bone; CH: Charcoal; CM: Carbonized material from pottery sherds  Object bearing traces of manufacture similar to those on the bird carving Although the bird figurine was found in a spoil heap, a number of contextual observations argue in favor of its Paleolithic origin. First, the spoil heap did not contain Neolithic or Bronze Age cultural remains. Likewise, no archeological remains attributed to Paleolithic ages were found in layers 1–4; these latter layers were archeologically sterile in the area where remnants of layer 5 and the associated spoil heap were identified. The distance separating the areas con- taining Neolithic/Bronze Age and LGM/Younger Dryas archeological material is too impor- tant, i.e., >50 m, to suggest the carving percolated into layer 5 from overlying more recent layers. The absence of evidence suggesting the mixing of both assemblages, i.e., from layer 5, on the one hand, and from layers 1–4 on the other, is further supported by the differences in the manufacture, decoration, and ages of the ceramic remains. While the sherds from layer 5 are crude, without decoration, and fired at low temperature, those from layers 1–4 are fine and bear stylistic features that allow their cultural attribution to the Yangshao Neolithic and Shang-Zhou Bonze Age. It is well-known that crude pottery appears in the Chinese archeologi- cal record at the end of the Upper Paleolithic and in Epipaleolithic contexts. The oldest known occurrences appear circa 20 ka BP in South China [47–49] and circa 12 ka BP in North China [40,50]. PLOS ONE Should the pottery sherds from layer 5 have had a Neolithic/Bronze Age origin, their expected ages would have ranged between 7 and 5 ka BP, not 12 to 9 ka BP. Finally, although the figurine could date to any of the three episodes of the human occupation identified by the 14C ages, none of the numerous dated faunal remains falls into the timeframe of the two most recent proposed human occupations, i.e., between 11 and 10 ka BP or between 9.6 and 8.7 ka BP; their radiocarbon ages are exclusively comprised between 13.8 and 13 ka BP regardless of the testing laboratory to which they were submitted for dating. Moreover, the figurine and associated faunal remains feature a similar color range and patina, which argues in favor of their syndeposition. These results and contextual information indicate that the most probable age of the figurine corresponds to that of the directly dated faunal remain bearing evidence of gouging, i.e., within the ~13.4–13.2 ka cal BP time interval. Archeological context and dating Lab ID Material 14C age Error Calibrated date (1σ) Calibrated date (2σ) References From To Probability From To Probability Beta-515951 BB 11290 30 13165 13090 68.2% 13211 13070 95.4% This study Beta-515952 BB 11470 30 13353 13275 68.2% 13412 13251 95.4% This study Beta-515953 BB 11520 40 13410 13315 68.2% 13448 13279 95.4% This study IAAA-92123 BB 11530 50 13424 13317 68.2% 13463 13276 95.4% 40–43 IAAA-92124 BB 11590 50 13473 13364 68.2% 13500 13300 90.7% 40–43 13548 13510 4.7% 40–43 IAAA-92125 BB 11940 40 13820 13719 68.2% 13967 13701 87.9% 40–43 13678 13585 7.5% 40–43 IAAA-100080 CH 28690 120 33050 32593 68.2% 33310 32313 95.4% 40–43 IAAA-100082 BB 11760 40 13611 13480 64.7% 13719 13470 95.4% 40–43 13696 13685 3.5% 40–43 IAAA-102634 BB 11480 50 13390 13276 68.2% 13443 13214 95.4% 40–43 IAAA-102635 BB 11720 50 13571 13472 68.2% 13641 13440 89.9% 40–43 13715 13664 5.5% 40–43 IAAA-102636 CM 8570 40 9553 9517 68.2% 9600 9483 95.4% 40–43 IAAA-102538 CH 10270 40 12123 11959 68.2% 12169 11821 93.8% 40–43 12232 12211 1.1% 40–43 12367 12357 0.5% 40–43 IAAA-102639 CH 11760 50 13618 13481 59.6% 13726 13466 95.4% 40–43 13703 13678 8.6% 40–43 IAAA-102640 CH 11930 50 13825 13705 59.1% 13864 13574 86.8% 40–43 13672 13629 9.1% 13958 13879 8.6% 40–43 IAAA-102641 BB 11520 50 13416 13310 68.2% 13460 13270 95.4% 40–43 IAAA-102642 CH 11120 50 13071 12935 68.2% 13089 12830 95.4% 40–43 IAAA-102643 BB 11370 50 13268 13155 68.2% 13307 13099 95.4% 40–43 IAAA-102644 CH 11610 50 13490 13374 68.2% 13559 13322 95.4% 40–43 IAAA-102645 BB 11610 50 13490 13374 68.2% 13559 13322 95.4% 40–43 IAAA-102647 CH 11300 50 13198 13094 68.2% 13263 13070 95.4% 40–43 IAAA-102648 CH 11600 50 13480 13370 68.2% 13550 13314 95.4% 40–43 IAAA-102649 BB 11930 50 13825 13705 59.1% 13864 13574 86.8% 40–43 13672 13629 9.1% 13958 13879 8.6% 40–43 IAAA-102650 CH 11920 50 13800 13702 51.8% 13856 13566 90.7% 40–43 13678 13617 16.4% 13957 13898 4.7% IAAA-150572 BB 11710 40 13558 13479 68.2% 13599 13441 95.4% 40 IAAA-150573 BB 11660 40 13498 13449 39.1% 13574 13421 95.4% 40 13545 13507 29.1% IAAA-150574 BB 11430 40 13316 13216 68.2% 13377 13155 95.4% 40 IAAA-150575 BB 11639 40 13495 13427 53.3% 13570 13395 95.4% 40 13543 13515 14.9% IAAA-150576 BB 11430 40 13316 13216 68.2% 13377 13155 95.4% 40 IAAA-150577 BB 11520 40 13410 13315 68.2% 13448 13280 95.4% 40 IAAA-150578 BB 11490 40 13386 13291 68.2% 13437 13259 95.4% 40 TKa-15555 CM 9330 60 10603 10486 50.1% 10703 10372 92.4% 40 10653 10621 9.6% 10326 10299 2.2% 10461 10433 8.4% 10354 10341 0.8% TKa-15556 CM 9530 70 10869 10711 37.5% 11133 10652 93.8% 40 (Continued) PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 7 / 24 https://doi.org/10.1371/journal.pone.0233370.t001 Materials and methods The Lingjing bird carving and the associated archeological material analyzed in this study are curated at the Henan Provincial Institute of Cultural Relics and Archaeology, Zhengzhou, Henan Province, China (Repository ID: 09L5鸟-01). No permits were required for the 8 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China described research, which complied with all relevant regulations. The Lingjing bird carving was 3D scanned using a General Electrics (GE) Vtome x|s microtomography housed at the PLACAMAT facilities, Bordeaux University. The figurine was scanned with the beak facing downward to minimize artifacts produced when X-rays are tangential to the surface. The acquisition was done with a 180 kV X-ray nanofocus tube and the following beam parameters: 100 kV and 200 μA. The scan was performed at a cubic voxel size of 11.5 μm. A copper filter of 0.1 mm was used and 1875 projections were taken at 360˚. The reconstruction of the volume was performed with the Datos | Rec software of GE. 3D visualizations were then performed within the Avizo 9.1 (FEI) workspace. A simplified 3D surface was also generated (S1 Data) with the MiKTeX Console 2.9.7076. The object was photographed with a Sony E 30 mm with a macro-lens, and examined and photographed with a motorized Leica Z6 APOA equipped with a DFC420 digital camera linked to LAS Montage and Leica Map DCM 3D computer software. Images showing various aspects of the figurine were imported into Adobe Illustrator and used to make a tracing of the areas bearing traces of manufacture identified under the microscope. This tracing was com- pared to the original specimen under a microscope and corrected as required. High-resolution surface topography of selected areas was obtained with a Sensofar S neox confocal microscope driven by SensoScan 6 software (Sensofar) in order to better visualize and characterize traces of manufacture and use wear. Data acquired from the various imaging methods were com- pared to establish the technique of manufacture used and the orientation of the motion on each identified area. Distinction between traces of manufacture and alterations resulting from use was based on diagnostic features identified on ethnographic, experimental and archeologi- cal bone items [51–62]. When two or more techniques were identified on the same area, partial obliteration of the original anthropogenic modifications by traces generated through the sub- sequent application of another technique clarified their chronological ordering. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Results Scales = 2 mm ournal.pone.0233370.g004 https://doi.org/10.1371/journal.pone.0233370.g004 the infilling of mineral deposits following physicochemical and biochemical alterations [65,68,69] after the action of microorganisms. Reticular fibrolamellar bone tissue is found in birds, large dinosaurs and in long bones of fast-growing juvenile mammals such as herbivores [70–75]. Considering the orientation of the bone structure and the figurine dimensions, the bird was likely carved from a diaphyseal fragment of a medium-size mammal limb bone. The bone figurine as well as other worked and unworked osseous fragments found in the spoil heap are completely blackened (Fig 3, Table 2). Recent and heat fractures present on some specimens illustrate the extent of the process in cross-section. The color gradient ranges from a dark brown, almost black, outer layer measuring between 0.2 mm to 0.8 mm in thick- ness to dark black at the center. None of the original bone coloration remains visible on the specimens. Although the staining of bones could result from a variety of processes [76], the the infilling of mineral deposits following physicochemical and biochemical alterations [65,68,69] after the action of microorganisms. Reticular fibrolamellar bone tissue is found in birds, large dinosaurs and in long bones of fast-growing juvenile mammals such as herbivores [70–75]. Considering the orientation of the bone structure and the figurine dimensions, the bird was likely carved from a diaphyseal fragment of a medium-size mammal limb bone. The bone figurine as well as other worked and unworked osseous fragments found in the spoil heap are completely blackened (Fig 3, Table 2). Recent and heat fractures present on some specimens illustrate the extent of the process in cross-section. The color gradient ranges from a dark brown, almost black, outer layer measuring between 0.2 mm to 0.8 mm in thick- ness to dark black at the center. None of the original bone coloration remains visible on the specimens. Although the staining of bones could result from a variety of processes [76], the The bone figurine as well as other worked and unworked osseous fragments found in the spoil heap are completely blackened (Fig 3, Table 2). Recent and heat fractures present on some specimens illustrate the extent of the process in cross-section. The color gradient ranges from a dark brown, almost black, outer layer measuring between 0.2 mm to 0.8 mm in thick- ness to dark black at the center. None of the original bone coloration remains visible on the specimens. Results The figurine depicts a small standing bird (length 19.2 mm, width 5.1 mm, height 12.5 mm). The subject’s morphology and proportions, i.e., short head and neck, robust, rounded bill and long tail, are reminiscent of Passeriformes. Passeriformes is an order that encompasses more than half of all known extant bird species. Unfortunately, the lack of minute details on the figu- rine prevents a more precise identification. Aside from the ostrich eggshell fragments found in the spoil heap, no avian remains were identified in the faunal assemblages of any archeological layers at Lingjing, i.e., layers 1–4, 5, 10, and 11. Nonetheless, our identification of the carving as representing a passerine is fully compatible with the paleoecology of the species comprised in this order as they were, and are, found on virtually all continents and climatic settings, including present-day and Late Pleistocene China [63,64]. In lieu of the passerine short legs, a large, rectangular pedestal allows the figurine to stand in the upright position (Fig 3, S1 Data). The oversized tail prevents the object from tilting forward. The lateral aspects of the body are flat and the wings are not represented. Openings left on the surface by capillaries and vascular canals demonstrate the use of corti- cal bone for the manufacture of the figurine (Fig 4a–4d, S1 Video). Microtomography reveals a densely vascularized fibrolamellar complex without evidence of remodeling. The orientation of the bone structure corresponds to the bill-tail axis of the carving (S1 Video). The lack of lamellated layers and the presence of the same pattern of vascularization throughout the figu- rine suggest that the bone outermost and innermost cortex were removed when carving the object. The presence of transverse layers of demineralized bone is consistent with the action of osteolytic bacteria [65–67] (Fig 4e). Increased bone density at the surface of the object indicates PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 9 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 4. Lingjing bird carving. (A) Photographs of the six aspects of the carvings. (B) 3D renderings of the carving obtained by CTscan. Scales = 2 mm. https://doi.org/10.1371/journal.pone.0233370.g004 Fig 4. Lingjing bird carving. (A) Photographs of the six aspects of the carvings. (B) 3D renderings of the carving obtained by CTscan. Scales = 2 mm. https://doi.org/10.1371/journal.pone.0233370.g004 arving. (A) Photographs of the six aspects of the carvings. (B) 3D renderings of the carving obtained by CTscan. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Results Based on the color gradient visible in cross-sec- tion, and considering that microtomography reveals an intact histological structure, the faunal fragments were most likely heated from 1 to 3 hours in an anaerobic environment at a temper- ature ranging between 300 and 500˚C. Controlled experiments are required, however, to reproduce this treatment accurately and assess its effects on easing the carving of osseous raw material. To study the manufacturing process of this exceptional object, we applied a novel approach consisting in surveying diagnostic traces of manufacture and post-depositional modifications by multifocus and confocal microscopy, while using high-resolution microtomography to detect edges between worked areas. We recorded sixty-eight areas distributed over the entire surface of the object, each corresponding to distinct carving episodes leading to the shaping of the figurine (Fig 5, Table 3). For each area, we documented the manufacturing technique applied and, when possible, the orientation of the gesture. In thirteen cases, we were able to identify the superimposition of techniques. Gouging was applied with the robust edge of a stone tool, such as a burin spall, producing flat facets covered by chatter marks, i.e., undula- tions perpendicular to the direction of the tool motion generated by pressure changes (Fig 6a). Prolonged use of a burin in a gouging activity may result in the microchipping of its flat edge, Ab: Abrading; Cm: Cut mark; Fl: Flaking; Go: Gouging; No: Notching; Pf: Pressure flaking; Po: Polishing; Re: Retouching; Sa: Sawing; St: Striations https://doi.org/10.1371/journal.pone.0233370.t002 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 11 / 24 observed color gradient and the density of the bone fragments suggest they might have been subjected to a controlled heat treatment. Exposure to open flame, and contact with calcitic ash, may cause the bone to crack, shrink and deform. However, the recognition of histological structure remains achievable when bone is burnt at less than 600˚C [77–85]. Experimental data highlights that the temperature of cremation, the duration of heat exposure, and the avail- ability of oxygen and organic compounds in the environment play key roles in modifying the aspect and structure of osseous remains [86]. Based on the color gradient visible in cross-sec- tion, and considering that microtomography reveals an intact histological structure, the faunal fragments were most likely heated from 1 to 3 hours in an anaerobic environment at a temper- ature ranging between 300 and 500˚C. Results Although the staining of bones could result from a variety of processes [76], the PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 10 / 24 PLOS ONE A Paleolithic bird figurine from China Table 2. Technological and morphometric data on burnt bone fragments associated with the bird carving. Figure Maximum Length Maximum Width Maximum Thickness Cortical Thickness Human modification Modification type Burnt Fracture Type Fig 3a 43.39 14.65 9.67 6.04 Yes Fl, Go, Po y Fig 3b 29.01 7.82 5.28 Yes Po, St y Recent Fig 3c 21.21 7.56 6.67 No y Fig 3d 13.31 7.61 3.64 Yes Ab y Fig 3e 24.53 11.98 4.04 No y Fig 3f 15.3 12.61 5.52 5.17 Maybe Fl, Go(?), St y Fig 3g 17.92 10.45 6.81 No y Fig 3h 10.99 3.7 3.21 Yes Cm, No, St y Fig 3i 14.43 11.61 6.64 Yes No, St y Fig 3j 15.23 6.21 4.75 No y Fig 3k 24.1 9.46 3.17 Yes Ab, Po y Fig 3l 16.84 2.87 2.79 Yes Po, Sa(?) y Old Fig 3m 10.25 4.8 3.23 No y Fig 3n 10.46 8.77 3.07 2.13 No y Fig 3o 19.03 5.09 4.2 Yes Pf, Re y Fig 3p 14.81 7.24 4.78 No y Fig 3q 14.3 6.81 3.81 No y Fig 3r 21.81 11.79 6.56 6.3 Yes Fl, Po y Heat Fig 3s 12.04 6.91 3.38 Maybe Fl y Fig 3t 18.74 5.75 2.93 No y Recent Ab: Abrading; Cm: Cut mark; Fl: Flaking; Go: Gouging; No: Notching; Pf: Pressure flaking; Po: Polishing; Re: Retouching; Sa: Sawing; St: Striations https://doi org/10 1371/journal pone 0233370 t002 Table 2. Technological and morphometric data on burnt bone fragments associated with the bird carving. Ab: Abrading; Cm: Cut mark; Fl: Flaking; Go: Gouging; No: Notching; Pf: Pressure flaking; Po: Polishing; Re Retouching Sa Sawing St Striations observed color gradient and the density of the bone fragments suggest they might have been subjected to a controlled heat treatment. Exposure to open flame, and contact with calcitic ash, may cause the bone to crack, shrink and deform. However, the recognition of histological structure remains achievable when bone is burnt at less than 600˚C [77–85]. Experimental data highlights that the temperature of cremation, the duration of heat exposure, and the avail- ability of oxygen and organic compounds in the environment play key roles in modifying the aspect and structure of osseous remains [86]. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 https://doi.org/10.1371/journal.pone.0233370.t002 Ab: Abrading; Cm: Cut mark; Fl: Flaking; Go: Gouging; No: Notching; Pf: Pressure flaking; Po: Polishing; Re: Retouching; Sa: Sawing; St: Striations Results Controlled experiments are required, however, to reproduce this treatment accurately and assess its effects on easing the carving of osseous raw material. To study the manufacturing process of this exceptional object, we applied a novel approach consisting in surveying diagnostic traces of manufacture and post-depositional modifications by multifocus and confocal microscopy, while using high-resolution microtomography to detect edges between worked areas. We recorded sixty-eight areas distributed over the entire surface of the object, each corresponding to distinct carving episodes leading to the shaping of the figurine (Fig 5, Table 3). For each area, we documented the manufacturing technique applied and, when possible, the orientation of the gesture. In thirteen cases, we were able to identify the superimposition of techniques. Gouging was applied with the robust edge of a stone tool, such as a burin spall, producing flat facets covered by chatter marks, i.e., undula- tions perpendicular to the direction of the tool motion generated by pressure changes (Fig 6a). Prolonged use of a burin in a gouging activity may result in the microchipping of its flat edge, PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 11 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 5. Microtomography of the Lingjing bird carving. (A) Location of the three sections. (B-D) Transverse sections showing that inner cortical bone was used to carve the object and that the bill-tail axis was oriented along the bone structure. (E) Longitudinal section showing diagenetic alterations. Increase density near the surface (arrows) indicate probable infilling of mineral deposits following physicochemical and biochemical alteration. Transversal layers of demineralized bone (arrow heads) suggesting the action of osteolytic bacteria. Scales = 2 mm. https://doi.org/10.1371/journal.pone.0233370.g005 Fig 5. Microtomography of the Lingjing bird carving. (A) Location of the three sections. (B-D) Transverse sections showing that inner cortical bone was used to carve the object and that the bill-tail axis was oriented along the bone structure. (E) Longitudinal section showing diagenetic alterations. Increase density near the surface (arrows) indicate b bl f ll f l d f ll h h l d b h l l l l f Fig 5. Microtomography of the Lingjing bird carving. (A) Location of the three sections. (B-D) Transverse sections showing that inner cortical bone was used to carve the object and that the bill-tail axis was oriented along the bone structure. (E) Longitudinal section showing diagenetic alterations. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Results Increase density near the surface (arrows) indicate probable infilling of mineral deposits following physicochemical and biochemical alteration. Transversal layers of demineralized bone (arrow heads) suggesting the action of osteolytic bacteria. Scales = 2 mm. https://doi.org/10.1371/journal.pone.0233370.g005 12 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China Table 3. Location and manufacturing technique applied to each area of the bird figurine. Results Area N˚ Techniques Visible on aspect: Area N˚ Techniques Visible on aspect: 1st 2nd Right Left Top Bottom Front Rear 1st 2nd Right Left Top Bottom Front Rear 1 Go • • • 35 Go Sc • • • 2 Go • • • 36 Sc • • • 3 Go • • • 37 Ab • • • • • 4 Go • • 38 Ab • • • 5 Go • • • 39 Go • • • 6 Go • • • • 40 Go • • • 7 Go • • • • 41 Go Sc, Po • • • 8 Go Sc • • • 42 Sc • • • 9 Go • • • 43 Ab • • • 10 Go Sc • • • 44 Ab Po • • • 11 Go • • • 45 Go • • • 12 Go • • • • 46 Go • • 13 Go Sc • • 47 Go • • 14 Go Sc • • • 48 Ab • • • 15 Go • • • 49 Go • • • 16 Go Sc • • • • 50 Go • • 17 Go Sc • • • 51 Ab • • • 18 Go • • • • 52 Go • • • 19 Ib • • • 53 Go • • • 20 Ib • • • 54 Go • • • 21 In • • • 55 Go • • • 22 In • • • 56 Go Sc • • • 23 In • • • 57 Go • • 24 In • • • 58 Go • • • 25 In • • 59 Go • • • 26 In • • 60 Go Sc • • • 27 Go • • • 61 Go Sc • • • 28 Go • • • 62 Ab • • • 29 Go • • • • 63 Go • • • 30 Ab • • • 64 Go Sc • • • 31 Go • • • 65 Ab • • • • 32 Sc • • • 66 Go • • 33 Sc • • 67 Go • • 34 Sc • 68 Go • • Ab: Abrading; Go: Gouging; Ib: Incising with burin; In: Incising; Po: Polishing; Sc: Scraping Table 3. Results Location and manufacturing technique applied to each area of the bird figurine. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Results Area N˚ Techniques Visible on aspect: Area N˚ Techniques Visible on aspect: 1st 2nd Right Left Top Bottom Front Rear 1st 2nd Right Left Top Bottom Front Rear 1 Go • • • 35 Go Sc • • • 2 Go • • • 36 Sc • • • 3 Go • • • 37 Ab • • • • • 4 Go • • 38 Ab • • • 5 Go • • • 39 Go • • • 6 Go • • • • 40 Go • • • 7 Go • • • • 41 Go Sc, Po • • • 8 Go Sc • • • 42 Sc • • • 9 Go • • • 43 Ab • • • 10 Go Sc • • • 44 Ab Po • • • 11 Go • • • 45 Go • • • 12 Go • • • • 46 Go • • 13 Go Sc • • 47 Go • • 14 Go Sc • • • 48 Ab • • • 15 Go • • • 49 Go • • • 16 Go Sc • • • • 50 Go • • 17 Go Sc • • • 51 Ab • • • 18 Go • • • • 52 Go • • • 19 Ib • • • 53 Go • • • 20 Ib • • • 54 Go • • • 21 In • • • 55 Go • • • 22 In • • • 56 Go Sc • • • 23 In • • • 57 Go • • 24 In • • • 58 Go • • • 25 In • • 59 Go • • • 26 In • • 60 Go Sc • • • 27 Go • • • 61 Go Sc • • • 28 Go • • • 62 Ab • • • 29 Go • • • • 63 Go • • • 30 Ab • • • 64 Go Sc • • • 31 Go • • • 65 Ab • • • • 32 Sc • • • 66 Go • • 33 Sc • • 67 Go • • 34 Sc • 68 Go • • Ab: Abrading; Go: Gouging; Ib: Incising with burin; In: Incising; Po: Polishing; Sc: Scraping https://doi.org/10.1371/journal.pone.0233370.t003 Table 3. Results Location and manufacturing technique applied to each area of the bird figurine. Ab: Abrading; Go: Gouging; Ib: Incising with burin; In: Incising; Po: Polishing; Sc: Scraping https://doi.org/10.1371/journal.pone.0233370.t003 https://doi.org/10.1371/journal.pone.0233370.t003 which can then produce uninterrupted superficial striations running parallel to the gouging motion and keeping a constant depth on the undulations of the chatter marks. Abrading was performed by displacing the object on an abrasive grindstone which resulted in broad, spin- dle-like grooves of variable depths (Fig 6b). Scraping was achieved with retouched and unre- touched lithic cutting edges resulting in groups of shallow grooves featuring internal parallel striations (Fig 6c). Incisions were done either by deeply engraving the surface with the dihe- dral-shaped tip of a stone tool, or by superficially marking the bone with a sharp point (Fig 6d). which can then produce uninterrupted superficial striations running parallel to the gouging motion and keeping a constant depth on the undulations of the chatter marks. Abrading was performed by displacing the object on an abrasive grindstone which resulted in broad, spin- dle-like grooves of variable depths (Fig 6b). Scraping was achieved with retouched and unre- touched lithic cutting edges resulting in groups of shallow grooves featuring internal parallel striations (Fig 6c). Incisions were done either by deeply engraving the surface with the dihe- dral-shaped tip of a stone tool, or by superficially marking the bone with a sharp point (Fig 6d). PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 13 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 6. Tracings of the six aspects of the Lingjing bird carving with the technique used to manufacture each area. Ab: Abrading; Go: Gouging; Ib: Incising with a burin; In: Incising; Po: Polishing; Sc: Scraping. Numbers refer to Table 3. Scale = 1 mm. https://doi.org/10.1371/journal.pone.0233370.g006 Fig 6. Tracings of the six aspects of the Lingjing bird carving with the technique used to manufacture each area. Ab: Abrading; Go: Gouging; Ib: Incising with a burin; In: Incising; Po: Polishing; Sc: Scraping. Numbers refer to Table 3. Scale = 1 mm. Fig 6. Tracings of the six aspects of the Lingjing bird carving with the technique used to manufacture each area. Ab: Abrading; Go: Gouging; Ib: Incising with a burin; In: Incising; Po: Polishing; Sc: Scraping. Numbers refer to Table 3. Scale = 1 mm. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Results https://doi.org/10.1371/journal.pone.0233370.g006 Although the initial stage of manufacture cannot be identified, it is possible that the first step entailed abrading the bone fragment. A large area on the right side of the figurine, covered by traces of abrasion on a coarse grindstone may represent the remnant of this first shaping event (Fig 6b). Gouging was used to rough out the figurine. This technique was vigorously applied to shape concave surfaces such as the throat and breast (Figs 6a and 8a, S1 Data), the back (Fig 7b, S1 Data), the undertail coverts and, in particular, the pedestal (Fig 7c, S1 Data). It was used more gently to carve the head (Fig 7d, S1 Data) and flat sides of the figurine (Fig 7e, S1 Data). Scraping was used on both sides of the throat (Fig 8a, S1 Data), the back (Figs 7b and 9a, S1 Data) and the vent (Fig 8b, S1 Data), to smooth out the chatter marks, and refine the final shape of the carving. The edges of the pedestal were carefully shaped by juxtaposing tiny facets of abrasion (S1 Data). The base of the pedestal was first carved by gouging and subse- quently even out by two episodes of scraping (Figs 7c and 8c, S1 Data). The purpose of the sec- ond episode, carried out with a retouched cutting edge, may have been to slightly change the orientation of the base to ensure the bird could stand upright. On the right aspect of the head, two groups of incisions may have served to identify the bird’s eye and bill (Fig 7d). The first group is composed of two deep incisions made with the same burin. The second includes six subparallel, superficial incisions made with the same sharp point. At the microscopic scale, prominent features left by the manufacture are smoothed at vary- ing degrees by an abrasion that has produced randomly oriented striations (Fig 9). The PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 14 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 7. Techniques used to carve the Lingjing bird figurine. (A) Gouging used to shape the throat and breast of the bird. Scale = 1 mm. (B) Abrading applied on the right side of the figurine. Scale = 1 mm. (C) Scraping used to even out the base of the pedestal. Results (D) Incising on the right aspect of the head, probably to depict the eye and the bill edge, producing two converging groups of incisions, one on the left composed of two deep incisions with v- shaped sections (see Fig 6 n.19-20), the other consisting of six superficial incisions with similar internal striations demonstrating the use of the same point (see Fig 6 n. 21–26). (A-B) black and white micrographs. (C-D) 3D renderings obtained with a confocal microscope. https://doi org/10 1371/journal pone 0233370 g007 Fig 7. Techniques used to carve the Lingjing bird figurine. (A) Gouging used to shape the throat and breast of the bird. Scale = 1 mm. (B) Abrading applied on the right side of the figurine. Scale = 1 mm. (C) Scraping used to even out the base of the pedestal. (D) Incising on the right aspect of the head, probably to depict the eye and the bill edge, producing two converging groups of incisions, one on the left composed of two deep incisions with v- shaped sections (see Fig 6 n.19-20), the other consisting of six superficial incisions with similar internal striations demonstrating the use of the same point (see Fig 6 n. 21–26). (A-B) black and white micrographs. (C-D) 3D renderings obtained with a confocal microscope. https://doi.org/10.1371/journal.pone.0233370.g007 absence of such smoothing associated with randomly oriented striations of different widths on the other small blackened bone fragments from the same assemblage rules out the possibility that this wear may result from natural mechanical abrasion. The presence of sediment in some striations indicates they are ancient in origin. Based on experimental criteria, these traces differ markedly from those produced by manipulation or intentional polishing with skins or furs. They are, however, entirely consistent with use wear pattern resulting from the experimental transportation of an osseous object in a leather bag [60]. Other striations cutting into sediment deposits developed following the deposition of the object. Finally, we also recorded traces of gouging, abrading, scraping and incising on small fragments of burnt bone recovered from the same context (Fig 10, Table 2). Aside from the specimen, bearing a deep notch produced by gouging (Fig 10a, Table 2) and directly dated to 13,448–13,279 cal BP (Beta-515953: 11,520±40 BP), the assemblage also contains fragments of bone rods shaped by scraping displaying inci- sions perpendicular and oblique to the main axis (Fig 10, Table 2). PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Results absence of such smoothing associated with randomly oriented striations of different widths on the other small blackened bone fragments from the same assemblage rules out the possibility that this wear may result from natural mechanical abrasion. The presence of sediment in some striations indicates they are ancient in origin. Based on experimental criteria, these traces differ markedly from those produced by manipulation or intentional polishing with skins or furs. They are, however, entirely consistent with use wear pattern resulting from the experimental transportation of an osseous object in a leather bag [60]. Other striations cutting into sediment deposits developed following the deposition of the object. Finally, we also recorded traces of gouging, abrading, scraping and incising on small fragments of burnt bone recovered from the same context (Fig 10, Table 2). Aside from the specimen, bearing a deep notch produced by gouging (Fig 10a, Table 2) and directly dated to 13,448–13,279 cal BP (Beta-515953: 11,520±40 BP), the assemblage also contains fragments of bone rods shaped by scraping displaying inci- sions perpendicular and oblique to the main axis (Fig 10, Table 2). PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 15 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 8. Manufacturing techniques applied to the Lingjing bird. (A) Traces of gouging followed by scraping on the bird back. (B) Traces of gouging on the bird throat leaving diagnostic chatter marks on juxtaposed facets. (C) Large notch produced by multiple vigorous gouging motions to shape the pedestal. Scale = 1 mm. (D) Gentle gouging applied to carve the head of the birds. Scale = 2 mm. (E) Superficial gouging applied to the left side of the figurine. Scale = 1 mm. (A-B) 3D renderings obtained with a confocal microscope. (C-E) Black and white micrographs. https://doi org/10 1371/journal pone 0233370 g008 Fig 8. Manufacturing techniques applied to the Lingjing bird. (A) Traces of gouging followed by scraping on the bird back. (B) Traces of gouging on the bird throat leaving diagnostic chatter marks on juxtaposed facets. (C) Large notch produced by multiple vigorous gouging motions to shape the pedestal. Scale = 1 mm. (D) Gentle gouging applied to carve the head of the birds. Scale = 2 mm. (E) Superficial gouging applied to the left side of the figurine. Scale = 1 mm. (A-B) 3D renderings obtained with a confocal microscope. (C-E) Black and white micrographs. Results https://doi org/10 1371/journal pone 0233370 g008 https://doi.org/10.1371/journal.pone.0233370.g008 https://doi.org/10.1371/journal.pone.0233370.g008 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Discussion A representation is generally defined as the use of signs that stand in for, and take the place of, something else [87,88]. In the domain of artistic expressions, a representation is a type of recording in which the sensory information about a physical object, or being, is recorded in a medium. The degree to which an artistic representation resembles the object, or the being, it represents is a function of resolution. Our contention for the Lingjing figurine representing a bird is based on four lines of evidence. First, its outline, with the exception of the pedestal, almost perfectly matches that of a bird and identifies several avian anatomical features, e.g., the tail, head, bill, throat, breast, and belly. Second, the edges of the outline are modified on both aspects to enhance the anatomical features of most birds, i.e., rounded volume of the body, conical morphology of the bill, etc. Third, marks were added on the head at the location of the eye and bill. Finally, the technological analysis of the modifications present on the carving demonstrates they were deliberately produced, and the carving techniques were coherently chosen in order to highlight the anatomical features of a bird. The fact that the wings are not carved does not represent an obstacle to identifying the carving as a representation of a bird since an artistic representation is by definition an operation of subtraction, addition, and/or modification of the real world, which depends on the chosen medium, the artist’s know-how PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 16 / 24 PLOS ONE A Paleolithic bird figurine from China Fig 9. Manufacturing techniques applied to Lingjing bird. (A) Traced of scraping on (B) Traces of scraping on the left side of the bird tail. (C) Base of the pedestal with trace those of gouging. (A-C) Black and white micrographs. Scales = 1 mm. https://doi.org/10.1371/journal.pone.0233370.g009 Fig 9. Manufacturing techniques applied to Lingjing bird. (A) Traced of scraping on the left side of the bird throat. (B) Traces of scraping on the left side of the bird tail. (C) Base of the pedestal with traces of scraping superimposed to those of gouging. (A-C) Black and white micrographs. Scales = 1 mm. https://doi.org/10.1371/journal.pone.0233370.g009 Fig 9. Manufacturing techniques applied to Lingjing bird. (A) Traced of scraping on the left side of the bird throat. (B) Traces of scraping on the left side of the bird tail. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Discussion The style of this diminutive representation is original and remark- ably different from all other known Paleolithic avian figurines. Avian representations, and passerine in particular, constitute a recurring theme in Chinese Neolithic art, the oldest exam- ple being a passerine made of jade dating back to circa 5 ka BP [89,90]. The Lingjing bird carv- ing predates previously known instances from this region by almost 8,500 years. The sophistication reflected by the object manufacturing process suggests this three-dimensional representation is several conceptual stages removed from the origin of a long-standing artistic tradition, extending well into the Paleolithic, that may be better characterized by future discoveries. Even though carving and painting are generally seen as activities demonstrating the acquisi- tion of symbolic thought, the ways in which they reify meaning in matter differ markedly. Each activity involves different spatial conceptualizations, sensorimotor experiences, analogi- cal reasonings, and skill-learning processes [91–93]. Pigment preparation and application obviously play a key role in painting. At times, particular morphological features of the canvas, e.g., natural protuberances or concavities of a cave wall, might have been exploited to enhance the perspective of a representation. Carving a figurine, on the other hand, requires the combi- nation of different techniques, e.g., scraping, grinding, polishing, gouging, incising, and notch- ing, their adaptation to the selected raw material, and the alternating application of different tools and motions. It also requires the ability to mentally visualize a volume in matter and cre- ate symmetries in a three-dimensional space. Unlike paintings, anchored to sites charged with symbolic meaning, Paleolithic carvings are representations made to be transported, curated, manipulated, and often hung on clothing [30]. In view of the above, the cognitive require- ments and technical skills needed to produce and perpetuate painted, engraved and drawn representations, on the one hand, and sculptures, on the other hand, may vary considerably and justifies approaching the emergence of these practices independently. The bird figurine from Lingjing constitutes the first carving found at an East Asian Paleolithic site and it differs technologically and stylistically from previous and contemporaneous representations of avi- fauna found in Europe and Siberia. The earliest known statuettes, made of mammoth ivory and including a flying waterfowl, are found in the Aurignacian of the Swabian Jura [29,30]. They are dated to c. 40–38 ka BP. Discussion (C) Base of the pedestal with traces of scraping superimposed to those of gouging. (A-C) Black and white micrographs. Scales = 1 mm. Fig 9. Manufacturing techniques applied to Lingjing bird. (A) Traced of scraping on the left side of the bird throat. (B) Traces of scraping on the left side of the bird tail. (C) Base of the pedestal with traces of scraping superimposed to those of gouging. (A-C) Black and white micrographs. Scales = 1 mm. https://doi.org/10.1371/journal.pone.0233370.g009 https://doi.org/10.1371/journal.pone.0233370.g009 https://doi.org/10.1371/journal.pone.0233370.g009 17 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China Fig 10. Microscopic wear. (A) Area on the back of the figurine showing traces of scraping smoothed by wear associated with randomly oriented striations. (B-C) Areas on the left side of the head (B) and the body (C) on which traces of manufacture are removed by wear associated with randomly oriented striations. (A-C) 3D renderings obtained with a confocal microscope. https://doi.org/10.1371/journal.pone.0233370.g010 Fig 10. Microscopic wear. (A) Area on the back of the figurine showing traces of scraping smoothed by wear associated with randomly oriented striations. (B-C) Areas on the left side of the head (B) and the body (C) on which traces of manufacture are removed by wear associated with randomly oriented striations. (A-C) 3D renderings obtained with a confocal microscope. https://doi.org/10.1371/journal.pone.0233370.g010 18 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China and skills, and cultural rules he/she wishes to comply to, or to transgress. Thus, the absence of wings may be explained by limitations inherent to the thickness of the bone fragment chosen to produce the carving, the diminutive size of the figurine itself, the difficulties to carve these features with the techniques and/or tools available to the artist, the stylistic canons the crafts- man was conforming to, or a combination of these reasons. Although the definition of an artistic tradition would ideally require the identification of shared technical, thematic, and stylistic traits on a number of asynchronous artistic produc- tions, we argue that, in the case of single items, striking concomitant differences in different characterizing domains, i.e., technological, thematic, and stylistic, may reasonably be used to infer original artistic traditions. Our analysis reveals that the Lingjing artist has chosen the appropriate techniques and applied them skillfully to faithfully reproduce the distinct anatomi- cal features of a passerine. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 Acknowledgments We thank Gauthier Devilder for his assistance in improving the graphic rendering of the carv- ing’s tracing. We are also grateful to Anna Rufa´ for her assistance in identifying the figurine as a passerine. Discussion Few other three-dimensional carvings representing birds, made of teeth and antler, come from West European late Upper Paleolithic sites [94–97]. The only Paleolithic bird carvings from Asia are those found at Mal’ta and Buret’, two neighboring Siberian sites located west of Lake Baikal [98]. They mainly consist of pendants made of ivory and antler representing flying waterfowls. The Lingjing figurine is the only Paleolithic three- dimensional object carved in burnt bone and representing a bird standing on a pedestal. It is also the only Paleolithic carving for which, thanks to its exceptional state of preservation, the final stages of manufacture could be documented in detail. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 19 / 24 PLOS ONE A Paleolithic bird figurine from China S1 Video. 3D model of the Lingjing bird with longitudinal and transverse sections showing the reticular fibrolamellar bone structure and vascularization pattern. (MP4) S1 Video. 3D model of the Lingjing bird with longitudinal and transverse sections showing the reticular fibrolamellar bone structure and vascularization pattern. (MP4) Funding acquisition: Zhanyang Li, Luc Doyon, Hui Fang, Francesco d’Errico. Investigation: Zhanyang Li, Luc Doyon, Alain Queffelec, Emeline Raguin, Francesco d’Errico. Methodology: Luc Doyon, Ronan Ledevin, Emeline Raguin, Francesco d’Errico. Project administration: Zhanyang Li, Hui Fang. Resources: Hui Fang, Emeline Raguin. Software: Ronan Ledevin, Alain Queffelec, Emeline Raguin. Software: Ronan Ledevin, Alain Queffelec, Emeline Raguin. Supervision: Zhanyang Li. Validation: Luc Doyon. Visualization: Luc Doyon, Francesco d’Errico. Writing – original draft: Luc Doyon, Ronan Ledevin, Emeline Raguin, Francesco d’Errico. Writing – review & editing: Luc Doyon, Alain Queffelec, Emeline Raguin, Francesco d’Errico. Supporting information S1 Data. Interactive 3D.pdf model of the Lingjing bird carving obtained by microtomogra- phy. (PDF) Author Contributions Conceptualization: Luc Doyon, Francesco d’Errico. Data curation: Zhanyang Li, Ronan Ledevin. Data curation: Zhanyang Li, Ronan Ledevin. Formal analysis: Luc Doyon, Ronan Ledevin, Alain Queffelec, Emeline Raguin, Francesco d’Errico. Formal analysis: Luc Doyon, Ronan Ledevin, Alain Queffelec, Emeline Raguin, Francesco d’Errico. Funding acquisition: Zhanyang Li, Luc Doyon, Hui Fang, Francesco d’Errico. References 1. Mellars P. Major issues in the emergence of modern humans. Curr Anthropol. 1989; 30: 349–385. https://doi.org/10.1086/203755 2. Gamble C. The Palaeolithic societies of Europe. Cambridge: Cambridge University Press; 1999. 3. Mithen S. The prehistory of mind: The cognitive origins of art, religion and science. New York: Thames and Hudson; 1999. 4. McBrearty S, Brooks AS. The revolution that wasn’t: A new interpretation of the origin of modern human behavior. J Hum Evol. 2000; 39: 453–563. https://doi.org/10.1006/jhev.2000.0435 PMID: 11102266 5. d’Errico F, Stringer CB. Evolution, revolution or saltation scenario for the emergence of modern cul- tures? Philos Trans R Soc B Biol Sci. 2011; 366: 1060–1069. https://doi.org/10.1098/rstb.2010.0340 PMID: 21357228 1. Mellars P. Major issues in the emergence of modern humans. Curr Anthropol. 1989; 30: 349–385. https://doi.org/10.1086/203755 2. Gamble C. The Palaeolithic societies of Europe. Cambridge: Cambridge University Press; 1999. 3. Mithen S. The prehistory of mind: The cognitive origins of art, religion and science. New York: Thames and Hudson; 1999. 4. McBrearty S, Brooks AS. The revolution that wasn’t: A new interpretation of the origin of modern human behavior. J Hum Evol. 2000; 39: 453–563. https://doi.org/10.1006/jhev.2000.0435 PMID: 11102266 5. d’Errico F, Stringer CB. Evolution, revolution or saltation scenario for the emergence of modern cul- tures? Philos Trans R Soc B Biol Sci. 2011; 366: 1060–1069. https://doi.org/10.1098/rstb.2010.0340 PMID: 21357228 20 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China 6. Colagè I, d’Errico F. Culture: The driving force of human cognition. Top Cogn Sci. 2018; Forthcoming. https://doi.org/10.1111/tops.12372 PMID: 30033618 7. Galway-Witham J, Cole J, Stringer C. Aspects of human physical and behavioural evolution during the last 1 million years. J Quat Sci. 2019; 34: 355–378. https://doi.org/10.1002/jqs.3137 8. Pettitt P. The Neanderthal dead. Exploring mortuary variability in Middle Palaeolithic Eurasia. Farming. 2002; 4: 1–26. 9. Soressi M, d’Errico F. Pigments, gravures, parures: Les comportements symboliques controverse´s des Ne´andertaliens. In: Vandermeersch B, Maureille B, editors. Les Ne´andertaliens: Biologie et cultures. Paris: E´ ditions du CTHS; 2007. pp. 297–309. 10. Henshilwood CS, d’Errico F, editors. Homo Symbolicus. The dawn of language, imagination and spiritu- ality. Amsterdam: John Benjamins Publishing Company; 2011. 11. Henshilwood CS, d’Errico F, van Niekerk KL, Coquinot Y, Jacobs Z, Lauritzen S-E, et al. A 100,000- year-old ochre-processing workshop at Blombos Cave, South Africa. Science. 2011; 334: 219–222. https://doi.org/10.1126/science.1211535 PMID: 21998386 12. Vanhaeren M, d’Errico F, van Niekerk KL, Henshilwood CS, Erasmus RM. References Thinking strings: Additional evidence for personal ornament use in the Middle Stone Age at Blombos Cave, South Africa. J Hum Evol. 2013; 64: 599–517. https://doi.org/10.1016/j.jhevol.2013.02.001 PMID: 23498114 13. Joordens JCA, d’Errico F, Wesselingh FP, Munro S, de Vos J, Wallinga J, et al. Homo erectus at Trinil on Java used shells for tool production and engraving. Nature. 2015; 518: 228–231. https://doi.org/10. 1038/nature13962 PMID: 25470048 14. Henshilwood CS, d’Errico F, van Niekerk KL, Dayet L, Queffelec A, Pollarolo L. An abstract drawing from the 73,000-year-old levels at Blombos Cave, South Africa. Nature. 2018; 562: 115. https://doi.org/ 10.1038/s41586-018-0514-3 PMID: 30209394 15. Li Z, Doyon L, Li H, Wang Q, Zhang Z, Zhao Q, et al. Engraved bones from the archaic hominin site of Lingjing, Henan Province. Antiquity. 2019; 93: 886–900. https://doi.org/10.15184/aqy.2019.81 16. Pitarch Martı´ A, d’Errico F, Turq A, Lebraud E, Discamps E, Gravina B. Provenance, modification and use of manganese-rich rocks at Le Moustier (Dordogne, France). PLOS ONE. 2019; 14: e0218568. https://doi.org/10.1371/journal.pone.0218568 PMID: 31314755 17. Aubert M, Brumm A, Ramli M, Sutikna T, Saptomo EW, Hakim B, et al. Pleistocene cave art from Sula- wesi, Indonesia. Nature. 2014; 514: 223–227. https://doi.org/10.1038/nature13422 PMID: 25297435 18. Aubert M, Setiawan P, Oktaviana AA, Brumm A, Sulistyarto PH, Saptomo EW, et al. Palaeolithic cave art in Borneo. Nature. 2018; 564: 254–257. https://doi.org/10.1038/s41586-018-0679-9 PMID: 30405242 19. Aubert M, Lebe R, Oktaviana AA, Tang M, Burhan B, Hamrullah, et al. Earliest hunting scene in prehis- toric art. Nature. 2019; 576: 442–445. https://doi.org/10.1038/s41586-019-1806-y PMID: 31827284 20. Hoffmann DL, Standish CD, Garcı´a-Diez M, Pettitt PB, Milton JA, Zilhão J, et al. U-Th dating of carbon- ate crusts reveals Neandertal origin of Iberian cave art. Science. 2018; 359: 912–915. https://doi.org/ 10.1126/science.aap7778 PMID: 29472483 21. Aubert M, Brumm A, Huntley J. Early dates for ‘Neanderthal cave art’ may be wrong. J Hum Evol. 2018; 125: 215–217. https://doi.org/10.1016/j.jhevol.2018.08.004 PMID: 30173883 22. Pearce DG, Bonneau A. Trouble on the dating scene. Nat Ecol Evol. 2018; 2: 925–926. https://doi.org/ 10.1038/s41559-018-0540-4 PMID: 29632350 23. Slimak L, Fietzke J, Geneste J-M, Ontaño´n R. Comment on “U-Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art”. Science. 2018; 361. https://doi.org/10.1126/science.aau1371 PMID: 30237321 24. White R, Bosinski G, Bourrillon R, Clottes J, Conkey MW, Rodriguez SC, et al. Still no archaeological evidence that Neanderthals created Iberian cave art. J Hum Evol. 2019; 102640. https://doi.org/10. 1016/j.jhevol.2019.102640 25. Aubert M, Brumm A, Tac¸on PSC. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 References The timing and nature of human colonization of Southeast Asia in the Late Pleistocene: A rock art perspective. Curr Anthropol. 2017; 58: S553–S566. https://doi.org/10. 1086/694414 26. Hoffmann DL, Standish CD, Pike AWG, Garcı´a-Diez M, Pettitt PB, Angelucci DE, et al. Dates for Nean- derthal art and symbolic behaviour are reliable. Nat Ecol Evol. 2018; 2: 1044–1045. https://doi.org/10. 1038/s41559-018-0598-z PMID: 29942018 27. Hoffmann DL, Standish CD, Garcı´a-Diez M, Pettitt PB, Milton JA, Zilhão J, et al. Response to Comment on “U-Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art”. Science. 2018; 362. https://doi.org/10.1126/science.aau1736 PMID: 30309914 21 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China 28. Hoffmann DL, Standish CD, Garcı´a-Diez M, Pettitt PB, Milton JA, Zilhão J, et al. Response to Aubert et al.’s reply ‘Early dates for “Neanderthal cave art” may be wrong’ [J. Hum. Evol. 125 (2018), 215–217]. J Hum Evol. 2019; 135: 102644. https://doi.org/10.1016/j.jhevol.2019.102644 29. Conard NJ. Palaeolithic ivory sculptures from southwestern Germany and the origins of figurative art. Nature. 2003; 426: 830–832. https://doi.org/10.1038/nature02186 PMID: 14685236 30. Dutkiewicz E, Wolf S, Floss H, Conard NJ. Les objets en ivoire du Jura souabe. L’Anthropologie. 2018; 122: 447–468. https://doi.org/10.1016/j.anthro.2018.05.003 31. Gaut B, Livingston P, editors. The creation of art: New essays in philosophical æsthetics. Cambridge: Cambridge University Press; 2003. 32. Chen C. Preliminary exploration of the typology and technology of microcore in China—Also of the cul- ture relationship between Northeast Asia and Northwestern North America. Acta Anthropol Sin. 1983; 2: 331–346. 33. Li Z, Wu X, Zhou L, Liu W, Gao X, Nian X, et al. Late Pleistocene archaic human crania from Xuchang, China. Science. 2017; 355: 969–972. https://doi.org/10.1126/science.aal2482 PMID: 28254945 34. Nian XM, Zhou LP, Qin JT. Comparisons of equivalent dose values obtained with different protocols using a lacustrine sediment sample from Xuchang, China. Radiat Meas. 2009; 44: 512–516. https://doi. org/10.1016/j.radmeas.2009.06.002 35. Li H, Li Z, Lotter MG, Kuman K. Formation processes at the early Late Pleistocene archaic human site of Lingjing, China. J Archaeol Sci. 2018; 96: 73–84. https://doi.org/10.1016/j.jas.2018.05.004 36. Trinkaus E, Wu X-J. External auditory exostoses in the Xuchang and Xujiayao human remains: Patterns and implications among eastern Eurasian Middle and Late Pleistocene crania. PLOS ONE. 2017; 12: e0189390. https://doi.org/10.1371/journal.pone.0189390 PMID: 29232394 37. Doyon L, Li H, Li Z, Wang H, Zhao Q. References Further evidence of organic soft hammer percussion and pressure retouch from Lingjing (Xuchang, Henan, China). Lithic Technol. 2019; 44: 100–117. https://doi.org/10. 1080/01977261.2019.1589926 38. Doyon L, Li Z, Li H, d’Errico F. Discovery of circa 115,000-year-old bone retouchers at Lingjing, Henan, China. PLOS ONE. 2018; 13: e0194318. https://doi.org/10.1371/journal.pone.0194318 PMID: 29529079 39. van Kolfschoten T, Li Z, Wang H, Doyon L. The Middle Palaeolithic site of Lingjing (Xuchang, Henan, China): Preliminary new results. Analecta Praehist Leiden. 2020; 50: 21–28. 40. Li Z, Kunikita D, Kato S. Early pottery from the Lingjing site and the emergence of pottery in northern China. Quat Int. 2017; 441: 49–61. https://doi.org/10.1016/j.quaint.2016.06.017 41. Li Z, Xing Y, Mu J, Wu X, Li Y, Kato S. Report of the excavation of microlithic implements on the 5th layer of Lingjing, Xuchang Man site from 2008 to 2013. Huaxia Kaogu / Huaxia Archaeol. 2018; 2018: 3–33. 42. Li Z, Ma H. Techno-typological analysis of the microlithic assemblage at the Xuchang Man site, Lingjing, central China. Quat Int. 2016; 400: 120–129. https://doi.org/10.1016/j.quaint.2015.08.065 43. Li Z, Li Y, Kato S. Observations on microblade core technologies from lever 5 of the Xuchang Man site, Lingjing. Acta Anthropol Sin. 2014; 33: 285–303. 44. Bronk Ramsey C. Methods for summarizing radiocarbon datasets. Radiocarbon. 2017; 59: 1809–1833. https://doi.org/10.1017/RDC.2017.108 45. Reimer PJ, Bard E, Bayliss A, Beck JW, Blackwell PG, Bronk Ramsey C, et al. IntCla13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon. 2013; 55: 1869–1887. 46. Snoeck C, Brock F, Schulting RJ. Carbon exchanges between bone apatite and fuels during cremation: Impact on radiocarbon dates. Radiocarbon. 2014; 56: 591–602. https://doi.org/10.2458/56.17454 47. Cohen DJ. The advent and spread of early pottery in East Asia: New dates and new considerations for the world’s earliest ceramic vessels. J Austronesian Stud. 2013; 4: 55–92. 48. Wu X, Zhang C, Goldberg P, Cohen D, Pan Y, Arpin T, et al. Early pottery at 20,000 years ago in Xian- rendong Cave, China. Science. 2012; 336: 1696–1700. https://doi.org/10.1126/science.1218643 PMID: 22745428 49. Cohen DJ, Bar-Yosef O, Wu X, Patania I, Goldberg P. The emergence of pottery in China: Recent dat- ing of two early pottery cave sites in South China. Quat Int. 2017; 441: 36–48. https://doi.org/10.1016/j. quaint.2016.08.024 50. Wang Y, Zhang S, Gu W, Wang S, He J, Wu X, et al. Lijiagou and the earliest pottery in Henan Province, China. Antiquity. 2015; 89: 273–291. https://doi.org/10.15184/aqy.2015.2 51. Provenzano N. References Techniques et proce´de´s de fabrication des industries osseuses terramaricoles de l’Aˆ ge du bronze. In: Julien M, Averbouh A, Ramseyer D, Bellier C, Buisson D, Cattelain P, et al., editors. Pre´- histoire d’os. Aix-en-Provence: Universite´ de Provence; 1999. pp. 273–288. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 22 / 24 PLOS ONE A Paleolithic bird figurine from China 52. Averbouh A, Provenzano N. Propositions pour une terminologie du travail pre´historique des matières osseuses: I—Les techniques. Pre´histoire Anthropol Me´diterrane´ennes. 1998; 7–8: 5–25. 53. Averbouh A, Bergoue¨n R, Clottes J. Technique et e´conomie du travail du bois de cervide´ chez les Mag- dale´niens d’Enlène (Montesquieu-Avantès, Ariège): Vers l’identification d’un cycle saisonnier de pro- duction? In: Julien M, Averbouh A, Ramseyer D, Bellier C, Buisson D, Cattelain P, et al., editors. Pre´histoire d’os. Aix-en-Provence: Universite´ de Provence; 1999. pp. 289–318. 54. David E´ . Transformation des matières dures d’origine animale dans le Me´solithique de l’Europe du Nord. In: Ramseyer D, editor. Fiches typologiques de l’industrie osseuse pre´historique; Cahier XI Matières et techniques. Paris, France: E´ ditions Socie´te´ Pre´historique Franc¸aise; 2004. pp. 113–149. 55. Le Dosseur G. Sens et contre sens. Re´flexions concernant l’orientation d’un geste technique observe´ sur des objets en matière osseuse du Levant. Pre´histoires Me´diterrane´ennes. 2003; 115–127. 56. Side´ra I, Legrand A. Trace´ologie fonctionnelle des matières osseuses: une me´thode. Bull Socie´te´ Pre´- historique Fr. 2006; 103: 291–304. 57. Gates St-Pierre C, Walker RB, editors. Bone as tools: Current methods and interpretations in worked bone studies. Oxford: Oxbow Books; 2007. 58. Buc N. Experimental series and use-wear in bone tools. J Archaeol Sci. 2011; 38: 546–557. https://doi. org/10.1016/j.jas.2010.10.009 59. LeMoine GM. Use wear on bone and antler tools from the Mackenzie Delta, Northwest Territories. Am Antiq. 1994; 59: 316–334. https://doi.org/10.2307/281935 60. d’Errico F. La vie sociale de l’art mobilier pale´olithique. Manipulation, transport, suspension des objets en os, bois de cervide´s, ivoire. Oxf J Archaeol. 1993; 12: 145–174. https://doi.org/10.1111/j.1468-0092. 1993.tb00289.x 61. Zhang S, d’Errico F, Backwell LR, Zhang Y, Chen F, Gao X. Ma’anshan cave and the origin of bone tool technology in China. J Archaeol Sci. 2016; 65: 57–69. https://doi.org/10.1016/j.jas.2015.11.004 62. Zhang S, Doyon L, Zhang Y, Gao X, Chen F, Guan Y, et al. Innovation in bone technology and artefact types in the late Upper Palaeolithic of China: Insights from Shuidonggou Locality 12. J Archaeol Sci. 2018; 93: 82–93. https://doi.org/10.1016/j.jas.2018.03.003 63. References Effects of temperature on bone tissue. Histo- logical study of the changes in the bone matrix. Forensic Sci Int. 2013; 226: 33–37. https://doi.org/10. 1016/j.forsciint.2012.11.012 PMID: 23287528 79. Castillo RF, Ubelaker DH, Acosta JAL, de la Rosa RJE, Garcia IG. Effect of temperature on bone tissue: Histological changes. J Forensic Sci. 2013; 58: 578–582. https://doi.org/10.1111/1556-4029.12093 PMID: 23458344 80. Cattaneo C, DiMartino S, Scali S, Craig OE, Grandi M, Sokol RJ. Determining the human origin of frag- ments of burnt bone: A comparative study of histological, immunological and DNA techniques. Forensic Sci Int. 1999; 102: 181–191. https://doi.org/10.1016/s0379-0738(99)00059-6 PMID: 10464934 81. Ellingham STD, Thompson TJU, Islam M, Taylor G. Estimating temperature exposure of burnt bone—A methodological review. Sci Justice. 2015; 55: 181–188. https://doi.org/10.1016/j.scijus.2014.12.002 PMID: 25934370 82. Hanson M, Cain CR. Examining histology to identify burned bone. J Archaeol Sci. 2007; 34: 1902– 1913. https://doi.org/10.1016/j.jas.2007.01.009 83. Huisman H, Ismail-Meyer K, Sageidet BM, Joosten I. Micromorphological indicators for degradation processes in archaeological bone from temperate European wetland sites. J Archaeol Sci. 2017; 85: 13–29. https://doi.org/10.1016/j.jas.2017.06.016 84. Imaizumi K. Forensic investigation of burnt human remains. Res Rep Forensic Med Sci. 2015; 5: 67– 74. https://doi.org/10.2147/RRFMS.S75141 85. Mulhern DM, Ubelaker DH. Differences in osteon banding between human and nonhuman bone. J Forensic Sci. 2001; 46: 220–222. https://doi.org/10.1520/JFS14952J PMID: 11305421 86. Walker PL, Miller KWP, Richman R. 7—Time, temperature, and oxygen availability: An experimental study of the effects of environmental conditions on the color and organic content of cremated bone. In: Schmidt CW, Symes SA, editors. The analysis of burned human remains. San Diego: Academic Press; 2008. pp. 129–135. 87. Mitchell WJT. Representation. In: Lentricchia F, McLaughlin T, editors. Critical terms for literary study. Chicago: The University of Chicago Press; 1995. pp. 11–22. 88. Vukcevich M. Representation. The Chicago School of Media Theory RSS. Chicago: Chicago School of Media Theory; 2002. http://csmt.uchicago.edu/glossary2004/representation.htm. 89. Shi X. Issues on bird images and bird worship of the ancient cultures in the eastern coastal area and southeast China. In: Tian C, Shi X, editors. Collection of papers on Chinese prehistoric cultures. Beijing: Wenwu Press; 1989. pp. 234–266. 90. Liu D. Art history and ancient civilization. Taibei: Yuncheng Culture Industrial Co., Ltd.; 1994. 91. Hopkins R. Painting, sculpture, sight, and touch. Br J Æsthet. 2004; 44: 149–166. https://doi.org/10. 1093/bjaesthetics/44.2.149 92. Seitamaa-Hakkarainen P, Huotilainen M, Ma¨kela¨ M, Groth C, Hakkarainen K. How can neuroscience help understand design and craft activity? References Orme CDL, Davies RG, Olson VA, Thomas GH, Ding T-S, Rasmussen PC, et al. Global patterns of geo- graphic range size in birds. PLOS Biol. 2006; 4: e208. https://doi.org/10.1371/journal.pbio.0040208 PMID: 16774453 64. Cracraft J, Barker FK. Passerine birds (Passeriformes). In: Hedges SB, Kumar S, editors. The timetree of life. New York: Oxford University Press; 2009. pp. 423–431. http://www.timetree.org/public/data/pdf/ Cracraft2009Chap61.pdf. 65. Turner–Walker G, Syversen U. Quantifying histological changes in archaeological bones using BSE– SEM image analysis. Archaeometry. 2002; 44: 461–468. https://doi.org/10.1111/1475-4754.t01-1- 00078 66. Bell LS. Palaeopathology and diagenesis: An SEM evaluation of structural changes using backscat- tered electron imaging. J Archaeol Sci. 1990; 17: 85–102. https://doi.org/10.1016/0305-4403(90) 90016-X 67. Bell LS, Boyde A, Jones SJ. Diagenetic alteration to teeth in situ illustrated by backscattered electron imaging. Scanning. 1991; 13: 173–183. https://doi.org/10.1002/sca.4950130204 68. Fairbridge RW. Diagenetic minerals. In: Frye K, editor. The encyclopedia of Mineralogy. Encyclopedia of Earth sciences series. Boston, MA: Springer US; 1981. pp. 121–123. 69. Hedges REM. Bone diagenesis: an overview of processes. Archaeometry. 2002; 44: 319–328. https:// doi.org/10.1111/1475-4754.00064 70. Francillon-Vieillot H, Buffrenil VD, Castanet J, Geraudie J, Meunier FJ, Sire JY, et al. Microstructure and mineralization of vertebrate skeletal tissues. In: Carter JG, editor. Skeletal biomineralization: Pat- terns, processes and evolutionary trends. New York: Van Nortrand Reinhold; 1990. pp. 471–530. 71. de Margerie E, Cubo J, Castanet J. Bone typology and growth rate: testing and quantifying ‘Amprino’s rule’ in the mallard (Anas platyrhynchos). C R Biol. 2002; 325: 221–230. https://doi.org/10.1016/s1631- 0691(02)01429-4 PMID: 12017770 72. Currey JD. Bones: Structure and mechanics. Oxford: Princeton University Press; 2002. http://library. wur.nl/WebQuery/clc/1854910. 73. Enlow DH, Brown SO. A comparative histological study of fossil and recent bone tissues. Part II. Tex J Sci. 1957; 9: 186–204. 74. De Ricqles A, Meunier FJ, Castanet J, Francillon-Vieillot H. Comparative microstructure of bone. In: Hall BK, editor. Bone, Vol 3: Bone matrix and bone specific products. Boston: CRC Press; 1991. pp. 1– 78. 23 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 PLOS ONE A Paleolithic bird figurine from China 75. Locke M. Structure of long bones in mammals. J Morphol. 2004; 262: 546–565. https://doi.org/10.1002/ jmor.10282 PMID: 15376271 76. Ferna´ndez-Jalvo Y, Andrews P. Atlas of taphonomic identifications. New York: Springer Dordrecht; 2016. 77. Bradtmiller B, Buikstra JE. Effects of burning on human bone microstructure: A preliminary study. J Forensic Sci. 1984; 29: 535–540. https://doi.org/10.1520/JFS11701J PMID: 6726157 78. Castillo RF, Ubelaker DH, Acosta JAL, de la Fuente GAC. PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020 References The promise of cognitive neuroscience in design studies. For- mAkademisk—Forskningstidsskrift Des Og Des. 2016; 9. https://doi.org/10.7577/formakademisk.1478 93. Cialone C, Tenbrink T, Spiers HJ. Sculptors, architects, and painters conceive of depicted spaces differ- ently. Cogn Sci. 2018; 42: 524–553. https://doi.org/10.1111/cogs.12510 PMID: 28656679 94. Mene´ndez Ferna´ndez M. Excavaciones arqueologicas en la Cueva del Buxu (Cardes. Cangas de Onis). Excavaciones Arqueol En Astur. 1992; 1992: 69–74. 95. Fortea Pe´rez FJ. Abrigo de la Viña. Informe de la campañas 1980–1986. Excavaciones Arqueol En Astur. 1986; 1986: 55–68. 96. Ortega I, Rios-Garaizar J, Garate Maidagan D, Arizaga J, Bourguignon L. A naturalistic bird representa- tion from the Aurignacian layer at the Cantalouette II open-air site in southwestern France and its rele- vance to the origins of figurative art in Europe. J Archaeol Sci Rep. 2015; 4: 201–209. https://doi.org/10. 1016/j.jasrep.2015.09.009 97. Buisson D, Pinc¸on G. Nouvelle lecture d’un galet grave´ de Gourdan et essai d’analyse des figurations d’oiseaux dans l’art pale´olithique franc¸ais. Antiq Natl. 1986; 18–19: 75–90. 98. Abramova ZA. L’art pale´olithique d’Europe orientale et de Sibe´rie. Grenoble: Je´roˆme Millon; 1995. 24 / 24 PLOS ONE | https://doi.org/10.1371/journal.pone.0233370 June 10, 2020
https://openalex.org/W3094328789
https://zenodo.org/records/4016818/files/MMJ-Vol-63-No-4-Oct-2020_61-64.pdf
English
null
Antimycotic activity of phenoxythiazolchloralum
Zenodo (CERN European Organization for Nuclear Research)
2,020
cc-by
2,746
Abstract Background: The therapeutic options in invasive candidiasis and aspergillosis are limited and don’t provide expected results. Introducing a new drug in the therapeutic practice can improve the quality of life of immunocompromised patients. The aim of this research is to study the antimycotic activity of new substance phenoxithiazolchloralum (MF-0010) on Aspergillus spp., Candida albicans, and Saccharomyces cerevisiae. Background: The therapeutic options in invasive candidiasis and aspergillosis are limited and don’t provide expected results. Introducing a new drug in the therapeutic practice can improve the quality of life of immunocompromised patients. The aim of this research is to study the antimycotic activity of new substance phenoxithiazolchloralum (MF-0010) on Aspergillus spp., Candida albicans, and Saccharomyces cerevisiae. Material and methods: Phenoxythiazolchoralum has been kindly offered by the Institute of Chemistry. The standards were offered by Nicolae Testemitanu State University of Medicine and Pharmacy Antifungal activity of the phenoxythiazolchloralum against Aspergillus spp was evaluated by microdilution Background: The therapeutic options in invasive candidiasis and aspergillosis are limited and don’t provide expected results. Introducing a new drug in the therapeutic practice can improve the quality of life of immunocompromised patients. The aim of this research is to study the antimycotic activity of new substance phenoxithiazolchloralum (MF-0010) on Aspergillus spp., Candida albicans, and Saccharomyces cerevisiae. Material and methods: Phenoxythiazolchoralum has been kindly offered by the Institute of Chemistry. The standards were offered by Nicolae Testemitanu State University of Medicine and Pharmacy. Antifungal activity of the phenoxythiazolchloralum against Aspergillus spp. was evaluated by microdilution method. The successive double dilution method was used for determination of in vitro susceptibility against Candida albicans, and Saccharomyces cerevisiae. Results: For the first time, it was studied in vitro susceptibility of MF-0010 against Aspergillus spp., Candida albicans, Saccharomyces cerevisiae. With at least 0 05 μM/ml difference of MF-0010 MIC value from standards it can be considered quite more potent than ketoconazole and bifonazole against Material and methods: Phenoxythiazolchoralum has been kindly offered by the Institute of Chemistry. The standards were offered by Nicolae Testemitanu State University of Medicine and Pharmacy. Antifungal activity of the phenoxythiazolchloralum against Aspergillus spp. was evaluated by microdilution method. The successive double dilution method was used for determination of in vitro susceptibility against Candida albicans, and Saccharomyces cerevisiae. Results: For the first time, it was studied in vitro susceptibility of MF-0010 against Aspergillus spp., Candida albicans, Saccharomyces cerevisiae. Antimycotic activity of phenoxythiazolchloralum 1Department of Pharmaceutical and Toxicological Chemistry, 2Drug Scientific Center Nicolae Testemitanu State University of Medicine and Pharmacy, Chisinau, the Republic of Moldova 3Laboratory of Organic Synthesis and Biopharmaceuticals, Institute of Chemistry, Chisinau, the Republic of Moldova Cite this article Cite this article Podgornii A, Valica V, Pogrebnoi S, Lupascu L, Uncu A, Macaev F. Antimycotic activity of phenoxythiazolchloralum. Mold Med J. 2020;63(4):61-64. doi: 10.5281/zenodo.4016818. N H N H S HO O N N S O Cl Cl Fig. 1. The chemical structure of phenoxythiazolchloralum (MF-0010) Authors’ ORCID iDs, academic degrees and contributions are available at the end of the article *Corresponding author: ana.podgornii@usmf.md Manuscript received July 21, 2020; revised manuscript August 28, 2020; published online September 15, 2020 A. Podgornii et al. Moldovan Medical Journal. October 2020;63(4):61-64 A. Podgornii et al. Moldovan Medical Journal. October 2020;63(4):61-64 ORIGINAL research DOI: 10.5281/zenodo.4016818 UDC: 615.282 DOI: 10.5281/zenodo.4016818 UDC: 615.282 Cite this article Podgornii A, Valica V, Pogrebnoi S, Lupascu L, Uncu A, Macaev F. Antimycotic activity of phenoxythiazolchloralum. Mold Med J. 2020;63(4):61-64. doi: 10.5281/zenodo.4016818. Abstract With at least 0.05 μM/ml difference of MF-0010 MIC value from standards, it can be considered quite more potent than ketoconazole and bifonazole against Aspergillus fumigatus, Aspergillus versicolor, Aspergillus ochramensis. The MIC/MCF ratios of MF-0010 are lower than nistatine ones with 0.08 μMol for both pathogens: Candida albicans, and Saccharomyces cerevisiae. p g y Conclusions: All analyzed pathogens were susceptible to MF-0010. According to experimental data on Aspergillus spp., the antimycotic activity of MF-0010 is quite better than the standards one. The MIC/MCF ratios showed in vitro significant susceptibility of MF-0010 against Candida albicans. Key words: Phenoxythazolchloralum, Aspergillus spp., Candida albicans, Saccharomyces cerevisiae, drug discovery. Introduction Patients with AIDS, organ transplants, cancer, diabetes, autoimmune disorders, biomedical-assist devices, or long- term antibiotic therapy are at major risk of developing my- cosis [1, 2]. Reports of invasive aspergillosis and candidiasis in immunocompromised patients have been increasing [3, 4]. Candida spp. can affect almost all organ systems in the human body. The genome plasticity and high reproductive capacity constitute a serious risk to human health [5, 6]. Ni- statine in one of the few topical drugs for the treatment of cutaneous mycosis caused by this pathogen [7]. Reports of resistance to antimycotic agents have been reported for sev- eral years [8]. Thus, every year it is more difficult to predict clinical success /failure and duration of antimycotic therapy. The therapeutic options are quite limited and don’t provide expected results [9, 10]. The introduction of new drugs could help to manage this critical situation. h Fig. 1. The chemical structure of phenoxythiazolchloralum (MF-0010) Acute toxicity tests reveal that MF-0010 in mice after in- traperitoneal injection and intragastric administration for a single dose suggests that compound has minimal toxicity, and can be placed in the toxicity class 5 according to TG 423 Acute Toxic Class Method (OECD) [13]. Organoleptic characteristics, solubility, melting temperature and UV ab- sorption spectra are described by O. Butescu [14]. It was observed that compound has poor solubility in polar sol- vents (water, methanol, ethanol, acetonitrile, acetone) and low bioavailability that limits the use as antitubercular agent. Following research of chemical structure showed that substance could have other biological activities. The The computer-aided study of novel 5-aryl-2thio- 1,3,4-oxadiazoles reported in vitro anti-tubercular activ- ity of phenoxythiazolchloralum (MF-0010) (fig. 1) against Mycobacterium tuberculosis H37Rv [11]. Pharmacophore groups –NH-, =C=O, =N-N=, -Cl produce three-dimen- sional arrangements that are required for anti-tubercular activity [12]. 61 A. Podgornii et al. Moldovan Medical Journal. October 2020;63(4):61-64 ORIGINAL research presence of the azole ring (pharmacophore group of azole antimycotic drug) suggests that the studied substance may have antimycotic activity.h 24 hours. On the second day, a preliminary analysis of the results was made. The last tube from the series in which no visible growth of microorganisms has been detected was considered to be the minimal inhibitory concentration (MIC) of the preparation. The aim of this research is to study the antimycotic acti- vity of MF-0010 on Aspergillus spp., Candida albicans, and Saccharomyces cerevisiae. Results For evaluating of the antifungal activity for Candida al- bicans, and Saccharomyces cerevisiae was used the successive double dilution method. For this, at the initial stage, 1 mL of Sabouraud broth for test fungus was introduced into a series of 10 tubes. Subsequently, 1 mL of the analyzed compound was dropped into the first test tube.ht Introduction For the estimation of the minimum fungicidal concen- tration (MFC), the contents of the test tubes with MIC and with higher concentrations were seeded on Sabouraud agar from Petri dishes with the use of the bacteriological loop. The seeded dishes were kept in the thermostat at 35oC for 24 hours. The concentration of preparation, which did not allow the growth of any colony of microorganisms, was con- sidered to be the minimal fungicidal concentration of the compound [16]. Aspergillus spp. For the first time, it was studied in vitro susceptibility of MF-0010 against A. fumigatus, A. versicolor, A. ochramensis and A. niger. The MIC and MFC values of MF-0010 against Aspergillus spp. ranged from 0.23 μM/ml – 0.62 μM/ml (fig. 2) and 0.62 μM/ml –1.24 μM/ml (fig. 3), respectively. The highest values of MIC and MCF of MF-0010 are related to A. niger. Thus, the use of MF-0010 against this pathogen is not appropriate. h i Then, the obtained mixture was pipetted, after that 1 mL of it was transferred to the next tube, so the procedure was repeated until the tube no 10 of the series. Thus, the con- centration of the initial preparation decreased 2-fold in each subsequent tube. The MIC values of MF-0010 on A. fumigatus, A. versi- color, A. ochramensis and A. niger are 0.23 μM/ml -0.31 μM/ ml, which are lower than MICs of standards: ketoconazole (0.28 μM/ml-0,38 μM/ml) and bifonazole (0.32 μM/ml-0.48 μM/ml). With at least 0.05 μM/ml difference from stan- dards, MF-0010 can be considered quite more potent than ketoconazole and bifonazole. At the same time, 24 hour test fungi of Candida albi- cans and Saccharomyces cerevisiae were prepared. Initially, suspensions of test fungi were prepared with optical densi- ties (D.O.) of 0.5 according to the McFarland index. Sub- sequently, 1 mL of the obtained fungal suspension was dropped in a tube containing 9 mL of sterile distilled water. The content of the tube was mixed, after which 1 mL was transferred to the tube no 2 of the 5-tube series containing 9 mL of sterile distilled water. The MCF value of MF-0010 against A. fumigatus, A. ver- sicolor, A. ochramensis and A. niger is 0.62 μM/ml, which is lower than MFC values of standards: ketoconazole – 0.94 μM/ml (exception is A.ochramensis – 0.38 μM/ml) and bi- fonazole – 0.64 μM/ml. With a 0.02 μM/ml difference from standards, MF-0010 has the same antimycotic activity as bi- fonazole, and better one than ketoconazole. From the 5-th tube of the series was taken 0.1 mL of the fungal suspension, which represents the seeded dose and added to each tube with titrated preparation. Subsequently, the tubes with titrated preparation and the seeded doses of the microorganisms were put in the thermostat at 35°C for Fig. 2. The Minimum Inhibitory Concentration (MIC) of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochramensis and A. niger Fig. Aspergillus spp. For evaluating of the antifungal activity of the phenoxy- thiazolchloralum compound against Aspergillus spp. was used the microdilution method described by E. Stingaci et al. [15]. References For the first time, it was studied the in vitro inhibition potential of MF-0010 against Candida albicans, Saccharo- myces cerevisiae (fig. 4). h 1. Calandra T, Roberts JA, Antonelli M, Bassetti M, Vincent JL. Diagnosis and management of invasive candidiasis in the ICU: an updated approach to an old enemy. Crit Care. 2016;20(1):1-6. doi: 10.1186/s13054-016- 1313-6. i The MIC/MCF ratios of MF-0010 for inhibition of Can- dida albicans, Saccharomyces cerevisiae are lower than ni- statine ones with 0.08 μMol for both pathogens. Thus, we can conclude that MF-0010 is more potent active molecule than nistatine against Candida albicans. Further research is needed to create topical forms and confirm by clinical trial the pharmacological action, pharmacokinetic profile and safety of formulations with MF-0010. Several types of topi- cal medication including creams, ointments, and pastes can be designed. 2. Vandeputte P, Ferrari S, Coste AT. Antifungal resistance and new strategies to control fungal infections. Int J Microbiol. 2012;1-26. doi: 10.1155/2012/713687. 3. Mosquera J, Denning DW. Azole cross-resistance in Aspergillus fumiga- tus. Antimicrob Agents Chemother. 2002;46(2):556-557. doi: 10.1128/ AAC.46.2.556-557.2002. 4. Pfaller JB, Messer SA, Hollis RJ, Diekema DJ, Pfaller MA. In vitro susceptibility testing of Aspergillus spp.: comparison of e-test and reference microdilution methods for determining voriconazole and itraconazole MICs. J Clin Microbiol. 2003;41(3):1126-1129. doi: 10.1128/ JCM.41.3.1126-1129.2003. 5. Srinivasan A, Lopez-Ribot JL, Ramasubramanian AK. Overcoming antifungal resistance. Drug Discov Today Technol. 2014;11:1-12. doi: 10.1016/j.ddtec.2014.02.005. Aspergillus spp. 2. The Minimum Inhibitory Concentration (MIC) Fig. 2. The Minimum Inhibitory Concentration (MIC) of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochramensis and A. niger of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochramensis and A. niger 62 ORIGINAL research A. Podgornii et al. Moldovan Medical Journal. October 2020;63(4):61-64 A. Podgornii et al. Moldovan Medical Journal. October 2020;63(4):61-64 ORIGINAL research Fig. 3. The Minimum Fungicidal Concentration (MCF) of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochramensis and A. niger Fig. 3. The Minimum Fungicidal Concentration (MCF) g g of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochramensis and A. niger of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochram of MF-0010, ketoconazole, bifonazole against A. fumigatus, A. versicolor, A. ochramensis and A. niger 0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 Candida albicans Saccharomyces cerevisiae MIC/ MCF ratios (in μMol) MF-0010 Nistatine Fig. 4. MIC/MCF ratios of MF-0010, nistatine against Candida albicans, Saccharomyces cerevisiae Fig. 4. MIC/MCF ratios of MF-0010, nistatine against Candida albicans, Saccharomyces cerevisiae Candida albicans and Saccharomyces cerevisiaei Conclusions 6. Oliveira Santos GC, Vasconcelos CC, Lopes AO, Sousa Cartagenes MS, Filho AK, Nascimento FR, Ramos RM, Pires ER, Andrade MS, Rocha FM, Andrade Monteiro C. Candida infections and therapeutic strate- gies: mechanisms of action for traditional and alternative agents. Front Microbiol. 2018;9:1-23. doi: 10.3389/fmicb.2018.01351. In this study, we found that all analyzed pathogens were susceptible to MF-0010. According to the experimental data, the antimycotic activity of MF-0010 is quite better than the standard one. The MIC and MCF values of MF-0010 show a good potency against Candida albicans, and new studies are warranted in order to design optimized formulations, to analyze in vivo the efficacy and quality assurance of for- mulations. 7. Lotan AM, Lysy I, Wiener-Well Y, Karni-Shahroor S, Gronovich Y. Topical Nystatin Treatment for Candida Infection Following Wound Reconstruction. Wounds. 2018;30(2):41-44. 8. Fisher MC, Hawkins NJ, Sanglard D, Gurr SJ. Worldwide emergence of resistance to antifungal drugs challenges human health and food security. Science. 2018;360(6390):739-742. doi: 10.1126/science.aap7999. 63 A. Podgornii et al. Moldovan Medical Journal. October 2020;63(4):61-64 ORIGINAL research 9. Roemer T, Krysan DJ. Antifungal drug development: challenges, unmet clinical needs, and new approaches. Cold Spring Harb Perspect Med. 2014;4:1-14. doi: 10.1101 /cshperspect.a019703. 13. Parii S, Valica V, Macaev F, Pogrebnoi S, Boldescu V, Stingaci E, Duca G, Ivancic A, Rusnac L, Nicolai E, Ungureanu A, Uncu L, Macaeva A. Determinarea toxicităţii acute a unor noi compuşi chimici cu proprietăţi antituberculoase [Determination of acute toxicity of chemical com- pounds with antituberculosis properties]. [Bull Acad Sci Mold. Med Sci]. 2015;(1/46):445-451. Romanian. 10. Verweij PE, Ananda-Rajah M, Andes D, Arendrup MC, Brüggemann RJ, Chowdhary A, Cornely OA, Denningh DW, Groll AH, Izumikawa K, Kullberg BJ, Lagrou K, Maertens J, Meis JF, Newton P, Page I, Seyed- mousavi S, Sheppard DC, Viscoli C, Warris A, Donnelly JP. International expert opinion on the management of infection caused by azole-resistant Aspergillus fumigatus. Drug Resist Updat. 2015;21-22:30-40. doi: 10.1016/j.drup.2015.08.001. 14. Butescu O, Doroschevici A. Determination of physicochemical pa- rameters of the phenoxithiazolchloralum. In: Nicolae Testemitanu State University of Medicine and Pharmacy. Collection of scientific summaries of students, residents and young researchers. Chisinau: Medicina; 2015. p. 292. j p 11. Macaev F, Rusu G, Pogrebnoi S, Gudima A, Stingaci E, Vlad L, Shvets N, Kandemirli F, Dimoglo A, Reynolds R. Synthesis of novel 5-aryl-2-thio- 1,3,4-oxadiazoles and the study of their structure-anti-mycobacterial activities. Bioorg Med Chem. 2005;13(16):4842-4850. doi: 10.1016/j. bmc.2005.05.011. 15. Authors’ contribution AP interpreted the data, drafted the first manuscript. VV formulated the research hypothesis, revised the manuscript. SP synthetized MF-0010, performed the technological part of laboratory work. LL designed the study, conducted the laboratory work. AU performed the technological part of laboratory work. FM interpreted the data, revised the manuscript. All the authors revised and approved the final version of the manu- script. Conclusions Stingaci E, Zveaghinteva M, Pogrebnoi S, Lupascu L, Valica V, Uncu L,  Smetanscaia A, Drumea M, Petrou A, Ciric A, Glamoclija J, Sokovic M, Kravtsov V, Geronikaki A, Macaev F. New vinyl-1,2,4-triazole derivatives as antimicrobial agents: Synthesis, biological evaluation and molecular docking studies. Bioorg Med Chem Lett. 2020;30(17):127368. doi: 10.1016/j.bmcl.2020.127368. 12. Macaev F, Ribkovskaia Z, Pogrebnoi S, Boldescu V, Rusu G, Shvets N, Dimoglo A, Geronikaki A., Reynolds R. The structure-antituberculosis activity relationships study in a series of 5-aryl-2-thio-1,3,4-oxadiazole derivatives. Bioorg Med Chem. 2011;19(22):6792-6807. doi: 10.1016/j. bmc.2011.09.038. j 16. Saint Petersburg Pasteur Scientific Research Institute of Epidemiology and Microbiology. Metod seriinykh razvedenii v bul’one [Serial broth dilution method] [Internet]. Sankt Petersburg, 2020 [cited 2020 july 27]. Available from: https://www.dntpasteur.ru/metodic2_4_2_2. Russian. Authors’ ORCID iDs and academic degrees Vladimir Valica, PharmD, PharmPhD, Professor – https://orcid.org/0000-0002-1068-5504. Serghei Pogrebnoi, ChemD, ChemPhD, Associate Professor – https://orcid.org/0000-0003-2827-505X. Fliur Macaev, ChemD, ChemPh, Professor – https://orcid.org/0000-0002-3094-1990. Ethics approval and consent to participate No approval was required for this study. No approval was required for this study. Ethics approval and consent to participate Ethics approval and consent to participate No approval was required for this study. Fundingh g This study was supported by Nicolae Testemitanu State University of Medicine and Pharmacy. The trial was the authors’ initiative. The authors are independent and take responsibility for the integrity of the data and accuracy of the data analysis. Conflict of Interests Conflict of Interests l No competing interests were disclosed. 64
https://openalex.org/W2278877312
https://tidsskrift.dk/rvt/article/download/22573/19863
Danish
null
Jehovas Vidner. I grænsefladen af den danske religionsmodel
Religionsvidenskabeligt tidsskrift
2,015
cc-by
6,446
RVT 6 2 (2 0 1 5 ) 1 2 5 -1 3 7 RVT 6 2 (2 0 1 5 ) 1 2 5 -1 3 7 1 Taleren fremstår både i stævneprogrammet og i politianmeldelsen anonymt. Politianmeldelsen var følgelig rettet mod Jehovas Vidner i Danmark (Jehovas Vidners områdestævne. Program 2013). Jehovas Vidner. I grænsefladen af den danske religionsmodel AN NI KA H VI TH AMAR ENGLISH ABSTRACT: Jehovah’s Witnesses have existed as a minority religion in Denmark for more than hundred years. I argue that even Jehovah’s Witnesses are clearly within, although situated in the borderland of the Danish model of religion, they are also located in a borderland position when seen from juridical, theological and sociological points of views. I discuss the borderland position on the basis of a media debate that took place in the summer of 2013, when Jehovah’s Witnesses were accused of racism due to speeches given at the yearly summer convent DANSK RESUMÈ: Jehovas Vidner har eksisteret som minoritetsreligion i Danmark i mere end hundrede år. Artiklen argumenterer for, at selv om Jehovas Vidner klart er indenfor, men i grænsefladen af den danske religionsmodel, er Jehovas Vidner stadig i en grænseposition ud fra andre juridiske, teologiske og sociologiske synsvinkler. Den- ne grænseposition har dog flyttet sig længere og længere ind mod midten i de senere år. Artiklen diskuterer grænsepositionen ud fra en mediedebat i sommeren 2013. Her blev Jehovas Vidner anklaget for racisme som følge af indhold i taler afholdt på det år- lige sommerstævne. 2 §266b lyder: Den, der offentligt eller med forsæt til udbredelse i en videre kreds fremsætter udtalelse eller anden meddelelse, ved hvilken en gruppe af personer trues, forhånes eller nedværdiges på grund af sin race, hudfarve, nationale eller etniske oprindelse, tro eller seksuelle orientering, straffes med bøde el- ler fængsel indtil 2 år. Stk. 2. Ved straffens udmåling skal det betragtes som en særligt skærpende om- stændighed, at forholdet har karakter af propagandavirksomhed (Straffeloven). KEYWORDS: Jehovah’s Witnesses; minority religion; inclusion; conflict Ved Jehovas Vidners årlige sommerstævne i sommeren 2013 var temaet for en af talerne “Tag dig i agt for frafaldne!”.1 Her omtalte taleren frafaldne Jehovas Vidner på denne måde: Roskildesyge. Mon ikke det er noget, der giver associationer i dig, når du hører det ord. Roskildesyge. Mon ikke det er noget, der giver associationer i dig, når du hører det ord. Måske har du gjort dig de smertelige erfaringer. Så er jeg ked af at minde dig om det. Hvis ikke du ved, hvad det er, så kan vi bare sige: "Du er ikke gået glip af noget!" Måske har du gjort dig de smertelige erfaringer. Så er jeg ked af at minde dig om det. Hvis ikke du ved, hvad det er, så kan vi bare sige: "Du er ikke gået glip af noget!" […]Den åndelige sygdom, der hedder "frafalden", den kan på mange måder sammen- lignes med en smittefarlig sygdom. Hvad er en frafalden? En frafalden er en, der har været i sandheden, men nu har forladt den og er blevet modstander af alt, der har med Jehova at gøre. Symptomerne på den sygdom er, at man udtaler sig imod sine tidligere trosfæller, og man har et brændende ønske om at trække andre ud af sandheden, ud af tilbedelsen for Jehova. Som du kan se, er det en dødsensfarlig sygdom. Hvilket åndeligt organ bliver angrebet, hvis du er angrebet af "frafaldensygen"? 1. Tim. 6:3 siger, at en der har forladt sandheden er syg i sindet eller fordærvet i sindet (”Tag dig i agt for fra- faldne” 2013). Talen er opbygget efter en retorisk model, som kendetegner Jehovas Vidners skrifter og foredrag. Den tager udgangspunkt i en konkret, nutidig situation, som i dette tilfælde omhandler de frafaldne medlemmer af Jehovas Vidner. Der bruges konkrete eksempler, som vækker genkendelse hos tilhøreren, her ved at sammenligne med roskildesyge. Derefter fortælles det, hvor i Bibelen der gives retningslinjer for at forholde sig til problemet, her i forhold til 1. Tim, hvis tema er, hvordan menigheder skal opbygges og drives. Herunder, hvordan frafaldne fra menigheden skal opfattes. Den klare afstandstagen fra eks-medlemmer, der er fjendtligt indstillede over for Jehovas Vidner, har været karakteristisk for bevægelsen helt tilbage fra dens konso- lideringstid i begyndelsen af det 20. århundrede (Beckford 1975, 15-16, 55; Watch Tower Bible and Tract Society 1993, 186-87). Støttegruppen for tidligere Jehovas Vidner valgte dog den 25. 126 Annika Hvithamar KEYWORDS: Jehovah’s Witnesses; minority religion; inclusion; conflict juli 2013, på baggrund af denne tale, at indgive en anmeldelse af Jehovas Vidner til politiet (Politianmeldelse 30.7.2013) i forhold til Straffelovens § 266b.2 I løbet af den næste uges tid nåede sagen rundt i de danske trykte medier, og avisernes lederskribenter, religionspolitiske aktører og danske politiske partier bi- drog med deres synspunkter på sagen. Efterfølgende blev anklagen ændret til kun at gælde to talere fra sommerstævnet. Disse blev afhørt i december 2013, men i august 2014 blev sagen henlagt af rigsadvokaten (Erik Jørgensen, personlig korrespondance, 8.8.2014). Det er denne sag og dennes udtryk for, hvordan Jehovas Vidner balancerer i grænseområdet af den danske religionsmodel, som er omdrejningspunktet for denne artikel. I det følgende vil jeg efter at have diskuteret den danske religionsmodel ana- lysere, hvordan Jehovas Vidner, historisk og i dag, bevæger sig i dette grænseområ- de. Den danske religionsmodel ”Religionsmodellen er det institutionaliserede og historisk betingede aftryk, som religionssystemets udvikling og interaktion med staten og samfundet har efterladt os med” (Christoffersen et al. 2012, 14). Den danske religionsmodel er dermed kende- tegnet ved en række træk af historisk, sociologisk, teologisk og mentalitetsmæssig art, som kendetegner Danmark. Særligt er den kendetegnet ved et tæt forhold mel- lem et bestemt religiøst samfund, folkekirken, og den danske stat. Dette forhold er igen baseret på en tæt historisk sammenknytning mellem den lutherske kirke og den danske stat. I modsætning til de andre nordiske lande har Danmark indtil for nylig ikke haft religiøse minoriteter med så stor indflydelse, at de har kunnet påvirke reli- gionsmodellen. Folkekirkeligheden er, måske derfor, kendetegnet ved liberalisme og sekularisering (Iversen 2012, 28, 34). Der kan også argumenteres for, at den danske religionsmodel er påvirket af Grundtvigs frihedsbegreb. For Grundtvig var tanke- og åndsfrihed tæt knyttet til kravet om religionsfrihed (Jørgensen 2012, 193-194). Grundtvigs religionsopfattelse var rettet mod individets frihed i forhold til staten. Grundtvigianismens sejrsgang i det danske kirkelige landskab har gennem de sidste 150 år betydet, at den frie religi- onsudøvelse, som ikke kontrolleres af et højkirkeligt miljø, kendetegner religions- udøvelse i Danmark (Iversen 2012, 28). I kort form kan det siges, at den danske religionsmodel er kendetegnet ved høj grad af religiøs autonomi blandt de praktiserende og en udbredt opfattelse af, at enhver er fri til at bestemme, hvordan deres religion skal praktiseres. Det udtrykkes direkte i grundlovens centrale paragraffer om religionsfrihed (§67) og frihed til at danne et trossamfund (§70). KEYWORDS: Jehovah’s Witnesses; minority religion; inclusion; conflict Det er desuden en religionsmodel, der er stærkt påvirket af den udgave af kristendom, som formidles af den danske folkekirke. Jehovas Vidner. I grænsefladen af den danske religionsmodel 127 3 Det var først i 1931, at Jehovas Vidner fik deres nuværende navn. Indtil da kaldte de sig selv Bibelstuden- terne. Det var bevægelsens anden leder, J. F. Rutherford (1869-1941), som gav Jehovas Vidner sit navn. Han stod for en centralisering af bevægelsen og en skærpelse af dens grundlæggende forestillinger og grænser udadtil (Chryssides 2008, 121). Jehovas Vidner som ‘matter out of place’ Religionssociologen Eileen Barker har påpeget, at nye religiøse bevægelser opfattes af majoritetssamfundet som det, Mary Douglas kaldte for “matter out of place” (Bar- ker 2001, 202). Med dette udtryk mener Barker en forurening, som opstår, når et fænomen ikke kan kategoriseres i kendte grupper. Douglas pointerede, at mennesket skaber orden ud af kaos ved at kategorisere fænomener, ordne dem i grupper og adskille dem fra hinanden. Men i denne kategorisering opstår fænomener som svært lader sig kategorisere. Det er Barkers pointe, at minoritetsreligioner ofte får denne position, idet de, i deres religiøst betingede normer og forestillinger balancerer på grænsen af, hvad majoriteten ser som acceptabelt (Barker 1994, 97-98; 2006, 202). I det følgende vil jeg argumentere for, at Jehovas Vidner bevæger sig i grænsefla- den af den danske religionsmodel. Jehovas Vidner er på mange måder en integreret del af det danske religiøse landskab, men på afgørende områder adskiller bevægel- Jehovas Vidner. I grænsefladen af den danske religionsmodel 127 sen sig fra de normer, som religiøse minoriteter forventes at leve op til i den danske religionsmodel. Disse normer er blandt andet udtrykt i grundlovens bestemmelser om religionsfrihedens begrænsning (§70) og om regler for godkendelse af trossam- fund, herunder retten til at melde sig ud. Herved bliver Jehovas Vidner i Douglask forstand out of place (Barker, 2006, 207). De er religiøse og de er danske, men ikke på den helt acceptable måde 4 Frikirker og Sekter udkom første gang i 1927 og blev senest genoptrykt i 1958. 128 Annika Hvithamar 5 Se Hvithamar (2007) for en længere behandling af danske Jehovas Vidners forhold til værnepligt og behandlinger med blod, herunder hvordan Jehovas Vidners oprettelse af Afdelingen for Hospitalsinfor- mation i 1979 sidenhen har påvirket udviklingen af blodløs kirurgi. Jehovas Vidner i Danmark Jehovas Vidner har eksisteret i Danmark siden 1895, og siden 1904 er bladet Vagttår- net regelmæssigt udkommet på dansk (Hvithamar 2007, 224-225).3 Bevægelsen blev i dens etableringsfase opfattet som kontroversiel, fordi Jehovas Vidner afgrænsede sig stærkt over for alle andre religioner, i Danmark over for den danske folkekirke. Modviljen var gensidig. Når Jehovas Vidner i det 20. århundrede sporadisk blev beskrevet i faglitteraturen, var det negativt. Jehovas Vidner blev betegnet som afvi- gende – både i religiøs og i psykologisk forstand. Fremstillingen af Jehovas Vidner i præsten og kirkehistorikeren Michael Neiiendams meget brugte bog Frikirker og Sekter, kan tjene som eksempel.4 I bogen, som blev genudgivet fire gange i løbet af århundredet, optræder Jehovas Vidner første gang i udgaven fra 1948 med følgende beskrivelse: I en Organisation som “Jehovas Vidner” findes adskillige uens Typer, hos hvem Gra- derne i Had til og Bitterhed mod Samfundet ligger højere og lavere. Der er Kynikere, udpræget polemisk betonede Propagandister, som føler Fryd ved at kredse om de fore- staaende Massakrer i Slaget ved Harmagedon. Der er nøjsomme, moralsk udadelige Mennesker, som i deres aandelige Jævnhed faktisk ikke kender alle Enkeltheder i Læ- ren; de har, hvad man kalder Barnetroen i Behold, er besveget i dette svigefulde Liv, men er til Gengæld blevet grebet af den glødende Forkyndelse om den forestaaende Verdenskatastrofe, som skal forvandle denne Jammerdal til et Paradis, og de vil være på Herrens Side, naar han nu snart kommer igen (Neiiendam 1948, 222). En opfattelse af Jehovas Vidner som mennesker, hvis medlemskab af bevægelsen er determineret af enten intellektuel, social eller psykisk uformåenhed, er karakteristisk for ovenstående citat. Det er en fremstilling, som ikke er ualmindelig for beskrivelser af minoritetsreligioner, men i denne sammenhæng er det også en fremstilling, som søger at placere Jehovas Vidner uden for den danske religionsmodel. Jehovas Vidner beskrives i ovenstående citat ikke som en del af det danske samfund, kendetegnet ved et frit forhold til religion, men som individer, som står udenfor, og er drevet af had til samfundet. Gennem størstedelen af det 20. århundrede betragtedes Jehovas Vidner som stå- ende udenfor med en fejlagtig, asocial religionsudøvelse. Men i den sidste tredjedel af århundredet skete der en gradvis accept. Det er denne udvikling, jeg vil beskrive nedenfor. Jehovas Vidner. I grænsefladen af den danske religionsmodel 129 Teologiske grænsedragninger Det er afvisningen af treenighedslæren, som mest markant adskiller Jehovas Vidner fra de store kirkesamfund. De anser treenighedslæren for at være et resultat af en senere udvikling og ikke i overensstemmelse med, hvad Bibelen fortæller om Jesus’ lære. Jesus opfattes i stedet som den første skabning (Hvad er det Bibelen virkelig lærer 2005, 41). Jehovas Vidners opfattelse kommer f.eks. til udtryk i Ny Verden- oversættelsen af Bibelen. Dette er blandt andet tilfældet i Joh 1, 1, der i Ny Verden- oversættelsen lyder “I begyndelsen var Ordet, og Ordet var hos Gud, og Ordet var en gud”. Mens der i Bibelselskabets oversættelse står “I begyndelsen var Ordet og Ordet var hos Gud og Ordet var Gud”. I Ny Verden-oversættelsen er “Ordet” såle- des en selvstændig guddom, mens “Ordet” i Bibelselskabets oversættelse er en del af Gud. Siden J. F. Rutherford i 1917 blev valgt som leder af bevægelsen (se ovenfor), har Jehovas Vidner været kritisk over for andre religioner, der identificeres som “Baby- lon den Store” i Joh Åb 17, 1, 15 (Holden 2002, 20; Chryssides 2008, 12-13). I bevægel- sens skrifter indgår derfor også regelmæssigt anklager mod især andre kirkesamfund i forhold til teologiske uenigheder f.eks. om helgeners status (Vågn op! 11, 2010, 20- 21), ikondyrkelse (Vågn op 8, 2008, 20-21) og præsters involvering i politik (Vågn op! 5, 2004, 3-5). Også i 1990’ernes religionssociologiske forskningsformidling er forfatterne forbe- holdne over for den teologiske legitimitet ved Jehovas Vidners afvisning af treenig- hedslæren. I gymnasiebogen Minoritetsreligioner i Danmark – Religionssociologisk set skriver lektor i kristendomshistorie Jørgen Skafte Jensen: Ifølge denne [treenighedslæren] er Jesus Kristus Gud, nemlig Gud Sønnen. Dette cen- trale kristne dogme afvises af Jehovas Vidner. Det står der nemlig ikke noget om i Bibe- len, siger de. Det korrekte i dette synspunkt kan naturligvis diskuteres, men vidnerne har naturligvis ret i, at Oldkirkens treenighedslære ikke findes i Bibelen i samme form, og at Bibelen i øvrigt er åben for fortolkninger, også når det gælder spørgsmålet om Je- sus som gud. Det er måske alligevel værd at bemærke, at det kun er de moderne, bibel- kritiske og “anti-fundamentalistiske” bibelforskere, der mener, at der kan være tvivl om, at forestillingen om Jesus som gud rent faktisk kan læses ud af teksterne (Jensen 1991, 156-57). Juridiske grænsedragninger I grænsefladen af den danske religionsmodel 129 Vidner som trossamfund og vedtagelsen af de to love er vigtige markeringer af, at Jehovas Vidner har bevæget sig fra at befinde sig langt ude i den danske religions- models grænseflade til at komme længere og længere indenfor. Vidner som trossamfund og vedtagelsen af de to love er vigtige markeringer af, at Jehovas Vidner har bevæget sig fra at befinde sig langt ude i den danske religions- models grænseflade til at komme længere og længere indenfor. Juridiske grænsedragninger Den gradvise ændring af majoritetssamfundets opfattelse af Jehovas Vidner giver sig bedst til udtryk i den juridiske anerkendelse af bevægelsen og dens særsynspunkter. Da reglerne for godkendelse af trossamfund blev ændret i 1969 søgte Jehovas Vidner om at blive et godkendt trossamfund og blev godkendt i 1970 og fik dermed den blåstempling, som er implicit i statens godkendelse (Warburg 1996, 21). Den kends- gerning, at Jehovas Vidner faktisk søgte om godkendelse er et betydeligt skridt fra trossamfundets side mod at anerkende spillereglerne i den danske religionsmodel. I løbet af det 20. århundrede har medlemmer af Jehovas Vidner siddet i fængsel som et resultat af deres afvisning af militær værnepligt. Men i 1998 blev det muligt at aftjene civil værnepligt, og siden da har Jehovas Vidners afstandstagen fra at bære våben ikke været problematisk for den danske stat (Militærnægterloven §1, 2006). Derudover betød en udvikling i patientlovgivningen, at det siden 1998 har været en rettighed for patienter, at der ikke må indledes behandlinger, som indebærer transfusion af blod eller blodprodukter, uden patientens informerede samtykke (Sundhedsloven §24, 2010). Denne ændring i patientlovgivningen betød, at Jehovas Vidner nu fik ret til at afvise behandlinger med blod.5 Da denne lovgivning specifikt retter sig mod netop Jehovas Vidners særlige forhold til blodtransfusioner må lovændringen siges at være det mest vægtige udtryk for, at det danske samfund har anerkendt Jehovas Vidners synspunkter som legitime. Den danske stat har gradvist accepteret særtræk ved Jehovas Vidner. Det skal dog understreges, at lovgivningen er et resultat af en forhandling gennem årene. På den ene side har Jehovas Vidner tilnærmet sig samfundet. I forhold til militærnæg- telse tillod Jehovas Vidner ikke sine medlemmer at udføre militærnægtertjeneste frem til 1996, selvom det foregik ved civile myndigheder. Men i 1996 ændrede Det Styrende Råd holdning og vedtog, at det kun var den militære tjeneste, som Jehovas Vidner skulle tage afstand fra (Vagttårnet 5, 1996, 19-20). På den anden side har det danske samfund udviklet sig i retning af at tillade borgerne mere frihed i forhold til værnepligten. Det giver sig til udtryk i militærnægterloven, hvilket bevidner en generel positiv indstilling overfor afvisning af værnepligt. Godkendelsen af Jehovas Jehovas Vidner. 130 Annika Hvithamar Sociologiske grænsedragninger Ifølge den danske minoritetsforsker Helen Krag bliver minoriteter til i ordenspara- digmer (Krag 2007, 61). Det vil sige som det, der er afvigende eller unormalt eller det som har behov for at blive tolereret. For i lighed med Barkers observation om minori- teter som grænsetilfælde, så kan minoriteter defineres som dem, der ikke rigtig hører til, blandt andet fordi de opfører sig på en anden måde og har andre normer og vær- dier end majoriteten. De repræsenterer en anden form for orden, og det gælder også for Jehovas Vidner. Ifølge den danske minoritetsforsker Helen Krag bliver minoriteter til i ordenspara- digmer (Krag 2007, 61). Det vil sige som det, der er afvigende eller unormalt eller det som har behov for at blive tolereret. For i lighed med Barkers observation om minori- teter som grænsetilfælde, så kan minoriteter defineres som dem, der ikke rigtig hører til, blandt andet fordi de opfører sig på en anden måde og har andre normer og vær- dier end majoriteten. De repræsenterer en anden form for orden, og det gælder også for Jehovas Vidner. Jehovas Vidner lever og arbejder blandt majoritetsbefolkningen og er, i en dansk sammenhæng, en forholdsvis stor bevægelse.6 Men deres normer og værdier er på kant med de nationale danske. De fejrer ikke nationale højtider som jul, personlige festdage som fødselsdage eller årstidshøjtider som fastelavn. De bruger ikke Danne- brog, og selv om Jehovas Vidner godt kan have venner og bekendte blandt ikke- medlemmer, er man påpasselig med at undgå skadelige påvirkninger fra den om- kringliggende verden. I bevægelsen bruges begrebet at kunne “holde menigheden ren” (Watch Tower Bible and Tract Society 1993, 187).7 Da seksualnormerne i bevægel- sen er konservative, tillades heller ikke sex før og uden for ægteskab eller praktiseret homoseksualitet (Vågn op! 1, 2012, 28-29; 9, 2013, 4-5). Set fra majoritetssamfundets side, afviser Jehovas Vidner fællesskabets normer og værdier. Medlemmer af Jehovas Vidner takker nej til fødselsdagsinvitationer, børn må ikke klæde sig ud til fastelavn, de tillader ikke kærester at bo sammen før ægteskab, og de afviser i det hele taget at deltage i fællesskabsritualer, som de anser som ikke-kristne. Derudover er det Jeho- vas Vidners praksis at udelukke medlemmer, som overtræder bevægelsens normer samt deres afvisning af at modtage blod, der medfører de største kontroverser i for- hold til majoritetens normer og værdier. 6 Ifølge Jehovas Vidners egne tal var der 14.484 forkyndere i Danmark i 2013 (Jehovas Vidners årbog 2014, 82). Tallet angiver Jehovas Vidner, som er aktive, hvorfor tal, som inkluderer udøbte børn og andre, som ikke er aktive, vil være noget højere. Som enkeltstående bevægelse gør det Jehovas Vidner til den største minoritetskristne gruppe i Danmark, bortset fra den katolske kirke. Teologiske grænsedragninger Det er interessant, at Jehovas Vidner her sættes på linje med “moderne, bibelkritiske” og “antifundamentalistiske bibelforskere”, idet bevægelsens læsning af Bibelen netop er fundamentalistisk; men pointen er her, at Jehovas Vidner gennem deres manglen- de accept af treenighedslæren placeres uden for de bevægelser, som anerkender et centralt kristent dogme. Sådan som den danske folkekirke gør det. Ud over treenighedslæren har der også været teologisk fokus på Jehovas Vidners eskatologi. At Jesus Kristus skal komme og dømme levende og døde er også en del af den apostolske trosbekendelse i folkekirkens højmesser. Men alligevel er Jehovas Vidners nærforventning et træk, der i majoritetskristendom opfattes som afvigende. 7 Begrebet bygger på formaningerne fra 1. Kor 5 1-13, hvor omgang med ‘utugtige mennesker’ forbydes. Jehovas Vidner. I grænsefladen af den danske religionsmodel 131 9 Dagbladet Struer 26. juli 2013. Politiken 27. juli 2013. Morsø Folkeblad 29. juli 2013 og Kristeligt Folkeblad 1. august 2013. Sociologiske grænsedragninger For ved ovenstående praksis sætter med- lemmer af bevægelsen de religiøse værdier over alle andre værdier, inklusiv familie- sammenhold og overlevelse, hvilket i en dansk kontekst er normbrydende. Jehovas Vidner. I grænsefladen af den danske religionsmodel 131 I forhold til andre danske religiøse grupper er Jehovas Vidners normer og værdi- er dog ikke altid enestående. Hvad angår seksualmoral er Jehovas Vidner på linje med maksimalistiske retninger inden for dansk islam og kristendom f.eks. Indre Mission. Ligesom man blandt mange religiøse minoriteter kan finde en påpasselig- hed mod at indgå i samme form for ungdomskultur som den dominerende danske. Det er i forhold til de nationale ritualer og fællesskabsritualer som fastelavn og fød- selsdage, at Jehovas Vidner har særstandpunkter. Med andre ord er det de ofte over- lappende ritualer, der dels knytter sig til det danske nationale fællesskab, dels til det folkekirkelige fællesskab, som gør Jehovas Vidner til en omstridt gruppe blandt andre religiøse mindretalsgrupper i den danske religionsmodels grænseflade. Desu- den indebærer Jehovas Vidners overbevisning om at sætte de religiøse værdier over alle andre værdier, at de sætter sig udover det frihedsbegreb, som karakteriserer religionsudøvelsen i Danmark og som er et implicit normativt element i den danske religionsmodel. 132 Annika Hvithamar 8 Søgningen er foretaget gennem Infomedia i perioden 16. juli 2013 – 30. august 2013. Sommerstævnet 2013 Jehovas Vidners sommerstævne i 2013 og den efterfølgende debat om, hvorvidt an- meldelsen for racisme var på sin plads, er et eksempel på, hvordan Jehovas Vidner overskrider grænsen for, hvad man finder er acceptabelt i det danske samfund. Som nævnt i artiklens begyndelse bevirkede talen ved sommerstævnet, at Jehovas Vidner blev anmeldt efter straffelovens §266b, fordi den sammenlignede frafaldne Jehovas Vidner med en smittefarlig sygdom . I løbet af de næste uger affødte sagen 83 artikler i danske trykte medier.8 Det var karakteristisk, at det i høj grad var artikler fra de regionale og lokale aviser, som tog sagen op, samt at sagen i flere tilfælde endte på lederplads.9 Den store spredning i mediedækningen afspejler, at Jehovas Vidner, som nævnt ovenfor, er en af de største religiøse minoriteter i Danmark og er geografisk spredt over hele landet. Tilstrækkeligt mange danskere har kendskab til bevægelsen til, at historier om den kan synes relevante for lederskribenter, både lokalt og natio- nalt. I løbet af de uger, hvor debatten fandt sted, skete der også en udvikling af artik- lernes karakter. I de første dage var det særligt selve nyheden, om at Jehovas Vidner blev anmeldt, som dominerede medierne. Derfor var det også især juridiske argu- menter og juridiske eksperter, der blev hørt. F.eks. udtalte lederen af Dokumentati- ons- og Rådgivningscentret mod Racediskrimination, Niels-Erik Hansen, at “Talerne er den samme slags retorik, som man hørte mod jøderne i Europa i 1930'erne. Det er præcis det, som racismeparagraffen er sat i verden for at dæmme op for” (Kristeligt Folkeblad 25. juli 2013). Vinklen i artiklerne var, at når sådanne udtalelser kom til udtryk i det offentlige rum, så faldt de ind under racismeparagraffen. Vinklen blev 132 Annika Hvithamar fulgt op af andre artikler, som efterlyste øget kontrol med de danske trossamfund, og herunder muligheden for at fratage Jehovas Vidner deres godkendelse som trossam- fund. Dette var blandt andet tilfældet da jurist og professor Lisbet Christoffersen udtalte: “Vi skal sætte nogle grænser inden for den plads, der skal være til religions- og ytringsfriheden, og jeg kan godt forstå de tidligere Jehovas Vidner, der siger, at nu er det nok, for det er meget grove ord, der bliver sagt om dem” (Politiken 29. juli 2013, 6). Sommerstævnet 2013 Efterhånden kom dog også andre aktører på banen, som, i lighed med repræsen- tanter fra bevægelsen selv, påpegede, at henvisningerne til sygdom kom fra Bibelen, og at et forbud dermed ville stå i vejen for religionsfriheden. F.eks. skrev sognepræst Katrine Lilleør, at “Det er kun totalitære stater, der forbyder tro. Det er Jehovas vid- ners demokratiske ret at tro, at deres frafaldne er giftig jord. De bør også have ret til at sige det højt. Så højt, at vi alle kan høre, hvad Jehovas Vidner er for nogle” (Ber- lingske Tidende 3. august 2013, 20). På samme linje skrev forhenværende kirkeminister Birthe Rønn Hornbech: Jamen det er da for galt, at Jehovas Vidner forfølger tidligere medlemmer. Ja, det er det. Men det er der allerede lovgivet for. Efter straffeloven kan der gives tilhold, hvis med- lemmer af trossamfundet chikanerer frafaldne og bliver ved med at trænge ind på dem (Kristeligt Folkeblad 16. august 2014, 7). Debatten viste, at der var røster i majoriteten, som ønskede at ekskludere Jehovas Vidner fra den danske religionsmodel, mens andre debattører påviste det juridisk og politiske uholdbare i at anfægte godkendelsen af Jehovas Vidner som trossamfund eller at foretage indskrænkninger i deres ytringsfrihed. Debatten viste, at der var røster i majoriteten, som ønskede at ekskludere Jehovas Vidner fra den danske religionsmodel, mens andre debattører påviste det juridisk og politiske uholdbare i at anfægte godkendelsen af Jehovas Vidner som trossamfund eller at foretage indskrænkninger i deres ytringsfrihed. Debatten inddrog også et sociologisk perspektiv, som omhandlede Jehovas Vid- ners praksis med at afskære kontakten til tidligere medlemmer. Temaet var her de personlige omkostninger, som er forbundet med denne praksis. F.eks. blev det radi- kale folketingsmedlem Marlene Borst Hansen citeret for at sige: “Jeg tror, at vi alle sammen har hørt historier eller kender mennesker, for hvem det har været meget svært at bryde ud af Jehovas Vidner. Det er sørgeligt, og det er klart, at når det er et krav i loven,10 så skal vi sikre os, at de anerkendte trossamfund lever op til det” (Poli- tiken 25. juli 2013). En lignende udtalelse kom fra Støttegruppen for tidligere Jehovas Vidners formand, Poul Dal. Han udtalte: Og netop dette vilkår – at miste sin familie – har de fleste på træffet tilfælles. 10 Det bør bemærkes, at der i den danske lovgivning ikke findes en lov, hvor uhindret udmeldelse er et krav for et trossamfunds godkendelse. I Vejledende retningsliner udarbejdet af det Rådgivende Udvalg vedr. Trossamfund (2011, opdateret 27. marts 2014), skrives der ikke om praksis for fratagelse af en anbe- faling til godkendelse. Klarere juridiske retningslinjer for trossamfund uden for folkekirkenvar et punkt, som professor Lisbet Christoffersen rejste i løbet af den ovennævnte mediedebat. Se fx Lisbet Christoffer- sen “Ingen holder øje med trossamfundene i Danmark”, Politiken, 29. juli 2013. Jehovas Vidner. I grænsefladen af den danske religionsmodel 133 Lederskribenten i Morsø Folkeblad var enig: Lederskribenten i Morsø Folkeblad var enig: Religion bør være et middel til at opnå et bedre liv for sig selv og sammen med andre. Manipulation og trusler hører ikke hjemme i den sammenhæng, og finder man belæg for den opførsel i Bibelen eller andre hellige skrifter, kan man kun sige: Vorherre beva- res! (Morsø Folkeblad 29. juli 2013, 2). I ovenstående indlæg er pointen, at religion er og bør være ikke-fundamentalistisk, og det som gør, at man får et godt liv sammen. Religiøse krav må således ikke sættes over frihed, over fællesskabet eller skabe ufred. I lederskribenternes udsagn under- streger de Jehovas Vidners omstridte position i den danske religionsmodels grænse- flade, idet at Jehovas Vidner lader religiøse bud påvirke deres handlinger på en må- de, som har negativ effekt på det danske samfund. Det er karakteristisk for alle artiklerne i (dags)pressen, at argumentationen er se- kulær. Artiklernes forfattere argumenterer i forhold til dansk jura, i forhold til indi- videts ve og vel og i forhold til danske normer. Der påkaldes ikke transcendente kræfter. Det synes selvfølgeligt i en dansk avis, men det indebærer, at Jehovas Vidner får mindre legitimitet. Ved ikke at anerkende religiøse argumenter bliver disse ar- gumenter underordnet et krav om, at de skal indpasses i en sekulær kontekst. Bevæ- gelser der, ligesom Jehovas Vidner, handler ud fra en religiøs legitimering, befinder sig i strid med de normer som udgør et sekulært samfund. 134 Annika Hvithamar Sommerstævnet 2013 Som tidli- gere Jehovas Vidne har man et vist behov for at møde andre som en selv, et behov, som næsten kan sammenlignes med det behov, overlevende fra koncentrationslejre har for at møde andre, der har samme skæbne (Kristeligt Folkeblad 17. august 2013, 4). næsten kan sammenlignes med det behov, overlevende fra koncentrationslejre har for at møde andre, der har samme skæbne (Kristeligt Folkeblad 17. august 2013, 4). Jehovas Vidner. I grænsefladen af den danske religionsmodel 133 I forhold til citater som disse er det spørgsmålet om, hvad der er acceptabel religion, som er temaet. Det forhold, at Jehovas Vidner lader religiøse normer gå over familie- sammenhold, vurderes, i ovenstående citater, negativt. Dette gælder også den tema- tik, som omhandler spørgsmålet om Jehovas Vidners “danskhed”. Dette emne berør- te chefredaktøren for Dagbladet Ringkøbing i følgende leder: I bund og grund er al fundamentalisme usund. Om den udspringer fra den islamiske verden, fra politiske partier eller Jehovas Vidner er sådan set underordnet, for med fundamentalismen følger fanatismen, og i min verden er det svært at komme i tanke om noget mere ubehageligt – eller udansk for den sags skyld” (Dagbladet Ringkøbing 26. juli 2013, 13). Lederskribenten i Morsø Folkeblad var enig: I grænsefladen af den danske religionsmodel Jehovas Vidner er i grænsefladen af den danske religionsmodel. Jeg har dog også undersøgt nogle andre aspekter end de juridiske i diskussionen om Jehovas Vidner som et grænsetilfælde. Historisk set, og sammenlignet med den danske folkekirke, er bevægelsen ny, men den har i løbet af de sidste hundrede år bevæget sig i retning af at blive accepteret som en del af det danske religiøse landskab. Det ses gennem be- vægelsens forholdsvis store udbredelse her i landet og især gennem den accept af bevægelsens særstandpunkter, som giver sig til udtryk i den danske lovgivning. Teologisk set er bevægelsen principielt stadig langt fra luthersk kristendom. Jehovas Vidners afvisning af treenighedslæren, deres eskatologi og andre religiøse minori- tets- særstandpunkter indebærer, at de er på kant med de store kirkesamfund. I en dansk sammenhæng bør det dog bemærkes, at de teologiske uenigheder fylder min- dre i dag end i bevægelsens tidlige historie. Den afteologisering, som er foregået i Danmark, indebærer, at den teologiske lære, som folkekirken bekender sig til, står uklar for mange danskere (Andersen & Lüchau 2011, 81). Dermed ikke sagt, at Jeho- vas Vidner er accepteret teologisk i Danmark i dag, blot at kontroverserne er blevet mindre. Sociologisk set er Jehovas Vidner en del af det danske samfund og lever og arbejder sammen med majoriteten. Bevægelsens normer er dog på kanten af de majo- ritetsdanske. Dette er mest udtalt i forhold til spørgsmålet om udelukkelse af med- lemmer, som bryder med Jehovas Vidners religiøse normer. Dette sidste hænger sammen med det overordnede kontrovers: at Jehovas Vidner lader deres religiøse normer være bestemmende for deres retorik og omgang med majoritetssamfundet. Som eksemplet med artiklerne i forbindelse med racisme-anklagen viser, accepteres en religiøs diskurs ikke. Som nævnt i begyndelsen af artiklen, er det danske samfund kendetegnet ved sekularisme, liberalisme og et udpræget frihedsbegreb. I forhold til disse punkter står Jehovas Vidner udenfor. Samtidig viser Jehovas Vidners udvikling gennem det sidste århundrede, at grænserne rykkes. Selv om Jehovas Vidner stadig er matter out of place i forhold til det danske samfund og den danske religionsmodel, bevæger de sig gradvist i retning af en inklusion. Jehovas Vidner. I grænsefladen af den danske religionsmodel 135 LITTER AT UR Andersen, Peter B. & Peter Lüchau 2011 “Individualisering og aftraditionalisering af danskernes religiøse værdier”, in: Peter Gundelach, ed, Små og store forandringer. Danskernes værdier siden 1981. Hans Reitzels forlag, København, 76-96. Barker, Eileen 2006 “We’ve Got to Draw the Line Somewhere: An Exploration of Boundaries that Define Locations of Religious Identity”, Social Compass 53 (2), 201-213. 1994 “But is it a Genuine Religion?", in: A. Greil & T. Robbins, eds., Between Sacred and Secular: Research and Theory on Quasi-religion. Religion and the Social Order 4, JAI Press, Inc., Greenwich & London. Beckford, James A. 1975 The Trumphet of Prophecy. A Sociological Study of Jehovah’s Witnesses, Basil Blackwell, Oxford. Bruun, Christoffer Emil 2013 “Retten til sludder: Frihed for Jehova såvel som for Thor”, Politiken 27. juli, 1. Christensen, Søren 2013 “Her hjælper politiet ikke”, Dagbladet Struer, 26. juli, 1. Christoffersen, Lisbet 2013 “Ingen holder øje med trossamfundene i Danmark”, Politiken, 29. juli 2013. Christoffersen, Lisbet, Hans Raun Iversen, Niels Kærgård og Margit Warburg 2012 “Den danske religionsmodel”, in: Lisbet Christoffersen at al., eds., Fremtidens danske Religionsmodel, Anis, København, 23-39. 2012 “Den danske religionsmodel på vej mod et tippepunkt”, in: Lisbet Christoffersen et al., eds., Andersen, Peter B. & Peter Lüchau Andersen, Peter B. & Peter Lüchau 2011 “Individualisering og aftraditionalisering af danskernes religiøse værdier”, in: Peter Gundelach, ed, Små og store forandringer. Danskernes værdier siden 1981. Hans Reitzels forlag, København, 76-96. 2006 “We’ve Got to Draw the Line Somewhere: An Exploration of Boundaries that Define Locations of Religious Identity”, Social Compass 53 (2), 201-213. 1994 “But is it a Genuine Religion?", in: A. Greil & T. Robbins, eds., Between Sacred and Secular: Research and Theory on Quasi-religion. Religion and the Social Order 4, JAI Press, Inc., Greenwich & London. Beckford, James A. 1975 The Trumphet of Prophecy. A Sociological Study of Jehovah’s Witnesses, Basil Blackwell, Oxford. Bruun Christoffer Emil Bruun, Christoffer Emil Bruun, Christoffer Emil 2013 “Retten til sludder: Frihed for Jehova såvel som for Thor”, Politiken 27. juli, 1. 3 “Retten til sludder: Frihed for Jehova såvel som for Thor”, Polit hristensen, Søren Christensen, Søren 2013 “Her hjælper politiet ikke”, Dagbladet Struer, 26. juli, 1. 2013 “Her hjælper politiet ikke”, Dagbladet Struer, 26. juli, 1. 2013 “Ingen holder øje med trossamfundene i Danmark”, Politiken, 29. juli 2013. 2013 “Ingen holder øje med trossamfundene i Danmark”, Politiken, 29. juli 2013. 136 Annika Hvithamar LITTER AT UR offersen, Lisbet, Hans Raun Iversen, Niels Kærgård og Margit Warbu Christoffersen, Lisbet, Hans Raun Iversen, Niels Kærgård og Margit Warburg 2012 “Den danske religionsmodel”, in: Lisbet Christoffersen at al., eds., Fremtidens danske Religionsmodel, Anis, København, 23-39. 2012 “Den danske religionsmodel på vej mod et tippepunkt”, in: Lisbet Christoffersen et al., eds., Jehovas Vidner. I grænsefladen af den danske religionsmodel 135 Fremtidens danske Religionsmodel, Anis, København, 381-84. Chryssides, George D. 2008 Hi t i l Di ti f J h h’ Wit Th S P L h M h Chryssides, George D. 2008 Historical Dictionary of Jehovah’s Witnesses, The Scarecrow Press, Lanham, Maryham. Hornbech, Birthe Rønn 2013 “Overvågningssamfundet: Der bør siges fra over for kontrol af vor religionsudøvelse”, Kristeligt Folkeblad, 16. august, 6. Hvithamar, Annika Hvithamar, Annika 2007 Verdensreligioner i Danmark – Kristendom, Gyldendal, København. 007 Verdensreligioner i Danmark – Kristendom, Gyldendal, Københa Iversen, Hans Raun 2012 “Den danske folkekirke i nordisk belysning”, in: Lisbet Christoffersen et al., eds., Fremtidens danske Religionsmodel, Anis, København, 23-39. Jensen, Jørgen Skafte 1991 “Jehovas Vidner”, in: Tim Jensen, ed, Minoritetsreligioner i Danmark – Religionssociologisk set, Columbus, Viborg, 151-177. Jørgensen, Jonas Adelin 2012 “Sindbilledsprog og hjertefrihed”, in: Lisbet Christoffersen et al., eds., Fremtidens danske Religionsmodel, Anis, København, 183-197. Korsgaard, Kristine 2013 “Ingen holder øje med trossamfundene i Danmark”, Politiken, 29. juli, 6. Krag, Helen 2007 Mangfoldighed, magt og minoriteter – introduktion til minoritetsforskningens teorier, Forlaget Samfundslitteratur, Frederiksberg. Lilleør, Katrine 2013 “Jehovas grovheder”, Berlingske Tidende, 3. august, 20. Lilleør, Katrine 2013 “Jehovas grovheder”, Berlingske Tidende, 3. august, 20. Madsen, Anders Ellebæk 2013 “Gør ikke Bibelen kriminel”, Kristeligt Folkeblad, 1. august, 2. Militærnægterloven 2001 https://www.retsinformation.dk/forms/R0710.aspx?id=6464 (set 30.6.2014). https://www.retsinformation.dk/forms/R0710.aspx?id=6464 (set 30.6 Neiiendam, Michael 1948 Frikirker og Sekter, Gad, København. Politianmelselse “Politianmeldelse af Vagttårnets Bibel- og Traktatselskab, danske afdeling, Stenhusvej 28, 4300 Holbæk, 30 juli”). Schnabel, Laura Elisabeth 2013 “Tidligere Jehovas Vidner holder møde”, Kristeligt Folkeblad, 17. august, 4. g j 2013 “Jehovas Vidner risikerer racismesag”, Kristeligt Folkeblad, 25. juli, 1. Straffeloven ff 2013 https://www.retsinformation.dk/Forms/r0710.aspx?id=152827 (set 30.6.2014). g pp g 2013 “Politianmeldelse af Vagttårnets Bibel- og Traktatselskab, danske afdeling, Stenhusvej 28, 4300 Holbæk, 30 juli.” http://www.kristeligt-dagblad.dk/modules/fsArticle/download.php?fileid=251 (set 23.3.2014). 2010 https://www.retsinformation.dk/forms/R0710.aspx?id=130455 (set 30.6.2014). Svarrer, Jens Peter 2013 “Religions(u)frihed”, Morsø Folkeblad, 29. juli, 1. , J 2013 “Religions(u)frihed”, Morsø Folkeblad, 29. juli, 1. 2013 “Religions(u)frihed”, Morsø Folkeblad, 29. juli, 1. 136 Annika Hvithamar Støttegruppen for tidligere Jehovas Vidner 2013 “Politianmeldelse af Vagttårnets Bibel- og Traktatselskab. Jehovas Vidner. I grænsefladen af den danske religionsmodel 137 LITTER AT UR Program”, danske afdeling, Stenhusvej 28, 4300 Holbæk, 30. juli. Warburg, Margit Warburg, Margit 1996 “Lige ret for Loke såvel som for Thor? Religionsbegreber og retspraksis i forbindelse med religioner uden for folkekirken”, Chaos. Dansk-norsk tidsskrift for religionshistoriske studier 26, 9-32. u e o o e i e , Chaos. Dansk norsk tidsskrift for religionshistoriske studier , Watch Tower Bible and Tract Society of Pennsylvania 2014 Jehovas Vidners årbog 2014. Wachtturm Bibel- und Traktatgesellschaft der Zeugen Jehovas, E. V. Selters/Taunus. 2013 “Hvad siger Bibelen? Sex før ægteskabet”, Vågn op! 9, 4-5. 2012 “Unge spørger: Hvordan kan jeg forklare Bibelens syn på homoseksualitet?”, Vågn op! 1, 28-29. 2010 “Hvad siger Bibelen?: Hvordan skal man se på det at bede til ‘helgener’?”, Vågn op! 11, 20-21. 2009 “Hvad siger Bibelen?: Bør der være et skel mellem præster og lægfolk?”, Vågn op! 8, 22-23. 2008 “Hvad siger Bibelen? Skal Gud tilbedes ved hjælp af billeder? Vågn op! 8, 20-21. 2005 Hvad er det Bibelen virkelig lærer? Watch Tower Bible and Tract Society of New York. 2004 “Skal præster blande sig i politik?”, Vågn op! 5, 3-5. 1996 “Det, der tilkommer kejseren”, Vagttårnet 5, 20-21. 1993 Jehovas Vidner. Forkyndere af Guds rige. Watch Tower Bible and Tract Society of New York. [N.a.] 2013 Jehovas Vidners områdestævne, Program, [N.p]. [N.a.] 2013 “Tag dig i agt for frafaldne”, Transskription af lydfil foretaget af David Bille Maagaard, http://www.x-jv.dk/nyt/transcript_frafaldne.html, (set 30.6.2014). Watch Tower Bible and Tract Society of Pennsylvania Watch Tower Bible and Tract Society of Pennsylvania 2014 Jehovas Vidners årbog 2014. Wachtturm Bibel- und Traktatgesellschaft der Zeugen Jehovas, E. V. Selters/Taunus. 2013 “Hvad siger Bibelen? Sex før ægteskabet”, Vågn op! 9, 4-5. g g g p 2012 “Unge spørger: Hvordan kan jeg forklare Bibelens syn på homoseksualitet?”, Vågn op! 1, 28-29. 2012 “Unge spørger: Hvordan kan jeg forklare Bibelens syn på homosek 2010 “Hvad siger Bibelen?: Hvordan skal man se på det at bede til ‘helgener’?”, Vågn op! 11, 20-21. g p g g p 2009 “Hvad siger Bibelen?: Bør der være et skel mellem præster og lægfolk?”, Vågn op! 8, 22-23. 8 “Hvad siger Bibelen? Skal Gud tilbedes ved hjælp af billeder? V 2005 Hvad er det Bibelen virkelig lærer? Watch Tower Bible and Tract Society of New York. 2005 Hvad er det Bibelen virkelig lærer? Watch Tower Bible an 2004 “Skal præster blande sig i politik?”, Vågn op! 5, 3-5. 1996 “Det, der tilkommer kejseren”, Vagttårnet 5, 20-21. 1993 Jehovas Vidner. Forkyndere af Guds rige. Warburg, Margit Watch Tower Bible and Tract Society of New York. [N.a.] 2013 “Tag dig i agt for frafaldne”, Transskription af lydfil foretaget af David Bille Maagaard, http://www.x-jv.dk/nyt/transcript_frafaldne.html, (set 30.6.2014). Institut for Tværkulturelle og Regionale Studier, Københavns Universitet
https://openalex.org/W4306861391
https://bjbas.springeropen.com/counter/pdf/10.1186/s43088-022-00313-0
English
null
Investigation of the corrosion inhibition potentials of some 2-(4-(substituted)arylidene)-1H-indene-1,3-dione derivatives: density functional theory and molecular dynamics simulation
Beni-Seuf University Journal of Basic and Applied Sciences /Beni-Suef University Journal of Basic and Applied Sciences
2,022
cc-by
10,255
*Correspondence: emmanueltoba90@gmail.com; oluwatoba. oyeneyin@aaua.edu.ng; dammynath@yahoo.com 1 Theoretical and Computational Chemistry Unit, Department of Chemical sciences, Adekunle Ajasin University, Akungba‑Akoko, Ondo state, Nigeria 2 Department of Chemistry, Faculty of Science, University of Ibadan, Oyo State, Nigeria F ll li t f th i f ti i il bl t th d f th ti l g Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. Investigation of the corrosion inhibition potentials of some 2‑(4‑(substituted) arylidene)‑1H‑indene‑1,3‑dione derivatives: density functional theory and molecular dynamics simulation Oluwatoba Emmanuel Oyeneyin1*   , Nathanael Damilare Ojo2*   , Nureni Ipinloju1   , Eric Bamidele Agbaffa3    and Abiodun Vestor Emmanuel1 Oluwatoba Emmanuel Oyeneyin1*   , Nathanael Damilare Ojo2*   , Nureni Ipinloju1   , Eric Bamidele Agbaffa3    and Abiodun Vestor Emmanuel1 Abstract Background:  Corrosion is a threat to material strength and durability. Electron-rich organic inhibitor may offer good corrosion mitigation potentials. In this work, anti-corrosion potentials of nine derivatives of 1H-indene-1,3-dione have been investigated using density functional theory (DFT) approach and molecular dynamics (MD) simulation. Chemi- cal reactivity descriptors like energies of lowest unoccupied molecular orbital ­(ELUMO), highest occupied molecular orbital ­(EHOMO), electron affinity (A), ionization potential (I), energy gap (ΔEgap), global hardness (η), global softness (σ), electronegativity (χ), electrophilicity (ω), number of transferred electrons (ΔN) and back-donation (ΔEback-donation) were computed at DFT/B3LYP/6-31G(d) theoretical level. The local reactive sites and the charge partitioning on the com- pounds were studied using Fukui indices and molecular electrostatic potential (MEP) surface analysis. The adsorption behavior and the binding energy of the inhibitors on Fe (110) surface in hydrochloric acid solution were investigated using MD simulation. Results:  The high chemical reactivity, kinetic instability and good corrosion inhibition potentials demonstrated by the inhibitors are rationalized based on their high ­EHOMO, A, σ, ΔN, ΔEback-donation, and low ΔEgap, ­ELUMO, I and η. A wide difference of approximately 2.4–3.2 eV between the electronegativities of iron and the 1H-inden-1,3-diones suggests good charge transfer tendency from the latter to the low-lying vacant d-orbitals of iron. The heteroatoms (O and N) and the aromatic moieties are the nucleophilic sites on the inhibitors for effective adsorption on the metal surface as shown by condensed Fukui dual functions and MEP analysis. The MD simulation shows good interaction and strong binding energy between the inhibitor and Fe (110) surface. Conclusions:  Effective surface coverage and displacement of ­H3O+, ­Cl− and water molecules from Fe (110) surface by the inhibitors indicate good corrosion inhibition properties of the inden-1,3-diones. Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 https://doi.org/10.1186/s43088-022-00313-0 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 https://doi.org/10.1186/s43088-022-00313-0 Beni-Suef University Journal of Basic and Applied Sciences Open Access 1  Background [10, 11]. Corrosion inhibitors can be divided into organic and inorganic corrosion inhibitor [12]. Organic inhibi- tors are preferably used over inorganic inhibitors due to their cheap availability and also, they are less toxic com- pare to inorganic inhibitors [13]. Organic inhibitors with heterocyclic organic structures containing pi conjugated structure and heteroatoms, like N, S, O and P, capable of adsorbing on the metal surfaces via physisorption or chemisorption adsorption, are often being most effective [14]. Commonly adopted corrosion inhibitors are organic compounds including imidazolines, azoles, Schiff bases, amines and their derived salts [15–17]. Metal surface protection from corrosion is a major con- cern in metallurgical industries, oil factories, environ- ment and chemical plants. Corrosion of metals is a chemical and electrochemical reaction that occurs when metal surfaces are exposed to acidic environment such as in acidic pickling, oil well acidizing and industrial clean- ing [1]. Destruction of metals and materials leads to deg- radation of mechanical properties which, consequently, leads to direct loss of materials strength, reduction in thickness and failure of the material [2, 3]. This failure of operating equipment and material properties due to cor- rosion results into an increase in production and mainte- nance cost and breaking down of operating equipment. The consequences of corrosion are tremendous ranging from health effect, economic effect and safety to environ- mental outlooks [4, 5]. Arylidene-indane-1,3-diones are crucial category of organic compounds with several potential applications in materials science to medical field [18–22]. Arylidene- indane-1,3-dione derivatives possess antibacterial, anti-coagulant and anti-diabetes properties [23]. Their applications have also been reported in electrolumines- cent devices, optoelectronics, eye lens clarification agents and nonlinear optical devices [18, 24, 25]. Although corrosion is a natural process and cannot be completely eliminated, it can be extensively reduced by adopting some methods. The method of corrosion control adopted by many industries includes material selection, rational design, cathodic protection and the use of chemical inhibitors [6–8]. Among the methods adopted for controlling corrosion, the use of chemical inhibitors has been one of the most practical and eco- nomical methods in combating corrosion [9]. Inhibitors mitigate corrosion by adsorbing onto the metal surface and impeding the anodic and/or cathodic electrochemi- cal reactions occurring at the metal-medium interface Pluskota et  al. [21] synthesized some 2-arylidene- indan-1,3-dione derivatives and checked their drug- like properties [26]. Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 2 of 14 2-((4,7-dimethylnaphthalen-1-yl)methylene)-1H-indene-1,3(2H)-dione display low energy gap, strongest binding interaction and most stabilized iron-inhibitor configuration, hence, the best anti-corrosion potential. Keywords:  Anti-corrosion, DFT, Fukui descriptor, Indenedione, Molecular dynamics simulation Graphical abstract 2-((4,7-dimethylnaphthalen-1-yl)methylene)-1H-indene-1,3(2H)-dione display low energy gap, strongest binding interaction and most stabilized iron-inhibitor configuration, hence, the best anti-corrosion potential. Keywords:  Anti-corrosion, DFT, Fukui descriptor, Indenedione, Molecular dynamics simulation Graphical abstract 2-((4,7-dimethylnaphthalen-1-yl)methylene)-1H-indene-1,3(2H)-dione display low energy gap, strongest binding interaction and most stabilized iron-inhibitor configuration, hence, the best anti-corrosion potential. Keywords:  Anti-corrosion, DFT, Fukui descriptor, Indenedione, Molecular dynamics simulation hi l b 1  Background However, the electron-rich structure of these organic systems owing to the pres- ence of π-conjugation and heteroatoms has attracted our group to investigate the corrosion inhibitive potential of these compounds in progress of our research efforts to discover organic molecule with high corrosion inhibition potential. Page 3 of 14 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Computational approaches such as DFT and MD simu- lations are considered to be effective, versatile and trend- ing methods of designing and rationalizing corrosion inhibition properties of small to large systems [27–29]. The computational method provides insight into the reactivity parameters of a molecule and the relationship between these parameters and corrosion inhibition effi- ciency [30]. Also, MD simulations provide insight into the orientation, adsorption and binding energy of the inhibitor molecules on metallic surface [31]. Understand- ing the relationship between corrosion inhibition poten- tials, molecular structure and adsorption mechanism of organic compounds helps in improving on exist inhibi- tors and/or designing new functional ones. Therefore, this work was aimed at investigating the corrosion inhibi- tion potential of 2-(4-(substituted)arylidene)-1H-indene- 1,3-dione derivatives via density functional theory and molecular dynamic (MD) simulation. The adsorption, orientation and interactions of nine indene-1,3-diones corrosion inhibitors on Fe (110) surface in hydrochloric acid solution were studied using MD simulation. and MEP surface analyses Recently, DFT has become a powerful tool in predict- ing effective organic corrosion inhibitors [40, 41]. Some quantum chemical parameters of a compound have been established to relate the corrosion inhibi- tion efficiency with the molecular properties [42]. These quantum chemical parameters were used to correlate the corrosion inhibition potentials of nine 2-arylidene-indan-1,3-diones with the structural vari- ations. The results of quantum chemical calculations of 2-arylidene-indan-1,3-diones are summarized in Table 1. (10) f+ k = qk(N + 1) −qk(N) nucleophilic attack (11) f− k =  qk(N) −qk(N −1)  electrophilic attack (12) fk(r) = f+ k −f− k (10) f+ k = qk(N + 1) −qk(N) nucleophilic attack (10) (11) f− k =  qk(N) −qk(N −1)  electrophilic attack The optimized structures of the corrosion inhibitors are shown in Fig. 2a. The HOMO and LUMO surfaces are shown in Fig. 2b and c, respectively. The molecular elec- trostatic potential (MEP) maps of the corrosion inhibi- tors are shown in Fig. 3. The MEP surface represents the local regions on the organic systems that are suscepti- ble to nucleophilic attack (blue region) and electrophilic attack (red, orange, yellow and green region) [43]. ( ) (12) fk(r) = f+ k −f− k (12) 2  Methods 2.1  Density functional theory (DFT) studies 2.1  Density functional theory (DFT) studies DFT is a computational tool used to predict the elec- tronic structure of a compound. The optimization of 2-arylidene-indan-1,3-dione derivatives (Fig. 1) was done by employing DFT technique with B3LYP functional and 6-31G(d) basis set after initially running the conformal distribution to obtain the most stable conformal (lowest energy conformal) using molecular mechanics force field (MMFF) [32]. This level of theory has been described to be consistent with experimental findings [29]. The fron- tier molecular orbital (FMO) energies such as highest occupied molecular orbital ­(EHOMO), lowest unoccupied molecular orbital ­(ELUMO), energy gap, ∆Egap (Eq. 1), electron affinity, A (Eq. 2), ionization potential, I (Eq. 3), electronegativity, χ (Eq. 4), chemical potential, CP (Eq. 4), chemical hardness, η (Eq. 5), softness, σ (Eq. 6), global electrophilicity index, ω (Eq. 7) and the back-donation (Eq. 8) were calculated [33–35]. All quantum chemical calculations were carried out using the Spartan 14 soft- ware program. Fig. 1  Structures of the 2-arylidene-indan-1,3-dione derivatives (Compounds 1 to 9) Structures of the 2-arylidene-indan-1,3-dione derivatives (Compounds 1 to 9) Page 4 of 14 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 (1) (2) (3) (4) (5) (6) (7) in the BIOVIA Materials Studio 2017. The simulation was carried out in a box of dimension (24.82 Å × 24.82 Å × 38.11 Å) with periodic boundary conditions con- sisting of six layers of cleaved iron Fe (110) expanded into a 10 × 10 supercell. The Fe (110) plane is consid- ered the most stable iron surface because it is densely packed [38]. A 30 Å vacuum layer was directly placed above the iron surface and filled with each inhibitor compound and 290 ­H2O, 10 ­H3O+ and 10 ­Cl- molecules to replicate a real corrosive medium. The Condensed- phase Optimized Molecular Potentials for Atomis- tic Simulation Studies (COMPASS) force field was employed to perform geometry optimization on all component structures in the box. Molecular dynamic simulations were performed in a Nose-Hoover thermo- stat using NVT canonical ensemble at 298 K at a time step of 1 fs and a simulation time of 300 ps. The energy of interaction, Einteraction of the compounds with the metal surface was obtained using Eq. 3.1  Quantum chemical calculations, HOMO, LUMO and MEP surface analyses The Fukui analysis plays an important role in predicting the site of nucleophilic and electrophilic attack [37]. The partial atomic charge is defined by qk(N+1) when a mol- ecule accepts electron and by qk(N-1) when it loses an electron. For a neutral molecule, the charge on each atom is defined by qk(N). The nucleophilic, electrophilic and dual Fukui parameters were calculated using Yao’s dual descriptors as shown in Eqs. 10, 11 and 12, respectively. 2  Methods 13: Egap = ELUMO −EHOMO (1) Egap = ELUMO −EHOMO (2) A = −ELUMO (3) I = −EHOMO (4) χ = −Cp = I + A 2 (5) η = I −A 2 (6) σ = 1 η (7) ω = χ2 2η (8) Eback−donation = −η 4 (7) (13) Einteraction = Etotal −  EFe+solution + Einhibitor (13) (8) Eback−donation = −η 4 (8) where Etotal is the single point energy of all components in the simulation box, EFe+solution is the energy of iron (110) surface and the corrosive medium and Einhibitor is the single point energy of the inhibitor. The binding energy (Ebinding) is given by Eq. 14 [39]: Furthermore, the charge transfer between the corro- sion inhibitor and the metal is driven by the difference in their electronegativities (and chemical potentials) when they are in contact until a balance in chemical potential is attained [36]. The number of electrons transferred (ΔN) is calculated using Eq. 9. (14) Ebinding = −Einteraction (14) (9) N = χFe −χinh 2(ηFe + ηinh) (9) 3.2  Fukui indicesh The Fukui indices and the Mulliken charge distribution describe the local reactive sites of an organic system that are susceptible to either nucleophilic or electro- philic attacks. Tables 2, 3, 4, 5, 6, 7, 8, 9 and 10 show selected Fukui indices ­(fk +, ­fk − and ∆fk) and Mulliken charges for the anionic, neutral and cationic forms of the corrosion inhibitors 1 to 9. The Mulliken charges depict asymmetric charge distribution of the atoms. Large values of the Fukui indices ­(fk + and ­fk −) indicate molecular softness and chemically reactive sites [44]. The condensed Fukui dual descriptor (∆fk) is a more explicit and accurate parameter than ­fk + and ­fk − for describing local reactivities [45, 46]. Atoms with posi- tive ∆fk are susceptible to electrophilic attack while those with negative ∆fk are susceptible to nucleophilic attack [47]. Molecular dynamics simulations were performed on each compound to explore the stable configurations of the different inhibitor molecules and the interaction modes between each of them and the Fe (110) surface [48]. The method tends to imitate the real-life corrosion process of mild steel in a corrosive environment and also explore the processes relating to the corrosion inhibition behav- ior of the molecules on the metal [49]. The interaction and the binding energies of the inhibitors on Fe (110) sur- face in hydrochloric acid solution are reported in Table 1. The top and side view of the Fe (110)-inhibitor system in ­HCl(aq) medium is shown in Fig. 4. 2.2  Molecular dynamics simulation Molecular dynamics (MD) simulations were performed on all compounds under consideration to probe their interactions with iron surface using the Forcite module Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 5 of 14 Table 1  The calculated quantum chemical parameters with DFT/B3LYP/6-31G(d) method Compounds EHOMO (eV) ELUMO (eV) ΔEgap (eV) I (eV) A (eV) χ (eV) η (eV) σ (eV-1) ω (eV) ΔN ΔEback-donation 1 −6.52 −2.55 3.97 6.52 2.55 4.54 1.99 0.503 5.180 0.620 −0.4962 2 −5.37 −2.09 3.28 5.37 2.09 3.73 1.64 0.609 4.241 0.996 −0.4100 3 −5.99 −2.35 3.64 5.99 2.35 4.17 1.82 0.549 4.777 0.777 −0.4550 4 −6.00 −2.36 3.64 6.00 2.36 4.18 1.82 0.549 4.800 0.774 −0.4550 5 −5.81 −2.35 3.46 5.81 2.35 4.08 1.73 0.578 4.811 0.843 −0.4325 6 −5.94 −2.61 3.33 5.94 2.61 4.28 1.67 0.600 5.488 0.818 −0.4162 7 −5.42 −2.27 3.15 5.42 2.27 3.85 1.58 0.634 4.693 1.001 −0.3937 8 −5.86 −2.28 3.58 5.86 2.28 4.07 1.79 0.559 4.620 0.826 −0.4475 9 −5.27 −2.29 2.98 5.27 2.29 3.78 1.49 0.671 4.791 1.080 −0.3725 Fig. 2  a Optimized structures, b HOMO and c LUMO surface maps of compounds 1 to 9 Table 1  The calculated quantum chemical parameters with DFT/B3LYP/6-31G(d) method Compounds EHOMO (eV) ELUMO (eV) ΔEgap (eV) I (eV) A (eV) χ (eV) η (eV) σ (eV-1) ω (eV) ΔN ΔEback-donation 1 −6.52 −2.55 3.97 6.52 2.55 4.54 1.99 0.503 5.180 0.620 −0.4962 2 −5.37 −2.09 3.28 5.37 2.09 3.73 1.64 0.609 4.241 0.996 −0.4100 3 −5.99 −2.35 3.64 5.99 2.35 4.17 1.82 0.549 4.777 0.777 −0.4550 4 −6.00 −2.36 3.64 6.00 2.36 4.18 1.82 0.549 4.800 0.774 −0.4550 5 −5.81 −2.35 3.46 5.81 2.35 4.08 1.73 0.578 4.811 0.843 −0.4325 6 −5.94 −2.61 3.33 5.94 2.61 4.28 1.67 0.600 5.488 0.818 −0.4162 7 −5.42 −2.27 3.15 5.42 2.27 3.85 1.58 0.634 4.693 1.001 −0.3937 8 −5.86 −2.28 3.58 5.86 2.28 4.07 1.79 0.559 4.620 0.826 −0.4475 9 −5.27 −2.29 2.98 5.27 2.29 3.78 1.49 0.671 4.791 1.080 −0.3725 Fig. 2  a Optimized structures, b HOMO and c LUMO surface maps of compounds 1 to 9 Table 1  The calculated quantum chemical parameters with DFT/B3LYP/6-31G(d) method Fig. 2  a Optimized structures, b HOMO and c LUMO surface maps of compounds 1 to 9 Fig. 2  a Optimized structures, b HOMO and c LUMO surface maps of compounds 1 to 9 4.1  Discussion on DFT calculationsh 4.1  Discussion on DFT calculationsh The energies of HOMO and LUMO are crucial quan- tum chemical parameters that describe the reactivity Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 6 of 14 Fig. 3  Molecular electrostatic potential (MEP) map of Compounds 1 to 9 Fig. 3  Molecular electrostatic potential (MEP) map of Compounds 1 to 9 Fig. 4.1  Discussion on DFT calculationsh Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 7 of 14 Table 4  Selected Mulliken atomic charges and Fukui functions for Compound 3 Comp 3 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.130 −0.118 −0.011 −0.012 0.001 C10 −0.319 −0.249 −0.242 −0.07 −0.007 −0.063 O1 −0.568 −0.490 −0.439 −0.078 −0.051 −0.027 O2 −0.574 −0.495 −0.463 −0.079 −0.032 −0.047 C11 +0.194 +0.166 +0.177 0.028 −0.011 0.039 C12 +0.343 +0.390 +0.420 −0.047 −0.03 −0.017 C14 −0.226 −0.187 −0.148 −0.039 −0.039 0 O3 −0.523 −0.496 −0.427 −0.027 −0.069 0.042 C17 −0.193 −0.222 −0.259 0.029 0.037 −0.008 Table 5  Selected Mulliken atomic charges and Fukui functions for Compound 4 Comp 4 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.140 −0.130 −0.121 −0.01 −0.009 −0.001 C10 −0.318 −0.250 −0.249 −0.068 −0.001 −0.067 O1 −0.563 −0.490 −0.448 −0.073 −0.042 −0.031 O2 −0.567 −0.493 −0.471 −0.074 −0.022 −0.052 C11 +0.197 +0.169 +0.179 0.028 −0.01 0.038 O3 −0.588 −0.569 −0.487 −0.019 −0.082 0.063 C17 +0.362 +0.321 +0.333 0.041 −0.012 0.053 C18 +0.179 +0.186 +0.188 −0.007 −0.002 −0.005 C22 −0.194 −0.189 −0.174 −0.005 −0.015 0.01 C23 −0.528 −0.532 −0.540 0.004 0.008 −0.004 Table 6  Selected Mulliken atomic charges and Fukui functions for Compound 5 Comp 5 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.130 −0.119 −0.011 −0.011 0 C10 −0.319 −0.249 −0.249 −0.07 0 −0.07 O1 −0.566 −0.488 −0.443 −0.078 −0.045 −0.033 O2 −0.574 −0.493 −0.471 −0.081 −0.022 −0.059 C11 +0.193 +0.168 +0.181 0.025 −0.013 0.038 C15 +0.294 +0.306 +0.346 −0.012 −0.04 0.028 C16 −0.212 −0.209 −0.186 −0.003 −0.023 0.02 O3 −0.560 −0.541 −0.488 −0.019 −0.053 0.034 C17 −0.193 −0.219 −0.255 0.026 0.036 −0.01 O4 −0.683 −0.652 −0.567 −0.031 −0.085 0.054 Table 4  Selected Mulliken atomic charges and Fukui functions for Compound 3 Table 5  Selected Mulliken atomic charges and Fukui functions for Compound 4 Comp 4 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.140 −0.130 −0.121 −0.01 −0.009 −0.001 C10 −0.318 −0.250 −0.249 −0.068 −0.001 −0.067 O1 −0.563 −0.490 −0.448 −0.073 −0.042 −0.031 O2 −0.567 −0.493 −0.471 −0.074 −0.022 −0.052 C11 +0.197 +0.169 +0.179 0.028 −0.01 0.038 O3 −0.588 −0.569 −0.487 −0.019 −0.082 0.063 C17 +0.362 +0.321 +0.333 0.041 −0.012 0.053 C18 +0.179 +0.186 +0.188 −0.007 −0.002 −0.005 C22 −0.194 −0.189 −0.174 −0.005 −0.015 0.01 C23 −0.528 −0.532 −0.540 0.004 0.008 −0.004 Table 6  Selected Mulliken atomic charges and Fukui functions for Compound 5 Comp 5 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.130 −0.119 −0.011 −0.011 0 C10 −0.319 −0.249 −0.249 −0.07 0 −0.07 O1 −0.566 −0.488 −0.443 −0.078 −0.045 −0.033 O2 −0.574 −0.493 −0.471 −0.081 −0.022 −0.059 C11 +0.193 +0.168 +0.181 0.025 −0.013 0.038 C15 +0.294 +0.306 +0.346 −0.012 −0.04 0.028 C16 −0.212 −0.209 −0.186 −0.003 −0.023 0.02 O3 −0.560 −0.541 −0.488 −0.019 −0.053 0.034 C17 −0.193 −0.219 −0.255 0.026 0.036 −0.01 O4 −0.683 −0.652 −0.567 −0.031 −0.085 0.054 and kinetic stability of a material [50–52]. 4.1  Discussion on DFT calculationsh 3  Molecular electrostatic potential (MEP) map of Compounds 1 to 9 Table 2  Selected Mulliken atomic charges and Fukui functions for Compound 1 Comp 1 qk (N + 1) qk (N ) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.129 −0.115 −0.012 −0.014 0.002 C4 +0.065 +0.050 +0.058 0.015 −0.008 0.023 C9 +0.304 +0.362 +0.386 −0.058 −0.024 −0.034 C10 −0.316 −0.248 −0.222 −0.068 −0.026 −0.042 O1 −0.566 −0.485 −0.422 −0.081 −0.063 −0.018 O2 −0.569 −0.487 −0.441 −0.082 −0.046 −0.036 C11 +0.186 +0.164 +0.171 0.022 −0.007 0.029 Table 3  Selected Mulliken atomic charges and Fukui functions for Compound 2 Comp 2 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.131 −0.121 −0.01 −0.01 0 C7 +0.289 +0.329 +0.343 −0.04 −0.014 −0.026 C10 −0.321 −0.253 −0.254 −0.068 0.001 −0.069 O1 −0.569 −0.497 −0.448 −0.072 −0.049 −0.023 O2 −0.574 −0.500 −0.470 −0.074 −0.03 −0.044 C11 +0.198 +0.170 +0.184 0.028 −0.014 0.042 C12 +0.328 +0.383 +0.389 −0.055 −0.006 −0.049 N1 −0.464 −0.470 −0.427 0.006 −0.043 0.049 C17 −0.293 −0.316 −0.350 0.023 0.034 −0.011 C18 −0.295 −0.317 −0.350 0.022 0.033 −0.011 Table 2  Selected Mulliken atomic charges and Fukui functions for Compound 1 Comp 1 qk (N + 1) qk (N ) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.129 −0.115 −0.012 −0.014 0.002 C4 +0.065 +0.050 +0.058 0.015 −0.008 0.023 C9 +0.304 +0.362 +0.386 −0.058 −0.024 −0.034 C10 −0.316 −0.248 −0.222 −0.068 −0.026 −0.042 O1 −0.566 −0.485 −0.422 −0.081 −0.063 −0.018 O2 −0.569 −0.487 −0.441 −0.082 −0.046 −0.036 C11 +0.186 +0.164 +0.171 0.022 −0.007 0.029 Table 2  Selected Mulliken atomic charges and Fukui functions for Compound 1 Table 3  Selected Mulliken atomic charges and Fukui functions for Compound 2 Comp 2 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.131 −0.121 −0.01 −0.01 0 C7 +0.289 +0.329 +0.343 −0.04 −0.014 −0.026 C10 −0.321 −0.253 −0.254 −0.068 0.001 −0.069 O1 −0.569 −0.497 −0.448 −0.072 −0.049 −0.023 O2 −0.574 −0.500 −0.470 −0.074 −0.03 −0.044 C11 +0.198 +0.170 +0.184 0.028 −0.014 0.042 C12 +0.328 +0.383 +0.389 −0.055 −0.006 −0.049 N1 −0.464 −0.470 −0.427 0.006 −0.043 0.049 C17 −0.293 −0.316 −0.350 0.023 0.034 −0.011 C18 −0.295 −0.317 −0.350 0.022 0.033 −0.011 Table 3  Selected Mulliken atomic charges and Fukui functions for Compound 2 Oyeneyin et al. 4.1  Discussion on DFT calculationsh Also, compound 2 presents well stabilized ­ELUMO (−2.09 eV) with good potential of back-bond C10 −0.320 −0.256 −0.258 −0.064 0.002 −0.066 O1 −0.563 −0.490 −0.449 −0.073 −0.041 −0.032 O2 −0.562 −0.490 −0.462 −0.072 −0.028 −0.044 C11 +0.130 +0.111 +0.137 0.019 −0.026 0.045 C12 −0.227 −0.198 −0.153 −0.029 −0.045 0.016 C13 −0.142 −0.143 −0.128 0.001 −0.015 0.016 C14 −0.221 −0.176 −0.141 −0.045 −0.035 −0.01 C15 +0.141 +0.128 +0.128 0.013 0 0.013 Table 8  Selected Mulliken atomic charges and Fukui functions for Compound 7 Comp 7 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.131 −0.122 −0.01 −0.009 −0.001 C8 +0.032 +0.035 +0.063 −0.003 −0.028 0.025 C10 −0.324 −0.260 −0.265 −0.064 0.005 −0.069 O1 −0.564 −0.496 −0.456 −0.068 −0.04 −0.028 O2 −0.569 −0.501 −0.471 −0.068 −0.03 −0.038 C11 +0.144 +0.118 +0.144 0.026 −0.026 0.052 C12 −0.250 −0.210 −0.175 −0.04 −0.035 −0.005 C13 −0.212 −0.219 −0.185 0.007 −0.034 0.041 C22 −0.280 −0.282 −0.252 0.002 −0.03 0.032 C17 −0.205 −0.183 −0.156 −0.022 −0.027 0.005 O3 −0.514 −0.504 −0.470 −0.01 −0.034 0.024 O4 −0.196 −0.502 −0.468 0.306 −0.034 0.34 C21 −0.194 −0.221 −0.250 0.027 0.029 −0.002 Table 9  Selected Mulliken atomic charges and Fukui functions for Compound 8 Comp 8 qk (N + 1) qk (N) qk (N - 1) fk + fk − ∆fk C1 −0.141 −0.130 −0.119 −0.011 −0.011 0 C8 +0.022 +0.020 +0.056 0.002 −0.036 0.038 C10 −0.319 −0.249 −0.245 −0.07 −0.004 −0.066 O1 −0.566 −0.492 −0.442 −0.074 −0.05 −0.024 O2 −0.569 −0.496 −0.466 −0.073 −0.03 −0.043 C11 +0.196 +0.164 +0.176 0.032 −0.012 0.044 O3 −0.577 −0.548 −0.475 −0.029 −0.073 0.044 C17 +0.333 +0.308 +0.260 0.025 0.048 −0.023 Table 8  Selected Mulliken atomic charges and Fukui functions for Compound 7 Comp 7 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.131 −0.122 −0.01 −0.009 −0.001 C8 +0.032 +0.035 +0.063 −0.003 −0.028 0.025 C10 −0.324 −0.260 −0.265 −0.064 0.005 −0.069 O1 −0.564 −0.496 −0.456 −0.068 −0.04 −0.028 O2 −0.569 −0.501 −0.471 −0.068 −0.03 −0.038 C11 +0.144 +0.118 +0.144 0.026 −0.026 0.052 C12 −0.250 −0.210 −0.175 −0.04 −0.035 −0.005 C13 −0.212 −0.219 −0.185 0.007 −0.034 0.041 C22 −0.280 −0.282 −0.252 0.002 −0.03 0.032 C17 −0.205 −0.183 −0.156 −0.022 −0.027 0.005 O3 −0.514 −0.504 −0.470 −0.01 −0.034 0.024 O4 −0.196 −0.502 −0.468 0.306 −0.034 0.34 C21 −0.194 −0.221 −0.250 0.027 0.029 −0.002 Table 9  Selected Mulliken atomic charges and Fukui functions for Compound 8 Comp 8 qk (N + 1) qk (N) qk (N - 1) fk + fk − ∆fk C1 −0.141 −0.130 −0.119 −0.011 −0.011 0 C8 +0.022 +0.020 +0.056 0.002 −0.036 0.038 C10 −0.319 −0.249 −0.245 −0.07 −0.004 −0.066 O1 −0.566 −0.492 −0.442 −0.074 −0.05 −0.024 O2 −0.569 −0.496 −0.466 −0.073 −0.03 −0.043 C11 +0.196 +0.164 +0.176 0.032 −0.012 0.044 O3 −0.577 −0.548 −0.475 −0.029 −0.073 0.044 C17 +0.333 +0.308 +0.260 0.025 0.048 −0.023 compound 2 can be related to the presence of an effec- tive electron-donating amine group. 4.1  Discussion on DFT calculationsh HOMO energy value of a molecule shows the tendency of the molecule to donate electron to low-lying vacant orbit- als of a metal. Therefore, the adsorption process of a molecule on a metal surface is facilitated and increased with increase in HOMO energy ­(EHOMO). Hence, higher HOMO energy of a molecule enhances the corrosion inhibition potential of the molecule. On the other hand, LUMO energy ­(ELUMO) value of a molecule indicate the tendency of a molecule to accept electron from metal atom. Decrease in ­(ELUMO) shows the ability of a mol- ecule to accept electron from a metal atom. Hence, lower LUMO energy increases the corrosion inhibition efficiency of a molecule. inhibition potential of the molecule. On the other hand, LUMO energy ­(ELUMO) value of a molecule indicate the tendency of a molecule to accept electron from metal atom. Decrease in ­(ELUMO) shows the ability of a mol- ecule to accept electron from a metal atom. Hence, lower LUMO energy increases the corrosion inhibition efficiency of a molecule. inhibition potential of the molecule. On the other hand, LUMO energy ­(ELUMO) value of a molecule indicate the tendency of a molecule to accept electron from metal atom. Decrease in ­(ELUMO) shows the ability of a mol- ecule to accept electron from a metal atom. Hence, lower LUMO energy increases the corrosion inhibition efficiency of a molecule. Oyeneyin et al. The ­EHOMO and ­ELUMO calculated for the compounds are shown in Table 1. From the results, the best com- pounds according to the ­EHOMO are compounds 2, 7 and 9 (with ­EHOMO values of −5.37, −5.42 and −5.27 eV, respectively). The ­EHOMO follows the order: 9 > 2 > 7 > 5 > 8 > 6 > 3 > 4 >1. High ­EHOMO of compound 9 and 4.1  Discussion on DFT calculationsh This indicates that compounds 2, 7 and 9 possess high tendency of donat- ing electrons to the low-lying unfilled d-orbital of the metal atom. 4.1  Discussion on DFT calculationsh Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 8 of 14 Table 7  Selected Mulliken atomic charges and Fukui functions for Compound 6 Comp 6 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.140 −0.130 −0.119 −0.01 −0.011 0.001 C10 −0.320 −0.256 −0.258 −0.064 0.002 −0.066 O1 −0.563 −0.490 −0.449 −0.073 −0.041 −0.032 O2 −0.562 −0.490 −0.462 −0.072 −0.028 −0.044 C11 +0.130 +0.111 +0.137 0.019 −0.026 0.045 C12 −0.227 −0.198 −0.153 −0.029 −0.045 0.016 C13 −0.142 −0.143 −0.128 0.001 −0.015 0.016 C14 −0.221 −0.176 −0.141 −0.045 −0.035 −0.01 C15 +0.141 +0.128 +0.128 0.013 0 0.013 Table 8  Selected Mulliken atomic charges and Fukui functions for Compound 7 Comp 7 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.141 −0.131 −0.122 −0.01 −0.009 −0.001 C8 +0.032 +0.035 +0.063 −0.003 −0.028 0.025 C10 −0.324 −0.260 −0.265 −0.064 0.005 −0.069 O1 −0.564 −0.496 −0.456 −0.068 −0.04 −0.028 O2 −0.569 −0.501 −0.471 −0.068 −0.03 −0.038 C11 +0.144 +0.118 +0.144 0.026 −0.026 0.052 C12 −0.250 −0.210 −0.175 −0.04 −0.035 −0.005 C13 −0.212 −0.219 −0.185 0.007 −0.034 0.041 C22 −0.280 −0.282 −0.252 0.002 −0.03 0.032 C17 −0.205 −0.183 −0.156 −0.022 −0.027 0.005 O3 −0.514 −0.504 −0.470 −0.01 −0.034 0.024 O4 −0.196 −0.502 −0.468 0.306 −0.034 0.34 C21 −0.194 −0.221 −0.250 0.027 0.029 −0.002 Table 9  Selected Mulliken atomic charges and Fukui functions for Compound 8 Comp 8 qk (N + 1) qk (N) qk (N - 1) fk + fk − ∆fk C1 −0.141 −0.130 −0.119 −0.011 −0.011 0 C8 +0.022 +0.020 +0.056 0.002 −0.036 0.038 C10 −0.319 −0.249 −0.245 −0.07 −0.004 −0.066 O1 −0.566 −0.492 −0.442 −0.074 −0.05 −0.024 O2 −0.569 −0.496 −0.466 −0.073 −0.03 −0.043 C11 +0 196 +0 164 +0 176 0 032 0 012 0 044 Table 7  Selected Mulliken atomic charges and Fukui functions for Compound 6 The ­EHOMO and ­ELUMO calculated for the compounds are shown in Table 1. From the results, the best com- pounds according to the ­EHOMO are compounds 2, 7 and 9 (with ­EHOMO values of −5.37, −5.42 and −5.27 eV, respectively). The ­EHOMO follows the order: 9 > 2 > 7 > 5 > 8 > 6 > 3 > 4 >1. High ­EHOMO of compound 9 and compound 2 can be related to the presence of an effec- tive electron-donating amine group. 4.1  Discussion on DFT calculationsh This indicates that compounds 2, 7 and 9 possess high tendency of donat- ing electrons to the low-lying unfilled d-orbital of the metal atom. Also, compound 2 presents well stabilized ­ELUMO (−2.09 eV) with good potential of back-bond formation with the metal. The ­EHOMO and ­ELUMO calculated for the compounds are shown in Table 1. From the results, the best com- pounds according to the ­EHOMO are compounds 2, 7 and 9 (with ­EHOMO values of −5.37, −5.42 and −5.27 eV, respectively). The ­EHOMO follows the order: 9 > 2 > 7 > 5 > 8 > 6 > 3 > 4 >1. High ­EHOMO of compound 9 and Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 9 of 14 Table 10  Selected Mulliken atomic charges and Fukui functions for Compound 9 Comp 9 qk (N + 1) qk (N) qk (N – 1) fk + fk − ∆fk C1 −0.140 −0.130 −0.124 −0.01 −0.006 −0.004 C10 −0.315 −0.253 −0.257 −0.062 0.004 −0.066 O1 −0.561 −0.493 −0.457 −0.068 −0.036 −0.032 O2 −0.563 −0.496 −0.477 −0.067 −0.019 −0.048 C11 +0.196 +0.172 +0.184 0.024 −0.012 0.036 C13 −0.163 −0.189 −0.162 0.026 −0.027 0.053 C14 +0.204 +0.323 +0.287 −0.119 0.036 −0.155 C15 −0.158 −0.184 −0.157 0.026 −0.027 0.053 C16 −0.211 −0.190 −0.183 −0.021 −0.007 −0.014 N1 −0.653 −0.648 −0.573 −0.005 −0.075 0.07 C17 +0.307 +0.260 +0.249 0.047 0.011 0.036 C23 +0.308 +0.258 +0.244 0.05 0.014 0.036 Table 10  Selected Mulliken atomic charges and Fukui functions for Compound 9 The ionization potential (I) and electron affinity (A) depict the electron-donating and electron-accepting ability of the inhibitors, respectively [53]. The ease of electron donation of a corrosion inhibitor is enhanced as the ionization potential decreases. From Table 1, the corrosion inhibitors display good ionization potential which decreases in the order: 1 > 4 > 3 > 6 > 8 > 5 > 7 > 2 > 9 and follows a reverse trend with ­EHOMO. It could be observed that para-substitution of diphenylamine (compound 9) and dimethylamine (compound 2) on the parent structure (compound 1) significantly lowers the ionization potential possibly by destabilizing the highest filled molecular orbital. On the other hand, the electron- accepting tendency of an inhibitor increases as its elec- tron affinity (A) increases. The A parameter is relatively high in compounds with bare phenyl and naphthalenyl substitution at 2-indene-1,3-dione position especially in compounds 1 and 6. 4.1  Discussion on DFT calculationsh Substitution of electron-donating groups (such as amine and methoxy moieties) on the phe- nyl and naphthalenyl rings decreased the electron affinity of the corrosion inhibitors possibly by destabilizing their LUMO level relative to the unsubstituted derivatives. substituents. Similar observation has been reported in literature [54]. Interestingly, para-substitution on the benzylidene moiety of 1H-indene-1,3-dione lowers the energy gap, global hardness and improves the ionization potential, global softness, the number of electrons trans- ferred and the back-donation properties of the organic inhibitors relative to the parent compound 1. The electronegativity (χ) specifies the direction of flow of electron between the metal surface and the inhibi- tor till a balance in chemical potential is attained. On adsorption of inhibitor on the metal surface, especially iron (with electronegativity of 7 eV), electrons should be transferred from the species with lower electronegative to the one with higher electronegativity. From Table 1, all the inhibitors display less electronegativity (χ = 3.78–4.54 eV) than that of iron which suggests that they are capable of electron transfer to the metal surface. The χ follows the order: 2 < 9 < 7 < 8 < 5 < 3 < 4 < 6 < 1. The farther the χ of the inhibitor is from that of the metal, the slower the attainment of equalization of chemical potential, the better the reactivity and the greater the corrosion inhi- bition efficiency. This suggests that compounds 2, 9 and 7 will present good kinetic interaction with the iron sur- face. Also, the number of transferred electron (ΔN) is an effective quantum chemical descriptor of metal-inhibitor interactions. High ΔN suggests good interaction between the corrosion inhibitor and the iron surface. The ΔN fol- lows an opposite trend with the energy gap, suggesting better corrosion inhibition potentials of the substituted organic systems than the parent structure (compound 1). Compound 9 displays largest ΔN indicating that it has a greater potential of releasing electron into low-lying vacant d-orbitals of the metal. Electrophilicity (ω) is a quantum chemical parameter that measures stabiliza- tion of energy as a result of maximum electron transfer The energy gap (ΔEgap) and the global hardness (η) of the compounds follow the same trend. Compounds with low ΔEgap and η possess low global hardness and exhibit good anti-corrosion properties. 4.2  HOMO, LUMO and molecular electrostatic potential (MEP) surface analysesh The HOMO and LUMO maps of compounds 1 to 9 show the distribution of the frontier molecular orbitals over the structures (Fig. 2b and c). The HOMO and LUMO of compound 1 are delocalized entirely over the struc- ture indicating that frontier orbitals available for elec- tron donation and reception are not essentially localized. Para-substitutions on the benzylidene moiety result in HOMO re-distribution majorly over the aryl group and partly over the indene carbon-2 suggesting the region of the compounds available for electron donation into the low-lying vacant orbitals of iron. On the other hand, the LUMO is delocalized almost over the entire molecules except on the methyl (compounds 2,3,5,7), phenyl (com- pounds 4 and 9) and tert-butyl (compound 8) moieties. These regions depicting the least unfilled molecular orbitals of the inhibitors are possible sites available for back-bonding with the metal surface. The molecular elec- trostatic potential surface analyses reveal the phenyl ring, and the oxygen atoms of the phenoxy, hydroxyl, methoxy and indene carbonyl moieties as local nucleophilic sites on the inhibitors available for adsorption on the metal surface (Fig. 3). The regions with positive electrostatic potential are the phenyl hydrogen, methyl, tert-butyl and dimethylamine moieties. 4.1  Discussion on DFT calculationsh The ω follows the trend: 2 < 8 < 7 < 3 < 9 < 4 < 5 < 1 < 6. Corrosion inhibitors with low electrophilicity exhibit good electron-releasing potential for effective interaction with and stabilization on a metal surface [55]. The back-donation character of the organic inhibitor signifies the synergistic tendency of the inhibitors to accept π-electron from the d-orbital of the iron into its vacant antibonding orbital. This property increases in the order: 1 < 3 < 4 < 8 < 5 < 6 < 2 < 7 < 9, indicating that compounds 9, 7 and 2 possess good back- bonding ability with metal surface. p y The most positive ∆fk for compound 2 is on amine nitrogen, N1, while the carbon of the methyl substituent on the amine displays maximum negative ∆fk (Table 3). Although the methoxy carbon (C17) of compounds 3 and 5 shows highly positive ­fk + and ­fk −, the positive ∆fk are localized on the heteroatoms, O3 and O4, respec- tively. Likewise, the maximum positive ∆fk values of 0.045 and 0.052 are observed for the naphthalenyl car- bon, C11, of compounds 6 and 7, respectively (Tables 7 and 8). The dual nucleophilic reactive sites are observed at the phenyl carbon, C11 and tert-butoxide oxygen, O3 (for compound 8), and at the phenyl carbon atoms, C13 and C15 (for compound 9), respectively. In essence, the nucleophilic sites on the corrosion inhibitors are located both on the heteroatoms and the ­sp2-hybridized carbon atoms, while the olefinic carbon bridge majorly behaves as an electrophilic center available for possible back- bonding with the metal surface. Effective metal-inhibitor interactions enhance adsorption of the inhibitor and, consequently, the isolation of the metal from the corrod- ing medium. Succinctly, p-methoxy and p-tolyloxy substituted inhibitors (compounds 3 and 4, respectively) display close chemical reactivity with the reference structure (compound 1), indicating little contribution of these moi- eties to corrosion inhibition potential while amine and dimethylnaphthalenyl moieties improved the chemical reactivity. 4.1  Discussion on DFT calculationsh Compounds 9 and 7 dis- play the lowest energy gap (2.98 and 3.15 eV, respectively) and least global hardness (1.49 and 1.58 eV) which sug- gest their high reactivity and kinetic instability. Also, the chemical reactivity, as deduced from ΔEgap and η, will increase in the order: 9 < 7 < 2 < 6 < 5 < 8 < 4 < 3 < 1. Conversely, global softness follows a reversed trend. This trend supports the high chemical reactivity, global soft- ness (σ), ease of electron donation and effective corrosion inhibition potentials of systems with electron-donating Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Page 10 of 14 Fig. 4  MD simulation graphics of most stable adsorption configurations of Compounds 1 to 9 in acidic (HCl) medium on Fe (110) surface Fig. 4  MD simulation graphics of most stable adsorption configurations of Compounds 1 to 9 in acidic (HCl) medium on Fe (110) surface Page 11 of 14 Page 11 of 14 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 the 1,3-indane-dione and the aryl moieties displays maxi- mum negative Fukui dual function (∆fk = −0.042) indi- cating that it is the preferred electrophilic site. Similarly, the electrophilic center on compounds 2, 3, 4, 6, 7 and 8 was located on olefinic carbon bridge, C10, with highly negative ∆fk as shown in Tables 3, 4, 5, 7, 8, 9, respec- tively. Compounds 5 and 9, however, exhibit active elec- trophilic centers at indenone oxygen O2 and phenyl carbon (C14), respectively as shown in Tables 6 and 10.h between a donor and acceptor. The ω follows the trend: 2 < 8 < 7 < 3 < 9 < 4 < 5 < 1 < 6. Corrosion inhibitors with low electrophilicity exhibit good electron-releasing potential for effective interaction with and stabilization on a metal surface [55]. The back-donation character of the organic inhibitor signifies the synergistic tendency of the inhibitors to accept π-electron from the d-orbital of the iron into its vacant antibonding orbital. This property increases in the order: 1 < 3 < 4 < 8 < 5 < 6 < 2 < 7 < 9, indicating that compounds 9, 7 and 2 possess good back- bonding ability with metal surface. between a donor and acceptor. 4.4  Adsorption and binding energy of iron‑inhibitor system Figure 4 represents the top and side views of the interac- tion of the molecules with the Fe (110) in the presence of the molecules and ions of the corrosive solvent ­(H2O, ­H3O+ and ­Cl−). It is noted that each of the inhibitor mol- ecules positioned itself closer to the surface of the metal and orientating itself in a flat configuration that is par- tially or completely parallel to the surface after equilibra- tion for 300 ps at 298 K. Thus, the flat configuration of the molecules ensures that the metal surface is well cov- ered and protected from the corrosive medium [56]. This is as a result of electron-donation from active centers of the molecules, such as N, O, phenyl rings, to the d-orbit- als of iron [57].h Author contributions OEO performed supervision; conceptualization; data curation; methodology; validation; visualization; software; formal analysis; roles/writing—original draft; and review and editing; NDO contributed to conceptualization; validation; formal analysis; roles/writing—original draft; and review and editing; NI performed data curation and roles/writing—original draft; EBA contributed to data curation; methodology; formal analysis; roles/writing—original draft and review; visualization; and software; AVE performed formal analysis and roles/ writing—original draft. All authors read and approved of the final manuscript. Availability of data and materials Availability of data and materials Not applicable. y Not applicable. Abbreviations DFT: Density functional theory; MD: Molecular dynamics; B3LYP: Becke-3–Lee- Yang-Parr; ELUMO: Lowest unoccupied molecular orbital energy; EHOMO: Highest occupied molecular orbital energy; A: Electron affinity; I: Ionization potential; ΔEgap: Energy gap; η: Global hardness; σ: Global softness; χ: Electronegativ- ity; ω: Electrophilicity; ΔN: Number of transferred electrons; ΔEback-donation: Back-donation; CP: Chemical potential; MEP: Molecular electrostatic potential; MMFF: Molecular mechanics force field; FMO: Frontier molecular orbital; fk +: Nucleophilic Fukui index; fk −: Electrophilic Fukui index; ∆fk(r): Condensed dual Fukui parameter; COMPASS: Condensed-phase Optimized Molecular Potentials for Atomistic Simulation Studies; Etotal: Total energy of substrate- adsorbate system; EFe+solution: Energy of iron (110) surface and the corrosive medium; Einhibitor: Energy of the inhibitor; Ebinding: Binding energy; Einteraction: Interaction energy; compound 1: 2-benzylidene-1H-indene-1,3(2H)-dione; compound 2: 2-(4-(dimethylamino)benzylidene)-1H-indene-1,3(2H)-dione; compound 3: 2-(4-methoxybenzylidene)-1H-indene-1,3(2H)-dione; com- pound 4: 2-(4-(p-tolyloxy)benzylidene)-1H-indene-1,3(2H)-dione; compound 5: 2-(3-hydroxy-4-methoxybenzylidene)-1H-indene-1,3(2H)-dione; compound 6: 2-(naphthalen-1-ylmethylene)-1H-indene-1,3(2H)-dione; compound 7: 2-((4,7-dimethylnaphthalen-1-yl)methylene)-1H-indene-1,3(2H)-dione; com- pound 8: 2-(4-(tert-butoxy)benzylidene)-1H-indene-1,3(2H)-dione; compound 9: 2-(4-(diphenylamino)benzylidene)-1H-indene-1,3(2H)-dione. energy, the more positive the binding energy and the more the molecule easily adsorbs on the steel surface. From Table 11, all the corrosion inhibitors exhibit good binding affinity for the iron (110) surface in the corrod- ing medium. Remarkably strong binding energies were obtained for compounds 2, 4, 7 and 8. However, com- pound 7 displays the largest binding energy compared to other selected molecules, indicating that it adsorbs most easily to the mild steel surface and provides the best protection for the surface against the corrosive chloride medium. This observation may be attributed to good structural planarity, effective back-donation tendency, high electron affinity, chemical softness of the naphtha- lene-based inhibitor and its proximal alignment on the iron surface (Fig. 4). Also, steric hindrance due to the diphenyl rings may hamper the iron-inhibitor interaction between the Fe (110) surface and compound 9, thereby relatively weakening inhibitor’s adsorption potential [58–61]. 4.3  Local reactivity and Fukui descriptors The interaction energy of each of the molecules is reported in Table  11. The negative values of the inter- action energies of the molecules indicate that the interaction between them and the metal substrate is exo- thermic in nature. On the other hand, the binding energy of adsorption to the metal surface is the negative of the interaction energy. The more negative the interaction In compound 1, aromatic carbon, C11, exhibits highest ­fk + and ­fk − indicating that it is reactive site for both elec- trophilic and nucleophilic attacks (Table 2). However, the highly positive ∆fk (0.029) of the carbon atom (C11) indi- cates that it is a preferred site for nucleophilic attack on a metal surface. The olefinic carbon bridge (C10) between Page 12 of 14 Page 12 of 14 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Table 11  Interaction (adsorption) and binding energies of inhibitor molecules with Fe (110) surface in hydrochloric acid medium System Einteraction (kJ/mol) Ebinding (kJ/mol) Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 1 −165.21 165.21 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 2 −207.17 207.17 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 3 −194.64 194.64 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 4 −229.32 229.32 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 5 −202.40 202.40 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 6 −207.15 207.15 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 7 −237.83 237.83 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 8 −208.68 208.68 Fe (1 1 0)/290 ­H2O/10H3O+/10 ­Cl-/Compound 9 −89.93 89.93 the iron surface and the 2-(4-(substituted)arylidene)- 1H-indene-1,3-diones and the effective surface cover- age indicate good anti-corrosion potentials. Compound 7 (2-((4,7-dimethylnaphthalen-1-yl)methylene)-1H-in- dene-1,3(2H)-dione), however, exhibits strongest binding with Fe(110) surface implying its effective adsorption and good anti-corrosion potentials which may be attributed to its good affinity for the iron surface and back-donation tendency. Although 2-(4-(diphenylamino)benzylidene)- 1H-indene-1,3(2H)-dione (compound 9) possesses low energy gap and good reactivity, its effective adsorption on the iron surface is reduced possibly by steric hindrance provided by the diphenyl rings. 5  Conclusions h b Corrosion inhibition potentials of nine derivatives of 1H-indene-1,3-dione have been studied using quantum chemical calculations and molecular dynamics approach. The compounds exhibit low energy gap, global hardness, ionization potential, electrophilicity and high electron affinity, global softness, back-donation and number of transferred electrons suggesting good corrosion mitiga- tion potentials. Compounds 2, 7 and 9 display uniquely low energy gap suggesting their high chemical reactivity and kinetic instability. The electrostatic potential map and the condensed Fukui dual descriptor suggest asym- metrical charge distribution with nucleophilic reac- tive sites located essentially on the heteroatoms and the carbon atoms of phenyl and naphthalenyl moieties. The strong binding energy (89.93–237.83 kJ/mol) between Acknowledgements h ld l k Authors would like to acknowledge Adekunle Ajasin University, Akungba- Akoko for creating an enabling environment to carry out this research work. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. References 1. Iroha NB, Dueke-Eze CU, Fasina TM, Anadebe VC, Guo L (2021) Anticorro- sion activity of two new pyridine derivatives in protecting X70 pipeline steel in oil well acidizing fluid: experimental and quantum chemical stud- ies. J Iran Chem Soc 19:1–16. https://​doi.​org/​10.​1007/​s13738-​021-​02450-2 1. Iroha NB, Dueke-Eze CU, Fasina TM, Anadebe VC, Guo L (2021) Anticorro- sion activity of two new pyridine derivatives in protecting X70 pipeline steel in oil well acidizing fluid: experimental and quantum chemical stud- ies. J Iran Chem Soc 19:1–16. https://​doi.​org/​10.​1007/​s13738-​021-​02450-2 20. Asadi S, Mohammadi Ziarani G (2016) The molecular diversity scope of 1,3-indandione in organic synthesis. Mol Divers 20(1):111–152. https://​ doi.​org/​10.​1007/​s11030-​015-​9589-z 20. Asadi S, Mohammadi Ziarani G (2016) The molecular diversity scope of 1,3-indandione in organic synthesis. Mol Divers 20(1):111–152. https://​ doi.​org/​10.​1007/​s11030-​015-​9589-z 21. Roy S, Piradhi V (2017) Formal [2+1] annulation (spiro-cyclopropanation) reaction between sulfur ylides derived from baylis-hillman bromides and arylidene indane-1,3-dione. ChemistrySelect 2(21):6159–6162. https://​ doi.​org/​10.​1002/​slct.​20170​1221 2. Songbo R, Ying G, Chao K, Song G, Shanhua X, Liqiong Y (2021) Effects of the corrosion pitting parameters on the mechanical properties of corroded steel. Constr Build Mater 272:121941. https://​doi.​org/​10.​1016/j.​ conbu​ildmat.​2020.​121941 2. Songbo R, Ying G, Chao K, Song G, Shanhua X, Liqiong Y (2021) Effects of the corrosion pitting parameters on the mechanical properties of corroded steel. Constr Build Mater 272:121941. https://​doi.​org/​10.​1016/j.​ conbu​ildmat.​2020.​121941 22. Saczewski J, Paluchowska A, Klenc J, Raux E, Barnes S, Sullivan S, Duszynska B, Bojarski, AJ, Strekowski L (2009) Synthesis of 4-substituted 2-(4-methylpiperazino) pyrimidines and quinazoline analogs as serotonin 5-HT2A receptor ligands. J Heterocycl Chem 46:1259–1265. https://​doi.​ org/​10.​1002/​jhet.​236 3. Zolkin AL, Galanskiy SA, Kuzmin AM (2021) Perspectives for use of com- posite and polymer materials in aircraft construction. IOP Conf Ser Mater Sci Eng 1047(1):1–7. https://​doi.​org/​10.​1088/​1757-​899X/​1047/1/​012023 3. Zolkin AL, Galanskiy SA, Kuzmin AM (2021) Perspectives for use of com- posite and polymer materials in aircraft construction. IOP Conf Ser Mater Sci Eng 1047(1):1–7. https://​doi.​org/​10.​1088/​1757-​899X/​1047/1/​012023 g p g 4. El Ibrahim B, Nardeli JV, Guo L (2021) An overview of corrosion. ACS Symp Ser 1403:1–19. https://​doi.​org/​10.​1021/​bk-​2021-​1403.​ch001 g p g 4. El Ibrahim B, Nardeli JV, Guo L (2021) An overview of corrosion. ACS Symp Ser 1403:1–19. https://​doi.​org/​10.​1021/​bk-​2021-​1403.​ch001 23. Pluskota R, Koba M (2018) Indandione and its derivatives—chemi- cal compounds with high biological potential. Mini-Rev Med Chem 18(15):1321–1330. https://​doi.​org/​10.​2174/​13895​57518​66618​03301​ 01809 5. Conradi, (2021) Development of mechanical, corrosion resistance, and antibacterial properties of steels. Materials 14(24):1–2. https://​doi.​org/​10.​ 3390/​ma142​47698 6. Consent for publication Not applicable. Consent for publication Not applicable. Page 13 of 14 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 Received: 19 March 2022 Accepted: 6 October 2022 19. Wössner JS, Esser B (2020) Spiroconjugated donor-σ-acceptor charge- transfer dyes: effect of the Ï-subsystems on the optoelectronic properties. J Org Chem 85(7):5048–5057. https://​doi.​org/​10.​1021/​acs.​joc.​0c005​67 There is no competing interest. There is no competing interest. References Akpoborie J, Fayomi OSI, Inegbenebor AO, Ayoola AA, Dunlami O, Samuel OD, Agboola O (2021) Electrochemical reaction of corrosion and its nega- tive economic impact. IOP Conf Ser Mater Sci Eng 1107(1):1–6. https://​ doi.​org/​10.​1088/​1757-​899X/​1107/1/​012071 24. Durand RJ, Gauthier S, Achelle S, Groizard T, Kahlal S, Saillard J, Barsella A, Le Poul N, Le Guen FR (2018) Push-pull D-π-Ru-π-A chromophores: syn- thesis and electrochemical, photophysical and second-order nonlinear optical properties. Dalt Trans 47(11):3965–3975. https://​doi.​org/​10.​1039/​ c8dt0​0093j g 7. Kolo AM, Yakubu A, Jibrin AA, Ibrahim A (2021) Investigation of corrosion inhibition of Zn metal in 0.1 M ­HNO3 by Terminalia catappa. Pharm Chem J 8(4):13–26 7. Kolo AM, Yakubu A, Jibrin AA, Ibrahim A (2021) Investigation of corrosion inhibition of Zn metal in 0.1 M ­HNO3 by Terminalia catappa. Pharm Chem J 8(4):13–26 25. Alizadeh A, Beiranvand Z, Safaei Z, Khodaei MM, Repo E (2020) Green and fast synthesis of 2-arylidene-indan-1,3-diones using a task-specific ionic liquid. ACS Omega 5(44):28632–28636. https://​doi.​org/​10.​1021/​acsom​ ega.​0c036​45 8. Dahal KP, Timilsena JN, Gautam M, Bhattarai J (2021) Investigation on probabilistic model for corrosion failure level of buried pipelines in Kir- tipur urban areas (Nepal). J Fail Anal Prev 21(3):914–926. https://​doi.​org/​ 10.​1007/​s11668-​021-​01138-2 8. Dahal KP, Timilsena JN, Gautam M, Bhattarai J (2021) Investigation on probabilistic model for corrosion failure level of buried pipelines in Kir- tipur urban areas (Nepal). J Fail Anal Prev 21(3):914–926. https://​doi.​org/​ 10.​1007/​s11668-​021-​01138-2 26. Pluskota R, Jaroch K, Kośliński P, Ziomkowska B, Lewińska A, Kruszewski S, Bojko B, Koba M (2021) Selected drug-likeness properties of 2-arylidene- indan-1,3-dione derivatives—chemical compounds with potential anti- cancer activity. Molecules 26(17):1–19. https://​doi.​org/​10.​3390/​molec​ ules2​61752​56 9. Nduma RC, Fayomi OSI, Nkiko MO, Inegbenebor AO, Udoye NE, Onyisi O, Sanni O. Fayomi J (2021) Review of metal protection techniques and application of drugs as corrosion inhibitors on metals. IOP Conf Ser Mater Sci Eng 1107(1):1–5. https://​doi.​org/​10.​1088/​1757-​899x/​1107/1/​012023 27. Sengupta S, Murmu M, Mandal S, Hirani H, Banerjee P (2021) Competitive corrosion inhibition performance of alkyl/acyl substituted 2-(2-hydroxy- benzylideneamino)phenol protecting mild steel used in adverse acidic medium: a dual approach analysis using FMOs/molecular dynamics simulation corroborated experimental fin. Colloids Surfaces A Phys- icochem Eng Asp. 617:126314. https://​doi.​org/​10.​1016/j.​colsu​rfa.​2021.​ 126314 10. Al-Amiery AA, Kadhim A, Al-Adili A, Tawfiq ZH (2021) Limits and develop- ments in ecofriendly corrosion inhibitors of mild steel: a critical review. Part 1: Coumarins. Int J Corros Scale Inhib 10(4):1355–1384. https://​doi.​ org/​10.​17675/​2305-​6894-​2021-​10-4-1 g 11. Competing interests 16. Sabzi M, Jozani AH, Zeidvandi F, Sadeghi M, Dezfuli SM (2022) Effect of 2-mercaptobenzothiazole concentration on sour-corrosion behavior of API X60 pipeline steel: electrochemical parameters and adsorption mechanism. Int J Miner Metall Mater 29(2):271–282. https://​doi.​org/​10.​ 1007/​s12613-​020-​2156-3 Author details 1 1 Theoretical and Computational Chemistry Unit, Department of Chemical sciences, Adekunle Ajasin University, Akungba‑Akoko, Ondo state, Nigeria. 2 Department of Chemistry, Faculty of Science, University of Ibadan, Oyo State, Nigeria. 3 Department of Physical Sciences, Wesley University, Ondo City, Nigeria. 17. Tanwer S, Shukla SK (2022) Recent advances in the applicability of drugs as corrosion inhibitor on metal surface: a review. Curr Res Green Sustain Chem 5:1–22. https://​doi.​org/​10.​1016/j.​crgsc.​2021.​100227 17. Tanwer S, Shukla SK (2022) Recent advances in the applicability of drugs as corrosion inhibitor on metal surface: a review. Curr Res Green Sustain Chem 5:1–22. https://​doi.​org/​10.​1016/j.​crgsc.​2021.​100227 18. Bogdanov G, Tillotson JP, Timofeeva T (2019) Crystal structures, syntheses, and spectroscopic and electrochemical measurements of two push-pull chromophores: 2-[4-(dimethylamino)benzylidene]-1H-indene-1,3(2H)- dione and (E)-2-{3-[4-(dimethylamino)phenyl]allylidene}-1H-indene- 1,3(2H)-dione. Acta Crystallogr Sect E Crystallogr Commun 75:1595–1599. https://​doi.​org/​10.​1107/​S2056​98901​90132​9X Received: 19 March 2022 Accepted: 6 October 2022 References Kadhim A, Betti N, Shaker LM (2022) Limits and developments in organic inhibitors for corrosion of mild steel : a critical review (Part two: 4-amino- antipyrine). Int J Corros Scale Inhib 11(1):43–63. https://​doi.​org/​10.​17675/​ 2305-​6894-​2022-​11-1-2 g 11. Kadhim A, Betti N, Shaker LM (2022) Limits and developments in organic inhibitors for corrosion of mild steel : a critical review (Part two: 4-amino- antipyrine). Int J Corros Scale Inhib 11(1):43–63. https://​doi.​org/​10.​17675/​ 2305-​6894-​2022-​11-1-2 28. Llache Robledo D, Berner Perez B, Gonzales Guiterrez AG, Flores-Moreno R, Casillas N (2021) DFT as a tool for predicting corrosion inhibition capac- ity. ECS Trans 101(1):277–290. https://​doi.​org/​10.​1149/​10101.​0277e​cst 12. Abdel-Karim AM, El-Shamy AM (2022) A review on green corrosion inhibi- tors for protection of archeological metal artifacts. J Bio- Tribo-Corros 8(2):1–21. https://​doi.​org/​10.​1007/​s40735-​022-​00636-6 12. Abdel-Karim AM, El-Shamy AM (2022) A review on green corrosion inhibi- tors for protection of archeological metal artifacts. J Bio- Tribo-Corros 8(2):1–21. https://​doi.​org/​10.​1007/​s40735-​022-​00636-6 29. Singh A, Ansari KR, Banerjee P, Murmu M, Quraishi MA, Lin Y (2021) Corro- sion inhibition behavior of piperidinium based ionic liquids on Q235 steel in hydrochloric acid solution: experimental, density functional theory and molecular dynamics study. Colloids Surfaces A Physicochem Eng Asp 623(May):1–12. https://​doi.​org/​10.​1016/j.​colsu​rfa.​2021.​126708 13. Hanoon MM, Resen AM, Al-Amiery AA, Kadhum AAH, Takriff MS (2021) Theoretical and experimental studies on the corrosion inhibition potentials of 2-((6-methyl-2-ketoquinolin-3-yl)methylene) hydrazinecar- bothioamide for mild steel in 1 M HCl. Prog Color Color Coat 15(1):21–33. https://​doi.​org/​10.​30509/​PCCC.​2020.​166739.​1095 y p g j 30. Abdelshafi NS, Sadik MA, Shoeib MA, Halim SA (2022) Corrosion inhibition of aluminum in 1 M HCl by novel pyrimidine derivatives, EFM measure- ments, DFT calculations and MD simulation. Arab J Chem 15(1):103459. https://​doi.​org/​10.​1016/j.​arabjc.​2021.​103459 p g 14. Obi-Egbedi NO, Ojo ND (2015) Computational studies of the corrosion inhibition potentials of some derivatives of 1H-imidazo[4,5-F][1,10]phen- anthroline. J Sci Res 14:50–56 31. Tazouti A, Errahmany N, Rbaa M, Galai M, Rouifi Z, Touir R, Zarrouk A, Kaya S, Touhami ME, El Ibrahimi B, Erkan S (2021) Effect of hydrocarbon chain length for acid corrosion inhibition of mild steel by three 8-(n-bromo-R- alkoxy)quinoline derivatives: experimental and theoretical investigations. J Mol Struct 1244:1–19. https://​doi.​org/​10.​1016/j.​molst​ruc.​2021.​130976 15. Shamsa A, Barmatov E, Hughes TL, Hua Y, Neville A, Barker R (2022) Hydrolysis of imidazoline based corrosion inhibitor and effects on inhibition performance of X65 steel in ­CO2 saturated brine. J Pet Sci Eng 208:109235. https://​doi.​org/​10.​1016/j.​petrol.​2021.​109235 Page 14 of 14 Oyeneyin et al. Beni-Suef Univ J Basic Appl Sci (2022) 11:132 32. References Sánchez-Márquez J (2019) Correlations between fukui indices and reactivity descriptors based on Sanderson’s principle. J Phys Chem A 123(40):8571–8582. https://​doi.​org/​10.​1021/​acs.​jpca.​9b055​71 54. Odewole OA, Ibeji CU, Oluwasola HO, Oyeneyin OE, Akpomie KG, Ugwu CM, Ugwu CG, Bakare TE (2021) Synthesis and anti-corrosive potential of Schiff bases derived 4-nitrocinnamaldehyde for mild steel in HCl medium: experimental and DFT studies. J Mol Struct 1223:1–9. https://​ doi.​org/​10.​1016/j.​molst​ruc.​2020.​129214 38. Shahraki M, Dehdab M, Elmi S (2016) Theoretical studies on the cor- rosion inhibition performance of three amine derivatives on carbon steel: Molecular dynamics simulation and density functional theory approaches. J Taiwan Inst Chem Eng 62:313–321. https://​doi.​org/​10.​ 1016/j.​jtice.​2016.​02.​010 55. Lotfy K (2015) Molecular modeling, docking and admet of dimethylthi- ohydantoin derivatives for prostate cancer treatment. J Biophys Chem. 6(4):91–117. https://​doi.​org/​10.​4236/​jbpc.​2015.​64010 39. Chen X, Chen Y, Cui J, Li Y, Liang Y, Cao G (2021) Molecular dynamics simulation and DFT calculation of “green” scale and corrosion inhibitor. Comput Mater Sci 188:110229. https://​doi.​org/​10.​1016/j.​comma​tsci.​2020. 110229 56. Saha SK, Ghosh P, Hens A, Murmu NC, Banerjee P (2015) Density func- tional theory and molecular dynamics simulation study on corrosion inhibition performance of mild steel by mercapto-quinoline Schiff base corrosion inhibitor. Phys E Low Dimens Syst Nanostruct 66:332–341. https://​doi.​org/​10.​1016/j.​physe.​2014.​10.​035 40. Oyeneyin OE, Obadawo BS, Ojo FK, Akerele DD, Akintemi EO, Ejelonu BC, Ipinloju N (2020) Experimental and theoretical study on the corrosion inhibitive potentials of schiff base of aniline and salicyaldehyde on mild steel in 0.5 M HCl. Adv J Chem B 2(4):197–208. https://​doi.​org/​10.​33945/​ SAMI/​AJCB.​2020.4.4 57. Obot IB, Obi-Egbedi NO, Ebenso EE, Afolabi AS, Oguzie EE (2013) Experimental, quantum chemical calculations, and molecular dynamic simulations insight into the corrosion inhibition properties of 2-(6-meth- ylpyridin-2-yl)oxazolo[5,4-f][1,10]phenanthroline on mild steel. Res Chem Intermed 39(5):1927–1948. https://​doi.​org/​10.​1007/​s11164-​012-​0726-3 41. Amoko SJ, Akinyele OF, Oyeneyin OE, Olayanju DS, Aboluwoye CO (2020) Experimental and theoretical investigation of corrosion inhibitive poten- tials of (E)-4-hydroxy-3-[(2,4,6-tribromophenyl)diazenyl]benzaldehyde on mild steel in acidic media. Phys Chem Res 8(3):399–416. https://​doi.​org/​ 10.​22036/​pcr.​2020.​217825.​1725 g 58. Nyijime TA, Iorhuna F (2022) Theoretical Study of 3-(4-hydroxyphenyl)- 1-(4-nitrophenyl)prop-2-en-1-one and 3-(4-hydroxyphenyl)-1-phenyl- prop-2-en-1-one as Corrosion Inhibitors on Mild Steel. Appl J Environ Eng Sci 8(2):177–86 58. Nyijime TA, Iorhuna F (2022) Theoretical Study of 3-(4-hydroxyphenyl)- 1-(4-nitrophenyl)prop-2-en-1-one and 3-(4-hydroxyphenyl)-1-phenyl- prop-2-en-1-one as Corrosion Inhibitors on Mild Steel. Appl J Environ Eng Sci 8(2):177–86 42. Oyeneyin OE, Ojo ND, Ipinloju N, James AC, Agbaffa EB (2022) Investiga- tion of corrosion inhibition potentials of some aminopyridine schiff bases using density functional theory and Monte Carlo simulation. References Benhiba F, Hsissou R, Benzikri Z, Echihi S, El-Blilak J, Boukhris S, Bellaou- chou A, Guenbour A, Oudda H, Warad I, Sebbar NK, Zarrouk A (2021) DFT/electronic scale, MD simulation and evaluation of 6-methyl-2-(p- tolyl)-1,4-dihydroquinoxaline as a potential corrosion inhibition. J Mol Liq 335:1–22. https://​doi.​org/​10.​1016/j.​molliq.​2021.​116539 48. Khadiri A, Ousslim A, Bekkouche K, Warad I, Elidrissi A, Hammouti B, Bentiss F, Bouachrine M, Zarrouk A (2018) 4-(2-(2-(2-(2-(Pyridine-4-yl) ethylthio)ethoxy)ethylthio)ethyl)pyridine as new corrosion inhibitor for mild steel in 1.0 M HCl Solution: experimental and theoretical studies. J Bio- Tribo-Corros. 4(4):64. https://​doi.​org/​10.​1007/​s40735-​018-​0179-3f 49. Lgaz H, Salghi R, Jodeh S, Hammouti B (2017) Effect of clozapine on inhi- bition of mild steel corrosion in 1.0 M HCl medium. J Mol Liq. 225:271– 280. https://​doi.​org/​10.​1016/j.​molliq.​2016.​11.​039 33. Akinyele OF, Adekunle AS, Olayanju DS, Oyeneyin OE, Durodola SS, Ojo ND, Akinmuyisitan AA, Ajayeoba TA, Olasunkanmi LO (2022) Synthesis and corrosion inhibition studies of (E)-3-(2-(4-chloro-2-nitrophenyl) diazenyl)-1-nitrosonaphthalen-2-ol on mild steel dissolution in 0.5 M HCl solution- experimental, DFT and Monte Carlo simulations. J Mol Struct 1268:1–14. https://​doi.​org/​10.​1016/j.​molst​ruc.​2022.​133738 50. Hayat F, Zia-ur-Rehman, Khan MH (2017) Two new heteroleptic ruthe- nium (II) dithiocarbamates: synthesis, characterization, DFT calculation and DNA binding. J Coord Chem 70(2), 279-295.https://​doi.​org/​10.​1080/​ 00958​972.​2016.​12553​28 34. Imran M, Rehman ZU, Hogarth G, Tocher DA, Chaudhry G, Butler IS, Bélanger-Gariepy F, Kondratyuk T (2020) Two new monofunctional platinum(ii) dithiocarbamate complexes: phenanthriplatin -type axial protection, equatorial-axial conformational isomerism, and anticancer and DNA binding studies. Dalt Trans 49(43):15385–15396. https://​doi.​org/​ 10.​1039/​d0dt0​3018j 51. Arulraj R (2022) Hirshfeld surface analysis, interaction energy calculation and spectroscopical study of 3-chloro-3-methyl-r(2), c(6)-bis(p-tolyl) piperidin-4-one using DFT approaches. J Mol Struct. 1248:131483. https://​ doi.​org/​10.​1016/j.​molst​ruc.​2021.​131483 51. Arulraj R (2022) Hirshfeld surface analysis, interaction energy calculation and spectroscopical study of 3-chloro-3-methyl-r(2), c(6)-bis(p-tolyl) piperidin-4-one using DFT approaches. J Mol Struct. 1248:131483. https://​ doi.​org/​10.​1016/j.​molst​ruc.​2021.​131483 52. Ojo ND, Krause RW, Obi-Egbedi NO (2020) Electronic and nonlinear opti- cal properties of 3-(((2-substituted-4-nitrophenyl)imino)methyl)phenol. Comput Theor Chem. 1192:1–8. https://​doi.​org/​10.​1016/j.​comptc.​2020.​ 113050 35. Khan SZ, Amir MK, Ullah I, Aamir A, Pezzuto JM, Kondratyuk T, Bélanger- Gariepy F,  Ali A, Khan S, Rehman Z (2016) New heteroleptic palladium(II) dithiocarbamates: synthesis, characterization, packing and anticancer activity against five different cancer cell lines. Appl Organomet Chem 30(6):392–398. https://​doi.​org/​10.​1002/​aoc.​3445 53. Obi-Egbedi NO, Ojo ND (2021) Synthesis, light harvesting efficiency, photophysical and nonlinear optical properties of 3-(5-(4-hydroxyben- zylideneamino)naphthalen-1-yliminomethyl)phenol: spectroscopic and quantum chemical approach. Res Chem Intermed 47(12):5249–5266. https://​doi.​org/​10.​1007/​s11164-​021-​04579-4 36. Chermette H (1999) Chemical reactivity indexes in density functional theory. J Comput Chem 20(1):129–154. https://​doi.​org/​10.​1002/​(SICI)​ 1096-​987X(19990​115)​20:1%​3c129::​AID-​JCC13%​3e3.0.​CO;2-A 37. References Chem Africa 5(2):319–332. https://​doi.​org/​10.​1007/​s42250-​021-​00304-1 59. Zheng T, Liu J, Wang M, Liu Q, Wang J, ChongY, Jia G (2022) Synergistic corrosion inhibition effects of quaternary ammonium salt cationic surfactants and thiourea on Q235 steel in sulfuric acid: experimental and theoretical research. Corros Sci. https://​doi.​org/​10.​1016/j.​corsci.​2022.​ 110199f 43. Oyeneyin OE, Ipinloju N, Ojo ND, Dada D (2021) Structural modification of ibuprofen as new NSAIDs via DFT, molecular docking and pharmacoki- netics studies. Int J Adv Eng Pure Sci 33(4):614–626. https://​doi.​org/​10.​ 7240/​jeps.​928422 60. Assad H, Kumar A (2021) Understanding functional group effect on corrosion inhibition efficiency of selected organic compounds. J Mol Liq 344:117755. https://​doi.​org/​10.​1016/j.​molliq.​2021.​117755 61. Yan Y, Dai L, Zhang L, Zhong S (2018) Investigation on the corrosion inhibition of two newly-synthesized thioureas to mild steel in 1 mol/L HCl solution. Res Chem Intermed 44:3437–3454. https://​doi.​org/​10.​1007/​ s11164-​018-​3317-0 44. Obot IB, Macdonald DD, Gasem ZM (2015) Density functional theory (DFT) as a powerful tool for designing new organic corrosion inhibitors. Part 1: an overview. Corros Sci 99:1–30. https://​doi.​org/​10.​1016/j.​corsci.​ 2015.​01.​037 45. Amoko JS, Akinyele OF, Oyeneyin OE, Olayanju DS (2021) Corrosion inhibitive potentials of (E)-5-((4-Benzoylphenyl) Diazenyl)-2-hydroxy- benzoic acid on mild steel surface in 0.5 M HCl-experimental and DFT calculations. J Turkish Chem Soc Sect A Chem 8(1):343–362. https://​doi.​ org/​10.​18596/​jotcsa.​821488 Publisher’s Note S i N i Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. 46. Martínez-Araya JI (2015) Why is the dual descriptor a more accurate local reactivity descriptor than Fukui functions? J Math Chem. 53(2):451–465. https://​doi.​org/​10.​1007/​s10910-​014-​0437-7 47. Oyeneyin O, Akerele D, Ojo N, Oderinlo O (2021) Corrosion inhibitive potentials of some 2H–1-benzopyran-2-one derivatives- DFT calcula- tions. Biointerface Res Appl Chem 11(6):13968–13981. https://​doi.​org/​10.​ 33263/​BRIAC​116.​13968​13981 47. Oyeneyin O, Akerele D, Ojo N, Oderinlo O (2021) Corrosion inhibitive potentials of some 2H–1-benzopyran-2-one derivatives- DFT calcula- tions. Biointerface Res Appl Chem 11(6):13968–13981. https://​doi.​org/​10.​ 33263/​BRIAC​116.​13968​13981
https://openalex.org/W4366817211
https://discovery.ucl.ac.uk/10168990/1/journal.pcbi.1011055.pdf
English
null
Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields
PLOS computational biology/PLoS computational biology
2,023
cc-by
13,296
Editor: Daniel A Beard, University of Michigan, UNITED STATES Received: November 4, 2022 Accepted: March 28, 2023 Published: April 24, 2023 Copyright: © 2023 Pajaziti et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields Endrit PajazitiID1*, Javier Montalt-Tordera1, Claudio Capelli1, Raphae¨l Sivera1, Emilie Sauvage1, Michael Quail2, Silvia Schievano1, Vivek Muthurangu1 Endrit PajazitiID1*, Javier Montalt-Tordera1, Claudio Capelli1, Raphae¨l S Emilie Sauvage1, Michael Quail2, Silvia Schievano1, Vivek Muthurangu1 1 University College London, Institution of Cardiovascular Science, London, United Kingdom, 2 G Ormond Street Hospital, Cardiac Unit, London, United Kingdom a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * endrit.pajaziti.13@ucl.ac.uk a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Abstract Computational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high computational resources, specialist experience needed for designing simulation set-ups, and long processing times. The aim of this study was to explore the use of machine learning (ML) to replicate conventional aortic CFD with auto- matic and fast regression models. Data used to train/test the model consisted of 3,000 CFD simulations performed on synthetically generated 3D aortic shapes. These subjects were generated from a statistical shape model (SSM) built on real patient-specific aortas (N = 67). Inference performed on 200 test shapes resulted in average errors of 6.01% ±3.12 SD and 3.99% ±0.93 SD for pressure and velocity, respectively. Our ML-based models performed CFD in *0.075 seconds (4,000x faster than the solver). This proof-of-concept study shows that results from conventional vascular CFD can be reproduced using ML at a much faster rate, in an automatic process, and with reasonable accuracy. OPEN ACCESS Citation: Pajaziti E, Montalt-Tordera J, Capelli C, Sivera R, Sauvage E, Quail M, et al. (2023) Shape- driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields. PLoS Comput Biol 19(4): e1011055. https://doi.org/ 10.1371/journal.pcbi.1011055 Editor: Daniel A Beard, University of Michigan, UNITED STATES PLOS COMPUTATIONAL BIOLOGY PLOS COMPUTATIONAL BIOLOGY Introduction e T r a a v M S T r d t Our primary aim was to show the feasibility of a shape-driven ML model for the accurate estimation of 3D CFD flow fields in a population of aortas with challenging shape features. The main aims of this study were: (i) to create a large synthetic cohort of 3D aortas based on real clinical cases with complex anatomies, (ii) to train ML models to predict aortic pressure and velocity fields by representing unstructured/large data types with low-dimensional vectors, and (iii) to compare results between our fast/automatic ML-based CFD solution and our con- ventional CFD method on both synthetic and real aortic cases. This is a PLOS Computational Biology Methods paper. This is a PLOS Computational Biology Methods paper. Medical Research, received by ES (GN2572), European Research Council, received by SS (ERC- 2017-StG-757923). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Medical Research, received by ES (GN2572), European Research Council, received by SS (ERC- 2017-StG-757923). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Author summary In the clinical management of pediatric disease (namely congenital heart defects), the indications for ‘when’ and ‘how’ to intervene are often unclear. It has been found that hae- modynamic modelling tools such as computational fluid dynamics (CFD) simulations are useful in assisting clinicians and surgeons to better understand patient conditions and establish any potential risk factors. While this tool remains useful in a research capacity, its separation from clinical settings is an ongoing hindrance which prevents the full adop- tion of CFD in healthcare. The translation of CFD towards clinics is a continuous chal- lenge, due to large time, computational and human resource requirements for running simulations. The application of machine learning (ML) for exploring potential methods to transform conventional CFD into clinically-suitable models is a recent phenomenon which is gaining significant momentum. Data Availability Statement: All code related to the study can be found in the GitHub repository CFD_Machine_Learning: https://github.com/ EndritPJ/CFD_Machine_Learning. Funding: We would like to acknowledge the following funders: UK Research and Innovation, received by SS (MR/S032290/1), Heart Research UK, received by VM (RG2661/17/20), British Heart Foundation, received by SS (NH/18/1/33511), the Engineering and Physical Sciences Research Council, received by SS (EP/N02124X/1), Action 1 / 20 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields i di l tti ith tic 3D pressure and velocity flow fields Introduction Competing interests: The authors have declared that no competing interests exist. Computational fluid dynamics (CFD) has significant potential in cardiovascular settings, with applications including modelling of complex flow patterns and non-invasive estimation of vas- cular pressure [1]. Importantly, several studies have validated CFD measurements against con- ventional clinical methods such as 4D cardiovascular magnetic resonance (CMR) imaging and catheter-based pressure measurements [2, 3]. A major strength of CFD lies in situations where it is difficult to acquire data with conventional clinical means. For example, studies have shown CFD can be used to assess the haemodynamic response to exercise as a non-invasive alternative to cardiac catheterisation with exercise or pharmacological stress [4]. Another example is the use of CFD for predicting the haemodynamic response to specific interventions, such as aortic stenting in patients with coarctation [5]. Furthermore, there is increasing inter- est in exploring CFD-derived metrics for risk stratification of patients. For example, Qiu et al. demonstrated that CFD derived helical flow patterns were associated with significantly increased risk of abdominal aortic aneurysm rupture [6]. Thus, integration of CFD into clini- cal settings may have important ramifications when used: (i) for detailed assessment of patient-specific haemodynamics in response to stress or interventions, and (ii) to estimate CFD-derived indices for supporting clinical decision-making and risk-stratification. Despite potential benefits, CFD is still not integrated into routine clinical practice. This is mainly due to long computation times, the requirement for large amounts of processing power, and the need for an experienced engineer to set up simulations correctly [7]. Recently, machine learning (ML) models have been successful in replacing time-consuming or compu- tationally intensive tasks, such as medical image segmentation [8]. Similarly, the use of ML for speeding up cardiac CFD tasks is also becoming increasingly studied [9, 10]. The main chal- lenges of training ML models using CFD data include: (i) poor availability of clinical data, (ii) unstructured meshes without point correspondence, and (iii) large meshes and resultant CFD flow fields. To overcome these problems, we used statistical shape modelling and dimensional- ity reduction techniques to produce dimensionality-reduced representations of both aortic shape and flow fields, and to enable creation of large amounts of synthetic training data [11, 12]. This approach relies on the fact that in the past it has been show haemodynamic flow structures can be regressed from shape features [13]. Fig 1. Segmentation and mesh-preprocessing pipeline. The aorta segmentation of each subject is re-meshed and smoothed in an automatic pipeline. This is followed by clipping of the inlets/outlets and head & neck vessels. https://doi.org/10.1371/journal.pcbi.1011055.g001 retrospectively collected data was approved by the local research ethics committee, and written consent was obtained from all subjects/guardians (Ref: 06/Q0508/124). Images for each subject were segmented and converted into surface meshes (Fig 1). This was followed by remeshing and smoothing using functions from the vascular modelling toolkit (VMTK) [14]. All geome- tries were aligned in the same local space and orientation through rigid registration using an iterative closest point algorithm in VMTK [15]. This ensured that shape modelling was not affected by any spatial misalignment. Surfaces were manually clipped above the aortic root for the inlet, and at the diaphragm for the outlet. An SSM was then built using these clipped aortic surfaces, using an approach previously described by Bruse et al [16]. g pp p y y The package Deformetrica 4 was used to build the SSM [17]. First, an average aortic shape (surface template) was computed, containing 2,541 nodes (Fig 2). A volume template was also created by meshing the surface template with tetrahedral elements (29,000 nodes). Each sub- ject could then be described as a non-linear deformation of 3D ambient space, relative to the template (Fig 2). In this case, each deformation is fully parameterised by a paired set of 3D control points (qi)i=1,. . .n and 3D momenta vectors (μi)i=1,. . .n using a Gaussian kernel of width σ which we set as 10 mm. The number and location of control points were optimised (n = 172) by initialising the model with a high resolution control point grid (n = 500) and truncating points which were observed to have little influence on the deformation (low variance). The final computed 3D deformations for a given subject were represented by a deformation vector with 516 coefficients—the number of control points (172) multiplied by the number of defor- mation directions (3). The deformation vectors for the whole population were collected in the 2D matrix M [67, 516], which was decomposed using principal component analysis (PCA) in order to identify low-dimensional deformations which account for most of the variance. This required standardisation for each column in M (i.e. removal of the mean and scaling to unit variance). Statistical shape modelling The dataset used for development of the statistical shape model (SSM) consisted of cardiac and respiratory gated steady state free precession CMR images (N = 67) from patients previously diagnosed with coarctation of the aorta (CoA). All patients were post-surgical repair, asymp- tomatic and underwent CMR imaging at a mean age of 22.4 ±6.2 years. The use of Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 2 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Fig 1. Segmentation and mesh-preprocessing pipeline. The aorta segmentation of each subject is re-meshed and smoothed in an automatic pipeline. This is followed by clipping of the inlets/outlets and head & neck vessels. https://doi.org/10.1371/journal.pcbi.1011055.g001 The deformation matrix L was then reshaped into a 3D momenta matrix [3000, 172, 3] and applied onto the aortic sur- face and volume template using Deformetrica, thus generating a surface and volume mesh for each new synthetic subject. Since all new meshes are derived from the same template, all syn- thetic aortas contained the same number of nodes/elements. Additionally, nodes can be thought to be lying within spatially correspondent locations within each aorta (see S1 Fig). This was vital for enabling the dimensionality reduction of derived flow fields, as described in later sections. The new synthetic population (n = 3000) was compared to the original population (n = 67) by computing geometric properties of the shapes based on a centreline approach. Mean cen- treline lengths and diameters were computed. Mean torsion is used to express how sharply the centreline is twisting in space. The parameter tortuosity describes the length ratio between the centreline and a rectilinear line between the endpoints. All parameters were computed using implementations within VMTK, as described by Piccinelli et al. [18]. https://doi.org/10.1371/journal.pcbi.1011055.g002 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 4 / 20 associated PCA modes, and the matrix product US [67, 516] contains the projection (weights) of each subject onto each PCA mode. It was found that the first 35 PCA modes were capable of approximating 99% of the variance in M. This meant that specific aortic shapes could be repre- sented by a lower-dimensional deformation vector with 35 coefficients rather than 516 (almost 15 times reduction in the size). associated PCA modes, and the matrix product US [67, 516] contains the projection (weights) of each subject onto each PCA mode. It was found that the first 35 PCA modes were capable of approximating 99% of the variance in M. This meant that specific aortic shapes could be repre- sented by a lower-dimensional deformation vector with 35 coefficients rather than 516 (almost 15 times reduction in the size). Using the SSM, new synthetic aortic shapes could then be created using synthetic lower dimensional deformation vectors. Specifically, each of the 35 coefficients in a synthetic vector was generated by randomly sampling a Gaussian distribution (within 2 standard deviations) based on the distribution of weights in the US matrix. Following this, singular value decomposition (SVD) was performed, with M being decomposed into three matrices: M = USVT. The V transpose matrix [516, 516] contained the principal component axes or PCA modes, S is a rectangular diagonal matrix [67, 516] where the singular values on the diagonal can be used to calculate the variance explained by the Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 3 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Fig 2. Mesh registration with SSM. An example aortic shape approximation using our SSM is shown. Individual surface or volumes can be reconstructed using a mean aortic shape and applied deformation field initialised on a set of control points (n = 172). https://doi.org/10.1371/journal.pcbi.1011055.g002 Fig 2. Mesh registration with SSM. An example aortic shape approximation using our SSM is shown. Individual surface or volumes can be reconstructed using a mean aortic shape and applied deformation field initialised on a set of control points (n = 172). https://doi.org/10.1371/journal.pcbi.1011055.g002 Fig 2. Mesh registration with SSM. An example aortic shape approximation using our SSM is shown. Individual surface or volumes can be reconstructed using a mean aortic shape and applied deformation field initialised on a set of control points (n = 172). https://doi.org/10.1371/journal.pcbi.1011055.g002 https://doi.org/10.1371/journal.pcbi.1011055.g002 associated PCA modes, and the matrix product US [67, 516] contains the projection (weights) of each subject onto each PCA mode. It was found that the first 35 PCA modes were capable of approximating 99% of the variance in M. This meant that specific aortic shapes could be repre- sented by a lower-dimensional deformation vector with 35 coefficients rather than 516 (almost 15 times reduction in the size). Using the SSM, new synthetic aortic shapes could then be created using synthetic lower dimensional deformation vectors. Specifically, each of the 35 coefficients in a synthetic vector was generated by randomly sampling a Gaussian distribution (within 2 standard deviations) based on the distribution of weights in the US matrix. Following concatenation of all the lower dimensional deformation vectors, the matrix X [3000, 35] was transformed into a matrix L [3000, 516] by matrix multiplication (L = XSVT), thus reversing PCA and mapping the matrix X onto the original axes. Standardisation was then reversed in all columns of L using the previ- ously computed standard deviations and means in the M matrix. Following concatenation of all the lower dimensional deformation vectors, the matrix X [3000, 35] was transformed into a matrix L [3000, 516] by matrix multiplication (L = XSVT), thus reversing PCA and mapping the matrix X onto the original axes. Standardisation was then reversed in all columns of L using the previ- ously computed standard deviations and means in the M matrix. The deformation matrix L was then reshaped into a 3D momenta matrix [3000, 172, 3] and applied onto the aortic sur- face and volume template using Deformetrica, thus generating a surface and volume mesh for each new synthetic subject. Since all new meshes are derived from the same template, all syn- thetic aortas contained the same number of nodes/elements. Additionally, nodes can be thought to be lying within spatially correspondent locations within each aorta (see S1 Fig). This was vital for enabling the dimensionality reduction of derived flow fields, as described in later sections. The new synthetic population (n = 3000) was compared to the original population (n = 67) by computing geometric properties of the shapes based on a centreline approach. Mean cen- treline lengths and diameters were computed. Mean torsion is used to express how sharply the centreline is twisting in space. The parameter tortuosity describes the length ratio between the centreline and a rectilinear line between the endpoints. All parameters were computed using implementations within VMTK, as described by Piccinelli et al. [18]. Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 4 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Computational fluid dynamics pipeline Volume meshes previously generated with the SSM were unsuitable for CFD computation. This was primarily because low mesh skewness could not be guaranteed, and remeshing was not an option since nodes were to be preserved in order to maintain point correspondence (see S1 Fig). Therefore, separate meshes solely for CFD computation were built, starting from the surface of each aorta. Firstly, each surface was extended by 40mm at the inlet (Fig 3). This was done to produce a flat and circular inlet upon which a velocity profile could be uniformly applied. Extending the inlet further than 40mm was avoided in order to reduce the likelihood of surface self-intersection. Following this, volume meshing with tetrahedral elements was per- formed [19] (*400,000 cells on average). Element/node counts were deemed to be in satisfac- tory ranges after a mesh sensitivity analysis (see S2 Fig). CFD (Fluent, Ansys Technologies) was performed on all 3,000 synthetic cases. The same boundary conditions were applied to each simulation, as part of adopting a simple model which would reliably converge for all subjects. Laminar, steady-state flow conditions were enforced. An inlet velocity of 1.3 m/s, corresponding to an average ascending aortic flow rate at peak systole, was set [20, 21]. A velocity boundary condition was preferred to a volumetric flow rate, since it is invariant to any differences in inlet surface area between subjects. Outlet gauge pressure was fixed at 0 Pa. Standard non-slip conditions were applied at the wall, and the fluid was assumed to be Newtonian with density and dynamic viscosity equal to 1,060 kg/ m3 and 0.004 Pas, respectively [22]. The set-up and simulation of all 3,000 cases was fully automated. Machine learning Data interpolation and principal component analysis. As CFD was performed on large unstructured meshes with inconsistent numbers of nodes/elements between cases, point corre- spondence had to be restored prior to PCA-based dimensionality reduction of the flow fields (needed for easier model training). This was done using the volume meshes previously gener- ated with the SSM by ‘shooting’ on the template. Since each of these ‘SSM volume meshes’ n the t Fig 3. CFD Pipeline. Surfaces have flow extensions added before volume meshing. The same simulation set-up is applied to the final volume mesh for each synthetic subject. Fig 3. CFD Pipeline. Surfaces have flow extensions added before volume meshing. The same simulation set-up is applied to the final volume mesh for each synthetic subject https://doi.org/10.1371/journal.pcbi.1011055.g003 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 5 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields inherited its properties from the template, relative nodal positions were preserved. Conse- quently, pressure/velocity data for all subjects were re-sampled from unstructured CFD meshes onto SSM volume meshes using a Voronoi kernel (padding of 5 mm, grid resolution of 1,000,000 voxels) in the software Paraview (Fig 4). This resulted in 3,000 newly resampled pressure/velocity fields, with all subjects containing 29,000 nodes in point correspondence. The data was concatenated into a feature vector and PCA was applied to reduce dimensional- ity. We aimed to capture 99% of variance with as few PCA modes as possible. Deep neural network architecture. The architecture we adopted was a standard sequen- tial, fully-connected deep neural network (DNN) with independent networks for pressure and velocity. The input for the model was the lower dimensional deformation vector, which is also referred to as a ‘shape vector’. The outputs of the trainable part of the model were the pressure/ velocity PCA scores (reduced order CFD field), referred to as a ‘pressure/velocity vector’. A non-trainable inverse PCA layer (implemented in Keras using a lambda layer) serves to recon- struct the pressure/velocity vector into the full 3D flow field with 29,000 nodes (see Fig 5). Rec- tified linear units (ReLU) were used in each hidden layer. Linear activation functions were set at the output. Model implementation was done using Keras and TensorFlow 2.0. Deep neural network training. Models were built separately for predicting the static pres- sure and the velocity-magnitude. The loss function used for training was the mean absolute error (MAE), computed on the entire 3D flow field (i.e. after inverse PCA) rather than on the output pressure/velocity vector. This provides a more granular measure of error and effectively weights the importance of each PCA mode in the network according to the amount of variance it explains. Model optimisation was carried out using the Adam optimiser [23]. The training Fig 4. CFD data interpolation. CFD results are interpolated onto a point-correspondence mesh (generated by the SSM), thus restoring node concordance. Fig 4 conc https Fig 4. CFD data interpolation. CFD results are interpolated onto a point-correspondence mesh (generated by the SSM), thus restoring node n rd n Fig 4. CFD data interpolation. CFD results are interpolated onto a point-correspondence mesh (generated by the SSM), thus restoring node concordance. https://doi.org/10.1371/journal.pcbi.1011055.g004 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 6 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Fig 5. DNN general architecture. The general sequential, fully-connected DNN set-up used to build both pressure and velocity predictors (’CFD vector’ can be either pressure or velocity PCA vectors). https://doi.org/10.1371/journal.pcbi.1011055.g005 Fig 5. DNN general architecture. The general sequential, fully-connected DNN set-up used to build both pressure and velocity predictors (’CFD vector’ can be either pressure or velocity PCA vectors). Fig 5. DNN general architecture. The general sequential, fully-connected DNN set-up used to build both pressure and velo vector’ can be either pressure or velocity PCA vectors). https://doi.org/10.1371/journal.pcbi.1011055.g005 https://doi.org/10.1371/journal.pcbi.1011055.g005 dataset was composed of 2800 randomly selected subjects, with the remaining (200) going into the test set. Hyperparameter tuning was conducted using 5-fold cross-validation to find model settings which generalise well for unseen data. For this, the training dataset was split into 5 groups (580 subjects each). The model was trained and tested five times, with each group taking turns to act as the validation set while the remaining four were used for training. The average validation error for all five training runs was used to represent the performance of that model. This pro- cess was repeated for 1,000 model configurations, which were sampled using a tree-structured Parzen estimator (TPE) algorithm [24]. The number of hidden layers, number of hidden layer neurons and initial learning rate were all explored. Batch size and epochs were set at 32 and 50, respectively. Hyperband pruning was used to terminate early training rounds if the model was deemed to be poorly fitting the validation data. After completing hyperparameter tuning, the best model was retrained for 1000 epochs using a larger portion of the training set (2,600 sub- jects) with 200 remaining for validation. Model training lasted 1,000 epochs and training/vali- dation loss was monitored to prevent overfitting. Model training was performed using an Nvidia GTX 1080Ti graphics card. Model evaluation. Once trained, the model was evaluated on the test set of synthetic aor- tas (n = 200). Absolute errors were computed for every node in all test cases by comparing the prediction value (ML) to the ground-truth value (CFD). Errors were then normalised accord- ing to subject CFD data range, as detailed in Liang et al. [11]. Normalisation was necessary to enable direct comparison between individual cases and also between CFD metrics, since pres- sure/velocity ranges widely differed per subject. Eq 1 details how normalised absolute error (NAE) is computed for either pressure or velocity at a node i, belonging to a subject j. Truei,j is the CFD nodal pressure/velocity value. Predi,j is the ML nodal pressure/velocity value. Range (Truej) is the difference between the maximum and minimum values in the CFD flow field for subject j. NAEði; jÞ ¼ jTruei; j Predi; jj RangeðTruejÞ  100% ð1Þ NAEði; jÞ ¼ jTruei; j Predi; jj RangeðTruejÞ  100% ð1Þ ð1Þ Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 7 / 20 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Mean node errors (MNAEN) were computed by averaging NAE values across the popula- tion for each node (n = 29,000). These were then plotted on the template mesh points in order to better visualise the magnitude of these errors with respect to their location. However, since NAE values are absolute errors, this provides no insight regarding any systematic over or under-estimation during model inference. Therefore, a Bland-Altman plot was used to exam- ine the bias and limits of agreement of the pressure and velocity DNNs. This was done for the overall aorta and for three separate regions; ascending aorta, transverse arch and descending aorta (anatomically defined). Mean subject errors (MNAES) were computed by averaging NAE values in each subject (n = 200). Cases with the best, median and worse mean subject error values were compared. A single population error for both pressure and velocity was given by averaging all MNAES val- ues. The relationship between shape mode scores and subject error (MNAES) was investigated with scatter plots and assessed using Pearson R coefficients and p values (p = 0.05 considered significant). In addition to evaluation of the models on the test set (n = 200), the models were also tested on real, patient-specific aortas with previously repaired CoA (n = 10), completely unseen from the SSM and the DNN. This was done in order to validate the robustness of the models for inferring accurate flow fields on real subjects outside our synthetic training/testing sets. CMR images of each patient were segmented. Surfaces were approximated by non-rigid registration (applying deformations on the template) using the SSM. Deformation matrices for each case were decomposed into PCA shape vectors and passed as inputs into the DNN models. Pressure and velocity-magnitude fields were inferred for each subject. Following this, all ten predictions were compared to CFD flow fields computed using both SSM derived geometries and real geometries. Comparison between ML and SSM derived flow fields were made using MNAES as previ- ously described. However, it was not possible to compare the predicted and real CFD flow fields with node-based metrics due to the lack of node-to-node correspondence and exact sur- face matching. Therefore, we used a gradient-based approach to enable direct comparison of pressure/velocity flow fields without shape correspondence. Subject centrelines were used to calculate plane-averaged pressure/velocity gradients at 99 locations over the length of the aorta (Fig 6). To compare gradients, the Fre´chet distance (FD) was used. The FD is a measure of similarity between two point-sets of curves, taking into account the location and ordering of the curve coordinates. Intuitively, it can be thought of as the shortest possible distance between two observers traversing different paths while remaining connected. An advantage of using the FD is that it does not neglect sharp spikes or differences between gradients, which some other metrics may diminish through averaging. Additionally, since FD is not a percentage error, it does not emphasise errors where values are close to zero (such as at the very end of the descending aorta in pressure flow fields). An algorithmic implementation for computing FD as described by Eiter et al. was used to calculate this metric [25]. Statistical shape modelling PCA decomposition was performed on the deformation matrices (momenta) computed by sta- tistical shape modelling. The first, second and third PCA modes captured 29.8%, 13.2% and 10.1% (total 53.1%) of the variability, respectively. Mode 1 relates to overall vessel size, mode 2 relates to ascending arch angulation/diameter, and mode 3 describes rounded versus triangu- lar arches. After PCA decomposition, 99% of the variance in the momenta could be repre- sented with the first 35 modes. Some examples of the 3,000 synthetic subjects produced by Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 8 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Deep neural networks for fast aortic 3D pressure and velocity flow fields randomly sampling and combining 35 PCA mode scores are shown in Fig 7. Anatomical char- acteristics of the synthetic aortas (length, diameter, tortuosity and torsion) were found to be statistically similar to those of the real patient cohort (see Table 1). Fig 6. CFD gradient extraction method. Using subject centrelines, 99 plane-averaged pressure or velocity points along the length of the aorta are extracted by sampling the 3D flow fields. The origin is always the aortic root (excluding the extension). https://doi.org/10.1371/journal.pcbi.1011055.g006 ent extraction method. Using subject centrelines, 99 plane-averaged pressure or velocity points along the length of the aorta a ling the 3D flow fields. The origin is always the aortic root (excluding the extension). 371/journal pcbi 1011055 g006 randomly sampling and combining 35 PCA mode scores are shown in Fig 7. Anatomical char- acteristics of the synthetic aortas (length, diameter, tortuosity and torsion) were found to be statistically similar to those of the real patient cohort (see Table 1). randomly sampling and combining 35 PCA mode scores are shown in Fig 7. Anatomical char- acteristics of the synthetic aortas (length, diameter, tortuosity and torsion) were found to be statistically similar to those of the real patient cohort (see Table 1). Training data Average MNAES in pressure and velocity fields were found to be 1.46% ± 0.59 Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 9 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields SD and 2.70% ±0.49 SD. The reconstructed test cases with the highest MNAES for pressure and velocity (4.32% and 4.93%, respectively) are shown in S3 Fig. Model architecture The model input layer size was set at 35 (number of shape modes). Output layer sizes were set at 20 and 55 for pressure and velocity, respectively (number of pressure/velocity modes). Hyperparameter tuning using cross-validation was performed 1,000 times to search for the optimal learning rate, number of neurons and number of layers, with tuning taking *4 hours per model. Pressure and velocity model architectures as a result of the optimisation process are shown in S4 Fig. Model predictive performance Pressure and velocity-magnitude fields were computed on the test set (n = 200) using the trained DNNs. Inference took 0.075 seconds per subject for both DNNs. In comparison, Fig 7. Modes of deformation. Left: first three modes of deformation from the SSM (SD = standard deviation). Right: examples of synthetic post-CoA aortas from the test set (using combinations of all 35 shape modes). https://doi.org/10.1371/journal.pcbi.1011055.g007 parison of aorta dimensions between original real cohort (n = 67) and synthetic cohort (n = 3000). Mean diameter (mm) Length (mm) Tortuosity Mean torsion erage (n = 67) 19.74 ±1.29 257.3 ±29.88 2.21 ±0.39 0.0044 ±0.061 rt average (n = 3000) 20.12 ±1.32 258.5 ±23.08 2.19 ±0.35 0.0012 ±0.053 h’s t-test) 0.25 0.75 0.74 0.65 0.1371/journal.pcbi.1011055.t001 MPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Fig 7. Modes of deformation. Left: first three modes of deformation from the SSM (SD = standard deviation). Right: examples of synthetic post-CoA aortas from the test set (using combinations of all 35 shape modes). SD and 2.70% ±0.49 SD. The reconstructed test cases with the highest MNAES for pressure and velocity (4.32% and 4.93%, respectively) are shown in S3 Fig. Fig 7. Modes of deformation. Left: first three modes of deformation from the SSM (SD = standard deviation). Right: examples of synthetic post-CoA aortas from the test set (using combinations of all 35 shape modes). https://doi.org/10.1371/journal.pcbi.1011055.g007 Fig 7. Modes of deformation. Left: first three modes of deformation from the SSM (SD = standard deviation). Training data After CFD was computed on all cases (n = 3000), values were interpolated from high resolu- tion meshes onto lower-resolution grids in point correspondence. The mean loss in accuracy due to interpolation was found to be 0.056% ±0.027 and 0.849% ±0.247 for pressure and veloc- ity-magnitude, respectively. This was computed by calculating the mean percentage error in centreline pressure and velocity gradients for all cases (n = 3000) and averaging the results. The data post-interpolation was used as the ‘ground-truth’ training and testing sets. PCA decomposition of the pressure and velocity training data matrices ([2800, 29000] each) was then performed, following standardisation. After PCA decomposition, 99% of the standardised pressure variance could be captured with 20 modes. Only 87% of the standard- ised velocity variance could be captured with 55 modes, and it was felt that adding more modes to capture greater variance was not feasible due to massively diminishing returns. Sub- ject errors resulting from PCA decomposition were tested on the 200 test cases (unseen by the PCA model). Average MNAES in pressure and velocity fields were found to be 1.46% ± 0.59 After CFD was computed on all cases (n = 3000), values were interpolated from high resolu- tion meshes onto lower-resolution grids in point correspondence. The mean loss in accuracy due to interpolation was found to be 0.056% ±0.027 and 0.849% ±0.247 for pressure and veloc- ity-magnitude, respectively. This was computed by calculating the mean percentage error in centreline pressure and velocity gradients for all cases (n = 3000) and averaging the results. The data post-interpolation was used as the ‘ground-truth’ training and testing sets. p p g g g PCA decomposition of the pressure and velocity training data matrices ([2800, 29000] each) was then performed, following standardisation. After PCA decomposition, 99% of the standardised pressure variance could be captured with 20 modes. Only 87% of the standard- ised velocity variance could be captured with 55 modes, and it was felt that adding more modes to capture greater variance was not feasible due to massively diminishing returns. Sub- ject errors resulting from PCA decomposition were tested on the 200 test cases (unseen by the PCA model). Training data Right: examples of synthetic post-CoA aortas from the test set (using combinations of all 35 shape modes). https://doi.org/10.1371/journal.pcbi.1011055.g007 SD and 2.70% ±0.49 SD. The reconstructed test cases with the highest MNAES for pressure and velocity (4.32% and 4.93%, respectively) are shown in S3 Fig. Model architecture The maximum MNAES for pressure and velocity were found to be 23.60% and 8.07%, respectively. The mini- mum MNAES for pressure and velocity were found to be 1.54% and 1.91%, respectively. The relationship between pressure/velocity MNAES and shape mode coefficients is pre- sented in Fig 10. The second and third mode showed statistically significant correlations with velocity prediction MNAES, with the third shape mode showing the highest correlation (R= −0.31). No significant correlations were found between any other shape mode and MNAES. Bland-Altman analysis showed negligible prediction biases for the overall aorta (Fig 8, right) and within selected regions (S5 Fig). Bland-Altman biases were found to be 0.19% and 0.28% for pressure and velocity, respectively. The limits of agreement were found to be mar- ginally wider for pressure when compared to velocity (15.65% vs 12.89%, respectively). Subject errors. The population error for pressure and velocity was 6.01 ±3.12% SD and 3.99 ±0.93% SD, respectively. The test cases with the best, median and worst subject error (MNAES) are shown in Fig 9 with corresponding pressure/velocity gradients. The maximum MNAE for pressure and velocity were found to be 23 60% and 8 07% respectively The mini g 3 (M M m se v − V C th p Model architecture Pressure and velocity-magnitude fields were computed on the test set (n = 200) using the trained DNNs. Inference took 0.075 seconds per subject for both DNNs. In comparison, 10 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Fig 8. Nodal error analysis. Left: Distribution of mean nodal errors (MNAEN), computed on the test set (n = 200). Errors are absolute values and are projected on the template aorta. Right: Bland-Altman plots for the overall aorta. Normalised error (%) refers to the NAE of each node in every test case, without taking the absolute value (n = 5,800,000). Only 1,000 randomly selected points were drawn to improve graph readability. https://doi.org/10.1371/journal.pcbi.1011055.g008 Fig 8. Nodal error analysis. Left: Distribution of mean nodal errors (MNAEN), computed on the test set (n = 200). Errors are absolute values and are projected on the template aorta. Right: Bland-Altman plots for the overall aorta. Normalised error (%) refers to the NAE of each node in every test case, without taking the absolute value (n = 5,800,000). Only 1,000 randomly selected points were drawn to improve graph readability. https://doi.org/10.1371/journal.pcbi.1011055.g008 https://doi.org/10.1371/journal.pcbi.1011055.g008 conventional CFD took *5 minutes on average for convergence, demonstrating an approxi- mate 4,000x speed-up with ML. Node errors. Average node prediction errors (MNAEN) were computed for all nodes (n = 29,000). Fig 8 (left) shows these values projected onto the template (average position of the nodes), allowing for the locations of the highest absolute errors to be assessed. Pressure errors were observed to be lower in the descending aorta, with the highest errors situated in the transverse arch. Velocity errors were notably more prevalent in the underside of the arch and descending aorta. The maximum MNAEN was observed to be 12.46% and 14.86% for pressure and velocity, respectively. y y Bland-Altman analysis showed negligible prediction biases for the overall aorta (Fig 8, right) and within selected regions (S5 Fig). Bland-Altman biases were found to be 0.19% and 0.28% for pressure and velocity, respectively. The limits of agreement were found to be mar- ginally wider for pressure when compared to velocity (15.65% vs 12.89%, respectively). Subject errors. The population error for pressure and velocity was 6.01 ±3.12% SD and 3.99 ±0.93% SD, respectively. The test cases with the best, median and worst subject error (MNAES) are shown in Fig 9 with corresponding pressure/velocity gradients. Validation with real patient data Coefficients for the first 10 shape modes for all new subjects (n = 10) were found to lie within the max-min ranges of the original PCA shape modes (n = 67, see S6 Fig). The average prediction error using the MNAES metric (ML vs CFD computed on SSM meshes) was Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 11 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Fig 9. Best, median and worst test-set predictions. Comparisons between ground truth (CFD) and predicted (ML) in the test set (n = 200). Best, median and worst cases for both pressure and velocity-magnitude are shown, ranked using the mean node-to-node error (MNAES). Pressure/velocity gradients are also displayed (black lines = CFD, red = ML). Fig 9. Best, median and worst test-set predictions. Comparisons between ground truth (CFD) and predicted (ML) in cases for both pressure and velocity-magnitude are shown, ranked using the mean node-to-node error (MNAES). Press lines = CFD, red = ML). https://doi.org/10.1371/journal.pcbi.1011055.g009 10.19% ±10.41 and 4.47% ±1.18 for pressure and velocity respectively. This corresponds to an increase in mean subject error by 4.18% for pressure and 0.48% for velocity, when compared to the population error in the test-set of synthetic cases (n = 200). Due to the lack of point correspondence and surface matching between the real and SSM shapes, Frechet distance of pressure and velocity gradients were used to compare the ML CFD results with the patient CFD data (computed using both SSM and real geometries). The FD computed between the ML and the CFD (SSM) gradients corresponded to the prediction error arising solely from the ML model (FD SSM). The FD computed between the ML and the CFD (real shape) gradients corresponds to the total error between the ML prediction and true CFD Fig 9. Best, median and worst test-set predictions. Comparisons between ground truth (CFD) and predicted (ML) in the test set (n = 200). Best, median and worst cases for both pressure and velocity-magnitude are shown, ranked using the mean node-to-node error (MNAES). Pressure/velocity gradients are also displayed (black lines = CFD, red = ML). https://doi.org/10.1371/journal.pcbi.1011055.g009 edian and worst test-set predictions. Comparisons between ground truth (CFD) and predicted (ML) in the test set (n = 200). pressure and velocity-magnitude are shown, ranked using the mean node-to-node error (MNAES). Validation with real patient data Testing on prospective data. Best and worst pressure and velocity predictions on the real patient test cohort (n = 10). FD (SSM) is the error between the di t d ( d) d SSM CFD (d h d bl k) di t FD ( l) i th b t th di t d ( d) d t CFD ( lid bl k) di t (FD real). The subjects with the best and worst FD (real) for pressure and velocity are shown in Fig 11. The mean FD SSM was 1680 ±629 Pa for pressure and 0.47 ±0.17 m/s for velocity. This compares to 4583 ±3210 Pa for pressure and 1.30 ±0.37 m/s for velocity between the ML and real-CFD (FD real). Discussion The main findings of this study were: i) statistical shape models (SSMs) and PCA are suitable for creating synthetic training data and dimensionality-reduced representations of 3D shape and flow, ii) DNNs based on these dimensionality-reduced representations can predict pres- sure and velocity fields with high accuracy. Fig 11. Testing on prospective data. Best and worst pressure and velocity predictions on the real patient test cohort (n = 10). FD (SSM) is the error between the predicted (red) and SSM CFD (dashed black) gradients. FD (real) is the error between the predicted (red) and true CFD (solid black) gradients. https://doi.org/10.1371/journal.pcbi.1011055.g011 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 13 / 2 Fig 11. Testing on prospective data. Best and worst pressure and velocity predictions on the real patient test cohort (n = 10). FD (SSM) is the error between the predicted (red) and SSM CFD (dashed black) gradients. FD (real) is the error between the predicted (red) and true CFD (solid black) gradients. https://doi.org/10.1371/journal.pcbi.1011055.g (FD real). The subjects with the best and worst FD (real) for pressure and velocity are shown in Fig 11. The mean FD SSM was 1680 ±629 Pa for pressure and 0.47 ±0.17 m/s for velocity. This compares to 4583 ±3210 Pa for pressure and 1.30 ±0.37 m/s for velocity between the ML and real-CFD (FD real). Validation with real patient data Pressure/velocity gradients ed = ML) 10.19% ±10.41 and 4.47% ±1.18 for pressure and velocity respectively. This corresponds to an increase in mean subject error by 4.18% for pressure and 0.48% for velocity, when compared to the population error in the test-set of synthetic cases (n = 200). 10.19% ±10.41 and 4.47% ±1.18 for pressure and velocity respectively. This corresponds to an increase in mean subject error by 4.18% for pressure and 0.48% for velocity, when compared to the population error in the test-set of synthetic cases (n = 200). Due to the lack of point correspondence and surface matching between the real and SSM shapes, Frechet distance of pressure and velocity gradients were used to compare the ML CFD results with the patient CFD data (computed using both SSM and real geometries). The FD computed between the ML and the CFD (SSM) gradients corresponded to the prediction error arising solely from the ML model (FD SSM). The FD computed between the ML and the CFD (real shape) gradients corresponds to the total error between the ML prediction and true CFD Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 12 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields (FD real). The subjects with the best and worst FD (real) for pressure and velocity are shown in Fig 11. The mean FD SSM was 1680 ±629 Pa for pressure and 0.47 ±0.17 m/s for velocity. This compares to 4583 ±3210 Pa for pressure and 1.30 ±0.37 m/s for velocity between the ML d l CFD (FD l) Fig 10. Shape modes vs. ML error. Scatter plot comparing the shape PCA mode values against subject error (MNAES) in the test set (n = 200). Pearson R coefficients and p-values were computed for each subplot. https://doi.org/10.1371/journal.pcbi.1011055.g010 Fig 11. Testing on prospective data. Best and worst pressure and velocity predictions on the real patient test cohort (n = 10). FD (SSM) is the error between the predicted (red) and SSM CFD (dashed black) gradients. FD (real) is the error between the predicted (red) and true CFD (solid black) gradients. https://doi.org/10.1371/journal.pcbi.1011055.g011 g 10. Shape modes vs. ML error. Scatter plot comparing the shape PCA mode values against subject error (MNAES) the test set (n = 200). Pearson R coefficients and p-values were computed for each subplot. Fig 11. Model performance Our ML models were observed to predict point clouds of both pressure and velocity flow fields with good accuracy, while being approximately 4,000x faster than our conventional CFD method. Node errors for pressure were seen to be larger at the inlet, while velocity errors were more skewed towards the distal regions of the aorta (Fig 8, left). This is most likely due to the CFD boundary conditions at the inlet and outlet constraining the pressure/velocity variability at these regions. The further away from the aortic inlet or outlet, the greater the variability in velocity or pressure, respectively. It should be noted that there were no significant biases in either pressure or velocity predictions, suggesting that there were no systematic errors with the models (Fig 8, right). In testing, we found that the mean pressure subject error was slightly higher than the mean velocity error. This is despite the pressure PCA model capturing more variance than the velocity model. A possible explanation for the higher pressure errors is that the association between shape and pressure is more complex than that between shape and velocity in aortic domains. This is supported by the observation that shape modes do not cor- relate with pressure errors (Fig 10). Interestingly, there was a strong negative correlation between shape mode 3 and velocity error. This suggests that more ‘gothic’ aortas (characterised by a more triangular arch) were less prone to velocity prediction errors. A possible explanation may be that the gothic arch constrains downstream flow patterns (where most velocity errors occur), hence making it easier for the model to characterise flow features associated to this sub- set of aortic shapes. An important element of this study was to apply trained DNN models to prospective, unseen cases in order to explore the feasibility of performing inference on real patient data (n = 10). It was shown that the average node-based prediction error (MNAES) on the prospec- tive cases did increase for pressure from 6.01% to 10.19%. Velocity errors increased only mar- ginally in the prospective test cases (3.99% to 4.47%). This aligns with the previous observation that the relationship between shape and pressure may be more complex due to the lack of an observed correlation between the shape PCA modes and pressure prediction errors. We also compared the ML predictions to the real CFD (performed on the raw segmentation mesh) for all cases. Discussion The main findings of this study were: i) statistical shape models (SSMs) and PCA are suitable for creating synthetic training data and dimensionality-reduced representations of 3D shape and flow, ii) DNNs based on these dimensionality-reduced representations can predict pres- sure and velocity fields with high accuracy. Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 13 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields h i i ure and velocity flow fields Synthetic data generation and dimensionality reduction A key element of our approach was to use an SSM and PCA to both generate a synthetic train- ing dataset (n = 3,000), and to parameterise aortic shape/flow fields for simplifying DNN model training. We demonstrated that volumetric meshes generated by our SSM were suitable for forcing point correspondence in our dataset through interpolation of 3D aortic CFD flow fields. It was also shown that less than 60 PCA modes were needed to capture the majority of variance within aortic shape deformations and 3D pressure/velocity fields. Mean PCA recon- struction errors of pressure/velocity were found to be low, but not insignificant in the worst observed test-set reconstructions (S3 Fig). Geometric properties of our synthetic data were shown to be mostly close to the real cohort (Table 1). In the future, newer approaches for gen- erating synthetic data and creating dimensionality-reduced representations of complex struc- tures could be explored, notably deep-learning methods such as autoencoders [11] and generative adversarial networks (GANs) [26]. In some studies, autoencoders have been shown to be superior to PCA-based methods (e.g. for 3D facial surface reconstruction) [27, 28]. Addi- tionally, GANs have shown promise for creating images of synthetic patients afflicted with CHD [29]. Finally, instead of sampling a Gaussian distribution to find combinations of shape PCA parameters, other methods for creating new DNN training data may be more appropri- ate, such as Latin hypercube sampling [30]. Model performance O ML d l b d t di t i t l d f b th d l it fl fi ld Limitations In order to translate our DNN-based CFD approach to clinics, there are two main modelling limitations which need to be bypassed. The first is related to the loss in surface accuracy when using SSM representations of aortic shapes. The second is the current simplicity of the CFD approach, which needs to be further developed in order to generate more meaningful DNN training data. Model performance This enabled the proportion of the total error due to the SSM and DNN to be Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 14 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields estimated. We showed *60% of the total gradient error in pressure/velocity was due to the SSM, which further strengthens the argument for improved shape parameterisation. The approach of using DNNs to model 3D aortic pressure and velocity flow fields has been described in other works [10, 11]. However, an important limitation of this data-driven approach is that flow fields cannot be assumed to satisfy the Navier-Stokes equations for incompressible flow. This may hinder the applicability of the approach for simulations where accurate flow field prediction is insufficient in of itself, and conservation of mass and momen- tum needs to be guaranteed. Future development should also include models for computing velocity x, y and z components, allowing visualisation of streamlines or possible derivation of parameters such as wall-shear stress. Although we used fully-connected DNNs, other studies have reported the use of long short-term memory (LSTM) networks or specialised architec- tures such as PointNet to build CFD-based ML models [9, 31, 32]. LSTM networks in particu- lar may be highly suited towards any ML transient flow applications, due to their inherent ability to learn temporal sequences of data. Other architectures such as probabilistic DNNs which output uncertainty intervals during inference should also be explored [31, 33, 34]. Addi- tionally, alternative methods for sampling synthetic data (e.g. Latin hypercube sampling) may produce a more diverse dataset for model training. This may prevent the occurrence of outli- ers, such as the worst pressure case in the test-set (MNAES=23.6%). Potential clinical utility We believe that the fast computation of haemodynamics using our method has multiple clini Potential clinical utility We believe that the fast computation of haemodynamics using our method has multiple clini- cal uses. However, this is a proof-of-concept study and further improvements are required prior to any clinical validation (particularly the inclusion of patient-specific boundary condi- tions and time varying flow fields—see limitations). Nevertheless, if this could be achieved we envisage several clinical uses, such as for supporting the identification of patients who need an intervention and predicting the outcome. Specific to our population, several studies have shown that CFD can be used to evaluate abnormal haemodynamics (particularly during stress) and predict normalisation of haemodynamics after stenting of coarctation. However, this approach is rarely used in the clinical environment because it is so time consuming. We believe our approach could be extrapolated to evaluate stress haemodynamics by simulating each training dataset case under elevated cardiac stress conditions. A second application could involve a fast and automatic pipeline to predict post-stenting haemodynamics in aortas. This could be implemented in a two-step solution using: (i) a surrogate finite-element model for predicting an ideal post-op aortic shape following stenting, and (ii) a surrogate CFD model for predicting haemodynamics on the post-op aortic shape (following the approach we present in this study). Use of such models would bring forth a new level of precision medicine that is cur- rently lacking in congenital heart disease. p p t r L I l u a t S I a Shape parameterisation Additionally, multiple SSMs based upon templates other than the mean aortic shape could be generated. These would enable closer non-rigid reg- istration for unique cases where the target aorta deviates significantly from the mean shape. Recently, Wiputra et al. demonstrated methods for augmenting SSMs to enable the inclusion of head and neck vessel geometry within the aortic shape parameterisation, while retaining high accuracy with low-dimensional PCA vectors [37]. A similar approach may be employed in future studies in order to be able to fully describe the patient-specific aorta with head and neck vessels. Of course, a simple initial improvement could be made by adding more subjects to our SSM to introduce more variability in the population. Computational fluid dynamics A standard CFD solver set-up was used, assuming steady-state conditions and incompressible flow. In order to account for the pulsati- lity involved in aortic flow, a transient solver may be better suited to real-life applications. It was also assumed that the flow through the great arteries at peak systole was laminar [38]. In the future, the Reynolds number may be computed for individual cases to allow for the inclu- sion of turbulence modelling where necessary. However, for complex morphologies (such as CoA), Reynolds number has been seen to be an inconsistent measure of turbulence [39]. Feiger et al. proposed an alternative solution to turbulence modelling when performing CFD on a large scale, which involves meshing the domain with extremely high numbers of nodes [10]. Boundary conditions selected included a fixed, flat velocity inlet and zero pressure outlet condition for all cases. However, idealised inlet velocity profiles (flat, parabolic etc.) have been shown to be ineffective for producing clinically relevant data [40, 41]. Thamsen et al. showed that a synthetic aortic population could be created with realistic accompanying 4D MRI- derived vector flow profiles [42]. In the future, a similar approach could be taken, allowing for an additional velocity vector field input parameter into the ML model. Alternatively, a more accessible approach could involve the use of a parabolic velocity inlet condition and a patient- specific unsteady flow profile (derived from phase-contrast MRI) in conjunction with a tran- sient solver for resolving the peak systolic flow field. In this study, it was decided to omit the head and neck vessels from the CFD model for simplicity, however this would be required when aiming to simulate realistic patient-specific aortic haemodynamics [36]. Indeed, Wiputra et al. showed that accurate modelling of the head and neck vessels is necessary for capturing local flow features in the arch and producing realistic downstream fluid flow forces [37]. Thus, the inclusion of head and neck vessels along with lumped parameter outlet models such as Windkessel models should be explored in the future [43, 44], as modelling downstream resis- tance has been seen to produce more clinically meaningful results [45–47]. c s t d a a s s h w e lo t W t C I p Computational fluid dynamics In this study, a simplified CFD pipeline was chosen in order to easily automate and ensure convergence for numerous simulations (n = 3,000). A standard CFD solver set-up was used, assuming steady-state conditions and incompressible flow. In order to account for the pulsati- lity involved in aortic flow, a transient solver may be better suited to real-life applications. It was also assumed that the flow through the great arteries at peak systole was laminar [38]. In the future, the Reynolds number may be computed for individual cases to allow for the inclu- sion of turbulence modelling where necessary. However, for complex morphologies (such as CoA), Reynolds number has been seen to be an inconsistent measure of turbulence [39]. Feiger et al. proposed an alternative solution to turbulence modelling when performing CFD on a large scale, which involves meshing the domain with extremely high numbers of nodes [10]. Boundary conditions selected included a fixed, flat velocity inlet and zero pressure outlet condition for all cases. However, idealised inlet velocity profiles (flat, parabolic etc.) have been shown to be ineffective for producing clinically relevant data [40, 41]. Thamsen et al. showed that a synthetic aortic population could be created with realistic accompanying 4D MRI- derived vector flow profiles [42]. In the future, a similar approach could be taken, allowing for an additional velocity vector field input parameter into the ML model. Alternatively, a more accessible approach could involve the use of a parabolic velocity inlet condition and a patient- specific unsteady flow profile (derived from phase-contrast MRI) in conjunction with a tran- sient solver for resolving the peak systolic flow field. In this study, it was decided to omit the head and neck vessels from the CFD model for simplicity, however this would be required when aiming to simulate realistic patient-specific aortic haemodynamics [36]. Indeed, Wiputra et al. showed that accurate modelling of the head and neck vessels is necessary for capturing local flow features in the arch and producing realistic downstream fluid flow forces [37]. Thus, the inclusion of head and neck vessels along with lumped parameter outlet models such as Windkessel models should be explored in the future [43, 44], as modelling downstream resis- tance has been seen to produce more clinically meaningful results [45–47]. In this study, a simplified CFD pipeline was chosen in order to easily automate and ensure convergence for numerous simulations (n = 3,000). Shape parameterisation It has been seen in previous studies that aortic CFD flow fields are highly sensitive to geometric and topological variation [35, 36]. For this reason, using shape vectors that are accurate Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 15 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields descriptors of the aortic surfaces is of critical importance. In the future, the relationship between the SSM registration error and the resultant CFD flow fields should be further investi- gated. Where possible, augmentations to the shape vector should be trialled in order to see if DNN prediction errors for real subjects can be reduced. This may require including additional shape information as inputs in the DNN models to act as a form of regularisation, such as a registration error or a geometric feature (e.g. centreline diameters). Such complimentary shape descriptors may be used to better inform the network of important features not fully captured by the SSM shape vector alone. Additionally, multiple SSMs based upon templates other than the mean aortic shape could be generated. These would enable closer non-rigid reg- istration for unique cases where the target aorta deviates significantly from the mean shape. Recently, Wiputra et al. demonstrated methods for augmenting SSMs to enable the inclusion of head and neck vessel geometry within the aortic shape parameterisation, while retaining high accuracy with low-dimensional PCA vectors [37]. A similar approach may be employed in future studies in order to be able to fully describe the patient-specific aorta with head and neck vessels. Of course, a simple initial improvement could be made by adding more subjects to our SSM to introduce more variability in the population. descriptors of the aortic surfaces is of critical importance. In the future, the relationship between the SSM registration error and the resultant CFD flow fields should be further investi- gated. Where possible, augmentations to the shape vector should be trialled in order to see if DNN prediction errors for real subjects can be reduced. This may require including additional shape information as inputs in the DNN models to act as a form of regularisation, such as a registration error or a geometric feature (e.g. centreline diameters). Such complimentary shape descriptors may be used to better inform the network of important features not fully captured by the SSM shape vector alone. Conclusion The requirements for this are two-fold; improving the accuracy of the shape representation methods while incorporating the head and neck vessels, and using a more real- istic CFD pipeline for generating training data with the inclusion of patient-specific boundary conditions. Comparison of the ML models against clinically acquired data (such as catheter- based pressure drops) should be also performed in the future for validation purposes. Conclusion In this proof-of-concept study, we have proposed a pipeline for building ML-based models to perform repetitive vessel-based CFD tasks. Generation of synthetic aortic training data by Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 16 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields means of shape modelling allowed ML techniques to be used, even where data scarcity is an issue (n = 67). Point correspondence was maintained between subject meshes in order to enable PCA. Our ML models were able to compute pressure and velocity flow fields much more rapidly (4,000x) than traditional CFD solvers, without large computational requirements or simulation setup. Comparison between predicted and ground truth test cases revealed good overall performance. Testing on prospective cases revealed that shape registration errors could produce misleading flow fields which deviated significantly from the ‘real’ CFD result, even in the presence of low ML errors. The approach described in this study is shape-driven and is applicable to any vascular structure which can be segmented from medical images. In the future, the models should be improved, so they can perform inference on prospective data from real patients. The requirements for this are two-fold; improving the accuracy of the shape representation methods while incorporating the head and neck vessels, and using a more real- istic CFD pipeline for generating training data with the inclusion of patient-specific boundary conditions. Comparison of the ML models against clinically acquired data (such as catheter- based pressure drops) should be also performed in the future for validation purposes. means of shape modelling allowed ML techniques to be used, even where data scarcity is an issue (n = 67). Point correspondence was maintained between subject meshes in order to enable PCA. Our ML models were able to compute pressure and velocity flow fields much more rapidly (4,000x) than traditional CFD solvers, without large computational requirements or simulation setup. Comparison between predicted and ground truth test cases revealed good overall performance. Testing on prospective cases revealed that shape registration errors could produce misleading flow fields which deviated significantly from the ‘real’ CFD result, even in the presence of low ML errors. The approach described in this study is shape-driven and is applicable to any vascular structure which can be segmented from medical images. In the future, the models should be improved, so they can perform inference on prospective data from real patients. Supporting information Supporting information S1 Fig. Volume mesh deformation: Mesh skewness. An example of template volume mesh deformation when generating the volume mesh for a new subject is shown. (EPS) S2 Fig. CFD sensitivity analysis. A test aortic shape with a sharp arch angulation was chosen to be used to perform a mesh sensitivity study. (EPS) S3 Fig. Best and worst PCA errors. PCA reconstructions in the test cohort (n = 200) were performed to assess the level of information loss due to dimensionality reduction. (EPS) S4 Fig. The final pressure and velocity DNN architectures. (EPS) S5 Fig. Regional Bland-Altman analysis. This was conducted on three different regions of the aorta (ascending, transverse arch, descending). (EPS) S6 Fig. PCA projections of real cases. All ten new cases had their first 35 shape mode scores plotted against the range of the original dataset scores. (EPS) Author Contributions Conceptualization: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphae¨l Sivera, Emilie Sauvage, Michael Quail, Silvia Schievano, Vivek Muthurangu. Data curation: Michael Quail. Formal analysis: Endrit Pajaziti. Investigation: Endrit Pajaziti. Methodology: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphae¨l Sivera, Silvia Schievano, Vivek Muthurangu. Supporting information S1 Fig. Volume mesh deformation: Mesh skewness. An example of template volume mesh deformation when generating the volume mesh for a new subject is shown. (EPS) S2 Fig. CFD sensitivity analysis. A test aortic shape with a sharp arch angulation was chosen to be used to perform a mesh sensitivity study. (EPS) S3 Fig. Best and worst PCA errors. PCA reconstructions in the test cohort (n = 200) were performed to assess the level of information loss due to dimensionality reduction. (EPS) S4 Fig. The final pressure and velocity DNN architectures. (EPS) S5 Fig. Regional Bland-Altman analysis. This was conducted on three different regions of the aorta (ascending, transverse arch, descending). (EPS) S6 Fig. PCA projections of real cases. All ten new cases had their first 35 shape mode scores plotted against the range of the original dataset scores. S5 Fig. Regional Bland-Altman analysis. This was conducted on three different regions of the aorta (ascending, transverse arch, descending). (EPS) S6 Fig. PCA projections of real cases. All ten new cases had their first 35 shape mode scores plotted against the range of the original dataset scores. (EPS) S p ( A C D F I M Methodology: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphae¨l Sivera, Silvia Schievano, Vivek Muthurangu. Supporting information Methodology: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphae¨l Sivera, Silvia Schievano, Vivek Muthurangu. Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 17 / 20 PLOS COMPUTATIONAL BIOLOGY Software: Endrit Pajaziti, Raphae¨l Sivera. Supervision: Claudio Capelli, Emilie Sauvage, Silvia Schievano, Vivek Muthurangu. Writing – original draft: Endrit Pajaziti. Writing – review & editing: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphae¨l Sivera, Emilie Sauvage, Michael Quail, Silvia Schievano, Vivek Muthurangu. References 1. Morris PD, Narracott A, Von Tengg-Kobligk H, Soto DAS, Hsiao S, Lungu A, et al. Computational fluid dynamics modelling in cardiovascular medicine. Heart. 2016; 102(1):18–28. https://doi.org/10.1136/ heartjnl-2015-308044 PMID: 26512019 2. Biglino G, Cosentino D, Steeden JA, De Nova L, Castelli M, Ntsinjana H, et al. Using 4D Cardiovascular Magnetic Resonance Imaging to Validate Computational Fluid Dynamics: A Case Study. Frontiers in Pediatrics. 2015; 3(December):1–10. https://doi.org/10.3389/fped.2015.00107 PMID: 26697416 3. Zhu Y, Chen R, Juan YH, Li H, Wang J, Yu Z, et al. Clinical validation and assessment of aortic hemody- namics using computational fluid dynamics simulations from computed tomography angiography. Bio- Medical Engineering Online. 2018; 17(1):1–12. https://doi.org/10.1186/s12938-018-0485-5 PMID: 29720173 4. LaDisa JF, Alberto Figueroa C, Vignon-Clementel IE, Jin Kim H, Xiao N, Ellwein LM, et al. Computa- tional simulations for aortic coarctation: representative results from a sampling of patients. Journal of Biomedical Engineering. 2011. https://doi.org/10.1115/1.4004996 PMID: 22010743 OGY Deep neural networks for fast aortic 3D pressure and velocity flow fields Deep neural networks for fast aortic 3D pressure and velocity flow fields Software: Endrit Pajaziti, Raphae¨l Sivera. Supervision: Claudio Capelli, Emilie Sauvage, Silvia Schievano, Vivek Muthurangu. Writing – original draft: Endrit Pajaziti. Writing – review & editing: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphae¨l Sivera, Emilie Sauvage, Michael Quail, Silvia Schievano, Vivek Muthurangu. References 1. Morris PD, Narracott A, Von Tengg-Kobligk H, Soto DAS, Hsiao S, Lungu A, et al. Computational fluid dynamics modelling in cardiovascular medicine. Heart. 2016; 102(1):18–28. https://doi.org/10.1136/ heartjnl-2015-308044 PMID: 26512019 2. Biglino G, Cosentino D, Steeden JA, De Nova L, Castelli M, Ntsinjana H, et al. Using 4D Cardiovascular Software: Endrit Pajaziti, Raphae¨l Sivera. References 1. Morris PD, Narracott A, Von Tengg-Kobligk H, Soto DAS, Hsiao S, Lungu A, et al. Computational fluid dynamics modelling in cardiovascular medicine. Heart. 2016; 102(1):18–28. https://doi.org/10.1136/ heartjnl-2015-308044 PMID: 26512019 2. Biglino G, Cosentino D, Steeden JA, De Nova L, Castelli M, Ntsinjana H, et al. Using 4D Cardiovascular Magnetic Resonance Imaging to Validate Computational Fluid Dynamics: A Case Study. Frontiers in Pediatrics. 2015; 3(December):1–10. https://doi.org/10.3389/fped.2015.00107 PMID: 26697416 3. Zhu Y, Chen R, Juan YH, Li H, Wang J, Yu Z, et al. Clinical validation and assessment of aortic hemody- namics using computational fluid dynamics simulations from computed tomography angiography. Bio- Medical Engineering Online. 2018; 17(1):1–12. https://doi.org/10.1186/s12938-018-0485-5 PMID: 29720173 3. Zhu Y, Chen R, Juan YH, Li H, Wang J, Yu Z, et al. Clinical validation and assessment of aortic hemody- namics using computational fluid dynamics simulations from computed tomography angiography. Bio- Medical Engineering Online. 2018; 17(1):1–12. https://doi.org/10.1186/s12938-018-0485-5 PMID: 29720173 4. LaDisa JF, Alberto Figueroa C, Vignon-Clementel IE, Jin Kim H, Xiao N, Ellwein LM, et al. Computa- tional simulations for aortic coarctation: representative results from a sampling of patients. Journal of Biomedical Engineering. 2011. https://doi.org/10.1115/1.4004996 PMID: 22010743 4. LaDisa JF, Alberto Figueroa C, Vignon-Clementel IE, Jin Kim H, Xiao N, Ellwein LM, et al. Computa- tional simulations for aortic coarctation: representative results from a sampling of patients. Journal of Biomedical Engineering. 2011. https://doi.org/10.1115/1.4004996 PMID: 22010743 5. Itu L, Sharma P, Ralovich K, Mihalef V, Ionasec R, Everett A, et al. Non-invasive hemodynamic assess- ment of aortic coarctation: validation with in vivo measurements. Annals of biomedical engineering. 2013; 41(4):669–81. https://doi.org/10.1007/s10439-012-0715-0 PMID: 23232558 5. Itu L, Sharma P, Ralovich K, Mihalef V, Ionasec R, Everett A, et al. Non-invasive hemodynamic assess- ment of aortic coarctation: validation with in vivo measurements. Annals of biomedical engineering. 2013; 41(4):669–81. https://doi.org/10.1007/s10439-012-0715-0 PMID: 23232558 6. Qiu Y, Wang J, Zhao J, Wang T, Zheng T, Yuan D. Association between blood flow pattern and rupture risk of abdominal aortic aneurysm based on computational fluid dynamics. European Journal of Vascu- lar and Endovascular Surgery. 2022; 64(2–3):155–64. https://doi.org/10.1016/j.ejvs.2022.05.027 PMID: 35605907 6. Qiu Y, Wang J, Zhao J, Wang T, Zheng T, Yuan D. Association between blood flow pattern and rupture risk of abdominal aortic aneurysm based on computational fluid dynamics. European Journal of Vascu- lar and Endovascular Surgery. 2022; 64(2–3):155–64. https://doi.org/10.1016/j.ejvs.2022.05.027 PMID: 35605907 7. References Deformetrica 4: an open-source software for statistical shape analysis. In: International Workshop on Shape in Medical Imaging. Springer; 2018. p. 3–13. 18. Piccinelli M, Veneziani A, Steinman DA, Remuzzi A, Antiga L. A framework for geometric analysis of vascular structures: application to cerebral aneurysms. IEEE transactions on medical imaging. 2009; 28 (8):1141–55. https://doi.org/10.1109/TMI.2009.2021652 PMID: 19447701 19. Hang S. TetGen, a Delaunay-based quality tetrahedral mesh generator. ACM Trans Math Softw. 2015; 41(2):11. 20. Garcia J, van der Palen RL, Bollache E, Jarvis K, Rose MJ, Barker AJ, et al. Distribution of blood flow velocity in the normal aorta: effect of age and gender. Journal of Magnetic Resonance Imaging. 2018; 47(2):487–98. https://doi.org/10.1002/jmri.25773 PMID: 28556277 21. Powell A, Maier S, Chung T, Geva T. Phase-velocity cine magnetic resonance imaging measurement of pulsatile blood flow in children and young adults: in vitro and in vivo validation. Pediatric cardiology. 2000; 21(2):104–10. https://doi.org/10.1007/s002469910014 PMID: 10754076 22. Bonfanti M, Balabani S, Greenwood JP, Puppala S, Homer-Vanniasinkam S, Dı´az-Zuccarini V. Compu- tational tools for clinical support: a multi-scale compliant model for haemodynamic simulations in an aor- tic dissection based on multi-modal imaging data. Journal of The Royal Society Interface. 2017; 14 (136):20170632. https://doi.org/10.1098/rsif.2017.0632 PMID: 29118115 23. Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprint arXiv:14126980. 2014. 23. Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprint arXiv:14126980. 2014. 24. Bergstra J, Bardenet R, Bengio Y, Ke´gl B. Algorithms for hyper-parameter optimization. Advances in neural information processing systems. 2011; 24. 25. Eiter T, Mannila H. Computing discrete Fre´chet distance. Technical Report CD-TR 94/64. 1994. 26. Creswell A, White T, Dumoulin V, Arulkumaran K, Sengupta B, Bharath AA. Generative adversarial net- works: An overview. IEEE signal processing magazine. 2018; 35(1):53–65. https://doi.org/10.1109/ MSP.2017.2765202 27. Ranjan A, Bolkart T, Sanyal S, Black MJ. Generating 3D faces using convolutional mesh autoencoders. In: Proceedings of the European Conference on Computer Vision (ECCV); 2018. p. 704–20. 28. Wang Y, Yao H, Zhao S. Auto-encoder based dimensionality reduction. Neurocomputing. 2016; 184:232–42. https://doi.org/10.1016/j.neucom.2015.08.104 29. Diller GP, Vahle J, Radke R, Vidal MLB, Fischer AJ, Bauer UM, et al. Utility of deep learning networks for the generation of artificial cardiac magnetic resonance images in congenital heart disease. BMC Medical Imaging. 2020; 20(1):1–8. https://doi.org/10.1186/s12880-020-00511-1 PMID: 33032536 30. Stein M. Large sample properties of simulations using Latin hypercube sampling. Technometrics. 1987; 29(2):143–51. https://doi.org/10.1080/00401706.1987.10488205 31. Abbas A, Rafiee A, Haase M, Malcolm A. References Huberts W, Heinen SGH, Zonnebeld N, van den Heuvel DAF, de Vries JPPM, Tordoir JHM, et al. What is needed to make cardiovascular models suitable for clinical decision support? A viewpoint paper. Jour- nal of Computational Science. 2018; 24:68–84. Available from: https://doi.org/10.1016/j.jocs.2017.07. 006 8. Avendi MR, Kheradvar A, Jafarkhani H. A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI. Medical image analysis. 2016; 30:108– 19. https://doi.org/10.1016/j.media.2016.01.005 PMID: 26917105 9. Yevtushenko P, Goubergrits L, Gundelwein L, Setio A, Heimann T, Ramm H, et al. Deep Learning Based Centerline-Aggregated Aortic Hemodynamics: An Efficient Alternative to Numerical Modelling of Hemodynamics. IEEE Journal of Biomedical and Health Informatics. 2021. 10. Feiger B, Gounley J, Adler D, Leopold JA, Draeger EW, Chaudhury R, et al. Accelerating massively par- allel hemodynamic models of coarctation of the aorta using neural networks. Scientific Reports. 2020; 10(1):1–13. Available from: https://doi.org/10.1038/s41598-020-66225-0 PMID: 32528104 11. Liang L, Mao W, Sun W. A feasibility study of deep learning for predicting hemodynamics of human tho- racic aorta. Journal of Biomechanics. 2020; 99. Available from: https://doi.org/10.1016/j.jbiomech.2019. 109544 PMID: 31806261 12. Hoeijmakers MJMM, Waechter-Stehle I, Weese J, Van de Vosse FN. Combining statistical shape modeling, CFD, and meta-modeling to approximate the patient-specific pressure-drop across the aortic valve in real-time. International Journal for Numerical Methods in Biomedical Engineering. 2020; 36 (10):1–18. https://doi.org/10.1002/cnm.3387 PMID: 32686898 13. Lee SW, Antiga L, Spence JD, Steinman DA. Geometry of the carotid bifurcation predicts its exposure to disturbed flow. Stroke. 2008; 39(8):2341–7. https://doi.org/10.1161/STROKEAHA.107.510644 PMID: 18556585 14. Antiga L, Piccinelli M, Botti L, Ene-Iordache B, Remuzzi A, Steinman DA. An image-based modeling framework for patient-specific computational hemodynamics. Medical & biological engineering & com- puting. 2008; 46(11):1097–112. https://doi.org/10.1007/s11517-008-0420-1 PMID: 19002516 15. Besl PJ, McKay ND. Method for registration of 3-D shapes. In: Sensor fusion IV: control paradigms and data structures. vol. 1611. Spie; 1992. p. 586–606. PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 18 / 20 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields 16. Bruse JL, McLeod K, Biglino G, Ntsinjana HN, Capelli C, Hsia TY, et al. A statistical shape modelling framework to extract 3D shape biomarkers from medical imaging data: assessing arch morphology of repaired coarctation of the aorta. BMC medical imaging. 2016; 16(1):1–19. https://doi.org/10.1186/ s12880-016-0142-z PMID: 27245048 17. Boˆne A, Louis M, Martin B, Durrleman S. References Geometric Convolutional Neural Networks–A Journey to Sur- rogate Modelling of Maritime CFD. In: The 9th Conference on Computational Methods in Marine Engi- neering (Marine 2021); 2022. p. 1–8. 32. Qi CR, Su H, Mo K, Guibas LJ. PointNet: Deep Learning on Point Sets for 3D Classification and Seg- mentation. CVPR 2017 Open Access. 2016. Available from: https://arxiv.org/abs/1612.00593 3 3 3 3 3 3 33. Umetani N, Bickel B. Learning Three-Dimensional Flow for Interactive Aerodynamic Design regression prediction for new shape. ACM Trans Graph. 2018; 37(4). Available from: https://doi.org/10.1145/ 3197517.3201325 34. Blundell C, Cornebise J, Kavukcuoglu K, Wierstra D. Weight uncertainty in neural network. In: Interna- tional conference on machine learning. PMLR; 2015. p. 1613–22. 35. Montalt-Tordera J, Pajaziti E, Jones R, Sauvage E, Puranik R, Singh AAV, et al. Automatic segmenta- tion of the great arteries for computational hemodynamic assessment. Journal of Cardiovascular Mag- netic Resonance. 2022; 24(1):1–14. https://doi.org/10.1186/s12968-022-00891-z PMID: 36336682 36. Tse KM, Chang R, Lee HP, Lim SP, Venkatesh SK, Ho P. A computational fluid dynamics study on geo- metrical influence of the aorta on haemodynamics. European Journal of Cardio-Thoracic Surgery. 2013; 43(4):829–38. https://doi.org/10.1093/ejcts/ezs388 PMID: 22766960 37. Wiputra H, Matsumoto S, Wagenseil JE, Braverman AC, Voeller RK, Barocas VH. Statistical shape representation of the thoracic aorta: accounting for major branches of the aortic arch. Computer meth- ods in biomechanics and biomedical engineering. 2022:1–15. https://doi.org/10.1080/10255842.2022. 2128672 PMID: 36165506 19 / 20 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 PLOS COMPUTATIONAL BIOLOGY Deep neural networks for fast aortic 3D pressure and velocity flow fields 38. Caballero AD, Laı´n S. A review on computational fluid dynamics modelling in human thoracic aorta. Car- diovascular Engineering and Technology. 2013; 4(2):103–30. https://doi.org/10.1007/s13239-013- 0146-6 39. Lantz J, Ebbers T, Engvall J, Karlsson M. Numerical and experimental assessment of turbulent kinetic energy in an aortic coarctation. Journal of biomechanics. 2013; 46(11):1851–8. https://doi.org/10.1016/ j.jbiomech.2013.04.028 PMID: 23746596 40. Youssefi P, Gomez A, Arthurs C, Sharma R, Jahangiri M, Alberto Figueroa C. Impact of patient-specific inflow velocity profile on hemodynamics of the thoracic aorta. Journal of biomechanical engineering. 2018; 140(1). https://doi.org/10.1115/1.4037857 PMID: 28890987 41. Morbiducci U, Ponzini R, Gallo D, Bignardi C, Rizzo G. Inflow boundary conditions for image-based computational hemodynamics: impact of idealized versus measured velocity profiles in the human aorta. Journal of biomechanics. 2013; 46(1):102–9. https://doi.org/10.1016/j.jbiomech.2012.10.012 PMID: 23159094 42. Thamsen B, Yevtushenko P, Gundelwein L, Setio A, Lamecker H, Kelm M, et al. References Synthetic Database of Aortic Morphometry and Hemodynamics: Overcoming Medical Imaging Data Availability. IEEE Trans- actions on Medical Imaging. 2021; 40(5):1438–49. https://doi.org/10.1109/TMI.2021.3057496 PMID: 33544670 43. Westerhof N, Lankhaar JW, Westerhof BE. The arterial windkessel. Medical & biological engineering & computing. 2009; 47(2):131–41. https://doi.org/10.1007/s11517-008-0359-2 PMID: 18543011 44. Romarowski RM, Lefieux A, Morganti S, Veneziani A, Auricchio F. Patient-specific CFD modelling in the thoracic aorta with PC-MRI–based boundary conditions: A least-square three-element Windkessel approach. International journal for numerical methods in biomedical engineering. 2018; 34(11):e3134. https://doi.org/10.1002/cnm.3134 PMID: 30062843 44. Romarowski RM, Lefieux A, Morganti S, Veneziani A, Auricchio F. Patient specific CFD modelling in the thoracic aorta with PC-MRI–based boundary conditions: A least-square three-element Windkessel approach. International journal for numerical methods in biomedical engineering. 2018; 34(11):e3134. https://doi.org/10.1002/cnm.3134 PMID: 30062843 45. Les AS, Shadden SC, Figueroa CA, Park JM, Tedesco MM, Herfkens RJ, et al. Quantification of hemo- dynamics in abdominal aortic aneurysms during rest and exercise using magnetic resonance imaging and computational fluid dynamics. Annals of biomedical engineering. 2010; 38(4):1288–313. https://doi. org/10.1007/s10439-010-9949-x PMID: 20143263 46. Madhavan S, Kemmerling EMC. The effect of inlet and outlet boundary conditions in image-based CFD modeling of aortic flow. Biomedical engineering online. 2018; 17(1):1–20. https://doi.org/10.1186/ s12938-018-0497-1 PMID: 29843730 47. Pirola S, Cheng Z, Jarral O, O’Regan D, Pepper J, Athanasiou T, et al. On the choice of outlet boundary conditions for patient-specific analysis of aortic flow using computational fluid dynamics. Journal of bio- mechanics. 2017; 60:15–21. https://doi.org/10.1016/j.jbiomech.2017.06.005 PMID: 28673664 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 20 / 20 45. Les AS, Shadden SC, Figueroa CA, Park JM, Tedesco MM, Herfkens RJ, et al. Quantification of hemo- dynamics in abdominal aortic aneurysms during rest and exercise using magnetic resonance imaging and computational fluid dynamics. Annals of biomedical engineering. 2010; 38(4):1288–313. https://doi. org/10.1007/s10439-010-9949-x PMID: 20143263 46. Madhavan S, Kemmerling EMC. The effect of inlet and outlet boundary conditions in image-based CFD modeling of aortic flow. Biomedical engineering online. 2018; 17(1):1–20. https://doi.org/10.1186/ s12938-018-0497-1 PMID: 29843730 47. Pirola S, Cheng Z, Jarral O, O’Regan D, Pepper J, Athanasiou T, et al. On the choice of outlet boundary conditions for patient-specific analysis of aortic flow using computational fluid dynamics. Journal of bio- mechanics. 2017; 60:15–21. https://doi.org/10.1016/j.jbiomech.2017.06.005 PMID: 28673664 PLOS Computational Biology | https://doi.org/10.1371/journal.pcbi.1011055 April 24, 2023 20 / 20
https://openalex.org/W2991095454
https://vestnik.guu.ru/jour/article/download/1800/951
Russian
null
USA AND WESTERN EUROPEAN SANCTIONS POLICY WITH RESPECT TO RUSSIA
Vestnik universiteta
2,019
cc-by
3,104
САНКЦИОННАЯ ПОЛИТИКА США И СТРАН ЗАПАДНОЙ ЕВРОПЫ В ОТНОШЕНИИ РОССИИ Аксенов Илья Антонович канд. экон. наук, ФГБОУ ВО  «Владимирский государственный университет имени Александра Григорьевича и Николая Григорь- евича Столетовых», г. Владимир, Российская Федерация ORCID: 0000-0003-0541-327X e-mail: il_aks@mail.ru Аксенов Илья Антонович канд. экон. наук, ФГБОУ ВО  «Владимирский государственный университет имени Александра Григорьевича и Николая Григорь- евича Столетовых», г. Владимир, Российская Федерация ORCID: 0000-0003-0541-327X e-mail: il_aks@mail.ru Аннотация. Проведен анализ санкционной политики США и стран Западной Европы в от- ношении Российской Федерации. Выявлены цели санкционной политики США и Европейского союза. Исследование построено на принципах трехуровневой классификации санкций, направ- ленных на дестабилизацию экономики Российской Федерации. Выявлены отрасли экономи- ки и компании, которые стали объектами санкционной политики и пострадали от санкций США и Европейского союза в большей мере. Определена роль санкций как инструментария, который способен воздействовать на политическую и экономическую расстановку сил в мире. В рамках статьи отражено мнение представителей бизнеса зарубежных стран по вопро- су формирования антироссийских санкций. Отмечены особенности проведения санкционной политики в «цифровом блоке» и киберпространстве. Ключевые слова: санкции, экономическое развитие, киберпространство, эмбарго, ограничение, дестабилизация, дипломатические отношения. Цитирование: Аксенов И.А. Санкционная политика США и стран Западной Европы в отношении России//Вестник университета. 2019. № 10. С. 147-151. Aksenov Iliya Candidate of Economic Sciences, Vladimir State University named after Alexander and Nikolay Stoletovs, Vladimir, Russia ORCID: 0000-0003-0541-327X e-mail: il_aks@mail.ru DOI 10.26425/1816-4277-2019-10-147-151 УДК338 JEL F10, H11, L10 Экономика: проблемы, решения и перспективы Экономика: проблемы, решения и перспективы Вестник университета № 10, 2019 Вестник университета № 10, 2019 Со стороны западных стран были выделены те отрасли российской экономики, которые «должны были понести наказание». Со стороны западных стран были выделены те отрасли российской экономики, которые «должны были По сути, санкции США и стран Западной Европы против Российской Федерации (далее – РФ) прошли через три уровня: сначала были применены дипломатические санкции, затем санкции в отношении физиче- ских лиц, а, в конечном, счете, были введены экономические санкции, которые оказались наиболее дорого- стоящими для обеих сторон [2]. Первый уровень санкций – дипломатические, были введены в марте-апреле 2014 г. Ограничения заклю- чались в том, что западные страны приостановили: –– упрощенный визовый режим; –– договора по партнерским отношениям между странами; –– договора по взаимным инвестициям. Дипломатические санкции последовали также и со стороны некоторых международных организаций. Так, Организация экономического сотрудничества и развития прервала процесс присоединения России, НАТО приостановлено практически все гражданское и военное сотрудничество с Россией, G8 перешла к формату G7 из-за исключения России из стран-участниц, а Парламентская ассамблея Совета Европы (ПАСЕ) замо- розила права российской делегации в ассамблее, в том числе и право голоса. Второй уровень санкций – санкции в отношении отдельных лиц и организаций, были приняты в 2014- 2015 г., но частично истекли в сентябре 2016 г. Санкции этого уровня наложили запрет на визы и привели к заморозке активов отдельных российских и украинских лиц и организаций. первые ограничения в отно- шении 119 физических и 23 юридических лиц, установившие запрет на покупку у них и продажу им това- ров, управление их имуществом, выплату дивидендов, были введены 14 марта 2014 г. Второй пакет санк- ций, принятый 20 марта 2014 г., запретил инвестиции в инфраструктуру Крыма. Третий уровень санкций – экономические санкции, принятые в июле-сентябре 2014 года. Частично санк- ции истекли в июле 2016 г., но большая их часть была продлена. Все виды санкций были введены США, а затем поддержаны и усилены со стороны стран Западной Европы. Так, санкции оказались направлены на топливно-энергетический, оборонно-промышленный и банковский секторы. Отличительная особенность антироссийских санкций 2014 г. состоит в том, что они направлены не полностью на всю отрасль экономи- ки, а на ее самую значимую для РФ часть. В случае с банковским сектором санкции ударили только по тем, которые сбыли созданы с государственным участием [4]. Банки, попавшие в санкционный список, нацеле- ны на ведение фундаментальных, долгосрочных системных макроэкономических проектов не только в РФ, но за рубежом. USA AND WESTERN EUROPEAN SANCTIONS POLICY WITH RESPECT TO RUSSIA Abstract. The implementation of the sanctions policy of the United States and Western Europe in relation to the Russian Federation has been analyzed. The objectives of the USA and Europe- an Union sanctions policies have been identified. The study has been based on the principles of a three-level classification of sanctions aimed at destabilizing the economy of the Russian Federa- tion. Sectors of the economy and companies have been identified, that, have become objects of sanc- tions policy and have suffered more from USA and European Union sanctions. The role of sanc- tions as a toolkit, which can influence the political and economic balance of power in the world, has been defined. The opinion of representatives of business of foreign countries on the formation of anti-Russian sanctions has been reflected within the article. The features of the implementation of the sanctions policy in the digital block and cyberspace have been noted. Keywords: sanctions, economic development, cyberspace, embargo, restriction, destabilization, diplomatic relations. For citation: Aksenov I.A. USA and Western European sanctions policy with respect to Russia (2019) Vestnik universiteta, I. 10, pp. 147-151. doi: 10.26425/1816-4277-2019-10-147-151 Политика антироссийской санкционной политики со стороны США и стран Западной Европы началась в 2014 г. в связи с украинскими событиями. Официальной причиной введения санкций против России стало ее вмешательство в политическую жизнь Украины [5]. Россия пыталась привлечь западные страны для уре- гулирования политического конфликта на Украине, но безуспешно. США и страны Западной Европы поэ- тапно развернули антироссийскую санкционную политику. Официальная цель санкций заключалась в том, чтобы Россия изменила свою политику в отношении Ук- раины, хотя многие эксперты сходятся во мнении, что основная цель санкций – политическая и экономиче- ская дестабилизация ситуации в России [6]. США и страны Европы трактуют цель антироссийской поли- тики по-разному. Европейский союз (далее – ЕС) при формировании санкционной политики ставил перед собой цель – разрешение украинского кризиса, путем давления на Россию экономическими и политически- ми инструментами. США при формировании санкционной политики руководствовались принципом наказа- ния России за незаконное присоединение Крыма. © Аксенов И.А., 2019. Статья доступна по лицензии Creative Commons «Attribution» («Атрибуция») 4.0. всемирная (http://creativecommons.org/licenses/by/4.0/). s), 2019. This is an open access article under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/). 147 Вестник университета № 10, 2019 В структуре активов этих банков большая часть принадлежит именно долгосрочным акти- вам, на ограничение которых как раз и нацелена западная санкционная политика. Предполагаемым эффек- том со стороны США и стран Западной Европы являются заморозка проектов на неопределенный срок или их полное исключение. Также осенью 2014 г. странами Европы предлагались более радикальные меры, на- правленные на блокирование российских банков от международной системы финансовых переводов SWIFT, но такие шаги так и не были предприняты. Объектом санкций в нефте- и газодобывающем секторе стали такие компании, как «Газпром», «Роснефть», «Газпромнефть», «Транснефть», «Лукойл» и «Сургутнефтегаз», а в оборонном секторе – Государственная корпорация «Ростехнологии» и входящие в нее компании, ОАО Концерн «Калашников», Объединенная ави- астроительная корпорация и др. В декабре 2014 г. сформирован новый пакет санкций США и стран Запад- ной Европы. Эти санкции коснулись в большей степени территории Крыма и запрещали экспорт, импорт, а также инвестиционные потоки. На поставки оружия было наложено эмбарго: запрещалось вести торговлю оружием с РФ, а также экс- портировать предметы двойного назначения военным клиентам. Ограничения в сотрудничестве с россий- ским энергетическим сектором были следующие: запрет на экспорт инновационных добывающих технологий, а также таких услуг, как бурение и тестирование, используемых российскими компаниями для разработки глубоководных, арктических и сланцевых запасов нефти. Все другие экспортные поставки энергии требу- ют специального одобрения. Санкции со стороны ЕС были продлены 21 декабря 2015 г. до 31 июля 2016 г. Сложившаяся ситуация как в России, так и за ее пределами, предопределила новые политические и эко- номические реалии, в которых пришлось существовать всем странам [3]. Именно поэтому многие страны, 148 Экономика: проблемы, решения и перспективы включая страны ЕС, были вынуждены поддержать политику США относительно антироссийских санкций. В СМИ на тот момент можно было заметить мнение о том, что США оказывало воздействие на страны Ев- ропы по поводу применения санкций против РФ. включая страны ЕС, были вынуждены поддержать политику США относительно антироссийских санкций. В СМИ на тот момент можно было заметить мнение о том, что США оказывало воздействие на страны Ев- ропы по поводу применения санкций против РФ. Поводом для ужесточения санкций против России стала катастрофа с Boeing 777 в Донецкой области 17 июля 2014 г., поскольку, по мнению ряда западных стран, крушение самолета было связано с действия- ми повстанцев, поддерживаемых РФ. Через год после введения санкций мнения европейских стран об антироссийской политике разделились: одна часть стран считала, что продлевать санкции не стоит, поскольку они не оказывают ожидаемого эффек- та, другая часть выступала за продление и преумножение санкций против России. Вестник университета № 10, 2019 Сторонники отмены санк- ций придерживались мнения, что целью вводимых санкций было не обрушение российской экономики, а из- менение политики РФ в отношении Украины. Санкции этой группой стран были признаны очень затратными, от применения которых потери несут обе стороны. Сторонники продления санкций приняли условие их отме- ны: соблюдение Россией Минских договоренностей. Но действительность оказалось таковой, что продление или приостановление санкций зависят от действий украинской стороны, которые носят ярко выраженный про- американский характер. В связи с этим такое положение дел дает возможность Европе и США манипулировать такими рычагами воздействия на российскую экономик, как санкции, которые продлеваются уже не первый раз. В июле 2015 г. США и ЕС ввели новые экономические санкции, направленные на ключевые отрасли эко- номики России, а именно государственные финансовые институты и энергетику, финансирование государст- венных банков со стороны США и стран Европы и предоставление технологий энергетическим предприятиям. В самом начале ведения антироссийской санкционной политики представители бизнеса зарубежных стран отзывались резко негативно об антироссийской политике в силу того, что, по их мнению, нельзя ре- шать политические вопросы за счет отечественных производителей. Поэтому американские компании опа- сались того, что их долю на рынке России займут европейские и китайские конкуренты, а компании евро- пейских стран считали своими конкурентами компании из стран Персидского залива и Китая. Но некоторые страны и, в том числе, Великобритания предоставили право своим компаниями самостоятельно выбирать бизнес-политику, поэтому, например, были все-таки реализованы соглашения между британской нефтяной компанией ВР и «Роснефтью». Чуть меньше свободы было предоставлено немецким промышленным ком- паниям, которые активно боролись за стабильное ведение бизнеса на территории России. В самом начале ведения антироссийской санкционной политики представители бизнеса зарубежных стран отзывались резко негативно об антироссийской политике в силу того, что, по их мнению, нельзя ре- шать политические вопросы за счет отечественных производителей. Поэтому американские компании опа- сались того, что их долю на рынке России займут европейские и китайские конкуренты, а компании евро- пейских стран считали своими конкурентами компании из стран Персидского залива и Китая. Но некоторые страны и, в том числе, Великобритания предоставили право своим компаниями самостоятельно выбирать бизнес-политику, поэтому, например, были все-таки реализованы соглашения между британской нефтяной компанией ВР и «Роснефтью». Чуть меньше свободы было предоставлено немецким промышленным ком- паниям, которые активно боролись за стабильное ведение бизнеса на территории России. Вплоть до 2019 г. против РФ расширялись, продлевались и вводились новые дипломатические, экономи- ческие и санкции против физических и юридических лиц. Библиографический список 1. Федеральный закон «О мерах воздействия (противодействия) на недружественные действия Соединенных Штатов Аме- рики и иных иностранных государств» от 04.06.2018 г. № 127-ФЗ // СПС «Консультант Плюс» [Электронный ресурс]. – Режим доступа: http://www.consultant.ru/document/cons (дата обращения: 20.08.2019). 2. Алешин, Д. В. Влияние санкций 2014-2016 гг. на экономику России//Финансы, налоги и учет в странах дальнего и ближ- него зарубежья: инновационные решения. – 2017. – № 1. – С. 120-125. 3. Алиев, Ф. Б. Антироссийские санкции – неправовые формы международно-правового воздействия//Современный уче- ный. – 2017. – № 6. – С. 213-219. 4. Казанцев, С.В. Антироссийские санкции – вчера и сегодня//ЭКО. – 2015. – № 3. – С. 23-29. 5. Спиридонов, В. В. К вопросу об антироссийских санкциях со стороны ведущих государств Западного мира В. В. Спи- ридонов//Геополитический журнал. – 2014. – № 7. – С. 56-63. 6. Травина, Л. А., Катушенко, С. А. Правомерность антироссийских санкций и ответных мер в рамках членства во Всемир- ной торговой организации//Актуальные проблемы экономики и права. – 2016. – № 4. – С. 135-141. Вестник университета № 10, 2019 На сегодняшний день в санкционный список вклю- чены 385 физических лиц: российские политики, чиновники, военные и крупнейшие российские бизнесме- ны, а также 511 организаций. Санкции против энергетического сектора продолжают логику ранее введенных санкций в отношении нефтяного сектора экономики РФ. При этом ограничения не касаются газового сектора, но разведка глубоководных и шельфовых месторождений, а также сланцевой нефти находится под санкциями. «Цифровой блок» санкций предусматривает наказание к любой российской информационной компании, которая будет осуществлять враждебные действия по отношению к США в киберпространстве. Оборон- ные санкции расширились введением пакета санкций от 27 августа 2018 г., которые включают ограничения по торговле оборонной продукцией и вооружением российского производства. Со стороны стран ЕС про- длены санкции в отношении физических и юридических лиц. Кроме того, 22 мая 2018 г. был принят Феде- ральный закон от 04.06.2018 № 127-ФЗ «О мерах воздействия (противодействия) на недружественные дей- ствия Соединенных Штатов Америки и иных иностранных государств» [1]. Таким образом, новый этап антироссийских запретов и ограничений со стороны США и стран Западной Европы связан с политикой, проводимой РФ в отношении Украины в 2014 г. и продолжающейся до сих пор. Современные антироссийские санкции складываются из трех уровней: дипломатические санкции, санкции против физических лиц и экономические санкции. Экономические санкции направлены на наиболее важ- ные отрасли российской экономики: финансово-банковский, нефте- и газодобывающий и оборонно-про- мышленный секторы. Введенные запреты и ограничения привели к введению контрмер российской сторо- ной, направленных, главным образом, на обеспечение продовольственной безопасности страны. Санкции и контрсанкции действуют и по сей день. Такая ситуация повлекла за собой кардинальные структурные и ко- личественные изменения в экономике России и во внешнеполитическом диалоге государств друг с другом. 149 Вестник университета № 10, 2019 Вестник университета № 10, 2019 На основе выявленных фактов, можно констатировать введенные санкции в отношении РФ, безуслов- но, нанесли ущерб конкретным отраслям и российской экономике в целом. Но реализация антироссийской санкционной политики не в полной мере оказала то воздействие на Российскую экономику, которое пред- полагалось странами, вводящими санкции. Такие результаты свидетельствуют о неоднозначности эффекта антироссийской санкционной политики и принимаемых ответных мер со стороны России. Используемые запреты и ограничения имеют, в конечном счете, как положительные, так и отрицатель- ные последствия. К числу положительных эффектов антироссийской санкционной политики можно отнести: –– выход на новые внешние рынки; –– повышение конкурентоспособности экспортеров на мировом рынке; –– укрепление связей с новыми торговыми партнерами; –– развитие и усиление позиций в агропромышленном комплексе; –– снижение зависимости от иностранных поставок. К числу отрицательных эффектов антироссийской санкционной политики можно отнести: –– девальвация рубля; –– разрыв связей с внешнеэкономическими партнерами; –– снижение покупательной способности; –– потеря иностранных инвестиций; –– снижение темпов роста валового внутреннего продукта. Санкции оказали значительное влияние на экономику Российской Федерации. Политика России, осно- ванная на идее импортозамещения, способствовала развитию, и с 2017 г. российская экономика начала вос- станавливаться после спадов в 2015-2016 гг. В настоящее время рост показателей российской экономики сохраняется. При этом политика импортозамещения не может оставаться единственным инструментом в обес- печении устойчивого развития экономики. Внимание должно быть уделено также таким моментам, как нор- мализация сотрудничества с другими государствами и объединениями, восстановление устойчивости кур- са рубля, а также развитие несырьевых отраслей. Главным образом, уход России от сырьевой экономики и налаживание внешнеэкономических связей позволит добиться стабилизации и экономического прогресса. 3. Aliev F. B., Antirossiiskie sanktsii – ne pravovye formy mezhdunarodno-pravovogo vozdeistviya [Anti-Russian sanctions are not legal forms of international legal influence], Sovremennyi uchenyi [Modern Scientist], 2017, I. 6, pp. 213-219. 6. Travina L.A., Katushenko S.A. Pravomernost’ antirossiiskikh sanktsii i otvetnykh mer v ramkakh chlenstva vo Vsemirnoi torgovoi organizatsii [The legality of anti-Russian sanctions and retaliation within the framework of membership in the World Trade Organization], Aktual’nye problemy ekonomiki i prava [Actual Problems of Economy and Law], 2016, I. 4, pp. 135-141. 5. Spiridonov V. V. K voprosu ob antirossiiskikh sanktsiyakh so storony vedushchikh gosudarstv Zapadnogo mira [On the issue of an- ti-Russian sanctions by the leading states of the Western world], Geopoliticheskii zhurnal [Geopolitical Journal], 2014, I. 7. pp. 56-63. 3. Aliev F. B., Antirossiiskie sanktsii – ne pravovye formy mezhdunarodno-pravovogo vozdeistviya [Anti-Russian sanctions are not legal forms of international legal influence], Sovremennyi uchenyi [Modern Scientist], 2017, I. 6, pp. 213-219. 4. Kazantsev S. V. Antirossiiskie sanktsii – vchera i segodnya [Anti-Russian sanctions – yesterday and today], EKO, 2015, I. 3, pp. 23-29. 5. Spiridonov V. V. K voprosu ob antirossiiskikh sanktsiyakh so storony vedushchikh gosudarstv Zapadnogo mira [On the issue of an- ti-Russian sanctions by the leading states of the Western world], Geopoliticheskii zhurnal [Geopolitical Journal], 2014, I. 7. pp. 56-63. 6. Travina L.A., Katushenko S.A. Pravomernost’ antirossiiskikh sanktsii i otvetnykh mer v ramkakh chlenstva vo Vsemirnoi torgovoi organizatsii [The legality of anti-Russian sanctions and retaliation within the framework of membership in the World Trade Organization], Aktual’nye problemy ekonomiki i prava [Actual Problems of Economy and Law], 2016, I. 4, pp. 135-141. g f f g fl y y 4. Kazantsev S. V. Antirossiiskie sanktsii – vchera i segodnya [Anti-Russian sanctions – yesterday and today 4. Kazantsev S. V. Antirossiiskie sanktsii – vchera i segodnya [Anti-Russian sanctions – yesterday and today], EKO, 2015, I. 3, pp. 23-29. 4. Kazantsev S. V. Antirossiiskie sanktsii vchera i segodnya [Anti Russian sanctions yesterday and today], EKO, 2015, I. 3, pp. 23 29. 5. Spiridonov V. V. K voprosu ob antirossiiskikh sanktsiyakh so storony vedushchikh gosudarstv Zapadnogo mira [On the issue of an- ti-Russian sanctions by the leading states of the Western world], Geopoliticheskii zhurnal [Geopolitical Journal], 2014, I. 7. pp. 56-63. 6 T i L A K t h k S A P t’ ti ii kikh kt ii i t t kh k kh hl t V i i 4. Kazantsev S. V. Antirossiiskie sanktsii – vchera i segodnya [Anti-Russian sanctions – yesterday and today], EKO, 2015, I. 3, pp. 23-29. 5. Spiridonov V. V. K voprosu ob antirossiiskikh sanktsiyakh so storony vedushchikh gosudarstv Zapadnogo mira [On the issue of an- ti-Russian sanctions by the leading states of the Western world], Geopoliticheskii zhurnal [Geopolitical Journal], 2014, I. 7. pp. 56-63. 6. Travina L.A., Katushenko S.A. Pravomernost’ antirossiiskikh sanktsii i otvetnykh mer v ramkakh chlenstva vo Vsemirnoi torgovoi organizatsii [The legality of anti-Russian sanctions and retaliation within the framework of membership in the World Trade Organization], Aktual’nye problemy ekonomiki i prava [Actual Problems of Economy and Law], 2016, I. 4, pp. 135-141. References 1. Federal’nyi zakon «O merakh vozdeistviya (protivodeistviya) na nedruzhestvennye deistviya Soedinennykh Shtatov Ameriki i inykh inostrannykh gosudarstv» ot 04.06.2018 g. № 127-FZ [Federal Law «On Measures of Impact (Counteraction) on Un- friendly Actions of the United States of America and Other Foreign States» dated on June 04, 2018 No. 127-FZ], SPS «Konsultant Plyus» [Legal reference system «Consultant Plus»]. Available at: http://www.consultant.ru/document/cons (accessed 20.08.2019). 2. Aleshin D. V. Vliyanie sanktsii 2014-2016 gg. na ekonomiku Rossii [The impact of the 2014-2016 sanctions. on the Russian economy], Finansy, nalogi i uchet v stranakh dal’nego i blizhnego zarubezh’ya: innovatsionnye resheniya [Finance, taxes and accounting in the far and near abroad: innovative solutions], 2017, I. 1, pp. 120-125. 150 Экономика: проблемы, решения и перспективы 151
https://openalex.org/W1689234104
https://dash.harvard.edu/bitstream/1/16120983/1/4439021.pdf
English
null
Surveillance for Neisseria meningitidis Disease Activity and Transmission Using Information Technology
PloS one
2,015
cc-by
11,069
Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA Permanent link http://nrs.harvard.edu/urn-3:HUL.InstRepos:16120983 Citation Ahmed, S. Sohail, Ernesto Oviedo-Orta, Sumiko R. Mekaru, Clark C. Freifeld, Gervais Tougas, and John S. Brownstein. 2015. “Surveillance for Neisseria meningitidis Disease Activity and Transmission Using Information Technology.” PLoS ONE 10 (5): e0127406. doi:10.1371/ journal.pone.0127406. http://dx.doi.org/10.1371/journal.pone.0127406. Published Version doi:10.1371/journal.pone.0127406 Published Version doi:10.1371/journal.pone.0127406 Methods Copyright: © 2015 Ahmed et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Annual analyses of N. meningitides disease alerts captured by HealthMap were compared to epidemiological data captured by the Centers for Disease Control’s Active Bacterial Core surveillance (ABCs) for N. meningitides. Morbidity and mortality case reports were mea- sured annually from 2010 to 2013 (HealthMap) and 2005 to 2012 (ABCs). Share Your Story The Harvard community has made this article openly available. Please share how this access benefits you. Submit a story . Accessibility RESEARCH ARTICLE Background While formal reporting, surveillance, and response structures remain essential to protecting public health, a new generation of freely accessible, online, and real-time informatics tools for disease tracking are expanding the ability to raise earlier public awareness of emerging disease threats. The rationale for this study is to test the hypothesis that the HealthMap in- formatics tools can complement epidemiological data captured by traditional surveillance monitoring systems for meningitis due to Neisseria meningitides (N. meningitides) by highlighting severe transmissible disease activity and outbreaks in the United States. Academic Editor: Chris T. Bauch, University of Waterloo, CANADA Academic Editor: Chris T. Bauch, University of Waterloo, CANADA Academic Editor: Chris T. Bauch, University of Waterloo, CANADA Received: December 22, 2014 Accepted: April 15, 2015 Published: May 20, 2015 Copyright: © 2015 Ahmed et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Academic Editor: Chris T. Bauch, University of Waterloo, CANADA Received: December 22, 2014 Accepted: April 15, 2015 Published: May 20, 2015 Copyright: © 2015 Ahmed et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. S. Sohail Ahmed1¤*, Ernesto Oviedo-Orta1¤, Sumiko R. Mekaru2, Clark C. Freifeld2, Gervais Tougas3, John S. Brownstein2,4 S. Sohail Ahmed1¤*, Ernesto Oviedo-Orta1¤, Sumiko R. Mekaru2, Clark C. Freifeld2, Gervais Tougas3, John S. Brownstein2,4 1 Global Clinical Sciences, Novartis Vaccines Srl, Siena, Italy, 2 Children’s Hospital Informatics Program at the Harvard-Massachusetts Institute of Technology Division of Health Sciences and Technology, Computational Epidemiology Group, Boston, Massachusetts, United States of America, 3 Drug Safety and Epidemiology, Novartis Pharma AG, Basel, Switzerland, 4 Department of Pediatrics, Harvard Medical School, Boston Children’s Hospital, Boston, Massachusetts, United States of America a1111 ¤ Current address: Global Clinical Sciences, GSK, Siena, Italy * sohail.q.ahmed@gsk.com OPEN ACCESS OPEN ACCESS Citation: Ahmed SS, Oviedo-Orta E, Mekaru SR, Freifeld CC, Tougas G, Brownstein JS (2015) Surveillance for Neisseria meningitidis Disease Activity and Transmission Using Information Technology. PLoS ONE 10(5): e0127406. doi:10.1371/journal.pone.0127406 Academic Editor: Chris T. Bauch, University of Waterloo, CANADA Received: December 22, 2014 Accepted: April 15, 2015 Published: May 20, 2015 Copyright: © 2015 Ahmed et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Citation: Ahmed SS, Oviedo-Orta E, Mekaru SR, Freifeld CC, Tougas G, Brownstein JS (2015) Surveillance for Neisseria meningitidis Disease Activity and Transmission Using Information Technology. PLoS ONE 10(5): e0127406. doi:10.1371/journal.pone.0127406 ¤ Current address: Global Clinical Sciences, GSK, Siena, Italy * sohail.q.ahmed@gsk.com Neisseria meningitidis and HealthMap Conclusions The specific roles of these authors are articulated in the “author contributions” section. The underreporting of disease cases in internet-based data streaming makes inadequate any comparison to epidemiological trends illustrated by the more comprehensive ABCs net- work published by the Centers for Disease Control. However, the expected delays in com- piling confirmatory reports by traditional surveillance systems (at the time of writing, ABCs data for 2013 is listed as being provisional) emphasize the helpfulness of real-time internet- based data streaming to quickly fill gaps including the visualization of modes of disease transmission in outbreaks for better resource and action planning. HealthMap can also con- tribute as an internet-based monitoring system to provide real-time channel for patients to report intervention-related failures. Competing Interests: Drs. Brownstein and Freifeld report being co-creators of the HealthMap system and hold an equity stake in, and are consultants to Epidemico, a spinoff company of Boston Children’s Hospital that handles licensing of HealthMap data. Dr. Mekaru declares no conflicts of interest. Drs. Ahmed, Oviedo-Orta, and Tougas are employees of Novartis and hold stock in Novartis AG. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. Findings HealthMap N. meningitides monitoring captured 80-90% of alerts as diagnosed N. menin- gitides, 5-20% of alerts as suspected cases, and 5-10% of alerts as related news articles. HealthMap disease alert activity for emerging disease threats related to N. meningitides were in agreement with patterns identified historically using traditional surveillance systems. HealthMap’s strength lies in its ability to provide a cumulative “snapshot” of weak signals that allows for rapid dissemination of knowledge and earlier public awareness of potential outbreak status while formal testing and confirmation for specific serotypes is ongoing by public health authorities. Funding: Drs. Brownstein, Mekaru, and Freifeld are supported by grant 5 R01 LM010812-04 from the National Library of Medicine. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No other sources of funding have supported this work. Novartis provided support in the form of salaries for authors (SSA, EOO, GT), but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 1 / 17 Introduction Neisseria meningitidis (N. meningitidis) is a human pathogen that spreads between individuals through direct contact with respiratory secretions [1]. The bacteria can cause septicemia, pneu- monia, and meningitis. N. meningitidis strains can be categorized, based on the composition of the bacteria’s capsule, into 13 different serogroups of which five are responsible for the great majority of invasive disease (serogroups A, B, C, W, and Y) [2]. Up to 14% of meningococcal infected individuals die [2,3] and among the survivors, up to 20% have serious sequelae includ- ing hearing loss, limb amputation, and neurologic disability [4]. Due to the disease’s rapid pro- gression (< 24–48 hours), serious sequelae, and potential for person-to-person spread, a single case of meningococcal disease will trigger immediate and increasing media attention during which public health officials are working to avoid panic, to identify whether there is an out- break, and to generate an effective response to treat disease and prevent further transmission. Recently, two United States (US) universities have experienced serogroup B strain outbreaks that were widely reported by both traditional media and internet-based news reporting. The widespread adoption of more sophisticated information technology has increased public awareness of potentially serious diseases as seen during the emergence and global spread of the 2009 pandemic influenza A/H1N1 virus [5]. While formal reporting, surveillance, and re- sponse structures remain essential to protecting public health, a new generation of online, free- ly accessible, and real-time disease surveillance-monitoring tools have become available. These tools complement and expand existing public health professionals’ capacities to detect weak signals and to raise earlier concerns regarding emerging diseases [6–8]. In this report, we used the HealthMap informatics tool [9] to investigate the burden of “meningitis” that can be cap- tured in the US. We aim to test the hypothesis that HealthMap complements data captured by traditional surveillance and provides insights related to the mode of disease transmission linked to emerging outbreaks. Methods HealthMap HealthMap (http://www.healthmap.org/) is a multistream real-time surveillance platform that continually aggregates reports on new and ongoing infectious disease outbreaks. The system performs extraction, categorization, filtration, and integration of these reports, facilitating knowledge management and earlier awareness of potential outbreak status [10]. 2 / 17 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Neisseria meningitidis and HealthMap HealthMap system capabilities. The HealthMap automated text processing system pro- ceeds in four stages: (1) Acquisition (2) Characterization (3) Filtering, and (4) Clustering. Trained public health analysts are employed to review content and correct misclassifications. Finally, the correctly classified content is used to improve the automated algorithms through a training feedback loop. The first step is data acquisition. The system collects data via five main channels: news aggregators (such as Google News, Yahoo News, Moreover); specific Real-time Simple Syndi- cation (RSS) feeds including WHO (World Health Organization), OIE (Office International des Epizooties/World Organisation for Animal Health), and others; e-mail subscriptions (GPHIN [Global Public Health Intelligence Network (developed by Health Camada)], ProMED, etc); custom-parsed HTML (HyperText Markup Language) scraping or CSV (Comma-separated Values)/XML (Extensible Markup Language) feeds; and user submitted re- ports. The framework allows one to add a new feed through a custom PHP (Personal Home Page Tools) Hypertext Preprocessor class where the feed-specific code fetches the content, parses it, and then extracts a set of common fields. In the second step, the common outputs of the feed framework then pass to the characteri- zation engine where we extract disease and location entities, flag the report as not-disease-relat- ed by rule (if applicable), and identify the source publication. This module loads large dictionaries of terms into memory and then matches them against the input text using a rapid matching algorithm. Algorithm performance is linear on the size of the input text and memory consumption is linear on the size of the dictionary. Under some conditions, the characteriza- tion module may make calls back to the acquisition module to request additional input text from the source. The dictionaries currently support identification of about 250 diseases. HealthMap has an internal location database of over 25,000 locations. The characterization en- gine generally assigns countries or first level administrative units; analysts generally change these to cities, hospitals, or schools, if that level of specificity is warranted. Once disease and location are assigned, the document then passes to a document-level Bayesian filtering module (third step). PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Methods HealthMap The current list of tags includes such diverse topics as schools, healthcare facilities, sports-associated, recall (of a product), and cruise. HealthMap data sources. HealthMap uses a large variety of inputs to detect outbreak news displayed on the main web page, to produce password protected sites for partners with sensitive data, and to do research on novel disease surveillance methods. The standard data inputs for the main webpage are: news aggregators (such as Google News, Yahoo News, Baidu), specific RSS feeds (WHO, OIE, FAO), constructed datasets of lan- guage-specific RSS feeds, email subscriptions (GPHIN, ProMED, etc), custom-parsed HTML scraping or CSV/XML feeds, and user submitted reports. HealthMap queries news aggregators at regular intervals with a list of queries designed to return outbreak news stories with excellent sensitivity while avoiding poor specificity. Post-collection processing ensures non-outbreak re- lated articles are not displayed on the map. Specific RSS feeds and e-mail subscriptions allow the system to receive high quality data from known providers. Custom-parsed scraping pro- grams allow the system to retrieve data that is otherwise difficult to access. User-submitted reports represent a diverse group of resources. Users may submit URLs or eyewitness descriptions of outbreaks to HealthMap via the website, e-mail, telephone, text mes- sage or the Outbreaks Near Me mobile app for Android and iPhone. These reports are pro- cessed through various means depending on the method of delivery. Social media sources are also included in HealthMap. Twitter and Facebook data are collect- ed and processed by the parser. The short nature of tweets does present challenges to automat- ed categorization, but enhanced dictionaries have helped to address that issue. HealthMap subject matter domains. Although HealthMap was developed specifically to provide a comprehensive real time display of infectious disease outbreak knowledge, the system has proven flexible and robust enough to support many other projects. HealthMap has been used in diverse government-sponsored and academic research-oriented projects including mon- itoring violence against civilians in the Syrian conflict, and enhancing situation awareness for re- sponse efforts for a natural disaster. The core architecture and processing capabilities are language and subject matter independent. HealthMap is constantly expanding its language capa- bilities. The core HealthMap languages have extensive language support: English, Spanish, Por- tuguese, French, Russian, Chinese, and Arabic. More recently added language capabilities are German, Korean, Japanese, Vietnamese, Malay and Indonesian. HealthMap also processes alerts in Italian and Thai for specific projects. Methods HealthMap The filter breaks the document into unigrams, digrams, trigrams (and tetragrams in some languages), and then computes a likelihood score based on training data from tens of thousands of previously recorded documents. It then assigns the document to one of five categories based on relevance; documents falling in the two most rele- vant categories, Breaking News and Warning, are posted to the map, whereas documents in the Old News, Context, or Not Disease Related categories are hidden from the map. These alerts are retained in the database for use in future research. The last step in the automated pipeline is document clustering to group duplicate reports to- gether. The clustering module compares the given document to each previously collected docu- ment over a fixed time window, and computes a similarity score based on the texts and classification outputs. If the maximum score is above a set threshold, the new document is as- signed as a “child” of the previous document, and hidden from the map. The similarity calcula- tion is based on six factors: (1) character-level headline comparison; (2) word-level headline comparison (fraction of words in common, with inverse term frequency taken into account); (3) disease and place comparison from the output of the named-entity extraction engine; (4) coun- try-level comparison (i.e., if the two reports have different locations within the same country) boost similarity score somewhat; (5) timestamp (reports closer in time are more likely to be simi- lar); and (6) filter category comparison based on the output of the Bayesian filtering module. Following automated processing, HealthMap alerts are posted directly to the public map, partner pages, and RSS feeds. However, at the same time, a team of trained analysts will review all incoming data and correct misclassifications and assign additional higher resolution meta- data classifications. For these curation tasks, a feature-rich analyst dashboard allows users to PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 3 / 17 Neisseria meningitidis and HealthMap search, review and annotate the alerts. For example, analysts can highlight text in a report and assign a named entity category such as executing a geocoding query with the highlighted text and bringing up a map for use in assigning precise location to a report. The analyst corrections feed back into the machine learning algorithms and improve the automated content analysis. Case counts data is entered and additional tags are assigned when relevant. Methods HealthMap HealthMap aims to continue to adding language capabil- ities to further expand its global coverage of hyper-local outbreaks. The relatively straightforward process for adding languages demonstrate the flexibility of the underlying automated parser. Study design and population HealthMap data capture and analyses. HealthMap became operational for N. meningiti- dis disease capture in 2010 and was used for this study to detect cases of meningococcal disease and potential outbreaks in the US from 2010 to 2013. Heat maps were generated using the fol- lowing search conditions: meningitis-Neisseria, United States, new & ongoing outbreaks, and yearly intervals. Alerts reflect online information from news, expert-curated discussions, and validated official reports (e.g., World Health Organization, ProMED Mail, and Google News feeds that were active during this time interval). Alerts were individually crosschecked PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 4 / 17 Neisseria meningitidis and HealthMap manually to ensure they reflected confirmed cases of meningococcal disease with citation of the appropriate state health departments or institutions. Accuracy of HealthMap capture was calculated based on downloading all the citations and individually counting the alerts to verify those capturing diagnoses, suspected cases, or general news. manually to ensure they reflected confirmed cases of meningococcal disease with citation of the appropriate state health departments or institutions. Accuracy of HealthMap capture was calculated based on downloading all the citations and individually counting the alerts to verify those capturing diagnoses, suspected cases, or general news. An “alert” in HealthMap is defined as any unique data source, typically a unique URL link. If the data source (i.e. Associated Press or Reuters) writes a story and 30 newspapers pick it up and write slightly different versions, then the system receives all of those URLs as distinct alerts, but the artificial intelligence of the system will group them together as having identical information despite existing in different locations in the Internet. The term alert is used sometimes inter- changeably as the presence of a case, but occasionally it could also be a warning. Centers for Disease Control data capture and analyses. Traditional surveillance data were obtained from the Centers for Disease Control (CDC) Active Bacterial Core surveil- lance (ABCs) Emerging Infections Program Network [11] for N. meningitidis from 2005 (to include assessment of the impact of vaccine introduction on disease incidence) to 2012 (most recent finalized report available online). Non-B serogroup and serogroup B case rates per 100,000 population reflects ABCs figures in ten US states representing 39 to 43 million inhabitants over eight years. Total cases in the US reflect CDC estimates based on applying ABC case rate to US population census. The HealthMap meningitis system—increasing visibility of meningococcal cases Geographic distribution of HealthMap alerts in the US. Fig 1 illustrates the distribution of HealthMap meningococcal alerts from 2010–2013 that are color-coded by year. The two pink shaded circles represent country levels alerts indicated by HealthMap for 2013 due to the serogroup B outbreaks at Princeton University and University of California Santa Bar- bara. Compared to a retrospective surveillance study reviewing outbreaks from July 1994 to Fig 1. HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Markers indicate state, province, or local alerts. The two pink-shaded circles represent country-level alerts for 2013 due to the serogroup B outbreaks at University of California Santa Barbara and Princeton University. Image generated using ZeeMaps. Reprinted under a CC BY license with permission from ZeeMaps. Fig 1. HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Markers indicate state, province, or l represent country-level alerts for 2013 due to the serogroup B outbreaks at University of California Santa Barbara a using ZeeMaps. Reprinted under a CC BY license with permission from ZeeMaps. Fig 1. HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Markers indicate state, province, or local alerts. The two pink-shaded circles represent country-level alerts for 2013 due to the serogroup B outbreaks at University of California Santa Barbara and Princeton University. Image generated using ZeeMaps. Reprinted under a CC BY license with permission from ZeeMaps. doi:10.1371/journal.pone.0127406.g001 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 5 / 17 Neisseria meningitidis and HealthMap Fig 2. Number of HealthMap meningococcal case alerts in twenty US states reporting disease activity from 2010–2013. The majority of internet- based case alerts derived from meningococcal disease activity in Oklahoma/Colorado (12.0% and 55.4% in 2010), Colorado/Oregon (16.7%/26.2% in 201 Oregon/California (34.6% and 32.1% in 2012), and California/New Jersey (50.0% and 32.6% in 2013). Bars indicate actual number of alerts. Blue color indicates case alerts in 2010, orange color indicates case alerts in 2011, light grey color indicates case alerts in 2012, and dark grey color indicates case alerts in 2013. Fig 2. Number of HealthMap meningococcal case alerts in twenty US states reporting disease activity from 2010–2013. The majority of internet- based case alerts derived from meningococcal disease activity in Oklahoma/Colorado (12.0% and 55.4% in 2010), Colorado/Oregon (16.7%/26.2% in 20 Oregon/California (34.6% and 32.1% in 2012), and California/New Jersey (50.0% and 32.6% in 2013). Bars indicate actual number of alerts. The HealthMap meningitis system—increasing visibility of meningococcal cases Blue color indicates case alerts in 2010, orange color indicates case alerts in 2011, light grey color indicates case alerts in 2012, and dark grey color indicates case alerts in 2013. doi:10.1371/journal.pone.0127406.g002 Fig 2. Number of HealthMap meningococcal case alerts in twenty US states reporting disease activity from 2010–2013. The majority of internet- based case alerts derived from meningococcal disease activity in Oklahoma/Colorado (12.0% and 55.4% in 2010), Colorado/Oregon (16.7%/26.2% in 2011), Oregon/California (34.6% and 32.1% in 2012), and California/New Jersey (50.0% and 32.6% in 2013). Bars indicate actual number of alerts. Blue color indicates case alerts in 2010, orange color indicates case alerts in 2011, light grey color indicates case alerts in 2012, and dark grey color indicates case alerts in 2013 Fig 2. Number of HealthMap meningococcal case alerts in twenty US states reporting disease activity from 2010–2013. The majority of internet- based case alerts derived from meningococcal disease activity in Oklahoma/Colorado (12.0% and 55.4% in 2010), Colorado/Oregon (16.7%/26.2% in 2011), Oregon/California (34.6% and 32.1% in 2012), and California/New Jersey (50.0% and 32.6% in 2013). Bars indicate actual number of alerts. Blue color indicates case alerts in 2010, orange color indicates case alerts in 2011, light grey color indicates case alerts in 2012, and dark grey color indicates case alerts in 2013. doi:10.1371/journal.pone.0127406.g002 June 2002 [12], there is a similar geographic distribution of N. meningitidis disease activity in HealthMap. The distribution of cases captured for the school setting (57.8%) and the commu- nity setting (42.2%) by HealthMap alerts were similar to that reported previously in the set- ting of N. meningitidis outbreaks (44.9% and 36.2%, respectively) [12]. Fig 2 is a breakdown of state-specific HealthMap case alerts for N. meningitidis annually from beginning of 2010 to end of 2013 (2010 = blue, 2011 = orange, 2012 = light grey, and 2013 = dark grey). From this breakdown, it shows that 20 states represented the case alert activity captured by HealthMap during this four-year period. The top-two states, by year, comprising the majority of internet- based case alerts were Oklahoma/Colorado (12.0%/55.4%) in 2010, Colorado/Oregon (16.7%/ 26.2%) in 2011, Oregon/California (34.6%/32.1%) in 2012, and California/New Jersey (50.0%/32.6%) in 2013. June 2002 [12], there is a similar geographic distribution of N. meningitidis disease activity in HealthMap. The HealthMap meningitis system—increasing visibility of meningococcal cases The distribution of cases captured for the school setting (57.8%) and the commu- nity setting (42.2%) by HealthMap alerts were similar to that reported previously in the set- ting of N. meningitidis outbreaks (44.9% and 36.2%, respectively) [12]. Fig 2 is a breakdown of state-specific HealthMap case alerts for N. meningitidis annually from beginning of 2010 to end of 2013 (2010 = blue, 2011 = orange, 2012 = light grey, and 2013 = dark grey). From this breakdown, it shows that 20 states represented the case alert activity captured by HealthMap during this four-year period. The top-two states, by year, comprising the majority of internet- based case alerts were Oklahoma/Colorado (12.0%/55.4%) in 2010, Colorado/Oregon (16.7%/ 26.2%) in 2011, Oregon/California (34.6%/32.1%) in 2012, and California/New Jersey (50.0%/32.6%) in 2013. PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 6 / 17 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Neisseria meningitidis and HealthMap Fig 3. Specificity of HealthMap case alerts for actual Neisseria meningitidis disease activity from 2010–2013. The specificity, on average, for HealthMap alerts during 2010 to 2013 were as follows: 83.3% were diagnosed cases, 10.5% were suspected cases, and 5.6% were related news but not actual cases of Neisseria meningitidis. Pie graphs illustrate percentages for alerts per year with numerical values indicating actual number of alerts. Blue color indicates diagnosed Neisseria meningitidis, orange color indicates suspected Neisseria meningitidis, and light grey color indicates general information associated with meningococcal disease (related news articles but not actual cases). doi:10 1371/journal pone 0127406 g003 Fig 3. Specificity of HealthMap case alerts for actual Neisseria meningitidis disease activity from 2010–2013. The specificity, on average, for HealthMap alerts during 2010 to 2013 were as follows: 83.3% were diagnosed cases, 10.5% were suspected cases, and 5.6% were related news but not actual cases of Neisseria meningitidis. Pie graphs illustrate percentages for alerts per year with numerical values indicating actual number of alerts. Blue color indicates diagnosed Neisseria meningitidis, orange color indicates suspected Neisseria meningitidis, and light grey color indicates general information associated with meningococcal disease (related news articles but not actual cases). doi:10.1371/journal.pone.0127406.g003 doi:10.1371/journal.pone.0127406.g003 Accuracy of HealthMap capture of meningococcal disease. Fig 3 includes information on the specificity of the disease cases captured by the HealthMap automated search algorithm. The HealthMap meningitis system—increasing visibility of meningococcal cases On average, 83.3% of reported cases identified by HealthMap represented diagnosed cases (blue color), 10.5% represented suspected cases (orange color), and 5.6% were related news but not actual cases of N. meningitidis (light grey color). Table 1 indicates the total cases captured through internet data sources utilized by HealthMap for 2010–2013 (30, 28, 22, and 40 cases, respectively) and illustrates that serogroup specific information for these cases was not avail- able in 53.3%, 82.1%, 68.1%, and 52.5% of alerts. PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Epidemiology of vaccine-preventable serogroup ACWY versus serogroup B Fig 4 illustrates the trend in meningococcal cases from 2005 to 2012 using ABCs surveillance data obtained from the CDC surveillance for N. meningitidis [11]. ABCs surveillance reflects ten US states representing 39 to 43 million inhabitants over eight years (Fig 5). There is a de- cline of total cases (Fig 4, solid black line) in the ABCs states from 1,025 cases in 2005 to what appears to be a plateau ranging from 475 to 620 cases during 2010 to 2012. The dotted blue line indicates the decreasing trend in case rates per 100,000 persons (all age groups) for non-B serogroups from 2007 to 2012. The dotted red line indicates a more variable trend in case rates per 100,000 persons (all age groups) for the serogroup B that appears to show an increasing trend since 2010. This variability in case rate for serogroup B is expected since the MenACWY vaccine introduced in 2005 would not be protective against serogroup B. No deviation from these trends occur when analyzing specifically case rates in 18–34 year old individuals (univer- sity age group) as indicated for non-B serogroups (solid blue line) and serogroup B (solid red line). Case rates from serogroup B meningococcus in children < 1 year of age were not includ- ed but are oscillating at a much higher case rate than all other age groups with values between PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 7 / 17 Year School (# of cases) Community (# of cases) Alive / Dead Distribution 2010 Red Bluff, CALIFORNIA (1) Red Bluff/Tehama County, 16/14 Total = 30 cases 2010 Denver/Fort Collins, COLORADO CALIFORNIA (1B+1) B = 23% (3C+1) Larimer County, COLORADO C = 23% 2010 Clare County, MICHIGAN (1) (4C+1B) Y = 0% 2010 Long Island, NEW YORK (1) Bullitt County, KENTUCKY (1) ? = 54% 2010 Athens/Franklin County, OHIO (4B) Franklin County, OHIO (1B) 2010 Oologah, OKLAHOMA (7) 2010 Roanoke College, VIRGINIA (1) 2010 Madison, WISCONSIN (2) 2011 Shasta County, CALIFORNIA (1) Craig, ALASKA (1) 19/9 Total = 28 cases 2011 Boulder, COLORADO (1) Fort Smith, ARKANSAS (1) B = 4% 2011 Washington, DISTRICT OF Los Angeles County, C = 7% COLUMBIA (1) CALIFORNIA (7) Y = 7% 2011 Canton, NEW YORK (1) Larimer County, COLORADO ? = 82% 2011 Columbus, OHIO (2) (1Y) 2011 Norman, OKLAHOMA (1) Hillsborough/Broward 2011 Prinville, OREGON (1C) County, FLORIDA (2) 2011 Bethlehem, PENNSYLVANIA (1) Morgan County, OHIO (1) 2011 Caroline County, VIRGINIA (1) Bend/Prinville/Deschutes 2011 Greenbay, WISCONSIN (1) County, OREGON (1C+1B+1Y+1) 2012 Sacramento/San Francisco/ Clearlake/Ukiah, CALIFORNIA 14/8 Total = 22 cases San Diego, CALIFORNIA (3+1B) (2) B = 5% 2012 Tallahassee/Miami, FLORIDA (2) New York, NEW YORK (4C) C = 27% 2012 East Lansing/Traverse City, Pawnee County, OKLAHOMA Y = 0% MICHIGAN (2) (1) ? = 68% 2012 St. Louis, MISSOURI (1) Prinville/Eugene, OREGON 2012 Prinville/Eugene, OREGON (1C+2) (1C+2) 2013 Maricopa County, ARIZONA (1B) Navajo/Maricopa County 34/6 Total = 40 cases 2013 Petaluma/San Diego/Shasta ARIZONA (2B) B = 43% County/Santa Barbara, Huntington Beach/San C = 5% CALIFORNIA (3+5B) Diego/West Hollywood/ Y = 0% 2013 Chicago, ILLINOIS (1C) Willits, CALIFORNIA ? = 52% 2013 South Bend/Muncie, INDIANA (2) (6+1C) 2013 Princeton/West Long Branch, New York, NEW YORK (4) NEW JERSEY (8B+1) 2013 Wester Chester/Reading/ University Park, PENNSYLVANIA (4) 2013 Charlottesville, VIRGINIA (1B) 2013 Madison, WISCONSIN (1) Example of community-identified and reported meningitis outbreaks in the United States in a four-year period. Based on the source of information and the stage of the outbreak, HealthMap is able to provide geographical and numerical information on cases and also displays information related to disease- causing serotype (when available). Table 1. Meningococcal disease cases identified by HealthMap alerts. Example of community-identified and reported meningitis outbreaks in the United States in a four-year period. doi:10.1371/journal.pone.0127406.t001 Neisseria meningitidis and HealthMap Table 1. Meningococcal disease cases identified by HealthMap alerts. Year School (# of cases) Community (# of cases) Alive / Dead Distribution 2010 Red Bluff, CALIFORNIA (1) Red Bluff/Tehama County, 16/14 Total = 30 cases 2010 Denver/Fort Collins, COLORADO CALIFORNIA (1B+1) B = 23% (3C+1) Larimer County, COLORADO C = 23% 2010 Clare County, MICHIGAN (1) (4C+1B) Y = 0% 2010 Long Island, NEW YORK (1) Bullitt County, KENTUCKY (1) ? = 54% 2010 Athens/Franklin County, OHIO (4B) Franklin County, OHIO (1B) 2010 Oologah, OKLAHOMA (7) 2010 Roanoke College, VIRGINIA (1) 2010 Madison, WISCONSIN (2) 2011 Shasta County, CALIFORNIA (1) Craig, ALASKA (1) 19/9 Total = 28 cases 2011 Boulder, COLORADO (1) Fort Smith, ARKANSAS (1) B = 4% 2011 Washington, DISTRICT OF Los Angeles County, C = 7% COLUMBIA (1) CALIFORNIA (7) Y = 7% 2011 Canton, NEW YORK (1) Larimer County, COLORADO ? = 82% 2011 Columbus, OHIO (2) (1Y) 2011 Norman, OKLAHOMA (1) Hillsborough/Broward 2011 Prinville, OREGON (1C) County, FLORIDA (2) 2011 Bethlehem, PENNSYLVANIA (1) Morgan County, OHIO (1) 2011 Caroline County, VIRGINIA (1) Bend/Prinville/Deschutes 2011 Greenbay, WISCONSIN (1) County, OREGON (1C+1B+1Y+1) 2012 Sacramento/San Francisco/ Clearlake/Ukiah, CALIFORNIA 14/8 Total = 22 cases San Diego, CALIFORNIA (3+1B) (2) B = 5% 2012 Tallahassee/Miami, FLORIDA (2) New York, NEW YORK (4C) C = 27% 2012 East Lansing/Traverse City, Pawnee County, OKLAHOMA Y = 0% MICHIGAN (2) (1) ? = 68% 2012 St. Louis, MISSOURI (1) Prinville/Eugene, OREGON 2012 Prinville/Eugene, OREGON (1C+2) (1C+2) 2013 Maricopa County, ARIZONA (1B) Navajo/Maricopa County 34/6 Total = 40 cases 2013 Petaluma/San Diego/Shasta ARIZONA (2B) B = 43% County/Santa Barbara, Huntington Beach/San C = 5% CALIFORNIA (3+5B) Diego/West Hollywood/ Y = 0% 2013 Chicago, ILLINOIS (1C) Willits, CALIFORNIA ? = 52% 2013 South Bend/Muncie, INDIANA (2) (6+1C) 2013 Princeton/West Long Branch, New York, NEW YORK (4) NEW JERSEY (8B+1) 2013 Wester Chester/Reading/ University Park, PENNSYLVANIA (4) 2013 Charlottesville, VIRGINIA (1B) 2013 Madison, WISCONSIN (1) Example of community-identified and reported meningitis outbreaks in the United States in a four-year period. Based on the source of information and the stage of the outbreak, HealthMap is able to provide geographical and numerical information on cases and also displays information related to disease- causing serotype (when available). doi:10.1371/journal.pone.0127406.t001 Table 1. Meningococcal disease cases identified by HealthMap alerts. Based on the source of information and the stage of the outbreak, HealthMap is able to provide geographical and numerical information on cases and also displays information related to disease- causing serotype (when available). PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 8 / 17 Neisseria meningitidis and HealthMap Fig 4. Invasive meningococcal disease (IMD) reported over the past eight years by Active Bacterial Core Surveillance. Data plotted from Active Bacterial Core Surveillance (ABCs) Emerging Infections Program Network (states highlighted in white). There is a decline of total cases (solid black line) in the ABCs states from 1,025 cases in 2005 to what appears to be a plateau ranging from 475 to 620 cases during 2010 to 2012. The dotted blue line indicates the decreasing trend in case rates per 100,000 persons (all age groups) for non-B serogroups from 2007 to 2012. The dotted red line indicates a more variable trend in case rates per 100,000 persons (all age groups) for the serogroup B that appears to show an increasing trend since 2010. This variability in case rate for serogroup B is expected since the MenACWY vaccine introduced in 2005 would not be protective against serogroup B. No deviation from these trends occur when analyzing specifically case rates in 18–34 year old individuals (university age group) as indicated for non-B serogroups (solid blue line) and serogroup B (solid red line). Case rates for serogroup B meningococcus in children < 1 year of age are not shown but are oscillating at a much higher case rate than all other age groups with values between 1.0 to 2.9 cases per 100,000 population compared to non-B serogroups that range from 0.6 to 1.2 cases per 100,000 population. Fig 4. Invasive meningococcal disease (IMD) reported over the past eight years by Active Bacterial Core Surveillance. Data plotted from Active Bacterial Core Surveillance (ABCs) Emerging Infections Program Network (states highlighted in white). There is a decline of total cases (solid black line) in the ABCs states from 1,025 cases in 2005 to what appears to be a plateau ranging from 475 to 620 cases during 2010 to 2012. The dotted blue line indicates the decreasing trend in case rates per 100,000 persons (all age groups) for non-B serogroups from 2007 to 2012. The dotted red line indicates a more variable trend in case rates per 100,000 persons (all age groups) for the serogroup B that appears to show an increasing trend since 2010. This variability in case rate for serogroup B is expected since the MenACWY vaccine introduced in 2005 would not be protective against serogroup B. No deviation from these trends occur when analyzing specifically case rates in 18–34 year old individuals (university age group) as indicated for non-B serogroups (solid blue line) and serogroup B (solid red line). Case rates for serogroup B meningococcus in children < 1 year of age are not shown but are oscillating at a much higher case rate than all other age groups with values between 1.0 to 2.9 cases per 100,000 population compared to non-B serogroups that range from 0.6 to 1.2 cases per 100,000 population. doi:10.1371/journal.pone.0127406.g004 1.0 to 2.9 cases per 100,000 population compared to non-B serogroups that range from 0.6 to 1.2 cases per 100,000 population. PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Detecting HealthMap signals across national borders for emerging N. meningitidis threats Fig 6 is an illustration of a close-up view from HealthMap illustrating the multiple N. meningi- tidis alerts for Southern California for 2013 including one in San Diego referring to a meningi- tis outbreak in Tijuana, Mexico [13]. What becomes readily apparent from the image is the proximity of Tijuana to the southern border of the United States. In the first three months of 2013, there had been 17 cases (including six deaths) in Tijuana [13]. If the Tijuana outbreak cases were counted as US cases, this would increase the total number of cases for 2013 from 40 to 57 patients (Table 1). In Fig 7, HealthMap is illustrating N. meningitidis disease alerts global- ly and illustrates N. meningitidis disease activity near the northern border of the United States with Canada (involving cities in British Columbia and Ontario). Of the eight cases captured by HealthMap in Canada between 2010 to 2013, the bordering cities reported one case in a student in Western Ontario (2010), one death in a student in Hamilton (2010), and one death in a stu- dent in Vancouver Island (2012). 9 / 17 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Neisseria meningitidis and HealthMap Fig 5. Population under suveillance by ABC for Neisseria meningitidis (adapted from CDC website). Non-B serogroup and serogroup B case rates reflects ABCs figures in ten US states (white shading of entire state indicates that total state population under surveillance while white shaded circle indicates that population of certain counties within the state are under surveillance). Numbers of inhabitants being followed per state are as follows: CT = 3,590,347; GA = 9,919,945; MD = 5,884,563; MN = 5,379,139; NM = 2,085,538; OR = 3,899,353; TN = 3,872,994 (20 urban counties); CA = 3,460,180 (3 county Bay area); CO = 2,532,982 (5 county Denver area); NY = 2,178,020 (7 county Rochester area and 8 county Albany area). Total population under surveillance by ABCs is 42,803,061. d i 10 1371/j l 0127406 005 Fig 5. Population under suveillance by ABC for Neisseria meningitidis (adapted from CDC website). Non-B serogroup and serogroup B case rates reflects ABCs figures in ten US states (white shading of entire state indicates that total state population under surveillance while white shaded circle indicates that population of certain counties within the state are under surveillance). Detecting HealthMap signals across national borders for emerging N. meningitidis threats Numbers of inhabitants being followed per state are as follows: CT = 3,590,347; GA = 9,919,945; MD = 5,884,563; MN = 5,379,139; NM = 2,085,538; OR = 3,899,353; TN = 3,872,994 (20 urban counties); CA = 3,460,180 (3 county Bay area); CO = 2,532,982 (5 county Denver area); NY = 2,178,020 (7 county Rochester area and 8 county Albany area). Total population under surveillance by ABCs is 42,803,061. p , , ; , , ; , , ; , , ; NM = 2,085,538; OR = 3,899,353; TN = 3,872,994 (20 urban counties); CA = 3,460,180 (3 county Bay area); CO = 2,532,982 (5 county Denver area); NY = 2,178,020 (7 county Rochester area and 8 county Albany area). Total population under surveillance by ABCs is 42,803,061. doi:10.1371/journal.pone.0127406.g005 “Connecting the dots” to disease transmission using HealthMap Fig 8 is an example of how real-time informatics can be used to connect what appear to be unre- lated meningococcal disease cases (due to serogroup C) that differ by state, city, site, and time. The data indicate one death in April (Metropolitan State University student in Denver, Fig 6. Local view of HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Multiple alerts in Southern California including one in San Diego refering to a meningitis outbreak across the border in Tijuana, Mexico. Marker color (light orange to darker purple) reflects increasing significance assigned by HealthMap users or news volume associated with the alert. Each marker may represent more than one Neisseria meningitidis alert (as seen in the break-out box). What becomes readily apparent from the image is the proximity of Tijuana to the southern border of the United States. In the first three months of 2013, there had been 17 cases (including six deaths) in Tijuana. Individuals today are more mobile than before which underscores the relevance of disease activity awareness in cities bordering countries. doi:10 1371/journal pone 0127406 g006 Fig 6. Local view of HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Multiple alerts in Southern California including one in San Diego refering to a meningitis outbreak across the border in Tijuana, Mexico. Marker color (light orange to darker purple) reflects increasing significance assigned by HealthMap users or news volume associated with the alert. Each marker may represent more than one Neisseria meningitidis alert (as seen in the break-out box). What becomes readily apparent from the image is the proximity of Tijuana to the southern border of the United States. In the first three months of 2013, there had been 17 cases (including six deaths) in Tijuana. Individuals today are more mobile than before which underscores the relevance of disease activity awareness in cities bordering countries. doi:10.1371/journal.pone.0127406.g006 10 / 17 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Neisseria meningitidis and HealthMap Fig 7. Global view of HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Multiple alerts globally including Neisseria meningococcus activity near the northern border of the US with Canada (cities in British Columbia and Ontario). Marker color (light orange to darker red) indicates increasing newsworthiness. Smaller marker size reflects local alert for city, state, or province while larger marker size reflects a country-level alert. “Connecting the dots” to disease transmission using HealthMap Colorado), two survivors and one death between May and December (Colorado State Universi- ty students at Fort Collins, Colorado), two deaths and one survivor between June and August (two men and one child in Fort Collins, Colorado), and one death in October (a man in Great Falls, Montana). Using the news data captured with HealthMap, what becomes apparent is that the disease transmission may have started at a college party in Boulder, Colorado attended by the student from Denver who died. This party was also attended by Colorado State University (CSU) students from Fort Collins. While it is not clear whether the three CSU students infected with meningococcus serogroup C attended the party in Boulder, the infection in one of the three students predated the infection resulting in the death of two men and the hospitalization of one child in Fort Collins. The infected child was related to one of the two men who died. These two men were players in a hockey match in June (which CSU students attended), and the man dying four months later in Great Falls, Montana was also a player in this hockey match. “Connecting the dots” to disease transmission using HealthMap Of the eight cases captured by HealthMap in Canada between 2010 to 2013, the bordering cities reported one case in a student in Western Ontario (2010), one death in a student in Hamilton (2010), and one death in a student in Vancouver Island (2012). HealthMap disease activity across borders could contribute to understanding disease transmission and subsequent outbreaks. Fig 7. Global view of HealthMap data for Neisseria meningitidis alerts from 2010 to 2013. Multiple alerts globally including Neisseria meningococcus activity near the northern border of the US with Canada (cities in British Columbia and Ontario). Marker color (light orange to darker red) indicates increasing newsworthiness. Smaller marker size reflects local alert for city, state, or province while larger marker size reflects a country-level alert. Of the eight cases captured by HealthMap in Canada between 2010 to 2013, the bordering cities reported one case in a student in Western Ontario (2010), one death in a student in Hamilton (2010), and one death in a student in Vancouver Island (2012). HealthMap disease activity across borders could contribute to understanding disease transmission and subsequent outbreaks. doi:10.1371/journal.pone.0127406.g007 Colorado), two survivors and one death between May and December (Colorado State Universi- ty students at Fort Collins, Colorado), two deaths and one survivor between June and August (two men and one child in Fort Collins, Colorado), and one death in October (a man in Great Falls, Montana). Using the news data captured with HealthMap, what becomes apparent is that the disease transmission may have started at a college party in Boulder, Colorado attended by the student from Denver who died. This party was also attended by Colorado State University (CSU) students from Fort Collins. While it is not clear whether the three CSU students infected with meningococcus serogroup C attended the party in Boulder, the infection in one of the three students predated the infection resulting in the death of two men and the hospitalization of one child in Fort Collins. The infected child was related to one of the two men who died. These two men were players in a hockey match in June (which CSU students attended), and the man dying four months later in Great Falls, Montana was also a player in this hockey match. Fig 8. Understanding disease transmission from disease alerts in HealthMap. The figure illustrates on the y-axis the geographic location from south to north and on the x-axis, monthly intervals between April and December of 2010. The circles indicate survivors of meningococcal infection while the cross- marks indicate deaths. Data captured from real-time informatics trace eight subjects infected with Neisseria meningococcus back to possible disease initiation at a college party in Boulder, Colorado, with subsequent amplification at a hockey match in Fort Collins, Colorado, two months later. d i 10 1371/j l 0127406 008 Fig 8. Understanding disease transmission from disease alerts in HealthMap. The figure illustrates on the y-axis the geographic location from south to north and on the x-axis, monthly intervals between April and December of 2010. The circles indicate survivors of meningococcal infection while the cross- marks indicate deaths. Data captured from real-time informatics trace eight subjects infected with Neisseria meningococcus back to possible disease initiation at a college party in Boulder, Colorado, with subsequent amplification at a hockey match in Fort Collins, Colorado, two months later. doi:10.1371/journal.pone.0127406.g008 current university-linked serogroup B outbreaks in 2013–2014 and the serogroup B outbreaks described previously in Oregon from 1994–1996 and in Ohio in 2010 [14] highlight the unpre- dictable nature of N. meningitidis outbreaks and the challenges facing public officials who need to rapidly identify whether an outbreak is occurring and trigger the necessary intervention to treat and prevent disease transmission. Given the historical data available for comparison and the recent advancement in the capabilities of informatics, we used HealthMap, an automated text processing system that collects data from diverse online sources including media reports and health department press releases, to detect cases of meningococcal disease and potential outbreaks in the US from 2010 to 2013. HealthMap application to emerging diseases was previ- ously proven during the global surveillance monitoring of H1N1 influenza pandemic in 2009 [5]. This real-time informatics system captured a national distribution of meningitis cases (Fig 1) and increased prevalence for school-based settings (Table 1) similar to what has been pub- lished in an older US study using traditional and more comprehensive surveillance [12]. PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Real-Time “snapshot” of meningococcal disease activity by HealthMap Surveillance monitoring systems based on Internet news can provide new dynamics to public health data analysis and, as illustrated in the 2009 A/H1N1 influenza pandemic, help to over- come certain limitations of existing public health surveillance systems including reporting de- lays, inconsistent population coverage, and poor accuracy in detecting emerging diseases. The PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 11 / 17 Neisseria meningitidis and HealthMap Detailed epidemiology captured by traditional surveillance systems The real-time informatics approach used by HealthMap has been active since 2006 and its ca- pacities have evolved significantly for disease monitoring [15–22] with the addition of data capture for N. meningitidis being fully operational as of 2010. However, internet-based data streaming cannot yet provide the level of epidemiological detail captured by traditional surveil- lance approaches which have also investigated patterns in meningococcal disease historically. Traditional surveillance has demonstrated that, since 1991, the frequency of local meningococ- cal outbreaks has increased with the majority of cases being due to serogroup C [23]. Since 1997, local outbreaks due to serogroups Y and B have also been reported [23]. A study pub- lished in 2006 characterized outbreak-associated and sporadic cases in the US during an eight- year period from 1994 through 2002 and identified 69 outbreaks (3–14 outbreaks per year) in- volving 229 patients from 30 states [12]. For all outbreaks, serogroup C was predominant (66.4%) followed by B (23.6%), Y (10.0%), and W (0%). During the same period, sporadic cases were predominantly caused by serogroup B (36.7%), followed by Y (32.5%), C (28.6%), and W (2.2%) [12]. This 2006 publication [12] concluded that 1) outbreak-associated cases were more likely to die compared to sporadic disease cases, 2) outbreak-associated cases most often in- volved serogroup C, and 3) institutional outbreaks generally last less than two weeks. However, in the case of the 2013 outbreaks involving Princeton University and University of California Santa Barbara, the serogroup involved was meningococcus B and these institu- tional outbreaks persisted for months as captured by HealthMap data-streaming. These recent outbreaks suggests a change in meningococcal serogroup distribution which could be expected due to the introduction of vaccines protective against A, C, W, and Y (decreasing incidence of non-serogroup B meningococcus captured by ABCs in Fig 4). Since the peak of meningococcal disease in the 1990s in the US, disease incidence has decreased to 0.2 cases per 100,000 popula- tion in all age groups in 2011 [24]. The burden of disease is highest among infants younger than 1 year of age (2.6 cases per 100,000 persons), adolescents 16 to 21 years of age (0.4 cases per 100,000 persons), and adults older than 65 years of age (0.3 cases per 100,000 persons) [24]. Detailed epidemiology captured by traditional surveillance systems While serogroups B, C, and Y are each responsible for approximately one third of menin- gococcal cases in the US, the proportion caused by each serogroup varies by age group. For ex- ample, 60% of disease in children younger than 60 months of age is caused by serogroup B. However, 73% of disease in persons aged  11 years is caused by serogroup C, Y, or W. While no serogroup B vaccine is licensed in the US for children younger than 60 months of age, quad- rivalent meningococcal glycoconjugate vaccines protective against serogroups A, C, W, and Y (MenACWY) have been licensed in the US since 2005 [24]. MenACWY vaccine coverage has increased to over 70% by 2011 in adolescents 13 to 17 years of age [24] with the estimated number of annual cases from serogroups C and Y decreasing by 75% in individuals 11 to 14 years of age from 2000–2004 to 2005–2009 [23]. However, for individuals 15 to 18 years of age, serogroup C and Y cases have decreased only by 27% and a peak in meningococcal disease has persisted among persons aged 18 to 21 years [23]. While the Internet-based data streaming approach used by HealthMap underestimates US disease burden (Table 1) when compared to estimates from ABCs (Fig 4), it is a very rapid method for acquiring a snapshot of meningococcal disease visibility as a function of time (Figs 2 and 3). Predicting when a sporadic case will evolve into an outbreak of the proportions seen at Prince- ton, New Jersey, and Santa Barbara, California, remains problematic with traditional surveil- lance and could be bridged by the application of real-time informatics such as HealthMap. current university-linked serogroup B outbreaks in 2013–2014 and the serogroup B outbreaks described previously in Oregon from 1994–1996 and in Ohio in 2010 [14] highlight the unpre- dictable nature of N. meningitidis outbreaks and the challenges facing public officials who need to rapidly identify whether an outbreak is occurring and trigger the necessary intervention to treat and prevent disease transmission. Given the historical data available for comparison and the recent advancement in the capabilities of informatics, we used HealthMap, an automated text processing system that collects data from diverse online sources including media reports and health department press releases, to detect cases of meningococcal disease and potential outbreaks in the US from 2010 to 2013. HealthMap application to emerging diseases was previ- ously proven during the global surveillance monitoring of H1N1 influenza pandemic in 2009 [5]. This real-time informatics system captured a national distribution of meningitis cases (Fig 1) and increased prevalence for school-based settings (Table 1) similar to what has been pub- lished in an older US study using traditional and more comprehensive surveillance [12]. While the Internet-based data streaming approach used by HealthMap underestimates US disease burden (Table 1) when compared to estimates from ABCs (Fig 4), it is a very rapid method for acquiring a snapshot of meningococcal disease visibility as a function of time (Figs 2 and 3). Predicting when a sporadic case will evolve into an outbreak of the proportions seen at Prince- ton, New Jersey, and Santa Barbara, California, remains problematic with traditional surveil- lance and could be bridged by the application of real-time informatics such as HealthMap. 12 / 17 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Neisseria meningitidis and HealthMap N. meningitidis outbreaks: potential of internet-based data streaming to de-convolute disease transmission or capture intervention-related signals The unpredictable nature of meningococcal outbreaks and the protracted timeline during which enough cases accumulate to trigger declaration of an outbreak were recently highlighted by the serogroup B meningococcus university-linked outbreaks that occurred on both the east and west coasts of the US in 2013. More recently, on March 10, 2014, a student at Drexel Uni- versity died from serogroup B meningococcal disease. CDC laboratory analysis, based on ge- netic fingerprinting, demonstrated that the strain infecting this student matched that in the Princeton University outbreak. This student had been in close contact with students from Princeton University about a week before illness, and the laboratory analysis indicated that the outbreak strain may still be present in the Princeton community and require vigilance for addi- tional cases [29]. These university outbreaks highlight the limitations of antibiotic chemopro- phylaxis interventions that require the identification and treatment of all individuals at risk in order to avoid subsequent disease transmission by untreated carriers of subclinical disease. This was evidenced during a serogroup B outbreak in a hotel [30] and another in a middle school [31] in 1995. However, real-time capture of cases by internet-based informatics can complement traditional surveillance monitoring approaches leading to increasing confidence and resourcing in outbreak settings by decreasing time to outbreak declaration and medical handling. As illustrated in Fig 8, modes of disease transmission could be traced to ensure tar- geted administration of antibiotic therapy or prophylactic interventions to the relevant at-risk population. Finally, internet-based monitoring systems like HealthMap can provide a real-time channel for patients to report intervention-related failures or can be applied to safety data min- ing by scanning social media channels like Twitter and Facebook on a macro-level for patient- reported safety insights. This could include the direct reporting to regulatory authorities of medication or vaccine-associated adverse events identified through social media and mobile phone apps. These widely accessible and user-friendly computer-based options can support the tracking of subject safety and could facilitate decision-making related to the use of unlicensed therapies or vaccines in emergency situations. One such emergency situation was the Princeton University serogroup B meningococcal disease outbreak that began in the US in March 2013. Using informatics to put into context emerging N. meningitidis disease threats and related public health interventions The ability of informatics-based approaches to filter through local, national, and global disease activity in real-time provides another feature not readily apparent from traditional surveillance approaches—modes of disease transmission. Individuals today are more mobile than before which underscores the relevance of disease activity awareness in cities bordering countries. From Fig 7, it becomes apparent that HealthMap disease activity across borders could contrib- ute to understanding disease transmission and subsequent outbreaks. As an example, the PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 13 / 17 Neisseria meningitidis and HealthMap HealthMap alerts for Southern California in 2013 brought to attention a meningitis outbreak in Tijuana, Mexico (Fig 6). According to the San Diego Deputy Public Health director, the out- break in Tijuana in 2013 was three times the incidence seen in Tijuana annually [13]. Tijuana is the largest city in the Mexican state of Baja California and is situated on the US-Mexico bor- der adjacent to San Diego, California. This is the most traversed international border in the world [25] with the two border crossing stations in Tijuana accounting for an estimated 300,000 border crossings each day. Notably, a subsequent epidemiological study on meningo- coccal disease in Latin America from 1945 to 2010 demonstrated that serogroup A has virtually disappeared, and that serogroups B and C are responsible for most cases [26]. The disease ac- tivity captured by HealthMap for Canadian cities (Fig 7) near the US border reflected only two cases in 2010 and one case in 2012 that may be reflecting the timely incorporation of vaccines into the national schedules by Health Canada. The quadrivalent conjugated vaccine protective for A,C,W,Y was licensed in Canada in 2007 [27] and, more recently, a recombinant serogroup B vaccine was approved on January 2014 [28]. N. meningitidis outbreaks: potential of internet-based data streaming to de-convolute disease transmission or capture intervention-related signals PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Acknowledgments The authors wish to acknowledge Mariagrazia Pizza and Rino Rappuoli for their critical review of the article and are grateful to Anna Tomasulo for her assistance with data preparation and Georgio Corsi for his assistance with artwork related to the figures. The authors affirm that all individuals who have contributed significantly to the article have been listed. Conclusions International events with the potential to cross borders (e.g., 2009 H1N1 pandemic influenza) are probably the best example of the application of HealthMap and highlight its potential use for other emerging worldwide threats (e.g., SARS, avian influenza, or Ebola virus). The under- reporting of disease cases by internet-based data streaming in specific N. meningitidis outbreak modes makes inadequate any comparison to epidemiological data captured by more compre- hensive traditional N. meningitidis surveillance such as the ABCs network published by the CDC. However, the natural delays in compiling reports from traditional surveillance (at the time of writing, ABCs N. meningitidis data for 2013 is listed as being provisional) emphasize the utility of real-time internet-based data streaming for providing a real-time “snapshot” of meningococcal disease. In this investigation on surveillance of N. meningitidis disease activity and transmission using the HealthMap information tool, what becomes apparent is that this information technology can complement epidemiological data captured by traditional surveil- lance monitoring systems by quickly filling gaps (including the visualization of modes of me- ningococcal disease transmission in outbreaks) allowing for better resource and action planning. For the general public and the public health community, HealthMap surveillance of meningococcal disease activity can increase disease awareness by collecting an immense amount of disease/outbreak reports and putting them in a widely accessible and user-friendly platform. This user-friendly platform of HealthMap has the potential to additionally comple- ment traditional surveillance monitoring by serving as an internet-based real-time channel for patients infected with N. meningitidis to report intervention-related failures. N. meningitidis outbreaks: potential of internet-based data streaming to de-convolute disease transmission or capture intervention-related signals An emergency vaccination campaign with an imported vaccine (Bexsero, approved in Austra- lia, Europe, and Canada but not in the US) was able to be implemented in the US in December 2013 (approximately nine months later) only under special regulatory procedures after the PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 14 / 17 Neisseria meningitidis and HealthMap CDC filed an Investigational New Drug application with the Food and Drug Administration requesting special permission for use [32]. With regards to invasive meningococcal disease (IMD), a potential limitation of using HealthMap may be related to the inability to identify and report the serotypes specifically linked to the disease outbreak. Traditionally, all US states have mandated reporting of IMD [33] and re- quire health care providers to immediately report suspected cases to assure rapid identification of close contacts and administration of antibiotic prophylaxis [23]. However, despite the swift execution of the protocol established to confirm these reports, there is always a window of delay secondary to the specialized public health molecular and microbiological laboratory testing nec- essary to confirm cases in establish IMD clusters. Usually, during an outbreak, the high case rates increase awareness in health authorities, but options for implementing the best tailored, prophylactic measure requires understanding the extent and mode of transmission involved in that particular outbreak. A national network of state epidemiologists and CDC maintain regular communication on the outbreak status through electronic alerts over the secure CDC Epi-X sys- tem and direct communication between states when out-of-state exposures occur. References 1. Bona G, Guidi C. Meningococcal vaccine evolution. J Prev Med Hyg 2012 September; 53(3):131–5. PMID: 23362617 2. Bilukha O, Messonnier N, Fischer M. Use of meningococcal vaccines in the United States. Pediatr In- fect Dis J 2007 May; 26(5):371–6. PMID: 17468644 3. Cohn AC, MacNeil JR, Harrison LH, Hatcher C, Theodore J, Schmidt M et al. Changes in Neisseria meningitidis disease epidemiology in the United States, 1998–2007: implications for prevention of me- ningococcal disease. Clin Infect Dis 2010 January 15; 50(2):184–91. doi: 10.1086/649209 PMID: 20001736 4. Erickson LJ, De WP, McMahon J, Heim S. Complications of meningococcal disease in college stu- dents. Clin Infect Dis 2001 September 1; 33(5):737–9. PMID: 11477523 5. Brownstein JS, Freifeld CC, Madoff LC. Influenza A (H1N1) virus, 2009—online monitoring. N Engl J Med 2009 May 21; 360(21):2156. doi: 10.1056/NEJMp0904012 PMID: 19423868 6. Brownstein JS, Freifeld CC, Madoff LC. Digital disease detection—harnessing the Web for public health surveillance. N Engl J Med 2009 May 21; 360(21):2153–5, 2157. doi: 10.1056/NEJMp0900702 PMID: 19423867 7. Hartley D, Nelson N, Walters R, Arthur R, Yangarber R, Madoff L et al. Landscape of international event-based biosurveillance. Emerg Health Threats J 2010; 3:e3. doi: 10.3134/ehtj.10.003 PMID: 22460393 8. Heymann DL, Rodier GR. Hot spots in a wired world: WHO surveillance of emerging and re-emerging infectious diseases. Lancet Infect Dis 2001 December; 1(5):345–53. PMID: 11871807 9. Brownstein JS, Freifeld CC, Reis BY, Mandl KD. Surveillance Sans Frontieres: Internet-based emerg- ing infectious disease intelligence and the HealthMap project. PLoS Med 2008 July 8; 5(7):e151. doi: 10.1371/journal.pmed.0050151 PMID: 18613747 10. Brownstein JS, Freifeld CC, Reis BY, Mandl KD. HealthMap: Internet-based emerging infectious dis- ease intelligence. In: Lemon SM, Hamburg MA, Sparling PF, Choffnes ER, Marck A, editors. Global In- fectious Disease Surveillance and Detection: Assessing the Challenges—Finding Solutions. Washington, D.C.: Institute of Medicine of the National Academies; 2007. doi: 10.1371/journal.pone. 0006954 PMID: 21391348 11. Centers for Disease Control and Prevention. Active Bacterial Core Surveillance Report, Emerging In- fections Program Network, Neisseria meningitidis, 2003–2012. 2013. Available: http://www.cdc.gov/ abcs/reports-findings/surv-reports.html. 12. Brooks R, Woods CW, Benjamin DK Jr, Rosenstein NE. Increased case-fatality rate associated with outbreaks of Neisseria meningitidis infection, compared with sporadic meningococcal disease, in the United States, 1994–2002. Clin Infect Dis 2006 July 1; 43(1):49–54. PMID: 16758417 13. Magee M. Meningitis warning in schools: outbreak results in six deaths in Tijuana and one in San Diego County. San Diego Union-Tribune 2013 Mar 21. 14. Author Contributions Conceived and designed the experiments: SSA JB. Performed the experiments: SSA SM. Ana- lyzed the data: SSA EOO SM CF GT JB. Contributed reagents/materials/analysis tools: SM CF JB. Wrote the paper: SSA EOO SM CF GT JB. 15 / 17 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 Neisseria meningitidis and HealthMap PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 References Mandal S, Wu HM, MacNeil JR, Machesky K, Garcia J, Plikaytis BD et al. Prolonged university outbreak of meningococcal disease associated with a serogroup B strain rarely seen in the United States. Clin In- fect Dis 2013 August; 57(3):344–8. doi: 10.1093/cid/cit243 PMID: 23595832 15. Brownstein JS, Freifeld CC. HealthMap: the development of automated real-time internet surveillance for epidemic intelligence. Euro Surveill 2007 November; 12(11):E071129. PMID: 18053571 16. Brownstein JS, Mandl KD. Reengineering real time outbreak detection systems for influenza epidemic monitoring. AMIA Annu Symp Proc 2006;866. PMID: 17238486 17. Chunara R, Freifeld CC, Brownstein JS. New technologies for reporting real-time emergent infections. Parasitology 2012 December; 139(14):1843–51. doi: 10.1017/S0031182012000923 PMID: 22894823 18. Freifeld CC, Brownstein JS, Menone CM, Bao W, Filice R, Kass-Hout T et al. Digital drug safety surveil- lance: monitoring pharmaceutical products in twitter. Drug Saf 2014 May; 37(5):343–50. doi: 10.1007/ s40264-014-0155-x PMID: 24777653 19. Freifeld CC, Chunara R, Mekaru SR, Chan EH, Kass-Hout T, Ayala IA et al. Participatory epidemiology: use of mobile phones for community-based health reporting. PLoS Med 2010; 7(12):e1000376. doi: 10. 1371/journal.pmed.1000376 PMID: 21151888 20. Freifeld CC, Mandl KD, Reis BY, Brownstein JS. HealthMap: global infectious disease monitoring through automated classification and visualization of Internet media reports. J Am Med Inform Assoc 2008 March; 15(2):150–7. PMID: 18096908 21. Scales D, Zelenev A, Brownstein JS. Quantifying the effect of media limitations on outbreak data in a global online web-crawling epidemic intelligence system, 2008–2011. Emerg Health Threats J 2013; 6:21621. doi: 10.3402/ehtj.v6i0.21621 PMID: 24206612 PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 16 / 17 Neisseria meningitidis and HealthMap 22. Wilson K, Brownstein JS. Early detection of disease outbreaks using the Internet. CMAJ 2009 April 14; 180(8):829–31. doi: 10.1503/cmaj.090215 PMID: 19364791 23. Centers for Disease Control and Prevention. Meningococcal Disease: Epidemiology and prevention of vaccine-preventable diseases. 12 ed. 2012. p. 193–204. 24. Cohn AC, MacNeil JR, Clark TA, Ortega-Sanchez IR, Briere EZ, Meissner HC et al. Prevention and control of meningococcal disease: recommendations of the Advisory Committee on Immunization Prac- tices (ACIP). MMWR Recomm Rep 2013 March 22; 62(RR-2):1–28. PMID: 23515099 25. Chacon-Cruz E, Sugerman DE, Ginsberg MM, Hopkins J, Hurtado-Montalvo JA, Lopez-Viera JL et al. Surveillance for invasive meningococcal disease in children, US-Mexico border, 2005–2008. Emerg In- fect Dis 2011 March; 17(3):543–6. doi: 10.3201/eid1703.101254 PMID: 21392455 26. Safadi MA, Gonzalez-Ayala S, Jakel A, Wieffer H, Moreno C, Vyse A. PLOS ONE | DOI:10.1371/journal.pone.0127406 May 20, 2015 References The epidemiology of meningo- coccal disease in Latin America 1945–2010: an unpredictable and changing landscape. Epidemiol In- fect 2012 August 9;1–12. doi: 10.1017/S0950268812002245 PMID: 23046921 27. Salvadori M, Bortolussi R. Meningococcal vaccines in Canada: An update. Paediatr Child Health 2011 October; 16(8):485–6. PMID: 23024587 28. Office of Regulatory Affairs BaGTD. Bexsero. www hc-sc gc ca 2014 April 10. Available: http://www.hc- sc.gc.ca/dhp-mps/prodpharma/sbd-smd/drug-med/sbd_smd_2014_bexsero_147275-eng.php?utm_ source = rss&utm_medium = rss&utm_campaign = summary-basis-of-decision-sbd-bexsero-2014. 29. Centers for Disease Control and Prevention. Meningococcal Disease Update. 2014 March 18. Avail- able: http://www.cdc.gov/meningococcal/outbreaks/princeton.html. 30. Centers for Disease Control and Prevention. Outbreaks of group B meningococcal disease—Florida, 1995 and 1997. MMWR Morb Mortal Wkly Rep 1998 October 9; 47(39):833–7. PMID: 9780239 31. Jackson LA, Alexander ER, DeBolt CA, Swenson PD, Boase J, McDowell MG et al. Evaluation of the use of mass chemoprophylaxis during a school outbreak of enzyme type 5 serogroup B meningococcal disease. Pediatr Infect Dis J 1996 November; 15(11):992–8. PMID: 8933547 32. Centers for Disease Control and Prevention. Princeton University meningococcal disease outbreak and University of California, Santa Barbara meningococcal disease outbreak. 2013 December 31. Available: http://www.cdc.gov/meningococcal/outbreaks/princeton.html. 33. Adams DA, Gallagher KM, Jajosky RA, Kriseman J, Sharp P, Anderson WJ et al. Summary of Notifiable Diseases—United States, 2011. MMWR Morb Mortal Wkly Rep 2013 July 5; 60(53):1–117. PMID: 23820934 17 / 17
https://openalex.org/W2122647535
https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0012582&type=printable
English
null
Ultrasound-Mediated DNA Transformation in Thermophilic Gram-Positive Anaerobes
PloS one
2,010
cc-by
10,291
Abstract This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by grants KSCX1-YW-11C2 and Hundred-Scholar Award from Chinese Academy of Sciences to JX, 30870572 from National Natural Science Foundation of China (NSFC) to JX and NSFC-30800014 to HS. It was also funded by the China-Israel Cooperation Project ‘‘Designer Cellulosome Systems’’ from the Ministry of Science and Technology of China, JQ200822 from the Young Investigator Award of Shandong Province, and Shandong Key Laboratory of Energy Genetics (China) to JX and HS. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * E-mail: xujian@qibebt.ac.cn (JX); jzhou@ou.edu (JZ) * E-mail: xujian@qibebt.ac.cn (JX); jzhou@ou.edu (JZ) . These authors contributed equally to this work. . These authors contributed equally to this work. DNA to high-temperature and reduce its stability and integrity. Thirdly, the fastidious requirements in cell handling and preparation of competent-cells for strictly anaerobic bacteria further exasperated this situation. Therefore, there is a tremendous need for developing a simple, rapid and effective method for genetically transforming TGPAs. Ultrasound-Mediated DNA Transformation in Thermophilic Gram-Positive Anaerobes 1 Qingdao Institute of BioEnergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao, Shandong, China, 2 Institute For Environmental Genomics, Department of Botany and Microbiology, University of Oklahoma, Norman, Oklahoma, United States of America Abstract Background: Thermophilic, Gram-positive, anaerobic bacteria (TGPAs) are generally recalcitrant to chemical and electrotransformation due to their special cell-wall structure and the low intrinsic permeability of plasma membranes. Methodology/Principal Findings: Here we established for any Gram-positive or thermophiles an ultrasound-based sonoporation as a simple, rapid, and minimally invasive method to genetically transform TGPAs. We showed that by applying a 40 kHz ultrasound frequency over a 20-second exposure, Texas red-conjugated dextran was delivered with 27% efficiency into Thermoanaerobacter sp. X514, a TGPA that can utilize both pentose and hexose for ethanol production. Experiments that delivered plasmids showed that host-cell viability and plasmid DNA integrity were not compromised. Via sonoporation, shuttle vectors pHL015 harboring a jellyfish gfp gene and pIKM2 encoding a Clostridium thermocellum b-1,4- glucanase gene were delivered into X514 with an efficiency of 66102 transformants/mg of methylated DNA. Delivery into X514 cells was confirmed via detecting the kanamycin-resistance gene for pIKM2, while confirmation of pHL015 was detected by visualization of fluorescence signals of secondary host-cells following a plasmid-rescue experiment. Furthermore, the foreign b-1,4-glucanase gene was functionally expressed in X514, converting the host into a prototypic thermophilic consolidated bioprocessing organism that is not only ethanologenic but cellulolytic. Conclusions/Significance: In this study, we developed an ultrasound-based sonoporation method in TGPAs. This new DNA- delivery method could significantly improve the throughput in developing genetic systems for TGPAs, many of which are of industrial interest yet remain difficult to manipulate genetically. Citation: Lin L, Song H, Ji Y, He Z, Pu Y, et al. (2010) Ultrasound-Mediated DNA Transformation in Thermophilic Gram-Positive Anaerobes. PLoS ONE 5(9): e12582. doi:10.1371/journal.pone.0012582 i Y, He Z, Pu Y, et al. (2010) Ultrasound-Mediated DNA Transformation in Thermophilic Gram-Positive Anaerobes. PLoS ONE 5(9): e12582 0012582 Editor: Shuang-yong Xu, New England Biolabs, Inc, United States of America Editor: Shuang-yong Xu, New England Biolabs, Inc, United States of America Received July 8, 2010; Accepted August 15, 2010; Published September 4, 2010 Received July 8, 2010; Accepted August 15, 2010; Published September 4, 2010 Copyright:  2010 Lin et al. This is an open-access article distributed under the terms of the Creative Commons use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright:  2010 Lin et al. PLoS ONE | www.plosone.org Introduction There have been relatively few reported successes in the transformation of thermophilic, Gram-positive, anaerobes (TGPAs) [1,2,3]. Firstly, some thermophilic bacteria are difficult to transform due to the unique features of their cell envelope, the formation of endospore and the low permeability of plasma membrane [4]. Secondly, most of the reported gene transfer protocols for these organisms were based on electroporation [1,2,5], where higher electroporation efficiency was limited to only a few laboratories that used sophisticated customer-built cuvettes and pulse generator [2]. Key drawbacks of electroporation include the need for an ion-free condition and the requirement of contact between electrodes and solution. Moreover, DNA electroporation through the membrane is not a mechanical penetration process, because interactions between pore and DNA are involved [6]. This interaction may slow down DNA penetration through cell wall after pulsing, which might expose the foreign double-stranded Thermoanaerobacter are a group of gram-positive anaerobes with optimal growth temperature of 60–65uC. They have attracted growing attention over the past few years due to their rare capability utilizing both hexose and pentose at high temperature for ethanol production [7–11]. The co-culture of the hexose- and pentose- fermenter Thermoanaerobacter and cellulose degraders such as Clostriditium thermocellum, both TGPAs, fulfills the key require- ments in a Consolidated BioProcessing scheme [12,13] and therefore has been considered one promising strategy for industrial-scale cellulosic ethanol production. One of the Thermo- anaerobacter strains, X514, was of particular interest. When partnering with selected C. thermocellum strains, the X514 strain PLoS ONE | www.plosone.org 1 September 2010 | Volume 5 | Issue 9 | e12582 Sonoporation of Thermophiles yield higher ethanol yields than other closely related Thermoanaer- obacter strains [8,9]. Furthermore, Thermoanaerobacter ethanolicus fermented pentose and hexose for ethanol in a simultaneous and unbiased manner [14], a feature crucial for cost-efficient cellulosic ethanol production but not widely found in currently available ethanologens such as Saccharomyces cerevisiae. Recently the genomic sequence of X514 has been completed (GenBank ID: NC010320) and the central metabolic pathways in X514 have been constructed via metabolic flux analysis [9]. However, the sugar to ethanol conversion rate in Thermoanaerobacter (,30%) remains low [7,9]. Along with ethanol, other fermentation by-products were usually generated, including acetate and lactate, which may significantly reduce ethanol yields [7,9]. Therefore, it is imperative that the genomic, metabolic and physiological understanding be leveraged towards genetic and genomic engineering of the cells for trait improvement. Effects of ultrasound-exposure duration and electroporation field strength on delivery efficiency of fluorescent dextran p Genetic manipulations of TGPAs in general, and of X514 in particular, are often hindered by the lack of a quick and efficient genetic transformation system. A few exceptions reported have involved custom-made electroporation-apparatus that are not accessible to most laboratories [1,3]. Thus, it is of paramount importance that quick and efficient transformation systems be developed for these organisms. Ultrasound-based sonoporation, an alternative to traditional electroporation for gene transfer, has been extensively studied in eukaryotic transformation and gene therapy [15,16]. Sonoporation employs acoustic cavitation of microbubbles to create pores on the cell membrane, which enhances delivery of DNA and other macromolecules [17]. When cell membrane is self-repaired, the DNA molecules are thus trapped inside cells. To date, all successful bacterial sonoporation experiments have been carried out only in mesophilic organisms or in Gram-negative organisms such as Fusobacterium nucleatum, Escherichia coli DH5a, Pseudomonas fluorescens and Pseudomonas putida [17,18]. Efforts that target thermophilic or Gram-positive organisms have not been successful. To develop and optimize the sonoporation method for TGPAs, we first tested the delivery of fluorescent Texas-red conjugated dextran into native Thermoanaerobacter sp. X514 cells using a Branson B200 Sonifier with fixed ultrasound frequency and power (40 kHz, 19 W). The delivery efficiency of Texas-red was measured as the proportion of fluorescent cells to all cells in a view-zone under the microscope [18]. The results showed that the population of fluorescent cells increased with extended ultrasound exposure, and reached the maximum after 20 seconds of exposure (Fig. 1A; Fig. 2). It is concluded that optimal ultrasound-exposure duration for sonoporation in the Branson B200 sonifier is 20 seconds (Fig. 1A; Fig. 2). Under this condition, intracellular delivery of the dextran was found in 27.5% of the Thermoanaer- obacter cells. This finding suggested that the delivery efficiency of DNA might also reach such a high level by optimizing the parameters such as temperature, frequency, intensity and duration of ultrasound exposure. In this study, we established a simple, rapid and effective transformation method for TGPAs based on ultrasound-mediated sonoporation. Transformation was confirmed by detecting the We next tested the Texas-red delivery using a BioRad Micro Pulser electroporator, in order to compare the transformation Figure 1. Effects of ultrasound exposure duration or electrical field strength on the delivery efficiency of Texas-red labeled dextran. A. Effects of sonoporation-exposure duration on delivery efficiency of Texas-red labeled dextran. B. Introduction presence or the activities of three introduced heterologous genes that include an antibiotic resistance marker gene, a green fluorescence protein gene and a b-1,4-glucanase gene. The introduced b-1,4-glucanase gene was functionally expressed and conveyed to the host-cell the ability to degrade cellulose, converting it into a prototypic thermophilic Consolidated BioProcessing organism that is not only ethanologenic but also cellulolytic [12]. The advantages of sonoporation-based transfor- mation, particularly the ease and simplicity of procedure, the preservation of cells in indigenous medium and the minimal perturbation to cell viability, suggested this method could be particularly suitable for high throughput development of genetic systems. Effects of ultrasound-exposure duration and electroporation field strength on delivery efficiency of fluorescent dextran Effects of electrical field strength on delivery efficiency of Texas-red labeled dextran. The ratio of Texas-red-featured, fluorescent X514 cells to total cells were quantified by five randomly selected view-zones under fluorescent microscope. Standard deviations are also shown. doi:10.1371/journal.pone.0012582.g001 Figure 1. Effects of ultrasound exposure duration or electrical field strength on the delivery efficiency of Texas-red labeled dextran. A. Effects of sonoporation-exposure duration on delivery efficiency of Texas-red labeled dextran. B. Effects of electrical field strength on delivery efficiency of Texas-red labeled dextran. The ratio of Texas-red-featured, fluorescent X514 cells to total cells were quantified by five randomly selected view-zones under fluorescent microscope. Standard deviations are also shown. doi:10.1371/journal.pone.0012582.g001 Figure 1. Effects of ultrasound exposure duration or electrical field strength on the delivery efficiency of Texas-red labeled dextran. A. Effects of sonoporation-exposure duration on delivery efficiency of Texas-red labeled dextran. B. Effects of electrical field strength on delivery efficiency of Texas-red labeled dextran. The ratio of Texas-red-featured, fluorescent X514 cells to total cells were quantified by five randomly selected view-zones under fluorescent microscope. Standard deviations are also shown. doi:10.1371/journal.pone.0012582.g001 September 2010 | Volume 5 | Issue 9 | e12582 PLoS ONE | www.plosone.org 2 Sonoporation of Thermophiles Figure 2. Delivery of Texas Red-conjugated dextran into Thermoanaerobacter sp. X514 by sonoporation. The sonoporation experiments were conducted with exposure durations ranged from 0 to 40s. Cells were visualized under Olympus-BX51 microscope with phase lights (upper-row figures) and blue lights (lower-row figures). Scale bar represents 10 mm. doi:10.1371/journal.pone.0012582.g002 Figure 2. Delivery of Texas Red-conjugated dextran into Thermoanaerobacter sp. X514 by sonoporation. The sonoporation experiments were conducted with exposure durations ranged from 0 to 40s. Cells were visualized under Olympus-BX51 microscope with phase lights (upper-row figures) and blue lights (lower-row figures). Scale bar represents 10 mm. doi:10.1371/journal.pone.0012582.g002 Methylation modification assay X514 could degrade foreign plasmid DNA, cell extracts of X514 were incubated with various forms of pHL015 (Fig. S2B). Methylation- competent E. coli DH5a (Dam+/Dcm+), but not the methylation- deficiency mutant E. coli JM110 (Dam2/Dcm2) cells, methylated pHL015 in vivo. The methylated pHL015 extracted from E. coli DH5a was not degraded by X514 cell extracts while the unmethylated pHL015 extracted from E. coli JM110 was readily cleaved by X514 cell extracts (Fig. 4A, Lane 4 and Lane 8 respectively). In vitro methylated pHL015, extracted from either species of E. coli, resulted in no degradation by X514 cell extracts (Fig. 4A, Lane 5, Lane 10 and Lane 11 respectively). These findings suggest that methylation of DNA is required for protection against degradation in X514 cells. To examine whether the enzyme activity underlying plasmid degradation in X514 is buffer dependent, the degradation of Dam2/Dcm2 pHL015 was evaluated in the presence of the New England Biolabs (NEB) restriction enzyme Buffer 1, 2 3 and 4. Nonspecific degradation of pHL015 was observed in the presence of NEB Buffer 1, 2 and 4 (Fig. 4B, Lane 2, 3 and 5), suggesting buffer dependency of the enzyme activity. To identify the putative restriction methylation (RM) system(s) in X514, we took advantage of the complete genome sequence of Thermoanaerobacter X514 (Genbank ID: NC010320). From the genome, a total of three putative genes encoding DNA methyltransferases (Teth514_0458/1737/1828) were identified, which, based on our functional annotations, all methylate adenines at their N6 positions in GATC sequences [20]. We hypothesized that the GATC sequences are one signal for cleavage recognized by the X514 RM systems. To test this hypothesis, we compared the gel electrophoresis patterns of pHL015 (methylated or unmethylated) resulted from two treatments, one by X514 cellular extracts and the other by DpnII, a commercially available enzyme However, the signature fluorescence of GFP was not observed in X514 under fluorescent microscope (data not shown), suggesting that the gfp gene encoded on pHL015 was either not expressed or was not functional in the hosting X514 cells. Causes for the later possibility might include protein structural-changes under high temperature or in the absence of molecular oxygen, both features of the Thermonanaerobacter growth [23]. To test this possibility, a plasmid rescue experiment was performed and the plasmid DNA extracts from the pHL015-transformed X514 were subsequently trans- formed into E. coli. GFP fluorescence was visualized in the E. coli transformants. Transformation and heterologous expression of green florescence protein (GFP) gene and b-1.4-endoglucanase gene in Thermoanaerobacter sp. X514 by sonoporation Transformation and heterologous expression of green florescence protein (GFP) gene and b-1.4-endoglucanase gene in Thermoanaerobacter sp. X514 by sonoporation Effects of ultrasound and electroporation pulse on cell viability and plasmid integrity efficiencies of sonoporation and electroporation. For the later, additional steps were required, including protoplast genera- tion, iterative centrifugation and multiple rounds of washing under anaerobic conditions, to ensure very low ion concen- tration and maximal efficiency (See Materials and Meth- ods). Optimal conditions were located by monitoring the delivery of fluorescent dextran into X514 upon pulsing under 0 to 1.75 Kv/mm electrical fields. The result showed that delivery efficiencies increased with elevated field strengths up to 1.25 Kv/mm and reached a maximum efficiency of 3.4% (Fig. 1B), which was 8-fold lower than the sonoporation method (Fig. 1A). To examine whether ultrasound has deleterious effects on cells, Thermoanaerobacter X514 cells were sonified for durations ranging from 0 to 40s and plated on agar plates without antibiotics. Results showed that cell viability was not compromised under the durations of exposure investigated (Fig. 3A). Interestingly, exposure durations between 10s and 20s resulted in 20% higher colony counts in cell viability assays, likely due to the dispersal of clumped cells under brief ultrasound treatments [17]. These data suggest that ultrasound can be used to efficiently deliver plasmid DNA into Thermoanaerobacters without incurring cell damage. Figure 3. Survival rate of Thermoanaerobacter sp. X514 after sonoporation or electroporation. A. Survival rate of X514 after sonoporation under exposure durations ranged from 0 to 40s. B. Survival rate of X514 after electroporation under field strengths ranged from 0 to 1.75 Kv/mm. Viable colonies were determined by plating cells on agar plates. The percent viability (survival rate) was calculated as described in Materials and Methods. doi:10.1371/journal.pone.0012582.g003 Figure 3. Survival rate of Thermoanaerobacter sp. X514 after sonoporation or electroporation. A. Survival rate of X514 after sonoporation under exposure durations ranged from 0 to 40s. B. Survival rate of X514 after electroporation under field strengths ranged from 0 to 1.75 Kv/mm. Viable colonies were determined by plating cells on agar plates. The percent viability (survival rate) was calculated as described in Materials and Methods. doi:10.1371/journal.pone.0012582.g003 September 2010 | Volume 5 | Issue 9 | e12582 3 Sonoporation of Thermophiles PLoS ONE | www.plosone.org September 2010 | Volume 5 | Issue 9 | e12582 Methylation modification assay gene in Thermoanaerobacter sp. X514 by sonoporation To test the efficacy of sonoporation in delivering heterologous double-stranded plasmids into Thermoanaerobacter, methylated pIKM2 (which encodes a clostridial b-1.4-endoglucanase gene, Fig S2A) or pHL015 (which encodes a jellyfish Aequorea Victoria gfp gene, Fig. S2B) were delivered into Thermoanaerobacter sp. X514 by sonoporation, and by electroporation in parallel. Efficiencies of sonoporation and electroporation were 66102 and 16102 CFU/mg plasmid DNA, respectively, based on the experimentally determined optimal conditions described above. Multiple experiments to confirm the transformation were carried out. The kanamycin resistance gene on pIKM2 was detected by PCR from the X514 colonies transformed via either sonoporation or electroporation (Fig. 5A). However, the pIKM2 DNA extracted from X514 transformants was not visible on EtBr-stained agarose gels (data not shown), likely due to low plasmid copy numbers in the host cell or inefficient recovery of pIKM2 from the Gram-positive X514 during plasmid DNA extraction. To address this ambiguity, ‘‘plasmid rescue’’ experiments was carried out. The plasmid pIKM2 extract from the Kmr X514 colonies was transformed into E. coli cells, followed by propagation and extraction from the E. coli. The XbaI/SmaI restriction digestion profile of the pIKM2 originally delivered into X514 was identical to that of the pIKM2 recovered from E. coli transformants (Fig. 5B), demonstrating that the pIKM2 derived from X514 was intact. The state of foreign DNA to be delivered also determines transformation efficiency. DNA methylation is a chemical modifica- tion system defending against foreign-DNA invasion. It was widely found in higher organisms, but varies in frequency in thermophilic bacteria. For instance, foreign DNA must be methylated before it can be introduced into C. thermocellum [2] via electroporation, while foreign-DNA methylation was not required for electro-transforming Thermoanaerobacterium saccharolyticum JW/SL-YS485 [1]. To determine whether enzymes from Thermoanaerobacter sp. X514 could degrade foreign plasmid DNA, cell extracts of X514 were incubated with various forms of pHL015 (Fig. S2B). Methylation- competent E. coli DH5a (Dam+/Dcm+), but not the methylation- deficiency mutant E. coli JM110 (Dam2/Dcm2) cells, methylated pHL015 in vivo. The methylated pHL015 extracted from E. coli DH5a was not degraded by X514 cell extracts while the unmethylated pHL015 extracted from E. coli JM110 was readily cleaved by X514 cell extracts (Fig. 4A, Lane 4 and Lane 8 respectively). In vitro methylated pHL015, extracted from either species of E. coli, resulted in no degradation by X514 cell extracts (Fig. Methylation modification assay 4A, Lane 5, Lane 10 and Lane 11 respectively). These findings suggest that methylation of DNA is required for protection against degradation in X514 cells. To examine whether the enzyme activity underlying plasmid degradation in X514 is buffer dependent, the degradation of Dam2/Dcm2 pHL015 was evaluated in the presence of the New England Biolabs (NEB) restriction enzyme Buffer 1, 2 3 and 4. Nonspecific degradation of pHL015 was observed in the presence of NEB Buffer 1, 2 and 4 (Fig. 4B, Lane 2, 3 and 5), suggesting buffer dependency of the enzyme activity. To identify the putative restriction methylation (RM) system(s) in X514, we took advantage of the complete genome sequence of Thermoanaerobacter X514 (Genbank ID: NC010320). From the genome, a total of three putative genes encoding DNA methyltransferases (Teth514_0458/1737/1828) were identified, which, based on our functional annotations, all methylate adenines at their N6 positions in GATC sequences [20]. We hypothesized that the GATC sequences are one signal for cleavage recognized by the X514 RM systems. To test this hypothesis, we compared the gel electrophoresis patterns of pHL015 (methylated or To determine whether enzymes from Thermoanaerobacter sp. X514 could degrade foreign plasmid DNA, cell extracts of X514 were incubated with various forms of pHL015 (Fig. S2B). Methylation- competent E. coli DH5a (Dam+/Dcm+), but not the methylation- deficiency mutant E. coli JM110 (Dam2/Dcm2) cells, methylated pHL015 in vivo. The methylated pHL015 extracted from E. coli DH5a was not degraded by X514 cell extracts while the unmethylated pHL015 extracted from E. coli JM110 was readily cleaved by X514 cell extracts (Fig. 4A, Lane 4 and Lane 8 respectively). In vitro methylated pHL015, extracted from either species of E. coli, resulted in no degradation by X514 cell extracts (Fig. 4A, Lane 5, Lane 10 and Lane 11 respectively). These findings suggest that methylation of DNA is required for protection against degradation in X514 cells. To examine whether the enzyme activity underlying plasmid degradation in X514 is buffer dependent, the degradation of Dam2/Dcm2 pHL015 was evaluated in the presence of the New England Biolabs (NEB) restriction enzyme Buffer 1, 2 3 and 4. Nonspecific degradation of pHL015 was observed in the presence of NEB Buffer 1, 2 and 4 (Fig. 4B, Lane 2, 3 and 5), suggesting buffer dependency of the enzyme activity. To determine whether enzymes from Thermoanaerobacter sp. Sonoporation of Thermophiles that specifically cleaves GATC when the A is not methylated. While X514 extracts did not degrade in vivo or in vitro methylated pHL015, DpnII similarly did not degrade methylated pHL015 from E. coli DH5a (Fig. 4A, Lane 3). However, treatments by X514 extracts or DpnII yield identical restriction profiles for unmethylated pHL015 from E. coli JM110 (Fig. 4A, Lane 7 and Lane 8 respectively). The results suggest that X514 cells harbor at least one restriction enzyme that digests unmethylated GATC sequences. This finding was consistent with the predicted function of the three N6 adenine-specific DNA methyltransferases found in X514 genome. A DpnII-like isoschizomer was previously predict- ed in REBASE of NEB (http://rebase.neb.com/rebase/rebase. html) [21] with an official name of TspX514ORF458P and its companion methylase M.TspX514ORF458P (The letter ‘‘P’’ indicates that the function is still putative). Taken together, it is concluded that the adenine methylation modification system in Thermoanaerobacter sp. X514 was active and is likely the major restriction modification system in this organism, suggesting that methylation of foreign DNA by specific methylases or cell extracts [22] is essential prior to introducing into X514 cells. The post-electroporation survival rate of X514 was also investi- gated by plating after electroporation under field strengths ranging from 0 to 1.75 Kv/mm. The result showed that the survival rate was drastically reduced under elevated voltages (Fig. 3B), indicating that cell membranes might undergo irreversible structural changes and cell damage ensued under high electrical field strength. Ultrasound, at higher intensity (e.g. 400,1000 W) or longer exposure, has been widely applied to DNA fragmentation in aqueous solutions as ultrasound can break hydrogen bonds and rapture single- or double-strands in the helix [19]. To test the impact of ultrasound on the plasmid DNA to be delivered, the plasmid pIKM1 was sonified for durations ranging from 0 to 40s. We found that plasmid integrity was not compromised by sonification at the durations of exposure investigated, as determined by pixel densitometry of supercoiled and linearized DNA bands on the gel (Fig. S1). Compared with electroporation, the higher cell viability (Fig. 3A) under sonification and the apparent preservation of post-sonification plasmid integrity (Fig. S1) were consistent with the much higher percentage of Taxas-red-transformed fluorescent cells observed under sonoporation (Fig. 1A). These results showed that sonoporation could be superior to electroporation in transformation efficiency and cell viability preservation. Methylation modification assay The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts (Lanes 4, 5, 8, 10 and 11) or DpnII (Lanes 3, 7 and 9). In vitro methylated (IvM) plasmids are marked with ‘‘+’’ on top of this figure (Lanes 5 and 11: methylated by X514 cell extracts; Lanes 9 and 10: methylated by a commercial Dam methyltransferase). B. The pHL015 plasmids (Lane 2 to 7) were extracted from Dam2/Dcm2 E. coli JM110. The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts in the NEB restriction enzyme buffer 1, 2, 3 and 4 in parallel (Lane 2 to 5). The NEB buffers used are illustrated on top of this figure. M: molecular weight marker. N/A: not applicable. doi:10.1371/journal.pone.0012582.g004 Figure 4. Agarose gel electrophoresis of pHL015 plasmids employed in methylation assay. A. The pHL015 plasmids (Lanes 2 to 11) were extracted from either Dam+/Dcm+ E. coli DH5a or Dam2/Dcm2 E. coli JM110. The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts (Lanes 4, 5, 8, 10 and 11) or DpnII (Lanes 3, 7 and 9). In vitro methylated (IvM) plasmids are marked with ‘‘+’’ on top of this figure (Lanes 5 and 11: methylated by X514 cell extracts; Lanes 9 and 10: methylated by a commercial Dam methyltransferase). B. The pHL015 plasmids (Lane 2 to 7) were extracted from Dam2/Dcm2 E. coli JM110. The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts in the NEB restriction enzyme buffer 1 2 3 and 4 in parallel (Lane 2 to 5) The NEB buffers used are illustrated on top of this figure M: molecular Figure 4. Agarose gel electrophoresis of pHL015 plasmids employed in methylation assay. A. The pHL015 plasmids (Lanes 2 to 11) were extracted from either Dam+/Dcm+ E. coli DH5a or Dam2/Dcm2 E. coli JM110. The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts (Lanes 4, 5, 8, 10 and 11) or DpnII (Lanes 3, 7 and 9). In vitro methylated (IvM) plasmids are marked with ‘‘+’’ on top of this figure (Lanes 5 and 11: methylated by X514 cell extracts; Lanes 9 and 10: methylated by a commercial Dam methyltransferase). B. The pHL015 plasmids (Lane 2 to 7) were extracted from Dam2/Dcm2 E. coli JM110. Methylation modification assay The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts in the NEB restriction enzyme buffer 1, 2, 3 and 4 in parallel (Lane 2 to 5). The NEB buffers used are illustrated on top of this figure. M: molecular weight marker. N/A: not applicable. doi:10.1371/journal.pone.0012582.g004 rule out this possibility, we sought direct evidence for successful transformation by expression of a heterologous (clostridial) b-1.4- endoglucanase gene in X514 and detection of the expressed endoglucanase activity in the presence of cellulose (CMC-Na). Result showed that X514 transformed with plasmid pIKM2 by sonopora- tion or by electroporation exhibited endoglucanase activities of 0.87 U/mg and 0.95 U/mg respectively, whereas the X514 wild type and X514 transformed with the pIKM1 plasmid (one without the clostridial b-1.4-endoglucanase gene) could not degrade cellulose (Fig. 6). These results demonstrated that the pIKM2 encoding b-1.4- endoglucanase was indeed delivered via sonoporation into and functionally expressed in Thermoanaerobacter sp. X514. conjugation. However, the lack of suitable donor strains makes conjugation difficult for thermophilic low G/C organisms such as Thermoanaerobacter. Thus most of the described gene transfer protocols for these types of organisms are based on electroporation [1,2,5,24]. However, success with these methods was limited to only a few laboratories, which used custom-built cuvettes and pulse generator [2,3] that are not accessible to most laboratories. In this study, we developed an efficient and quick transformation system for Gram-positive thermophiles via ultrasound-based sonoporation. The method was successfully employed in Thermo- anaerobacter X514 with a transformation efficiency of 66102 CFU/ mg plasmid DNA. The efficiency is much higher than that from a parallelly performed, yet much slower and more tedious, electroporation method (16102 CFU/ug plasmid). PLoS ONE | www.plosone.org Methylation modification assay This result suggests that albeit GFP protein is not a suitable marker gene for thermoanaerobes, the transformation of X514 might have been successful (Fig. 5C). To identify the putative restriction methylation (RM) system(s) in X514, we took advantage of the complete genome sequence of Thermoanaerobacter X514 (Genbank ID: NC010320). From the genome, a total of three putative genes encoding DNA methyltransferases (Teth514_0458/1737/1828) were identified, which, based on our functional annotations, all methylate adenines at their N6 positions in GATC sequences [20]. We hypothesized that the GATC sequences are one signal for cleavage recognized by the X514 RM systems. To test this hypothesis, we compared the gel electrophoresis patterns of pHL015 (methylated or unmethylated) resulted from two treatments, one by X514 cellular extracts and the other by DpnII, a commercially available enzyme However, the possibility that plasmids were surface-attached with, but not ‘into’, the X514 cell could not be excluded. In fact, excess amounts of plasmid DNA (5 mg) were used for each X514 transformation and spread together with cells on agarose plates. To PLoS ONE | www.plosone.org September 2010 | Volume 5 | Issue 9 | e12582 September 2010 | Volume 5 | Issue 9 | e12582 4 Sonoporation of Thermophiles Figure 4. Agarose gel electrophoresis of pHL015 plasmids employed in methylation assay. A. The pHL015 plasmids (Lanes 2 to 11) were extracted from either Dam+/Dcm+ E. coli DH5a or Dam2/Dcm2 E. coli JM110. The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts (Lanes 4, 5, 8, 10 and 11) or DpnII (Lanes 3, 7 and 9). In vitro methylated (IvM) plasmids are marked with ‘‘+’’ on top of this figure (Lanes 5 and 11: methylated by X514 cell extracts; Lanes 9 and 10: methylated by a commercial Dam methyltransferase). B. The pHL015 plasmids (Lane 2 to 7) were extracted from Dam2/Dcm2 E. coli JM110. The plasmids were separated on 1% agarose gel after incubation with X514 cell extracts in the NEB restriction enzyme buffer 1, 2, 3 and 4 in parallel (Lane 2 to 5). The NEB buffers used are illustrated on top of this figure. M: molecular weight marker. N/A: not applicable. doi:10.1371/journal.pone.0012582.g004 Figure 4. Agarose gel electrophoresis of pHL015 plasmids employed in methylation assay. A. The pHL015 plasmids (Lanes 2 to 11) were extracted from either Dam+/Dcm+ E. coli DH5a or Dam2/Dcm2 E. coli JM110. The advantages of ultrasound-based sonoporation The most commonly used methods of DNA transformation in bacteria are conjugation and electroporation. Typically, a specific DNA donor strain is required to achieve DNA transfer in There are several advantages of sonoporation-based over electroporation-based transformation. Firstly, the procedure and September 2010 | Volume 5 | Issue 9 | e12582 5 Figure 5. Confirmation of transformation of Thermoanaerobacter cells using PCR analysis and plasmid rescue experiments. A. PCR amplication of the kanamycin resistance marker-gene from pIKM2. The PCR products were amplified from DNA preparations of pIKM2-transformed X514 cells by sonoporation (sono-) or electroporation (electro-). The plasmid pIKM2 and wild-type X514 DNA were used as PCR template for positive (P) and negative (N) controls, respectively. B. Plasmid-pIKM2 rescue experiments. The plasmid DNA were extracted from X514-pIKM2 transformants, and ‘‘rescued’’ in E. coli (see Methods). The E. coli-rescued plasmids were digested by XbaI /SmaI and separated on 1% agarose gel. Plasmid pIKM2 was used as positive control (P). C. Plasmid-pHL015 rescue experiments. The plasmid DNA were extracted from X514-pHL015 transformants, and rescued in E. coli. The E. coli cells harboring rescued plasmids were visualized under Olympus-BX51 microscope equipped with a CCD. Scale bar represents 2 mm. E. coli/pHL015-S: the E.coli-rescued plasmid that was transformed into X514 by sonoporation. E. coli/pHL015-E: the E.coli-rescued plasmid that was transformed into X514 by electroporation. ‘‘M’’: molecular weight marker. d i 10 1371/j l 0012582 005 Sonoporation of Thermophiles Sonoporation of Thermophiles Figure 5. Confirmation of transformation of Thermoanaerobacter cells using PCR analysis and plasmid rescue experiments. A. PCR amplication of the kanamycin resistance marker-gene from pIKM2. The PCR products were amplified from DNA preparations of pIKM2-transformed X514 cells by sonoporation (sono-) or electroporation (electro-). The plasmid pIKM2 and wild-type X514 DNA were used as PCR template for positive (P) and negative (N) controls, respectively. B. Plasmid-pIKM2 rescue experiments. The plasmid DNA were extracted from X514-pIKM2 transformants, and ‘‘rescued’’ in E. coli (see Methods). The E. coli-rescued plasmids were digested by XbaI /SmaI and separated on 1% agarose gel. Plasmid pIKM2 was used as positive control (P). C. Plasmid-pHL015 rescue experiments. The plasmid DNA were extracted from X514-pHL015 transformants, and rescued in E. coli. The E. coli cells harboring rescued plasmids were visualized under Olympus-BX51 microscope equipped with a CCD. Scale bar represents 2 mm. E. coli/pHL015-S: the E.coli-rescued plasmid that was transformed into X514 by sonoporation. E. September 2010 | Volume 5 | Issue 9 | e12582 The advantages of ultrasound-based sonoporation Therefore, the copy number is dependent on the specific strains and plasmids used. Thirdly, the paucity of thermostable antibiotics and their corresponding resistance marker-genes have limited the genetic manipulation of thermophiles. Without such thermostable genetic marker systems, false positive colonies would grow on plates after transformation (data not shown). Thus, in this study we attempted a fluorescence-based reporter system by con- structing pHL015 which harbors a gfp reporter gene. However, GFP was not functional in the thermoanaerobic X514 due to the high temperature or the lack of molecular oxygen [23]. Another plasmid encoding the fluorescent marker PpFbFP, which was reported functional at both aerobic and anaerobic conditions [30], quickly quenched at temperatures above 40uC (data not shown). Eventually, we developed the reporter system consisting of clostridial b-1,4-glucanase gene whose expression was driven by a Thermoanaerobacter-compatible promoter. With this strategy, not only transformation but also heterologous expression were directly and conclusively validated via measur- ing the enzymatic activity in the transformed host cells. It is note-worthy though, that as detection and measurement of enzyme activities is typically indirect and much more cumber- some than those of in situ fluorescent signals, identification and development of fluorescent-protein marker-genes that are functional in thermophilic anaerobes is warranted. Figure 6. Heterologous endoglucanase activity of Thermoanaer- obter sp. X514 transfomants. The endoglucanse activity was determined as described in Materials and Methods. The X514 was transformed by electroporation (E) or sonoporation (S) respectively. Wild-type X514 and pIKM1 transformants were used as negative controls. All experiments were performed in triplicate, with standard deviations shown. doi:10.1371/journal.pone.0012582.g006 doi:10.1371/journal.pone.0012582.g006 moreover, the whole process has to take place in a specialized anaerobic glove box. Thirdly, supported by our study, ultrasound treatment does not reduce viability as cells can quickly repair membranes, whereas cell survival is drastically reduced after electroporation due to irreversible membrane damage [25]. Fourthly, sonoporation eliminates the need for expensive or specialized instruments and facilities. The only essential equipment is a commercially available and inexpensive ultrasound bath, in this study a Branson B200 (40 kHz, 19W). While in electropora- tion of obligate Gram-positive anaerobes such as some thermo- philic Clostridia, custom-made electroporators that are until now not accessible to most laboratories are required [1,3], and a large anaerobic workstation for competent cells preparation and applying electro-pulses has to be equipped. The advantages of ultrasound-based sonoporation These advantageous features of sonoporation-based transfor- mation suggested that this method can be readily adapted for high- throughput or remotely controlled screening and testing of genetic transformation schemes, where a large number of transformation experiments are performed in parallel, with or without robots. Furthermore, the ultrasonic DNA transfer is highly scalable and can be applied to large bioreactors and natural environments. This key feature could find potential applications for outdoors, large- scale, DNA-transformation-based modulation of microbial-com- munity structure and function. Additional barriers for DNA transformation towards TGPAs PLoS ONE | www.plosone.org The advantages of ultrasound-based sonoporation With this strategy, not only transformation but also heterologous expression were directly and conclusively validated via measur- ing the enzymatic activity in the transformed host cells. It is note-worthy though, that as detection and measurement of enzyme activities is typically indirect and much more cumber- some than those of in situ fluorescent signals identification and increased transformation efficiency using unmethylated DNA is rare [28], the foreign genetic materials should be methylated prior to delivery to prevent cleavage by the RM systems. Without a complete genome, identification of the specific methylase for DNA modification to resist restriction enzymes in the host cells is typically a tedious and cumbersome trial- and-error exercise. Here, we demonstrated a whole-genome- sequence-based strategy to quickly identify and confirm the particular restriction systems. A DpnII-like restriction system which digests unmethylated 59-GATC-39 sequence was identi- fied in Thermoanaerobacter sp. X514. Together with the previously reported MboI-like RM systems in C. thermocellum [2,29], our finding suggested such RM systems might be one common barrier to transformation in TGPAs. The exonuclease activities in thermophiles are species/strain, temperature and buffer dependent. For instance, in contrast to Clostridium thermosacchar- olyticum ATCC 31960, little nonspecific exonuclease activity was observed in C. thermocellum ATCC27405 (clear bands on agarose gel; [29]). Even for C. thermocellum ATCC27405, nonspecific degradation of linear DNA was observed at 63uC but not at 55uC or 60uC [29]. In our study, we observed nonspecific degradation of unmethylated plasmid by X514 cell extracts in the presence of NEB Buffer 1, 2 and 4, but not 3 (Fig. 4B). Secondly, the replication origin of transformed plasmid DNA is another important factor influencing plasmid stability inside the hosting thermophilic cells. To date, no native plasmid replication origin has been identified in thermophiles. The exogenous plasmid replicon used in Thermoanaerobacter and C. thermocellum was isolated from the mesophilic Bacillus subtilis plasmid pIM13 [1], which might result in low copy-number of the plasmids in thermophiles. The low copy-number of pIKM1, which was derived from pIM13, could underlie our inability to visualize the isolated plasmid from transformed Thermoanaer- obacter on agrose gel. This result was consistent with the observations in C. thermocellum [2]. However, the isolated pIKM1 could be gel-visualized from the transformed Thermoanaer- obacterium [1]; while plasmid pTE16, constructed not from Bacillus subtilis but from the replicon origin of Clostridium perfringens plasmid pIP404, could be abundantly isolated from transformed Thermoanaerobacter [5]. The advantages of ultrasound-based sonoporation coli/pHL015-E: the E.coli-rescued plasmid that was transformed into X514 by electroporation. ‘‘M’’: molecular weight marker. doi:10.1371/journal.pone.0012582.g005 to remote-control. The acoustic cavitation of microbubbles can be directly applied to indigenous cells in their native growth media, where the microbubbles permeate cell membranes and facilitate the delivery of DNA and other macromolecules [17]. Cellular manipulation can be achieved outside of an anaerobic glove box simply by immersing a sealed glass vial into a sonifier chamber, without any cell transfers or direct contact of the cells with anything other than the indigenous growth media. In contrast, in electroporation, cells must be removed from their native growth medium, exposed to low-ionic strength buffer, transferred to and held in a specialized electro-pulse cuvette for electro-pulsing; protocol for the former is much less cumbersome and easer to follow (See Materials and Methods). In contrast, the later necessitates adding glycine to the medium to induce spheroplast formation [5], or by using cell wall-weakening agents such as isonicotinic acid hydrazide [1,2]; moreover, laborious pretreat- ment steps involving chilling during cell collection and repetitious washing under anaerobic conditions are required. Thus sonopora- tion is particularly advantageous for handling strictly anaerobic bacteria, as the likelihood of exposing the cells to aerobic environments is minimized. Secondly, ultrasound sonoporation is a non-cell-contacting and less invasive method that is thus friendly PLoS ONE | www.plosone.org September 2010 | Volume 5 | Issue 9 | e12582 6 Sonoporation of Thermophiles Figure 6. Heterologous endoglucanase activity of Thermoanaer- obter sp. X514 transfomants. The endoglucanse activity was determined as described in Materials and Methods. The X514 was transformed by electroporation (E) or sonoporation (S) respectively. Wild-type X514 and pIKM1 transformants were used as negative controls. All experiments were performed in triplicate, with standard deviations shown. doi:10.1371/journal.pone.0012582.g006 increased transformation efficiency using unmethylated DNA is rare [28], the foreign genetic materials should be methylated prior to delivery to prevent cleavage by the RM systems. Without a complete genome, identification of the specific methylase for DNA modification to resist restriction enzymes in the host cells is typically a tedious and cumbersome trial- and-error exercise. Here, we demonstrated a whole-genome- sequence-based strategy to quickly identify and confirm the particular restriction systems. A DpnII-like restriction system which digests unmethylated 59-GATC-39 sequence was identi- fied in Thermoanaerobacter sp. X514. Together with the previously reported MboI-like RM systems in C. The advantages of ultrasound-based sonoporation thermocellum [2,29], our finding suggested such RM systems might be one common barrier to transformation in TGPAs. The exonuclease activities in thermophiles are species/strain, temperature and buffer dependent. For instance, in contrast to Clostridium thermosacchar- olyticum ATCC 31960, little nonspecific exonuclease activity was observed in C. thermocellum ATCC27405 (clear bands on agarose gel; [29]). Even for C. thermocellum ATCC27405, nonspecific degradation of linear DNA was observed at 63uC but not at 55uC or 60uC [29]. In our study, we observed nonspecific degradation of unmethylated plasmid by X514 cell extracts in the presence of NEB Buffer 1, 2 and 4, but not 3 (Fig. 4B). Secondly, the replication origin of transformed plasmid DNA is another important factor influencing plasmid stability inside the hosting thermophilic cells. To date, no native plasmid replication origin has been identified in thermophiles. The exogenous plasmid replicon used in Thermoanaerobacter and C. thermocellum was isolated from the mesophilic Bacillus subtilis plasmid pIM13 [1], which might result in low copy-number of the plasmids in thermophiles. The low copy-number of pIKM1, which was derived from pIM13, could underlie our inability to visualize the isolated plasmid from transformed Thermoanaer- obacter on agrose gel. This result was consistent with the observations in C. thermocellum [2]. However, the isolated pIKM1 could be gel-visualized from the transformed Thermoanaer- obacterium [1]; while plasmid pTE16, constructed not from Bacillus subtilis but from the replicon origin of Clostridium perfringens plasmid pIP404, could be abundantly isolated from transformed Thermoanaerobacter [5]. Therefore, the copy number is dependent on the specific strains and plasmids used. Thirdly, the paucity of thermostable antibiotics and their corresponding resistance marker-genes have limited the genetic manipulation of thermophiles. Without such thermostable genetic marker systems, false positive colonies would grow on plates after transformation (data not shown). Thus, in this study we attempted a fluorescence-based reporter system by con- structing pHL015 which harbors a gfp reporter gene. However, GFP was not functional in the thermoanaerobic X514 due to the high temperature or the lack of molecular oxygen [23]. Another plasmid encoding the fluorescent marker PpFbFP, which was reported functional at both aerobic and anaerobic conditions [30], quickly quenched at temperatures above 40uC (data not shown). Eventually, we developed the reporter system consisting of clostridial b-1,4-glucanase gene whose expression was driven by a Thermoanaerobacter-compatible promoter. In vitro methylation of DNA The DNA modification assay was performed as described elsewhere [22] by combining 53 ml TNE buffer (50 mM Tris pH 7.5, 50 mM NaCl, 10 mM EDTA), 10 ml SAM (0.8 mM), 1 ml BSA (10 mg/ml), 25 ml X514 cell extract and 3 mg plasmid DNA (pHL015) prepared from E. coli DH5a (Dam+/Dcm+) or JM110 (Dam2/Dcm2) in a final volume of 100 ml. The mixture was incubated at 60uC for 16 h. As a control for methylation experiments, a commercial Dam methyltransferase was purchased from New England Biolabs. When the commericial Dam methyltransferase was used, the reaction condition was 37uC for 4 h (based on the enzyme product manual). Methylated DNA fragments were purified by the QIAquick PCR purification kit (Qiagen). Application of sonoporation pp p The sonoporation method and reporter systems developed and validated in this study can be used as a generalized strategy to over-express foreign or native genes in thermophilic anaerobes, for example, by replenishing the carbon flow to ethanol fermentation pathway [9]. Apart from that, they might also be a useful for creating gene deletions involved in branched fermentation pathways. Traditional methods use suicide vectors to achieve in- frame deletions via homologous recombination. However, this requires much higher transformation efficiencies, usually above ,105 CFU/mg DNA [31]. Although the transformation efficiency of sonoporation is comparatively low, sonoporation has been successfully used to construct allelic-exchange mutant in the Gram-negative bacterium Fusobacterium nucleatum [18]. Further- more, there are gene-knock-out methods that might exploit the characteristics of sonoporation-based transformation. For exam- ple, the recently developed ClosTron, a gene-knock-out system alternative to homologous recombination, is based on Group II introns and does not require a high transformation efficiency [32– 35]. To date, however, this knock-out system has merely been used in mesophilic organisms [33,34,35]. Sonoporation-based transfor- mation and ClosTron, when coupled together, might be a promising strategy for gene-knockouts in thermophiles. Further optimization of this protocol to increase transformation efficiencies in thermophiles or Gram-positive bacteria may include optimizing sonoporation parameters (e.g, acoustic pressure amplitude, sonoporation frequency and intensity) or addition of calcium and Definity to control pore size of cell membrane [36,37,38]. Delivery of fluorescent dextran into Thermoanaerobacter sp. X514 by electroporation For the preparation of electro-competent host cells, Thermo- anaerobacter sp. X514 was grown in 100 ml N2-flushed medium. Glycine and sucrose were added into the early log-phase culture with final concentrations of 0.053 M and 0.27 M respectively. The culture was propagated at 60uC until the spheroplast morphology could be visualized under a light microscopy (Fig. S3). Cells were harvested by centrifugation at 3,500 g for 10 min at 4uC, washed three times in an anaerobic chamber using chilled electroporation buffer (0.27 M sucrose and 10% glycerol). Such electro-competent cells were resuspended in 2 ml of the same buffer. Texas Red- conjugated dextran was added to the cell suspension for a final concentration of 1.25 mg/ml. Approximately 100 ml of the mixture was transferred to a cuvette (Bio-Rad, 2 mm) and electro-pulsed with 10 mF, 600 V and voltages ranging from 0.75 Kv/mm to 1.75 Kv/mm, using a Bio-Rad Micro-Pulser. The cells were then washed four times in electroporation buffer and examined under a fluorescence microscope, as described above. In conclusion, this study marks the development of a sonoporation-based transformation method for TGPAs. Our work expands the presently limited repertoire of genetic tools for modifying and engineering TGPAs. Furthermore, the advantages of sonoporation suggested it could be readily adapted and particularly suitable for applications where in situ cell-transforma- tion, high throughput or remote control is preferred or even required. In the present days, for the microbial world in general and for thermophilic Gram-positive bacteria in particular, the much fewer number of currently available genetics systems contrasts sharply with the rapid and readily availability of sequenced genomes. Sonoporation-based transformation hold the promise to significantly elevate the throughput of genetic system development and contribute to bridging the enormous gap between genome sequences and manipulation of the genomes. Examination of restriction endonuclease activity in host cells Cell extracts of Thermoanaerobacter sp. X514 were prepared from stationary-phase cells as described previously with minor modifi- cations [22]. Briefly, cells from 100 ml stationary-phase culture of X514 were collected by centrifugation (3,500 g for 10 min), washed once with 100 ml chilled PENP buffer (10 mM KH2PO4- K2HPO4, 10 mM EDTA, 50 mM NaCl and 0.2 mM PMSF, pH 7.0), and then resuspended in 4 ml of the same buffer. The cells were chilled and maintained in an ice bath during cell- structure disruption in a Branson 450 sonifier (400W). Cellular debris was separated by centrifugation (10,000 g for 15 min) and the extract (supernatant) collected. Aliquots of 3 ml extract were mixed with 3 ml 100% glycerol and 0.6 ml of BSA (1 mg/ml) and then stored at 220uC before analysis. The DNA digestion assay was performed by mixing 0.5–1.6 mg of plasmid DNA, 4 ml cell extract and appropriate restriction endonuclease buffer to a final volume of 25 ml, and then incubated at 60uC for 4 h. Bacterial strains and growth conditions Bacterial strains and growth conditions Thermoanaerobacter sp. X514 was cultivated in reinforced clostridial medium (RCM, Oxoid) anaerobically at 60uC. For growth on plates, 0.7% Gelrite Gellan Gum (Sigma) was added to RCM medium. Escherichia coli strains were grown in Luria-Bertani (LB) medium. When required, the medium was supplemented with kanamycin, 250 mg/ml for Thermoanaerobacter sp. X514 and 50 mg/ml for E. coli. Additional barriers for DNA transformation towards TGPAs The unique physiological features of TGPAs represent additional barriers for DNA transformation. Firstly, the restriction-methylation systems (RM) of bacterial cells may digest improperly methlyated incoming DNA [26,27], which renders low transformation efficiency in recalcitrant hosts. As September 2010 | Volume 5 | Issue 9 | e12582 September 2010 | Volume 5 | Issue 9 | e12582 7 Sonoporation of Thermophiles device (Branson B200, 40 kHz, 19W) and exposed to ultrasonic waves for durations ranging from 0 to 40 seconds at room temperature. Cells were washed for four times in 500 ml PBS buffer, collected by centrifugation at 4,000 g 610 min and then were examined under an Olympus-BX51 fluorescence micro- scope. Viable counts were determined as the number of colony forming units (CFU) after plating diluted cells on RCM-Gelrite agar plates. Percent viability was calculated as viable counts divided by counts in a control, i.e., without sonoporation exposure. Materials and Methods Bacterial strains and growth conditions Delivery of methylated pIKM2 and pHL015 plasmids into Thermoanaerobacter sp. X514 by sonoporation and electroporation Delivery of methylated pIKM2 and pHL015 plasmids into Thermoanaerobacter sp. X514 by sonoporation and electroporation Found at: doi:10.1371/journal.pone.0012582.s001 (0.46 MB TIF) p For sonoporation-based transformation, 500 ml cell suspensions at approximately 16109 CFU/ml were mixed with in vitro methylated pIKM2 or pHL015 (5 mg) plasmids in a flat-bottom glass vial. The vial was placed at the center of the ultrasound device and exposed to 40 kHz ultrasound for 20 seconds. The mixture was then transferred to a Hungate tube containing 5 ml fresh RCM medium. Post-sonoporation cells were allowed to recover at 45uC for 5 h and then were diluted, mixed with melted warm RCM-agar medium supplemented with 250 mg/ml kana- mycin and poured onto Petri dishes inside an anaerobic glove box (Coy Lab). The agar plates were incubated for 5–7 days in an anaerobic jar at 45uC. The temperature is not raised to the optimal growth temperature of X514 (60uC), in order to preserve the activity of kanamycin. Figure S2 Physical maps of plasmid pIKM2 and pHL015. A. pIKM2. The plasmid pIKM2 was constructed by inserting the beta-1, 4-endoglucanase gene into Xba I/SmaI sites of pIKM1 (1) as described in text. B. pHL015. The plasmid pHL015 was constructed by inserting the NcoI/NsiI flanked fragment of pIKM1 into a similarly digested plasmid pML523. The pHL015 harbors a kanamycin resistance gene (aph, from pIKM1), a Bacillus subtilis replicon (pIM-ori, from pIKM1), a hygromycin resistance gene (hyg), a green fluorescence protein gene (gfp), a catechol 2,3-dioxygenase gene (xylE), a levansucrase gene (sacB) and an E. coli replicon (pUC-ori). Found at: doi:10.1371/journal.pone.0012582.s002 (0.24 MB TIF) For electroporation-based transformation, methylated pIKM2 or pHL015 DNA (5 mg) was mixed with 100 ml X514 competent cells as describe above, added into a cuvette (Bio-Rad, 2 mm) and pulsed in a Bio-Rad micropulser (1.25 Kv/mm, 10 mF and 600 V). The cells were then immediately transferred into Hungate tubes with 5 ml fresh RCM medium, incubated at 45uC for 5 h, diluted and mixed with melted warm RCM-agar medium supplemented with 250 mg/ml kanamycin, and then plated onto Petri dishes inside an anaerobic glove box. The agar plates were incubated for 5–7 days in an anaerobic jar at 45uC. Figure S3 Morphology of Thermoanaerobacter sp. X514 cells. A: X514 cells at early growth phase. B: Spheroplast formation of X514 cells in the presence of glycine and sucrose. Photographs were taken using Olympus-BX51 microscope equipped with a CCD. Delivery of fluorescent dextran into Thermoanaerobacter sp. X514 by sonoporation Thermoanaerobacter sp. X514 cells were harvested at mid-log phase, washed and resuspended in phosphate-buffered saline (PBS, pH 7.0) buffer supplemented with 0.1 mM MgCl2 and 0.1 mM CaCl2 [18]. The cell suspension at 109 cells/ml was then mixed with Texas Red-conjugated dextran (70 kDa; Invitrogen) to a final concentration of 1.25 mg/ml in a flat-bottom glass vial. The vials were immersed in the water bath at the center of the ultrasound PLoS ONE | www.plosone.org September 2010 | Volume 5 | Issue 9 | e12582 8 Sonoporation of Thermophiles km-rev (59-CAAATTCCTCGTAGGCGCTC-39) with an expect- ed PCR product size of 1470 bp. In addition, the presence of pIKM2 in Kmr transformants of X514 was verified by ‘‘DNA rescue’’ experiments. In these experiments, all extra-chromosomal elements were isolated from X514 and then introduced into and propagated in E. coli DH5a; all plasmid DNA was isolated from the E. coli using the Qiagen plasmid mini kit, digested by XbaI/ SmaI, and separated on 1% agarose gels. Construction of plasmid pIKM2 and pHL015 Construction of plasmid pIKM2 and pHL015 p p p E. coli-Thermoanaerobacter shuttle plasmid pIKM1 was kindly provided by Dr. Juergen Wiegel (University of Georgia, Athens). A 5.4 kb PCR fragment encoding the 4.5 kb endo-b-1, 4-glucanase gene flanked by its 651 bp promoter and 142 bp transcription terminator region was amplified from the genome of Clostridium thermocellum ATCC27405 by a primer-pair (fwd: 59- CCGTCTA- GAATTTTTGCGTATGGATTGTG-39 and rev: 59- GGTC- CCGGGCTGTTTCCCGTTTTTGTTTT-39) and subsequently ligated into pIKM1 through XbaI and SwaI. The resulting plasmid was named pIKM2 (Fig. S2A). The plasmid pHL015 (Fig. S2B) was constructed by inserting the NcoI/NsiI flanked fragment of pIKM1 into similarly digested plasmid pML523, which was a gift from Dr. Michael Niederweis (University of Alabama at Birmingham). Delivery of methylated pIKM2 and pHL015 plasmids into Thermoanaerobacter sp. X514 by sonoporation and electroporation Scale bar represents 10 mm. Found at: doi:10.1371/journal.pone.0012582.s003 (1.13 MB TIF) Acknowledgments We thank Yunfeng Yang for helpful discussion and Christoper L. Hemme, Jason Huff and Aifen Zhou for critical reading of the manuscript. After 5–7 days, single colonies were picked, transferred into RCM-kanamycin liquid medium and incubated at 45uC for 3 days. Plasmid DNA was extracted from Thermoanaerobacter sp. X514 transformants using a Qiagen plasmid mini-preparation kit. PCR was performed to detect the kanamycin marker gene in pIKM2 transformants using the extracted DNA as template and the primers km-fwd (59-CGATAAACCCAGCGAACCAT-39) and Plasmid DNA integrity assay To examine whether sonoporation compromised the integrity of the to-be-delivered foreign DNA, approximately 5 mg of pIKM1 DNA were added to 500 ml PBS buffer in a flat-bottom glass vial. The vials were exposed to 40 KHz ultrasound for durations from 0s to 40s. After sonoporation, DNA samples were transferred to 1.5 ml Eppendorf tubes containing 50 ml 3M NaAC and 1 ml 100% ethanol. Samples were incubated at 220uC for 2 h and then centrifuged at 12,000 g for 20 min. Supernatants were discarded and DNA pellets were dried in air. The DNA pellets were then dissolved in 20 ml nuclease-free water and separated on a 1% agarose gel. Enzyme activity assay of the introduced heterologous endo-b-1, 4-glucanase gene The b-endoglucanase activity was measured as the amount of reduced sugars released from sodium carboxymethyl cellulose (CMC-Na) in the presence of supernatant of the X514/pIKM2 cell-culture filtrate, as the enzyme was secreted. Concentrations of the released sugar were determined by the dinitrosalicylic acid (DNS) method using glucose as the standard, as previously described [39]. One enzyme unit was defined as the amount of protein required to release 1 mmol reduced sugar per minute from the CMC at 50uC. Specific activities are given in units per mg of protein. Protein concentrations were measured by the Bradford assay [40]. Supporting Information Figure S1 Impact of ultrasound exposure times on plasmid integrity. The plasmid pIKM1 was examined on 1% agarose gel via electrophoresis, after ultrasound exposure with durations ranged from 0 to 40s in a Branson B200 sonifier. The optimal duration of exposure investigated in this study is indicated with arrow. PLoS ONE | www.plosone.org References He Q, Lokken PM, Chen S, Zhou J (2009) Characterization of the impact of acetate and lactate on ethanolic fermentation by Thermoanaerobacter ethanolicus. Bioresour Technol 100: 5955–5965. y gy 28. Macaluso A, Mettus AM (1991) Efficient transformation of Bacillus thuringiensis requires nonmethylated plasmid DNA. J Bacteriol 173: 1353–1356. 29. Klapatch TR, Demain AL, Lynd LR (1996) Restriction endonuclease activity in Clostridium thermocellum and Clostridium thermosaccharolyticum. Appl Microbiol Biotechnol 45: 127–131. 9. Feng X, Mouttaki H, Lin L, Huang R, Wu B, et al. (2009) Characterization of the central metabolic pathways in Thermoanaerobacter sp. strain X514 via isotopomer-assisted metabolite analysis. Appl Environ Microbiol 75: 5001–5008. 30. Drepper T, Eggert T, Circolone F, Heck A, Krauss U, et al. (2007) Reporter proteins for in vivo fluorescence without oxygen. Nat Biotechnol 25: 443–445. 10. Larsen L, Nielsen P, Ahring BK (1997) Thermoanaerobacter mathranii sp. nov., an ethanol-producing, extremely thermophilic anaerobic bacterium from a hot spring in Iceland. Arch Microbiol 168: 114–119. 31. Wilson JH, Leung WY, Bosco G, Dieu D, Haber JE (1994) The frequency of gene targeting in yeast depends on the number of target copies. Proc Natl Acad Sci U S A 91: 177–181. 11. Roh Y, Liu SV, Li G, Huang H, Phelps TJ, et al. (2002) Isolation and characterization of metal-reducing thermoanaerobacter strains from deep subsurface environments of the Piceance Basin, Colorado. Appl Environ Microbiol 68: 6013–6020. 32. Heap JT, Pennington OJ, Cartman ST, Carter GP, Minton NP (2007) The ClosTron: a universal gene knock-out system for the genus Clostridium. J Microbiol Methods 70: 452–464. 12. Lynd LR, Laser MS, Bransby D, Dale BE, Davison B, et al. (2008) How biotech can transform biofuels. Nat Biotechnol 26: 169–172. 33. Bradshaw M, Marshall KM, Heap JT, Tepp WH, Minton NP, et al. (2010) Construction of a nontoxigenic Clostridium botulinum strain for food challenge studies. Appl Environ Microbiol 76: 387–393. 13. Lynd LR, van Zyl WH, McBride JE, Laser M (2005) Consolidated bioprocessing of cellulosic biomass: an update. Curr Opin Biotechnol 16: 577–583. 34. Dong H, Zhang Y, Dai Z, Li Y (2010) Engineering clostridium strain to accept unmethylated DNA. PLoS ONE 5: e9038. 14. Jones CR, Ray M, Strobel HJ (2002) Transcriptional analysis of the xylose ABC transport operons in the thermophilic anaerobe Thermoanaerobacter ethanolicus. Curr Microbiol 45: 54–62. 35. Heap JT, Kuehne SA, Ehsaan M, Cartman ST, Cooksley CM, et al. References 1. Mai V, Lorenz WW, Wiegel J (1997) Transformation of Thermoanaerobacterium sp. strain JW/SL-YS485 with plasmid pIKM1 conferring kanamycin resistance. FEMS Microbiol Lett 148: 163–167. 20. Geier GE, Modrich P (1979) Recognition sequence of the dam methylase of Escherichia coli K12 and mode of cleavage of Dpn I endonuclease. J Biol Chem 254: 1408–1413. 2. Tyurin MV, Desai SG, Lynd LR (2004) Electrotransformation of Clostridium thermocellum. Appl Environ Microbiol 70: 883–890. 2. Tyurin MV, Desai SG, Lynd LR (2004) Electro 21. Roberts RJ, Vincze T, Posfai J, Macelis D (2005) REBASE—restriction enzymes and DNA methyltransferases. Nucleic Acids Res 33: D230–232. thermocellum. Appl Environ Microbiol 70: 883–890. 3. Tyurin MV, Sullivan CR, Lynd LR (2005) Role of spontaneous current oscillations during high-efficiency electrotransformation of thermophilic anaer- obes. Appl Environ Microbiol 71: 8069–8076. 22. Alegre MT, Rodriguez MC, Mesas JM (2004) Transformation of Lactobacillus plantarum by electroporation with in vitro modified plasmid DNA. FEMS Microbiol Lett 241: 73–77. pp 4. Averhoff B (2004) DNA transport and natural transformation in mesophilic and thermophilic bacteria. J Bioenerg Biomembr 36: 25–33. 4. Averhoff B (2004) DNA transport and natural transformat 23. Heim R, Prasher DC, Tsien RY (1994) Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc Natl Acad Sci USA 91: 12501–12504. ( ) p thermophilic bacteria. J Bioenerg Biomembr 36: 25–33. 5. Peng H, Fu B, Mao Z, Shao W (2006) Electrotransformation of Thermoanaer- obacter ethanolicus JW200. Biotechnol Lett 28: 1913–1917. 24. Klapatch TR, Guerinot ML, Lynd LR (1996) Electrotransformation of Clostridium thermosaccharolyticum. J Ind Microbiol 16: 342–347. 6. Rittich B, Spanova´ A (1996) Electrotransformation of bacteria by plasmid DNAs: statistical evaluation of a model quantitatively describing the relationship between the number of electrotransformants and DNA concentration. Bioelectrochem Bioenerg 40: 233–238. 25. Weaver JC (1993) Electroporation: a general phenomenon for manipulating cells and tissues. J Cell Biochem 51: 426–435. 26. Kwak J, Jiang H, Kendrick KE (2002) Transformation using in vivo and in vitro methylation in Streptomyces griseus. FEMS Microbiol Lett 209: 243–248. g 7. Shaw AJ, Podkaminer KK, Desai SG, Bardsley JS, Rogers SR, et al. (2008) Metabolic engineering of a thermophilic bacterium to produce ethanol at high yield. Proc Natl Acad Sci USA. 27. Jennert KC, Tardif C, Young DI, Young M (2000) Gene transfer to Clostridium cellulolyticum ATCC 35319. Microbiology 146 Pt 12: 3071–3080. cellulolyticum ATCC 35319. Microbiology 146 Pt 12: 3071–308 y 8. Author Contributions Conceived and designed the experiments: HS. Performed the experiments: LL YJ YP. Analyzed the data: LL HS. Contributed reagents/materials/ analysis tools: ZH JZ JX. Wrote the paper: LL HS JX. PLoS ONE | www.plosone.org September 2010 | Volume 5 | Issue 9 | e12582 9 Sonoporation of Thermophiles Sonoporation of Thermophiles References (2010) The ClosTron: Mutagenesis in Clostridium refined and streamlined. J Microbiol Methods 80: 49–55. 15. Manome Y, Nakayama N, Nakayama K, Furuhata H (2005) Insonation facilitates plasmid DNA transfection into the central nervous system and microbubbles enhance the effect. Ultrasound in Medicine & Biology 31: 693–702. 36. Fan Z, Kumon RE, Park J, Deng CX (2010) Intracellular delivery and calcium transients generated in sonoporation facilitated by microbubbles. J Control Release 142: 31–39. 16. Newman CM, Bettinger T (2007) Gene therapy progress and prospects: ultrasound for gene transfer. Gene Ther 14: 465–475. 37. Zhou Y, Kumon RE, Cui J, Deng CX (2009) The size of sonoporation pores on the cell membrane. Ultrasound Med Biol 35: 1756–1760. 17. Song Y, Hahn T, Thompson IP, Mason TJ, Preston GM, et al. (2007) Ultrasound-mediated DNA transfer for bacteria. Nucleic Acids Res 35: e129. 18 H YW Ik i A Ch P Zh L D CX (2007) S i i 38. Zhou Y, Shi J, Cui J, Deng CX (2008) Effects of extracellular calcium on cell membrane resealing in sonoporation. J Control Release 126: 34–43. 38. Zhou Y, Shi J, Cui J, Deng CX (2008) Effects of extracel nd-mediated DNA transfer for bacteria. Nucleic Acids Res 35: e129. Ultrasound-mediated DNA transfer for bacteria. Nucleic Acids Res membrane resealing in sonoporation. J Control Release 126: 34–4 18. Han YW, Ikegami A, Chung P, Zhang L, Deng CX (2007) Sonoporation is an efficient tool for intracellular fluorescent dextran delivery and one-step double- crossover mutant construction in Fusobacterium nucleatum. Appl Environ Microbiol 73: 3677–3683. 39. Ghose TK (1987) Measurement of Cellulase activities. Pure & Appl Chem 59: 257–268. 40. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254. 19. Elsner HI, Lindblad EB (1989) Ultrasonic degradation of DNA. DNA 8: 697–701. PLoS ONE | www.plosone.org September 2010 | Volume 5 | Issue 9 | e12582 10
https://openalex.org/W4386461280
https://www.mdpi.com/2311-7524/9/9/1002/pdf?version=1693887648
English
null
Effect of Zinc Excess on Some Physiological Parameters and on the Fatty Acids Profile of Sinapis alba L. and Brassica juncea L. (Czern)
Horticulturae
2,023
cc-by
10,320
Citation: Repkina, N.; Murzina, S.A.; Voronin, V.P.; Kaznina, N. Effect of Zinc Excess on Some Physiological Parameters and on the Fatty Acids Profile of Sinapis alba L. and Brassica juncea L. (Czern). Horticulturae 2023, 9, 1002. https://doi.org/10.3390/ horticulturae9091002 Citation: Repkina, N.; Murzina, S.A.; Keywords: yellow mustard; oriental mustard; lipids; heavy metals; photosynthesis; plant biomass horticulturae Article Effect of Zinc Excess on Some Physiological Parameters and on the Fatty Acids Profile of Sinapis alba L. and Brassica juncea L. (Czern) Natalia Repkina * , Svetlana A. Murzina , Viktor P. Voronin and Natalia Kaznina Institute of Biology of the Karelian Research Centre of the Russian Academy of Sciences (IB KarRC RAS), 11 Pushkinskaya St., Petrozavodsk185910, Russia; murzina.svetlana@gmail.com (S.A.M.); voronen-viktor@mail.ru (V.P.V.); kaznina@krc.karelia.ru (N.K.) * Correspondence: nrt9@ya.ru Institute of Biology of the Karelian Research Centre of the Russian Academy of Sciences (IB KarRC RAS), 11 Pushkinskaya St., Petrozavodsk185910, Russia; murzina.svetlana@gmail.com (S.A.M.); voronen-viktor@mail.ru (V.P.V.); kaznina@krc.karelia.ru (N.K.) * Correspondence: nrt9@ya.ru Abstract: Zinc (Zn) is a vital micronutrient for all living organisms, but in high concentrations, it is a major anthropogenic pollutant. In this study, it was investigated the effect of zinc excess concentrations (50, 100, and 150 mg kg−1) in the substrate on some physiological parameters and the dynamics of fatty acid (FA) content in yellow mustard (Sinapis alba L. cv. Belgia) and oriental mustard (Brassica juncea L. (Czern) cv. Slavanka). The Zn concentration of 50 mg kg−1 did not affect the physiological parameters of yellow mustard, whereas some physiological parameters slightly decreased in oriental mustard. Moreover, this concentration stimulated an increase in total and unsaturated FAs content in leaves of both species, along with high Zn accumulation by both species. The Zn concentrations of 100 and 150 mg kg−1 in substrate had a negative effect on the state of plants as measured by the studied physiological parameters and caused different changes in FAs content in yellow and oriental mustard. Particularly, the total amount of FA increased in comparison to the control, but this increase was due to an elevation in saturated FA (SFA) content. At the same time, the total FA content in oriental mustard slightly decreased while the proportion of unsaturated FAs to SFA kept rising. According to the present data and analysis of the available literature, two strategies of adaptation to heavy metals were suggested through the changes in FAs: height-resistant horticultural plant species on the one side and sensitive species on the other. 1. Introduction The gradual buildup of pollution with heavy metals, especially in agricultural ar- eas, is one of the greatest worldwide concerns due to the hazards of pollutants for all living organisms [1,2]. Heavy metals include elements with an atomic mass greater than 40 Da and a density of 5.31–22.0 g cm−3 [3]. Some of these elements are essential for plants’ metabolism, whereas others are non-essential and can be toxic even at low con- centrations [4]. Among metals, zinc (Zn) is an essential micronutrient required at low concentrations (up to 150 ppm) for the growth and development of the majority of plant species [5]. Zn is a valuable element for the metabolism of all living organisms [6]. For instance, Zn is a structural component of a number of proteins. Among the structural domains of proteins, Zn finger domains have a major physiological relevance [7,8]. Zn- binding enzymes (RNA polymerase, alcohol dehydrogenase, carbonic anhydrase, and Cu/Zn-superoxide dismutase) are involved in translation, transpiration, photosynthesis, and the metabolism of reactive oxygen species [9]. In addition to this, the role of Zn in cell proliferation, membrane stabilization, prevention of oxidative stress, regulation of transcription, and autophagy has also been reported [10,11]. Although Zn is an essential microelement, elevated concentrations (more than 100 mg kg−1) of this metal in soils can lead to toxicity symptoms in plants [5]. Received: 7 August 2023 Revised: 30 August 2023 Accepted: 2 September 2023 Published: 5 September 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/horticulturae Horticulturae 2023, 9, 1002. https://doi.org/10.3390/horticulturae9091002 Horticulturae 2023, 9, 1002 2 of 13 The source of heavy metal contamination can be natural or anthropogenic as a result of agricultural activities [3,4]. Particularly, oversupplementation of Zn-containing fertilizers into soils causes a higher Zn ion concentration and affects crop yield is affected [2,12]. However, plants have a wide range of mechanisms of tolerance for preventing the damage effect of metal excess [3]. The defense reactions include: an increase in lignification of the cell wall for prevention of the transport of metal ions into the cell; the synthesis of chelating agents that are conjugated with ions, further inactivating them; a rise in the activity of antioxidant enzymes, preventing the development of oxidative stress, etc. [3,13,14]. It should be noted that data about the effects of heavy metals on fatty acid (FA) profiles is practically nonexistent. Furthermore, quality and quantity changes of FAs are important for maintaining the structure of membranes at the cellular level under optimum and stress growth conditions [15,16]. It is well known that FAs are constituents of lipids, which play various roles in cells as essential and basement components of membrane structure and participate in various signaling pathways [17]. Moreover, the content of FAs is important in estimating food’s nutritional value [18]. According to this, the study of changes in FAs profile under heavy metal influence is important for the assessment of agricultural crops. Sinapis alba (yellow mustard) and Brassica juncea (oriental mustard) plants belong to the Brassicaceae family. Some plant species from this family are capable of accumulating metal ions in high amounts and are used in phytoremediation (the technology of cleaning up the substrate by plants) [19–22]. Usually, B. juncea is classified as a “hyperaccumulator” (a group of plant species that accumulate high concentrations of metal ions in their aboveground parts) [23,24]. partly due to the fact that oriental mustard is able to accumulate high concentrations of metals and has a high activity of transport proteins involved in the translocation of ions through the plasmalemma to the root and further to the aboveground organs [25]. It was also demonstrated that B. juncea has higher levels of salicylic and glutamic acids, which facilitate the effective binding of metal ions and further sequester them into vacuoles [26]. At the same time, there are opposite data that demonstrate no significant metal ion accumulation by oriental mustard in comparison to other species of the Brassicaceae family [27,28]. With respect to yellow mustard, the data is limited, but some fragmentary research showed its capability to accumulate significant metal ions [29]. Both plant species are valuable for the food industry [30]. So the accumulation of heavy metal ions will have important consequences. Yellow and oriental mustard is a source for oil and spice [31], fodder [32], as well as biofuel production [33,34]. One of the key criteria for assessing the quality of plant raw materials is the determination of the qualitative and quantitative characteristics of the FA profile. Nowadays, there are a few reports demonstrating the changes in FAs profiles in plants under heavy metal effects, but the role of FAs in mechanisms of adaptation to heavy metals is still poorly studied. The present study was undertaken to evaluate the response of yellow mustard and oriental mustard, particularly changes in FAs profiles under growing at high Zn concen- trations in substrate. It has been suggested that, like any stressor, an excess of zinc can lead to quantitative changes in the composition of FAs, which can improve or worsen the nutritional value of plant raw materials. At the same time, a change in the composition of FAs, as one of the mechanisms of adaptation, can be implemented differently in two ways under the influence of excess zinc. 2.2.2. Growth-Related Parameters and Photosynthetic Rate Measurement 2.2.2. Growth-Related Parameters and Photosynthetic Rate Measurement Fresh weight (FW) (g) was measured immediately after harvest and the separation of plants into shoots and roots. The leaf area was analyzed by AreaS 2.01 (Gust, Moscow, Russia). The net photosynthetic rate (Pn) (µmol CO2 m−2s−1) was measured on the first true leaves using the portable photosynthesis system HCM-1000 (Walz, Effeltrich, Germany). 2. Materials and Methods 2.1. Plant Material, Growth Conditions, and Organization of the Experiment Seeds of yellow mustard (Sinapis alba L. cv. Belgia) and oriental mustard (Brassica juncea L. (Czern) cv. Slavyanka) were acquired from the Federal Research Center N. I. Vavilov All-Russian Institute of Plant Genetic Resources (VIR), Ministry of Science and Higher Education, Russia. This study was conducted at the research base of the Agricultural Experiment Station (61◦45′6′′ N, 119◦25′10.9′′ E) (WGS84), Petrozavodsk, Republic of Karelia, Russia. The day before sowing, mustard seeds were soaked in distilled water. Two-day-old seedlings (12 seedlings per pot) were placed in plastic pots (1 L). Each pot Horticulturae 2023, 9, 1002 3 of 13 contains 0.8 kg of pure quartz sand. Appropriate amounts of the sulfate salt of Zn (ZnSO4 × H2O) were once added to the sand to maintain the required level of Zn: 5 (control), 50, 100, and 150 mg kg−1 of the substrate. These concentrations are in comparison to the optimum, which is two-foldhigher than maximum allowable concentrations (MAC), four-foldhigher than MAC, and six-foldhigher than MAC in sand. The standards of MAC were published by the Ministry of Agriculture (GN 2.1.7.2041-06 from 2006). Before the seeds were sown, the sand in the pots was mixed in order to ensure a uniform distribution of the elements. During the growing period, pots have been irrigated with Hoagland-Arnon nutrient solution (pH 6.2), excluding Zn addition. All pots were placed in a greenhouse under natural day/night conditions with a day/night temperature of 21/14 ± 2 ◦C and 77 ± 5% relative humidity during the growing period. The experimental design was completely randomized with three replications. Plants were harvested and analyzed after 3 weeks (stage of 3–4 true leaves). 2.2.4. Photosynthetic Pigments Content Measurement Chlorophyll a (Chla), chlorophyll b (Chlb), total chlorophylls, and carotenoids were determined spectrophotometrically on an SF2000 spectrophotometer (Spectrum, Moscow, Russia) [36]. For the extraction of assimilatory pigments, 85% (v/v) aqueous acetone was used. The absorbance of the supernatant was measured at 665, 649, and 440.5 nm alongside a blank of untainted 85% liquid acetone. Photosynthetic pigment concentrations were analyzed using known formulas [36]. The percentage of Chl in light harvesting complex II (LHCII) was calculated by accepting that almost all Chlb is in LHCII and that the ratio of Chla and Chlb in LHCII is 1.2 [37]. 2.2.5. Lipids Extraction and Fatty Acids Analysis 2.2.3. Analysis of Relative Electrolyte Leakage The relative electrolyte leakage (REL) was measured by the electrical conductivity method as described by [35] with slight modifications. Fresh leaves were cut into pieces of 0.5 cm2 square, rinsed with deionized water, and put into 30 mL of deionized water in a test tube. After vacuuming for 10 min, tubes were stored at room temperature for 4 h. The initial electrolyte leakage (EL0) was measured using a conductometer (HANNA, Woonsocket, Italy). The tubes were subsequently boiled at 100 ◦C for 30 min to release all the electrolytes into the solution, cooled to 25 ◦C, and the final electrolytic leakage (EL1) was measured. REL was calculated according to the following formula: (EL0/EL1) × 100 and expressed as a percentage of total conductivity. 2.2.1. Determination of Zinc Content An inductively coupled plasma-mass spectrometry (ICP-MS) (Agilent 7900, Santa Clara, CA, USA) analysis was performed to determine the concentration of Zn in the roots and shoots of plants. The samples were digested by microwave using the Speedwave Digestion System (Berghof, Eningen, Germany). The extraction was performed in a solution of nitric acid. The data was expressed as mg kg−1DW, where DW is the dry weight that was evaluated after drying the aboveground parts of plants to a constant weight at 80 ◦C. 2.2.5. Lipids Extraction and Fatty Acids Analysis The total lipids (TL) from 0.2 g of fresh leaves (a mixed sample from 10 leaves of seedlings) were extracted using the Folch method with a mixture of chloroform and methanol (2:1 v/v) [38]. In brief, the samples were filtered using a paper filter, Red Ribbon Horticulturae 2023, 9, 1002 4 of 13 (pore size 5–8 µm) with 20 mL of the mixture, after which the precipitate on the filter was washed with 10 mL of pure chloroform. To remove water-soluble impurities, 10 mL of purified deionized water (Simplicity, Millipore, Merck, Darmstadt, Germany) was added to the filtered mixture and left for 2 h in separating funnels (Schott Duran, Mainz, Ger- many) until the organic phases were completely separated. Lipids remained in the lower chloroform layer, whereas non-lipid substances moved to the upper aqueous methanol phase. Then, the chloroform layer was withdrawn to evaporate under a vacuum on a rotary evaporator—Hei-VAP Advantage HL/G3 (Heidolph, Schwalbach, Germany)—and dried in a vacuum over phosphoric anhydride to a constant weight. The dried samples were redissolved with chloroform-methanol (1:1, v/v) and stored at −20 ◦C until further processing. p g The qualitative and quantitative fatty acid (FA) profile of the TL was analyzed through gas chromatography coupled with mass spectrometry (GC–MS); we had previously sub- jected the TL mixture to acid methylation. To obtain the fatty acid methyl esters (FAMEs) from the TLs, 0.2 mL of a solution of TLs, 2 mL of methanol, and 0.2 mL of chlorate acetyl (CH3COCl) as catalyst were added to a glass retort (Schott Duran, Mainz, Germany). The obtained solution was heated for 90 min at a fixed temperature of 70–80 ◦C. After methyla- tion (followed by cooling), the hexane was poured into 5 mL of each sample. For phase separation, 2 mL of deionized water was poured into each glass retort and transferred into the separatory glass funnels for 15 min. FAMEs remained in the upper hexane layer, while residue substances were concentrated in the lower aqueous phase. The hexane layer was evaporated using a rotary evaporator, Hei-VAP Advantage HL/G3 (Heidolph, Schwabach, Germany), to collect pure FAMEs. Then, 1 mL of hexane for GC (Sigma Aldrich, St. Louis, MO, USA) was added to the collected FAMEs in glass retorts, and the gathered and final solutions were accumulated in GC vials for the following GC analysis. 2.2.5. Lipids Extraction and Fatty Acids Analysis g y FAMEs were separated using GC with a mono-quadrupole mass-selective detector–α- Maestro (Saitegra, Moscow, Russia). The separation of FAs was carried out in a gradient thermal configuration (tstart = 140 ◦C—hold 5 min; increase in t from 140 ◦C to 240 ◦C at a rate of 4 ◦C per min; and tfinal = 240 ◦C—hold 2 min) with an HP-88 (60 m × 0.25 mm × 0.20 µm) capillary column (Agilent Technologies, Santa Clara, CA, USA) using helium as a mobile phase. FAMEs were detected in SCAN and SIM modes. The SCAN mode was used for searching and identifying FAs (qualitative analysis) with scan parameters of 50 to 400 mz−1. The SIM mode was used for FAs according to analytical standards with Supelco 37 (Sigma Aldrich, St. Louis, MO, USA), with subsequent quantitative analysis of the components that make up the mixture of Supelco 37. Quantitative individual FAs in the samples were detected using the calibration method using the Supelco 37 mixture (Sigma Aldrich, St. Louis, MO, USA). The data were analyzed using Maestro Analytic v. 1.026 software (Saitegra, Moscow, Russia) with the NIST library. 2.3. Statistical Analysis All experiments were repeated three times. The data were subjected to an analysis of variance (ANOVA). The data were processed using Excel 2007 (Microsoft Corp., Redmond, WA, USA) and analyzed with Statgraphics Plus 5.0 (Statgraphics Technologies, Inc., The Plains, VA, USA) statistical software. The normality of the distribution was tested using the Shapiro-Wilk test. Data are presented as mean values ± standard error (SE). The Fisher’s least significant difference (LSD) test was used to compare the treatment means. Differences at p< 0.05 were considered statistically significant. The statistical analysis of FA content was made using the GC-MSD method (n = 3) with high precision and convergence. 3.1. Amount of Zn in the Roots and Shoots of S. alba and B. juncea Seedlings Along with the increase in concentration of Zn in the substrate, its content increased in the roots and shoots of both S. alba and B. juncea seedlings (Table 1). The yellow mustard seedlings subjected to 50 mg kg−1of Zn mostly accumulated the metal ions in shoots, Horticulturae 2023, 9, 1002 5 of 13 whereas under higher concentrations of Zn (100 and 150 mg kg−1), their content between shoots and roots was similar. The oriental mustard seedlings at Zn concentrations of 50 and 100 mg kg−1 accumulated Zn ions in roots, while at Zn concentrations of 150 mg kg−1, there was no statistical difference in this metal content between shoots and roots (Table 1). Table 1. The Zn content in S. alba and B. juncea seedlings under exposure to Zn excess. Plant Species Zn Concentration, mg kg−1 Substrate Zn Content, mg kg−1 DW Shoot Root S. alba 5 187.32 ± 20.60 f 115.01 ± 14.80 f 50 2505.31 ± 206.02 c 1504.60 ± 125.96 e 100 2620.35 ± 215.22 b 2708.28 ± 222.26 b 150 3118.52 ± 311.08 a 3340.07 ± 272.80 a B. juncea 5 77.48 ± 10.78 g 170.59 ± 19.25 f 50 1450.23 ± 121.62 e 2169.80 ± 179.18 c 100 1849.71 ± 153.58 d 2174.27 ± 179.54 c 150 2239.21 ± 184.74 c 2379.11 ± 195.93 c Values are means ± SE (n = 3). Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least significant difference (LSD) test. 3.2. Effect of Zn Excess on Leaf Area, Fresh Weight, and Relative Electrolyte Leakage in S. alba and Table 1. The Zn content in S. alba and B. juncea seedlings under exposure to Zn excess. 3.2. Effect of Zn Excess on Leaf Area, Fresh Weight, and Relative Electrolyte Leakage in S. alba and B. juncea Seedlings Height Zn concentrations led to a decrease in leaf areas of S. alba and B. junceaseedlings (Table 2). It was noted that a Zn concentration of 50 mg kg−1stimulated the leaf area of yellow mustard (Table 2). The Zn concentration of 150 mg kg−1caused almost twice as much leaf area loss in yellow mustard and even more leaf area loss in oriental mustard. The fresh weight (FW) of both plant species decreased only at 100 and 150 mg kg−1 Zn concentrations (Table 2). 3.1. Amount of Zn in the Roots and Shoots of S. alba and B. juncea Seedlings a a c d a a b c 0 1 2 3 4 5 5 50 100 150 Pn (µmol CO2 m–2 s–1) Zn concentration, mg kg–1 S. alba B. juncea Figure 1. Effect of Zn excess on photosynthesis rate (Pn) of S. alba and B. juncea. Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least significant difference (LSD) test. a a c d a a b c 0 1 2 3 4 5 5 50 100 150 Pn (µmol CO2 m–2 s–1) Zn concentration, mg kg–1 S. alba B. juncea Zn concentration, mg kg–1 Figure 1. Effect of Zn excess on photosynthesis rate (Pn) of S. alba and B. juncea. Different letters ndicate significant differences between treatments (p < 0.05), determined by Fisher’s least signifi- ant difference (LSD) test. Figure 1. Effect of Zn excess on photosynthesis rate (Pn) of S. alba and B. juncea. Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least significant difference (LSD) test. The Zn excess also caused a decrease in photosynthetic pigment content in S. alba and B. juncea (Table 3). Particularly in yellow mustard, the decrease in total chlorophylls and carotenoids content was observed at 100 and 150 mg kg–1 Zn concentration impacts, whereas in oriental mustard seedlings, the drop in photosynthetic pigment content was caused by 50 mg kg–1 Zn concentration. Similar changes were detected for chlorophyll content in LHCII in both plant species. The Zn excess also caused a decrease in photosynthetic pigment content in S. alba and B. juncea (Table 3). Particularly in yellow mustard, the decrease in total chlorophylls and carotenoids content was observed at 100 and 150 mg kg−1 Zn concentration impacts, whereas in oriental mustard seedlings, the drop in photosynthetic pigment content was caused by 50 mg kg−1 Zn concentration. Similar changes were detected for chlorophyll content in LHCII in both plant species. Table 3. Effect of Zn excess on photosynthetic pigments content in S. alba and B. juncea. Plant Species Zn Concentration, mg kg–1 Substrate Total Chl (a + b), mg g–1 FW Carotenoids, mg g–1 FW Chl b LHCII, % S. 3.1. Amount of Zn in the Roots and Shoots of S. alba and B. juncea Seedlings alba 5 1.36 ± 0.04 a 0.28 ± 0.01 a 0.18 ± 0.02 a 50 1.35 ± 0.04 a 0.25 ± 0.01 a 0.20 ± 0.01 a 100 0.88 ± 0.03 bc 0.18 ± 0.01 b 0.09 ± 0.01 b 150 0.53 ± 0.03 e 0.13 ± 0.01 c 0.03 ± 0.01 d B. juncea 5 1.37 ± 0.02 a 0.25 ± 0.01 a 0.20 ± 0.01 a 50 0.94 ± 0.03 b 0.18 ± 0.01 b 0.09 ± 0.01 b 100 0.71 ± 0.03 d 0.15 ± 0.01 c 0.05 ± 0.01 dc 150 0.84 ± 0.01 c 0.18 ± 0.01 b 0.07 ± 0.01 c Values are means ± SE (n 12) Different letters indicate significant differences between treatments Table 3. Effect of Zn excess on photosynthetic pigments content in S. alba and B. juncea. Plant Species Zn Concentration, mg kg−1 Substrate Total Chl (a + b), mg g−1 FW Carotenoids, mg g−1 FW Chl b LHCII, % S. alba 5 1.36 ± 0.04 a 0.28 ± 0.01 a 0.18 ± 0.02 a 50 1.35 ± 0.04 a 0.25 ± 0.01 a 0.20 ± 0.01 a 100 0.88 ± 0.03 bc 0.18 ± 0.01 b 0.09 ± 0.01 b 150 0.53 ± 0.03 e 0.13 ± 0.01 c 0.03 ± 0.01 d B. juncea 5 1.37 ± 0.02 a 0.25 ± 0.01 a 0.20 ± 0.01 a 50 0.94 ± 0.03 b 0.18 ± 0.01 b 0.09 ± 0.01 b 100 0.71 ± 0.03 d 0.15 ± 0.01 c 0.05 ± 0.01 dc 150 0.84 ± 0.01 c 0.18 ± 0.01 b 0.07 ± 0.01 c Values are means ± SE (n = 12). Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least significant difference (LSD) test. Table 3. Effect of Zn excess on photosynthetic pigments content in S. alba and B. juncea. Table 3. Effect of Zn excess on photosynthetic pigments content in S. alba and B. juncea. Values are means ± SE (n 12). Different letters indicate significant differences between treat p< 0.05), determined by Fisher’s least significant difference (LSD) test. 3.4. Fatty Acids Content in S. alba and B. juncea Seedlings under Exposure to Zn Excess ( ) g (p< 0.05), determined by Fisher’s least significant difference (LSD) test. 3.4. Fatty Acids Content in S. alba and B. juncea Seedlings under Exposure to Zn Excess 3.4. 3.1. Amount of Zn in the Roots and Shoots of S. alba and B. juncea Seedlings The most negative effect of Zn excess on FW was observed under a 150 mg kg−1Zn concentration in both S. alba and B. juncea seedlings. However, the relative electrolyte leakage was not significantly different at exposure to all Zn concentrations in S. alba and B. juncea seedlings (Table 2). Table 2. Effect of Zn excess on leaf area, fresh weight (FW), and relative electrolyte leakage (REL) in S. alba and B. juncea seedlings. Plant Species Zn Concentration, mg kg−1 Substrate Leaf Area, cm2 FW, g REL, % S. alba 5 4.00 ± 0.61 b 1.27 ± 0.14 a 10.78 ± 0.82 a 50 5.70 ± 0.17 a 1.36 ± 0.15 a 11.33 ± 0.77 a 100 2.70 ± 0.55 d 0.46 ± 0.04 b 13.51 ± 1.23 a 150 1.99 ± 0.48 e 0.27 ± 0.03 c 12.29 ± 0.06 a B. juncea 5 4.09 ± 0.34 b 1.28 ± 0.07 a 12.52 ± 0.64 a 50 3.13 ± 0.22 c 1.09 ± 0.12 a 13.00 ± 0.81 a 100 1.77 ± 0.22 e 0.67 ± 0.10 b 14.61 ± 1.59 a 150 1.35 ± 0.20 f 0.39 ± 0.05 c 14.00 ± 1.67 a Values are means ± SE (n = 30). Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least significant difference (LSD) test. Table 2. Effect of Zn excess on leaf area, fresh weight (FW), and relative electrolyte leakage (REL) in S. alba and B. juncea seedlings. Values are means ± SE (n = 30). Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least significant difference (LSD) test. 3.3. Effect of Zn Excess on Some Photosynthesis Parameters of S. alba and B. juncea Seedlings At a Zn concentration of 50 mg kg−1 the rate of photosynthesis (Pn) of yellow mustard and oriental mustard was not significantly different in comparison to the control (Figure 1). At the higher Zn concentrations (100 and 150 mg kg−1), Pn decreased, which was observed to a greater extent in yellow mustard. 6 of 13 as ob- Horticulturae 2023, 9, 1002 Figure 1. Effect of Zn excess on photosynthesis rate (Pn) of S. alba and B. juncea. Different letters indicate significant differences between treatments (p < 0.05), determined by Fisher’s least signifi- cant difference (LSD) test. 3.1. Amount of Zn in the Roots and Shoots of S. alba and B. juncea Seedlings Fatty Acids Content in S. alba and B. juncea Seedlings under Exposure to Zn Excess. Along with the rise in Zn ion concentration, the content of palmitic acid (PA) and stearic acid (SA) in yellow mustard increased, as did the total content of saturated fatty acids (SFA) (Table 4). For instance, the total content of SFA increased by 8 and 59% at 50 and 100 mg kg−1 Zn concentrations, respectively, compared with control. However, the influence of 150 mg kg−1 Zn concentration on seedlings of yellow mustard caused a rise in SFA content about 2.5-fold (Table 4). The changes in unsaturated fatty acid (USFA) content were not as marked in yellow mustard compared with control (Table 4). The 100 mg kg−1 Zn concentration caused a slight decrease in all detected USFA, but under the influence of other Zn concentrations (50 and 150 mg kg−1), the total content of USFAs was similar to the control. Approximate 70% of all USFA is presented by α-linolenic (ALA) FA in yellow mustard under exposure to all concentrations of Zn. Also, in control plants of yellow mustard and seedlings subjected to 50 mg kg−1of Zn, the total content of USFAs was higher than SFAs. However, in seedlings Horticulturae 2023, 9, 1002 7 of 13 of yellow mustard treated with 100 and 150 mg kg−1 of Zn, the content of SFAs was higher than USFAs. Table 4. The fatty acids content in S. alba under Zn excess. Table 4. The fatty acids content in S. alba under Zn excess. Table 4. The fatty acids content in S. alba under Zn excess. Fatty Acids, µL mg−1 Zn Concentration in Substrate, mg kg−1 p Value 5 50 100 150 Saturated Myristic C 14:0 8.42 8.78 8.31 9.23 0.05 Palmitic C 16:0 84.37 90.11 120.64 145.01 ≤0.05 Stearic C 18:0 29.66 33.97 67.21 153.99 ≤0.05 Total content 122.45 132.86 196.16 308.23 ≤0.05 Unsaturated Oleic C 18:1n-9 20.36 18.40 13.44 16.71 ≤0.05 Linoleic C 18:2n-6 25.41 28.25 26.90 37.83 ≤0.05 Linolenic C 18:3n-3 147.66 141.24 112.74 145.86 0.05 Total content 193.43 187.89 153.08 200.40 ≤0.05 Unsturated/Saturated FAs 1.58 1.41 0.78 0.65 ≤0.05 Values are presented as means, the SE were low and approximately 0.003–0.009 since the analysis was made using GC-MSD method (n = 3) with high precision and convergence. 3.1. Amount of Zn in the Roots and Shoots of S. alba and B. juncea Seedlings The content of SFAs in oriental mustard slightly dropped at 50 mg kg−1 of Zn impact, whereas under higher Zn concentration influences (100 and 150 mg kg−1), the total content of SFAs was similar to control (Table 5). During exposure to Zn excess, the content of myristic acid (MA) increased while the PA content slightly dropped in oriental mustard. The content of SA stayed the same (Table 5). The influence of 50 mg kg−1 of Zn concen- tration caused a significant increase in the content of total USFAs, whereas at higher Zn concentrations (100 and 150 mg kg−1), the amount of total content dropped compared with control (Table 5). The increase in total USFA content at a Zn concentration of 50 mg kg−1 mostly depends on the rise of ALA FA. It is necessary to note that despite the decrease in USFA’s total content at high concentrations of Zn impact, its amount was higher than the total SFA’s content. This proportion was also observed in control plants and seedlings of oriental mustard treated by Zn excess (50 mg kg−1) (Table 5). Table 5. The fatty acid content in B. juncea under Zn excess. Fatty Acids, µL mg−1 Zn Concentration in Substrate, mg kg−1 p Value 5 50 100 150 Saturated Myristic C 14:0 5.81 8.19 11.48 9.10 ≤0.05 Palmitic C 16:0 100.05 88.31 98.11 94.17 ≤0.05 Stearic C 18:0 36.57 33.50 33.49 34.55 0.05 Total content 142.43 130.00 143.08 137.82 ≤0.05 Unsaturated Oleic C 18:1n-9 21.26 17.09 25.05 18.73 ≤0.05 Linoleic C 18:2n-6 24.29 27.47 28.14 32.31 ≤0.05 Linolenic C 18:3n-3 174.81 271.57 123.76 118.39 ≤0.05 Total content 220.36 316.13 176.95 169.43 ≤0.05 Unsturated/Saturated FAs 1.55 2.43 1.24 1.23 ≤0.05 Values are presented as means, the SE were low and approximately 0.003–0.009 since the analysis was made using GC-MSD method (n = 3) with high precision and convergence. Table 5. The fatty acid content in B. juncea under Zn excess. 4. Discussion It is well known that different plant species absorb and accumulate metal ions in organs and tissues [29,39]. Particularly, plants that are “hyperaccumulators” intake a high amount of metal ions and mostly keep them in the aboveground part, whereas plants that are “excluders” absorb less concentrations of metals and accumulate them mostly in the Horticulturae 2023, 9, 1002 8 of 13 8 of 13 roots [40]. In the present study, it was demonstrated that both yellow and oriental mustards accumulated Zn proportionally to the concentrations in the substrate and that both species contain similar amounts of Zn ions. At the same time, the Zn content in yellow mustard at a metal concentration of 50 mg kg −1 of substrate was higher in shoots, and at the higher Zn concentrations (100 and 150 mg kg−1), it turned out to be almost equal in underground and aboveground organs. In contrast, in oriental mustard, at a zinc concentration of 50 and 100 mg kg−1 of substrate, the metal accumulated mainly in the roots, and with an increase in the metal concentration to 150 mg kg−1 of substrate, it accumulated in the roots and shoots almost equally. It is correlated with data demonstrated previously on yellow mustard [29] and oriental mustard [28]. Despite the rather large amount of data on the high ability of oriental mustard to accumulate metal ions [25,41], their content is significantly lower than in “hyperaccumulator” species, such as Thlaspi caerulescens and Arabidopsis halleri, which accumulate up to 9000 mg kg−1 DW of Zn [42,43]. There are practically no data on yellow mustard; however, a fragmentary work demonstrated its ability to accumulate metal ions no less than oriental mustard [44] that corresponded with present data. p In addition to differences in the accumulation of heavy metals, different plant species also have different metal tolerances, which, as a rule, are assessed by growth and photosyn- thesis activity. As with other heavy metals, Zn has a negative effect on plant growth [45]. Under Zn excess, a decrease in growth was observed in plants: B. juncea and B. napus [28], Zea mays [46], Polypogon monspeliensis [47], Ocimum basilicum [48]. In the present study, the Zn concentration of 50 mg kg−1 caused a slightly stimulated effect on the growth of yellow mustard. 4. Discussion whereas the increase in Zn concentration in substrate to 100 and 150 mg kg−1 resulted in a negative effect on the growth of both plant species. It is suggested that growth inhibition under Zn excess concentration can be a result of limiting cell division due to an increase in the duration of mitotic phases as well as the full mitotic cycle [49], impaired cell elongation due to a decrease in the elasticity of cell walls, or a decrease in their turgor [39]. g y g In addition, the inhibition of plant growth under these conditions can be a result of changes in the photosynthesis activity of plants, leading to the inhibition of photosynthesis. The present study showed that zinc at a concentration of 50 mg kg−1 of substrate had no effect on the photosynthesis rate, while at higher concentrations (100–150 mg kg−1 of substrate), the rate of photosynthesis significantly decreased in both species, but a more pronounced effect was noted in yellow mustard, indicating its lower tolerance to metal influence. One of the reasons for the drop in photosynthesis activity is a decrease in photosynthetic pigment content, which we also observed in other species such as wheat [50], Polypogon monspeliensis [47], Ocimum basilicum [48], etc. On the other hand, the slowdown in the rate of photosynthesis under stress conditions can be a result of a decrease in the activity of carbonic anhydrase. Carbonic anhydrase (EC: 2.4.1.1) catalyzes the rapid conversion of carbon dioxide plus water into a proton and the bicarbonate ion (HCO 3−)and provides the RuBisCO with CO2 molecules [51]. There are no available data about the effect of Zn at high concentrations on the activity of carbonic anhydrase in plants. Whereas, the negative effect of an excess of these metal ions on the other enzymes’ activity that participate in photosynthesis, for instance, RuBisCO, was demonstrated [52,53] as well as a decrease in the activity of photosystems [50,54]. Moreover, Segardoy et al. [55] showed the effect of excess zinc on carboxylate metabolism, which leads to an increase in respiration due to using the carboxylates from roots as a substrate, in parallel with a decrease in photosynthesis rate. y p p y To ensure the process of photosynthesis, it is necessary to maintain the integrity of chloroplast membranes. One of the main roles in membrane stability is played by the USFAs. 4. Discussion It is interesting that a similar mechanism of biochemical adaptation was described for animals (fish) [17,63] At this point, it is possible to discuss the general patterns—evolutionarily determined ones—of stress-induced reactions. It is necessary to note that changes in FAs content in yellow mustard under the Zn excess effect were unlike those in oriental mustard. In particular, as the proportion of zinc in the substrate increased, the total content of FAs in the aerial parts of plants rose. The Zn concentration of 50 mg kg−1increased the total amount of SFAs but was lower than the USFAs content. At a zinc concentration of 150 mg kg−1, the total content of FAs increased by 60% compared to the control. But in this case, this increase was a result of a rise in SFAs, particularly palmitic and stearic acids. The increase in palmitic acid concentration under heavy metal influence was reported in several articles [59,64–66]. Moreover, the possible role of palmitic acid in the stress tolerance of plants was described by Zhukov [67]. The accumulation of SFAs is characteristic of plants that mainly accumulate metals in the roots (excluders), for example, wheat under Cd ions [66], caraway [68], coriander [64], and Arabidopsis lyrata under Zn ions [59]. The data from the present study demonstrated that both species (yellow and oriental mustard) accumulate significant Zn concentrations and maintain the capability for growth and development under this condition. This indicates successful adaptation of both species to high concentrations of Zn in the substrate, despite different changes in the FA composi- tion. Probably, the accumulation of SFAs is an alternative way of adapting plants (which are not hyperaccumulators or more sensitive species to metal ions’ influence). This also supports the fact that caraway plants under Zn excess in parallel with SFA accumulation are capable of forming natural seeds [68]. Earlier, we already made this assumption when studying changes in the FA composition of wheat under the influence of Cd ions [66]. Probably the increase in particularly palmitic and stearic acids caused a “raft” formation that leads to obturating membranes that prevent the overintake of metal ions into cells. At the same time, one of the possible functions of palmitic acid is “anchorage” in membrane transporter proteins to optimize their activity [67]. 4. Discussion In the present study, it was shown that the Zn concentration of 50 mg kg−1 stimulated the accumulation of total FAs, including USFA. Whereas, despite the drop in total FAs content in oriental mustard under Zn concentration (100 and 150 mg kg−1) influence, the content of USFAs was higher than SFA. At the same time, the photosynthesis rate decreased slightly in comparison to yellow mustard. It was assumed that the higher USFA content pointed to a compensatory rearrangement of membrane structure, which was also suggested by Horticulturae 2023, 9, 1002 9 of 13 the absence of electrolyte leakage. The functioning of membrane-bound enzymes and processes in membrane is maintained by an increase in the unsaturation of FAs, one of the mechanisms of adaptation in living systems, resulting in a change in the structure and physicochemical properties of membrane [56]. Due to this, it is possible to suggest that maintenance of unsaturation of lipids by accumulation of USFAs in the thylakoids membrane of oriental mustard provides for optimal activity of photosystem II under Zn excess. Also, it was shown that the chlorophyll content in LCH II remained stable in oriental mustard at Zn concentrations of 100 and 150 mg kg−1. Moreover, the connection between USFA accumulation and photosystem II functioning in oriental mustard under cadmium (Cd) influence was previously observed [57]. More recently, it was demonstrated the increase in USFAs content in “hyperaccumu- lator” plant species, such as Noccaea caerulescens under Cd ions [58] and Arabidopsis halleri under Zn ions [59]. The accumulation of high amounts of USFAs is an important mechanism that determines the tolerance of these species and their ability to grow in conditions of high concentrations of metals in the substrate. Probably, it is associated with the multifunctional activity of USFAs in cells, particularly those involved in maintaining membrane permeabil- ity. It was discussed that USFAs protect cells from reactive oxygen species, and these FAs are precursors for the manifestation of secondary metabolites [60]. More USFAs are known as “signal messengers” because they consist of nitro-FAs, which are the donors of nitrogen dioxide (NO2) [60,61]. As a constituent of triacylglycerides, USFAs are deposited in their structure (the energy cost of such TAGs is high) and actively used in energy-dependent processes under changing environments; thus, fast-growing plants are able to react to stressconditions [62]. 4. Discussion In addition, an increase in the total content of fatty acids is possibly associated with the creation of energy storage, which can subsequently be spent on energy-consuming cell reactions to maintain vital activity. 10 of 13 10 of 13 Horticulturae 2023, 9, 1002 5. Conclusions Both species (S. alba and B. juncea) are capable of accumulating sufficient amounts of Zn ions. Zn concentrations of 100 and 150 mg kg−1 caused the most negative effect on the physiological and biochemical parameters of mustard. The greater negative effect of Zn excess observed in yellow mustard, particularly in photosynthesis rate, in comparison with oriental mustard The changes in the quantity content of FAs in two plant species were different under the Zn effect. The Zn concentration of 50 mg kg−1 caused a stimulated effect on the total content of FAs and the amount of USFA to SFA, while the FAs content changes were rather different under higher Zn concentrations. Particularly in yellow mustard, the total content of FAs increased due to a rise in SFA content, whereas in oriental mustard, the total content of FAs decreased but the ratio of USFAs to SFAs was maintained. According to the present data and some recently available data, it is possible to conclude that there are two ways of adapting to heavy metal influences. On the one hand, there is an increase in USFAs content by “hyperaccumulators”, or highly resistant plants to metal stress, and on the other hand, there is an increase in SFAs content by “excluders”, or sensitive plant species to metal stress. Author Contributions: Conceptualization, N.R. and S.A.M.; Methodology, N.R. and S.A.M.; Investi- gation, N.K. and V.P.V.; Data Curation, N.R. and S.A.M.; Writing—Original Draft Preparation, N.R.; Writing—Review and Editing, N.K. and S.A.M. All authors have read and agreed to the published version of the manuscript. Funding: The study was carried out with the financial support of RSF (project No. 22-24-00668). This study was carried out using the equipment of the Core Facility of the Institute of Biology at KarRC RAS. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: All data and tables in this manuscript are original. Data Availability Statement: All data and tables in this manuscript are original. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. References 1. Zunaidi, A.A.; Lim, L.H.; Metali, F. Transfer of heavy metals from soils to curly mustard (Brassica ju agricultural farm in Brunei Darussalam. Heliyon 2021, 7, e07945. [CrossRef] [PubMed] 2. Rai, P.K.; Sonne, C.; Kim, K.-H. Heavy metals and arsenic stress in food crops: Elucidating antioxidative defense mechanisms in hyperaccumulators for food security, agricultural sustainability, and human health. Sci. Total Environ. 2023, 874, 162327. [CrossRef] [PubMed] in hyperaccumulators for food security, agricultural sustainability, and human health. Sci. Total Environ. 2023, 874, 162327. [CrossRef] [PubMed] 3. Ghuge, S.A.; Nikalje, G.C.; Kadam, U.S.; Suprassana, P.; Hong, J.C. Comprehensive mechanisms of heavy metal toxicity in plants, detoxification, and remediation. J. Hazard. Mater. 2023, 450, 131039. [CrossRef] 3. Ghuge, S.A.; Nikalje, G.C.; Kadam, U.S.; Suprassana, P.; Hong, J.C. Comprehensive mechanisms of heavy metal toxicity in plants, detoxification, and remediation. J. Hazard. Mater. 2023, 450, 131039. [CrossRef] 3. Ghuge, S.A.; Nikalje, G.C.; Kadam, U.S.; Suprassana, P.; Hong, J.C. Comprehensive mechanisms of h detoxification, and remediation. J. Hazard. Mater. 2023, 450, 131039. [CrossRef] 4. Patel, A.; Tiwari, S.; Raju, A.; Pandey, N.; Singh, M.; Prasad, S.M. Heavy metal contamination of environment and crop plants. In Sustainable Solutions for Elemental Deficiency and Excess in Crop Plants; Mishra, K., Tandon, P.K., Srivastava, S., Eds.; Springer: Singapore, 2020. [CrossRef] g p 5. Mehrota, D.K. Effect of deficiency of essential elements and toxicity of metals on human health. In Sustainable Solutions for Elemental Deficiency and Excess in Crop Plants; Mishra, K., Tandon, P.K., Srivastava, S., Eds.; Springer: Singapore, 2020. [CrossRef] 6. Ackland, M.L.; Michalczyk, A.A. Zinc and infant nutrition. Archiv. Biochem. Biophys. 2016, 611, 51–57. [CrossRef] 6. Ackland, M.L.; Michalczyk, A.A. Zinc and infant nutrition. Archiv. Biochem. Biophys. 2016, 611, 51–57. Michalczyk, A.A. Zinc and infant nutrition. Archiv. Biochem. Biophys. 2016, 611, 51–57. [CrossRef] S.A.; Krämer, U. The zinc homeostasis network of land plants. Biochem. Biophys. Acta 2012, 1823, 1553–1567. 7. Sinclair, S.A.; Krämer, U. The zinc homeostasis network of land plants. Biochem. Biophys. Acta 2012, 1823, 1553–1567. [CrossRef] 8. Hafeez, B.; Khanif, Y.M.; Saleem, M. Role of zinc in plant nutrition—A review. Am. J. Exp. Agricul. 2013, 3, 374–391. [CrossRef] 9. Stanton, C.; Sanders, D.; Krämer, U.; Podar, D. Zinc in plants: Integrating homeostasis and biofortification. Mol. Plant 2022, 15, 65–85. [CrossRef] 7. Sinclair, S.A.; Krämer, U. The zinc homeostasis network of land plants. Biochem. Biophys. Acta 2012 8. Hafeez, B.; Khanif, Y.M.; Saleem, M. References Plant 2020, 13, 955–983. [CrossRef] 17. Murzina, S.A.; Pekkoeva, S.N.; Kondakova, E.A.; Nefedova, Z.A.; Filippova, K.A.; Nemova, N.N.; Orlov, A.M.; Berge, J.; FalkPetersen, S. Tiny but Fatty: Lipids and Fatty Acids in the Daubed Shanny (Leptoclinus maculatus), a Small Fish in Svalbard Waters. Biomolecules 2020, 10, 368. [CrossRef] 18. Murzina, S.A.; Voronin, V.P.; Ruokolainen, T.R.; Artemenkov, D.V.; Orlov, A.M. Comparative analysis of lipids and fatty acids in beaked redfish Sebastes mentella Travin, 1951 collected in wild and in commercial products. J. Mar. Sci. Eng. 2022, 10, 59. [CrossRef] 19. Rizwan, M.; Ali, S.; ur Rehman, M.Z.; Rinklebe, J.; Tsang, D.C.W.; Bashir, A.; Maqbool, A.; Tack, F.M.G.; Ok, Y.S. Cadmium phytoremediation potential of Brassica crop species: A review. Sci. Total Environ. 2018, 631–632, 1175–1191. [CrossRef] [PubMed] 20. Pedron, T.; Procópio, V.A.; Rocha, B.A.; Batista, B.L. Agronomic management practices to tackle toxic metal entry into crop plants. In Sustainable Solutions for Elemental Deficiency and Excess in Crop Plants; Mishra, K., Tandon, P.K., Srivastava, S., Eds.; Springer: Singapore, 2020. [CrossRef] g p [ ] 21. Guo, K.; Yan, L.; He, Y.; Li, H.; Lam, S.S.; Peng, W.; Sonne, C. Phytoremediation as a potential technique for vehicle hazardous pollutants around highways. Environ. Pollut. 2023, 322, 121130. [CrossRef] [PubMed] 22. Li, Y.; Rahman, S.U.; Qiu, Z.; Shahzad, S.M.; Nawaz, M.F.; Huang, J.; Naveed, S.; Li, L.; Wang, X.; Cheng, H. Toxic effects of cadmium on the physiological and biochemical attributes of plants, and phytoremediation strategies: A review. Environ. Pollut. 2023, 325, 121433. [CrossRef] 23. Pasricha, S.; Mathur, V.; Garg, A.; Lenka, S.; Verma, K.; Agarwal, S. Molecular mechanisms underlying heavy metal uptake, translocation and tolerance in hyperaccumulators—An analysis Heavy metal tolerance in hyperaccumulators. Environ. Chall. 2021, 4, 100197. [CrossRef] , , [ ] 24. Niu, L.; Li, C.; Wang, W.; Zhang, J.; Scali, M.; Li, W.; Liu, H.; Tai, F.; Hu, X.; Wu, X. Cadmium tolerance and hyperaccumulation in plants—A proteomic perspective of phytoremediation. Ecotoxic. Environ. Saf. 2023, 256, 114882. [CrossRef] 24. Niu, L.; Li, C.; Wang, W.; Zhang, J.; Scali, M.; Li, W.; Liu, H.; Tai, F.; Hu, X.; Wu, X. Cadmium tolerance a plants—A proteomic perspective of phytoremediation. Ecotoxic. Environ. Saf. 2023, 256, 114882. [Cross 25. Chaudhry, H.; Nisar, N.; Mehmood, S.; Iqbal, M.; Nazir, A. Indian mustard Brassica juncea efficiency for the accumulation, tolerance and translocation of zinc from metal contaminated soil. Biocat. Agricul. Biotech. 2020, 23, 101489. References Role of zinc in plant nutrition—A review. Am. J. Exp. Agricul. 2013, 3, 374–391. [CrossRef] 9. Stanton, C.; Sanders, D.; Krämer, U.; Podar, D. Zinc in plants: Integrating homeostasis and biofortification. Mol. Plant 2022, 15, 8. Hafeez, B.; Khanif, Y.M.; Saleem, M. Role of zinc in plant nutrition—A review. Am. J. Exp. Agricul. 2013, 3, 374–391. [CrossRef] 8. Hafeez, B.; Khanif, Y.M.; Saleem, M. Role of zinc in plant nutrition—A review. Am. J. Exp. Agricul. p p g 9. Stanton, C.; Sanders, D.; Krämer, U.; Podar, D. Zinc in plants: Integrating homeostasis and biofortification. Mol. Plant 2022, 15, 65–85. [CrossRef] 10. Gupta, N.; Ram, H.; Kumar, B. Mechanism of zinc absorption in plants: Uptake, transport, transloc Environ. Sci. Biotechnol. 2016, 15, 89–109. [CrossRef] 11. Moreira, A.; Moraes, L.A.C.; dos Reis, A.R. The molecular genetics of zinc uptake and utilization efficiency in crop plants. In Plant Micronutrient Use Efficiency; Hossain, M.A., Kamiya, T., Burritt, D.J., Tran, L.-S.P., Fujiwara, T., Eds.; Elsevier: London, UK, 2018. [CrossRef] 12. Haroon, B.; Irshad, M.; Ebadi, A.G.; An, P. Soil Metal Contamination and Its Mitigation. In Sustainable Plant Nutrition under Contaminated Environments. Sustainable Plant Nutrition in a Changing World; Mahmood, Q., Ed.; Springer: Cham, Switzerland, 2022. [CrossRef] 11 of 13 Horticulturae 2023, 9, 1002 13. Berni, R.; Luyckx, M.; Xu, X.; Legay, S.; Sergeant, K.; Hausman, J.-F.; Lutts, S.; Cai, G.; Guerriero, G. Reactive oxygen species and heavy metal stress in plants: Impact on the cell wall and secondary metabolism. Environ. Exp. Bot. 2019, 161, 98–106. [CrossRef] 14. Bortoloti, G.A.; Baron, D. Phytoremediation of toxic heavy metals by Brassica plants: A biochemical and physiological approach. Environ. Adv. 2022, 8, 100204. [CrossRef] 13. Berni, R.; Luyckx, M.; Xu, X.; Legay, S.; Sergeant, K.; Hausman, J.-F.; Lutts, S.; Cai, G.; Guerriero, G. Reactive oxygen species and heavy metal stress in plants: Impact on the cell wall and secondary metabolism. Environ. Exp. Bot. 2019, 161, 98–106. [CrossRef] y p p y p 14. Bortoloti, G.A.; Baron, D. Phytoremediation of toxic heavy metals by Brassica plants: A biochemical and physiological approach. Environ. Adv. 2022, 8, 100204. [CrossRef] 15. Correa, S.M.; Fernie, A.R.; Nikoloski, Z.; Brotman, Y. Towards model-driven characterization and manipulation of plant lipid metabolism. Progress Lipid Res. 2020, 80, 101051. [CrossRef] [PubMed] 16. Wan, X.; Wu, S.; Li, Z.; An, X.; Tian, Y. Lipid metabolism: Critical roles in male fertility and other aspects of reproductive development in plants. Mol. References [CrossRef] h l l l d 26. Zhang, Z.-W.; Deng, Z.-L.; Tao, Q.; Peng, H.-Q.; Wu, F.; Fu, Y.-F.; Yang, X.-Y.; Xu, P.-Z.; Li, Y.; Wang, C.-Q.; et al. Salicylate and glutamate mediate different Cd accumulation and tolerance berween Brassica napus and B. juncea. Chemosphere 2022, 292, 133466. [CrossRef] [PubMed] 27. Gisbert, C.; Clemente, R.; Navarro-Aviˇnó, J.; Baixauli, C.; Ginér, A.; Serrano, R.; Walker, D.J.; Bernal, M.P. Tolerance and accumulation of heavy metals by Brassicaceae species grown in contaminated soils from Mediterranean regions of Spain. Environ. Experim. Bot. 2006, 56, 19–27. [CrossRef] 28. Feigl, G.; Kolbert, Z.; Lehotai, N.; Molnár, Á.; Ördög, A.; Bordé, Á.; Laskay, G.; Erdei, L. Different zinc sensitivity of Brassica organs is accompanied by distinct responses in protein nitration level and pattern. Ecotoxic. Environoment. Saf. 2016, 125, 141–152. [CrossRef] 29. Płociniczak, T.; Sinkkonen, A.; Romantschuk, M.; Piotrowska-Seget, Z. Characterization of Enterobacter intermedius MH8b and its use for the enhancement of heavy metals uptake by Sinapis alba L. Appl. Soil Ecol. 2013, 63, 1–7. [CrossRef] 30. Torrijos, R.; Righetti, L.; Cirlini, M.; Calani, L.; Manes, J.; Meca, G.; Dall’Astra, C. Phytochemical profiling of volatile and bioactive compounds in yellow mustard (Sinapis alba) and oriental mustard (Brassica juncea) seed flour and bran. LWT 2023, 173, 114221. [CrossRef] 31. Devi, P.A.; Pandiyan, V.; Kumar, T.M.A.S.; Kumar, G.V.P.P.S.R.; Padmanath, K. Dietary supplementation of mustard oil reduces blood glucose levels by triggering insulin receptor signaling pathway. Inter. J. Diab. Develop. Ctries. 2021, 42, 126–137. [CrossRef] 32. Repkina, N.; Nilova, I.; Kaznina, N. Effect of zinc excess in substrate on physiological responses of Sinapis alba L. Plants 2023, 12, 211. [CrossRef] 31. Devi, P.A.; Pandiyan, V.; Kumar, T.M.A.S.; Kumar, G.V.P.P.S.R.; Padmanath, K. Dietary supplementation of mustard oil reduces blood glucose levels by triggering insulin receptor signaling pathway. Inter. J. Diab. Develop. Ctries. 2021, 42, 126–137. [CrossRef] 32 R ki N Nil I K i N Eff t f i i b t t h i l i l f Si i lb L Pl t 2023 12 32. Repkina, N.; Nilova, I.; Kaznina, N. Effect of zinc excess in substrate on physiological responses of 211. [CrossRef] 32. Repkina, N.; Nilova, I.; Kaznina, N. Effect of zinc excess in substrate on physiological responses of Sinapis alba L. Plants 2023, 12, 211. [CrossRef] p , ; , ; , p y g p p , , 211. References Comparative impacts of heavy metals on root growth as related to their specificity and selectivity. Rus. J. Plant Sci. 2003, 50, 398–406. [CrossRef] y 50. Li, S.-P.; Zeng, L.-S.; Su, Z.-L. Wheat growth, photosynthesis and physiological characteristics under different soil Zn levels. J. Integrat. Agricul. 2022, 21, 1927–1940. [CrossRef] 51. Medina-Puche, L.; Castelló, M.J.; Canet, J.V.; Lamilla, J.; Colombo, M.L.; Tornero, P. β-carbonic anhydrases play a role in salicylic acid perception in Arabidopsis. PLoS ONE 2017, 12, e0181820. [CrossRef] 52. Son, J.A.; Narayanankutty, D.P.; Roh, K.S. Influence of exogenous application of glutathione on rubisco and rubisco activase in heavy metal-stressed tobacco plant grown in vitro. Saudi J. Biol. Sci. 2014, 21, 89–97. [CrossRef] Son, J.A.; Narayanankutty, D.P.; Roh, K.S. Influence of exogenous application of glutathione on rubisco an heavy metal-stressed tobacco plant grown in vitro. Saudi J. Biol. Sci. 2014, 21, 89–97. [CrossRef] .A.; Narayanankutty, D.P.; Roh, K.S. Influence of exogenous application of glutathione on rubisco and rubi y metal-stressed tobacco plant grown in vitro. Saudi J. Biol. Sci. 2014, 21, 89–97. [CrossRef] p g hu, G.P.S.; Araniti, F.; Bali, A.S.; Shahzad, B.; Tripathi, D.K.; Brestic, M.; Skalicky, M.; Landi, M. The Rol Plants Exposed to Heavy Metals. Molecules 2020, 25, 540. [CrossRef] Sharma, A.; Sidhu, G.P.S.; Araniti, F.; Bali, A.S.; Shahzad, B.; Tripathi, D.K.; Brestic, M.; Skalicky, M.; L 53. Sharma, A.; Sidhu, G.P.S.; Araniti, F.; Bali, A.S.; Shahzad, B.; Tripathi, D.K.; Brestic, M.; Skalicky, M Salicylic Acid in Plants Exposed to Heavy Metals. Molecules 2020, 25, 540. [CrossRef] 53. Sharma, A.; Sidhu, G.P.S.; Araniti, F.; Bali, A.S.; Shahzad, B.; Tripathi, D.K.; Brestic, M.; Skalicky, M Salicylic Acid in Plants Exposed to Heavy Metals. Molecules 2020, 25, 540. [CrossRef] 54. Moustakas, M.; Dobrikova, A.; Sperdouli, I.; Han’c, A.; Adamakis, I.-D.S.; Moustaka, J.; Apostolova, E. A Hormetic Spatiotemporal Photosystem II Response Mechanism of Salvia to Excess Zinc Exposure. Int. J. Mol. Sci. 2022, 23, 11232. [CrossRef] 55. Sagardoy, R.; Morales, F.; Rellán-Álvarez, R.; Abadía, A.; López-Millán, A.F. Carboxylate metabolism in sugar beet plants grown with excess Zn. J. Plant Physiol. 2011, 168, 730–733. [CrossRef] 56. Baccouch, R.; Shi, Y.; Vernay, E.; Mathelié-Guinlet, M.; Taib-Maamar, N.; Villette, S.; Feuillie, C.; Rascol, E.; Nuss, P.; Lecomte, S.; et al. The impact of lipid polyunsaturation on the physical and mechanical properties of lipid membranes. BBA-Biomembr. 2023, 1865, 184084. [CrossRef] [PubMed] 57. Nouairi, I.; Ghnaya, T.; Youssef, N.B.; Zarrouk, M.; Ghorbel, M.H. References [CrossRef] 33 L i K L H F M G t T J B fii l f hit d I di t d d l t h l t th d t i t 211. [CrossRef] 33. Lewis, K.L.; Hons, F.M.; Gentry, T.J. Beneficial use of white and Indian mustard seed meals to enhance plant growth and nutrient k A G i E i 2019 2 180042 [C R f] [ ] 33. Lewis, K.L.; Hons, F.M.; Gentry, T.J. Beneficial use of white and Indian mustard seed meals to enhance plant growth and nutrient uptake. Agrosyst. Geosci. Environ. 2019, 2, 180042. [CrossRef] 33. Lewis, K.L.; Hons, F.M.; Gentry, T.J. Beneficial use of white and Indian mustard seed meals to enhance p uptake. Agrosyst. Geosci. Environ. 2019, 2, 180042. [CrossRef] p g y 34. Ciubota-Roise, C.; Macoveanu, M.; Fernández, C.M.; Ramos, M.J.; Pérez, A.; Moreno, A. Sinapis alba seed as a prospective biodiesel source. Biomass Bioenergy 2013, 51, 83–90. [CrossRef] gy 35. Luo, Q.; Yu, B.; Liu, Y. Differential sensitivity to chloride and sodiumions in seedlings of Glycine max and G. soja under NaClstress. J. Plant Physiol. 2005, 162, 1003–1012. [CrossRef] 36. Lichtenthaler, H.K.; Wellburn, A.R. Determinations of total carotenoids and chlorophylls a and b of leaf extracts in different solvents. Biochem. Soc. Trans. 1983, 11, 591–592. [CrossRef] 37. Lichlenthaler, H.K. Chlorophylls and carotenoids—Pigments of photosynthetic biomembranes. Method. Enzymol. 1987, 148, 350–382. 12 of 13 12 of 13 Horticulturae 2023, 9, 1002 38. Folch, J.; Lees, M.; Sloane Stanley, G.H. A simple method for the isolation and purification of total lipids from animal tissue. J. Biol. Chem. 1957, 226, 497–509. [CrossRef] 39. Lišˇcáková, P.; Nawaz, A.; Molnárová, M. Reciprocal effects of copper and zinc in plants. Int. J. Environment. Sci. Technol. 2022, 19, 9297–9312. [CrossRef] 40. Belafrej, H.; Bogusz, D.; Triqui, Z.-E.A.; Guedira, A.; Bendaou, N.; Smouni, A.; Fahr, M. Zinc hyperaccumulation in plants: A review. Plants 2020, 9, 562. [CrossRef] 41. Choudhury, M.R.; Islam, M.S.; Ahmed, Z.U.; Nayar, F. Phytoremediation of heavy metal contaminated Buriganga riverbed sediment by Indian mustard and Marigold plants. Environ. Prog. Sustain. Energy 2015, 35, 117–124. [CrossRef] 42. Vázquez, M.D.; Barceló, J.; Poschenrieder, C.; Mádico, J.; Hatton, P.; Baker, A.J.M.; Cope, G.H. Localization of Zinc and cadmium in Thlaspi caerulescens (Brassicaceae), a metallophyte that can hyperaccumulate both metals. J. Plant Physiol. 1992, 140, 350–355. [CrossRef] 43. Kushwaha, P.; Neilson, J.W.; Maier, R.M.; Babst-Kostecka, A. References Soil microbial community and abiotic soil properties influence Zn and Cd hyperaccumulation differently in Arabidopsis halleri. Sci. Total Environ. 2022, 803, 150006. [CrossRef] [PubMed] 44. Solelmannejad, Z.; Sadeghlpour, H.R.; Abdolzadeh, A.; Golallpour, M. Physiological responses of white mustard grown in Zn-contaminated soils. Acta Physiol. Plant. 2020, 42, 131–145. [CrossRef] 43. Kushwaha, P.; Neilson, J.W.; Maier, R.M.; Babst-Kostecka, A. Soil microbial community and abiotic soil properties influence Zn and Cd hyperaccumulation differently in Arabidopsis halleri. Sci. Total Environ. 2022, 803, 150006. [CrossRef] [PubMed] yp y p 44. Solelmannejad, Z.; Sadeghlpour, H.R.; Abdolzadeh, A.; Golallpour, M. Physiological responses of white mustard grown in Zn-contaminated soils. Acta Physiol. Plant. 2020, 42, 131–145. [CrossRef] 44. Solelmannejad, Z.; Sadeghlpour, H.R.; Abdolzadeh, A.; Golallpour, M. Physiological responses of Zn-contaminated soils. Acta Physiol. Plant. 2020, 42, 131–145. [CrossRef] y 45. Chowardhara, B.; Borgohain, P.; Saha, B.; Awasthi, J.P.; Moulick, D.; Panda, S.K. Phytotoxicity of Cd Indian mustard varieties during germination and early seedling growth. Biocat. Agricul. Biotech. 2019, 45. Chowardhara, B.; Borgohain, P.; Saha, B.; Awasthi, J.P.; Moulick, D.; Panda, S.K. Phytotoxicity of Cd and Zn on three popular Indian mustard varieties during germination and early seedling growth. Biocat. Agricul. Biotech. 2019, 21, 101349. [CrossRef] 46. Islam, F.; Yasmeen, T.; Riaz, M.; Arif, M.S.; Ali, S.; Raza, S.H. Proteus mirabilis alleviates zinc toxicity by preventing oxidative stress in maize (Zea mays) plants. Ecotox. Environ. Saf. 2014, 110, 143–152. [CrossRef] [PubMed] g g y g g g 46. Islam, F.; Yasmeen, T.; Riaz, M.; Arif, M.S.; Ali, S.; Raza, S.H. Proteus mirabilis alleviates zinc toxicity by preventing oxidative stress in maize (Zea mays) plants. Ecotox. Environ. Saf. 2014, 110, 143–152. [CrossRef] [PubMed] ( y ) p f 47. Ouni, Y.; Mateos-Naranjo, E.; Abdelly, C.; Lakhdar, A. Interactive effect of salinity and zinc stress on growth and photosynthetic responses of the perennial grass, Polypogon monspeliensis. Ecol. Engineer. 2016, 95, 171–179. [CrossRef] 47. Ouni, Y.; Mateos-Naranjo, E.; Abdelly, C.; Lakhdar, A. Interactive effect of salinity and zinc stress on growth and photosynthetic responses of the perennial grass Polypogon monspeliensis Ecol Engineer 2016 95 171 179 [CrossRef] 48. Mahmoudi, H.; Salah, I.B.; Zaouali, W.; Zorrig, W.; Smaoul, A.; Ali, T.; Gruber, M.; Ouerghi, Z.; Hosni, K. Impact of zinc excess on germination, growth parameters and oxidative stress of sweet basil (Ocimum basilicum L.). Bull. Environ. Contam. Toxic. 2021, 106, 899–907. [CrossRef] [PubMed] 49. Ivanov, V.B.; Bystrova, E.I.; Seregin, I.V. References Changes in content and fatty acid profiles of total lipids of two halophytes: Sesuvium portulacastrum and Mesembryanthemum crystallinum under cadmium stress. J. Plant Physiol. 2006, 163, 1198–1202. [CrossRef] 58. Zamanová, V.; Pavlík, M.; Kyjaková, P.; Pavlíková, D. Fatty acid profiles of ecotypes of hyperaccumulator Noccaeacaerulescens growing under cadmium stress. J. Plant Physiol. 2015, 180, 27–34. [CrossRef] [PubMed] 59. Seregin, I.V.; Ivanova, T.V.; Voronkov, A.S.; Kozhevnikova, A.D.; Schat, H. Zinc- and nickel-induced changes in fatty acid profiles in the zinc hyperaccumulator Arabidopsis halleri and non-accumulator Arabidopsis lyrata. Plant Physiol. Biochem. 2023, 197, 107640. [CrossRef] 60. He, M.; Ding, N.-Z. Plant unsaturated fatty acids: Multiple roles in stress response. Front. Plant Sci. 202 [PubMed] 61. Sánchez-Calvo, B.; Barroso, J.B.; Corpas, F.J. Hypothesis: Nitro-fatty acids a role in plant metabolism. Plant Sci. 2013, 199–200, 1–6. [CrossRef] [ ] 62. Buist, P.H. Unsaturated fatty acids. In Comprehensive Natural Products II. Chemistry and Biology; Liu, H.-W., Mander, L., Eds.; Elsevier: London, UK, 2010; pp. 5–33. 63. Voronin, V.P.; Artemenkov, D.V.; Orlov, A.M.; Murzina, S.A. Lipids and fatty acids in some mesopelagic fish species: General characteristics and peculiarities of adaptive response to deep-water habitat. J. Mar. Sci. Eng. 2022, 10, 949. [CrossRef] 64. Marichali, A.; Dallali, S.; Ouerghemmi, S.; Sebei, H.; Hosni, K. Germination, morpho-physiological and biochemical responses of coriander (Coriandrum sativum L.) to zinc excess. Indus. Crop Prod. 2014, 55, 248–257. [CrossRef] 13 of 13 13 of 13 Horticulturae 2023, 9, 1002 65. Beniwal, V.; Nehra, K.S.; Chhokar, V. Cadmium induced alteration in lipid profile of developing mustard (Brassica juncea L.) seed. Biocat. Agric. Biotech. 2015, 4, 416–422. [CrossRef] 65. Beniwal, V.; Nehra, K.S.; Chhokar, V. Cadmium induced alteration in lipid profile of developing mustard (Brassica juncea L.) seed. Biocat. Agric. Biotech. 2015, 4, 416–422. [CrossRef] g 66. Repkina, N.; Murzina, S.A.; Voronin, V.P.; Kaznina, N. Does methyl jasmonate effectively protect plants under heavy metal contamination? Fatty acid content in wheat leaves exposed to cadmium with or without exogenous methyl jasmonate application. Biomolecules 2023, 13, 582. [CrossRef] 67. Zhukov, A.V. Palmitic acid and its role in the structure and functions of plant cell membranes. Rus. J. Plant Physiol. 2015, 62, 706–713. [CrossRef] 706 713. [CrossRef] 68. M’Rah, S.; Marichall, A.; M’Rabet, Y.; Chatti, S.; Casablanca, H.; Hosni, K. Morphology, physiology, and biochemistry of zinc-stressed caraway plants. Protoplasma 2023, 260, 853–868. [CrossRef] [PubMed] 68. M’Rah, S.; Marichall, A.; M’Rabet, Y.; Chatti, S.; Casablanca, H.; Hosni, K. 65. Beniwal, V.; Nehra, K.S.; Chhokar, V. Cadmium induced alteration in lipid profile of developing mustard (Brassica juncea L.) seed. Biocat. Agric. Biotech. 2015, 4, 416–422. [CrossRef] g 66. Repkina, N.; Murzina, S.A.; Voronin, V.P.; Kaznina, N. Does methyl jasmonate effectively protect plants under heavy metal contamination? Fatty acid content in wheat leaves exposed to cadmium with or without exogenous methyl jasmonate application. Biomolecules 2023, 13, 582. [CrossRef] References Morphology, physiology, and biochemistry of zinc-stressed caraway plants. Protoplasma 2023, 260, 853–868. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W4322713214
https://zenodo.org/records/8074450/files/Distinctia_dintre_falsificare_si_respingere_in_problema_demarcatie_%20la_Karl_Popper-Nicolae_Sfetcu-GTWK.pdf
Romanian, Moldavan
null
Distincția dintre falsificare și respingere în problema demarcației la Karl Popper
null
2,018
cc-by
13,333
Distincția dintre falsificare și respingere în problema demarcației la Karl Popper Popper Nicolae Sfetcu 03.06.2018 Colecția ESEURI Publicat de MultiMedia Publishing, Drobeta Turnu Severin, 2019, www.telework.ro/ro/editura Sfetcu, Nicolae, "Distincția dintre falsificare și respingere în problema demarcației la Karl Popper", SetThings (3 iunie 2018), MultiMedia Publishing (ed.), DOI: 10.58679/TW55590, ISBN: 978-606-033-139-1, URL = https://www.telework.ro/ro/e-books/distinctia-dintre-falsificare-si-respingere-in- problema-demarcatiei-la-karl-popper/ Email: nicolae@sfetcu.com Această carte este publicată sub licenţă Creative Commons Attribution-NoDerivatives 4.0 International. Pentru a vedea o copie a acestei licenţe, vizitaţi http://creativecommons.org/licenses/by-nd/4.0/. Colecția ESEURI Publicat de MultiMedia Publishing, Drobeta Turnu Severin, 2019, www.telework.ro/ro/editura Sfetcu, Nicolae, "Distincția dintre falsificare și respingere în problema demarcației la Karl Popper", SetThings (3 iunie 2018), MultiMedia Publishing (ed.), DOI: 10.58679/TW55590, ISBN: 978-606-033-139-1, URL = https://www.telework.ro/ro/e-books/distinctia-dintre-falsificare-si-respingere-in- problema-demarcatiei-la-karl-popper/ Email: nicolae@sfetcu.com Această carte este publicată sub licenţă Creative Commons Attribution-NoDerivatives 4.0 International. Pentru a vedea o copie a acestei licenţe, vizitaţi http://creativecommons.org/licenses/by-nd/4.0/. 2 Abstract În această lucrare argumentez faptul că, în ciuda criticilor teoriei falsificabilității propuse de Karl Popper pentru demarcarea între știință și ne-știință, în principal pseudoștiință, acest criteriu este încă foarte util, și perfect valabil după perfecționarea lui de către Popper și adepții lui. Mai mult, chiar și în versiunea sa inițială, considerată de Lakatos ca ”dogmatică”, Popper nu a afirmat că această metodologie este un criteriu absolut de demarcare: un singur contra-exemplu nu este suficient pentru a falsifica o teorie; mai mult, o teorie poate fi salvată în mod legitim de falsificare prin introducerea unei ipoteze auxiliare. În comparație cu teoria lui Kuhn a revoluțiilor, de care el însuși s-a dezis ulterior transformând-o într-o teorie a ”micro-revoluțiilor”, consider că metodologia de demarcare a lui Popper, împreună cu dezvoltarea ulterioară propusă de acesta, inclusiv coroborarea și verosimilitudinea, deși imperfectă, nu numai că este valabilă și azi, dar este încă cea mai bună metodologie de demarcare. Pentru argumentare, m-am folosit de principalele lucrări ale lui Popper care tratează această problemă, și a principalilor săi critici și susținători. După o scurtă prezentare a lui Karl Popper, și o introducere în problema demarcației și metodologia falsificabilității, trec în revistă principalele critici aduse și argumentele susținătorilor săi, accentuând pe ideea că Popper nu a pus niciodată semnul egalității între falsificare și respingere. În final prezint propriile concluzii în această problemă. Cuvinte cheie: Karl Popper, falsificabilitate, falsificare, problema demarcației, pseudoștiința 3 Introducere Pentru această discuție am apelat la cărțile lui Karl Popper, Filosofia socială și filosofia științei, (K. Popper 1985), Logica cercetării, (K. Popper 1934), Conjecturi și infirmări,(K. Popper 1963), și Viitorul este deschis, (K. Popper and Lorenz 1985), și la lucrările lui Sven Ove Hansson, "Science and Pseudo-Science"(Hansson 2017), Stephen Thornton, "Karl Popper" (Thornton 2017), Paul Newall, ”Falsificationism” (Newall 2005), Imre Lakatos, ”Falsification and the Methodology of Scientific Research Programs” (Imre Lakatos 1970), Brendan Shea, ”Karl Popper: Philosophy of Science Brendan Shea” (Shea 2017), David Miller, ”Some Hard Questions for Critical Rationalism” (Miller 2009b), Alan Musgrave și Charles Pigden, "Imre Lakatos" (Musgrave and Pigden 2016), Suddhachit Mitra, ”What Constitutes Science: Falsifiability as a Criterion of Demarcation” (Mitra 2016), Carl G. Hempel, ”Empirical Statements and Falsifiability” (Hempel 1958), Milos Taliga, ”Against Watkins: From a Popperian point of view” (Taliga 2004), D. C. Stove, ”Popper on Scientific Statements” (Stove 1978), A. A. Derksen, ”The Alleged Unity of Popper's Philosophy of Science: Falsifiability as Fake Cement” (Derksen 1985), Lansana Keita, ”Are Universal Statements Falsifiable? A Critique of Popper's Falsifiability Criterion” (Keita 1989), și Stephen Toulmin, ”Conceptual Revolutions in Science” (Toulmin 1967). Traducerile citatelor aparțin autorului acestui eseu. Traducerile citatelor aparțin autorului acestui eseu. După o scurtă prezentare a lui Popper, relevantă în contextul discuției, prezint principalele puncte de vedere în problema demarcației în capitolul cu același nume. În Pseudoștiință delimitez mai clar această noțiune de cea de ne-știință și non-știință, iar în capitolul Falsificabilitatea prezint pe scurt metodologia propusă de Popper. Următorul capitol, Falsificare și respingere, este cel în care argumentez distincția clară, semnalată și de Popper, dintre cele două noțiuni. În Extinderea falsificabilității prezint evoluția acestui concept, cu îmbunătățirile propuse atât de Popper cât și de 4 alți cercetători. Următoarele două capitole, Critici ai falsificabilității și Susțineri ale falsificabilității, prezint argumentele pro și contra expuse de cercetători față de metodologia falsificabilității. După un scurt capitol privind Tendințele actuale în problema demarcării științei, prezint opiniile mele în Concluzii, Karl Popper, ca raționalist critic, a fost un oponent al tuturor formelor de scepticism, convenționalism și relativism în știință. În 1935 a scris Logica cercetării (Logik der Forschung. Zur Erkenntnistheorie der modernen Naturwissenschaft) (K. Popper 1934), traducând ulterior cartea în engleză și publicând-o sub titlul The Logic of Scientific Discovery (1959) considerată ca o lucrare de pionierat în domeniu. Introducere Multe dintre argumentele din această carte sunt îndreptate împotriva membrilor "Cercului Vienez", precum Moritz Schlick, Otto Neurath, Rudolph Carnap, Hans Reichenbach, Carl Hempel și Herbert Feigl. Popper este de acord cu aceștia cu privire la aspectele generale ale metodologiei științifice și neîncrederea lor în metodologia filosofică tradițională, dar soluțiile sale au fost semnificativ diferite. Popper a contribuit semnificativ la dezbaterile privind metodologia științifică generală, demarcarea științei de pseudoștiință, natura probabilității și metodologia științelor sociale. Popper a fost profund impresionat de diferențele dintre presupusele teorii "științifice" ale lui Freud și Adler și de revoluția declanșată de teoria relativității lui Einstein în fizică în primele două decenii ale secolului 20. În timp ce teoria lui Einstein era extrem de "riscantă", în sensul că era posibil să se deducă consecințe din ea care, dacă s-ar fi dovedit a fi false, ar fi falsificat întreaga teorie, nimic nu putea, în principiu, să falsifice teoriile psihanalitice, acestea nefiind cu adevărat predictive. (Thornton 2017) Popper a început să fie criticat filosofic pentru abordarea sa prescriptivă față de știință și accentul pus pe logica falsificării. Teoriei sale i s-a opus abordarea socio-istorică a lui Thomas 5 Kuhn în "Structura revoluțiilor științifice" (1962), (Kuhn 1962) care, în susținerea a reintrodus ideea că schimbarea științei este în esență dialectică și depinde de stabilirea unui consens în cadrul comunităților de cercetători. teorii. Popper susține că nu există o metodologie unică specifică științei. teorii. Popper susține că nu există o metodologie unică specifică științei. teorii. Popper susține că nu există o metodologie unică specifică științei. ”Este necesar să se rezolve problema demarcării științei de metafizică. Dar ar trebui să recunoaștem că multe sisteme metafizice au condus la rezultate științifice importante. Amintesc numai sistemul lui Democrit; și cel al lui Schopenhauer care este foarte asemănător cu cel al lui Freud. Și unele, de exemplu cele ale lui Platon sau Malebranche sau Schopenhauer sunt construcții minunate ale gândirii. Dar cred, în același timp, că ar trebui să ne opunem acelor sisteme metafizice care au tendința de a ne fermeca și nedumeri. Dar evident, ar trebui să procedăm la fel chiar cu sisteme nemetafizice sau anti-metafizice, dacă afișează această tendință periculoasă. Și cred că nu o putem face dintr-o singură mișcare. Mai degrabă trebuie să facem efortul de a analiza sistemele în detaliu; trebuie să arătăm că înțelegem ce vrea autorul să spună, dar că ce spune el nu merită efortul de a înțelege.” (K. Popper 1985, 109) Popper propune în schimb, ca metodă pentru investigației științifică, falsificabilitatea. Popper propune în schimb, ca metodă pentru investigației științifică, falsificabilitatea. Pentru el, o teorie este științifică numai dacă este falsificabilă de un eveniment conștient. Teoria de demarcație a lui Popper se bazează pe percepția sa asupra asimetriei logice pe care o are între verificare și falsificare: este logic imposibil să se verifice definitiv o propoziție universală prin referire la experiență (cum afirmă și Hume), dar un singur contra-exemplu falsifică definitiv legea universală corespunzătoare. Într-un cuvânt, o excepție, departe de a se "dovedi" o excepție de la regulă, o respinge definitiv: (Thornton 2017) ”e vorba de oameni ai ideilor curajoase, deși foarte critici față de propriile lor idei, ei încearcă să afle dacă ideile lor sunt corecte, încercând să afle mai întâi dacă nu cumva sunt greșite. Ei operează cu conjecturi curajoase și încercări severe de a-și respinge propriile lor conjecturi . Criteriul de demarcație între știință și ne-știință pe care îl propun este o simplă analiză logică a acestei imagini. Dacă acesta este bun sau rău, acest lucru va fi arătat de fertilitatea lui. Ideile curajoase reprezintă ipoteze sau conjecturi noi și îndrăznețe. Iar încercările severe de respingere sunt discuțiile critice și testele empirice severe. 1 Problema demarcației Au existat încercări de demarcare a științei de neștiință încă din perioada antică: ”Pentru a fi științific, a spus Aristotel, trebuie să ne ocupăm de cauze, trebuie să folosim demonstrații logice și trebuie să identificăm universalii care "intră" în particularitățile sensului.” (Laudan 1983) Demarcarea științei de pseudoștiință are rațiuni atât teoretice (problema delimitării este o perspectivă iluminantă care contribuie la filozofia științei în același mod în care analiza erorilor contribuie la studiul logicii informale și a argumentării raționale), cât și practice (distincția este importantă pentru îndrumarea decizională atât în viața privată, cât și în cea publică). (Mahner 2007) Pozitivismul logic, prin teoria verificabilității semnificației (verificaționism) a considerat că numai afirmațiile despre chestiuni de fapt sau relațiile logice dintre concepte sunt semnificative. (Grayling 2001) Dar "propunerile de verificare aveau ca scop rezolvarea unei probleme de demarcare distinctă, și anume între știință și metafizică". (Hansson 2017) Conform lui Popper, problema centrală a filozofiei științei este cea a demarcării, a distingerii dintre știință și ceea ce el numește "non-știință" (incluzând logica, metafizica, psihanaliza, etc). ”Orice demarcație, în sensul meu, trebuie să fie inexactă. (Aceasta este una dintre marile diferențe față de orice criteriu al semnificației formale al oricărui "limbaj al științei" artificial). Pentru că linia de despărțire dintre metafizică și știință nu este una clară: ceea ce era ieri o idee metafizică poate deveni mâine o teorie științifică testabilă, ceea ce se întâmplă de multe ori.” (K. Popper 1985, 131) ”Vor exista teorii bine testabile, teorii testabile cu greu și teorii netestabile. Acelea care sunt netestabile nu prezintă niciun interes pentru știința empirică. Ele pot fi caracterizate 6 drept teorii metafizice.... ar fi inadecvat să trasăm linia de demarcație dintre știință și metafizică astfel încât să excludem metafizica din domeniul limbajului cu semnificație pe motivul că este lipsită de sens.” (K. Popper 1963, 333) Un argument principal al lui Popper este critica lui Hume a inducției (Hume 1738), argumentând că inducția nu trebuie niciodată folosită în știință. Dar nu este de acord cu scepticismul asociat cu Hume și nici cu susținerea "observației" pure ale lui Bacon și Newton ca punct de pornire în formarea teoriilor, întrucât nu există observații pure, care să nu implice anumite teorii. Popper susține că nu există o metodologie unică specifică științei. Dar când ”e vorba de oameni ai ideilor curajoase, deși foarte critici față de propriile lor idei, ei încearcă să afle dacă ideile lor sunt corecte, încercând să afle mai întâi dacă nu cumva sunt greșite. Ei operează cu conjecturi curajoase și încercări severe de a-și respinge propriile lor conjecturi . Criteriul de demarcație între știință și ne-știință pe care îl propun este o simplă analiză logică a acestei imagini. Dacă acesta este bun sau rău, acest lucru va fi arătat de fertilitatea lui. Ideile curajoase reprezintă ipoteze sau conjecturi noi și îndrăznețe. Iar încercările severe de respingere sunt discuțiile critice și testele empirice severe. Dar când 7 este o conjectură îndrăzneață, în sensul propus aici , și când nu? Este îndrăzneață dacă și numai dacă își asumă un mare risc de a fi falsă - dacă lucrurile ar sta altfel, și dacă par în acel moment că stau altfel.” (K. Popper 1985, 127–28) O teorie științifică veritabilă este restrictivă, și deci poate fi testată și falsificată, dar niciodată verificată logic. Astfel, dacă o teorie a rezistat testelor, nu înseamnă că a fost verificată, ea are doar un grad mai mare de coroborare, putând fi înlocuită oricând de o teorie mai bună. Popper folosește falsificarea drept criteriu de demarcare pentru a evalua teoriile. Criteriul popperian nu exclude din domeniul științei declarații care nu pot fi falsificate, ci doar teorii întregi care nu conțin nicio declarație falsificabilă, fără a preciza totuși foarte clar ce constituie o "teorie întreagă" și ce face o afirmație să fie "semnificativă". Verificaționismul 1 a fost o caracteristică esențială a pozitivismului logic al așa-numitului Cerc de la Viena. Popper a observat că filozofii Cercului de la Viena au amestecat două probleme diferite, cele de semnificație și de demarcare, și au propus ca verificare o singură soluție pentru ambele. Popper a afirmat că există teorii semnificative care nu sunt științifice, și deci un criteriu de semnificație nu coincide cu un criteriu de delimitare, propunând înlocuirea verificabilității cu falsificabilitatea ca criteriu de delimitare. Pe de altă parte, el s-a opus strict opiniei că afirmațiile care nu sunt falsificabile sunt lipsite de sens sau greșite. (K. Popper 1934, sec. 6) Popper susține că singura tehnică logică parte integrantă din metoda științifică este cea a testării deductive, concluziile fiind deduse dintr-o ipoteză și apoi comparate între ele și cu alte declarații relevante pentru a determina dacă acestea falsifică sau coroborează ipoteza. 1 Verificaționismul susține că o declarație trebuie, în principiu, să fie verificată empiric pentru a fi atât semnificativă, cât și științifică. 1 Verificaționismul susține că o declarație trebuie, în principiu, să fie verificată empiric semnificativă cât și științifică p ; ( ) p p p pentru a verifica dacă teoria este coroborată (dar nu verificată). (K. Popper 1934, sec. 1.3) eventualele contradicții; (2) Axiomizarea teoriei pentru a distinge între elementele empirice și ce Compararea noii teorii cu cele existente; (4) Testarea teoriei prin aplicarea empirică a concluziil pentru a verifica dacă teoria este coroborată (dar nu verificată) (K Popper 1934 sec 1 3) eventualele contradicții; (2) Axiomizarea teoriei pentru a distinge între elementele empirice și cele logice; (3) Compararea noii teorii cu cele existente; (4) Testarea teoriei prin aplicarea empirică a concluziilor derivate din ea, pentru a verifica dacă teoria este coroborată (dar nu verificată). (K. Popper 1934, sec. 1.3) alele contradicții; (2) Axiomizarea teoriei pentru a distinge între elementele empirice și cele logice; 2 Etapele pentru procedura deductivă, conform lui Popper: (1) O testare a coerenței interne pentru a vedea e ent alele contradicții; (2) A iomi area teoriei pentr a distinge între elementele empirice și cele logice; (3) eventualele contradicții; (2) Axiomizarea teoriei pentru a distinge între elementele empirice și cele l Compararea noii teorii cu cele existente; (4) Testarea teoriei prin aplicarea empirică a concluziilor d 2 Etapele pentru procedura deductivă, conform lui Popper: (1) O testare a coerenței interne ț ; ( ) p g p ș g ; ( ) ea noii teorii cu cele existente; (4) Testarea teoriei prin aplicarea empirică a concluziilor derivate din Compararea noii teorii cu cele existente; (4) Testarea teoriei prin aplicarea empirică a concluziilor derivate din ea, pentru a verifica dacă teoria este coroborată (dar nu verificată). (K. Popper 1934, sec. 1.3) teorii. Popper susține că nu există o metodologie unică specifică științei. Astfel de concluzii nu sunt în mod direct comparate cu faptele, pur și simplu pentru că nu există fapte "pure" disponibile; toate observațiile-declarații sunt încărcate de teorie și sunt la fel de mult o funcție de 8 factori pur subiectivi (interese, așteptări, dorințe etc.), deoarece acestea sunt o funcție a ceea ce este realmente obiectiv. (Thornton 2017) factori pur subiectivi (interese, așteptări, dorințe etc.), deoarece acestea sunt o funcție a ceea ce este realmente obiectiv. (Thornton 2017) Popper specifică patru etape pentru procedura deductivă 2: ”eu am propus ... să considerăm drept criteriu de demarcație caracterul infirmabil sau falsificabil al sistemelor teoretice. Corespunzător acestei concepții, pe care o mai susțin și acum, un sistem trebuie să fie considerat științific numai dacă face aserțiuni care pot intra în conflict cu observațiile; și în fapt, un sistem este testat prin încercări de a produce asemenea conflicte, altfel spus, prin încercări de a-l infirma. Așadar, testabilitatea este același lucru cu infirmabilitatea, și, în consecință, poate fi considerată drept un criteriu de demarcație. .... (după cum am descoperit ulterior), există grade de testabilitate: unele teorii se expun la infirmări posibile cu mai mult curaj decât altele.” (K. Popper 1963, 332) Popper consideră că filosofia lui Hume demonstrează că există o contradicție implicită în empirismul tradițional, care susține atât că toate cunoștințele provin din experiență, cât și că propozițiile universale (inclusiv legile științifice) sunt verificabile prin referire la experiență. Contradicția derivă din încercarea de a arăta că, în ciuda caracterului deschis al experienței, legile științifice pot fi interpretate ca generalizări empirice, care într-un fel confirmă în final o experiență "pozitivă". Popper elimină contradicția prin respingerea primului dintre aceste principii și eliminând impunerea verificării empirice în falsificare în al doilea principiu. El afirmă că teoriile științifice nu sunt deduse inductiv din experiență și nici nu se efectuează experimente științifice pentru verificarea sau stabilirea adevărului lor; toate cunoștințele sunt provizorii, conjecturale, ipotetice - nu putem niciodată să dovedim definitiv teoriile, putem doar să le confirmăm (provizoriu) sau să le refuzăm. De aceea trebuie să facem o alegere între teoriile care explică setul de fenomene investigate, eliminând numai acele teorii care sunt falsificate și alegem rațional între teoriile rămase, nefalsificate, pe cea care posedă cel mai înalt nivel de forță explicativă și de putere 9 predictivă. Popper pune accent pe importanța spiritului critic al științei - gândirea critică este însăși esența raționalității. teorii. Popper susține că nu există o metodologie unică specifică științei. (Thornton 2017) Au existat diferite propuneri privind demarcarea: aceasta trebuie să se refere la un program de cercetare (I. Lakatos 1974, 248–49), un domeniu epistemic sau o disciplină cognitivă, reprezentând scopuri comune de cunoaștere și practici (Bunge 1982) (Mahner 2007) o teorie (K. Popper 1963), o practică (Lugg 1992) (Morris 1987), o problemă sau o întrebare științifică (Siitonen 1984) și o anchetă anume (Kuhn 1970) (Mayo 1996). Dificultatea constă în selecția metodei de demarcare. Derksen (Derksen 1993) pune accentul în demarcare pe omul de pseudoștiință (persoana care promovează pseudoștiința), în ideea că pseudoștiința are pretenții științifice și astfel de pretenții sunt asociate cu o persoană, nu cu o teorie. 3 Pseudoştiinţa este o metodologie, credinţă sau practică, considerată de susţinătorii ei drept ştiinţifică, sau care pare a fi ştiinţifică dar nu aderă la o metodologie ştiinţifică adecvată, lipsindu-i susţinerea prin evidenţă sau caracterul plauzibil, sau statutul ştiinţific confirmat. (Sfetcu 2015, 1) 2 Pseudoștiința Delimitarea dintre știință și pseudoștiință face parte din sarcina mai generală de a determina care credințe sunt justificate epistemic. Știința poate fi descrisă ca fiind parțial descriptivă, parțial normativă. O definiție a științei se poate concentra pe conținutul descriptiv și specifică modul în care termenul este utilizat efectiv, sau, se poate concentra asupra elementului normativ și poate clarifica sensul mai fundamental al termenului. 3 Cea mai veche utilizare a acestei denumiri se consideră ca aparținând fiziologului francez François Magendie, (Magendie 1843) care este considerat unul din pionierii fiziologiei experimentale. Există o divergență între filozofii științei și unii membri ai comunității științifice referitor la posibilitatea unei distincții obiective între “pseudoștiință” și “știință”. 10 Profesorul Paul DeHart Hurd (Hurd 1998) consideră că o mare parte din oamenii de știință sunt în stare să distingă între știință și diferitele pseudoștiințe precum astrologia (Sfetcu 2015, 257), șarlatania, ocultismul, (Sfetcu 2015, 229) sau superstiția. Pseudoștiința este oricare subiect care pare a fi științific la prima vedere, sau ai cărui susținători susțin că este științific, dar care contravine condițiilor de testare sau deviază mult de la alte aspecte fundamentale ale metodelor științifice. Standardele pentru demarcare pot varia în funcție de domeniu, dar o serie de principii de bază sunt unanim acceptate. Toate rezultatele experimentale ar trebui să fie reproductibile, metoda științifică să poată fi aplicată peste tot, prejudecățile să poată fi controlate sau eliminate, experimentele să fie corecte, studiile să fie obiective, etc. Datele să fie documentate pentru a fi reproduse și pentru studii suplimentare. Cuantificarea statistică a importanței, încrederii, și erorii sunt, de asemenea, instrumente importante pentru metoda științifică. (Sfetcu 2015, 2) Conflictul dintre știință și pseudoștiință se regăsește, pe de o parte, în comunitatea de discipline de cunoaștere care include științele naturale și sociale și umaniste, și pe de altă parte în o mare varietate de mișcări și doctrine, cum ar fi creaționismul, (Sfetcu 2015, 112) astrologia, homeopatia și denialismul holocaustului, aflate în conflict cu rezultatele și metodele general acceptate în comunitatea cunoașterii. Nu toate non-științele sunt pseudoștiință, precum metafizica sau religia. Mahner (Mahner 2007, 548) a propus termenul "para-știință" pentru practicile ne-științifice care nu sunt pseudoștiințe. "Neștiințific" este un concept mai restrâns decât "non-științific", deoarece primul termen implică o anumită formă de contradicție sau conflict cu știința, iar "pseudoștiințific" este mai restrâns decât "neștiințific", deoarece implică o intenționalitate. 2 Pseudoștiința 11 Mulți autori presupun că pentru a fi pseudoștiințifică, o activitate sau o învățătură trebuie să satisfacă după două criterii (Hansson 1996): (1) să nu fie științifică și (2) principalii săi susținători să încerce să creeze impresia că este științifică. Există însă fenomene care satisfac ambele criterii, dar nu sunt pseudoștiințifice, precum frauda în știință. Unele forme de pseudo-știință au ca obiectiv lupta împotriva unei teorii științifice (denialismul științific), precum negarea holocaustului nazist (Gleberzon 1984), sau negarea schimbărilor climatice (Williams 2005). De asemenea, teoria "design-ului inteligent" (Sfetcu 2015, 103) susține o interpretare fundamentalistă a genezei negând evoluția. Grove (Grove 1985, 219) a inclus printre doctrinele pseudoștiințifice pe acelea care "intenționează să ofere explicații alternative celor din știință sau să pretindă să explice ce știința nu poate explica". Într-un sens mai larg, se presupune că pseudoștiința include nu numai doctrine contrare științei proclamate a fi științifice, ci și doctrine contrare științei în instanță, indiferent dacă sunt sau nu prezentate în numele științei. (Hansson 1996) 3 Falsificabilitatea Karl Popper a propus falsificabilitatea ca un criteriu important în diferențierea dintre știință și pseudoștiință. El susține că verificarea și confirmarea nu pot juca niciun rol în formularea unui criteriu satisfăcător de delimitare. În schimb, propune ca teoriile științifice să se deosebească de teoriile ne-științifice prin pretenții testabile pe care viitoarele observații le-ar putea dezvălui a fi false. Popper atrage atenția asupra faptului că teoriile științifice sunt caracterizate prin existența unor falsificatori potențiali - declarații care ar putea fi descoperite ca fiind false. Popper este un realist care susține că teoriile științifice urmăresc adevărul, nu crede că dovezile empirice ne pot oferi vreodată motive pentru a crede că o teorie este adevărată sau poate fi adevărată. În acest sens, Popper este un falibilist, afirmând că este imposibil să se justifice 12 confirmând adevărul diferitelor revendicări particulare, Popper descrie știința ca progresând pe un model evolutiv, observațiile selectând împotriva teoriilor nepotrivite prin falsificarea lor. (Shea 2017) Astfel, termenul falsificabilitate este sinonim cu testabilitatea. ”Propunerea mea se bazează pe o asimetrie între verificabilitate și falsificabilitate; o asimetrie care rezultă din forma logică a afirmațiilor universale. Deoarece acestea nu sunt derivate niciodată din afirmații singulare, ele pot fi contrazise de afirmații singulare.” ”Propunerea mea se bazează pe o asimetrie între verificabilitate și falsificabilitate; o asimetrie care rezultă din forma logică a afirmațiilor universale. Deoarece acestea nu sunt derivate niciodată din afirmații singulare, ele pot fi contrazise de afirmații singulare.” - Karl Popper, Logica descoperirii științifice, (K. Popper 1934, 19) - Karl Popper, Logica descoperirii științifice, (K. Popper 1934, 19) Popper a subliniat faptul că afirmațiile nefalsificabile sunt importante în știință. Contrar intuiției, afirmațiile nefalsificabile pot fi încorporate în - și deductiv generate de - teorii falsificabile. Popper a inventat noțiunea de programe de cercetare metafizică pentru a numi astfel de idei nefalsificabile. Spre deosebire de pozitivism, care a considerat că declarațiile sunt lipsite de sens dacă nu pot fi verificate sau falsificate, Popper a susținut că falsificabilitatea este doar un caz special al noțiunii mai generale de criticabilitate, chiar dacă a recunoscut că respingerea empirică este una dintre metodele cele mai eficiente prin care teoriile pot fi criticate. Criticabilitatea, spre deosebire de falsificabilitate și, deci, de raționalitate, poate fi cuprinzătoare (fără limite logice), deși această afirmație este controversată, chiar și printre susținătorii filozofiei și raționalismului lui Popper. 3 Falsificabilitatea Falsificarea deductivă inițială (naivă, dogmatică, sau naturalistă) a declarațiilor are loc prin modus tollens, prin intermediul unei anumite observații. Logica falsificării naive este valabilă dar limitată, datorită posibilelor "ajustări compensatorii". Popper recunoscut aceste limitări (K. Popper 1934) ca răspuns la criticile lui Pierre Duhem. W. V. Quine a numit acest argument holism de confirmare. Pentru a falsifica logic un universal se poate găsi o afirmație singulară adevărată falsificabilă, dar este întotdeauna posibilă schimbarea declarației universale sau a afirmației 13 existențiale, astfel încât să nu se producă falsificarea. Astfel, falsificarea naivă nu permite existențiale, astfel încât să nu se producă falsificarea. Astfel, falsificarea naivă nu permite oamenilor de știință să prezinte o falsificare definitivă a declarațiilor universale. Popper a negat că ar fi imaginat o astfel de teorie naivă a falsificării: ”Ceea ce aș dori să contest este faptul că aș fi elaborat vreodată o asemenea teorie naivă a falsificării. De la bun început am subliniat în lucrarea mea publicată în 1933, și mai ales în cea din 1934, că poți să te sustragi oricărei infirmări, dar că este foarte important să încerci să-ți clădești teoria în așa fel încât ea să poată fi infirmată.... niciodată nu trebuie să se renunțe la o teorie. Se mai poate încerca - și asta am spus-o explicit în Logica cercetării - salvarea teoriei cu ajutorul unor ipoteze ajutătoare sau cu alte mijloace.” (K. Popper and Lorenz 1985, 51) Pentru a depăși aceste probleme, Popper a imaginat că știința progresează prin respingerea succesivă a teoriilor falsificate păstrând pe cele cu putere explicativă mai mare, mai degrabă decât prin declarații falsificate. A doua formă de falsificare luată în considerare a fost cea metodologică. Falsificatorul face aceleași ipoteze de bază ca în falsificarea dogmatică, dar se numesc tentative, un set de presupuse cu ajutorul cărora se încearcă falsificarea teoriilor. Falsificarea metodologică susține deciziile riscante. Deși alegerea pe care o facem ar putea fi greșită, falsificatorul metodologic vede acest lucru ca pe o problemă a celor mai puține două rele. Paul Newall afirmă că este greu de criticat falsificaționismul metodologic pentru simplul motiv că este nefalsificabil. (Newall 2005) Pentru a reduce convenționalismul din falsificarea metodologică, Popper a conceput o versiune sofisticată a falsificării pe baza unor condiții impuse, respectiv noua teorie să aibă un conținut empiric în exces, să explice tot ce a fost explicat anterior și unele dintre aceste predicții noi să fi fost confirmate prin experiment. 3 Falsificabilitatea O teorie nu trebuie respinsă ca fiind falsă până când nu se dezvoltă una mai bună. Aceasta ne conduce la o noțiune de creștere sau dezvoltare a teoriilor în locul falsificării dogmatice care le acceptă sau le respinge în cazuri singulare, și niciun experiment nu poate fi crucial, dacă nu este interpretat ca atare după eveniment în lumina unei noi teorii pentru care oferă coroborarea. În cele din urmă, ideea de proliferare a teoriilor 14 (pluralism) este importantă pentru falsificarea sofisticată, spre deosebire de varianta dogmatică. (Newall 2005) Conflictul în știință nu este astfel între teorii și experimente, ci întotdeauna între teorii rivale. ”în timp ce eu nu pretind științei nici un fel de certitudine finală prin urmare nici nu o obțin, convenționalistul caută în știință „un sistem de cunoaștere bazat pe temeiuri-ultime“, pentru a folosi o frază a lui Dingler” (K. Popper 1985, 152) ”în timp ce eu nu pretind științei nici un fel de certitudine finală prin urmare nici nu o obțin, convenționalistul caută în știință „un sistem de cunoaștere bazat pe temeiuri-ultime“, pentru a folosi o frază a lui Dingler” (K. Popper 1985, 152) Lakatos afirmă că falsificaționismul sofisticat se deosebește de cel naiv atât în regulile de acceptare (sau "criteriul de delimitare"), cât și în regulile sale de falsificare sau eliminare. Nu există nici o falsificare înainte de apariția unei teorii mai bune. Dar atunci caracterul distinctiv negativ al falsificării naive dispare; critica devine mai dificilă și, de asemenea, pozitivă, constructivă. (Imre Lakatos 1970) 4 Falsificare și respingere Falsificarea dogmatică se referă la declarațiile științifice considerate în mod individual Acestea pot fi falsificate și, în funcție de rezultat, teoriile științifice pot fi considerate falsificabile și acceptate, respinse, sau pot fi menținute prin adăugarea de ipoteze ad-hoc. și acceptate, respinse, sau pot fi menținute prin adăugarea de ipoteze ad-hoc. ”Există o metodă semnificativă de a evita sau de a elimina respingerile: este metoda ipotezelor auxiliare sau ipotezelor ad hoc . Dacă vreuna din conjecturile noastre nu funcționează atunci trebuie să schimbăm teoria. Există însă în principal două genuri de schimbări, conservatoare și revoluționare. Iar în cadrul schimbărilor cele mai conservatoare, mai există încă două: ipoteze ad hoc și ipoteze auxiliare... Numesc o conjectură "ad hoc" dacă este introdusă pentru a explica o anumită dificultate, dar care nu poate fi testată în mod independent.” (K. Popper 1985, 136) ”Există o metodă semnificativă de a evita sau de a elimina respingerile: este metoda ipotezelor auxiliare sau ipotezelor ad hoc . Dacă vreuna din conjecturile noastre nu funcționează atunci trebuie să schimbăm teoria. Există însă în principal două genuri de schimbări, conservatoare și revoluționare. Iar în cadrul schimbărilor cele mai conservatoare, mai există încă două: ipoteze ad hoc și ipoteze auxiliare... Numesc o conjectură "ad hoc" dacă este introdusă pentru a explica o anumită dificultate, dar care nu poate fi testată în mod independent.” (K. Popper 1985, 136) ”să nu dăm un edict prea sever împotriva ipotezelor ad hoc, ele pot deveni până la urmă testabile, lucru care se poate la fel de bine întâmpla și cu o ipoteză metafizică. În general însă, criteriul nostru de testabilitate ne semnalează ipotezele ad hoc,... Ipotezele ad hoc - adică ipotezele auxiliare netestabile la momentul respectiv - pot salva aproape orice teorie de orice respingere specifică. Aceasta nu înseamnă Însă că putem să folosim o ipoteză ad hoc oricât de mult dorim. Ea poate deveni testabilă, iar un test negativ ne poate obliga fie să renunțăm la ea, fie să introducem o nouă ipoteză secundară ad hoc , și tot așa ad infinitum . Acesta este de fapt un lucru pe care îl evităm aproape întotdeauna. Lakatos 1970) Problema falsificării sofisticate este însă tocmai multitudinea teoriilor luate în considerare. În cazul a două teorii incompatibile, trebuie să ne întoarcem la aspectele convenționale ale falsificării metodologice sau la ipotezele incontestabile ale falsificării dogmatice pentru a face o alegere. Apelarea la fapte noi coroborate presupune o delimitare clară între termeni observaționali și teoretici, cu decizii convenționale cu privire la ceea ce constituie cunoștințe "de bază". Pentru falsificarea naivă știința se dezvoltă prin respingerea experimentală succesivă a teoriilor; proliferarea constantă a teoriilor este opțională, dar nu obligatorie. Pentru falsificarea sofisticată, proliferarea teoriilor nu poate aștepta până când teoriile acceptate sunt "respinse". În timp ce falsificaționismul naiv accentuează "urgența înlocuirii unei ipoteze falsificate cu una mai bună", falsificarea sofisticată subliniază urgența înlocuirii oricărei ipoteze cu cea mai bună. (Imre Lakatos 1970) Popper afirmă că Popper afirmă că ”Sistemul de enunțuri de bază. în sensul în care folosesc eu termenul, include mai degrabă toate enunțurile singulare auto-consistente care au o anumită formă logică — toate 15 enunțurile singulare factuale care pot t concepute ca atare. Astfel, sistemul enunțurilor de bază va conține multe enunțuri reciproc incompatibile.” (K. Popper 1985, 156) ”O teorie va fi numită „empirică“ sau ..falsificabilă“ dacă împarte în mod ambiguu mulțimea tuturor enunțurilor de bază posibile în următoarele două subclase nevide. În primul rând, mulțimea tuturor acelor enunțuri de bază cu care aceasta este inconsistentă (sau pe care îl elimină sau interzice): numim această mulțime, mulțimea falsificatorilor potențiali ai teoriei, iar în al doilea rând mulțimea acelor enunțuri elementare pe care nu le contrazice (sau pe care le „permite“). Putem spune mai concis că o teorie este falsificabilă dacă mulțimea falsificatorilor ei potențiali nu este vidă.” (K. Popper 1985, 157) 4 Falsificare și respingere (spun "aproape", pentru că regulile metodologice nu sunt rigide) Mai mult, posibilitatea de realiza multe lucruri folosind ipoteze ad hoc nu trebuie exagerată, există multe respingeri care nu pot fi evitate în acest fel, chiar dacă se poate recurge întotdeauna la o tactică de imuni zare ca și ignorarea respingerii.” (K. Popper 1985, 137) Popper admite că, în practică, un singur contra-exemplu nu este suficient pentru a falsifica Popper admite că, în practică, un singur contra-exemplu nu este suficient pentru a falsifica o teorie; de aceea teoriile științifice sunt reținute în multe cazuri, în ciuda dovezilor anormale. (Mitra 2016) (Mitra 2016) 16 ”Spunem că o teorie este falsificată numai atunci când am acceptat enunțuri care o contrazic... dacă teoria este contrazisă de enunțuri de bază răzlețe, nu o vom considera încă, din această cauză, ca fiind falsificată . Aceasta o facem numai atunci când a fost găsit un efect reproductibil care falsifică teoria ; cu alte cuvinte, dacă a fost formulată și coroborată o ipoteză empirică, de un nivel mai scăzut de generalitate, care descrie un asemenea efect.” (K. Popper 1934, 117) ”Spunem că o teorie este falsificată numai atunci când am acceptat enunțuri care o contrazic... dacă teoria este contrazisă de enunțuri de bază răzlețe, nu o vom considera încă, din această cauză, ca fiind falsificată . Aceasta o facem numai atunci când a fost găsit un efect reproductibil care falsifică teoria ; cu alte cuvinte, dacă a fost formulată și coroborată o ipoteză empirică, de un nivel mai scăzut de generalitate, care descrie un asemenea efect.” (K. Popper 1934, 117) O teorie științifică, conform lui Popper, poate fi salvată în mod legitim de falsificare prin introducerea unei ipoteze auxiliare care să permită generarea de predicții noi, falsificabile. De asemenea, dacă există suspiciuni de părtinire sau eroare, cercetătorii ar putea introduce o ipoteză falsificabilă auxiliară, care să permită testarea. Dar această tehnică nu poate rezolva problema în general, deoarece orice ipoteză auxiliară poate fi contestată în același mod, ad infinitum. Pentru a rezolva această regresiune, Popper introduce ideea unei declarații de bază, o afirmație empirică care poate fi folosită atât pentru a determina dacă o teorie dată este falsificabilă și, dacă este cazul, pentru a corobora ipotezele de falsificare. Declarațiile de bază trebuie să fie atât singulare, cât și existențiale (cerința formală) și să fie testabile prin observarea inter-subiectivă (cerința materială). 4 Pentru compararea teoriilor în termeni de verosimilitudine, Popper a oferit două metode de comparare a teoriilor, definiții calitative și cantitative. ”Acceptând că atât conţinutul de adevăr, cât şi conţinutul de falsitate a două teorii t1 şi t2 sunt comparabile, putem spune că t2 este în mai mare măsură asemănătoare adevărului, sau corespunde mai bine faptelor decât t1 dacă şi numai dacă fie 1)conţinutul de adevăr al lui t2 însă nu şi conţinutul lui de falsitate, depăşeşte conţinutul de adevăr al lui t1, 2) conţinutul de falsitate al lui tl dar nu şi conţinutul său de adevăr, depăşeşte conţinutul de falsitate al lui t2” (K. Popper 1985, 202) 5 1) t2 face aserţiuni mai precise decât t1 iar aceste aserţiuni mai precise 'rezistă unor teste mai precise; 2) t2 descrie şi explică mai multe fapte decât tI (care va include spre exemplu cazul următor că, dat fiind că ordinea lucrurilor este aceeaşi, aserţiunile lui t2 sunt mai precise)'; 3) t2 descrie sau explică faptele mai detaliat decât o face t1; 4) t2 a trecut teste la care t1 a căzut; 5) t2 a sugerat noi teste experimentale, care nu au fost luate în consideraţie înainte de construirea lui t2 (şi care nu au fost sugerate de t1 ' şi probabil că nici nu sunt aplicabile lui t1' iar t2 a trecut aceste teste; 6) t2 a unificat sau conectat diferite probleme necorelate până atunci.” (K. Popper 1985, 201) 4 Falsificare și respingere 18 Deși Popper respinge consecvent ideea că teoriile științifice nefalsificate și bine coroborate, cu conținuturi ridicate de conținut informativ, pot fi considerate adevărate sau susceptibile de a fi adevărate, conceptul lui de verosimilitudine explorează ideea că astfel de teorii sunt mai aproape de adevăr decât teoriile falsificate pe care le-au înlocuit: ”ideea de verosimilitudine este cea mai importantă în cazurile în care știm că avem de lucru cu teorii ce sunt în cel mai bun caz aproximări - ceea ce înseamnă teorii despre care știm de fapt că nu pot fi adevărate.” (K. Popper 1985, 204) În răspunsul său (K. Popper 1974, sec. 12.iii) către Lakatos (1974), Popper a insistat asupra unei distincții clare între falsificare și respingere. Scopul testării empirice nu este doar să falsifice ipotezele individuale, ci și să discrimineze ipoteze concurente; în special, să falsifice hotărârea implicită că sunt la fel de apropiate de adevăr. El a subliniat în repetate rânduri că judecățile de verosimilitudine sunt inevitabil comparative. Dar ceea ce nu a făcut în mod explicit a fost să conecteze verosimilitudinea sau să aproximeze adevărul cu rezultatele (atât pozitive cât și negative) ale testelor empirice: În răspunsul său (K. Popper 1974, sec. 12.iii) către Lakatos (1974), Popper a insistat asupra unei distincții clare între falsificare și respingere. Scopul testării empirice nu este doar să falsifice ipotezele individuale, ci și să discrimineze ipoteze concurente; în special, să falsifice hotărârea implicită că sunt la fel de apropiate de adevăr. El a subliniat în repetate rânduri că judecățile de verosimilitudine sunt inevitabil comparative. Dar ceea ce nu a făcut în mod explicit a fost să conecteze verosimilitudinea sau să aproximeze adevărul cu rezultatele (atât pozitive cât și negative) ale testelor empirice: ”Nu sugerez că introducerea explicită a ideii de verosimilitudine va duce la orice schimbare în teoria metodei... teoria mea de testare sau de coroborare prin teste empirice este echivalentul metodei corespunzătoare acestei noi idei metodologice. Singura îmbunătățire este una de clarificare.” (K. Popper 1963, 10, § xii) ”Nu sugerez că introducerea explicită a ideii de verosimilitudine va duce la orice schimbare în teoria metodei... teoria mea de testare sau de coroborare prin teste empirice este echivalentul metodei corespunzătoare acestei noi idei metodologice. Singura îmbunătățire este una de clarificare.” (K. 4 Falsificare și respingere El subliniază faptul că cunoștințele de bază nu sunt cunoștințe în sensul stabilirii concludente; ele pot fi contestate în orice moment, mai ales dacă se suspectează că acceptarea lor necritică poate fi responsabilă de dificultățile întâmpinate ulterior. Pentru a evita regresul, Popper apelează la rolul jucat de convenție și ceea ce el numește "relativitatea declarațiilor de bază": Fiecare încercare a unei teorii, fie că are ca rezultat coroborarea sau falsificarea, trebuie să se oprească la o declarație de bază sau la alta pe care noi decidem să o acceptăm. Dacă nu ajungem la nicio decizie și nu acceptăm nicio declarație de bază, atunci testul nu va conduce nicăieri ... Această procedură nu are un scop natural. Astfel, dacă testul este de a ne conduce oriunde, nu rămâne nimic altceva decât să ne oprim la un moment dat și să spunem că, pentru moment, suntem mulțumiți. (K. Popper 1934) În acest caz, Popper consideră că este necesar consensul comunității științifice relevante. 17 Popper respinge ideea selectării unei teorii mai probabile din două teorii la fel de puternice. El consideră că teoriile cu un conținut informativ înalt sunt de preferat posedă o putere predictivă ridicată și sunt mai testabile. Dar, poate părea paradoxal, Popper susține că cu cât o teorie este mai puțin probabilă, cu atât este mai bună științific, fiind mai falsificabilă. În Conjecturi și infirmări (K. Popper 1963) Popper a integrat conceptele de adevăr și de conținut pentru a construi conceptul metalogic de "similitudine a adevărului" sau "verosimilitudine". El a explicat acest concept prin referire la consecințele logice ale teoriilor, "conținutul de adevăr" și "conținutul de falsitate". 4 În Filosofie socială și filosofia științei (K. Popper 1985, 201) Popper propune o listă de ”șase tipuri de cazuri în care trebuie să fim înclinați să spunem despre o teorie t1 că este înlocuită de t2 în sensul că t2 pare - după câte știm - să corespundă, într-un sens sau în altul, mai bine faptelor decât t1” 5 Verosimilitudinea a permis lui Popper să atenueze ceea ce mulți au văzut ca pesimismul unei filozofii anti-inductiviste a științei, (Thornton 2017) care a susținut că majoritatea, dacă nu toate, teoriile științifice sunt false. Astfel, progresul științific ar putea fi acum reprezentat ca progres către adevăr, iar coroborarea experimentală ar putea fi văzută ca un indicator al verosimilitudinii. 4 Falsificare și respingere Popper 1963, 10, § xii) În contextul dezbaterii asupra mecanicii cuantice, Popper susține că probabilitățile sunt pretenții obiective referitoare la lumea exterioară independentă de minte, și propune teoria propensității ca o variantă a teoriilor de frecvență relativă a probabilităților apărate de pozitiviștii logici precum Richard von Mises și Hans Reichenbach. Popper susține că teoriile logice sau subiective interpretează incorect afirmațiile științifice despre probabilitate, propunând ca probabilitățile să fie tratate ca tendințe ale setărilor experimentale pentru a produce anumite rezultate, mai degrabă decât ca fiind derivate din clasa de referință a rezultatelor obținute prin rularea acestor experimente. (Shea 2017) Rezultatele experimentelor sunt importante deoarece ne 19 permit să testam ipoteze privind valorile anumitor probabilități; totuși, rezultatele nu sunt ele însele parte a probabilității în sine. permit să testam ipoteze privind valorile anumitor probabilități; totuși, rezultatele nu sunt ele însele parte a probabilității în sine. Poziția finală a lui Popper este de recunoaștere a imposibilității discriminării științei de non-știință pe baza falsificabilității afirmațiilor științifice; el recunoaște că teoriile științifice sunt predictive și, în consecință, prohibitive, numai atunci când sunt luate împreună cu ipotezele auxiliare, și că reajustarea sau modificarea acestora constituie o parte integrantă a practicii științifice. (Thornton 2017) 5 Extinderea falsificabilității Criteriul de delimitare al lui Popper se referă la structura logică a teoriilor. Imre Lakatos a afirmat că dacă o teorie este științifică sau ne-științifică se poate determina independent de fapte. (Imre Lakatos 1973, 117). El a propus o modificare a criteriului lui Popper pe care el l-a numit "falsificaționism sofisticat (metodologic)", unde criteriul de delimitare nu ar trebui să se aplice unei ipoteze sau unei teorii izolate, ci unui întreg program de cercetare. La Lakatos apare un nucleu dur al tezelor centrale care sunt considerate rezistente, constituind un program de cercetare. Astfel, un program de cercetare este progresiv dacă noile teorii fac predicții surprinzătoare care sunt confirmate. Progresul în știință este posibil numai dacă fiecare teorie nouă dezvoltată în program are un conținut empiric mai mare decât predecesorul său. Altfel, programul este pseudoștiințific. Știința bună este progresivă, știința proastă este degenerativă, iar dacă un program de cercetare nu prezice nimic nou sau presupune predicții noi care nu se întâmplă niciodată, atunci s-ar fi putut ajunge la o asemenea degenerare transformându-se într-o pseudoștiință. 20 Metodologia lui Lakatos a fost văzută, ca o încercare de a reconcilia falsificarea lui Popper cu paradigmele lui Thomas Kuhn. Lakatos a propus o cale de mijloc, în care abordările socio- psihologice ale lui Kuhn au fost înlocuite de cele logico-metodologice. Paul Thagard consideră că o teorie este pseudoștiințifică dacă nu reușește să progreseze și "comunitatea practicienilor face puține încercări de a dezvolta teoria către soluții ale problemelor, nu arată îngrijorarea încercărilor de evaluare a teoriei față de ceilalți și este selectivă în considerarea confirmărilor și a dezacordurilor." (Thagard 1978, 228). El a propus un alt set de principii pentru a încerca să depășească aceste dificultăți și consideră că este important pentru societate să găsească o modalitate de a face acest lucru. 6 Thagard precizează că uneori teoriile vor petrece ceva timp doar ca "nepromițătoare" înainte de a merita cu adevărat titlul de pseudoștiință. Daniel Rothbart (Rothbart 1990) stabilește criterii de eligibilitate conform cărora teoria trebuie să cuprindă succesul explicativ al rivalului său și să obțină implicații testabile care sunt incompatibile cu cele ale rivalului. George Reisch a propus ca demarcarea să se bazeze integrarea teoriei în mod corespunzător în celelalte științe. În general, conform lui Reisch, un domeniu epistemic este pseudoștiințific dacă nu poate fi încorporat în rețeaua existentă de științe stabilite (Reisch 1998). Sociologul Robert K. 6 Conform metodei lui Thagard, o teorie nu este științifică dacă îndeplinește două condiții: 1) Teoria a fost mai puțin progresivă decât teoriile alternative pe o perioadă lungă de timp și se confruntă cu multe probleme nerezolvate; și... 2) Comunitatea practicienilor face puține încercări de a dezvolta teoria față de soluțiile problemelor, nu arată îngrijorarea încercărilor de evaluare a teoriei față de ceilalți și este selectivă în privința confirmărilor și a dezacordurilor. țin progresivă decât teoriile alternative pe o perioadă lungă de timp și se confruntă cu multe problem vate; și... 2) Comunitatea practicienilor face puține încercări de a dezvolta teoria față de soluțiile 6 Conform metodei lui Thagard, o teorie nu este științifică dacă îndeplinește două condiții: 1) Teoria a fost problemelor, nu arată îngrijorarea încercărilor de evaluare a teoriei față de ceilalți și este selectiv confirmărilor și a dezacordurilor. 5 Extinderea falsificabilității Merton (Merton 1973) a propus criteriile de demarcare pe baza valorii științei, caracterizată printr-un spirit care poate fi rezumat ca patru seturi de imperative instituționale: universalismul (afirmațiile trebuie supuse unor criterii impersonale prestabilite), comunismul (descoperirile sunt produse de colaborare socială), dezinteresul (control instituțional 21 pentru a reduce efectele motivelor personale sau ideologice ), și scepticismul organizat (examinarea detașată a credințelor). pentru a reduce efectele motivelor personale sau ideologice ), și scepticismul organizat (examinarea detașată a credințelor). Mulți alți autori au propus criterii de demarcare pentru a identifica știința de pseudoștiință. De obicei acestea includ credința în autoritate, experimentele nerepetabile, exemple selectate, ne- dorința de a testa, nerespectarea informațiilor de respingere, subterfugii încorporate, explicațiile abandonate fără înlocuire. (Hansson 2017) Într-un pasaj notoriu, Popper a sugerat că "[deși] gradul de coroborare a unei teorii. . . nu poate fi interpretată doar ca măsură a verosimilitudinii sale,. . . ea poate fi luată ca o indicație a felului în care apare verosimilitudinea ei în acel moment " (Karl R. Popper 1979, chaps. 2, § 33). La urma urmei, afirmă Miller (Miller 2009b), gradul de coroborare a oricărei teorii falsificate nu dă nicio indicație, oricât de mică, fie despre verosimilitudinea teoriei, fie despre cât de bine aproximează Adevărul. Popper s-a gândit că o ipoteză care a eșuat în unele teste, dar nu a eșuat foarte rău, va apărea o ipoteză cu unele predicții testate cu certitudine dincolo de limitele erorilor experimentale, dar nu greșit a fi mai aproape de adevăr decât de un rival radical eșuat, chiar dacă ambele sunt falsificate. Dar lipsa unei soluții la această dificultate nu este o scuză pentru o retragere în instrumentalism, inductivism sau iraționalism și nu ar trebui să ne împiedice să căutăm un răspuns mai modest la faptul incontestabil că "nu toate cazurile de falsificarea sunt aceleași." (Kvasz 2004, 263) 6 Critici ale falsificabilității Thomas Kuhn a criticat falsificabilitatea pentru că a caracterizat "întreaga întreprindere științifică în termeni care se aplică doar părților revoluționare ocazionale" (Kuhn 1970). Potrivit lui Kuhn, modul în care știința funcționează în astfel de ocazii nu poate fi folosit pentru a caracteriza întreaga întreprindere științifică. În opinia lui Kuhn, un criteriu de delimitare trebuie să 22 se refere la funcționarea științei normale (Kuhn 1970, 802). Dar Kuhn a ignorat falsificarea sofisticată a lui Popper și programul de cercetare pe care l-a inițiat. se refere la funcționarea științei normale (Kuhn 1970, 802). Dar Kuhn a ignorat falsificarea sofisticată a lui Popper și programul de cercetare pe care l-a inițiat. Kuhn obiectează împotriva întregii teorii a lui Popper și exclude orice posibilitate de reconstrucție rațională a dezvoltării științei. În opinia lui Kuhn, nu poate exista logică, ci doar psihologia descoperirii. Într-o comparație succintă a lui Hume, a lui Carnap și a lui Popper, Watkins subliniază că dezvoltarea științei este inductivă și irațională conform lui Hume, inductivă și irațională conform lui Carnap, neinductivă și rațională după Popper. (J. W. N. Watkins 1968) Extinzând această comparare, se poate adăuga că dezvoltarea științei este neinductivă și irațională conform lui Kuhn. Popper a criticat criteriul de demarcație al lui Kuhn, afirmând că criteriul lui Kuhn conduce la un "dezastru major ... înlocuirea unui criteriu rațional al științei cu cel sociologic.". (Hansson 2017) Stephen Toulmin afirmă că T.S. Kuhn a expus practic trei teorii succesive ale schimbării științifice, depărtându-se de teoria originală a "revoluțiilor științifice". Kuhn interpretează contrastul între schimbarea "normală" și "revoluționară" în două moduri alternative: uneori ca o analiză filosofică, uneori ca o ipoteză sociologică, astfel încât "paradigma" este în mod corespunzător ambiguă. (Toulmin 1967) Paul Feyerabend, au susținut că nu este nici posibil, nici de dorit, o distincție între știință și non-știință. (Feyerabend 2010) El a respins și argumentul lui Lakatos pentru ipotezele ad-hoc, argumentând că știința nu ar fi progresat fără a folosi toate metodele disponibile pentru a susține noile teorii. Pentru Feyerabend, un statut special al științei poate deriva doar din valoarea socială și fizică a rezultatelor sale, și nu din metoda sa. 23 Imre Lakatos afirmă că, prin falsificabilitate, Popper a realizat o deconectare între jocul științei (falsificabilitatea) și scopul științei (dezvoltarea unor teorii adevărate). 7 ”un principiu inductiv care corelează metafizica realistă cu aprecieri metodologice, verosimilitudinea cu coroborarea, care reinterpretează regulile "jocului științific" ca o teorie conjecturală despre semnele creșterii cunoașterii, adică despre semnele verosimilitudinii crescânde a teoriile noastre științifice.” (Champion 1985, 156) 6 Critici ale falsificabilității Pentru a restabili legătura dintre joc și scopul său, Lakatos afirmă că Popper a introdus în joc o "adiere de "inductivism"." 7 De reținut că această critică a lui Lakatos împotriva lui Popper, este la fel de valabilă și împotriva lui însuși. (Musgrave and Pigden 2016) Falsificabilitatea lui Popper a fost criticată atât pentru excluderea științei legitime (Hansson 2006), cât și pentru acordarea statutului științific unor pseudo-științe (Agassi 1991) (Mahner 2007, 518–19). Conform lui Larry Laudan (Laudan 1983, 121), "are consecința nefericită de a considera drept "științific" orice revendicare care face afirmații false concludente". W. W. Bartley în 1978 a susținut, (Bartley 1976) că Popper a distrus dialogul. ”Distanța dintre modul de a face filosofie al lui Popper și cea a majorității filozofilor profesioniști contemporani este la fel de mare ca și cea dintre astronomie și astrologie". Rafe Champion afirmă (Champion 1985) că ”teoria sa a cunoștințelor conjecturale nici măcar nu pretinde că oferă fundații pozitive justificate.” Putnam (Putnam, Gasper, and Trout 1974) susține că acceptarea inițială pe scară largă a mecanicii newtonieni a avut puțin sau nimic de-a face cu predicții falsificabile, deoarece oamenii de știință au acceptat mai degrabă succesul teoriei în explicarea fenomenelor stabilite anterior. Hacking (Hacking 1983) afirmă că multe aspecte ale practicii științifice, inclusiv experimente, nu pot fi interpretate ca încercări de falsificare sau coroborare. 24 Fizicienii Alan Sokal și Jean Bricmont au criticat falsificabilitatea că nu descrie cu exactitate felul în care funcționează știința, (Sokal and Bricmont 1999) și că falsificabilitatea nu poate face distincția între astrologie și astronomie. Unii economiști, precum cei ai școlii austriece, consideră că macroeconomia este nefalsificabilă din punct de vedere empiric. (Heath 2015) (von Mises 2014) De asemenea, mulți filozofi afirmă că matematica nu este falsificabilă experimental, și deci nu e o știință conform definiției lui Karl Popper. (Shasha and Lazere 1998) Unii critică la Popper faptul că teoria lui nu oferă o alternativă legitimă față de propunerile inductiviste pe care le critică. Jeffrey (Jeffrey 1975) susține că bayesianismul, cu accentul pus pe gradul în care dovezile empirice susțin o ipoteză, este mult mai apropiat de practica științifică decât falsificabilitatea lui Popper. Una din marile provocări ale falsificabilității este teza Duhem-Quine. 6 Critici ale falsificabilității (Quine 1953) Referitor la prima lege de mișcare a lui Newton, a inerției, se presupune că un corp nu se găsește nici în repaus, nici în mișcare uniformă într-o linie dreaptă și, aparent, nu este acționat de o forță externă. Această observație pare să respingă legea lui Newton, dar în realitate acest lucru nu este neapărat adevărat. Newton însuși a constat că, de ex., orbitele eliptice ale planetelor, sunt acționate de forțele gravitaționale. ”Fizicianul nu poate supune niciodată o ipoteză izolată testului experimental, ci doar un întreg grup de ipoteze; atunci când experimentul este în dezacord cu previziunile sale înseamnă că cel puțin una dintre ipotezele care constituie acest grup este inacceptabilă și ar trebui să fie modificată, dar experimentul nu desemnează care dintre ele ar trebui schimbată". (Ariew 2014) Deoarece prima lege este folosită împreună cu multe presupuneri, nu este posibilă respingerea legii în cazul în care ceea ce legea prevede nu este realizat, deoarece presupunerile sau ipotezele suplimentare ar putea fi de vină. Prin urmare, prima lege a lui Newton 25 este nefalsificabilă. (Mitra 2016) Popper a răspuns la problema menționată mai sus, utilizând un model de trei niveluri de tipuri de declarații împărțite pe baza falsificabilității și confirmării lor. D.C. Stove (Stove 1978) consideră că teoria lui Popper despre declarațiile științifice (non- statistice) și neștiințifice sunt simple negări sau simple afirmații ale existenței sau existenței locale. Ceea ce Stove consideră "simple afirmații ale existenței locale", Popper le numește "afirmații ale existențiale singulare". Practic, spune el, "criteriul de falsificabilitate" al lui Popper cere ca o declarație, pentru a fi empirică, să fie inconsistentă cu o anumită declarație de bază în sensul precizat de Stove. Iar o ”simplă negare a existenței locale", Popper o numește "afirmație de non- existență singulară", care, atunci când este empirice, este "declarație instanțială". Conform lui Stove, o lege (non-statistică) sau teorie a științei empirice poate fi inconsecventă cu o alta, dar filozofia lui Popper despre declarații științifice este incompatibilă cu acest fapt evident, pentru că Popper identifică legi sau teorii cu simple negări ale existenței, și simpla negare a existenței nu poate fi inconsecventă cu o alta. Mai mult, Stove afirmă că criteriul de falsificabilitate al lui Popper exclude din știința empirică toate acele afirmații care, conform lui Popper însuși, constituie baza de observație a științei. A. 6 Critici ale falsificabilității O'Hear consideră că epistemologia lui Popper duce la un scepticism inacceptabil, neputând evita un angajament față de procedurile inductive. Similar, W. Salmon (Salmon and Hitchcock 2017) susține că ideea lui Popper de coroborare a teoriei presupune referirea la procedurile inductive. Mulți alți cercetători, precum Miller, Tichý și Grünbaum, au argumentat defectele definițiilor oficiale ale teoriei lui Popper. Astfel, se consideră că verosimilitudinea este în mare măsură importantă în sistemul lui Popper datorită aplicării sale la teorii despre care se știe că sunt false. În acest sens, Popper scrisese: 26 ”În cele din urmă, ideea de verosimilitudine este cea mai importantă în cazurile în care știm că trebuie să lucrăm cu teorii care sunt cel mai bine aproximative - adică teorii despre care știm că ele nu pot fi adevărate. (Acesta este adesea cazul în științele sociale). În aceste cazuri, putem vorbi despre aproximări mai bune sau mai rele cu adevărul (și, prin urmare, nu avem nevoie să interpretăm aceste cazuri într-un sens instrumentalist).” (K. Popper 1963, 235) Deficiențele descoperite de critici în definițiile formale ale lui Popper au fost văzute ca fiind importante, pentru că au fost legate de nivelurile de veridicitate ale teoriilor false. În 1974, Miller și Tichý au demonstrat că condițiile de verosimilitudine specificate de Popper pentru compararea conținutului teoriilor cu conținut de adevăr și falsitate pot fi satisfăcute numai atunci când teoriile sunt adevărate. În cazul crucial important al teoriilor false, definițiile lui Popper sunt oficial defecte. (Thornton 2017) Prin urmare, condițiile lui Popper de a compara nivele de verosimilitate nu pot fi îndeplinite niciodată. După eșecul definițiilor lui Popper în 1974, unii critici au ajuns să considere că întregul edificiu al falsificabilității a fost subminat. Popper a recunoscut deficiențele ("eroarea mea principală a fost că nu am văzut imediat că ... dacă conținutul unei declarații false a depășește valoarea unei afirmații b, atunci conținutul de adevăr al lui a depășește conținutul adevărului b, și aceleași aspecte ale conținutului lor de falsitate ", (Karl R. Popper 1979, 371) dar a susținut că "cred că nu ar trebui să concluzionăm din eșecul încercărilor mele de a rezolva problema [de definire a verosimilitudinii] că problema nu poate fi rezolvată." (Karl R. Popper 1979, 372) El a mutat sarcina de a defini în mod formal conceptul din stadiul central în filosofia sa a științei, afirmând că nu a intenționat niciodată să sublinieze că "gradele de verosimilitudine ... 6 Critici ale falsificabilității pot fi determinate valoric, cu excepția anumitor cazuri limitate", (Karl R. Popper 1979, 59) și susținând în schimb că valoarea principală a conceptului este euristică și intuitivă, în care absența unei definiții formale adecvate nu este un impediment insurmontabil al utilizării sale în evaluarea reală a teoriilor relativizate la probleme în care avem un interes. Efortul acestei strategii pare să reflecte 27 cu adevărat semnificația conceptului de verosimilitudine în sistemul lui Popper, dar nu a satisfăcut pe toți criticii lui. (Thornton 2017) 7 Susțineri ale falsificabilității Suporterii lui Popper au susținut că cele mai multe critici se bazează pe o interpretare neînțeleasă a ideilor sale. Ei afirmă că Popper nu ar trebui interpretat în sensul că falsificabilitatea este o condiție suficientă pentru delimitarea științei. Unele pasaje par să sugereze că el o consideră doar o condiție necesară (Feleppa 1990, 142). Alte pasaje ar sugera că, pentru ca o teorie să fie științifică, Popper impune (pe lângă falsificabilitate) și alte teste, și că rezultatele negative ale testelor sunt acceptate (Cioffi 1985, 14–16). Un criteriu de delimitare bazat pe falsificare care include aceste elemente va evita cele mai evidente contra-argumente ale unui criteriu bazat numai pe falsificabilitate. (Hansson 2017) David Miller consideră că problema delimitării și a problema inducției la Popper sunt încă uneori ”lamentabil de greșit înțelese... Problema demarcării este rezolvată așa cum a rezolvat-o Popper.” (Miller 2009b) Mulți critici înțeleg în mod greșit filozofia lui Popper în problema demarcării. Sarcina sa nu este de a "distinge problemele științifice și ne-științifice într-un mod care să prezinte un mandat epistemic mai sigur sau un temei doveditor pentru știință decât pentru neștiință", conform lui Laudan (Laudan 1983, 118), și nici "explicarea uzanțelor paradigmatice ale "științificului" (Laudan 1983, 122). Problema nu ține de siguranță, mandat și motive pentru cei care trăiesc cu temerea că nu pot "avea dreptul să creadă nici o teorie științifică" (Papineau 2006, 63), și nici de utilizare, clasificare și statut. Și, contrar lui Grunbaum, (Grünbaum 1989) doar în mod accidental preocupată de ratificarea statutului neștiințific. Principala problemă a teoriei cunoașterii, cel puțin pentru un empirist, este destul de diferită în natură: Popper a descris principala problemă a teoriei cunoașterii ca "analiza critică a apelului la autoritatea experienței" 28 (K. Popper 1934, chap. 10). Filosofia lui Popper se opune în mod puternic și în mod expres tuturor (K. Popper 1934, chap. 10). Filosofia lui Popper se opune în mod puternic și în mod expres tuturor acestor tendințe la modă și tuturor viziunilor științei ca "un corp de cunoaștere": acestor tendințe la modă și tuturor viziunilor științei ca "un corp de cunoaștere": acestor tendințe la modă și tuturor viziunilor științei ca "un corp de cunoaștere": ”În contextul actual, nu prea contează dacă sunt sau nu drept în ceea ce privește incontestabilitatea oricăreia dintre aceste trei teorii [cele ale lui Freud, Adler și Marx]: aici ele servesc doar ca exemple, ca ilustrații. 7 Susțineri ale falsificabilității Scopul meu este de a arăta că "problema demarcării" mea a fost de la început problema practică de evaluare a teoriilor și de judecare a pretențiilor lor. Cu siguranță nu a fost o problemă de clasificare sau distincție a unor subiecte numite "știință" și "metafizică". A fost mai degrabă o problemă practică urgentă: în ce condiții este un apel critic la experiență posibil - unul care ar putea purta niște fructe?” (Karl Raimund Popper 1983, secs. I, § 18) Problema clară filosofică, și chiar logică, ce rezultă de aici, este: în ce circumstanțe o investigație empirică merită să se angajeze? cu soluția clară: ”din moment ce formularea unei ipoteze, acceptarea ei ca fiind un candidat pentru adevăr, trebuie să preceadă examinarea ei, sarcina unei investigații empirice nu poate fi să promoveze ipoteze, ci doar să le diminueze. Empirismul cere ca o ipoteză să fie păstrată dacă nu se ciocnește într-un mod adecvat cu experiența. Prin urmare, o ipoteză acceptată rămâne acceptată până când este respinsă. Nu sunt necesare alte acțiuni” (Miller 2009a, chaps. 4, § 1) Problema clară filosofică, și chiar logică, ce rezultă de aici, este: în ce circumstanțe o investigație empirică merită să se angajeze? cu soluția clară: ”din moment ce formularea unei Miller consideră că problema inducției este de asemenea rezolvată în modul în care a făcut- o Popper. Miller consideră că problema inducției este de asemenea rezolvată în modul în care a făcut- o Popper. Milos Taliga, în Against Watkins: From a Popperian point of view, (Taliga 2004), argumentează împotriva criticii lui John Watkins (John W. N. Watkins 1997) care susține că teoria verosimilitudinii lui Popper (împreună cu teoria sa de coroborare) introduce elemente justificative și inductiviste. Pe scurt, Watkins susține că afirmația lui Popper că "putem ști sau cel puțin avem motive să credem că facem progrese în ceea ce privește adevărul" (Karl Raimund Popper 1983, chap. § 16) este cu siguranță un element justificativ, și afirmația că "dacă două teorii concurente au fost criticate și testate, și gradul de coroborare al uneia dintre ele este mai mare decât a celeilalte, avem, în general, motive să credem că prima este o aproximare mai bună a adevărului decât a 29 doua" (Karl Raimund Popper 1983, 58) este suficientă pentru Watkins pentru a ajunge la concluzia că "evaluările de coroborare oferă o justificare pentru evaluările corespunzătoare de verosimilitudine... 7 Susțineri ale falsificabilității Se pare clar că un element inductiv a apărut aici" (John W. N. Watkins 1997, § 16-17). După ce prezintă în detaliu argumentația lui Watkins, Popper o analizează și ajunge în final la concluzia că esența criticii este ”întrebarea "De ce cea mai bună teorie coroborată este cea mai bună teorie?" Watkins crede că răspunsul lui Popper ar fi: "Pentru că este mai adevărată decât toate celelalte considerate și o putem cunoaște sau cel puțin avem motive serioase să credem în ea". Dar adevărul este că Popper a subliniat mereu că evaluarea noastră de verosimilitudine a ipotezelor concurente este doar o presupunere. El a insistat, de asemenea, că o putem apăra cu ajutorul coroborării-aprecierii și a altor motivații critice. Dar nu putem să o justificăm.” (Taliga 2004) Watkins transformă motivele critice ale lui Popper (oferite pentru a apăra, dar nu pentru a justifica) în motive pozitive (oferite pentru a justifica). Carl Hempel, în Empirical Statements and Falsifiability, (Hempel 1958) critică și el afirmațiile lui Watkins împotriva lui Popper. Watkins tinde să ascundă natura problemei la îndemână argumentând ca și cum criteriul de falsificabilitate al afirmațiilor empirice este el însuși o declarație calificabilă drept adevărată sau falsă, atribuindu-i lui Popper "o încercare de a-și falsifica criteriul falsificabilității științei". (John W. N. Watkins 1997, 122) Dar Popper, departe de a considera criteriul său de falsificabilitate drept o declarație falsificabilă, este foarte explicit în a- și caracteriza propunerea ca o "propunere pentru o stipulare", care trebuie judecată prin adecvarea ei la scopul său teoretic. 8 Tendința actuală După decăderea conceptului de falsificabilitate, s-a pus din nou problema metodei de demarcare între știință și pseudoștiință: să alegem cea mai probabilă teorie în scopuri educaționale, 30 cea mai bine coroborată teorie de dragul adevărului, sau cea mai informativă și mai explicativă teorie pentru a ne apropia cel mai mult de realitate? Este posibil ca cele mai informative și mai explicative teorii să nu fie cea mai riscantă estimare, sau să nu aibă cea mai mare probabilitate anterioară înainte de testare, deci nu ar oferi neapărat cea mai mare șansă de a învăța. O teorie care ar corespunde în oarecare măsură falsificabilității ar fi cea mai probabilă teorie, având și cea mai bună șansă de a învăța din propriile greșeli. (Derksen 1985) Dar este nevoie și de coroborare. Promovarea pluralismului metodologic ar fi o soluție, putând să ușureze alegerea dificilă între teoria cea mai probabilă și cea mai bună coroborată teorie, dar dacă se alege cea mai bine coroborată teorie se pune în pericol caracterul empiric și raționalitatea științei, iar prin alegerea celei mai probabile teorii riscăm să pierdem cea mai bună aproximare a adevărului . În 1978, Paul Thagard a propus ca pseudoștiința să se distingă de știință în primul rând prin faptul că ar fi mai puțin progresivă decât teoriile alternative pe o perioadă lungă de timp, iar susținătorii săi nu reușesc să recunoască sau să rezolve probleme cu ajutorul teoriei. (Thagard 1978) În 1983, Mario Bunge a sugerat categoriile de "domenii de credință" și "domenii de cercetare", pentru a ajuta la a distinge între pseudoștiință și știință, în cazul în care prima este în principal personală și subiectivă, și cea de a doua implică o anumită abordare sistematică. (Bunge 1982) Importanța delimitării pare să fi scăzut după ce Laudan (Laudan 1983) a argumentat că nu există nicio șansă de a găsi un criteriu necesar și suficient pentru ceva atât de eterogen ca metodologia științifică. Criteriile de delimitare ar fi fost folosite istoric ca "mașini de război" în disputele polemice dintre "oameni de știință" și "pseudo-oameni de știință". În opinia sa, delimitarea dintre știință și ne-știință a fost o pseudo-problemă, mai importantă ar fi concentrarea pe distincția dintre cunoștințele fiabile și cele nesigure. (Laudan 1983). Sebastian Lutz, în schimb, 31 susține că demarcarea nu trebuie să fie o singură condiție necesară și suficientă; trebuie să existe, mai degrabă, un criteriu necesar și un criteriu suficient posibil diferit. 8 Tendința actuală (Lutz 2011) Alți critici au susținut criterii multiple de delimitare, specifice pentru fiecare ramură principală a științei. Problema de demarcare a fost comparată de Michael LeVine cu problema diferențierii știrilor false de știrile reale, care au devenit proeminente în alegerile prezidențiale din 2016 ale Statelor Unite (LeVine 2016). Concluzii Deși extrem de criticat, contribuțiile lui Popper la filozofia științei sunt imense. Abordarea metodologiei sale științifice, bazată pe falsificabilitate, deși nu mai este acceptată pe scară largă, a jucat un rol vital în dezvoltarea filosofiei științei și a influențat toți filosofii care au abordat ulterior problema demarcației dintre știință și ne-știință, inclusiv pe Kuhn, Lakatos și Feyerabend. Popularitatea teoriei falsificabilității a avut un rol important în consolidarea imaginii științei ca o activitate empirică, inclusiv în justiție. Lucrările lui Popper din filosofia științei, inclusiv verosimilitudinea, mecanica cuantică, teoria probabilității și individualismul metodologic, continuă și astăzi să influențeze cercetătorii. Popper a făcut o distincție clară între logica falsificabilității și metodologia aplicată. Din punct de vedere logic, o lege poate fi falsificabilă, chiar dacă nu este verificabilă în mod concludent. Metodologic, nicio observație nu este scutită de posibilitatea erorii. Astfel, Popper admite în mod explicit faptul că, în practică, o singură situație conflictuală sau contra-exemplu nu este niciodată suficient metodologic pentru a falsifica o teorie, și că teoriile științifice sunt adesea păstrate chiar dacă o mare parte din dovezile disponibile sunt în conflict cu teoriile. (Thornton 2017) Modul în care este formulată o anumită teorie nu are nicio consecință în ceea ce privește filosofia științei. Popper subliniază că nu există o metodă unică pentru teoria științifică, fiind de 32 acord cu Einstein că "Nu există o cale logică care să conducă la [ legea universală a științei]. La ea se poate ajunge numai prin intuiție, bazată pe ceva de genul iubirii intelectuale a obiectelor de experiență". Popper nu a considerat că o teorie care nu este științifică nu este neapărat explicativă, și nici lipsită de sens; o teorie care nu este științifică (pentru că nu este înțeleasă) într-un anumit moment poate deveni falsificabilă și, astfel, științifică, odată cu dezvoltarea tehnologiei sau cu perfecționarea ulterioară a teoriei. (Thornton 2017) O teorie falsificată de observații poate fi revizuită, respinsă, sau menținută așa cum este prin schimbarea unor ipoteze auxiliare, cu condiția producerii de noi predicții falsificabile. Popper admite că oamenii de știință pot păstra teoriile chiar dacă predicțiilor eșuează, atunci când nu există o teorie mai bună din punct de vedere predictiv. În schimb, teoriile care sunt imunizate permanent împotriva falsificării prin introducerea unor ipoteze ad-hoc netestabile nu mai pot fi clasificate drept științifice. (Shea 2017) ad-hoc netestabile nu mai pot fi clasificate drept științifice. (Shea 2017) ”în fața respingerilor putem totdeauna adopta tactici evazive. Concluzii Din motive istorice, am numit inițial aceste tactici "stratageme convenționaliste (sau ocoluri), acum însă le voi numi "stratageme sau tactici de imunizare": întotdeauna trebuie să imunizăm o teorie împotriva respingerii. Există multe asemenea tactici evazive de imunizare, și dacă nu apare nimic mai bun, putem nega totdeauna obiectivitatea - sau chiar existența - observațiilor ce conduc la respingere.” (K. Popper 1985, 133) ”Cel care, în fața unor aparente respingeri, renunță cu prea multă ușurință la teoria sa nu va descoperi niciodată posibilitățile inerente teoriei sale. În știință există spațiu pentru discuție, pentru atac și, prin urmare, și pentru apărare. Numai dacă încercăm să apărăm teoriile putem descoperi diferitele posibilități inerente teoriei noastre. Știința este, ca întotdeauna, conjecturală. Trebuie să facem conjecturi legate de momentul în care ne vom opri din apărarea unei teorii favorite și vom încerca una nouă.” (K. Popper 1985, 134) În științele sociale, falsificabilitatea lui Popper rămâne un criteriu puternic, cu ajutorul Bibliografie primară Popper, Karl. 1934. Logica Cercetării. Popper, Karl. 1934. Logica Cercetării. ———. 1963. Conjecturi Și Infirmări. ———. 1974. “Replies to My Critics.” 1974. http://www.theopensociety.net/category/popper- karl/replies-to-my-critics/. . 1963. Conjecturi Și Infirmări. ———. 1974. “Replies to My Critics.” 1974. http://www.theopensociety.net/category/popper- karl/replies-to-my-critics/. ———. 1985. “Filosofie Socială Şi Filosofia Ştiinţei.” 1985. Ş Ş ţ http://www.edituratrei.ro/carte/karl-r-popper-filosofie-sociala-si-filosofia-stiintei/1085/. l d d 198 l h http://www.edituratrei.ro/carte/karl-r-popper-filosofie-sociala-si-filosofia http://www.edituratrei.ro/carte/karl-r-popper-filosofie-sociala-si-filosofia-stiintei/1085/. Popper, Karl, and Konrad Lorenz. 1985. Viitorul Este Deschis. Popper, Karl, and Konrad Lorenz. 1985. Viitorul Este Deschis. Popper, Karl R. 1979. Objective Knowledge: An Evolutionary Approach. Oxford University Press. Popper, Karl Raimund. 1983. Realism and the Aim of Science. Hutchinson. ipotezelor cu valoare adăugată. ipotezelor cu valoare adăugată. Deși mulți cercetători contemporani consideră că este dificil de comparat conținutului empiric, conținutul adevărului și conținutul de falsitate al diferitelor teorii, sunt de acord cu Popper că trebuie să existe o modalitate de a face acest lucru, folosind o abordare intuitivă. În acest sens, 33 A. A. Derksen (Derksen 1985) afirmă că nu este necesar ca cea mai bună teorie coroborată să fie cea mai falsificabilă în sensul testării mai severe. În opinia lui Popper, scăderea severității unui test explică de ce "un test repetat nu va mai fi considerat semnificativ sau ca sever: există ceva de genul legii scăderii randamentelor din testele repetate ". (K. Popper 1963) A. A. Derksen (Derksen 1985) afirmă că nu este necesar ca cea mai bună teorie coroborată să fie cea mai falsificabilă în sensul testării mai severe. În opinia lui Popper, scăderea severității unui test explică de ce "un test repetat nu va mai fi considerat semnificativ sau ca sever: există ceva de genul legii scăderii randamentelor din testele repetate ". (K. Popper 1963) Popper afirmă că "ceea ce încercăm în știință este să descriem (și pe cât posibil) să explicăm realitatea" (K. Popper 1974, 40). Deși Popper susține că realismul este o viziune metafizică incontestabilă despre natură, el crede totuși că avem motive bune pentru a accepta realismul și pentru a respinge ideile antirealiste, cum ar fi idealismul sau instrumentalismul. Teoriile științifice ar trebui interpretate ca încercări de a descrie o realitate independentă de minte. În ultimii ani însă teoria falsificabilității lui Popper a fost revitalizată. Susținătorii săi afirmă că conceptul poate fi clarificat prin alte mijloace decât criteriile necesare și suficiente (Pigliucci 2013) (Mahner 2013) sau că o astfel de definiție este într-adevăr posibilă, deși trebuie completată cu criterii specifice disciplinei pentru a deveni pe deplin operativ. (Hansson 1996) 34 Bibliografie secundară Agassi, Joseph. 1991. “Popper’s Demarcation of Science Refuted.” Methodology an Agassi, Joseph. 1991. “Popper’s Demarcation of Science Refuted.” Methodology and Science 24. Ariew, Roger. 2014. “Pierre Duhem.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Fall 2014. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/fall2014/entries/duhem/. B tl W W 1976 “III Bi l E l ti E i t l ” Phil hi 6 Ariew, Roger. 2014. “Pierre Duhem.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Fall 2014. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/fall2014/entries/duhem/. Bartley, W. W. 1976. “III: Biology – Evolutionary Epistemology.” Philosophia 6. Bunge, Mario. 1982. “Demarcating Science from Pseudoscience.” Fundamenta Sci Bunge, Mario. 1982. “Demarcating Science from Pseudoscience.” Fundamenta Scientiae 3. Champion, Rafe. 1985. “Agreeing to Disagree: Bartley’s Critique of Reason.” 1985. http://www the-rathouse com/bartagree html Bunge, Mario. 1982. Demarcating Science from Pseudoscience. Fundamenta Scientiae 3. Champion, Rafe. 1985. “Agreeing to Disagree: Bartley’s Critique of Reason.” 1985. http://www.the-rathouse.com/bartagree.html. Champion, Rafe. 1985. “Agreeing to Disagree: Bartley’s Critique of Reason.” http://www.the-rathouse.com/bartagree.html. Cioffi, Frank. 1985. “Psychoanalysis, Pseudo-Science and Testability.” In Popper and the Human Sciences, edited by Gregory Currie and Alan Musgrave, 13–44. Kluwer Academic Publishers. Derksen, A. A. 1985. “The Alleged Unity of Popper’s Philosophy of Science: Falsifiability as Fake Cement.” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition 48 (3): 313–36. http://www.jstor.org/stable/4319794. ———. 1993. “The Seven Sins of Pseudo-Science.” Journal for General Philosophy of Science / Zeitschrift Für Allgemeine Wissenschaftstheorie 24 (1): 17–42. ———. 1993. “The Seven Sins of Pseudo-Science.” Journal for General Philosophy of Science / Zeitschrift Für Allgemeine Wissenschaftstheorie 24 (1): 17–42. Feleppa, Robert. 1990. “Kuhn, Popper, and the Normative Problem of Demarcation.” Philosophy of Science and the Occult 2. Feyerabend, Paul K. 2010. Against Method. 4th edition. London ; New York: Verso. Gleberzon, William. 1984. “Academic Freedom and Holocaust Denial Literature: Dealing with Infamy.” Interchange on Education. G li A C 2001 A Sh d O O f d y g Grayling, A. C. 2001. Wittgenstein: A Very Short Introduction. OUP Oxford. Grove, J W. 1985. “Rationality at Risk: Science Against Pseudoscience.” ResearchGate. 1985. https://www.researchgate.net/publication/11730117_Rationality_at_Risk_Science_Again st_Pseudoscience. Grünbaum, Adolf. 1989. “Degeneration of Popper’s Theory of Demarcation.” 1989. http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780199989928.001.0001/a cprof-9780199989928-chapter-3. Hacking, Ian. 1983. Representing and Intervening: Introductory Topics in the Philosophy of Natural Science. Cambridge University Press. Hansson, Sven Ove. 1996. “Defining Pseudo-Science.” Philosophia Naturalis 33 (1 n Ove. 1996. “Defining Pseudo-Science.” Philosophia Naturalis 33 (1): 169–176. 35 ———. Bibliografie secundară 2006. “Falsificationism Falsified.” Foundations of Science 11 (3): 275–86. https://doi.org/10.1007/s10699-004-5922-1. ———. 2017. “Science and Pseudo-Science.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Summer 2017. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/sum2017/entries/pseudo-science/. Heath, Joseph. 2015. “Methodological Individualism.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Spring 2015. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/spr2015/entries/methodological- individualism/. Hempel, Carl G. 1958. “Empirical Statements and Falsifiability.” Philosophy 33 (127): 342–48. http://www.jstor.org/stable/3748661. p j g Hume, David. 1738. A Treatise of Human Nature. Oxford University Press. Hurd, Paul DeHart. 1998. “Scientific Literacy: New Minds for a Changing World.” Science Education 82 (3): 407–16. https://doi.org/10.1002/(SICI)1098- 237X(199806)82:3<407::AID-SCE6>3.0.CO;2-G. Jeffrey, Richard C. 1975. “Probability and Falsification: Critique of the Popper Program.” Synthese 30 (1–2): 95–117. Keita, Lansana. 1989. “Are Universal Statements Falsifiable?” Journal for General Philosophy of Science 20 (2): 351–66. https://doi.org/10.1007/BF01801484. Kuhn, Thomas. 1962. “Structura Revolutiilor Stiintifice.” 1962. http://www.humanitas.ro/humanitas/structura-revolutiilor-stiintifice-0. ———. 1970. “Logic of Discovery or Psychology of Research?” Criticism and the Growth of Knowledge. Kvasz, Ladislav. 2004. “How Can A Falsified Theory Remain Corroborated?” In Induction and Deduction in the Sciences, 263–71. Vienna Circle Institute Yearbook. Springer, Dordrecht. https://doi.org/10.1007/978-1-4020-2196-1 19. Lakatos, I. 1974. “Popper on Demarcation and Induction.” In The Philosophy of Karl Popper, edited by Karl R. Popper and Paul Arthur Schilpp, 1st ed. Vol. The Library of living philosophers. La Salle, Ill: Open Court. p p , p Lakatos, Imre. 1970. “Falsification and the Methodology of Scientific Research Programs.” 1970. http://www.stephenjaygould.org/ctrl/lakatos_prediction.html. ———. 1973. “Science and Pseudoscience.” Scribd. 1973. https://www.scribd.com/document/324484267/Science-and-Pseudoscience-1973-Imre- Lakatos. Laudan, Larry. 1983. “The Demise of the Demarcation Problem.” In Physics, Philosophy and Psychoanalysis, 111–27. Boston Studies in the Philosophy of Science. Springer, Dordrecht. https://doi.org/10.1007/978-94-009-7055-7_6. LeVine, Michael V. 2016. “Science Has Experience Fighting Fake News — and Facebook Should Take Note.” 2016. https://mic.com/articles/161376/science-has-experience- fighting-fake-news-and-facebook-should-take-note. g g Lugg, Andrew. 1992. “Pseudoscience as Nonsense.” Methodology and Science 25. Lutz, Sebastian. 2011. “On an Allegedly Essential Feature of Criteria for the Demarcation of Science.” Published Article or Volume. The Reasoner. 2011. http://www.kent.ac.uk/secl/philosophy/jw/TheReasoner/vol5/TheReasoner-5(8).pdf. 36 Magendie, Francois. 1843. “An Elementary Treatise on Human Physiology.” Bokus.com. 1843. https://www.bokus.com/bok/9781334413049/an-elementary-treatise-on-human- physiology/. Mahner, Martin. 2007. “Demarcating Science from Non-Science.” ResearchGate. 2007. https://www.researchgate.net/publication/286895878_Demarcating_Science_from_Non- Science. ———. 2013. “Science and Pseudoscience. How to Demarcate after the (Alleged) Demise of the Demarcation Problem.” In Pigliucci and Boudry. artin. 2007. “Demarcating Science from Non-Science.” ResearchGate. 2007. https://www.researchgate.net/publication/286895878_Demarcating_Science_from_Non Science. ———. 2013. “Science and Pseudoscience. Salmon, Wesley C., and Christopher Hitchcock. 2017. The Foundations of Scientific Inference: 50th Anniversary Edition. University of Pittsburgh Press. https://muse.jhu.edu/book/52596. Bibliografie secundară How to Demarcate after the (Alleged) Demise of the Demarcation Problem.” In Pigliucci and Boudry. ———. 2013. “Science and Pseudoscience. How to Demarcate after the (Alleged) Demise of the Demarcation Problem.” In Pigliucci and Boudry. Mayo, Deborah G. 1996. “Ducks, Rabbits, and Normal Science: Recasting the Kuhn’s-Eye View of Popper’s Demarcation of Science.” The British Journal for the Philosophy of Science 47 (2): 271–90. http://www.jstor.org/stable/687948. ( ) p j g Merton, Robert. 1973. “Science and Technology in a Democratic Order.” Journal of Legal and Political Sociology 1. Miller, David. 2009a. “Out of Error: Further Essays on Critical Rationalism.” In Rethinking Popper, edited by Zuzana Parusniková and R. S. Cohen, 417–423. Springer. ———. 2009b. “Some Hard Questions for Critical Rationalism.” https://philpapers.org/rec/MILSHQ. Miller, David. 2009a. “Out of Error: Further Essays on Critical Rationalism.” In Rethinking Popper, edited by Zuzana Parusniková and R. S. Cohen, 417–423. Springer. Mises, Ludwig von. 2014. “Human Action.” Text. Mises Institute. 2014. https://mises.org/library/human-action-0. Mitra, Suddhachit. 2016. “What Constitutes Science: Falsifiability as a Criterion of Demarcation.” ResearchGate. 2016. https://www.researchgate.net/publication/304462826_What_Constitutes_Science_Falsifia bility_as_a_Criterion_of_Demarcation. Morris, Robert L. 1987. “Parapsychology and the Demarcation Problem.” Inquiry 30 (3): 241– 51. https://doi.org/10.1080/00201748708602122. Musgrave, Alan, and Charles Pigden. 2016. “Imre Lakatos.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Winter 2016. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/win2016/entries/lakatos/. Newall, Paul. 2005. “Falsificationism.” 2005. https://web.archive.org/web/20070708134433/galilean-library.org/falsificationism.html. Papineau, D. 2006. “Three Scenes and a Moral.” The Philosophers’ Magazine 38. Pigliucci, Massimo. 2013. “The Demarcation Problem: A (Belated) Response to Laudan.” In Philosophy of Pseudoscience: Reconsidering the Demarcation Problem, edited by Massimo Pigliucci and Maarten Boudry, 9. University of Chicago Press. Putnam, Hilary, Philip Gasper, and J. D. Trout. 1974. The ‘Corroboration’ of Theories - In The Philosophy of Science. Vol. The ‘corroboration’ of Theories. MIT Press. Quine, W. V. O. 1953. “Two Dogmas of Empiricism.” Wikipedia. https://en.wikipedia.org/w/index.php?title=Two_Dogmas_of_Empiricism&oldid=840214 769. g p p https://en.wikipedia.org/w/index.php?title=Two_Dogmas_of_Empiricism&oldid=840214 769. Reisch, George A. 1998. “Pluralism, Logical Empiricism, and the Problem of Pseudoscience.” Philosophy of Science 65 (2): 333–48. https://doi.org/10.1086/392642. p y f ( ) p g Rothbart, Daniel. 1990. “Demarcating Genuine Science from Pseudoscience.” Philosophy of Science and the Occult 2. Salmon, Wesley C., and Christopher Hitchcock. 2017. The Foundations of Scientific Inference: 50th Anniversary Edition. University of Pittsburgh Press. https://muse.jhu.edu/book/52596. 37 Sfetcu, Nicolae. 2015. Pseudoştiinţă? Dincolo de noi... Lulu.com. Shasha, Dennis Elliot, and Cathy A. Lazere. 1998. “Out of Their Minds: The Lives and Discoveries of 15 Great Computer Scientists.” 1998. https://books.google.ro/books/about/Out_of_their_Minds.html?id=-0tDZX3z- 8UC&redir_esc=y. Bibliografie secundară _ y Shea, Brendan. 2017. “Karl Popper.” 2017. http://www.iep.utm.edu/pop-sci/. Siitonen, Arto. 1984. “Demarcation of Science From the Point of View of Problems and Problem-Stating.” Philosophia Naturalis 21: 339–353. Sokal, Alan, and Jean Bricmont. 1999. Fashionable Nonsense: Postmodern Intellectuals’ Abuse of Science. 1st edition. New York: Picador. f Stove, D. C. 1978. “Popper on Scientific Statements.” Philosophy 53 (203): 81–88. https://doi org/10 1017/S0031819100016326 Stove, D. C. 1978. “Popper on Scientific Statements.” Philosophy 53 (203): 81–88. https://doi.org/10.1017/S0031819100016326. Taliga, Miloš. 2004. “Against Watkins: From a Popperian Point of View.” Organon F: M d i á d ý A i P A l ti kú Fil fi 11 (2) 143 157 Taliga, Miloš. 2004. “Against Watkins: From a Popperian Point of View.” Organon F: Medzinárodný Asopis Pre Analytickú Filozofiu 11 (2): 143–157. a, Miloš. 2004. “Against Watkins: From a Popperian Point of View.” Organon F: Medzinárodný Asopis Pre Analytickú Filozofiu 11 (2): 143–157 Thagard, Paul R. 1978. “Why Astrology Is a Pseudoscience.” PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association 1978: 223–234. Thornton, Stephen. 2017. “Karl Popper.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Summer 2017. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/sum2017/entries/popper/. Toulmin, Stephen. 1967. “Conceptual Revolutions in Science.” Synthese 17 (1): 75–91. http://www.jstor.org/stable/20114536. Toulmin, Stephen. 1967. “Conceptual Revolutions in Science.” Synthese 17 (1): 75–91. http://www.jstor.org/stable/20114536. Watkins, J. W. N. 1968. “Hume, Carnap and Popper.” The Problem of Inductive Logic. Watkins, John W. N. 1997. “Popperian Ideas on Progress & Rationality in Science.” The Critical Rationalist 2. Watkins, John W. N. 1997. “Popperian Ideas on Progress & Rationality in Science.” The Critical Rationalist 2. Watkins, John W. N. 1997. “Popperian Ideas on Progress & Rationality in Science.” The Critical Rationalist 2. Williams, Nigel. 2005. “Heavyweight Attack on Climate-Change Denial.” Current Biology 15 (4): R109–10. https://doi.org/10.1016/j.cub.2005.02.002. Williams, Nigel. 2005. “Heavyweight Attack on Climate-Change Denial.” Current Biology 15 (4): R109–10. https://doi.org/10.1016/j.cub.2005.02.002.
https://openalex.org/W3214266937
https://air.uniud.it/bitstream/11390/1215981/1/ijerph-18-11918.pdf
English
null
Variation of the Occurrence of Physical Restraint Use in the Long-Term Care: A Scoping Review
International journal of environmental research and public health/International journal of environmental research and public health
2,021
cc-by
7,291
Article Variation of the Occurrence of Physical Restraint Use in the Long-Term Care: A Scoping Review , Martina Debiasi 2, Jessica Longhini 3, Lorenzo Giori 4, Luisa Saiani 1, Elisabetta Mezzalira 1 1 Elisa Ambrosi 1,* , Martina Debiasi 2, Jessica Longhini 3, Lorenzo Giori 4, Luisa Saiani 1, Elisabetta Mezzalira 1 and Federica Canzan 1 1 Department of Diagnostics and Public Health, University of Verona, Strada Le Grazie 8, 37134 Verona, Italy; luisa saiani@univr it (L S ); elisabetta mezzalira@univr it (E M ); federica canzan@univr it (F C ) 1 Department of Diagnostics and Public Health, University of Verona, Strada Le Grazie 8, 37134 Verona, Italy; luisa.saiani@univr.it (L.S.); elisabetta.mezzalira@univr.it (E.M.); federica.canzan@univr.it (F.C.) 2 Centre of Higher Education for Health Sciences, Azienda Provinciale per i Servizi Sanitari, Via Briamasco 2, 38121 Trento, Italy; martina.debiasi@univr.it luisa.saiani@univr.it (L.S.); elisabetta.mezzalira@univr.it (E.M.); federica.canzan@univr.it (F.C.) 2 Centre of Higher Education for Health Sciences, Azienda Provinciale per i Servizi Sanitari, Via Briamasco 2, 38121 Trento, Italy; martina.debiasi@univr.it ( ) ( ) ( ) 2 Centre of Higher Education for Health Sciences, Azienda Provinciale per i Servizi Sanitari, Via Briamasco 2, 38121 Trento, Italy; martina.debiasi@univr.it 3 Department of Medical Sciences, University of Udine, Viale Ungheria 20, 33010 Udine, Italy; jessica.longhini@uniud.it 3 Department of Medical Sciences, University of Udine, Viale Ungheria 20, 33010 Udine, Italy; jessica.longhini@uniud.it 4 Santa Maria Del Carmine Hospital, Azienda Provinciale per i Servizi Sanitari, Corso Verona, 4, 38068 Rovereto, Italy; lorenzo.giori@apss.tn.it * Correspondence: elisa.ambrosi_01@univr.it; Tel.: +39-045-8027288 Abstract: Physical restraints in the long-term care setting are still commonly used in several countries with a prevalence ranging from 6% to 85%. Trying to have a broad and extensive overlook on the physical restraints use in long-term care is important to design interventions to prevent and/or reduce their use. Therefore, the aim of this scoping review was to analyze the range of occurrence of physical restraint in nursing homes, long-term care facilities, and psychogeriatric units. Pubmed, CINAHL, Ovid PsycINFO- databases were searched for studies with concepts about physical restraint use in the European long-term care setting published between 2009 and 2019, along with a hand search of the bibliographies of the included studies. Data on study design, data sources, clinical setting and sample characteristics were extracted. A total of 24 studies were included. The median occurrence of physical restraint in the European long-term care setting was still high (26.5%; IQR 16.5% to 38.5%) with a significant variability across the studies.   Citation: Ambrosi, E.; Debiasi, M.; Longhini, J.; Giori, L.; Saiani, L.; Mezzalira, E.; Canzan, F. Variation of the Occurrence of Physical Restraint Use in the Long-Term Care: A Scoping Review. Int. J. Environ. Res. Public Health 2021, 18, 11918. https:// doi.org/10.3390/ijerph182211918 Keywords: long-term care; physical restraints; prevalence; incidence; scoping review Academic Editor: Paul B. Tchounwou Received: 28 September 2021 Accepted: 10 November 2021 Published: 13 November 2021 Received: 28 September 2021 Accepted: 10 November 2021 Published: 13 November 2021 Int. J. Environ. Res. Public Health 2021, 18, 11918. https://doi.org/10.3390/ijerph182211918 Article Variation of the Occurrence of Physical Restraint Use in the Long-Term Care: A Scoping Review The heterogeneity of data varied according to study design, data sources, clinical setting, physical restraint’s definition, and patient characteristics, such as ADLs dependence, presence of dementia and psychoactive drugs prescription. International Journal of Environmental Research and Public Health International Journal of Environmental Research and Public Health 1. Introduction Physical restraint (PR) is a complex practice, which involves the different health professionals involved in the daily routine of nursing homes; its most shared definition is “any device, material or equipment attached to or near a person’s body and which cannot be controlled or easily removed by the person, and which deliberately prevents or is deliberately intended to prevent a person’s free body movement to a position of choice and/or a person’s normal access to their body” [1,2]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. This definition includes a wide range of devices that may be used to contain patients, such as: bilateral bed rails, limb or trunk belts, fixed tables on a chair or chairs that prevent patients from getting up, containment sheets and pajamas [2]; the most frequently reported reasons for their application are essentially three: management of patient agitation [3,4], prevention of falls [5] and for medical devices safety [3,4,6,7] Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). It is essential to underline that the use of PR does not come without risks: adverse effects can be observed, ranging from bodily injuries, mortality, decreased mobility, to reduced physiological well-being. Nevertheless, it is not associated with a reduction of falls, even if PR use is still justified in clinical practice as a prevention fall strategy [8,9] j p p gy Despite this, PR is still commonly used in in long-term care settings of several coun- tries, with a range of prevalence from 6% to 85%. The estimated prevalence of PR use nt. J. Environ. Res. Public Health 2021, 18, 11918. https://doi.org/10.3390/ijerph182211918 https://www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2021, 18, 11918 2 of 11 may vary considerably due to several reasons such as: (1) different definitions of physical restraint, (2) different ways to collect and report the data, (3) different characteristics of the care settings (4) clinical variability of patients and (5) different laws and policies [9,10]. The reported prevalence may then further vary depending on the type of restraint used, the estimated time of permanence, and the contextual administration of drugs [9,11–13]. 2. Materials and Methods This scoping review was based on the design by Arksey and O’Malley [17] with enhancements from Levac et al. [18]. A scoping review is a knowledge synthesis approach focusing on an exploratory research question. It is aimed at mapping key concepts, types of evidence, and gaps in research related to a defined area or field [19]. The method included the following steps: specifying the research question; identifying relevant studies; extracting relevant information; collating and summarizing the information. g g g The present scoping review was guided by the following research questions: (1) What is the occurrence of the use of physical restraint in the European long-term care setting? (2) What factors contribute to the heterogeneity of data reported in the literature? 1. Introduction Therefore, there is a significant potential for discrepancy in the occurrence of physical restraint use in long-term care that may not always be considered in studies and these relevant differences constitute a challenge for researchers and clinicians, making more complex to identify which factors affect PR use in long-term care. p y g Many efforts have been made to reduce PR use in a wide range of countries, such as in the United States where a well-designed reform regarding nursing homes (Nursing Home Reform Act OBRA ‘87) [14] has led to a drastic reduction of restraint use. However, the European Union does not currently have a shared law or policy that may be used as a guidance for clinical practice on physical restraints. Thus, as highlighted by Lee et al. (2021) [15], it is essential to understand how culture and economic aspects may have an impact on physical restraint policies of any country. As this premise suggests, trying to have a broad and extensive overlook on the physical restraint use in long-term care is an important step to accurately measuring and subsequently designing interventions to prevent and/or reduce their use. As Munn and colleagues (2018) [16] stated in their manuscript, performing a scoping review is useful and effective to have an extensive overview of a body of literature. Such a broad and extensive overlook regarding the use of physical restraints in European long term care settings constitutes an essential step to accurately measure and subsequently design interventions to prevent and/or reduce PR use. Therefore, the aim of this scoping review was to examine the variation of the oc- currence (prevalence/incidence) of physical restraint use in the European long-term care settings (nursing homes, long term care facilities, and psychogeriatric units). 2.1. Search Strategy The search strategy included both free text and controlled vocabulary for the concepts of “physical restraint” and “long-term care” (nursing homes, long-term care facilities and psychogeriatric units). The original search was conducted in April 2019. The databases searched were Ovid Medline, PubMed, CINAHL, Ovid PsycINFO. A hand search was conducted of the bibliographies of the included studies. The databases were searched using the following terms: ((nursing home* OR residential care facilit* OR long-term care) and physical restraint*) in combination. 2.2. Identification of Relevant Studies Studies that defined or assessed physical restraint use in residents living in nursing homes, long-term facilities and psychogeriatric units as an aim, outcome, or covariate, were included. Studies conducted in European countries and published between 1 January 2009 and 30 April 2019, were selected. The motivation for including only the studies published in Int. J. Environ. Res. Public Health 2021, 18, 11918 3 of 11 the last decade was to obtain an updated picture of the phenomenon, to provide guidance to design and act on specific aspects of physical restraint use in the clinical contexts. the last decade was to obtain an updated picture of the phenomenon, to provide guidance to design and act on specific aspects of physical restraint use in the clinical contexts. Qualitative studies, systematic reviews, meta-syntheses, case series, case reports, conference abstracts, letters, comments, and opinion pieces were excluded; additional exclusion criteria included studies on pediatric population, conducted in hospital setting, geriatric and medical department, intensive care units, home care and residential care facilities, and not written in English, Italian, or Spanish. Studies in which physical restraint prevalence was not reported or could not be computed were also excluded. p p p Consistently with the scoping review methodology, the quality of included studies was not assessed [17]. Title/abstract and full text screening was performed by two reviewers independently. Disagreements were solved through discussion between the two reviewers or among the full research group. 2.3. Data Extraction Information was extracted regarding the study characteristics including design, year of publication, origin/country of origin (where the study was published or conducted), study population (gender, clinical setting, average age of resident/patients, disease conditions) and sample size (if applicable), data collection, definition, and occurrence of physical restraint. PR occurrence was computed when it was not reported but all the necessary information was available. For this calculation, the definition provided by Bleijlevens and colleagues (“any device, material or equipment attached to or near a person’s body and which cannot be controlled or easily removed by the person, and which deliberately prevents or is deliberately intended to prevent a person’s free body movement to a position of choice and/or a person’s normal access to their body”) [1] was used when a study did not report a definition. Data extraction was completed by one reviewer and checked by a secondary reviewer for accuracy. 2.4. Statistical Analysis The normal distribution of prevalence occurrence was checked using the quantile– quantile plot, Shapiro-Wilk test, considering as acceptable a value lower than |0.5| of Skew- ness and Kurtosis. The prevalence occurrence was not distributed normally, thus median, and interquartile range have been used as measures to report data. The Mann-Whitney test and the Kruskal-Wallis test have been used when appropriate as non-parametric statistical methods in sub-groups analysis, at a statistical significance level of <0.05, to detect possible associations between prevalence occurrence and categorical variables extracted by study included. The information from the studies was extracted and imported into a Microsoft Excel 2010 database. Data analysis was conducted using R-statistical software (R Core Team, 2021). 3.1. Summary of Included Studies 3.1. Summary of Included Studies Six hundred and nineteen titles were identified and screened, of whom 40 qualified for review and data extraction. Of the 40 studies, 10 that did not report overall occurrence of physical restraint use and six that were study protocols, were excluded. The final sample was constituted by 24 [11,13,20–40] studies (Figure 1). The distribution of the characteristics of the 24 studies that estimated overall preva- lence is synthesized in Table 1. The studies were mostly conducted in the Netherlands (33.3%, n = 8), published between 2009 and 2013 (58.3%, n = 14), cross-sectional or RCT by design (58%, n = 14), and predominantly used observation to collect data (38%, n = 9). 4 of 11 Int. J. Environ. Res. Public Health 2021, 18, 11918 Figure 1. Flow diagram of studies identified, screened, and extracted Records identified through database searching: (n = 1006) Additional records identified through hand search: (n = 20) Records removed before screening: Duplicate records removed (n = 407) Records screened (n = 619) Records excluded (n = 579) Reports sought for retrieval (n = 46) Reports not retrieved (n = 6) Full-text articles assessed for eligibility (n = 40) Reports excluded: Occurrence not reported or could not be computed (n = 10) Study protocol (n = 6) Studies included in review (n =24) Identification Screening Included Figure 1. Flow diagram of studies identified, screened, and extracted, PRISMA 2020. Records excluded (n = 579) Reports sought for retrieval (n = 46) Reports not retrieved (n = 6) Study protocol (n = 6) Studies included in review (n =24) Studies included in review Figure 1. Flow diagram of studies identified, screened, and e Figure 1. Flow diagram of studies identified, screened, and extracted, PRISMA 2020. g g The distribution of the characteristics of the 24 studies that estimated ov lence is synthesized in Table 1. The studies were mostly conducted in the (33.3%, n = 8), published between 2009 and 2013 (58.3%, n = 14), cross-section design (58%, n = 14), and predominantly used observation to collect data (38% Table 1. Characteristics of included studies and occurrence of physical restraint use. Characteristic N. Studies Median Occur- rence (%) IQR p Overall 24 100% 26.39 (6.7–40.4) European Coun- tries Table 1. Characteristics of included studies and occurrence of physical restraint use. Characteristic N. 3.1. Summary of Included Studies Studies Median Occurrence (%) IQR p-Value Overall 24 100% 26.39 (6.7–40.4) European Countries - Netherlands 8 33.3% 41.7 (26.9–57.3) Germany 4 16.7% 27.2 (21.6–29.3) Norway 2 8.3% 26.5 (21.5–31.6) Spain 2 8.3% 14.8 (13.8–15.8) Italy 2 8.3% 19.0 (13.7–24.4) Belgium 1 4.2% 47.5 (47.5–47.5) Sweden 1 4.2% 25.3 (25.3–25.3) Swiss 1 4.2% 26.7 (26.7–26.7) The distribution of the characteristics of the 24 stu Table 1. Characteristics of included studies and occurrence of physical restraint use. Int. J. Environ. Res. Public Health 2021, 18, 11918 5 of 11 Table 1. Cont. Characteristic N. Studies Median Occurrence (%) IQR p-Value Finland 1 4.2% 52.0 (52.0–52.0) Combinations 2 8.3% 18.3 (17.7–18.9) Year of publication 0.841 $ 2009–2013 14 58.3% 27.2 (17.6- 35.7) 2014–2019 10 41.7% 23.2 (16.7–43.4) Study design 0.125 † Cross-sectional 8 33% 29.9 (23.8–40.7) Randomized Controlled Trial 6 25% 26,7 (18.7–34.5) Quasi-experimental 5 21% 45.0 (29.5–59.2) Cohort 5 21% 16.8 (14.5–19.1) Data sources 0.149 † Chart review/Electronic records 7 29% 36.6 (20.1–43.0) Observation 9 38% 29.5 (26.0–56.6) Standardized scale and interview 4 17% 16.8 (15.2– 17.5) Combinations 4 17% 18.2 (16.2–25.9) Sample size 0.473 † 100–299 8 33% 33.0 (16.0–47.9) 300–999 5 21% 25.3 (19.6–47.5) 1000+ 11 46% 26.3 (15.1–30.5) Average age of enrolled patients 0.742 $ <85 years 16 66.7% 27.0 (16.2–46.7) >85 years 8 33.3% 23.2 (18.4–33.9) Sex percentage - Mostly females > 50% 24 100% 26.4 (16.7–40.1) Mostly males >50% – – - - Health care setting 0.04 $ Nursing home and long-term care 18 75% 22.5 (16.6–31.8) Psychogeriatric units 6 25% 50.8 (33.4–58.5) Physical restraint role 0.009 $ Outcome 18 75% 31.2 (25.6–46.9) Covariate 6 25% 15.7 (13.7–18.5) Definition of physical restraint 0.002 † Bed rails included 11 45.8% 36.6 (28.8–54.3) Bed rails excluded 10 41.7% 16.9 (13.7–23.8) Not specified 3 12.5% 14.5 (13.9–17.1) Presence of dementia in enrolled patients - Dementia/cognitive impairment (from mild to severe) 16 66.7% 93.9 (62.1–100.0) Not reported 8 33.3% - - Presence of ADLs dependence in enrolled patients 0.002 † No 4 16.67% 8.0 (5.7–13.2) Mild 9 37.5% 7.2 (5.0–30.0) Severe 8 33.3% 39.7 (37.1- 42.5) Very severe 11 45.8% 53.8 (21.9- 64.5) Not reported 13 54.2% - - Use of drugs in enrolled patients 0.014 † Antipsychotics/neuroleptics 7 29.1% 44.7 (36.-67.5) Benzodiazepines 7 29.1% 29.1 (20.8–33.9) Psychoactive drugs (not specified) 9 37.5% 68.2 (53.5–70.2) Not reported 8 33.3% - - Abbreviations: IQR, interquartile range; ADLs, activities of daily living. Abbreviations: IQR, interquartile range; ADLs, activities of daily living. $ Mann-Whitney test; † Kruskal-Wallis test. 3.2. Occurrence of Physical Restraint Use 3.2. Occurrence of Physical Restraint Use As shown in Figure 2, occurrence of physical restraint use in long-term care was highly variable, ranging from 7.7% to 60.5%. The median occurrence was 26.5% (IQR 16.5–38.5%). 21, 18, x 7 of 12 As shown in Figure 2, occurrence of physical restraint use in long term care was highly variable, ranging from 7.7% to 60.5%. The median occurrence was 26.5% (IQR 16.5–38.5%). es. ublic ealth 0 , 8, o Figure 2. Distribution of overall occurrence of physical restraint use in long-term care. 1 9 5 3 2 3 1 0 1 2 3 4 5 6 7 8 9 10 0 -9 10 -19 20-29 30-39 40-49 50-59 60-69 Number of studies Occurrence of physical restraint use Figure 2. Distribution of overall occurrence of physical restraint use in long-term care. Figure 2. Distribution of overall occurrence of physical restraint use in long-term care. Figure 2. Distribution of overall occurrence of physical restraint use in long-term care. Table 1 shows the median, first and third quartiles of the occurrence of physical re- straint use in relation to different characteristics of the included studies. Table 1 shows the median, first and third quartiles of the occurrence of physical restraint use in relation to different characteristics of the included studies. Concerning European continent, the study conducted in Finland had the highest me- dian occurrence of 52.0% (IQR 52.0%–52.0%), while studies (n = 2) conducted in Spain had the lowest median occurrence of 14.8% (IQR 13.8%–15.8%). In the last decade, a decrease in the physical restraint use has been occurred (27.2%; IQR 17.6%–35.7% in the 2009–2013 period vs. 23.2%; IQR 16.6%–43.4% in the 2014–2019 period). Nevertheless, no statistically significant difference was found (p = 0.841) (Table 1). Concerning European continent, the study conducted in Finland had the highest median occurrence of 52.0% (IQR 52.0–52.0%), while studies (n = 2) conducted in Spain had the lowest median occurrence of 14.8% (IQR 13.8–15.8%). In the last decade, a decrease in the physical restraint use has been occurred (27.2%; IQR 17.6–35.7% in the 2009–2013 period vs. 23.2%; IQR 16.6–43.4% in the 2014–2019 period). Nevertheless, no statistically significant difference was found (p = 0.841) (Table 1). 3.1. Summary of Included Studies $ Mann-Whitney test; † Kruskal-Wallis test. Table 1. Cont. 6 of 11 Int. J. Environ. Res. Public Health 2021, 18, 11918 Moreover, most studies enrolled more than 1000 patients (46%, n = 11), predominantly women (100%, n = 24), with an average age younger than 85 years (66.7%, n = 16) and were conducted in nursing homes and long-term care (75% n = 18). The frequency of physical restraint use was measured both as prevalence (37.5%, n = 9) and as incidence (62.5%, n = 15); these two measures, for this analysis, were treated the same way. About 50% of the studies (n = 11) considered bed rails as a type of physical restraint. Dementia or cognitive impairment (from mild or severe) and the use of psychotropic drugs (antipsychotics/neuroleptics, benzodiazepines) were both assessed in the 66.6% (n = 16) of the studies. In the 45.8% (n = 11) of the studies, dependence in activities of daily living (ADLs), was also measured. 4. Discussion Using the scoping review approach, the present study found the occurrence of physical restraint use in long-term care within the European context varying by study setting, methodology, definition of physical restraint, or clinical conditions (e.g., level of ADLs dependence, presence of dementia/cognitive impairment, use of psychotropic drugs). We found studies done in Northern Europe (e.g., Finland and Belgium) or conducted in psychogeriatric units to have a median occurrence higher than 45% while, on the other hand, studies conducted in Southern Europe (e.g., Spain and Italy) or in nursing homes, had a median occurrence of less than 25%. Study methodology, including design and data sources, had a great impact on the reported occurrence of physical restraint. It is relevant to highlight that most of the included studies in this review had a cross-sectional design. Results from a cross- sectional study may be less able to deepen the phenomenon described as the measurement is conducted only once, compared to longitudinal studies where a series of measurements are performed over a period of time. On the other hand, the opportunity to have six RCTs allowed us to analyze incidence data. Future research on this topic should include prospective longitudinal studies aiming to obtain an accurate occurrence analysis. The use of the electronic records allows for retrospective studies in healthcare and in our results, the highest median occurrence was observed in the studies which used electronic records, while the lowest median prevalence was found in the exploratory longitudinal studies exploiting direct observation of the patients. It is undeniable that the measurement approach has an impact on the results, and this is confirmed even by the work of Lee and colleagues [15], which suggested to use a direct observation approach or a combination of multiple measurement approaches, in order to not underestimate the occurrence of physical restraint use. p y Our review observed a decrease in the occurrence of physical restraint use over the last decade, which may be explained by the publication and dissemination of the guidelines “Promoting Safety: Alternative Approaches to the Use of Restraint”, developed by the Registered Nursing Association of Ontario [41]. Indeed, these guidelines recommend a preventive approach to promote patient safety and, at the same time, guarantee freedom. It is therefore necessary to focus on the assessment of at-risk patients and to implement alternative strategies to the use of physical restraint. 3.2. Occurrence of Physical Restraint Use sig i ica di e e ce as ou d (p 0 8 ) ( ab e ) With respect to the study design, the quasi-experimental study had the highest me- dian occurrence of 45% (IQR 29.5%–59.2%). Among the other study designs, the cross- sectional studies had the highest median occurrence of 29.9% (IQR 23.8%–40.7%), while the cohort studies had the lowest median occurrence of 16.8% (IQR 14.5%–19.1%), p = g p With respect to the study design, the quasi-experimental study had the highest median occurrence of 45% (IQR 29.5–59.2%). Among the other study designs, the cross-sectional studies had the highest median occurrence of 29.9% (IQR 23.8–40.7%), while the cohort studies had the lowest median occurrence of 16.8% (IQR 14.5–19.1%), p = 0.125. the cohort studies had the lowest median occurrence of 16.8% (IQR 14.5% 19.1%), p 0.125. Studies using the chart review and electronic health records (36.6%; IQR 20.1%–43%) had higher median occurrence than studies using other data sources, with standardized scale and interview based studies reporting the lowest occurrence of 16 8% (IQR 15 2% Studies using the chart review and electronic health records (36.6%; IQR 20.1–43%) had higher median occurrence than studies using other data sources, with standardized scale and interview-based studies reporting the lowest occurrence of 16.8% (IQR 15.2–17.5%), p = 0.149. 7 of 11 Int. J. Environ. Res. Public Health 2021, 18, 11918 Studies enrolling a small sample size (33.05%; IQR 16.0–47.9% in the 100–299 patients group) had higher median occurrence than studies with a large sample size (26.3%; IQR 15.1–30.5% in the 1000+ group), p = 0.473. g p p The median occurrence of physical restraint use tended to decrease when the average age of the sample was higher, going from 27.0% (IQR 16.2–46.7%) of the <85 years class, to 23.2% (IQR 18.4–33.9%) of the >85 years class; nevertheless, no statistically significant differences were detected (p = 0.742). p Studied conducted in psychogeriatric units (50.8%; IQR 33.4–58.5%) had a statistically significant higher median occurrence than studies conducted in nursing homes and long- term care (22.5%; IQR 16.6–31.8%), p = 0.04. Studies that considered bed rails as a physical restraint had a statistically significant higher median occurrence 36.6%; IQR 28.8–54.3%) than studies excluding them (16.9%; IQR 13.7–23.8%), p = 0.002. There has been a steady increase in the physical restraint use with the worsening of ADLs dependence (7.15%, 39.7%, 53.8%, p = 0.002). 3.2. Occurrence of Physical Restraint Use Regarding medications, the highest median occurrence of physical restraint use was observed in patients prescribed with psychotropic drugs (not specified) (68.2%; IQR 53.5– 70.2%, p = 0.014). Studies that defined drug classification reported a median occurrence of physical restraints of 44.7% (IQR 36.0–67.5%) and of 29.1% (IQR 20.8–33.9%) in patients prescribed with antipsychotics/neuroleptics and benzodiazepines, respectively. 4. Discussion In order to implement these changes in clinical practice, it is essential to analyze every professional’s own working context and Int. J. Environ. Res. Public Health 2021, 18, 11918 8 of 11 8 of 11 to design basic, advanced and continuing education interventions relevant to the learning needs of the recipients. However, education represents a necessary, but not sufficient, element to create a restraint-free environment. In this sense, it becomes crucial to imple- ment strategies that promote an organizational culture supporting the implementation of evidence-based guidelines, the active involvement of patients and their caregivers in the care process, the consultation of expert nurses, the availability of alternative strategies, such as low beds or electronic alarm systems. y We found that, compared to younger patients, older patients were more often put on physical restraints, although no statistically significant difference was detected between the three age groups. The health care setting has a great impact on the occurrence of physical restraints as demonstrated in our study. The psychogeriatric units had the highest magnitude of physical restraint use; this result may be explained by the different clinical characteristics of the patients admitted to this setting, who generally present dementia and behavioral disorders, such as aggressiveness, agitation, disinhibition, and wandering, which are strictly associated to physical restraint use [13,42]. Another possible explanation of the variation of physical restraint use in the different settings may be related to organizational conditions, such as nurse ratio and competence, staff mix and ward characteristics [42]. However, this aspect is very controversial in the international literature; some studies did not find a statistically significant difference between staff intensity and staff mix in relation to the use of restraint [11], other studies found that high workloads and patient’s physical and cognitive disability were moderately associated with the use of physical restraint [43]. The median occurrence of physical restraint use increased in case bed side rails were also considered. Many studies did not consider bed rails as means of restraint although they are extensively used in the clinical practice. Indeed, they are sometimes considered as safety tools, used to reduce the risk of accidentally slipping, rolling, or falling out of bed [43]. In order to improve future research, a shared definition of physical restraints should be provided [40]. 4. Discussion Almost all the patients involved in the studies presented dementia and/or cognitive impairment, were prescribed with at least one antipsychotic drug, and had some degree of dependence in ADLs. According to previous studies, being cognitive impaired, highly dependent in ADLs and being prescribed with psychoactive drugs increased the risk of being restrained [43]. It is relevant to highlight that it might be interesting to produce a summary of the PR use occurrence combining the individual effects of each included study. Nevertheless, due to the nature of the scoping review (which is not expected to pool the findings in the form of a meta-analysis) and the extremely high heterogeneity of the included studies, this has not been possible to perform with our study. Future studies need to focus on conducting systematic reviews and specifically to combine the results and give more information in physical restraint use in relation to specific populations and settings, in order to be able to inform policy. Strengths and Limitations The scoping review methodology allows to perform an extensive literature overview according to a rigorous approach. Moreover, our review has included several studies on the phenomenon conducted in a variety of settings, which permitted us to analyze the occurrence of physical restraint in different patients’ subgroups. However, due to the heterogeneity of the included studies, median and interquartile range has been used to summarize our findings, although they are not the most suitable metrics. The median was preferred over the mean as it is less influenced by the significative range variability founded in the literature reporting the prevalence of physical restraint use. An assessment of methodological limitations or risk of bias of the evidence included in this scoping review was not performed, this could be a limitation. Although, according to the scoping review method, a critically appraisal of the included studies is not expected. 9 of 11 9 of 11 Int. J. Environ. Res. Public Health 2021, 18, 11918 Moreover, some studies published into national medical or health care journals in country languages might have been missed as we did not search in all the available databases. Finally, in this scoping review six studies were unretrieved and this may have affected the occurrence assessment. 5. Conclusions The occurrence of physical restraint in the European long-term care setting is still high (26.4% median). A significant variability emerged across the studies published between 2009 and 2019, with a median occurrence of restraint use ranging from 16.5% (IQR1) to 38.5% (IQR3). The heterogeneity of data reported in the included studies varied according to study design, data sources, clinical setting, physical restraint’s definition, and patient characteris- tics, such as ADLs dependence, presence of dementia and psychoactive drugs prescription. p p p y g p p The wide range of methodological and clinical differences make an effective compari- son across the involved studies difficult. Further studies need to be focused on the clear description of the factors that are associated with a variability on physical restraint use. To this aim, particular attention in sample description may be helpful for defining and explaining the different restraint occurrence in the European long-term care. Moreover, particular attention for future research needs to be focused on the ethical issues concerning physical restraint policy. Indeed, patients or legal representees must be involved in the making decision process and the opportunity of having available data on how many times this happen in real life is essential as it better clarifies who are the health professionals involved in the decision process. Author Contributions: L.G., M.D.: searched the literature; data extraction and analysis; writing— review and editing; J.L., E.M.: data analysis; writing—review and editing L.S.: research supervision; writing—review and editing; F.C., E.A.: conceptualization; methodology; data extraction and analysis; visualization; research supervision, writing—original draft preparation. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data are stored at the Department of Diagnostics and Public Health, University of Verona. Data Availability Statement: Data are stored at the Department of Diagnostics and Public Health, University of Verona. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. 7. Ben Natan, M.; Akrish, O.; Zaltkina, B.; Noy, R.H. Physically Restraining Elder Residents of Long-Term Care Facilities from a Nurses’ Perspective. Int. J. Nurs. Pract. 2010, 16, 499–507. [CrossRef] 5. Via-Clavero, G.; Sanjuán-Naváis, M.; Romero-García, M.; de la Cueva-Ariza, L.; Martínez-Estalella, G.; Plata-Menchaca, E.; Delgado-Hito, P. Eliciting Critical Care Nurses’ Beliefs Regarding Physical Restraint Use. Nurs. Ethics 2019, 26, 1458–1472. [CrossRef] References [CrossRef] [ ] [ ] 9. Hamers, J.P.; Huizing, A.R. Why Do We Use Physical Restraints in the Elderly? Z. Gerontol. Geriatr. 2005, 38, 19–25. [CrossRef] 10. Huizing, A.R.; Hamers, J.P.; de Jonge, J.; Candel, M.; Berger, M.P. Organisational Determinants of the Use of Physical Restraints: 10. Huizing, A.R.; Hamers, J.P.; de Jonge, J.; Candel, M.; Berger, M.P. Organisational Determinants of the U A Multilevel Approach. Soc. Sci. Med. 2007, 65, 924–933. [CrossRef] pp , , [ ] 11. Hofmann, H.; Schorro, E.; Haastert, B.; Meyer, G. Use of Physical Restraints in Nursing Homes: A Multicentre Cross-Sectional Study. BMC Geriatr. 2015, 15, 129. [CrossRef] 12. Dunn, K.S. The Effect of Physical Restraints on Fall Rates in Older Adults Who Are Institutionalized. J. Gerontol. Nurs. 2001, 27, 40–48. [CrossRef] [PubMed] 13. Feng, Z.; Hirdes, J.P.; Smith, T.F.; Finne-Soveri, H.; Chi, I.; Du Pasquier, J.N.; Gilgen, R.; Ikegami, N.; Mor, V. Use of Physical Restraints and Antipsychotic Medications in Nursing Homes: A Cross-National Study. Int. J. Geriatr. Psychiatry 2009, 24, 1110–1118. [CrossRef] 14. Marek, K.D.; Rantz, M.J.; Fagin, C.M.; Krejci, J.W. Obra ’87: Has It Resulted in Better Quality of Care? J. Gerontol. Nurs. 1996, 22, 28–36. [CrossRef] 15. Lee, D.A.; Robins, L.M.; Bell, J.S.; Srikanth, V.; Möhler, R.; Hill, K.D.; Griffiths, D.; Haines, T.P. Prevalence and Variability in Use of Physical and Chemical Restraints in Residential Aged Care Facilities: A Systematic Review and Meta-Analysis. Int. J. Nurs. Stud. 2021, 117, 103856. [CrossRef] [PubMed] 16. Munn, Z.; Peters, M.D.J.; Stern, C.; Tufanaru, C.; McArthur, A.; Aromataris, E. Systematic Review or Scoping Review? Guidance for Authors When Choosing between a Systematic or Scoping Review Approach. BMC Med. Res. Methodol. 2018, 18, 143. [CrossRef] ey, L. Scoping Studies: Towards a Methodological Framework. Int. J. Soc. Res. Methodol. 2005, 8, 19–32 17. Arksey, H.; O’Malley, L. Scoping Studies: Towards a Methodological Framework. Int. J. Soc. Res. [CrossRef] 18. Levac, D.; Colquhoun, H.; O’Brien, K.K.; Levac, D.; Colquhoun, H.; O’Brien, K.K. Scoping Studies: Advancing the Methodology. Implement. Sci. 2010, 5, 69. [CrossRef] p 19. Daudt, H.M.L.; van Mossel, C.; Scott, S.J. Enhancing the Scoping Study Methodology: A Large, Inter-Professional Team’s Experience with Arksey and O’malley’s Framework. BMC Med. Res. Methodol. 2013, 13, 48. [CrossRef] 20. Meyer, G.; Köpke, S.; Haastert, B.; Mühlhauser, I. Restraint Use among Nursing Home Residents: Cross-Sectional Study and Prospective Cohort Study. J. Clin. Nurs. 2009, 18, 981–990. [CrossRef] 21. References Aranda-Gallardo, M.; Morales-Asencio, J.M.; Enriquez de Luna-Rodriguez, M.; Vazquez-Blanco, M.J.; Morilla-Herrera, J.C.; Rivas- Ruiz, F.; Toribio-Montero, J.C.; Canca-Sanchez, J.C. Characteristics, Consequences and Prevention of Falls in Institutionalised Older Adults in the Province of Malaga (Spain): A Prospective, Cohort, Multicentre Study. BMJ Open 2018, 8, e020039. [CrossRef] [PubMed] 22. Attanasio, A.; Bedin, M.; Stocco, S.; Negrin, V.; Biancon, A.; Cecchetto, G.; Tagliapietra, M. Clinical Outc of Enteral Nutrition among Older Adults. Minerva Med. 2009, 100, 159–166. [PubMed] 23. Te Boekhorst, S.; Depla, M.F.; Francke, A.L.; Twisk, J.W.; Zwijsen, S.A.; Hertogh, C.M. Quality of Life of Nursing-Home Residents with Dementia Subject to Surveillance Technology Versus Physical Restraints: An Explorative Study. Int. J. Geriatr. Psychiatry 2013, 28, 356–363. [CrossRef] 24. Foebel, A.D.; Onder, G.; Finne-Soveri, H.; Lukas, A.; Denkinger, M.D.; Carfi, A.; Vetrano, D.L.; Brandi, V.; Bernabei, R.; Liperoti, R. Physical Restraint and Antipsychotic Medication Use among Nursing Home Residents with Dementia. J. Am. Med. Dir. Assoc. 2016, 17, 184.e9–184.e14. [CrossRef] 25. Gulpers, M.J.; Bleijlevens, M.H.; Ambergen, T.; Capezuti, E.; Van Rossum, E.; Hamers, J.P. Belt Restraint Reduction in Nursing Homes: Effects of a Multicomponent Intervention Program. J. Am. Geriatr. Soc. 2011, 59, 2029–2036. [CrossRef] 26. Gulpers, M.J.; Bleijlevens, M.H.; Capezuti, E.; Van Rossum, E.; Ambergen, T.; Hamers, J.P. Preventing Be Admitted Residents in Nursing Homes: A Quasi-Experimental Study. Int. J. Nurs. Stud. 2012, 49, 1473 Admitted Residents in Nursing Homes: A Quasi Experimental Study. Int. J. Nurs. Stud. 2012, 49, 1473 1479. [CrossRef] 27. Gulpers, M.J.; Bleijlevens, M.H.; Ambergen, T.; Capezuti, E.; Van Rossum, E.; Hamers, J.P. Reduction of Belt Restraint Use: Long-Term Effects of the Exbelt Intervention. J. Am. Geriatr. Soc. 2013, 61, 107–112. [CrossRef] [PubMed] 27. Gulpers, M.J.; Bleijlevens, M.H.; Ambergen, T.; Capezuti, E.; Van Rossum, E.; Hamers, J.P. Reducti Long-Term Effects of the Exbelt Intervention. J. Am. Geriatr. Soc. 2013, 61, 107–112. [CrossRef] [PubMe g 28. Heeren, P.; Van de Water, G.; De Paepe, L.; Boonen, S.; Vleugels, A.; Milisen, K. Staffing Levels and the Use of Physical Restraints in Nursing Homes: A Multicenter Study. J. Gerontol. Nurs. 2014, 40, 48–54. [CrossRef] g y 29. Heinze, C.; Dassen, T.; Grittner, U. Use of Physical Restraints in Nursing Homes and Hospitals and Related Factors: A Cross- Sectional Study. J. Clin. Nurs. 2012, 21, 1033–1040. [CrossRef] 30. Huizing, A.R.; Hamers, J.P.; Gulpers, M.J.; Berger, M.P. References 1. Bleijlevens, M.H.; Wagner, L.M.; Capezuti, E.; Hamers, J.P. Physical Restraints: Consensus of a Research Definition Using a Modified Delphi Technique. J. Am. Geriatr. Soc. 2016, 64, 2307–2310. [CrossRef] 1. Bleijlevens, M.H.; Wagner, L.M.; Capezuti, E.; Hamers, J.P. Physical Restraints: Consensus of a Research Definition Using a Modified Delphi Technique. J. Am. Geriatr. Soc. 2016, 64, 2307–2310. [CrossRef] 2. Evans, D.; Wood, J.; Lambert, L. A Review of Physical Restraint Minimization in the Acute and Residen Nurs. 2002, 40, 616–625. [CrossRef] [PubMed] 3. Gallinagh, R.; Nevin, R.; Mc Ilroy, D.; Mitchell, F.; Campbell, L.; Ludwick, R.; McKenna, H. The Use of Physical Restraints as a Safety Measure in the Care of Older People in Four Rehabilitation Wards: Findings from an Exploratory Study. Int. J. Nurs. Stud. 2002, 39, 147–156. [CrossRef] 4. Luk, E.; Burry, L.; Rezaie, S.; Mehta, S.; Rose, L. Critical Care Nurses’ Decisions Regarding Physical Re Icus: A Prospective Observational Study. Can. J. Crit. Care Nurs. 2015, 26, 16–22. 5. Via-Clavero, G.; Sanjuán-Naváis, M.; Romero-García, M.; de la Cueva-Ariza, L.; Martínez-Estalella, G.; Plata-Menchaca, E.; Delgado-Hito, P. Eliciting Critical Care Nurses’ Beliefs Regarding Physical Restraint Use. Nurs. Ethics 2019, 26, 1458–1472. [CrossRef] 5. Via-Clavero, G.; Sanjuán-Naváis, M.; Romero-García, M.; de la Cueva-Ariza, L.; Martínez-Estalella, G.; Plata-Menchaca, E.; Delgado-Hito, P. Eliciting Critical Care Nurses’ Beliefs Regarding Physical Restraint Use. Nurs. Ethics 2019, 26, 1458–1472. [CrossRef] dam, S.; Endacott, R. Physical Restraint Use in Intensive Care Units across Europe: The Price Study. Intensiv 0, 26, 241–245. [CrossRef] [PubMed] 7. Ben Natan, M.; Akrish, O.; Zaltkina, B.; Noy, R.H. Physically Restraining Elder Residents of Long-Term Care Facilities from a Nurses’ Perspective. Int. J. Nurs. Pract. 2010, 16, 499–507. [CrossRef] 10 of 11 10 of 11 Int. J. Environ. Res. Public Health 2021, 18, 11918 8. Capezuti, E. Minimizing the Use of Restrictive Devices in Dementia Patients at Risk for Falling. Nurs. Clin. N. Am. 2004, 39, 625–647. [CrossRef] [PubMed] [ ] [ ] 9. Hamers, J.P.; Huizing, A.R. Why Do We Use Physical Restraints in the Elderly? Z. Gerontol. Geriatr. 2005, 38, 19–25. [CrossRef] 10. Huizing, A.R.; Hamers, J.P.; de Jonge, J.; Candel, M.; Berger, M.P. Organisational Determinants of the Use of Physical Restraints: A Multilevel Approach. Soc. Sci. Med. 2007, 65, 924–933. [CrossRef] [ ] [ ] 9. Hamers, J.P.; Huizing, A.R. Why Do We Use Physical Restraints in the Elderly? Z. Gerontol. Geriatr. 2005, 38, 19–25. References A Cluster-Randomized Trial of an Educational Intervention to Reduce the Use of Physical Restraints with Psychogeriatric Nursing Home Residents. J. Am. Geriatr. Soc. 2009, 57, 1139–1148. [CrossRef] [PubMed] 31. Koczy, P.; Becker, C.; Rapp, K.; Klie, T.; Beische, D.; Büchele, G.; Kleiner, A.; Guerra, V.; Rissmann, U.; Kurrle, S.; et al. Effectiveness of a Multifactorial Intervention to Reduce Physical Restraints in Nursing Home Residents. J. Am. Geriatr. Soc. 2011, 59, 333–339. [CrossRef] [PubMed] 32. Köpke, S.; Mühlhauser, I.; Gerlach, A.; Haut, A.; Haastert, B.; Möhler, R.; Meyer, G. Effect of a Guideline-Based Multicomponent Intervention on Use of Physical Restraints in Nursing Homes: A Randomized Controlled Trial. JAMA 2012, 307, 2177–2184. [CrossRef] 11 of 11 Int. J. Environ. Res. Public Health 2021, 18, 11918 33. Morichi, V.; Fedecostante, M.; Morandi, A.; Di Santo, S.G.; Mazzone, A.; Mossello, E.; Bo, M.; Bianchetti, A.; Rozzini, R.; Zanetti, E.; et al. A Point Prevalence Study of Delirium in Italian Nursing Homes. Dement. Geriatr. Cogn. Disord. 2018, 46, 27–41. [CrossRef] 34. Muñiz, R.; Gómez, S.; Curto, D.; Hernández, R.; Marco, B.; García, P.; Tomás, J.F.; Olazarán, J. Reducing Physical Restraints in A f W lff d S A d i A 1 [C f] 33. Morichi, V.; Fedecostante, M.; Morandi, A.; Di Santo, S.G.; Mazzone, A.; Mossello, E.; Bo, M.; Bianchetti, A.; Rozzini, R.; Zanetti, E.; et al. A Point Prevalence Study of Delirium in Italian Nursing Homes. Dement. Geriatr. Cogn. Disord. 2018, 46, 27–41. [CrossRef] y g g 34. Muñiz, R.; Gómez, S.; Curto, D.; Hernández, R.; Marco, B.; García, P.; Tomás, J.F.; Olazarán, J. Reducing Physical Restraints in Nursing Homes: A Report from Maria Wolff and Sanitas. J. Am. Med. Dir. Assoc. 2016, 17, 633–639. [CrossRef] 34. Muñiz, R.; Gómez, S.; Curto, D.; Hernández, R.; Marco, B.; García, P.; Tomás, J.F.; Olazarán, J. Reduci Nursing Homes: A Report from Maria Wolff and Sanitas. J. Am. Med. Dir. Assoc. 2016, 17, 633–639. [Cr g p Pellfolk, T.J.; Gustafson, Y.; Bucht, G.; Karlsson, S. Effects of a Restraint Minimization Program on Staff K d P ti A Cl t R d i d T i l J A G i t S 2010 58 62 69 [C R f] [P bM d] g p 35. Pellfolk, T.J.; Gustafson, Y.; Bucht, G.; Karlsson, S. Effects of a Restraint Minimization Program on Staff Knowledge, Attitudes, and Practice: A Cluster Randomized Trial. J. Am. Geriatr. Soc. 43. Hofmann, H.; Hahn, S. Characteristics of Nursing Home Residents and Physical Restraint: A Systematic Literature Review. J. Clin. Nurs. 2014, 23, 3012–3024. [CrossRef] [PubMed] References 2010, 58, 62–69. [CrossRef] [PubMed] 35. Pellfolk, T.J.; Gustafson, Y.; Bucht, G.; Karlsson, S. Effects of a Restraint Minimization Program on Staff Knowledge, Attitudes, and Practice: A Cluster Randomized Trial. J. Am. Geriatr. Soc. 2010, 58, 62–69. [CrossRef] [PubMed] , J ; , ; , ; , g and Practice: A Cluster Randomized Trial. J. Am. Geriatr. Soc. 2010, 58, 62–69. [CrossRef] [PubMed] 36. Testad, I.; Ballard, C.; Brønnick, K.; Aarsland, D. The Effect of Staff Training on Agitation and Use of Restraint in Nursing Home Residents with Dementia: A Single-Blind, Randomized Controlled Trial. J. Clin. Psychiatry 2010, 71, 80–86. [CrossRef] 37. Testad, I.; Mekki, T.E.; Førland, O.; Øye, C.; Tveit, E.M.; Jacobsen, F.; Kirkevold, Ø. Modeling and Evaluating Evidence-Based Continuing Education Program in Nursing Home Dementia Care (Medced)—Training of Care Home Staff to Reduce Use of Restraint in Care Home Residents with Dementia. A Cluster Randomized Controlled Trial. Int. J. Geriatr. Psychiatry 2016, 31, 24–32. [CrossRef] 38. Verbeek, H.; Zwakhalen, S.M.; van Rossum, E.; Ambergen, T.; Kempen, G.I.; Hamers, J.P. Effects of Small-Scale, Home-Like Facilities in Dementia Care on Residents’ Behavior, and Use of Physical Restraints and Psychotropic Drugs: A Quasi-Experimental Study. Int. Psychogeriatr. 2014, 26, 657–668. [CrossRef] y 39. Willemse, B.M.; de Jonge, J.; Smit, D.; Dasselaar, W.; Depla, M.F.; Pot, A.M. Is an Unhealthy Work Environment in Nursing Home Care for People with Dementia Associated with the Prescription of Psychotropic Drugs and Physical Restraints? Int. Psychogeriatr. 2016, 28, 983–994. [CrossRef] [PubMed] 40. Kuronen, M.; Kautiainen, H.; Karppi, P.; Hartikainen, S.; Koponen, H. Physical Restraints and Associations with Neuropsychiatric Symptoms and Personal Characteristics in Residential Care: A Cross-Sectional Study. Int. J. Geriatr. Psychiatry 2017, 32, 1418–1424. [CrossRef] [PubMed] 41. (RNAO), The Registered Nurses’ Association of Ontario. Promoting Safety: Alternative Approaches to the Use of Restraints; RNAO: Toronto, ON, Canada, 2012. 42. Øye, C.; Jacobsen, F.F.; Mekki, T.E. Do Organisational Constraints Explain the Use of Restraint? A Comparative Ethnographic Study from Three Nursing Homes in Norway. J. Clin. Nurs. 2017, 26, 1906–1916. [CrossRef] [PubMed] 43. Hofmann, H.; Hahn, S. Characteristics of Nursing Home Residents and Physical Restraint: A Systematic Literature Review. J. Clin. Nurs. 2014, 23, 3012–3024. [CrossRef] [PubMed]
https://openalex.org/W4233653239
https://figshare.com/articles/thesis/Examining_components_of_decision_making_in_obsessive-compulsive_disorder/14644500/1/files/28123410.pdf
English
null
Examining components of decision making in obsessive-compulsive disorder
null
2,023
cc-by
25,834
Running head: DECISION MAKING IN OCD Running head: DECISION MAKING IN OCD AUTHOR'S DECLARATION FOR ELECTRONIC SUBMISSION OF A THESI I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, includin any required final revisions, as accepted by my examiners. I authorize Ryerson University to lend this thesis to other institutions or individuals for the purpose of scholarly research. I further authorize Ryerson University to reproduce this thesis by photocopying or by other means, in total or in part, at the request of other institutions or individuals for the purpose of scholarly research. I understand that my thesis may be made electronically available to the public. ii ii DECISION MAKING IN OCD Examining components of decision making in obsessive-compulsive disorder Master of Arts 2015 Jeanine Lane BHSc BA Abstract The purpose of this study was to examine decision making in individuals with OCD as compared to healthy controls. To the author's knowledge, this was the first study to evaluate decision making in OCD using multiple neuropsychological measures to assess proposed components of the construct, including ambiguity, impulsivity, risk perception, and organizational strategies. It was also believed to be the first study to compare subjective and objective decision making in OCD. Subjective decision making showed poor convergent validity with the three objective measures, as only self-reported indecision was significantly different between groups. The purpose of this study was to examine decision making in individuals with OCD as compared to healthy controls. To the author's knowledge, this was the first study to evaluate decision making in OCD using multiple neuropsychological measures to assess proposed components of the construct, including ambiguity, impulsivity, risk perception, and organizational strategies. It was also believed to be the first study to compare subjective and objective decision making in OCD. Subjective decision making showed poor convergent validity with the three objective measures, as only self-reported indecision was significantly different between groups. Performance on a memory task, as a proxy for organizational strategies, indicated differences between groups on strategic learning and the effect of interference. Although the presence of decision making deficits remains inconclusive, this study provided support for the conclusion that individuals with OCD have perceived difficulty with decision making, but not necessarily objective deficits. Treatment implications are discussed. Performance on a memory task, as a proxy for organizational strategies, indicated differences between groups on strategic learning and the effect of interference. Although the presence of decision making deficits remains inconclusive, this study provided support for the conclusion that individuals with OCD have perceived difficulty with decision making, but not necessarily objective deficits. Treatment implications are discussed. Keywords: obsessive-compulsive disorder, decision making, subjective decision making, ambiguity, impulsivity, perception of risk, organizational strategies iii Acknowledgements First, I would like to thank my supervisor, Dr. Tisha Ornstein, for her ongoing, gracious support and insight throughout this study. I would also like to extend sincere gratitude to Dr. Kristin Vickers for her enduring assistance and dedication to the study, and to Dr. Martin Antony for agreeing to serve on my examination committee. I am honoured to have worked with such knowledgeable psychologists. Thank you to Stephanie Taillefer, who was not only a partner on this journey, but also a mentor and friend. I would also like to thank Laurence Lai and Kayla Edwards for all their hard work as research assistants on this project. Finally, thank you to my fiancé, Robert Sales and my parents, George and Marjorie Lane, for their endless support, patience, guidance, and love. To all my other family members and friends, thank you for making this journey worthwhile. iv DECISION MAKING IN OCD Dedication Dedication This thesis is dedicated to my parents, George and Marjorie Lane, who have taught me so much about life, have spent countless hours helping me with anything and everything, and who have made all of my education and success possible. Your never ending love and unwavering support is an aspiration I hope to fulfill for my own family one day. v v DECISION MAKING IN OCD Table of Contents List of Tables ..............................................................................................................................vii Introduction .................................................................................................................................1 Hypotheses ..................................................................................................................................18 Method ........................................................................................................................................19 Results .........................................................................................................................................34 Discussion ...................................................................................................................................48 References ...................................................................................................................................65 List of Tables ..............................................................................................................................vii Introduction .................................................................................................................................1 Hypotheses ..................................................................................................................................18 Method ........................................................................................................................................19 Results .........................................................................................................................................34 Discussion ...................................................................................................................................48 References ...................................................................................................................................65 vi List of Tables Table 1: Summary of Studies Examining Decision Making in OCD ....................................6 Table 2: Mean Demographic Characteristics Between Groups .............................................35 Table 3: Mean Clinical Information Scores Between Groups ...............................................36 Table 4: Mean Decision Making Performance Between Groups ..........................................39 Table 5: Mean Performance on Memory Task Between Groups ..........................................40 Table 6: Correlations (r) Among Decision Making Measures ..............................................45 Table 7: Other Neuropsychological Measures ......................................................................47 vii DECISION MAKING IN OCD Examining Components of Decision Making in Obsessive-compulsive Disorder Obsessive-compulsive disorder (OCD) is a chronic and disabling condition with a lifetime prevalence of approximately 2.3% among Canadian adults (Fogel, 2003; Stein, Forde, Anderson, & Walker, 1997). OCD is characterized by obsessions (unwanted, repetitive and intrusive thoughts, images or impulses that cause marked anxiety or distress) or compulsions (repetitive behaviour or mental acts intended to prevent, reduce or neutralize unbearable anxiety and distress (American Psychiatric Association [APA], 2013). Common obsessions include thoughts about being contaminated by germs or dirt, persistent doubts regarding one's actions or behaviours, the need for symmetry or exactness, and aggressive, sexual and disturbing thoughts or images (APA, 2013; Olley, Malhi, & Sachdev, 2007; Thobaben, 2012). Associated compulsions often include excessive hand washing and cleaning, repetitive checking behaviours, ordering and arranging, excessive reassurance seeking, and mental acts such as praying and counting, which the person feels driven to perform (APA, 2013; Olley et al., 2007; Thobaben, 2012). Obsessions and compulsions often co-occur in those with OCD (Thobaben, 2012). The disturbing thoughts and time-consuming behaviours in OCD have a profound negative impact on quality of life, and interfere considerably with academic, occupational and social functioning (APA, 2013). In a study conducted by Sorensen, Kirkeby and Thomsen (2004), 57% of participants with OCD enrolled in school reported that their symptoms impacted their ability to attend school (n = 70), and of those who had a job, 53% reported an impact on their ability to attend work (n = 115). Of the 219 individuals with OCD in the study, 72% reported that their symptoms affected their social and everyday life functioning, and 27% indicated that they were dissatisfied with their quality of life (Sorensen et al., 2004). Recent evidence suggests that quality of life may be even poorer in those with comorbid psychological 1 1 DECISION MAKING IN OCD disorders (Saleem, Gul, & Khalily, 2012). OCD is commonly comorbid with sleep difficulties, depression, anxiety disorders, eating disorders, and personality disorders (APA, 2013). Evidence also suggests that individuals with OCD may exhibit deficits in cognitive functioning (Dettore & O'Connor, 2013; Shin, Lee, Kim & Kwon, 2014), which likely contributes to the difficulties in day-to-day functioning. Thus, it is not surprising that the World Health Organization (WHO; 2001) reported OCD as one of the ten leading causes of medical disability worldwide. Decision Making Defined Decision making is defined as the action or process of making decisions, and involves executive functioning and the ability to evaluate environmental information (Dittrich & Johansen, 2013). Choosing from a set of options is a fundamental part of day-to-day activities. In healthy populations this process occurs over time, while the individual searches for relevant information and then evaluates alternatives in the context of his or her social environment (Raynard, Crozier, & Svenson, 1997). There is also a postdecision process during which individuals further evaluate their decisions through feedback and the resulting outcome. When examining decision making in clinical populations, it is important to highlight the process of arriving at a decision, rather than the decision itself (Raynard et al., 1997). Overall findings comparing decision making outcomes between individuals with OCD and healthy controls are mixed; therefore, examining the process that leads to outcomes, and the many important variables involved in this process, is important. Examining decision making using various measures can provide insight into these deficits. Theory Underlying the Neurobiology of OCD Deficits in cognitive functions, particularly executive control processes (i.e., decision making, response inhibition, cognitive flexibility and planning) have been identified in OCD (Cavedini, Gorini, & Bellodi, 2006; Olley et al., 2007; Tukel et al., 2012). Evidence from neuroimaging studies supported the involvement of fronto-striato-thalamic circuits, which mediate these cognitive processes (Cavedini et al., 2006; Ping et al., 2013; Menzies, Chamberlain, Laird, Thelen, Sahakian, & Bullmore, 2008; Whiteside, Port & Abramowitz, 2004). Specifically, differences in neuronal activity and functional connectivity have been reported in the orbitofrontal cortex in OCD, as compared to healthy controls (Ping et al., 2013; Stern, Welsh, Gonzalez, Fitzgerald, Abelson, & Taylor, 2013; Whiteside et al., 2004). The orbitofrontal cortex is an important structure for the process of making decisions in terms of evaluating reward and emotion in the context of various outcomes. Evidence from neuropsychological studies has corroborated neuroimaging research in that the orbitofrontal cortex is involved in the pathophysiology of OCD. This was done by demonstrating impaired performance on tests that are known to be sensitive to deficits in the orbitofrontal cortex (Cavedini et al., 2006, Menzies et al., 2008; Shin et al., 2014). Thus, aberrant structural changes appear related to the presentation of cognitive dysfunction. 2 2 Decision Making in OCD The literature on OCD is largely focused on deficits in executive functioning more generally, and examining possible endophenotypes (i.e., a term used to describe symptoms of behaviour that may be closer to the underlying pathophysiology of the disorder than the clinical phenomena; Rao, Reddy, Kumar, Kandavel, & Chandrashekar, 2008). There is little published research on decision making in this population, despite the suggestion that it is a possible endophenotype of OCD (Olley et al., 2007; Vandenbroucke & Gabriels, 2012). Further identifying the decision-making process may enhance the understanding of the pathophysiology of OCD, given its potential as an endophenotype. Importantly, decision making has been 3 3 DECISION MAKING IN OCD reported as a potential predictor of treatment outcomes (Cavedini et al., 2006; Olley et al., 2007; Vandenbroucke & Gabriels, 2012). A better understanding of decision making in OCD could provide diagnostic information to add to the case conceptualization of individuals with this disorder, and has the potential to enhance the development of more specific and effective treatment interventions. Although some findings have highlighted possible deficits in decision making in OCD (e.g., time required to make a decision may be increased; Dittrich & Johansen, 2013; Stern et al., 2013), the nature of these deficits remains unclear (Burdick, Robinson, Malhotra, & Szeszko, 2008; Vandenbroucke & Gabriels, 2012). Few studies have explored the neurocognition of decision making in OCD (Cavedini et al., 2006; Dittrich & Johansen, 2013; Olley et al., 2007). For a summary of these studies, see Table 1. Of those published studies, some demonstrated impairments in decision making in adults with OCD compared to healthy control participants (Cavallaro et al., 2003; Cavedini, Riboldi, D'Annucci, Belotti, Cisima, & Bellodi, 2002; Da Rocha, Alvarenga, Malloy-Diniz, & Correa, 2011; Long, Kai, Yi-Fu, Chun-Yan, Ping and Hui-Juan, 2012; Starcke, Tuschen-Caffier, Markowitsch, & Brand, 2010). Other studies, similarly using the highly popular Iowa Gambling Task (IGT; Bechara, Damasio, Damasio, & Anderson, 1994) as a measure of decision making, failed to find such differences between OCD and control participants (Dittrich, Johansen, Landro, & Fineberg, 2011; Lawrence, Wooderson, Mataix-Cols, David, & Speckens, 2006; Nielen, Veltman, de Jong, Mulder, & den Boer, 2002). These mixed results may be due to various factors. For example, inconsistent findings may be due to the fact that the IGT taps multiple components of decision making (Buelow & Suhr, 2009). Studies to date have largely relied on one or two tasks to examine decision making, and there is also a lack of research examining the process using multimodal methods (e.g., 4 DECISION MAKING IN OCD subjective and objective measures). Also, differences in sex may be a confounding variable, as males tend to perform better than females on the IGT (Overman et al., 2006). Medication status is another important factor to consider, as those who do not respond to medication tend to exhibit poorer performance on decision-making tasks (Cavedini et al., 2002; Chamberlain, Fineberg, Blackwell, Clark, Robbins, & Sahakian, 2007). In addition, severity of symptoms has been suggested as important when examining decision making, as those with more severe symptoms tend to have poorer performance on the IGT (Nielen et al., 2002). Inconsistent results may be due to the clinical heterogeneity of OCD, as previous research has revealed symptom subtypes (e.g., primary symptoms of In addition, severity of symptoms has been suggested as important when examining decision making, as those with more severe symptoms tend to have poorer performance on the IGT (Nielen et al., 2002). Inconsistent results may be due to the clinical heterogeneity of OCD, as previous research has revealed symptom subtypes (e.g., primary symptoms of contamination/cleaning, symmetry/ordering, obsessions/checking and hoarding) and the preliminary association to certain neurocognitive impairments (Leckman et al., 1997; McKay et al., 2004). Importantly, previous research may have included participants with primary hoarding phenomena (with and without obsessive-compulsive symptoms), which is now considered its own diagnostic category in DSM-5 (APA, 2013). Decision making deficits are considered to be a core feature of hoarding (Lawrence et al., 2006; Steketee & Frost, 2003); thus, poor performance on decision-making tasks in previous studies may be better accounted for by the hoarding disorder rather than OCD. Finally, sample size may be another contributing factor to inconsistent results, as previous studies have varied in their sample sizes. Overall, there is a paucity of published work that has evaluated the various features that encompass the complex process of decision making in OCD, with largely inconsistent results. 5 5 DECISION MAKING IN OCD Summary of studies examining decision making in OCD Study Primary aim Sample size Age, M (SD) % Male YBOCS, M (SD) Medication Status in OCD Decision- making Measures Result OCD C OCD C OCD C Cavallaro et al. (2003)1 Examined the link of prefrontal cortex dysfunction to neuroanatomic systems in OCD and schizophrenia patients 67 56 30.50 (8.90) 31.20 (6.00) 49 39 N/A Medication free for at least 2 weeks IGT p < .001*** Cavendini et al. DECISION MAKING IN OCD (2002)2 Examined OFC function in OCD and panic disorder patients and whether this predicted response to medication 34 34 33.70 (11.50) 29.50 (8.90) 53 44 28.30 (5.30) and 28.70 (5.90) for the two OCD groups Medication free for at least 2 weeks IGT p= .004** Chamberlain, Fineberg, Blackwell, et al. (2007)3 Examined neurocognitive functioning in trichotillomania and OCD 20 20 35.30 (14.07) 32.15 (7.70) N/A N/A 20.40 (4.07) 80% on SSRIs IST CGT p= ns p= ns Cavendini et al. (2002)2 Examined OFC function in OCD and panic disorder patients and whether this predicted response to medication 34 34 33.70 (11.50) 29.50 (8.90) 53 44 28.30 (5.30) and 28.70 (5.90) for the two OCD groups Medication free for at least 2 weeks IGT p= .004** Chamberlain, Fineberg, Blackwell, et al. (2007)3 Examined neurocognitive functioning in trichotillomania and OCD 20 20 35.30 (14.07) 32.15 (7.70) N/A N/A 20.40 (4.07) 80% on SSRIs IST CGT p= ns p= ns 6 6 DECISION MAKING IN OCD Chamberlain, Fineberg, Menzies, et al. (2007)4 Examined genetic contributions to inhibitory control processes in OCD patients and unaffected relatives 20 20 32.10 (11.90) 33.10 (10.50) 20 35 22.70 (6.60) 90% on SSRIs CGT p= ns Da Rocha et al. (2011) Examined decision making in OCD patients 107 107 28.40 (14.12) 29.33 (13.22) 54 52 26.61 (7.70) 78% on medication, 52% of those on SSRIs IGT p < .01 Dittrich & Johansen (2013) Examined cognitive deficits in OCD patients 13 13 43.20 (9.50) 36.90 (10.20) 30 39 16.20 (6.80) 100% on SSRIs CGT p = .049* Dittrich et al. (2011) Examined risky and ambiguous decision making in OCD by symptom dimension 72 66 41.80 (12.50) 37.60 (14.40) 39 32 Ranged from 15.80 (3.60) for sexual/ religious symptoms to 23.30 (8.20) for aggressive symptoms 90% on SSRIs IGT CGT p = ns p = .004** - .059 depending on variable selected Da Rocha et al. (2011) Examined decision making in OCD patients 107 107 28.40 (14.12) 29.33 (13.22) 54 52 26.61 (7.70) 78% on medication, 52% of those on SSRIs IGT p < .01 Dittrich & Johansen (2013) Examined cognitive deficits in OCD patients 13 13 43.20 (9.50) 36.90 (10.20) 30 39 16.20 (6.80) 100% on SSRIs CGT p = .049* Dittrich et al. DECISION MAKING IN OCD (2011) Examined risky and ambiguous decision making in OCD by symptom dimension 72 66 41.80 (12.50) 37.60 (14.40) 39 32 Ranged from 15.80 (3.60) for sexual/ religious symptoms to 23.30 (8.20) for aggressive symptoms 90% on SSRIs IGT CGT p = ns p = .004** - .059 depending on variable selected 7 DECISION MAKING IN OCD Foa et al. (2003) Examined amount of information, time and risk in decision making in OCD patients 18 18 35.60 (12.10) 28.50 (8.90) 39 44 26.77 (5.24) N/A Decision- making scenarios of low and high risk p =ns for high risk p <.01** Lawrence et al. (2006) Examined OCD subtypes and decision making 39 40 36.10 (11.07) 33.48 (10.40) 51 50 25.13 (6.86) 77% on medication IGT p =ns Long et al. (2012) Examined ambiguity and risk in OCD patients 45 48 N/A N/A N/A N/A N/A N/A IGT GDT p = .023* p =ns Nielen et al. (2002) Examined decision making in OCD patients 27 26 34.90 (9.90) 31.20 (8.30) 26 31 25.10 (5.20) Medication free for at least 4 weeks IGT p =ns Starcke et al. (2009) Examined decision making and SCRs in ambiguous and risky situations in OCD patients 14 15 36.36 (8.54) 36.60 (10.84) 50 67 N/A 64% on SSRIs. 33% of those on additional medication IGT GDT p < .001*** p = ns Starcke et al. (2010) Examined decision making under ambiguity and risk in OCD patients 23 22 35.25 (7.35) 36.50 (10.23) 56 55 N/A 61% on SSRIs 43% of those on additional medication IGT GDT p < .001*** p = ns Foa et al. (2003) Examined amount of information, time and risk in decision making in OCD patients 18 18 35.60 (12.10) 28.50 (8.90) 39 44 26.77 (5.24) N/A Decision- making scenarios of low and high risk p =ns for high risk p <.01** Lawrence et al. (2006) Examined OCD subtypes and decision making 39 40 36.10 (11.07) 33.48 (10.40) 51 50 25.13 (6.86) 77% on medication IGT p =ns Long et al. (2012) Examined ambiguity and risk in OCD patients 45 48 N/A N/A N/A N/A N/A N/A IGT GDT p = .023* p =ns Nielen et al. (2002) Examined decision making in OCD patients 27 26 34.90 (9.90) 31.20 (8.30) 26 31 25.10 (5.20) Medication free for at least 4 weeks IGT p =ns Starcke et al. DECISION MAKING IN OCD (2009) Examined decision making and SCRs in ambiguous and risky situations in OCD patients 14 15 36.36 (8.54) 36.60 (10.84) 50 67 N/A 64% on SSRIs. 33% of those on additional medication IGT GDT p < .001*** p = ns Starcke et al. (2010) Examined decision making under ambiguity and risk in OCD patients 23 22 35.25 (7.35) 36.50 (10.23) 56 55 N/A 61% on SSRIs 43% of those on additional medication IGT GDT p < .001*** p = ns ( ) g and SCRs in ambiguous and risky situations in OCD patients ( ) ( ) 33% of those on additional medication p Starcke et al. (2010) Examined decision making under ambiguity and risk in OCD patients 23 22 35.25 (7.35) 36.50 (10.23) 56 55 N/A 61% on SSRIs 43% of those on additional medication IGT GDT p < .001*** p = ns 8 8 DECISION MAKING IN OCD Stern et al. (2013) Examined decision making and level of certainty during fMRI's in OCD patients 25 29 25.06 (6.04) 27.11 (8.42) 48 41 Medication free group: 27.70 (4.90), Medication group: 20.10 (4.10) 64% medicated with SSRIs Decision- making task measuring uncertainty (see author for description) Decisions with no objective uncertainty: p < .001*** Decisions with objective uncertainty: p =ns Watkins et al. (2005)5 Examined cognitive performance in Tourette's and OCD patients 20 20 38.20 (13.40) 36.60 (12.80) 40 45 19.30 (14.30) 40% on SSRIs. 10% on other medication CGT p = ns Note. C = Controls; M = Mean; SD = Standard Deviation; IGT = Iowa Gambling Task; GDT = Game of Dice Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; YBOCS = Yale-Brown Obsessive-Compulsive Inventory (as a measure of symptom severity); SSRIs = Selective Serotonin Reuptake Inhibitors; OFC = Orbitofrontal cortex; SCRs = Skin Conductance Responses; fMRI = Functional Magnetic Resonance Imaging; *p <.05; **p < .01;*** p < .001, indicating significant differences between OCD and control participants on decision-making measures, with poorer performance in the OCD group than in controls. Note. 5. Watkins et al., 2005: Included 20 Tourette's Syndrome patients (n= 60). Chamberlain, Fineberg, Menzies, et al. (2007): Included 20 first degree relatives of OCD p p p mberlain, Fineberg, Blackwell, et al. (2007): Included 20 trichotillomania patients (n= 60) p p Cavendini et al. (2002): Included 16 panic disorder patients (n= 84). Components of Decision Making Although decision making may be generally thought of as a unitary construct, there are various components of decision making that contribute to this cognitive process. Previous studies appear to have largely focused on ambiguous decision making (i.e., results and probabilities are unknown; Clark, Bechara, Damasio, Aitken, Sahakian, & Robbins, 2008), with very few studies examining other components of decision making. Examples of other components of decision making include impulsivity, the perception of risk (i.e., uncertain results with known Although decision making may be generally thought of as a unitary construct, there are various components of decision making that contribute to this cognitive process. Previous studies appear to have largely focused on ambiguous decision making (i.e., results and probabilities are unknown; Clark, Bechara, Damasio, Aitken, Sahakian, & Robbins, 2008), with very few studies examining other components of decision making. Examples of other components of decision making include impulsivity, the perception of risk (i.e., uncertain results with known probabilities; Clark et al., 2008), and organizational strategies (Dittrich & Johansen, 2013; Foa et al., 2002; Olley et al., 2007; Starcke, Tuschen-Caffier, Markowitsch, & Brand, 2009; Stern et al., 2013). The limited body of literature supported further investigation of the nature and pattern of decision-making performance in OCD. probabilities; Clark et al., 2008), and organizational strategies (Dittrich & Johansen, 2013; Foa et al., 2002; Olley et al., 2007; Starcke, Tuschen-Caffier, Markowitsch, & Brand, 2009; Stern et al., 2013). The limited body of literature supported further investigation of the nature and pattern of decision-making performance in OCD. Ambiguity. The commonly used IGT is largely studied as a decision-making task that examines ambiguous decision making. However, the construct validity of this measure has been criticized for not clearly identifying which component of decision making the test is attempting to capture (Buelow & Suhr, 2009). It appears that early on in the task, decisions are based on ambiguous information, whereas later in the task, it measures risky decision making (Brand, Fujiwara, Borsutzky, Kalbe, Kessler, & Markowitsch, 2005; Brand, Recknor, Grabenhorst, & Bechara, 2007). However, the point during the task when the concept switches is not clear or identified (Brand et al., 2007). This is problematic as studies that use this task as a measure of ambiguous decision making compare it to tasks that measure risky decision making (Chamberlain, Fineberg, Blackwell, et al., 2007). Dittrich et al., 2011; Long et al., 2012; Starcke et al., 2009; 2010). DECISION MAKING IN OCD C = Controls; M = Mean; SD = Standard Deviation; IGT = Iowa Gambling Task; GDT = Game of Dice Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; YBOCS = Yale-Brown Obsessive-Compulsive Inventory (as a measure of symptom severity); SSRIs = Selective Serotonin Reuptake Inhibitors; OFC = Orbitofrontal cortex; SCRs = Skin Conductance Responses; fMRI = Functional Magnetic Resonance Imaging; *p <.05; **p < .01;*** p < .001, indicating significant differences between OCD and control participants on decision-making measures, with poorer performance in the OCD group than in controls. Some studies included clinical control groups, which were not described above. 1. Cavallaro et al. (2003): Included 110 schizophrenic patients (n= 233). 2 2. Cavendini et al. (2002): Included 16 panic disorder patients (n= 84). 3 3. Chamberlain, Fineberg, Blackwell, et al. (2007): Included 20 trichotillomania pati g p 4. Chamberlain, Fineberg, Menzies, et al. (2007): Included 20 first degree relatives o 5 9 Components of Decision Making Further to this, the authors of the measure also claim that performance on the IGT requires the use of organizational strategies (Bechara et al., 1994), another proposed (Chamberlain, Fineberg, Blackwell, et al., 2007). Dittrich et al., 2011; Long et al., 2012; Starcke et al., 2009; 2010). Further to this, the authors of the measure also claim that performance on the IGT requires the use of organizational strategies (Bechara et al., 1994), another proposed 10 DECISION MAKING IN OCD of uncertainty was an important cognitive process in OCD, as scores measuring this construct predicted unique variance beyond that of other factors, such as perfectionism. of uncertainty was an important cognitive process in OCD, as scores measuring this construct predicted unique variance beyond that of other factors, such as perfectionism. When making decisions, a greater subjective experience of uncertainty has been reported in OCD as compared to controls (Stern et al., 2013). This study also reported that compared to controls, patients with OCD only had increased difficulty on tasks with subjective uncertainty (i.e., no objective uncertainty inherent in the task) and that there was no differences on tasks that included objective uncertainty (Stern et al., 2013). It is well known that the IGT is a task of ambiguity, and it is possible that scores on this measure, and other measures of decision making, may be related to subjective reports of intolerance of uncertainty. Impulsivity. It is also suggested that impulsivity is associated with the process of decision making (Bayard, Raffard, & Gely-Nargeot, 2011). Impulsive behaviours may involve the inability to forgo immediate reward in favour of long term gain, or the inability to use information to predict possible consequences of behaviour (Torregrossa, Quinn, & Taylor, 2008). In real-life decision making, individuals must consider immediate as well as delayed reward and punishment, all in the context of risk and uncertainty (Bechara et al., 1994). Impulsivity. It is also suggested that impulsivity is associated with the process of decision making (Bayard, Raffard, & Gely-Nargeot, 2011). Impulsive behaviours may involve the inability to forgo immediate reward in favour of long term gain, or the inability to use information to predict possible consequences of behaviour (Torregrossa, Quinn, & Taylor, 2008). In real-life decision making, individuals must consider immediate as well as delayed reward and punishment, all in the context of risk and uncertainty (Bechara et al., 1994). Impulsivity can be measured via self-report questionnaires, such as the Barratt Impulsiveness Scale (BIS-II; Patton, Stanford, & Barratt, 1995). Using the BIS-II, previous research found that individuals with OCD reported higher impulsivity than controls (Summerfeldt, Hood, Antony, Richter, & Swinson, 2004; Gupta, Khess, & Venkatesan, 2013). Previous research also found that high impulsivity resulted in poorer performance on the IGT, as compared to individuals with low impulsivity (Franken, van Strien, Nijs, & Muris, 2008). DECISION MAKING IN OCD component of decision making, yet there are no outcome variables to measure this. Although the current study considered the IGT to measure ambiguous decision making, the poor construct validity should be noted, and likely contributed to the previously inconsistent results. Despite this, the IGT was included in the current study to compare to previous research, as it is a commonly used measure of decision making. The presence of ambiguity as a component of decision making may be related to intolerance of uncertainty. Intolerance of uncertainty can be defined as a habitual negative reaction to uncertain situations or events (Dugas & Koerner, 2005). The self-report Intolerance of Uncertainty Scale (IUS; Buhr & Dugas, 2002) is the most commonly used method to assess this construct. Individuals who are intolerant of uncertainty may exhibit extreme emotional, cognitive or behavioural responses when faced with uncertainty (Buhr & Dugas, 2002), and thus try to avoid ambiguous situations. Some research has suggested that a high intolerance of uncertainty, in the context of ambiguous situations, underlies difficulties with decision making, as evidenced by compulsive behaviours to gather information to reduce uncertainty (Stern et al., 2013). Intolerance of uncertainty has been studied in OCD, although to a lesser extent than that seen in other anxiety disorders, such as generalized anxiety disorder (GAD). For example, Gentes and Ruscio (2011) reported that across multiple studies, individuals with OCD reported levels of intolerance of uncertainty similar to those with GAD. Intolerance of uncertainty was also shown to be significantly related to symptom severity among those with specific symptoms of OCD, such as checking (Tolin, Abramowitz, Brigidi, & Foa, 2003) and hoarding (Oglesby, Medley, Norr, Capron, Korte, & Schmidt, 2013). Fergus and Wu (2010) reported that intolerance Medley, Norr, Capron, Korte, & Schmidt, 2013). Fergus and Wu (2010) reported that intolerance 11 11 DECISION MAKING IN OCD Impulsivity can be measured via self-report questionnaires, such as the Barratt Impulsiveness Scale (BIS-II; Patton, Stanford, & Barratt, 1995). Using the BIS-II, previous research found that individuals with OCD reported higher impulsivity than controls (Summerfeldt, Hood, Antony, Richter, & Swinson, 2004; Gupta, Khess, & Venkatesan, 2013). Previous research also found that high impulsivity resulted in poorer performance on the IGT, as compared to individuals with low impulsivity (Franken, van Strien, Nijs, & Muris, 2008). The Information Sampling Task (IST; Clark, Robbins, Ersche, & Sahakian, 2006) has been previously reported as an objective measure that evaluates impulsivity in relation to 12 DECISION MAKING IN OCD decision making by evaluating the tendency to gather information before making a decision. The IST has been previously used to examine cognitive flexibility (which is another component of executive functioning) in OCD, and no differences between patients and controls were reported (Fadda, Sottocorno, Martoni, Galimberti, & Bellodi, 2011). However, it appears that only one study has examined the tendency to gather and evaluate information before making a decision on the IST in OCD (Chamberlain, Fineberg, Blackwell, et al., 2007). This study found a significant difference between individuals with OCD and control participants on the mean number of boxes opened before making a decision in the first condition (p = .041), but this result was no longer significant after corrections for multiples analyses were employed. No differences between groups were found on the same dependent variable in the second condition, even before corrections were made (p = .090). Interestingly, in this study, participants with OCD did not adapt their behaviour between the two conditions (p =.064), but the control group did (p < .001). This result was suggested as indicative of a lack of cognitive flexibility (or cognitive rigidity). The finding suggested possible differences between groups on the task and thus future research is There are also components of self-reported impulsivity that need to be considered in the context of decision making. These include urgency, which refers to acting impulsively when experiencing negative affect; lack of premeditation, which involves not considering potential outcomes of behaviour; lack of perseverance, which refers to difficulty with continued focus on boring or challenging tasks; and sensation seeking, which refers to openness to new, exciting, potentially dangerous experiences (Bayard et al., 2011; Gupta et al., 2013; Zermatten & Van der 13 13 DECISION MAKING IN OCD Linden, 2008). In healthy undergraduate students, only a lack of premeditation was moderately associated with poorer performance on the IGT (Zermatten, Van der Linden, d'Acremont, Jermann, & Bechara, 2005). In a more recent follow-up study, higher levels of urgency were associated with poor IGT performance (Billieux, Gay, Rochat, & Van der Linden, 2010). There also appeared to be a strong association between symptoms of OCD and impulsive urgency (Zermatten & Van der Linden, 2008). The foundation of urgency is based on difficulties inhibiting responses. Response inhibition can be measured using tasks such as the Stop-Signal Paradigm (Logan, Cowan, & Davis, 1984) and the Stroop Color-Word Interference Test (Golden, 1976). Poor inhibition has been related to self-reported impulsivity (Bayard et al., 2011). As summarized by Menzies and colleagues (2008), consistent results indicated deficits in response inhibition in OCD, which provided further evidence that impulsivity may be a factor influencing decision making. Therefore, it is possible that even if impulsivity is low, individuals with OCD may still demonstrate difficulty in decision making due to an inability to inhibit choices based on compulsive symptoms (Zermatten & Van der Linden, 2008). Differences in the results of studies measuring decision making using the IGT may be related to the influence of impulsivity. Thus, examining the association between impulsivity and decision making is warranted, and inhibition as a factor contributing to differences between OCD and control participants should be examined. Using the BIS-II, Gupta and colleagues (2013) reported that patients with OCD had higher attention-impulsiveness (similar to a lack of perseverance), and higher nonplanning impulsiveness (similar to a lack of premeditation). These authors concluded that high impulsivity in OCD may be a reason for poor performance on the Cambridge Gambling Task (CGT; Rogers et al., 1999), which is a popular measure of risky decision making (with known probabilities). As 14 14 DECISION MAKING IN OCD mentioned, ambiguity and risk perception are also components of decision making. Bayard and colleagues (2011) concluded that decision making under ambiguous situations (as in the IGT) is not likely impacted by impulsivity, but that decision-making performance in situations of risk, is predicted by sensation seeking and urgency. Thus, it appears that further examination of these components is needed. Risk taking. Research examining the perception of risk in decision making has also compared low and high-risk situations (Foa et al., 2003). This study reported that patients with OCD demonstrated significant differences compared to healthy controls in decision making, but only in low risk situations. For example, OCD patients required more information and more time than controls before making a decision in low risk situations (Foa et al., 2003). Recent studies by Dittrich and Johansen (2013), and Dittrich and colleagues (2011) found significant differences between participants with OCD and controls using the CGT, whereas the majority of other studies found intact decision making in OCD under risky situations (Chamberlain, Fineberg, Blackwell, et al., 2007; Chamberlain, Fineberg, Menzies, et al., 2007; Walkins et al., 2005). Two studies (Long et al., 2012; Starcke et al., 2009) also reported no differences in decision making using the Game of Dice (Brand et al., 2005), which is similar to the CGT in that it measures risky decision making with known probabilities. However, both studies reported significant differences between participants with OCD and controls under ambiguous decision-making situations using the IGT. Another study reported that during risky decision making, high levels of sensation seeking and urgency predicted poorer outcomes on the Game of Dice (Bayard et al., 2011). These studies provide further support that examining decision making using multiple measures to tap different components is important. Although results are inconsistent, it appears 15 15 DECISION MAKING IN OCD performance on the CVLT-II functioned as a proxy measure for this important component of decision making. Although organizational strategies have been indicated as an important part of decision making (Olley et al., 2007), no studies to date have evaluated the relationship of strategic learning deficits and decision making. In summary, the existing literature evaluating decision making in OCD is largely mixed. This is likely due to various factors, including the fact that multiple components of decision making are rarely considered. The decision-making process has been shown to be negatively affected in individuals with OCD, but more research is needed to determine which components of the process are specifically impaired and under what circumstances (Starcke et al., 2009; Olley et al., 2007). Furthermore, limited published work has examined other possible factors, such as intolerance of uncertainty and impulsivity, which may contribute to the construct of decision making. To the author's knowledge, there are no published studies evaluating decision- making performance in the context of all of the aforementioned process components. DECISION MAKING IN OCD that decision making under high-risk situations may not be impaired in OCD; however, studies tapping risky decision making, impulsivity, and ambiguity in the same sample are warranted. that decision making under high-risk situations may not be impaired in OCD; however, studies tapping risky decision making, impulsivity, and ambiguity in the same sample are warranted. Organizational strategies. Following the process of gathering information to make a decision, individuals must apply organizational strategies to evaluate alternatives (Olley et al., 2007). Previous research on executive functioning in OCD consistently highlighted deficits in organizational strategies between OCD patients and healthy controls, using multiple neuropsychological measures (Markarian et al., 2010; Shin et al., 2014). For example, relative to controls, participants with OCD showed significant impairments on the California Verbal Learning Test 2nd Edition (CVLT-II; Delis, Kramer, Kaplan, & Ober, 2000), which is a test of verbal memory (Shin et al., 2014; Tukel et al., 2012). The CVLT-II taps organizational strategies in the context of strategic learning. Savage and colleagues (2000) reported that organizational strategies mediate performance on memory tasks in OCD. It is therefore possible that deficits in the use of organizational strategies may be implemented in decision making. The CVLT-II also has two outcome variables that directly assess organizational strategies (Delis et al., 2000). Semantic clustering involves the extent to which individuals use the strategy of grouping categorically similar words together, in order to better recall words presented in a list. Alternatively, serial clustering involves the extent to which individuals attempt to recall the words in the same order that they were presented Alternatively, serial clustering involves the extent to which individuals attempt to recall the words in the same order that they were presented. According to Bechara and colleagues (1994), the IGT requires the use of organizational strategies during the task, but findings associated with overall performance on the IGT have been inconsistent in the literature thus far, and there are no outcome variables that tap this component of decision making. In addition, there are no decision-making measures designed to assess organizational strategies in the context of decision making. Thus, in the current study, 16 DECISION MAKING IN OCD yet to be directly compared in OCD. Examining the relationship between subjective and objective decision making would be a novel contribution to the literature on OCD. yet to be directly compared in OCD. Examining the relationship between subjective and objective decision making would be a novel contribution to the literature on OCD. Subjective Decision Making Indecisiveness is often a prominent clinical symptom for those with OCD (Steketee & Frost, 2003), yet studies examining subjective reports of decision making are scarce in the literature. The clinical phenomena evidenced in OCD, such as a difficulty discarding items, especially in those with hoarding symptoms (Steketee & Frost, 2003; Tolin, Kiehl, Worhunsky, Book, & Maltby, 2008) supports the need to study decision making further. More generally, individuals with OCD have been found to require more information, and need additional time to consider alternative possibilities, before making a decision (Dittrich & Johansen, 2013; Stern et al., 2013). It is unclear whether decision making deficits are consistent in subjective and objective reporting. To the author's knowledge, subjective and objective decision making have Indecisiveness is often a prominent clinical symptom for those with OCD (Steketee & Frost, 2003), yet studies examining subjective reports of decision making are scarce in the literature. The clinical phenomena evidenced in OCD, such as a difficulty discarding items, especially in those with hoarding symptoms (Steketee & Frost, 2003; Tolin, Kiehl, Worhunsky, Book, & Maltby, 2008) supports the need to study decision making further. More generally, 17 17 Study Goals and Hypotheses The primary objective of this study is to explore differences in decision making between individuals with OCD and healthy controls. Further to this, the study aims to elucidate the different components of decision making by determining if there are differences between OCD and control participants across three objective tasks. Based on previous literature, it is hypothesized that: 1) participants with OCD will demonstrate deficits on the IGT as compared to controls; 2) participants with OCD will demonstrate impairments on the IST as compared to controls; and 3) participants with OCD will perform similarly to controls on the CGT. Since organizational strategies appear to be important in decision making, the study aims to examine this component of decision making using a task that taps strategic learning and memory. It is hypothesized that: 4) there will be differences between participants with OCD and controls on strategic learning, as measured by the CVLT-II. The study also aims to determine if there are differences in subjective and objective decision making. It is hypothesized that: 5) subjective decision making based on the FIS will be significantly poorer in participants with OCD compared to controls; and 6) there will be a relationship between subjective decision-making scores and objective measures. Finally, the influence of other psychological factors, including intolerance of uncertainty and components of impulsivity, on decision making will also be examined. It is hypothesized that: 7) there will be a relationship between decision-making outcomes and self-reported intolerance of uncertainty and impulsivity. 18 DECISION MAKING IN OCD Participants This study included a case-control design comparing participants with diagnosed OCD (n = 22) and healthy controls (n = 22), ages 18 to 65 years. Inclusion criteria for participants with OCD was a principal diagnosis of OCD (minimum symptom duration of 1 year) according to the Structured Clinical Interview for DSM-IV (SCID-IV; First, Spitzer, Gibbon, & Williams, 1997). Participants in the study were required to endorse symptoms meeting criteria for a principal diagnosis of OCD, which may or may not have included symptoms of hoarding. Exclusion criteria included comorbid bipolar disorder or psychosis, current or recent (within the previous 12 months) substance use disorders, and a history of traumatic brain injury, as identified in the overview section of the SCID-IV. Clinical participants with other comorbid diagnoses, including depression and anxiety disorders, were not excluded from the study, as requiring such would severely limit eligibility and generalizability due to the high rates of comorbidity in this population. However, depression severity is known to hinder cognitive performance (Menzies et al., 2008), which is why individuals with severe depression, as indicated by a score of 29 or greater on the Beck Depression Inventory-II (BDI-II; Beck, Steer, & Brown, 1996), were excluded. English-speaking control participants with no medical, neurological, or psychiatric conditions were included in the study based on the diagnostic criteria of the SCID-IV. Recruitment and assessment took place at Ryerson University and the Clinic for OCD and Related Disorders at Sunnybrook Health Sciences Centre. Patients with OCD at the clinic were asked by a research coordinator whether they wished to hear about a research study. If they agreed, they were provided with a description of the study and asked if they would like to participate. A total of n = 19 participants with OCD were recruited from the hospital. Further 19 Power Between-group differences on various components of decision making and cognitive processes were examined. According to a power analysis, 22 patients with OCD and 22 controls were needed to achieve a moderate effect size (η2 = .44; α = .05; 1-β = .95). Power estimates were calculated with G-power calculator using repeated measures analysis of variance (ANOVA), between two groups and across 5 measurements (as there are 5 trials in the IGT). Effect size estimates were calculated based on the results of a study by Da Rocha and colleagues (2011), which compared differences on the IGT between participants with OCD (n = 107) and controls (n = 107). Due to the limited number of studies comparing decision making between OCD and controls, a conservative approach to power was used estimating sample size based on a η2 = .44, which is larger than the recommended minimum moderate effect size estimate of squared association indices (.25). DECISION MAKING IN OCD participants with OCD (n = 3), and healthy controls (n = 22), were recruited via advertisements at Ryerson University. The study received ethical approval from the Research Ethics Boards at Ryerson University and Sunnybrook Health Sciences Centre. A paid research assistant and a volunteer assisted with the project at the university. Measures Participants completed a battery of psychometrically validated tests and questionnaires that included semistructured clinical interviews, self-report questionnaires, paper-and-pencil tasks, and computerized measures. Clinical measures. OCD symptom severity. Diagnoses and comorbidities were assessed using the SCID-IV (First et al., 1997). The SCID-IV is a highly popular semistructured interview administered by a trained clinician. OCD symptoms and symptom severity were assessed using 20 20 DECISION MAKING IN OCD The scale can also be divided into six subscales based on factors measuring dimensions of OCD symptoms, including washing, checking, ordering, obsessing, hoarding, and neutralizing. Each statement was rated on a 5-point Likert scale from not at all disturbed (0) to extremely disturbed (4). Scores range between 0-90, with higher scores indicating greater severity of OCD The scale can also be divided into six subscales based on factors measuring dimensions of OCD symptoms, including washing, checking, ordering, obsessing, hoarding, and neutralizing. Each statement was rated on a 5-point Likert scale from not at all disturbed (0) to extremely disturbed (4). Scores range between 0-90, with higher scores indicating greater severity of OCD symptoms. The OCI-R has been shown to have good internal consistency in OCD patients (α = .83), test-retest reliability (r = .74-.91), convergent validity as indicated by moderate to large correlations with other measures of OCD, and good discriminant validity in distinguishing OCD from other anxiety disorders (Abramowitz & Deacon, 2006; Foa et al., 2002). symptoms. The OCI-R has been shown to have good internal consistency in OCD patients (α = .83), test-retest reliability (r = .74-.91), convergent validity as indicated by moderate to large correlations with other measures of OCD, and good discriminant validity in distinguishing OCD from other anxiety disorders (Abramowitz & Deacon, 2006; Foa et al., 2002). Depression. Depressive symptoms were evaluated using the Beck Depression Inventory- 2nd Edition (BDI-II; Beck et al., 1996). The scale is a 21-item self-report questionnaire that asks individuals to indicate which statement best describes the way they felt in the past 2 weeks. Each statement was rated on a 4-point Likert scale ranging from not at all (0) to extreme (3). Scores range between 0 and 63, with severity of symptoms delineated as follows: 0-13 indicates minimal symptoms, 14-19 is mild depression, 20-28 is moderate, and 29-63 is severe. The BDI- II has been demonstrated to have excellent internal consistency (α = .92), and test-retest reliability (r = .93). Convergent validity was also shown, as indicated by moderate to large correlations with other measures of depression, such as the Hamilton Depression Rating Scale, at pretreatment (r = .30) and posttreatment (r = .81; Beck et al.,1996; Schneibel, Brakemeier, Wilbertz, Dykierek, Zobel, & Schramm, 2012). Anxiety. Symptoms of anxiety were assessed using the Beck Anxiety Inventory (BAI; Beck, Epstein, Brown, & Steer, 1988), which primarily measures physical symptoms of anxiety. DECISION MAKING IN OCD reliable and valid measures of OCD. The Yale-Brown Obsessive-Compulsive Scale (YBOCS; Goodman, Price, & Rasmussen, 1989) is a semistructured clinical interview administered by a trained clinician. The interview consisted of both a symptom checklist and a severity of symptoms scale, which yielded a total score, and obsessions and compulsions subscale scores. The symptom checklist consisted of a list of common obsessions and compulsions, which were read to the participants with OCD. They then indicated if the symptom was currently present, or was present in the past, and then which of those symptoms were considered principal symptoms that would be targeted in treatment. The list of the principal symptoms was transferred to the severity scale, where participants with OCD rated these symptoms. The severity scale on the YBOCS inquired about the time spent, interference, distress, resistance, and control of symptoms for both obsessions and compulsions. Each symptom was rated on a 5-point Likert scale from no symptoms (0) to extreme symptoms (4). All items were added for a total score on the YBOCS. Severity of symptoms were scored as follows: 0-7 = subclinical symptoms; 8-15 = mild; 16-23 = moderate; 24-31 = severe; and 32-40 = extreme. The YBOCS has been demonstrated to have acceptable to excellent internal consistency (α = .69- .91), interrater reliability (r = .93), and test-retest reliability (r =.61; Goodman et al., 1989; Woody, Steketee, & Chambless, 1995). The scale has also shown good convergent validity with other measures of OCD (Woody et al., 1995). The majority of participants with OCD completed the SCID-IV and YBOCS at Sunnybrook Health Sciences Centre, as part of an ongoing genetics study examining OCD and related disorders. Participants with OCD, recruited from the community, completed these interviews as part of the test battery at Ryerson University. The Obsessive-Compulsive Inventory-Revised (OCI-R; Foa et al., 2002) is an 18-item self-report questionnaire that measures the presence and severity of common OCD symptoms. 21 21 DECISION MAKING IN OCD The scale is a 21-item self-report questionnaire in which individuals rate the extent to which the statement applies to them (Beck & Steer, 1993). Each statement is rated on a 4-point Likert scale 22 DECISION MAKING IN OCD ambiguous situations (i.e., with no known probabilities). This measure was designed to mimic real-life decision making, due to the uncertainty involved in the choices, as results and probabilities are unknown or cannot be precisely calculated (Bechara et al., 1994). Participants were asked to select a card from one of four possible decks (A, B, C, D) for 100 trials, which was divided in 5 blocks. They were told that they can select cards from the decks in any order. Each time they selected a card, they either received money ($100 for decks A and B, and $50 for decks C and D), or had to pay money. There were random reward and punishment schedules for each deck, that the participant is initially unaware of, as they were only told that some decks are "worse" than others. Decks A and B were considered disadvantageous decks as they resulted in large immediate cash rewards, but had larger losses over time. Decks C and D were considered advantageous desks as they resulted in smaller more consistent immediate cash rewards, with fewer losses. The IGT net score was calculated by the number of cards selected from the advantageous decks minus the number of cards selected from the disadvantageous decks [(C+D)- (A+B)]. The test measured learning over time, where participants should gradually choose less cards from A and B and more cards from C and D. A higher score indicated better performance on the task, as it meant the participant was choosing more cards from the advantageous decks. The participants were asked to try and earn as much money as possible, or to simply avoid losing money. Impaired performance on this task has been shown to be sensitive to patients with orbitrofrontal cortex damage (Cavedini et al., 2002). ambiguous situations (i.e., with no known probabilities). This measure was designed to mimic real-life decision making, due to the uncertainty involved in the choices, as results and probabilities are unknown or cannot be precisely calculated (Bechara et al., 1994). Participants were asked to select a card from one of four possible decks (A, B, C, D) for 100 trials, which was divided in 5 blocks. They were told that they can select cards from the decks in any order. Each time they selected a card, they either received money ($100 for decks A and B, and $50 for decks C and D), or had to pay money. DECISION MAKING IN OCD ranging from not at all (0) to severely (3). Scores range between 0-63, and were interpreted as follows: 0-7 indicates minimal or normal level of anxiety, 8-15 is mild anxiety, 16-25 is moderate, and 26-63 is severe. The BAI has been shown to have excellent internal consistency (α = .92), and good test-retest reliability (r = .75). Convergent validity was also demonstrated as indicated by moderate to large correlations with other measures of anxiety, such as the Hamilton Anxiety Rating Scale-Revised (r = .51), and both the state (r = .47) and trait (r = .58) subscales on the State-Trait Anxiety Inventory (STAI; Beck et al.,1988; Beck & Steer, 1993). Quality of Life. The Quality of Life Enjoyment and Satisfaction Questionnaire-Short Form (QLES; Stevanovic, 2011) is a 16-item self-report questionnaire, that measures the extent to which an individual is satisfied with their life and current circumstances based on physical health, mood, work, activities, relationships, functioning, sexual desire, economic status, overall well-being, and medication. Each statement was rated on a 5-point Likert scale ranging from very poor (1) to very good (5) in the past week. Scores range between 16-70 (as the item for medication status may be not applicable and is thus rated separately, and the item regarding overall well-being is rated separately), with higher scores indicating better quality of life. The QLES has demonstrated excellent internal consistency (α = .90), and test-retest reliability (r = .93), as well as convergent validity with related measures assessing quality of life (Stevanovic, 2011). Objective decision making measures. Three objective neuropsychological tests of decision making were included in the battery of tests. These included the IGT (Bechara et al., 1994), the IST (Clark et al., 2006), and the CGT (Rogers et al., 1999). Ambiguity in decision making. The IGT is a well-known neuropsychological measure of decision making. The task tests the ability to forgo immediate reward for long-term gain, under 23 23 DECISION MAKING IN OCD cognitive flexibility, which was not examined in this study. The IST took about 12 minutes to complete. Participants started by seeing a grid of 25 grey boxes on the screen. Participants were required to decide which colour was more prevalent (e.g., yellow or blue) hidden under the grey boxes. By touching the grey boxes, the colours were revealed. In the first condition (referred to as the no cost condition), participants were instructed that they could open as many boxes as they wish before deciding which colour there was more of. To make a decision, they simply touched the coloured panel at the bottom of the screen. For each correct decision, participants were awarded 100 points. In the second condition (the cost condition), as the number of boxes opened increased, the amount of points won decreased. Participants were told that on these trials, the amount you can win starts at 250 points, and decreases by 10 points for every box that is touched. In both conditions, there were 10 trials, and there was a 100-point penalty for wrong decisions. Six dependent variables on the IST were included in the study (three variables, each calculated for both the cost and no cost conditions). The mean number of boxes opened per trial is the most commonly reported dependent variable on this test. It was expected that in the no cost condition, more boxes would be opened than in the cost condition. A greater number of boxes opened indicated the need to gather more information before making a decision, out of a total of 25 boxes. The number of sampling errors indicated the number of incorrect decisions made, Six dependent variables on the IST were included in the study (three variables, each calculated for both the cost and no cost conditions). The mean number of boxes opened per trial is the most commonly reported dependent variable on this test. It was expected that in the no cost condition, more boxes would be opened than in the cost condition. A greater number of boxes opened indicated the need to gather more information before making a decision, out of a total of 25 boxes. The number of sampling errors indicated the number of incorrect decisions made, which was measured in each condition. It was expected that more errors would be made in the cost condition than the no cost condition. DECISION MAKING IN OCD There were random reward and punishment schedules for each deck, that the participant is initially unaware of, as they were only told that some decks are "worse" than others. Decks A and B were considered disadvantageous decks as they resulted in large immediate cash rewards, but had larger losses over time. Decks C and D were considered advantageous desks as they resulted in smaller more consistent immediate cash rewards, with fewer losses. The IGT net score was calculated by the number of cards selected from the advantageous decks minus the number of cards selected from the disadvantageous decks [(C+D)- (A+B)]. The test measured learning over time, where participants should gradually choose less cards from A and B and more cards from C and D. A higher score indicated better performance on the task, as it meant the participant was choosing more cards from the advantageous decks. The participants were asked to try and earn as much money as possible, or to simply avoid losing money. Impaired performance on this task has been shown to be sensitive to patients with orbitrofrontal cortex damage (Cavedini et al., 2002). Impulsivity in decision making. The IST is a computerized measure, part of the Impulsivity in decision making. The IST is a computerized measure, part of the Cambridge Neuropsychological Test Automated Battery (CANTAB; Cambridge Cognition Ltd, 2006). This task assessed decision making and impulsivity by examining the tendency to gather and evaluate information before a decision was made. The test was also designed to measure 24 DECISION MAKING IN OCD A greater number of errors indicated more impulsiveness in making decisions. Finally, the total number of trials correct was a measure of overall performance on the test. A greater number of trials correct indicated better performance on the test, out of a total of 10 trials per condition. 25 DECISION MAKING IN OCD measured the proportion of trials that the more likely outcome was chosen (i.e., the colour with the greater probability). A higher score indicated better performance. Finally, the probability of risk taking measured the mean proportion of the total points gambled. Lower scores indicated greater self-control. Strategic learning. As there are no decision-making tasks (or outcome variables on these tasks) that tap organizational strategies, a memory measure was included in the study to examine organizational strategies. The CVLT-II (Delis et al., 2000) measures verbal memory by having participants recall words presented in a list. The assessor began by reading a list of 16 words to participants. Participants were then asked to recall as many words as they could from memory, which was repeated for 5 trials. The list contained an imbedded semantic structure, consisting of four different categories of words (vegetables, ways of travelling, furniture, and animals). The words were presented in a random order (where two words from the same category were never in sequential order), and participants were not primed to notice this semantic structure. Next, a second list of 16 words was presented with another four categories, and participants were instructed to recall words from the second list for one trial. This trial acted as interference, as participants were then asked to again recall words from the first list, which was read several times (considered a short delay). After a 20-minute delay (long delay), participants were asked to recall words from the first list, and then were subsequently primed to recall words from each of the four semantic categories. Finally, after a 10-minute delay, a recognition test was presented, where participants were instructed to identify which words were presented in the first list. The greater number of words recalled in each trial, the better the verbal memory performance. participants were then asked to again recall words from the first list, which was read several times (considered a short delay). After a 20-minute delay (long delay), participants were asked to recall words from the first list, and then were subsequently primed to recall words from each of the four semantic categories. Finally, after a 10-minute delay, a recognition test was presented, where participants were instructed to identify which words were presented in the first list. The greater number of words recalled in each trial, the better the verbal memory performance. DECISION MAKING IN OCD Risky decision making. The CGT is a computerized measure, part of the CANTAB (Cambridge Cognition Ltd, 2006). This task was also employed to measure risk-taking behaviour in decision making, based on the perception of risk with known probabilities. The CGT took about 25 minutes to complete. On each trial, the participant was presented with 10 boxes, with a varying proportion of red and blue boxes. Participants were instructed that the computer had hidden a yellow token behind one of the boxes. They were expected to indicate which coloured box (blue or red) they believed the token was hidden under, and then wager a desired number of points to bet on the answer. The ratio of blue to red boxes changed randomly and participants were expected to use probabilities to guess the answer. In the first set of 36 trials, the number of points to bet was shown on the screen in ascending order, so participants could choose the size of their bet by touching the screen. In the second set of 36 trials, points were presented in descending order, starting with the highest possible bet. After each response, participants were told whether they are right or wrong, and the points were added or subtracted from the total score. An individual with greater impulsivity was expected to choose low bets in the ascending condition and high bets in the descending condition, whereas someone with high risk taking behaviour would likely wait for the bets to increase in the ascending condition, but would quickly select bets in the descending condition. Four dependent variables on the CGT were included in the study. The probability of rational decisions, measured the overall proportion of the bet placed, in relation to the probability that the colour selected was correct. A lower score on this variable indicated greater self-control. Deliberation time was the mean time elapsed in seconds between when the coloured boxes were presented and when the choice was selected, which measured a component of impulsivity. Lower scores indicated greater impulsiveness in responses. The probability of quality decisions 26 DECISION MAKING IN OCD Performance on the CVLT-II improves when organizational strategies are implemented during the learning trials when participants are presented the same 16 words, five times. A 27 DECISION MAKING IN OCD dependent variable included in this study was a score of retroactive interference, which examined the extent to which the presentation of a second word list interfered with recall of words from the first list during the short delay. Although this measure taps memory performance, the focus in the current study was on the two outcomes variables that assessed organizational strategies: semantic and serial clustering. A semantic clustering score measured the degree to which participants grouped words from the same category together in order to facilitate recall. The more words from the same category that were recalled in order, the better the score. A serial clustering score measured the degree to which participants attempted to recall words in the same order as initially presented. The more words recalled in sequential order, the better the score. All outcomes were calculated and scored using the CVLT-II computerized scoring software, which accounted for age, sex, education and ethnicity. ubjective decision making. The Frost Indecisiveness Scale (FIS; Frost & Shows, ubjective decision making. The Frost Indecisiveness Scale (FIS; Frost & Shows, Subjective decision making. The Frost Indecisiveness Scale (FIS; Frost & Shows, 1993), is a 15-item self-report questionnaire that measures perceived everyday decision making. Each statement was rated on a 5-point Likert scale ranging from strongly disagree (1) to strongly agree (5). Six items were reverse coded (Questions 2, 3, 5, 6, 8, and 9). Scores range between 15 and 75, with higher scores indicating greater difficulty with decision making or indecisiveness. The scale has been shown to have excellent internal reliability (α = .90; Frost & Shows, 1993). Other psychological measures. In order to further examine components of decision making, two related self-report questionnaires were included in the study. The purpose of these measures was to measure intolerance of uncertainty and impulsivity. Intolerance of Uncertainty. The Intolerance of Uncertainty Scale (IUS; Buhr & Dugas, 2002) is a 27-item self-report questionnaire that measures differing levels of negative beliefs and reactions to uncertainty, ambiguous situations, and beliefs about the future. Each statement was 28 DECISION MAKING IN OCD rated on a 5-point Likert scale from not at all characteristic of me (1) to entirely characteristic of me (5). Scores range between 27 and 135, with higher scores indicating greater intolerance of uncertainty. The IUS exhibited excellent internal consistency (α = .94), good test-retest reliability (r = .74), and convergent validity with related cognitive constructs, including worry, neuroticism, and anxiety (Buhr & Dugas, 2002; McEvoy & Mahoney, 2011). Impulsivity. The Barratt Impulsiveness Scale-II (BIS-II; Patton et al., 1995) is a 30-item self-report questionnaire that measures the behavioural construct/personality trait of impulsivity. The scale was broken down into higher order factors (attentional impulsivity, motor impulsivity and nonplanning impulsivity). These factors can be further isolated into lower order factors: attentional impulsivity is composed of attention and cognitive instability subscales; motor impulsivity is composed of motor and perseverance subscales; and nonplanning impulsivity is composed of self-control and cognitive complexity. Each statement was rated on a 4-point Likert scale ranging from rarely/never (1) to almost always/always (4). Ten items were reverse coded (Questions 1, 7, 8, 9, 10, 12, 13, 15, 20 and 29). Scores range between 30 and 120, with higher scores indicating greater impulsivity. The scale has been shown to have good internal consistency on the total scale (α = .83), and on the attentional (α = .74), and nonplanning (α = .72), subscales (Stanford, Mathias, Dougherty, Lake, Anderson, & Patton, 2009). However, the motor subscale demonstrated poor internal consistency (α = .59). The BIS-II has also shown high concurrent validity with other commonly used measures of impulsivity (Stanford et al., 2009). Other cognitive measures. An intelligence measure was included to determine premorbid functioning and to acquire a baseline level of functioning. In addition, in order to better say that any performance decrements found among OCD patients were not the result of slowed motor speed or thinking speed, or an inability to inhibit a response, measures of 29 29 DECISION MAKING IN OCD psychomotor speed, information processing speed, and inhibition of motor and cognitive responses were also employed. psychomotor speed, information processing speed, and inhibition of motor and cognitive responses were also employed. The Wechsler Test of Adult Reading (Holdnack, 2001) is a proxy measure for premorbid functioning, assessed by examining reading ability. Participants were asked to read a list of words out loud, and were given a score of 1 if the word was pronounced correctly or 0 if it was pronounced incorrectly. The words on the list progressively became more difficult over time. There were 50 words in total and thus scores range between 0-50, with higher scores indicating higher reading ability. The Finger Tapping Test (Heaton, Grant, & Matthews, 1992) is a measure of psychomotor speed. Participants tapped their index finger on a lever as rapidly as possible for 10 seconds. The number of taps was recorded for each trial, with a greater number of taps indicating greater speed. The task was repeated until 5 consecutive trials were within 5 points of each other, up to a maximum of 10 trials. The dependent variables were the mean scores of the combined trials for both the dominant and nondominant hands. Digit Symbol-Coding (Wechsler, 1997) is a measure of cognitive processing speed, and is part of the Wechsler Adult Intelligence Scale-III. This task was completed on a sheet of paper using a pencil. Across the top of the page there were numbers associated with corresponding symptoms (e.g., 1= O, 2= >). On the rest of page, there were boxes with numbers in a random order printed in the upper portion, and blank spaces in the lower portion. Participants were instructed to copy the corresponding symbol in the boxes below the numbers. They were further instructed to complete the boxes in order, without skipping any, and to work as fast as possible. The dependent outcome was the number of correct symbols copied in 2 minutes, with a greater 30 30 Procedure Once recruited, participants were provided with a description of the study, including purpose, length of time, risks, etc. Participants completed a phone screen interview conducted by a research assistant to determine eligibility. Questions were related to background information and symptoms of psychological disorders. Individuals for whom inclusion criteria were met were invited to participate and those for whom inclusion criteria were not met were told the reason for their exclusion (e.g., that the tests required colour vision and English fluency). Participants were also required to provide informed consent to participate in the study. Testing was conducted by two trained graduate students at the Master's and Doctoral levels in Clinical Psychology. Each assessment took place over the course of one session of approximately 4 hours, with offered breaks between measures, as required. The two interviews (SCID, YBOCS) were completed first. In order to avoid interference effects, tests measuring similar modalities (e.g., verbally mediated tests) were not placed in sequential order of administration. All participants received a fixed administration order and participants were compensated $25 for their time. DECISION MAKING IN OCD number of items completed indicating better performance. Incorrectly drawn symbols were subtracted from the total score. number of items completed indicating better performance. Incorrectly drawn symbols were subtracted from the total score. The Stop-Signal Paradigm (Logan, Cowan, & Davis, 1984) tested the speed for executing and voluntarily inhibiting a motor response. Participants were asked to watch a computer screen that presented a fixation point in the middle of the screen. Participants were asked to press corresponding buttons in response to "X"s and "O"s sequentially presented in the middle of the screen, as quickly as they could. The stop signal is an audible tone that occurred at random intervals between the presentation of letters. Participants were asked to inhibit their response of pressing the corresponding button when they heard the tone. The primary dependent variable was the stop signal reaction time (SSRT), measured in seconds, where a shorter time indicated better performance. This outcome measures the latency between the presentation of the sound, and the initiation of inhibiting the response, which is calculated by subtracting the mean delay from the mean reaction time. The Stroop Color-Word Interference Test (Golden, 1976) evaluated cognitive response inhibition across 3 trials. The first trial (word trial) required participants to read a list of words (colours) as fast as they could in 45 seconds. The second trial required that they read the colour of the ink that a series of "XXX"s were printed in (colour trial). The final trial required participants to read the colour of the ink that the word was in, inhibiting the response to read the word itself (word/colour interference). On this trial, there was a discrepancy between the word (e.g., blue) and the colour of the ink (e.g., red). The scores for each trial were based on the number of words read, where a greater number of words indicated better performance. 31 31 Data Analysis The data were analyzed using the Statistical Package for Social Sciences, Version 21 (SPSS). Analyses consisted of a comparison of demographic information between participants with OCD and controls using independent samples t-tests and chi-square analyses. Independent samples t-tests were also conducted to determine if there were significant differences between OCD and control participants on measures of decision making, psychological measures, and neuropsychological measures. Between-group (OCD, control) differences on tasks where there were several levels of difficulty were analyzed using mixed ANOVA. To ascertain if there were 32 Demographics Demographic characteristics for participants with OCD (n= 22) and healthy controls (n= 22) are shown in Table 2. Independent samples t-tests were conducted to compare groups on age, level of education, and premorbid functioning. The OCD and control groups did not differ in age, t(42) = -.86, p = .398, or years of education, t(42) = -.72, p = .476. There were no significant differences between groups on the WTAR, which is a measure of premorbid function, t(42) = - 1.99, p = .053, indicating that both groups had comparable reading ability. To compare sex and ethnicity between groups, Pearson's chi-square tests were conducted. The two groups also did not differ for sex, χ2(1) = 1.47, p = .226. The assumption of expected frequencies was not met for ethnicity, indicating the need to collapse across the groups. The majority of the sample identified as Caucasian (59%), and therefore the groups were collapsed into Caucasian and other. There was a significant association between group status (OCD, control) and ethnicity, χ2 (1) = 6.02, p = .031, φ = -.370. participants with OCD were 1.89 times more likely to be Caucasian than healthy controls, and thus the influence of ethnicity on outcomes was considered in subsequent analyses. DECISION MAKING IN OCD relationships among the decision-making variables and psychological measures, Pearson's correlation coefficients were computed. Further analyses included regression and moderator analyses to examine the relationship between decision making, intolerance of uncertainty and impulsivity. All assumptions of parametric tests were met, including normality, independent groups, interval level data, and homogeneity of variance, with a few exemptions. In the instances where an assumption had been violated, this was reported within the Results section. Participants were excluded pair-wise from analyses for missing data points, which reduced the sample size for that test. However, these participants were still included in the analyses for the tests for which they had complete data. As the statistical tests completed were designed to examine All assumptions of parametric tests were met, including normality, independent groups, interval level data, and homogeneity of variance, with a few exemptions. In the instances where an assumption had been violated, this was reported within the Results section. Participants were excluded pair-wise from analyses for missing data points, which reduced the sample size for that test. However, these participants were still included in the analyses for the tests for which they had complete data. As the statistical tests completed were designed to examine neuropsychological performance based on a priori hypotheses, each measure was viewed as a separate analysis for statistical purposes (Lawrence et al., 2006). Thus, significant results were based on a p value of < .05 and all tests were two-tailed. neuropsychological performance based on a priori hypotheses, each measure was viewed as a separate analysis for statistical purposes (Lawrence et al., 2006). Thus, significant results were based on a p value of < .05 and all tests were two-tailed. There is high comorbidity between OCD and depression, and depression is well known to have an impact on cognitive performance (Menzies et al., 2008). Thus, it is possible that depression may impact decision-making ability among those with OCD. To examine if OCD diagnostic status maintained significance in terms of predicting performance when depression scores were taking into account, hierarchical regression analyses were conducted. 33 Clinical Information Clinical information, including scores on measures assessing severity of anxiety and depression are presented in Table 3. According to the SCID-IV, all clinical participants had a principal diagnosis of OCD. The majority of the sample also had comorbid diagnoses, including major depressive disorder/dysthymic disorder (23%), trichotillomania/impulse control disorder not otherwise specified (18%), hoarding (based on an insert of questions added to the SCID-IV; 14%), and other anxiety disorders, including specific phobia (14%), panic disorder (1%), social 34 DECISION MAKING IN OCD anxiety disorder (1%) and generalized anxiety disorder (1%). No psychological diagnoses were present in the control group. The mean YBOCS score in the OCD group indicated a moderate severity of symptoms (M = 22.41, SD = 7.04). As expected, there were significant differences between groups for severity of OCD symptoms. On the OCI-R, participants with OCD (M = 26.27, SD = 9.71) reported greater symptoms than controls (M = 7.27, SD = 7.69), t(42) = -7.20, p < .001. As expected, few OCD symptoms were reported by healthy controls. p < .001. As expected, few OCD symptoms were reported by healthy controls. p < .001. As expected, few OCD symptoms were reported by healthy controls. Table 2 Mean Demographic Characteristics Between Groups Characteristic OCD (n = 22) M (SD) Control (n = 22) M (SD) Statistic p Effect Size Age (years) 37.23 (13.53) (range = 21-65) 33.50 (15.34) (range = 19-64) t(42) = -0.86 .398 r = .13 Females: Male 12:10 8:14 χ2(1) = 1.47 .364 φ = .18 Education (years) 16.39 (2.60) 15.89 (1.96) t(42) = -0.72 .476 r = .11 Ethnicity χ2(1) = 6.02 .031* φ = -.37 Caucasian (%) 77 41 Asian (%) 9 41 African American (%) 0 9 Hispanic/Latino (%) 5 5 Native American (%) 5 0 Other (%) 5 5 WTAR 44.82 (3.45) 41.86 (6.06) t(42) = -1.99 .053 r = .29 Note. M = Mean; SD = Standard deviation; * p < .05; WTAR = Wechsler Test of Adult Reading. Mean Demographic Characteristics Between Groups There were also differences between groups on other measures of psychopathology. On the measures of anxiety and depression, Levene’s tests, BAI: F(1, 41) = 5.85, p = .020, BDI-II: F(1, 41) = 5.43, p = .025, were significant, indicating that the assumption of homogeneity of variances was not met. Therefore, an analysis that did not assume homogeneity of variance was adopted, which revealed a significant group effect on the BAI, t(32.63) = -4.49, p < .001, and BDI-II, t(35.77) = -5.01, p < .001. The range of anxiety scores for participants with OCD was 0 to 33. As a group, participants with OCD (M = 15.52, SD = 9.82) reported a mild anxiety There were also differences between groups on other measures of psychopathology. On the measures of anxiety and depression, Levene’s tests, BAI: F(1, 41) = 5.85, p = .020, BDI-II: 35 Table 3 Mean Clinical Information Scores Between Groups Mean Clinical Information Scores Between Groups Mean Clinical Information Scores Between Groups Mean Clinical Information Scores Between Groups Measure OCD (n = 22) M (SD) Control (n = 22) M (SD) Statistic p Effect Size (r) Age of Onset 17.75 (13.12) N/A OCI-R 26.27 (9.71) 7.27(7.69) t(42) = -7.20 < .001 *** .74 YBOCS total 22.41 (7.04) N/A YBOCS-O 11.64 (3.57) N/A YBOCS-C 10.68 (4.55) N/A BDI-II 17.18 (9.18) 5.48 (5.83) t(35.77) = -5.01 < .001 *** .64 BAI 15.52 (9.82) 4.36(5.94) t(32.63) = -4.49 < .001 *** .62 QLES 44.63 (9.47) 52.63 (8.88) t(42) = 2.47 .020* .36 Note. M = Mean; SD = Standard deviation; OCI-R = Obsessive-Compulsive Inventory-Revised; YBOCS = Yale-Brown Obsessive-Compulsive Scale; YBOCS-O = Yale-Brown Obsessive- Compulsive Scale Obsessions subscale; YBOCS-C = Yale-Brown Obsessive-Compulsive Scale Compulsions subscale; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Inventory; QLES = Quality of Life and Enjoyment Survey; * p < .05; ** p < .01; *** p <.001. Levene's test was significant (p < .05) for the BDI-II and BAI, thus equal variances were not assumed. ; ; p y ; YBOCS = Yale-Brown Obsessive-Compulsive Scale; YBOCS-O = Yale-Brown Obsessive- Compulsive Scale Obsessions subscale; YBOCS-C = Yale-Brown Obsessive-Compulsive Scale Compulsions subscale; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Inventory; QLES = Quality of Life and Enjoyment Survey; * p < .05; ** p < .01; *** p <.001. Levene's test was significant (p < .05) for the BDI-II and BAI, thus equal variances were not assumed. YBOCS = Yale-Brown Obsessive-Compulsive Scale; YBOCS-O = Yale-Brown Obsessive- Compulsive Scale Obsessions subscale; YBOCS-C = Yale-Brown Obsessive-Compulsive Scale Compulsions subscale; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Inventory; QLES = Quality of Life and Enjoyment Survey; * p < .05; ** p < .01; *** p <.001. Levene's test was significant (p < .05) for the BDI-II and BAI, thus equal variances were not assumed. YBOCS = Yale-Brown Obsessive-Compulsive Scale; YBOCS-O = Yale-Brown Obsessive- Compulsive Scale Obsessions subscale; YBOCS-C = Yale-Brown Obsessive-Compulsive Scale Compulsions subscale; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Inventory; QLES = Quality of Life and Enjoyment Survey; * p < .05; ** p < .01; *** p <.001. Levene's test was significant (p < .05) for the BDI-II and BAI, thus equal variances were not assumed. DECISION MAKING IN OCD symptom presentation, whereas controls reported minimal symptoms (M = 4.36, SD = 5.94). The range of depression scores for participants with OCD was 0 to 28. As a group, participants with OCD (M = 17.18, SD = 9.18) also reported mild symptoms of depression, whereas controls reported minimal symptoms (M = 5.48, SD = 5.83). To examine self-reported quality of life, the difference between groups on the QLES was examined. As expected, results indicated that participants with OCD (M = 44.63, SD = 9.47) reported poorer quality of life than controls (M = 52.63, SD = 8.88), t(42) = 2.47, p = .020. Table 3 DECISION MAKING IN OCD taking additional medications (one was also on an investigational new drug, one was on cyclohexan, one was on buproprion, and one was on mirtazapine and risperidone). Of those participants on medication, 14% were on selective norepinephrine reuptake inhibitors (SNRIs), including venlafaxine and duloxetine. Of those taking SNRIs, one participant was also taking buproprion and lisdexamfetamin dimesylate. An additional 9% of participants were taking benzodiazepines, including clonazepam (this participant was also on aripiprazole) and alprazolam. Finally, an additional participant (5%) was on an unknown medication. Only one control participant was on a psychotropic medication (alprazolam), which was being prescribed as a muscle relaxant. The typical adult daily dose of this medication ranges between 0.75 mg and 4 mg (Stahl, 1993). The control participant was only taking 0.12 mg every other day, did not have any psychological disorders, and thus was not excluded. Table 3 In total, 59% of participants in the OCD group were on psychotropic medication. Of those participants, 36% were on selective serotonin reuptake inhibitors (SSRIs), including sertraline, escitalopram, paroxetine, and fluoxetine. Of those taking SSRIs, four participants were 36 Decision-making Outcomes Results of decision making outcomes based on the four decision-making measures in OCD and control participants are presented in Table 4. This included three objective measures of decision making (IGT, IST and CGT), and one subjective measure (FIS). Next, an examination of a memory measure (CVLT-II) as a proxy for organizational strategies is presented, as seen in Table 5. Subjective decision-making outcomes are then reported, followed by an examination of the three objective measures of decision making in relation to the self-report FIS, as seen in Table 6. Finally, the findings associated with self-report measures of ambiguity and impulsivity are examined in relation to decision-making performance. Objective decision-making tasks. A mixed ANOVA was conducted on the IGT to examine ambiguous decision making and learning over time between groups, with group (OCD, healthy controls) as the between-subjects factor and blocks (5 blocks) as the within-subjects 37 DECISION MAKING IN OCD factor. Results revealed that on the IGT, there was a significant block effect, F(4, 168) = 5.35, p < .001, η2 p = .113, indicating that both groups showed learning over time. However, no effect of group was found, F(4, 168) = 0.40, p = .808. An independent samples t-test was conducted to examine differences between groups on the IGT Net outcome, which is defined by the number of cards selected from the advantageous decks (C+D) minus the number of cards of disadvantageous decks (A+B). No significant differences between groups on IGT Net were found, t(42) = -1.05, p = .301. For the remaining two objective measures of decision making (IST and CGT), data for six and five participants with OCD, respectively, and two controls, were unusable due to technical errors. A mixed ANOVA was conducted on the IST measuring impulsivity and decision making with the outcome measuring the mean number of boxes opened per trial with group (OCD, healthy controls) as the between-subjects factor and condition (cost, no cost) as the within-subjects factor. Results revealed no significant differences between groups on the mean number of boxes opened, F(1, 34) = 0.41, p = .525. Similar models were also conducted on the outcomes measuring total sampling errors and total number of correct answers. There were also no differences found between groups on sampling errors, F(1, 34) = 1.21, p = .280, or correct answers, F(1, 34) = 1.17, p = .287. Independent samples t-tests were conducted to compare the two groups on the dependent outcomes associated with the CGT as a measure of risky decision making. No significant differences emerged between groups on outcomes related to the probability of rational decisions, t(35) = -0.46, p = .652, deliberation time, t(35) = -1.18, p = .247, the probability of quality decisions, t(35) = -0.55, p = .586, or the probability of risk taking behaviour, t(35) = -0.88, p = 38 38 DECISION MAKING IN OCD P = Probability; M = Mean; SD = Standard Deviation; s = Seconds; IGT = Iowa Gambling Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; 1 = no cost condition; 2 = cost condition; FIS = Frost Indecisiveness Scale. *** p < .001. Due to technical difficulties, n = 20 for the control group and n = 17 for the OCD group on the CGT; n = 20 for the control group and n = 16 for the OCD group on the and IST. DECISION MAKING IN OCD .384. Overall, no significant differences emerged between groups on any of the dependent variables for the IGT, IST, and CGT. Table 4 Mean Decision-making Performance Between Groups Measure OCD M (SD) Control M (SD) Statistic p Effect Size FIS 50.50 (9.10) 35.91 (9.79) t(42) = -5.12 <.001*** r = .62 IGT Net (CD-AB) 21.77 (25.20) 13.77 (25.44) t(42) = -1.10 .301 r = .17 CGT p of Rational Decisions .53 (0.13) .51 (0.14) t(35) = -.46 .652 r = .08 CGT Deliberation time (s) 2343.00 (805.00) 2021.00 (850.00) t(35) = -1.18 .247 r = .20 CGT p of Quality Decisions .90 (0.16) .87 (.21) t(35) = -.55 .586 r = .09 CGT p of Risk Taking .57 (.12) .53 (.16) t(35) = -.88 .384 r = .15 IST M Boxes Opened per Trial (1) 16.84 (6.59) 15.66 (6.62) F(1, 34) = .02 .883 η2 =.001 IST M Boxes Opened per Trial (2) 12.56 (5.73) 11.17 (6.40) IST Errors (1) 0.88 (1.15) 0.75 (0.97) F(1, 34) = 1.21 .280 η2 =.034 IST Errors (2) 1.44 (1.15) 1.75 (1.25) IST Total Trials Correct (1) 8.88 (1.26) 9.15 (1.09) F(1, 34) = 1.17 .287 η2 =.033 IST Total Trials Correct (2) 8.25 (1.18) 8.10 (1.29) Note. P = Probability; M = Mean; SD = Standard Deviation; s = Seconds; IGT = Iowa Gambling Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; 1 = no cost condition; 2 = cost condition; FIS = Frost Indecisiveness Scale. *** p < .001. Due to technical difficulties, n = 20 for the control group and n = 17 for the OCD group on the CGT; n = 20 for the control group and n = 16 for the OCD group on the and IST. Organizational strategies as measured by a verbal memory task. Results on the Mean Decision-making Performance Between Groups Note. P = Probability; M = Mean; SD = Standard Deviation; s = Seconds; IGT = Iowa Gambling Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; 1 = no cost condition; 2 = cost condition; FIS = Frost Indecisiveness Scale. *** p < .001. Due to technical difficulties, n = 20 for the control group and n = 17 for the OCD group on the CGT; n = 20 for the control group and n = 16 for the OCD group on the and IST. Note. Organizational strategies as measured by a verbal memory task. Results on the verbal memory measure (CVLT-II), as a representation of organizational strategies in OCD and verbal memory measure (CVLT-II), as a representation of organizational strategies in OCD and 39 DECISION MAKING IN OCD control participants, are presented in Table 5. Independent samples t-test were conducted to examine both semantic and serial clustering, which are two outcome variables on the CVLT-II used to measure learning and organizational strategies. For semantic clustering, Levene's test, control participants, are presented in Table 5. Independent samples t-test were conducted to examine both semantic and serial clustering, which are two outcome variables on the CVLT-II used to measure learning and organizational strategies. For semantic clustering, Levene's test, F(1, 42) = 7.33, p = .010, was significant, indicating that the assumption of homogeneity of variances was not met. Therefore, an analysis that did not assume homogeneity of variance was adopted. Examination of semantic learning revealed that there were no differences in the use of semantic clustering strategies between participants with OCD and controls, t(37.50) = 1.49, p = .146. In contrast, results revealed that participants with OCD (M = 0.95, SD = 1.46) used serial clustering strategies significantly more than control participants (M = 0.13, SD = 0.87), t(42) = - 2.26, p = .029. Further analyses, using independent samples t-tests, examined interference on the CVLT-II. Results revealed that after the presentation of a second word list, participants with OCD (M = -17.65, SD = 21.25) remembered significantly less words from the first list than controls (M = -7.08, SD = 12.16), t(42) = 2.02, p = .049. Table 5 Table 5 Mean Performance on Memory Task Between Groups Outcome on CVLT-II OCD (n = 22) M (SD) Control (n = 22) M (SD) Statistic p Effect Size (r) Semantic Clustering 1.97 (2.22) 3.21 (3.19) t(37.50) = 1.49 .146 .24 Serial Clustering 0.95 (1.46) 0.13 (0.87) t(42) = -2.26 .029* .33 Interference -17.65 (21.25) -7.08 (12.16) t(42) = 2.02 .049* .30 Note. M = Mean; SD = Standard deviation; CVLT-II = California Verbal Learning Test; Levene's test was significant (p< .05) for Semantic Clustering, thus equal variances were not assumed; * p < .05 Mean Performance on Memory Task Between Groups Mean Performance on Memory Task Between Groups Note. M = Mean; SD = Standard deviation; CVLT-II = California Verbal Learning Test; Levene's test was significant (p< .05) for Semantic Clustering, thus equal variances were not assumed; * p < .05 In summary, participants with OCD were significantly more likely to apply serial strategies to remembering the word list than controls; however, there did not appear to be a In summary, participants with OCD were significantly more likely to apply serial strategies to remembering the word list than controls; however, there did not appear to be a 40 DECISION MAKING IN OCD significant difference between groups for semantic clustering. In addition, participants with OCD showed greater interference than controls when a second word list was presented. significant difference between groups for semantic clustering. In addition, participants with OCD showed greater interference than controls when a second word list was presented. Subjective decision making. To examine subjective decision making between groups, an independent samples t-test was conducted using the self-report FIS. Despite the lack of differences between groups on objective measures, the self-report measure revealed statistically significant differences between groups, t(42) = -5.12, p < .001, with a large effect, r = .62. Participants with OCD (M = 50.50, SD = 9.10) reported greater levels of subjective difficulty with decision making than controls (M = 35.91, SD = 9.79). Scores on both the BAI (r = .65) and BDI-II (r = .64) were significantly correlated with FIS scores (p < .001); therefore, to see if diagnostic status would maintain significance even if depression and anxiety were controlled for, a hierarchical linear regression was conducted. Given the significant differences between groups, ethnicity was also included in the model to see if diagnostic group status could explain the model above and beyond these factors. As ethnicity is a demographic characteristic, it was entered into the model first. The second level consisted of clinical symptoms, including depression and anxiety. The third level of the model was the variable of interest, diagnostic group status. Results revealed that ethnicity did not account for any significant variance in self-reported indecision, F(1, 40) = 1.87, p = .179. Although symptoms of anxiety and depression accounted for 45% of the variance in the model, R2= .45, adding group status significantly improved the fit of the model, F(4, 37) = 10.36, p < .001. Incorporating group status into the model accounted for 53% of the variance in self-reported indecision, R2= .53, which is an increase of 8% from depression and anxiety alone. Thus, it appears that neither ethnicity, anxiety nor depression, had an impact on whether individuals with OCD reported greater perceived difficulty with decision making than controls. 41 DECISION MAKING IN OCD Relationship between objective and subjective decision making. Pearson's correlation coefficients were computed to examine the relationship between decision-making measures. See Table 6 for a summary. Results revealed no statistically significant associations between objective decision-making outcomes on the IGT, IST and CGT, and self-reported indecision. Moderator analyses were conducted to further examine the relationship between decision- making measures. Self-reported indecision, and objective decision making (as measured by IGT Net) were the variables of interest, with diagnostic group status proposed as a moderator. The analyses showed that diagnostic group status significantly moderated the relationship between IGT Net and FIS, F(3, 39) = 11.83, p = .001, R = .69. β = .22, t(1, 39) = 1.99, p = .040. Specifically, introducing diagnostic group status into the model demonstrated that as self- reported indecisiveness increased, performance on the IGT improved in those with OCD (r = .45, p = .034). The relationship between FIS and IGT scores was nonsignificant for controls (r = -.17, p = .438). Overall, results indicate poor convergent validity between objective and subjective measures of decision making, as participants with OCD reported perceived difficulties with decision making, despite the lack of findings to indicate impairments in objective decision making. Relationship between self-reported intolerance of uncertainty and decision-making performance. Pearson's correlation coefficients were computed and independent samples t-tests were conducted in order to examine the association between OCD, intolerance of uncertainty (IUS), and decision making. For intolerance of uncertainty scores, Levene’s test, F(1, 41) = 4.96, p = .032, was significant, indicating that the assumption of homogeneity of variances was not met. Therefore, an analysis that did not assume homogeneity of variance was adopted, which revealed a significant group effect of intolerance of uncertainty, t(36.75) = -2.78, p = .009. Relationship between self-reported intolerance of uncertainty and decision-making 42 DECISION MAKING IN OCD DECISION MAKING IN OCD Participants with OCD (M = 74.52, SD = 15.51) reported greater intolerance of uncertainty than controls (M = 57.86, SD = 23.26). Participants with OCD (M = 74.52, SD = 15.51) reported greater intolerance of uncertainty than controls (M = 57.86, SD = 23.26). Further to this, results revealed that self-reported indecisiveness was significantly correlated with intolerance of uncertainty, r = .62, p = .001, with approximately 39% of the variance accounted for (r2 = .39). As previous research suggested that intolerance of uncertainty may underlie difficulties with decision making, this was examined using linear regression. Intolerance of uncertainty scores significantly predicted perceived difficulty with decision making, b =.616, t(42) = 5.01, p <.001. Intolerance of uncertainty also explained 38% of the variance in self-reported indecision scores, R2 =.38, F(1, 41) = 25.11, p <.001. This indicated that as intolerance of uncertainty scores increased by 1 standard deviation, self-reported indecision also increased by 0.616 of a standard deviation. There were no significant associations between intolerance of uncertainty and objective measures of decision making. In summary, results indicated that intolerance of uncertainty is likely related to perceived difficulties with decision making reported in those with OCD. Further to this, results revealed that self-reported indecisiveness was significantly correlated with intolerance of uncertainty, r = .62, p = .001, with approximately 39% of the variance accounted for (r2 = .39). As previous research suggested that intolerance of uncertainty may underlie difficulties with decision making, this was examined using linear regression. Self-reported impulsivity related to decision-making performance. Pearson's correlation coefficients were computed and independent samples t-tests were conducted in order to examine the association between OCD, self-reported impulsivity (BIS-II), and decision making. Self-reported impulsivity was significantly greater in participants with OCD (M = 64.86, SD = 13.61) than in controls (M = 57.36, SD = 10.11), t(41) = -2.06, p = .046. The BIS-II was broken down into components of impulsivity based on subscales of the measure, which also resulted in significant differences between groups. Individuals with OCD reported greater nonplanning impulsivity (M = 25.91, SD = 5.86 ) than controls (M = 21.32, SD = 4.69), t(41) = - 2.84, p =.007. Greater attentional impulsivity in those with OCD (M = 18.62, SD = 4.40) as Self-reported impulsivity related to decision-making performance. Pearson's 43 DECISION MAKING IN OCD compared to controls (M = 15.59, SD = 5.97) approached significance, t(41) = -1.89, p =.066. There were no differences between groups on the third factor related to motor impulsivity, t(41)= 0.09, p = .931. There were no differences between groups on the third factor related to motor impulsivity, t(41)= 0.09, p = .931. Self-reported impulsivity was also examined in relation to self-reported indecision. Self-reported impulsivity was also examined in relation to self-reported indecision. Results revealed that self-reported indecisiveness was significantly correlated with impulsivity, r = .35, p = .022, with approximately 12% of the variance accounted for (r 2 = .12). As previous research suggested that a lack of premeditation (nonplanning impulsivity), and a lack of perseverance (attentional-impulsivity) is related to decision-making, this was examined using linear regression. Nonplanning impulsivity significantly predicted perceived difficulty with decision making, b =.352, t(41) = 2.41, p = .020. Nonplanning impulsivity also explained 12% of the variance in self-reported indecision scores, R2 =.12, F(1, 41) = 5.82, p = .020. Results also revealed that attentional impulsivity significantly predicted perceived difficulty with decision making, b =.447, t(41) = 3.20, p = .003. Nonplanning impulsivity also explained 20% of the variance in self-reported indecision scores, R2 =.20, F(1, 41) = 10.22, p = .003. The objective decision-making measures showed a significant moderate association between performance on the IGT and self-reported impulsivity r = .318, p = .038. Upon further examination, results revealed that attentional impulsivity was moderately associated with IGT performance, r = .386, p = .011, but nonplanning impulsivity was not, r = .206, p = .186. No associations were found between total impulsivity scores, subscales of impulsivity, and the variables on the CGT or the IST. Overall, there appeared to be a relationship between self- reported impulsivity and decision making, specifically as measured by the IGT. Despite the fact that the IST is a decision-making task that examines impulsivity, no relationship between self- reported impulsivity and the IST was noted. The objective decision-making measures showed a significant moderate association between performance on the IGT and self-reported impulsivity r = .318, p = .038. Upon further examination, results revealed that attentional impulsivity was moderately associated with IGT performance, r = .386, p = .011, but nonplanning impulsivity was not, r = .206, p = .186. DECISION-MAKING IN OCD DECISION-MAKING IN OCD DECISION MAKING IN OCD 44 DECISION MAKING IN OCD ure FIS IGT Net CGT p of Rational Decisions CGT Del- iberation time (s) CGT p of Quality Decisions CGT p of Risk Taking IST M Boxes Opened (1) IST M Boxes Opened (2) IST Samplin g Errors (1) IST Sampling Errors (2) IST Total Trials Correct (1) Net 20 x Measure FIS IGT Net CGT p of Rational Decisions CGT Del- iberation time (s) CGT p of Quality Decisions CGT p of Risk Taking IST M Boxes Opened (1) IST M Boxes Opened (2) IST Samplin g Errors (1) IST Sampling Errors (2) IST Total Trials Correct (1) IGT Net .20 x CGT p of Rational Decisions -.02 -.05 x CGT Deliberation time (s) -.09 -.05 .04 x CGT p of Quality Decisions -.10 -.01 .22 -.18 x CGT p of Risk Taking .03 -.02 .97** .02 .33* x IST M Boxes Opened (1) .24 -.06 -.27 -.11 -.04 -.17 x IST M Boxes Opened (2) .28 .03 -.53*** -.10 -15 -.43** .77*** x IST Sampling Errors (1) -.16 .17 .19 .12 .10 .13 -.81*** -.50** x IST Sampling Errors (2) -.29 -.22 .43** .03 .26 .32^ -.63*** -.83*** .44** x IST Total Trials Correct (1) .08 -.14 -.12 -.07 -.05 -.07 .75*** .44** -.95*** -.44** x IST Total Trials Correct (2) .23 .22 -.36* .03 -.20 -.25 .71*** .82*** -.53*** -.94*** .52*** Note. p = Probability; M = Mean; s =Seconds; FIS = Frost Indecisiveness Scale; IGT = Iowa Gambling Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; 1 = no cost condition; 2 = cost condition; * p <.05; **=.01;***p < .001. Note. p = Probability; M = Mean; s =Seconds; FIS = Frost Indecisiveness Scale; IGT = Iowa Gambling Task; CGT = Cambridge Gambling Task; IST = Information Sampling Task; 1 = no cost condition; 2 = cost condition; * p <.05; **=.01;***p < .001. 45 Other Neuropsychological Measures A summary of other neuropsychological measures used in the study to examine motor speed, cognitive speed, response inhibition and cognitive inhibition are presented in Table 7. All of these outcomes were examined using independent samples t-tests. Results revealed no statistically significant differences between OCD and control participants on both psychomotor speed, as measured by Finger Tapping, for the dominant hand, t(42) = 1.55, p = .128, and nondominant hand, t(42) = 1.84, p =.072. Similarly, there were no differences between groups for cognitive information processing speed, as measured by the Digit Symbol Test, t(42) = 1.31, p = .199. This indicated that participants with OCD are not slower than controls when processing information. In terms of response inhibition, no significant differences in performance on the Stop- Signal Paradigm, t(42) = -0.27, p= .789, as measured by SSRT, were found between groups. Furthermore, an examination of cognitive inhibition, using the Stroop Color-Word Interference Test, indicated no significant differences between OCD and control participants on the word trial, t(42) = 1.40, p = .168, colour trial, t(42) = -0.06, p = .955, or the interference trial, t(42) = - 0.07, p = .944. These results indicated that participants with OCD did not demonstrate difficulty with motor or cognitive inhibition as compared to controls. 46 DECISION-MAKING IN OCD Other Neuropsychological Measures Outcome OCD (n = 22) M (SD) Control (n = 22) M (SD) Statistic p Effect Size (r) Finger Tapping Test Dominant Hand 40.31 (8.66) 43.94 (6.70) t(42) = 1.55 .128 .23 Nondominant Hand 37.06 (6.78) 40.88 (6.99) t(42) = 1.84 .072 .27 Digit Symbol # of Symbols 70.00 (19.73) 78.83 (20.09) t(42) = 1.31 .199 .20 Stop-Signal Paradigm SSRT 329.00 (108.40) 318.69 (129.40) t(38) = -0.27 .789 .04 Stroop Color-Word Interference Test Word 45.79 (9.35) 50.21 (11.44) t(42) = 1.40 .168 .21 Colour 44.49 (8.28) 44.35 (8.16) t(42) = -0.057 .955 .01 Word/colour interference 49.73 (10.44) 49.50 (10.86) t(42) = -0.071 .944 .01 Note. M = Mean; SD = Standard deviation; SSRT = Stop signal reaction time; Due to technical difficulties, n = 18 for participants with OCD on the Stop-Signal Paradigm. 47 Discussion The primary purpose of this study was to examine decision making in individuals with OCD as compared to healthy control participants. A review of the findings suggested that some of the a priori hypotheses were supported by the results. Overall, no significant differences between groups on measures of objective decision making were found (IGT, IST and CGT). Performance on a memory task (CVLT-II) as a proxy for organizational strategies indicated differences between groups on serial clustering strategies and the interference caused by the presentation of a second word list. Furthermore, the results of the self-report measure of decision making (FIS) supported the clinical phenomena of complaints of indecisiveness in OCD. However, the self-report measure showed poor convergent validity with the objective measures. Although the presence of decision making deficits remains inconclusive in this population, this study provided evidence that individuals with OCD have perceived deficits in decision making, which likely has implications related to treatment for OCD patients. It is suggested that rather than targeting decision-making ability, treatment should focus on the beliefs and perceptions about one's decision-making ability. Alternatively, it is possible that the objective measures of decision making are not capturing the deficits, as these measures may be tapping a different construct than that related to the subjective measure. Hypothesis 1 stated that participants with OCD would demonstrate impairments on the IGT as compared to controls. This hypothesis was not supported by the findings of the current study. An analysis across blocks, and a comparison between groups on the Net outcome, Hypothesis 1 stated that participants with OCD would demonstrate impairments on the IGT as compared to controls. This hypothesis was not supported by the findings of the current study. An analysis across blocks, and a comparison between groups on the Net outcome, indicated that all participants made comparable decisions on the task. Both groups learned to choose cards from the advantageous decks (C and D) and to avoid the disadvantageous decks (A and B) across blocks. These results are supported by some previous research that found no indicated that all participants made comparable decisions on the task. Both groups learned to choose cards from the advantageous decks (C and D) and to avoid the disadvantageous decks (A and B) across blocks. These results are supported by some previous research that found no 48 DECISION-MAKING IN OCD differences in IGT performance between individuals with OCD and controls (Dittrich et al., 2011; Lawrence et al., 2006; Nielen et al., 2002), but are inconsistent with other published studies (Cavallaro et al., 2003; Cavedini et al., 2002; Da Rocha et al., 2011; Long et al., 2012; Starcke et al., 2010; Starcke et al., 2010). Previous research has demonstrated differences in neuronal activity in the orbitofrontal cortex in OCD patients as compared to healthy controls (Ping et al., 2013; Stern et al., 2013; Whiteside et al., 2004). The orbitofrontal cortex has been indicated as important during the context of decision making. For example, impairment on the IGT has been suggested as indicative of abnormal orbitofrontal cortex functioning (Cavedini et al., 2002). Although there appears to be a relationship between difficulties with decision making and OCD, not all studies have consistently found deficits on decision-making measures. Therefore, future neuroimaging studies are warranted to better decipher the neuropsychological functioning of decision making in OCD, and the extent of involvement of the orbitofrontal cortex. One possible reason for the inconsistent findings using the IGT may be explained by the measure. It is possible that the ecological validity of the IGT is problematic due to the poor reliability and construct validity that has been reported for the measure (Brand et al., 2005; Brand et al., 2007; Buelow & Suhr, 2009). Most studies used the IGT as a measure of ambiguous decision making; however, it is unclear if the test taps both ambiguous and risky decision making. Thus, contradictory findings may be explained by concerns with the measure that is often used to examine decision making. The failure of the current study to find a statistically significant difference between OCD and control participants on the IGT may also be due to the older mean age of participants with OCD (M = 37.23, SD = 13.53), relative to other studies. Specifically, Da Rocha and colleagues 49 49 DECISION-MAKING IN OCD (2011) found that participants with OCD with a mean age of 28.40 (SD = 14.12) had significant impairment on the IGT, compared to controls. In contrast, the study by Dittrich and colleagues used older participants with OCD (M = 41.80, SD = 12.50; 2011) and, like the current study, also did not find impairments on the IGT. There is some evidence that performance on the IGT improves with age, until older adulthood (mid-50s to 60s) when performance declines (Beitz, Salthouse, & Davis, 2014). Therefore, age may be an important factor contributing to the inconsistent findings in the literature. Another possible explanation for the differences in findings between studies using the IGT may be severity of symptoms. According to the YBOCS, participants with OCD in the current study were considered to have a moderate severity of symptoms as a group. Similarly, in the study by Dittrich and colleagues (2011), mean symptom severity ranged from mild (for those with sexual/religious obsessions) to moderate (for aggressive symptoms), and no significant difference between groups was found. In contrast, participants in the study by Da Rocha and colleagues (2011) had a mean symptom presentation falling in the severe range on the YBOCS, and participants with OCD demonstrated poorer performance than controls. It has been previously reported that OCD patients with more severe symptoms demonstrated poorer performance on the IGT (Nielen et al., 2002). Thus, there is some indication that the level of severity of obsessive-compulsive symptoms is important to consider when examining decision- making deficits. As no impairment on this task was found in the current study, it is possible that in more severe cases of OCD, impairment on the IGT would be reported. Thus, future studies should examine symptom severity as a variable of interest in order to better determine whether this factor contributes to performance decrement. 50 50 DECISION-MAKING IN OCD Furthermore, inconsistent results may also be due to the clinical heterogeneity of OCD, as previous research has revealed that symptom subtypes (e.g., primary symptoms of contamination/cleaning, symmetry/ordering, obsessions/checking and hoarding) are associated with certain neurocognitive impairments (Leckman et al., 1997; McKay et al., 2004). For example, Lawrence and colleagues (2006) reported impaired performance on the IGT in those with primary hoarding symptoms compared to healthy controls, and compared to patients low on hoarding symptoms (Lawrence et al., 2006). Due to the small sample size of the current study, an analysis comparing decision-making performance between subtypes of OCD was not possible, and thus future studies should attempt to examine subtypes as a possible factor in whether deficits in decision making occur. Overall, given the large inconsistencies in performance on the IGT in the literature, more research is needed to determine if deficits on this task exist. Hypothesis 2 stated that participants with OCD would demonstrate impairments on the IST (a measure that taps decision making and impulsivity) as compared to controls. This was not supported. Since impulsivity was suggested as an important component of the decision-making process (Bayard et al., 2011; Torregrossa et al., 2008), and previous research found greater perceived impulsivity in those with OCD compared to controls (Gupta et al., 2013; Summerfeldt et al., 2004), impaired performance on the IST was expected. The current study’s nonsignificant result was supported by Chamberlain, Fineberg, Blackwell, and colleagues (2007), where after controlling for multiple comparisons, no differences between groups on decision-making outcomes on the IST were reported. However, inconsistent with the current study, these authors also found that performance across conditions differed between OCD and control participants. These authors found that when an incentive to choose a majority colour earlier by revealing fewer boxes before making a decision was introduced in the second condition (cost condition), 51 51 DECISION-MAKING IN OCD current decision-making measures that tap organizational strategies. Although the IGT claims to require the use of organizational strategies on the task (Bechara et al., 1994), there are no dependent outcome variables to assess this component of decision making. As previous research has found that organizational strategies mediate performance on certain memory measures (Savage et al., 2000), the CVLT-II was employed to examine this component of decision making. Hypothesis 4 stated that strategic learning on the CVLT-II would be impaired in those with OCD. This hypothesis was partially supported by the results, as there were differences between groups on outcome variables that directly tapped organizational strategies. The two variables examined included serial clustering (attempting to remember the words in the order they were presented), and semantic clustering (attempting to group words by categories in order to facilitate memory). In the current study, participants with OCD were more likely to use serial clustering strategies than controls, as demonstrated by a medium effect size (r = .33). No Hypothesis 4 stated that strategic learning on the CVLT-II would be impaired in those with OCD. This hypothesis was partially supported by the results, as there were differences between groups on outcome variables that directly tapped organizational strategies. The two variables examined included serial clustering (attempting to remember the words in the order they were presented), and semantic clustering (attempting to group words by categories in order to facilitate memory). In the current study, participants with OCD were more likely to use serial clustering strategies than controls, as demonstrated by a medium effect size (r = .33). No significant differences between OCD and control participants were found on the use of semantic clustering strategies; however, an examination of the means found that participants with OCD used semantic strategies less than controls. The effect size for semantic clustering was close to the effect seen in serial clustering (r = .24), but was not large enough to bring about significant differences. Thus, it is possible that the current study was under powered to detect differences in semantic clustering between groups. Previous research has found that participants with OCD implemented semantic clustering strategies less than controls (Tukel et al., 2012), and it is possible that this would have been found in the current study with a larger sample size. DECISION-MAKING IN OCD controls adjusted their behaviour and opened fewer boxes than in the first condition, which was not the case for the OCD group (Chamberlain, Fineberg, Blackwell, et al., 2007). The authors concluded that this result was likely representative of the cognitive rigidity often seen in OCD, and thus may not be indicative of impaired decision making. Given that only two studies (including the current study) have examined decision making in OCD using the IST, future studies are warranted. Hypothesis 3 stated that participants with OCD would perform similarly to controls on the CGT. This hypothesis was supported. Analyses that compared groups on the four dependent variables (probability of rational decisions, deliberation time, probability of quality decisions, and probability of risk taking behaviour) demonstrated comparable performance between OCD and control participants. Results of the current study were consistent with other studies that examined risky decision making using the CGT, GDT, and high-risk scenarios (Chamberlain, Fineberg, Blackwell, et al., 2007; Chamberlain, Fineberg, Menzies, et al., 2007; Foa et al., 2003; Long et al., 2012; Starcke et al., 2009; Starcke et al., 2010; Watkins et al., 2005), which suggested intact decision-making performance in OCD during risky decision making. However, Dittrich and colleagues (2011) reported impairment on rational risky decision making for those with symmetry/ordering and sexual/religious symptoms as compared to other symptom subtypes. Given that the current study did not examine subtypes of OCD, it is possible that different symptom presentations between the samples contributed to the inconsistent findings. Overall, previous literature appeared to largely support the findings that individuals with OCD did not demonstrate impairments on tasks of risky decision making. It has been suggested that organizational strategies are important in the context of decision making, as individuals are required to evaluate alternatives during this process (Olley et al., 2007). Despite this, there are no 52 DECISION-MAKING IN OCD Given that serial clustering is not as beneficial as semantic clustering in improving performance (Delis et al., 2000), it is interesting that participants with OCD demonstrated a greater use of this 53 DECISION-MAKING IN OCD strategy than controls. Since serial clustering strategy is not a primary outcome on the CVLT-II, future research should aim to report this finding in the results. Interestingly, participants with OCD experienced significantly more interference than controls when remembering the first list of words after the presentation of the second list of words, which may be related to executive functioning. This result supported previous research that suggested deficits in executive functioning in OCD patients (Olley et al., 2007; Tukel et al., 2012). Overall, these findings suggested that organizational strategies involved in verbal learning may be important factors to consider in future research. It should be noted that the current study examined strategic learning using a verbal memory measure. The Rey-Osterrieth Complex Figure Test (RCFT; Meyers & Meyers, 1995) tests visuospatial memory, which asks participants to copy and recall a complex figure. The RCFT also taps organizational strategies in the context of strategic learning by examining the manner in which participants copy the figure, as well as subsequent performance on memory outcomes. Savage and colleagues (2000) reported that organizational strategies also mediated performance on nonverbal memory tasks in OCD. In addition, it was reported that greater organizational strategies during the copy condition resulted in improved performance on the RCFT. Thus, future research comparing verbal and nonverbal organizational strategies in OCD would likely add to the limited literature on strategy in decision making. Hypothesis 5, which stated that self-reported decision making as measured by the FIS would be poorer in those with OCD compared to controls, was supported. This result remained significant even after considering the variance accounted for by ethnicity, anxiety, and depression. This result makes theoretical sense, given that indecisiveness is often reported as an issue in those with OCD (Steketee & Frost, 2003). Findings from the current study add to the 54 DECISION-MAKING IN OCD limited literature on subjective reports of decision making in OCD patients. Taken together, the findings indicated that perception of decision-making ability may be impaired in OCD, rather than having actual deficits in decision making on objective measures. Alternatively, it may be that we are not measuring the same construct or process as that being captured by the self-report measure. A functional Magnetic Resonance Imaging study by Stern and colleagues (2013) indicated that individuals with OCD only demonstrated impairments in decision making as compared to controls when there was no uncertainty inherent in the task. When uncertainty was present as part of the measure, participants with OCD performed similarly to controls. The authors concluded that greater activation in the limbic and paralimbic brain regions (important for emotion and behaviour) during decision making may have exacerbated the subjective experience of uncertainty and indecision in OCD, even when there was no uncertainty inherent in the task. These authors concluded that targeting the perception of decision-making difficulties and associated anxiety during decision making should be a focus of treatment, rather than focusing on "how to" make decisions. Given the lack of objective deficits in decision making and the self-reported difficulties with decision making, the findings of the current study supported this previous research. Hypothesis 6, which stated that subjective and objective decision-making performance would be related, was partially supported. Correlational analyses computed between the various decision-making measures demonstrated that there was very little relationship between these measures. In fact, the only significant correlations between measures were the probability of rational decisions on the CGT, and three of the dependent variables on the cost condition of the IST: mean number of boxes opened, sampling errors, and total trials correct. This suggested that 55 DECISION-MAKING IN OCD although all of the measures are supposed to be assessing decision making, they are likely tapping different components of this complex construct. although all of the measures are supposed to be assessing decision making, they are likely tapping different components of this complex construct. To further examine the association among decision-making measures, whether group status was a significant moderator of the relationship between the IGT and FIS, was examined. Within the OCD group, performance on the IGT increased as subjective indecision increased, which was contrary to expectations. For controls, there was no significant relationship between scores on the IGT and FIS. A possible reason for these results was based on the perception of difficulty with decision making that is examined in the self-report measure (i.e., how decision- making performance was perceived by the patient may not be related to actual decision-making ability). Again, this finding further supported the idea that the maladaptive beliefs about decision-making ability should be the target in OCD treatment, rather than providing specific decision-making strategies. Another explanation for the unexpected relationship between the FIS and IGT may be due to the fact that objective and subjective tests tap different constructs. The self-report FIS inquired about the process of decision making (e.g., "I do not get assignments done on time because I cannot decide what to do first," "It seems that deciding on the most trivial thing takes me a long time"), and anxiety associated with decision making (e.g., "I become anxious when making a decision," "I often worry about making the wrong choice"). The FIS also attempted to capture what the decision-making process was like in real-life settings (e.g., "When ordering from a menu, I usually find it difficult to decide what to get," "I have a hard time planning my free time"). These questions were in contrast to the behavioural neuropsychological measures of decision making that largely rely on gambling-like scenarios, as in the IGT, IST, and CGT. Although tests like the IGT purport to mimic real-life decision making (Bechara et al., 1994), it 56 DECISION-MAKING IN OCD is possible that these objective measures may not evaluate the construct in this manner as well as previously thought. Overall, results on the IGT, CGT and IST suggested intact decision making in OCD patients. The findings from the current study also demonstrated that participants with OCD, as a group, had differences in organizational strategies during the memory task. A perceived difficulty with decision making was present in the OCD sample, suggesting poor convergent validity between objective and subjective tests. Decision making is a complex neuropsychological process that involves many components and can be affected by many factors. Thus, a task designed to incorporate various components of decision making, or more studies examining measures that tap different components, are needed. Further to this, future research that highlights the process of making decisions, rather than outcomes (Raynard et al., 1997), is warranted. In addition, perhaps the development of objective measures of decision making with greater ecological validity would provide better insight into this process in OCD. Hypothesis 7 stated that there would be a relationship between decision-making outcomes and intolerance of uncertainty in OCD. This hypothesis was supported. Intolerance of uncertainty is a negative reaction to uncertain or unforeseen circumstances or events (Dugas & Koerner, 2005). Evidence has suggested intolerance of uncertainty as a maintaining factor contributing to various anxiety and depressive disorders, including GAD, OCD, social anxiety disorder, panic disorder, agoraphobia, and depression (McEvoy, & Mahoney, 2011; 2012). In the current study, participants with OCD reported greater intolerance of uncertainty than controls, which was consistent with previous research (Gentes & Ruscio, 2011; Fergus & Wu, 2010). 57 57 DECISION-MAKING IN OCD In the current study, intolerance of uncertainty predicted perceived difficulty with decision making, indicating that it is possible that this transdiagnostic construct of anxiety may be related to the subjective indecision in OCD; however, intolerance of uncertainty was not related to objective decision-making performance. The results of previous research on uncertainty in decision making (Stern et al., 2013), and the findings that perceived decision- making ability was poor in OCD, have important treatment implications for OCD patients. As mentioned, it can be suggested that maladaptive beliefs about one's ability to make decisions and tolerating uncertainty should be a focus in the treatment of OCD, rather than an explicit focus on decision making. One of the primary targets in cognitive-behavioural therapy (CBT) is to understand unrealistic thoughts, and appraisals of those thoughts, that contribute to symptoms of anxiety and associated dysfunctional emotions (Clark, 2004). This is done by first identifying the maladaptive beliefs and cognitive processes (Clark, 2004). Patients are taught, through goal- oriented therapy, that these thoughts and the associated anxiety are not harmful and can be tolerated. Behavioural techniques such as exposure and response prevention (ERP) have been shown to be highly effective in the treatment of OCD (Clark, 2004). ERP is designed to focus on avoidance (as compulsive acts are often used to reduce anxiety from associated obsessions) by exposing patients to their fears. A combination of both cognitive and behavioural techniques is often used in the treatment of OCD, and tailoring these strategies to the cognitive biases unique to the patient may be important. Thus, if a patient endorses high intolerance of uncertainty and perceived difficulties with decision making, CBT strategies may be useful in targeting these problems. Hypothesis 7 further stated that there would be a relationship between decision-making In the current study, intolerance of uncertainty predicted perceived difficulty with decision making, indicating that it is possible that this transdiagnostic construct of anxiety may be related to the subjective indecision in OCD; however, intolerance of uncertainty was not related to objective decision-making performance. The results of previous research on Hypothesis 7 further stated that there would be a relationship between decision-making outcomes and impulsivity, and this was also supported. Although there were no differences 58 DECISION-MAKING IN OCD between groups on the IST, participants with OCD reported greater perceived impulsivity than controls on the self-report BIS-II, which supported previous research (Gupta et al., 2013; controls on the self-report BIS-II, which supported previous research (Gupta et al., 2013; Summerfeldt et al., 2004). Consistent with the study by Gupta and colleagues (2013), the results of the current study revealed that participants with OCD reported higher nonplanning impulsivity compared to controls, which indicated poorer planning and less enjoyment on difficult tasks (Patton et al., 1995). There was also a trend toward higher attention-impulsiveness (involved in making quick decisions), but no differences in motor impulsiveness (involved in engaging in behaviours without thinking). Scores on both the nonplanning and attentional impulsivity subscales of the BIS-II predicted scores of self-reported indecision on the FIS, which suggested that severity of certain components of impulsivity may be related to perceived difficulties with decision making. In the current study, as overall self-reported impulsivity increased, performance on the IGT improved, which was contrary to expectations. In addition, the results showed that the attentional facet of impulsivity was most related to performance on the IGT, an effect that was of moderate size and statistically significant. As scores on attentional impulsivity increased, so did performance on the IGT. However, no associations between nonplanning or motor impulsivity and IGT performance were found. Previous studies found that those with high impulsivity, or high scores on the nonplanning subscale, had poorer performance on the IGT than controls, and those with low impulsivity, or low nonplanning scores, had better performance (Franken et al., 2008; Zermatten et al., 2005). One possible reason for the inconsistent results is that previous research used healthy nonclinical samples while the current study used OCD patients. However, the observed positive relation between impulsivity and decision-making performance was in line with an earlier study conducted by Franken and Muris (2005). This 59 DECISION-MAKING IN OCD study found no association between decision-making performance on the IGT and dysfunctional impulsivity, but the researchers indicated that a higher level of a reward sensitivity trait associated with impulsivity resulted in improved performance on the IGT. Therefore, the association between high impulsivity and poorer decision-making performance did not appear to be consistent across studies, given the findings of a positive association between impulsivity and IGT performance, and the lack of association between impulsivity and other decision-making measures. A possible reason for the inconsistent results is the differences in the measures used. For example, Franken and colleagues (2008) used one of the computerized versions of the IGT that progressively increased the cost of punishment, whereas the current study used the card version where the cost of punishment did not increase. In addition, multiple self-report questionnaires were used to assess impulsivity. For example, on the Impulsiveness Scale used by Franken and colleagues (2008), scores were rated on a dichotomous yes/no scale that focused on behavioural control and delay in gratification. Zermatten and colleagues (2005) used the French version of the Impulsive Behavior Scale, which measured items on a 4-point Likert scale, similarly to the BIS-II. It is possible that different versions of the questionnaire resulted in differences in the findings between studies, and thus more research in the area of impulsivity and decision making is needed. The inclusion of multiple measures of impulsivity in future studies appears warranted in order to determine if different approaches to measuring the construct relate to performance on various decision-making tasks. In addition, differentiating between different subtypes of OCD to further understand the relationship between impulsivity and symptoms of OCD is important (Zermatten et al., 2008), and future research should also investigate this. 60 DECISION-MAKING IN OCD An examination of other neuropsychological measures was included in the current study in order to rule out confounding variables and other possible explanations of the findings. It was found that participants with OCD and control participants did not differ on level of premorbid functioning as measured by a proxy for intelligence, which indicated that the findings of the study were not related to intelligence, as measured by reading ability. In addition, participants with OCD in the current sample were not slower than controls during tasks, as indicated by similar performance between groups on measures of psychomotor and cognitive information processing speed. Finally, findings indicated that performance between OCD and control participants did not differ on measures of motor or cognitive inhibition, and thus it is not likely that these factors influenced the results. This study contributed to the limited literature on decision making in OCD, which is currently fraught with inconsistent results. This study employed a large battery of tests composed of clinical and neurocognitive measures to assess for comorbidities, severity of clinical This study contributed to the limited literature on decision making in OCD, which is currently fraught with inconsistent results. This study employed a large battery of tests composed of clinical and neurocognitive measures to assess for comorbidities, severity of clinical symptoms, and cognitive impairment. This was the first study to evaluate decision making in OCD using multiple measures that assessed proposed components of the construct, including ambiguity, impulsivity, perception of risk, and organizational strategies. It was also presumed to be the first study to compare subjective and objective decision making in OCD, using a multi- modal method of assessment (i.e., tests and questionnaires), and to measure decision making using three objective measures. Further to this, all participants had a principal diagnosis of OCD, and those for whom criteria were not met (i.e., those with hoarding disorder only) were excluded from the sample. It is also unlikely that depression contributed to the results of the current study as: 1) only patients with mild or moderate depressive severity were included in the study; 2) where high correlations between cognitive performance and depression were found, a 61 61 DECISION-MAKING IN OCD research should attempt to match participants on ethnicity to ascertain if these variables impact the results. research should attempt to match participants on ethnicity to ascertain if these variables impact the results. The majority of clinical participants had comorbid mood and anxiety disorders, which is common is OCD; however, given the small sample size, we were unable to compare performance across these groups. As the study was not adequately powered to analyze differences between subtypes of OCD, future research should also compare performance on decision-making measures across theoretically supported subtypes. Studies with clinical control groups (e.g., primary depression, other anxiety disorders) and subtypes of OCD may help elucidate differences in decision making between these clinical presentations. Despite the large number of statistical analyses performed, a Bonferroni correction for multiple comparisons was not used. As the statistical tests completed were designed to examine neuropsychological performance based on a-priori hypotheses, each measure was viewed as a separate analysis for statistical purposes, and p < .05 was reported. DECISION-MAKING IN OCD hierarchical Regression was performed to control for these symptoms; and 3) no correlations were found between depressive scores and other significant differences in cognitive functioning. The results of the current study suggested two possible explanations for the findings: 1) the need to target beliefs about decision-making ability, rather than actual deficits, or 2) the need to develop a measure of decision making that objectively captures the process of decision making, as we may not be currently tapping the construct in the manner that we thought. This finding has implications for the treatment of individuals with OCD, as a better understanding of decision-making difficulties unique to a patient could provide diagnostic information to add to the case conceptualization of that patient. Despite its many strengths, the current study was not without limitations. First, due to technical difficulties, data from participants in both the OCD and control groups were lost on the IST, CGT and Stop-Signal Paradigm, and thus results on these measures were analyzed with a sample size that was smaller than that estimated for power a priori. Given the small sample size, it is possible that there was not adequate power to detect differences in cognitive performance. Thus, the results should be interpreted with caution, and findings should be replicated with a larger sample size. Some other potentially confounding variables, such as medication, were not controlled for in this study due to the small sample size. Since the clinical participants were largely recruited from an outpatient hospital setting, varying levels of medication status were present in the sample. Furthermore, ethnicity was significantly different between groups, with a higher proportion of individuals identifying as Caucasian in the OCD sample and a higher proportion of individuals identifying as Asian in the control sample. However, this difference did not appear to affect the results on decision-making tasks, as shown by the significant differences between groups on subjective decision making, even after ethnicity was controlled for. Future 62 Conclusion The purpose of the current study was to examine the neurocognitive process of decision making in individuals with OCD, using multiple measures. The presence of decision-making deficits in this population remained inconclusive, due to poor convergent validity between objective and subjective measures of decision making. Nevertheless, this study provided support that individuals with OCD have perceived deficits in decision making, which has implications related to treatment in this population. Furthermore, differences in the application of organizational strategies on a memory measure were reported, but more research in this area is warranted. Intolerance of uncertainty and impulsivity may be related to perceived difficulties with decision making in OCD, but more research examining these constructs is also needed. The 63 DECISION-MAKING IN OCD results of this study have added to the limited literature on neurocognition and decision making in OCD, and multiple suggestions for future research were discussed. 64 64 References Abramowitz, J. S., & Deacon, B. J. (2006). Psychometric properties and construct validity of the Obsessive-Compulsive Inventory- Revised: Replication and extension with a clinical sample. Journal of Anxiety Disorders, 20, 1016-1035. doi:10.1016/j.janxdis.2006.03.001 American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders, 5th ed. Washington, DC: Author. disorders, 5th ed. Washington, DC: Author. Bayard, S., Raffard, S., & Gely-Nargeot, M. (2011). Do facets of self-reported impulsivity predict decision-making under ambiguity and risk? Evidence from a community sample. Psychiatry Research, 190, 322-326. doi:10.1016/j.psychres.2011.06.013 Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994). Insensitivity to future consequences following damage to human prefrontal cortex. Cognition, 50, 7-15. doi: 10.1016/0010-0277(94)90018-3 Beck, A.T., Epstein, N., Brown, G., & Steer, R.A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893-897. doi: 10.1037/0022-006X.56.6.893 Beck, A. T., & Steer, R.A. (1993). Beck Anxiety Inventory: Manual. San Antonio, TX: Pearson Assessment. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck Depression-II: Manual. San Antonio, TX: Pearson Assessment. Beitz, K. M., Salthouse, T. A., & Davis, H. P. (2014). Performance on the Iowa Gambling Task: From 5 to 89 years of age. Journal of Experimental Psychology: General, 143, 1677- 1689. doi: 10.1037/a0035823 65 DECISION-MAKING IN OCD Billieux, J., Gay, P., Rochat, L., & Van der Linden, M. (2010). The role of urgency and its underlying psychological mechanisms in problematic behaviours. Behaviour Research and Therapy, 48, 1085-1096. doi:10.1016/j.brat.2010.07.008 Brand, M., Fujiwara, E., Borsutzky, S., Kalbe, E., Kessler, J., & Markowitsch, H.J. (2005). Decision-making deficits of Korsakoff patients in a new gambling task with explicit rules-associations with executive functions. Neuropsychology, 19, 267–277. doi: 10 1037/0894 4105 19 3 267 Brand, M., Recknor, E. C., Grabenhorst, F., & Bechara, A. (2007). Decisions under ambiguity and decisions under risk: Correlations with executive functions and comparisons of two different gambling tasks with implicit and explicit rules. Journal of Clinical and Experimental Neuropsychology, 29, 86-99. doi: 10.1080/13803390500507196 Buelow, M., & Suhr, J. (2009). Construct validity of the Iowa Gambling Task. Neuropsychology Review, 19, 102-114. doi: 10.1007/s11065-009-9083-4 Burdick, K. E., Robinson, D G. Malhotra, A. K., & Szeszko, P. R. (2008). Neurocognitive profile analysis in obsessive-compulsive disorder. Journal of the International Neuropsychological Society, 14, 640-645. doi: 10.1017/S1355617708080727 Cambridge Cognition Ltd. (2006). Cambridge Neuropsychological Test Automated Battery (CANTAB). Author: Cambridge: UK. Cavallaro, R., Cavedini, P., Mistretta, P., Bassi, T., Angelone, S. M., Ubbiali, A., & Bellodi, L. (2003). Basal-corticofrontal circuits in schizophrenia and obsessive-compulsive 66 DECISION-MAKING IN OCD disorder: A controlled, double dissociation study. Society of Biological Psychiatry, 54, 437-443. doi:10.1016/S0002-3223(03)01814-0 Cavedini, P., Gorini, A., & Bellodi, L. (2006). Understanding OCD: Focus on decision-making. Neuropsychology Review, 16, 3-15. doi: 10.1007/s11065-006-9001-y Cavedini, P., Riboldi, G., D'Annucci, A., Belotti, P., Cisima, M., & Bellodi, L. (2002). Decision- making heterogeneity in obsessive-compulsive disorder: Ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia, 40, 205-211. doi: 10.1016/S0028-3932(01)00077-X Chamberlain, S, R., Fineberg, N. A., Blackwell, A. D., Clark, L., Robbins, T. W. & Sahakian, B. J. (2007). A neuropsychological comparison of obsessive-compulsive disorder and trichotillomania. Neuropsychologia, 45, 654–662. doi: 10.1016/j.neuropsychologia.2006.07.016 Chamberlain, S. R., Fineberg, N. A., Menzies, L. A., Blackwell, A. D., Bullmore, E. T., Robbins, T. W. & Sahakian, B. J. (2007). Impaired cognitive flexibility and motor inhibition in unaffected first-degree relatives of patients with obsessive-compulsive disorder. American Journal of Psychiatry, 164, 335–338. doi: 10.1176/appi.ajp.164.2.335 Clark, D. A. (2004). Cognitive-Behavioral Therapy for OCD. New York, NY: Guilford Press. Clark, L., Bechara, A., Damasio, H., Aitken, M. R. F., Sahakian, B. J. & Robbins, T. W. (2008). Differential effects of insular and ventrome-dial prefrontal cortex lesions on risky decision-making. Brain, 131, 1311–1322. doi: 10.1093/brain/awn066 Clark, L., Robbins, T. W., Ersche, K. D., & Sahakian, B. J. (2006). Reflection impulsivity in current and former substance users. Biological Psychiatry, 60, 515–522. doi: 10.1016/j.biopsych.2005.11.007 disorder: A controlled, double dissociation study. Society of Biological Psychiatry, 54, 437-443. doi:10.1016/S0002-3223(03)01814-0 disorder: A controlled, double dissociation study. Society of Biological Psychiatry, 54, 437-443. doi:10.1016/S0002-3223(03)01814-0 Cavedini, P., Gorini, A., & Bellodi, L. (2006). Understanding OCD: Focus on decision-making. Neuropsychology Review, 16, 3-15. doi: 10.1007/s11065-006-9001-y Cavedini, P., Gorini, A., & Bellodi, L. (2006). Understanding OCD: Focus on decision-making. Neuropsychology Review, 16, 3-15. doi: 10.1007/s11065-006-9001-y Cavedini, P., Riboldi, G., D'Annucci, A., Belotti, P., Cisima, M., & Bellodi, L. (2002). Decision- making heterogeneity in obsessive-compulsive disorder: Ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia, 40, 205-211. doi: 10.1016/S0028-3932(01)00077-X Chamberlain, S, R., Fineberg, N. A., Blackwell, A. D., Clark, L., Robbins, T. W. & Sahakian, B. J. (2007). A neuropsychological comparison of obsessive-compulsive disorder and trichotillomania. Neuropsychologia, 45, 654–662. doi: Chamberlain, S. R., Fineberg, N. A., Menzies, L. A., Blackwell, A. D., Bullmore, E. T., Robbins, T. W. & Sahakian, B. J. (2007). Impaired cognitive flexibility and motor inhibition in unaffected first-degree relatives of patients with obsessive-compulsive disorder. American Journal of Psychiatry, 164, 335–338. doi: 10.1176/appi.ajp.164.2.335 Clark, D. A. (2004). Cognitive-Behavioral Therapy for OCD. DECISION-MAKING IN OCD New York, NY: Guilford Press. Clark, L., Bechara, A., Damasio, H., Aitken, M. R. F., Sahakian, B. J. & Robbins, T. W. (2008). Differential effects of insular and ventrome-dial prefrontal cortex lesions on risky decision-making. Brain, 131, 1311–1322. doi: 10.1093/brain/awn066 Clark, L., Robbins, T. W., Ersche, K. D., & Sahakian, B. J. (2006). Reflection impulsivity in current and former substance users. Biological Psychiatry, 60, 515–522. doi: 10.1016/j.biopsych.2005.11.007 67 DECISION-MAKING IN OCD Da Rocha, F., Alvarenga, N., Malloy-Diniz, L., & Correa, H. (2011). Decision-making impairment in OCD as measured by the Iowa Gambling Task. Arquivos de Neuro- Psiquiatria, 69, 642-647. doi: 10.1590/S0004-282X2011000500013 Delis, D.C., Kramer, J.H., Kaplan, E., & Ober, B.A. (2000). California Verbal Learning Test: Second Edition. San Antonio, TX: Pearson Assessment. Dittrich, W. H., Johansen, T., Landrø, N. I. & Fineberg, N. A. (2011). Cognitive performance and specific deficits in OCD symptom dimensions: III. Decision-making and impairments in risky choices. German Journal of Psychiatry, 14,13–25. Dugas, M. J., & Koerner, N. (2005). Cognitive-behavioral treatment for generalized anxiety disorder: Current status and future directions. Journal of Cognitive Psychotherapy, 19, 61-81. doi: 10.1891/jcop.19.1.61.66326 Fadda, E., Sottocorno, S., Martoni, R., Galimberti, E., & Bellodi, L. (2011). No evidence for deficient set-shifting in obsessive-compulsive disorder. European Psychiatry, 26, 153. doi: 10.1016/S0924-9338(11)71864-8 Fergus, T. A., & Wu, K. D. (2010). Do symptoms of generalized anxiety and obsessive- compulsive disorder share cognitive processes? Cognitive Therapy and Research, 34, 168-176. doi: 10.1007/s10608-009-9239-9 68 DECISION-MAKING IN OCD First, M., Spitzer, R., Gibbon, M., Williams, J. (1997). Structured Clinical Interview for DSM- IV, Research Version. Washington, DC: American Psychiatric Press. Foa, E. B., Huppert, J., D., Leiberg, S., Langner, R., Kichic, R., Hajcak, G.,... Salkovskis, P.M. (2002). The obsessive-compulsive inventory: Development and validation of a short version. Psychological Assessment, 14, 485-496. doi: 10.1037/1040-3590.14.4.485 Foa, E. B., Mathews, A., Abramowitz, J. S., Amir, N., Przeworski, A., Riggs, D. S.,...Alley, A. (2003). Do patients with obsessive-compulsive disorder have deficits in decision- making? Cognitive Therapy and Research, 27, 431-445. doi: 10.1023/A:1025424530644 Fogel, J. (2003). An epidemiological perspective of OCD in children and adolescents. Canadian Child Adolescent Psychiatry Review, 12, 33-36. Franken, I. H. A., & Muris, P. (2005). Individual differences in decision-making. Personality and Individual Differences, 39, 159-253. doi: 10.1016/j.paid.2005.04.004 Franken , I. H. A., van Strien, J. W., Nijs, I., & Muris, P. (2008). Impulsivity is associated with behavioral decision-making deficits. Psychiatry Research, 158, 155-163. doi: 10.1016/j.psychres.2007.06.002 Frost, R. O., & Shows, D. L. (1993). The nature and measurement of compulsive indecisiveness. Behaviour Research and Therapy, 31, 683-692. doi: 10.1016/0005-7967(93)90121-A Gentes, E. L., & Ruscio, A. M. (2011). A meta-analysis of the relation of intolerance of uncertainty to symptoms of generalized anxiety disorder, major depressive disorder, and obsessive–compulsive disorder. Clinical Psychology Review, 31, 923-933. doi:10.1016/j.cpr.2011.05.001 First, M., Spitzer, R., Gibbon, M., Williams, J. (1997). Structured Clinical Interview for DSM- IV, Research Version. Washington, DC: American Psychiatric Press. Foa, E. B., Huppert, J., D., Leiberg, S., Langner, R., Kichic, R., Hajcak, G.,... Salkovskis, P.M. (2002). The obsessive-compulsive inventory: Development and validation of a short version. Psychological Assessment, 14, 485-496. doi: 10.1037/1040-3590.14.4.485 Foa, E. B., Mathews, A., Abramowitz, J. S., Amir, N., Przeworski, A., Riggs, D. S.,...Alley, A. (2003). Do patients with obsessive-compulsive disorder have deficits in decision- making? Cognitive Therapy and Research, 27, 431-445. doi: 10.1023/A:1025424530644 Fogel, J. (2003). An epidemiological perspective of OCD in children and adolescents. Canadian Child Adolescent Psychiatry Review, 12, 33-36. Franken, I. H. A., & Muris, P. (2005). Individual differences in decision-making. Personality and Individual Differences, 39, 159-253. doi: 10.1016/j.paid.2005.04.004 Franken , I. H. A., van Strien, J. W., Nijs, I., & Muris, P. (2008). Impulsivity is associated with behavioral decision-making deficits. Psychiatry Research, 158, 155-163. doi: 10.1016/j.psychres.2007.06.002 Frost, R. O., & Shows, D. L. (1993). The nature and measurement of compulsive indecisiveness. Behaviour Research and Therapy, 31, 683-692. doi: 10.1016/0005-7967(93)90121-A Gentes, E. L., & Ruscio, A. M. (2011). A meta-analysis of the relation of intolerance of uncertainty to symptoms of generalized anxiety disorder, major depressive disorder, and obsessive–compulsive disorder. Clinical Psychology Review, 31, 923-933. doi:10.1016/j.cpr.2011.05.001 69 DECISION-MAKING IN OCD Golden, C. J. (1976). Identification of brain disorders by the Stroop Color and Word Test. Journal of Clinical Psychology, 32, 654-658. Heaton, R. K., Grant, I., & Matthews, C. G. (1992). Comprehensive norms for an expanded Halstead-Reitan battery. Lutz, FL: Psychological Assessment Resources. Holdnack, H.A. (2001). Wechsler Test of Adult Reading: WTAR. San Antonio, TX: Pearson Assessment. Lawrence, N. S., Wooderson, D., Mataix-Cols, R., David, A., Speckens, M. L., & Phillips, M. L. (2006). Decision making and set shifting impairments are associated with distinct symptom dimensions in OCD. Neuropsychology, 20, 409-419. doi: 10.1037/0894- 4105.20.4.409 Leckman, J., Grice, D. E., Boardman, J., Zhang, H., Vitale, A., Bondi, C.,..Pauls, D. L. (1997). Symptoms of obsessive-compulsive disorder. American Journal of Psychiatry, 154, 911-917. Logan, G. D., Cowan, W. B., & Davis, K. A. (1984). On the ability to inhibit simple and choice reaction time responses: A model and a method. Journal of Experimental Psychology: Human Perception and Performance, 10, 276-291. doi: 10.1037/0096-1523.10.2.276 Logan, G. D., Cowan, W. B., & Davis, K. A. (1984). On the ability to inhibit simple and choice reaction time responses: A model and a method. Journal of Experimental Psychology: Human Perception and Performance, 10, 276-291. doi: 10.1037/0096-1523.10.2.276 70 DECISION-MAKING IN OCD Long, Z., Kai, W., Yi-Fu, J., Chun-Yan, Z., Ping, Y., & Hui-Juan, M. (2012). Dissociation of decision making under ambiguity and decision making under risk in patients with obsessive-compulsive disorder. Chinese Mental Health Journal, 26, 209-214. McEvoy, P. M., & Mahoney, A. E. J. (2011). Achieving certainty about the structure of intolerance of uncertainty in a treatment-seeking sample with anxiety and depression. Journal of Anxiety Disorders, 25, 112-122. doi:10.1016/j.janxdis.2010.08.010 McKay, D., Abramowitz, J. S., Calamari, J. E., Kyrios, M., Radomsky, A., Sookman, Menzies, L., Chamberlain, S. R., Laird, A. R., Thelen, S. M., Sahakian, B. J., & Bullmore, E. T. (2008). Integrating evidence from neuroimaging and neuropsychological studies of obsessive-compulsive disorder: The orbitofronto-striatal model revisited. Neuroscience and Biobehavioural Reviews, 32, 525-549. doi: 10.1016/j.neubiorev.2007.09.005 Meyers, J. E., & Meyers, K. R. (1995). Rey Complex Figure Test under four different administration procedures. The Clinical Neuropsychologist, 9, 63-67. doi:10.1080/13854049508402059 71 10.1016/j.jad.2007.02.023 Overman, W., Graham, L., Redmond, A., Eubank, R., Boettcher, L., Samplawski, O., & Walsh, K. (2006). Contemplation of moral dilemmas eliminates sex differences on the Iowa Gambling Task. Behavioral Neuroscience, 120, 817-825. doi: 10.1037/0735- 7044.120.4.817 DECISION-MAKING IN OCD Nielen, M. M. A., Veltman, D. J., de Jong, R., Mulder, G., & den Boer, J. A. (2002). Decision- making performance in obsessive compulsive disorder. Journal of Affective Disorders, 69, 257-260. doi: 10.1016/S0165-0327(00)00381-5 Nielen, M. M. A., Veltman, D. J., de Jong, R., Mulder, G., & den Boer, J. A. (2002). Decision- making performance in obsessive compulsive disorder. Journal of Affective Disorders, 69, 257-260. doi: 10.1016/S0165-0327(00)00381-5 7044.120.4.817 Patton, J.H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt Patton, J.H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51, 768-774. Impulsiveness Scale. Journal of Clinical Psychology, 51, 768-774. Impulsiveness Scale. Journal of Clinical Psychology, 51, 768-774. Ping, L., Su-Fang, L., Hai-Ying, H., Zhang-Ye, D., Jia, L., Zhi-Hua, G.,... & Zhan-Jiang, L. (2013). Abnormal spontaneous neural activity in obsessive-compulsive disorder: A resting-state functional magnetic resonance imaging study. Plos one, 8. doi: 10.1371/journal.pone.0067262 Ping, L., Su-Fang, L., Hai-Ying, H., Zhang-Ye, D., Jia, L., Zhi-Hua, G.,... & Zhan-Jiang, L. (2013). Abnormal spontaneous neural activity in obsessive-compulsive disorder: A resting-state functional magnetic resonance imaging study. Plos one, 8. doi: 10.1371/journal.pone.0067262 72 72 DECISION-MAKING IN OCD Rao, N. P., Reddy, Y. C. J., Kumar, K. J., Kandavel, T., & Chandrashekar, C. R. (2008). Are neuropsychological deficits trait markers in OCD? Progress in Neuro- Psychopharmacology and Biological Psychiatry, 32, 1574-1579. doi: 10.1016/j.pnpbp.2008.05.026 Raynard, R., Crozier, W. R., & Svenson, O. (1997). Decision-making: Cognitive models and explanations. London, UK: Routledge. Rogers, R. D., Owen, A. M., Middleton, H. C., Williams, E. J., Pickard, J. D.,... Sahakian, B. J. (1999). Choosing between small, likely rewards and large, unlikely rewards activates inferior and orbital prefrontal cortex. Journal of Neuroscience, 19, 9029–9038. Saleem, T., Gul, S., & Khalily, M. T. (2012). Quality of life of individuals having OCD with and without comorbid psychiatric disorders. Pakistan Journal of Psychology, 43. Savage, C. R., Deckersbach, T., Wilhelm, S. L., Rauch, S. L., Baer, L., Reid, T., & Jenike, M. A. (2000). Strategic processing and episodic memory impairment in obsessive-compulsive disorder. Neuropsychology, 14, 141-151. doi: 10.1037//0894-4105.14.1.141 Schneibel, R., Brakemeier, E., Wilbertz, G., Dykierek, P., Zobel, I., & Schramm, E. (2012). Sensitivity to detect change and the correlation of clinical factors with the Hamilton Depression Rating Scale and the Beck Depression Inventory in depressed inpatients. Psychiatry Research, 198, 62-67. doi: 10.1016/j.psychres.2011.11.014 Shin , N. Y., Lee, T. Y., Kim, E., & Kwon, J. S. (2014). Cognitive functioning in obsessive- compulsive disorder: A meta-analysis. Psychological Medicine, 44, 1121-1130. doi:10.1017/S0033291713001803 Rao, N. P., Reddy, Y. C. J., Kumar, K. J., Kandavel, T., & Chandrashekar, C. R. (2008). Are neuropsychological deficits trait markers in OCD? Progress in Neuro- Psychopharmacology and Biological Psychiatry, 32, 1574-1579. doi: 10.1016/j.pnpbp.2008.05.026 Raynard, R., Crozier, W. R., & Svenson, O. (1997). Decision-making: Cognitive models and explanations. London, UK: Routledge. Raynard, R., Crozier, W. R., & Svenson, O. (1997). Decision-making: Cognitive models and explanations. London, UK: Routledge. Rogers, R. D., Owen, A. M., Middleton, H. C., Williams, E. J., Pickard, J. D.,... Sahakian, B. J. (1999). Choosing between small, likely rewards and large, unlikely rewards activates inferior and orbital prefrontal cortex. Journal of Neuroscience, 19, 9029–9038. el, R., Brakemeier, E., Wilbertz, G., Dykierek, P., Zobel, I., & Schramm, E. (2012). Depression Rating Scale and the Beck Depression Inventory in depressed inpatient Psychiatry Research, 198, 62-67. doi: 10.1016/j.psychres.2011.11.014 Shin , N. Y., Lee, T. Y., Kim, E., & Kwon, J. S. (2014). Cognitive functioning in obsessive- compulsive disorder: A meta-analysis. Psychological Medicine, 44, 1121-1130. doi:10.1017/S0033291713001803 73 DECISION-MAKING IN OCD Sorensen, C. B., Kirkeby, L., & Thomsen, P. H. (2004). Quality of life with OCD: A self- reported survey among members of the Danish OCD Association. Nordic Journal of Psychiatry, 58, 231-236. doi: 10.1080/08039480410006287 Stahl, S. (1993). Alprazolam-XR: Dosage considerations. Psychiatric Annals, 27-31. Stanford, M. S., Mathias, C. W., Dougherty, D. M., Lake, S. L., Anderson, N. E., & Patton, J. H. (2009). Fifty years of the Barratt Impulsiveness Scale: An update and review. Personality and Individual Differences, 47, 385-395. doi: 10.1016/j.paid.2009.04.008 Starcke, K., Tuschen-Caffier, B., Markowitsch, H. J., & Brand, M. (2010). Dissociation of decisions in ambiguous and risky situations in obsessive-compulsive disorder. Psychiatry Research, 175, 114-120. doi:10.1016/j.psychres.2008.10.022 Starcke, K., Tuschen-Caffier, B., Markowitsch, H. J., & Brand, M. (2009). Skin conductance responses during decisions in ambiguous and risky situations in obsessive-compulsive disorder. Cognitive Neuropsychiatry, 14, 199-216. doi: 10.1080/13546800902996831 Stein, M.B., Forde, D.R., Anderson, G., & Walker, J.R. (1997). Obsessive-compulsive disorder in the community: An epidemiologic survey with clinical reappraisal. American Journal of Psychiatry, 154, 1120-1126. Steketee, G., & Frost, R. (2003). Compulsive hoarding; Current status of the research. Clinical Psychology Review, 23, 905-927. doi:10.1016/j.cpr.2003.08.002 Stern, E. R., Welsh, R. C., Gonzalez, R., Fitzgerald, K. D., Abelson, J. L., & Taylor, S. F. (2013) Subjective uncertainty and limbic hyperactivation in obsessive-compulsive disorder. S. (1993). Alprazolam-XR: Dosage considerations. Psychiatric Annals, 27-31. Steketee, G., & Frost, R. (2003). Compulsive hoarding; Current status of the research. Clinical Psychology Review, 23, 905-927. doi:10.1016/j.cpr.2003.08.002 Stern, E. R., Welsh, R. C., Gonzalez, R., Fitzgerald, K. D., Abelson, J. L., & Taylor, S. F. (2013) Subjective uncertainty and limbic hyperactivation in obsessive-compulsive disorder. Human Brain Mapping, 34, 1956-1970. doi: 10.1002/hbm.22038 Human Brain Mapping, 34, 1956-1970. doi: 10.1002/hbm.22038 Stevanovic, D. (2011). Quality of Life Enjoyment and Satisfaction Questionnaire- short form for quality of life assessments in clinical practice: A Psychometric study. Journal of Stevanovic, D. (2011). Quality of Life Enjoyment and Satisfaction Questionnaire- short form for quality of life assessments in clinical practice: A Psychometric study. Journal of 74 DECISION-MAKING IN OCD DECISION-MAKING IN OCD Psychiatric and Mental Health Nursing, 18, 744-750. doi: 10.1111/j.1365- 2850.2011.01735.x Summerfeldt, L. J., Hood, K., Antony, M. M., Richter, M. A., & Swinson, R. P. (2004). Impulsivity in obsessive-compulsive disorder: Comparisons with other anxiety disorders and within tic-related subgroups. Personality and Individual Differences, 36, 539-553. doi:10.1016/S0191-8869(03)00113-2 Thobaben, M. (2012). Obsessive-compulsive disorder (OCD): Symptoms and interventions. Home Health Care Management and Practice, 24, 211-213. doi: 10.1177/1084822312441364 Tolin, D. F., Abramowitz, J. S., Brigidi, B. D., & Foa, E. B. (2003). Intolerance of uncertainty in obsessive-compulsive disorder. Journal of Anxiety Disorders, 17, 233-242. doi: 10.1016/S0887-6185(02)00182-2 Tolin, D. F., Kiehl, K. A., Worhunsky, P., Book, G.A., & Maltby, N. (2008). An exploratory study of the neural mechanisms of decision-making in compulsive hoarding. Psychological Medicine, 39, 325-336. doi:10.1017/S0033291708003371 Torregrossa, M. M., Quinn, J. J., & Taylor, J. R. (2008). Impulsivity, compulsivity, and habit: The role of orbitofrontal cortex revisited. Society of Biological Psychiatry, 63, 253-255. doi:10.1016/j.biopsych.2007.11.014 Tukel, R., Gurvit, H., Ertekin, B. A., Oflaz, S., Ertekin, E., Baran, B.,...Atalay, F. (2012). Neuropsychological functioning in obsessive-compulsive disorder. Comprehensive Psychiatry, 53, 167-175. doi: 10.1016/j.comppsych.2011.03.007 Vandenbroucke, C.L., & Gabriëls, L. (2012). The decision-making process in obsessive- compulsive disorder. Tijdschrift voor Psychiatrie, 54, 39-49. Psychiatric and Mental Health Nursing, 18, 744-750. doi: 10.1111/j.1365- 2850.2011.01735.x merfeldt, L. J., Hood, K., Antony, M. M., Richter, M. A., & Swinson, R. P. (2004). Impulsivity in obsessive-compulsive disorder: Comparisons with other anxiety disorders and within tic-related subgroups. Personality and Individual Differences, 36, 539-553. doi:10.1016/S0191-8869(03)00113-2 Thobaben, M. (2012). Obsessive-compulsive disorder (OCD): Symptoms and interventions. Home Health Care Management and Practice, 24, 211-213. doi: Tukel, R., Gurvit, H., Ertekin, B. A., Oflaz, S., Ertekin, E., Baran, B.,...Atalay, F. (2012). Neuropsychological functioning in obsessive-compulsive disorder. Comprehensive Psychiatry, 53, 167-175. doi: 10.1016/j.comppsych.2011.03.007 Vandenbroucke, C.L., & Gabriëls, L. (2012). The decision-making process in obsessive- compulsive disorder Tijdschrift voor Psychiatrie 54 39-49 ., Gurvit, H., Ertekin, B. A., Oflaz, S., Ertekin, E., Baran, B.,...Atalay, F. (2012). Neuropsychological functioning in obsessive-compulsive disorder. Comprehensive Vandenbroucke, C.L., & Gabriëls, L. (2012). The decision-making process in obsessive- compulsive disorder. Tijdschrift voor Psychiatrie, 54, 39-49. 75 DECISION-MAKING IN OCD Watkins, L. H., Sahakian, B. J., Robertson, M. M., Veale, D.M., Rogers, R. D., Pickard, K. M.,...Robbins, T. W. (2004). Executive function in Tourette's syndrome and obsessive- compulsive disorder. Psychological Medicine, 35, 571-582. doi: H., Sahakian, B. J., Robertson, M. M., Veale, D.M., Rogers, R. D., Pickard, K. Watkins, L. H., Sahakian, B. J., Robertson, M. M., Veale, D.M., Rogers, R. D., Pickard, K. M.,...Robbins, T. W. (2004). Executive function in Tourette's syndrome and obsessive- compulsive disorder. Psychological Medicine, 35, 571-582. doi: 10.1017/S0033291704003691 Wechsler, D. (1997). Wechsler Adult Intelligence Scale, 3rd ed. (WAIS-III). San Antonio, TX: Pearson Assessment. neuroimaging in obsessive-compulsive disorder. Psychiatry Research: Neuroimaging, 132, 69-79. doi:10.1016/j.pscychresns.2004.07.001 Zermatten, A., & Van der Linden, M. (2008). Impulsivity in non-clinical persons with obsessive- compulsive symptoms. Personality and Individual Differences, 44, 1824-1830. doi:10.1016/j.paid.2008.01.025 Zermatten, A., Van der Linden, M., d'Acremont, M., Jermann, F., & Bechara, A. (2005). Impulsivity and decision-making. Journal of Nervous and Mental Disease, 193, 647-650. doi: 10.1097/01.nmd.0000180777.41295.65 76
https://openalex.org/W4395686605
https://link.springer.com/content/pdf/10.1007/978-3-031-59091-7_24.pdf
English
null
Securing Hybrid Wireless Body Area Networks (HyWBAN): Advancements in Semantic Communications and Jamming Techniques
Communications in computer and information science
2,024
cc-by
7,034
M. S¨arest¨oniemi et al. (Eds.): NCDHWS 2024, CCIS 2084, pp. 369–387, 2024. https://doi.org/10.1007/978-3-031-59091-7_24 Securing Hybrid Wireless Body Area Networks (HyWBAN): Advancements in Semantic Communications and Jamming Techniques Simone Soderi1(B) , Mariella S¨arest¨oniemi2,3 , Syifaul Fuada3 , Matti H¨am¨al¨ainen3 , Marcos Katz3 , and Jari Iinatti3 Simone Soderi1(B) , Mariella S¨arest¨oniemi2,3 , Syifaul Fuada3 Matti H¨am¨al¨ainen3 , Marcos Katz3 , and Jari Iinatti3 simone.soderi@imtlucca.it 2 Health Sciences and Technology, Faculty of Medicine, University of Oulu, Oulu, Finland mariella.sarestoniemi@oulu.fi 3 Centre for Wireless Communications, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, Finland { if l f d i h l i k j i ii i}@ l fi simone.soderi@imtlucca.it 2 Health Sciences and Technology, Faculty of Medicine, University of Oulu, Oulu, Finland mariella.sarestoniemi@oulu.fi 3 Centre for Wireless Communications, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, Finland { if l f d tti h l i k t j i ii tti}@ l fi 2 Health Sciences and Technology, Faculty of Medicine, University of Oulu, Oulu, Finland 3 Centre for Wireless Communications, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, Finland {syifaul.fuada,matti.hamalainen,marcos.katz,jari.iinatti}@oulu.fi Abstract. This paper explores novel strategies to strengthen the secu- rity of Hybrid Wireless Body Area Networks (HyWBANs), which are essential in smart healthcare and Internet of Things (IoT) applica- tions. Recognizing the vulnerability of HyWBAN to sophisticated cyber- attacks, we propose an innovative combination of semantic communi- cations and jamming receivers. This dual-layered security mechanism protects against unauthorized access and data breaches, particularly in scenarios involving in-body to on-body communication channels. We conduct comprehensive laboratory measurements to understand hybrid (radio and optical) communication propagation through biological tis- sues. We utilize these insights to refine a dataset for training a Deep Learning (DL) model. These models, in turn, generate semantic con- cepts linked to cryptographic keys for enhanced data confidentiality and integrity using a jamming receiver. The proposed model signif- icantly reduces energy consumption compared to traditional crypto- graphic methods, like Elliptic Curve Diffie-Hellman (ECDH), especially when supplemented with jamming. Our approach addresses the primary security concerns and sets the baseline for future secure biomedical com- munication systems advancements. Keywords: Heterogeneous · WBAN · energy · security · optical · RF · near-infrared communications c ⃝The Author(s) 2024 1 Introduction The advent of wireless and mobile communications technologies has been essen- tial in enhancing healthcare, marking a paradigm shift towards more proactive c ⃝The Author(s) 2024 S. Soderi et al. 370 and personalized medical interventions. The concept of smart healthcare is at the forefront of this transformation, offering many opportunities to address the growing needs of an ageing population and the increasing prevalence of chronic diseases [1]. Remote health monitoring, a cornerstone of modern healthcare, has emerged as a cost-effective and efficient approach to disease prevention and healthcare provision, especially with the integration of 5G and 6G technologies. These advancements are pivotal in supporting in-body communications with implanted medical devices, enabling real-time health provisioning, virtual con- sultations, better diagnostics, and telesurgeries, among other benefits [1]. His- torically, information transmission through biological tissues has predominantly relied on radio and acoustic waves [2]. However, these conventional methods are fraught with challenges, including security, safety, privacy, and interference, necessitating the exploration of alternative communications media. The vulnera- bilities of implantable or in-body devices to hacks and unauthorized access have underscored the urgent need for enhanced security measures [3,4]. Optical Wireless Communications (OWC) has emerged as a promising alter- native, utilizing light, especially in the near-infrared range, to transmit informa- tion through biological tissues. This method offers many advantages, including high security, privacy, safety and low complexity, as well as low power con- sumption. It has been used to successfully establish connectivity to electronic devices embedded under the skin [5]. Going further, a hybrid solution which is merging both radio-based and optical-based technologies in Wireless Body Area Network(WBAN) context can open a new, more secure way to implement per- sonalized healthcare services and transfer personal health data. This is also a way to reduce radio signal emission towards the human body. Hybrid Wireless Body Area Networks (HyWBANs) stand at the forefront of innovation in healthcare and Internet of Things (IoT) applications, merg- ing radio and OWC. These networks offer remarkable advantages such as data throughput enhancement, enhanced security, and improved reliability, making them ideal for critical healthcare applications and various services, from patient monitoring to advanced diagnostics. Reconfigurable HyWBANs takes adaptabil- ity to the next level with a dynamic architecture, ensuring consistent perfor- mance in diverse environments. Preliminary studies on HyWBANs underscore their potential, showcasing notable performance and energy efficiency improve- ments [6]. 1 Introduction Energy harvesting is crucial to HyWBANs [7], focusing on developing energy-autonomous nodes that enhance sustainability and reduce maintenance. The networks’ advanced sensing capabilities support single and dual-mode sens- ing, enabling comprehensive data collection for diverse applications. Moreover, HyWBANs’ design promotes sustainable operations, which is essential in today’s environmentally conscious landscape. Optimized data transmission functionality in HyWBANs caters to the high demands of medical and IoT applications. A significant feature is the ability to transfer energy to in-body devices, ensuring continuous operation. Additionally, HyWBANs’ advanced sensing capabilities are essential in medical diagnostics, allowing for detailed tissue analysis and health monitoring, thus revolutionizing healthcare and IoT applications [1]. Securing Hybrid Wireless Body Area Networks (HyWBAN) 371 Over the years, academia has shown interest in Physical Layer Security (PLS) solutions that aim to protect communications by exploiting the properties of the communication media [8–11]. These techniques consist of processing the signal sent over a channel in such a way as to obtain certain security properties without resorting to specific primitives, typically cryptography, offered by layers above the physical level. In this paper, we show how to combine PLS techniques with Deep Learning (DL) algorithms to improve the security of HyWBANs. Data Optical Radio channel state information protocol information HYBRID TRANSMITTER Optical Radio HYBRID RECEIVER Data CODING WIRELESS CHANNEL DECODING channel state information protocol information Fig. 1. Coding strategy for hybrid networks. Fig. 1. Coding strategy for hybrid networks. Motivation. In this article, we have decided to use a model already used in the literature that involves defining operating modes based on the combina- tions of the two communications channels [6,12]. Figure 1 depicts how hybrid radio-optical wireless networks utilize Shannon’s theory, which defines the max- imum channel capacity for communications. It shows the dynamic selection of the device’s operating modes based on factors like channel state information (radio/optical) and user context. In our view, HyWBANs can improve secu- rity by integrating in the radio transmissions the OWC, which is known for the localised and secure transmission of signals. Encoding signals across radio and optical channels maximises secrecy, in line with recent theoretical work on conventional networks. This approach exploits the inherent security features of optical communications, addressing the vulnerabilities of WBANs. This paper uses data measured in the laboratory to implement an innovative hybrid network security scheme using semantic communications and intentional interference. Contribution. 2 Background This section introduces the radio and optical technologies used for wireless sensor communication. The aim is to provide some notions before discussing how hybrid networks combine these two technologies. S. Soderi et al. The remainder of the paper is organized as follows. Section 2 briefly recalls the concepts useful for understanding the paper. Section 3 discusses the major security threats in this scenario. Then, Sect. 4 presents the proposed scheme to enhance the security of hybrid networks. Section 5 presents the results achieved in terms of performance and energy cost. Finally, Sect. 6 concludes the paper by discussing our findings. 1 Introduction In particular, hybrid communications have been of great inter- est for sensor networks. In a digital healthcare scenario, protecting these com- munications and doing so effectively while consuming as little energy as possi- ble makes significance. The contributions of this article can be summarised as follows. (i) We present a novel concept that exploits the combination of Com- munications (SC) with a jamming receiver to improve the confidentiality and integrity of these wireless communications. (ii) We performed measurements in the laboratory to study the propagation of hybrid (radio and optical) commu- nications in biological tissue. These measurements allowed us to define part of the dataset used for the DL model. Finally, (iii) we evaluated and performed a security analysis of the HyWBANs. 372 S. Soderi et al. 2.1 Radio-Based WBAN Technologies WBAN is a way to link various wearable sensor nodes wirelessly into one indi- vidual and personalized network used to monitor a person’s psycho-physiological vital signs. Depending on the need, the vital sensors can be distributed all around the human body. Low-power consumption, small size, and lightweight are the requirements set for the nodes to enable user acceptance. In principle, the amount of connected sensors within one WBAN can be high, but a realistic number is less than five for the sake of usability. The basic idea behind a WBAN is that dedicated sensors are collecting vital information and transmitting it wirelessly to the central node (called a hub), which then pre-process the data or conveys it further. Figure 2 shows the variability of the vital sensor nodes, which can be used in the WBAN context (the list of sensors is not exhaustive) [13]. In addition to sensors which are attached to the skin, so called on-body sensors, WBAN can utilize smart implants, such as pacemakers, or other in-body sen- sors/devices, such as Wireless Capsule Endoscope (WCE). In WBAN, all the nodes are connected to the on-body hub to enable real-time information trans- mission towards backbone infrastructure. Typically, WBAN is using a one-hop star network topology. Currently, de facto wireless standard in WBAN is Bluetooth Low Energy (BLE) but there are also other dedicated WBAN standards available, such as ETSI SmartBAN [14], IEEE 802.15.6 [15], or IEEE 802.15.4 [16]. The latter one is better known via its higher layer protocols ZigBee and 6LoWPAN. From a radio technology point-of-view, WBAN connectivity can be based on narrowband (NB) signals, which are used, e.g., in BLE and SmartBAN, or ultra wideband (UWB), which is adopted by [15]. The most common frequency band at the moment for NB signal is Industrial, Scientific and Medical (ISM) band at about 2.4 GHz. On the other hand, e.g., [15] defines several NB frequency bands for WBAN use also occupying sub-GHz frequencies. As operating in a highly populated frequency range, ISM band is typically subject to high inter- ference originated from other radio equipment nearby. The selected frequency band, as well as signal bandwidth, also have an impact on the observed signal Securing Hybrid Wireless Body Area Networks (HyWBAN) 373 propagation properties through/along tissues, positioning accuracy, throughput, etc., depending on the application and use-case. 2.2 Optical Communications in Wireless Sensor Networks The utilization of optical communications, including Visible Light Communica- tion (VLC) and Infrared (IR) technologies, in wireless sensor networks is gaining interest for various IoT and body network applications. Optical communication offers security, bandwidth, and energy efficiency advantages, which are crucial for IoT deployments. VLC utilizes Light Emitting Diodes (LED) to transmit data using the visible spectrum. This approach is inherently secure due to the limited light propagation and offers high data rates, making it suitable for indoor IoT applications [19]. VLC’s potential in hybrid optical-wireless networks for next-generation commu- nications, especially in 5G and beyond, is highlighted in [20]. IR communica- tion leverages the non-visible spectrum for data transmission, offering benefits in terms of device miniaturization and reduced interference with existing RF systems. Its suitability for low-data-rate IoT applications, especially in hybrid networks combining IR and VLC, is explored in [7]. IR in hybrid radio-optical wireless networks offers innovative solutions for versatile IoT applications, as discussed in [21]. Integrating optical and wireless technologies in a hybrid frame- work opens new avenues for enhancing IoT network performance. The synergy of RF and optical communication technologies in hybrid networks is investigated in [22], which outlines the implementation and advantages of such an approach. 2.1 Radio-Based WBAN Technologies If high resolution, real-time data is needed, then UWB can be the best option from radio-based technolo- gies. Lower performance requirements and deeper in-body penetration, however, are favouring NB technology. Reference [15] also defines Human Body Commu- nications(HBC) technology operating around 21 MHz, but this is omitted in this review due to its deviation from the conventional Radio Frequency (RF)-based communications as being a coupling-based solution. g p g The original network topology in WBAN is based on a star topology, where all the data flows are going through the central node, the hub. In this case, the hub is also a bridge from the body domain to the backbone network. The recent development, e.g., in ETSI SmartBAN has introduced and defined a hub-to-hub communications to transfer information between adjacent WBAN networks [17]. In addition, a two-hop relay functionality is included in the SmartBAN techni- cal specification [18]. From the security viewpoint, all the hops between WBAN nodes, although being short, should be reliable but also secure as the communi- cations chain is as reliable as its weakest link. This highlights the importance of light security protocols to be used also in the WBAN context. Fig. 2. Variability of the possible sensors that can be used in the WBAN context. Fig. 2. Variability of the possible sensors that can be used in the WBAN context. S. Soderi et al. 374 3 Security Analysis of HyWBANs Developing next-generation networks to support better biomedical applications presents an opportunity. However, cyber-security risks arise mainly from this technology’s highly interconnected and ubiquitous nature [1]. Therefore, the cybersecurity analysis of these hybrid communications begins with choosing a system model that best represents the problem. WBANs are components of cyberspace that assist people in their daily activ- ities and collect data from persons. WBANs and, more broadly, wearable wire- less networks (WWNs) have three communication layers, according to the tier model [23]. As shown in Fig. 3, wearable sensors capture data in Tier 1 and transmit it to Tier 2 for aggregation and data processing. Finally, data is sent to Tier 3 and made available for remote access. The HyWBANs follow the same system model, where radio link, optical link, or both can be used for each type of communication in Tier 1 and Tier 2. As illustrated in Fig. 3, we have different types of communications in HyWBAN: on-body to in-body devices (labelled as On-In) and on-body to on-body (labelled as On-On) devices that operate in Tier 1. Instead, at Tier 2, all communications are off-body, including on-body to off-body devices (labelled On-Off). We assume that HyWBANs operate up to Tier 2, as depicted in Fig. 3. One of the main security problems of this communication chain is that Eve, the adversary (attacker) shown in Fig. 3, can carry out several attacks. We can assume that she has complete control to intercept and modify all messages Securing Hybrid Wireless Body Area Networks (HyWBAN) 375 Fig. 3. Communications tiers system model, in which hybrid communications operate in the first two tiers. Fig. 3. Communications tiers system model, in which hybrid communications operate in the first two tiers. exchanged between HyWBAN nodes [24]. In the rest of the paper, we analyse the possible attacks and their mitigation. exchanged between HyWBAN nodes [24]. In the rest of the paper, we analyse the possible attacks and their mitigation. 3.1 Security Threats Overview The complexity of HyWBANs, which combines RF and OWC, far exceeds tra- ditional communications systems due to their dual-channel nature. This sophis- tication poses significant challenges for attackers attempting to compromise the network, as they must navigate radio and optical channels. In the HyWBANs domain, the security of communication channels is essen- tial, especially when considering transmitting sensitive health data. This section delves into the nuanced vulnerabilities inherent to RF and OWC, laying the groundwork for understanding the superior security posture of optical commu- nications in specific scenarios. RF communications, by their very nature, are susceptible to eavesdropping due to their omnidirectional signal propagation. This characteristic allows malicious entities to intercept signals without necessi- tating a direct line of sight, thereby posing a significant risk to the confidentiality of transmitted data. Conversely, optical communications demand a line-of-sight for effective transmission, inherently restricting the potential for unauthorized interception. Despite this advantage, optical channels are not impervious to secu- S. Soderi et al. 376 Fig. 4. Security threats to HyWBAN. Fig. 4. Security threats to HyWBAN. rity threats. A breach in the line-of-sight or sophisticated techniques to capture reflected optical signals can compromise data integrity and confidentiality. In the context of HyWBANs, an attacker would primarily focus on tactics that enable eavesdropping on biomedical device communications. These tactics could include exploiting network security protocol vulnerabilities, conducting Man-in-the-Middle (MitM) attacks to intercept data, or using sophisticated tech- niques to bypass encryption. Reconnaissance plays a crucial role, as the attacker must gather detailed information about the network’s configuration and security mechanisms to successfully deploy malware or other attack vectors. Figure 4 presents a multifaceted evaluation of security threats in HyWBANs. Each threat is analyzed based on three critical dimensions: relevance to the net- work’s security, ease of implementation by potential attackers, and the potential impact on network integrity and functionality. This brief assessment enables a slight understanding of each threat’s effectiveness and helps prioritize security measures. To fortify the security framework of HyWBANs against these vulnerabilities, we introduce a semantic communication method that significantly enhances the security of transmitted data. Securing Hybrid Wireless Body Area Networks (HyWBAN) 377 4 Enhancing the Security of HyWBAN Through Semantic Communications Understanding the implications for security in the dynamic landscape of HyW- BANs is essential. Adversaries exploiting vulnerabilities in these networks could potentially gain unauthorized access to the human body, leading to critical threats like hijacking pacemakers, reconfiguring smart pill dispensers, or even creating novel types of diseases. The dual nature of these networks, encompass- ing radio and optical wireless channels, adds a layer of complexity to potential attacks. This study proposes a novel security mechanism combining the princi- ples of semantic communications with the strategic deployment of a jamming receiver (see Fig. 5), enhancing the confidentiality and integrity of HyWBANs. Semantic communications [25], an emerging paradigm in network secu- rity [26], involves generating semantic concepts related to biomedical applications or patient health status. This approach utilizes a DL model trained on a dataset comprising measured, augmented, and synthetic biological signals [1,27]. Dur- ing an enrollment phase, assumed free from adversarial presence, each semantic concept is associated with a secret, such as a cryptographic key, stored in the nodes’ memory. The transmission of semantic concepts over the wireless channel, although susceptible to interception by malicious adversaries, is protected through a jam- ming receiver. As shown in Fig. 5, this receiver introduces intentional interference on either the light or radio channel, or both, effectively interfering with the trans- mitted data from the in-body device. Consequently, an adversary attempting to decode the data encounters altered signal characteristics, such as decreased Signal-to-Noise Ratio (SNR) for radio and input power for Near-Infrared (NIR) signals; this leads to an erroneous classification of semantic concepts. However, the legitimate receiver, Bob, knows the jamming pattern and can reverse the artificially induced bias to correctly decode the transmitted semantic Data Optical Radio HYBRID TRANSMITTER (ALICE) CODING SEMANTIC- CODING Message Semantic Information SEMANTIC COMMUNICATIONS CLASSICAL COMMUNICATIONS channel features bio-signals features channel state information protocol information Data Optical Radio HYBRID TRANSMITTER (ALICE) Radio HYBRID RECEIVER (BOB) Data CODING DECODING SEMANTIC- CODING Message Semantic Information SEMANTIC COMMUNICATIONS WIRELESS CHANNEL SEMANTIC- DECODING Message Semantic Information CLASSICAL COMMUNICATIONS channel features bio-signals features channel features bio-signals features channel state information protocol information channel state information protocol information Optical EVE Fig. 5. AI-based data stream encoding in HyWBANs using semantic communications and a jamming receiver to harden the security of wireless communications. 4 Enhancing the Security of HyWBAN Through Semantic Communications Radio HYBRID RECEIVER (BOB) Data DECODING n WIRELESS CHANNEL SEMANTIC- DECODING Message Semantic Information channel features bio-signals features channel state information protocol information Optical EVE Fig. 5. AI-based data stream encoding in HyWBANs using semantic communications and a jamming receiver to harden the security of wireless communications. S. Soderi et al. 378 concept. In contrast, an adversary, referred to as Eve, who lacks this knowl- edge, faces significant challenges in decoding the data accurately [9,11]. This approach, leveraging semantic communication and controlled jamming, offers a dual-layered defence mechanism, enhancing the resilience of HyWBANs against sophisticated cyber threats. The rest of the section describes how the data were prepared and how we propose to use a DL algorithm on devices with constrained resources. 4.1 Radio and Optical Channels Data Measurement In this study, we investigate the efficacy of hybrid communications that uti- lize optical and radio channels, explicitly investigating their capacity to pene- trate biological tissues. Two distinct experimental setups were designed to assess the performance characteristics necessary for effective and secure communica- tions through such mediums. For optical communications, NIR frequencies were employed, selected for their proven proficiency in penetrating biological tissues. On the other hand, UWB technology, recognized for its superior transmission capabilities, particularly noise-like signals, was chosen for radio communications. This dual-faceted approach allows for a comprehensive evaluation of the poten- tial of hybrid communication systems in medical applications. NEAR-IR LED (TX) PHOTODETECTOR (RX) LED'S CURRENT DRIVER d OSCILLOSCOPE Vout I USRP RADIO External Modulation LAPTOP POWER SUPPLY BIAS TEE UDP data CHANNEL: BIOLOGICAL TISSUE (SAMPLE XX) Fig. 6. Measurement set-up with NIR communications (as OWC) through the biolog- ical tissue. NEAR-IR LED (TX) I UDP data LED'S CURRENT DRIVER Fig. 6. Measurement set-up with NIR communications (as OWC) through the biolog- ical tissue. The experimental setup, depicted in Fig. 6, is an optical communication part; it comprises various components that can be divided into two subsys- tems: transmitter and receiver front end. The transmitter unit includes a NIR LED (M810L3, THORLABS) with 810 nm wavelength, a bias-tee, and a LED driver. The receiver unit utilizes a Photo-Detector (PD) (PDA 36A-EC switch- able gain detector, THORLABS). A sample of biological tissue was used, acting as the communications channel. The LED is driven by a current driver module (DC2200, THORLABS), which is controlled by an external modulation source. Securing Hybrid Wireless Body Area Networks (HyWBAN) Securing Hybrid Wireless Body Area Networks (HyWBAN) 379 The modulation of the NIR LED is essential for transmitting data through the biological tissue. The PD, positioned at a specific distance (d is the thickness of the meat sample used in the measurements) from the NIR LED, captures the transmitted light after it has passed through the tissue. The output from the PD (Vout) is then analyzed using an oscilloscope (with 50 Ω impedance) to assess the effectiveness of data transmission through the tissue sample. Using a laptop, we sent the same ASCII character with the User Datagram Protocol (UDP) protocol to a software-defined radio USRP that modulated the signal before sending it to the LED driver. Two NI USRPs (2920 model) were employed in this study. We also used a bias-tee (ZFBT-4R2GW-FT+, Mini-Circuits) to combine the mod- ulation signal and the bias current to feed the driver. We measured the peak of the received burst signal for each character sent from the laptop. The Vout is then converted into an input power unit by using the equation provided in the datasheet of PDA36A-EC. This setup is crucial for evaluating the feasibility of NIR communications in scenarios where signals need to penetrate biological tissues, such as in implantable medical device applications. CHANNEL: BIOLOGICAL TISSUE (SAMPLE XX) d TIMEDOMAIN UWB PULSON P410 RCM LAPTOP Serial COM 3 Serial COM 4 TIMEDOMAIN UWB PULSON P410 RCM TRANSMITTER ID #100 RECEIVER ID #106 RADIO ABSORBER CHAMBER IN-BODY ANTENNA or ON-BODY ANTENNA 1 ON-BODY ANTENNA 2 Fig. 7. Measurement set-up with UWB radio through the biological tissue. Fig. 7. Measurement set-up with UWB radio through the biological tissue. As illustrated in Fig. 7, the radio measurement setup was meticulously designed to evaluate the performance of radio communications in HyWBANs. It consists of a UWB transmitter (P410 PulsON, Time Domain) inside the body (i.e., in-body device) that communicates with a UWB receiver (P410 PulsON, Time Domain) that has its antenna positioned on the porcine skin (i.e., on-body). S. Soderi et al. 380 Using Time Domain’s Channel Analysis Tool (CAT) software, we simulated the communications scenario inside the body by sending signals from the transmitter to the receiver. We enclosed the antennas inside a box of RF absorber material to avoid external interference. The received signals were saved on a laptop using CAT software and analyzed later using MATLAB. Securing Hybrid Wireless Body Area Networks (HyWBAN) This system makes it pos- sible to accurately measure the radio signal’s capability to penetrate biological tissue and the effectiveness of UWB technology in an in-body communications scenario, essential for developing reliable hybrid WBANs. Fig. 8. NIR received power varying the temperature of two biological tissue samples (i.e., sample #6 and #7) and the gain of the PD . Fig. 8. NIR received power varying the temperature of two biological tissue samples (i.e., sample #6 and #7) and the gain of the PD . Figure 8 shows the power (expressed in mW) of the NIR signal received after passing through the two biological tissue samples with a maximum thickness of 37 mm and 39 mm for samples #6 and #7, respectively. From Fig. 8, it is evident how the propagation capabilities improve when the temperature reaches 37◦C, which is considered almost typical for a human body. Selecting a higher gain (i.e., from 0 dB to 10 dB) in the receiver (PDA 36A-EC offers this option using a rotary switch) does not lead to a significant advantage regarding receiver sensitivity. Figure 9 shows the UWB SNR measured by the on-body antenna placed on the skin. Meanwhile, the transmitter and receiver were aligned for light. For the UWB measurements, we left the in-body antenna at the fixed position while we moved the on-body antenna in 1 cm steps to investigate the communication limit. Securing Hybrid Wireless Body Area Networks (HyWBAN) 381 Fig. 9. SNR measurements of UWB transmissions through biological tissue. Fig. 9. SNR measurements of UWB transmissions through biological tissue. Fig. 9. SNR measurements of UWB transmissions through biological tissue. 4.2 Dataset: Measured and Synthetic Data for Medical Applications Using HyWBAN Developing and optimizing semantic communications and strengthening security within hybrid networks necessitate a comprehensive dataset for training and testing DL models. This dataset encompasses both measured features, such as SNR and the power received by the PD in NIR communications, as well as synthetic features. These synthetic features are conceptualized on the premise that the constituent devices of HyWBAN can acquire and process biological signals from individuals. This dual approach in dataset formulation facilitates a realistic assessment of the HyWBAN’s operational capabilities and aids in simulating a wide range of scenarios for advanced medical applications. To refine the dataset for DL models in the context of HyWBANs for medical applications, we employed a dual-strategy approach involving both the augmen- tation of measured data and the generation of synthetic data. The augmentation process for the measured attributes, specifically SNR for UWB and received power for NIR, employs a statistical methodology in which new values are gen- erated based on a Gaussian distribution. This distribution is centred on the measured mean and standard deviation. This statistical rigour ensures the aug- mented data are closely aligned with realistic measurement variations. We gener- ated values for a few parameters: acceleration, heart rate and body temperature to generate synthetic data to emulate the ability of HyWBAN devices to measure S. Soderi et al. 382 biological signals. We assumed that a hybrid device could access these quantities (as a knowledge base for semantic communications) in a medical application or at least a part of it. These synthetic values are derived from a Gaussian distri- bution, adhering to predefined mean and standard deviations to ensure they fall within physiologically plausible ranges. The Table 1 summarises the specification of statistics for the data generation process. Our semantic communication model uses a binary classification approach, simplifying complex data into categories like ’HIGH SNR’ or ’LOW SNR’ using the thresholds defined in Table 1. This method efficiently filters out noise, focusing on key data aspects and significantly reducing computational load. This binary representation accelerates model train- ing and enhances interpretability, facilitating fast and decisive communications analysis. We can then define the labels to be associated with each type of com- munication in a supervised manner (see Table 2). These labels are the semantic concepts that represent data the device measures in a compressed manner. This particular approach to dataset preparation supports the robustness of the devel- oped models. 5 Proposed Model Evaluation Our study developed a deep learning model for semantic communication in HyW- BAN: an autoencoder with a 64 −32 −64 neuron structure and a classification model with dense layers and dropout regularization. The purpose of the autoen- coder model within our semantic analysis is to reduce the dimensionality of the input data, including SNR and heart rate, thereby enabling more efficient pro- cessing and transmission. The autoencoder helps identify the most significant features crucial for semantic analysis by transforming the data into a lower- dimensional space. This process not only aids in preserving essential information but also contributes to the system’s security by minimizing the amount of data exposed to potential threats. We have conducted a series of experiments to determine the optimal architec- ture for our model. Our choice was guided by a grid search approach, where we evaluated various configurations and selected the one that minimized the recon- struction error on a validation set: the 64-32-64 structure balanced model com- plexity and the ability to capture the underlying patterns in the data. We also experimented with different activation functions and learning rates, ultimately choosing a Rectified Linear Unit (ReLU) activation for its efficiency and a learn- ing rate of 0.001 for stable convergence. The training process of our autoencoder was carried out over 50 epochs, with early stopping implemented to prevent overfitting. We used a batch size of 256, which was determined to be optimal through experimentation, balancing the trade-offbetween training speed and memory constraints. The dataset (after an augmentation of the data by a factor of 50) comprised 2040 samples, split into 80% for training and 20% for valida- tion. This information aims to enhance the transparency and reproducibility of our model evaluation. To visualize the effectiveness of the autoencoder in capturing semantic rela- tionships, we apply the t-Distributed Stochastic Neighbor Embedding (t-SNE) technique (see Fig. 10). This method is noted for its ability to represent high- dimensional data in lower dimensions while preserving data structures, allowing us to visually inspect the clustering of data points based on their semantic sim- ilarities. We have expanded our discussion on the interpretability of clusters formed in the t-SNE visualization. The clusters represent distinct data patterns that the autoencoder has learned to encode. By examining the characteristics of samples within each cluster, we can infer the model’s ability to discern different features in the data, which supports its effectiveness. 4.2 Dataset: Measured and Synthetic Data for Medical Applications Using HyWBAN It ensures the simulation of diverse scenarios, which is critical for applying HyWBAN in medical settings. Table 1. Statistical summary of augmented and synthetic features. Feature Mean Stand. Deviation [Min, Max] Threshold SNR (dB)a 23.6 4.23 [17.57, 33.32] 19 dB Input Power (mW)b 0.07 0.03 [0.02, 0.09] 0.05 mW Acceleration (m/s2) 0 0.1 [-0.5, 0.5] 0.1 m/s2 Heart Rate (bpm) 60 25 [50, 120] <60 bpm, >110 bpm Body Temperature (◦C) 36 2 [34, 42] 37◦C a UWB measured data. b NIR measured data. Table 1. Statistical summary of augmented and synthetic features. Table 2. Classification labels for semantic communications Label Condition Full Communications HIGH SNR and HIGH LPW Wide Communications HIGH SNR and LOW LPW Communications in Motion (HIGH SNR or HIGH LPW) and HIGH ACC Critical Communications (HIGH HR or HIGH TMP) and LOW LPW Unstable Communications LOW SNR or LOW LPW Reduced Communications Other scenarios not covered by above conditions Table 2. Classification labels for semantic communications Securing Hybrid Wireless Body Area Networks (HyWBAN) 383 5 Proposed Model Evaluation The model is optimized using Adam optimizer and trained to categorize the data into predefined semantic classes, as described in our data preprocessing phase. Performance evaluation using a confusion matrix and accuracy metrics confirmed the model’s efficacy. Finally, the model was converted to TensorFlow Lite [28] (i.e., a Tiny Machine Learning framework that supports the conversion of ML models into a format that can be run on microcontrollers), aligning with low-power, edge-based IoT device requirements, ensuring privacy, energy effi- ciency, and real-time processing. This approach signifies a substantial advance- ment in semantic communication for smart healthcare applications. S. Soderi et al. 384 Fig. 10. Low-dimensional representation of data preserving semantic similarities. For example, the figure shows the effect of NIR jamming on the semantic concepts. Fig. 10. Low-dimensional representation of data preserving semantic similarities. For example, the figure shows the effect of NIR jamming on the semantic concepts. In our evaluation, we compared the energy efficiency of our semantic commu- nication with a jamming solution to the Elliptic Curve Diffie-Hellman (ECDH) key exchange [29]. The comparison focused on various configurations, assessing the energy consumption for different key lengths in ECDH (160 and 256 bits) against our semantic communication model that can use 8 or 16 bits to repre- sent the semantic concepts, and it is enhanced with jamming up to 8 and 16 bits (i.e., worst case for our proposal). We assumed 0.1 µJ as the energy per bit and 0.2 µJ as the energy cost to jam a bit. This analysis, crucial for understanding the practicality of deploying these methods in energy-constrained environments like HyWBANs, is visualized Fig. 11, illustrating the total energy consumption of each method. Such comparisons highlight the efficiency of semantic communi- Fig. 11. Energy consumption comparison between SC with jamming and ECDH. Fig. 11. Energy consumption comparison between SC with jamming and ECDH. Securing Hybrid Wireless Body Area Networks (HyWBAN) 385 cations, especially when supplemented with jamming, in contrast to traditional cryptographic approaches like ECDH. The classification performance of our TensorFlow model for semantic com- munication in HyWBANs demonstrated good precision and recall across most classes, with an overall accuracy of 94%. However, the corresponding Tensor- Flow Lite model, optimized for low-power devices, showed a variation in perfor- mance, particularly in precision and recall for specific communication classes like Communication in Motion and Critical Communication, resulting in an overall accuracy of 86%. 5 Proposed Model Evaluation This variation underscores the challenges in collecting more data with measurements and balancing model complexity with the constraints of edge computing devices. References 1. Batista, E., Lopez-Aguilar, P., Solanas, A.: Smart health in the 6G Era: bringing security to future smart health services. IEEE Commun. Mag. 1–7 (2023) 2. Ahmed, I., Halder, S., Bykov, A., Popov, A., Meglinski, I.V., Katz, M.: In-body communications exploiting light: A proof-of-concept study using ex vivo tissue samples. IEEE Access 8, 190378–190389 (2020) ( ) 3. Zafar, S., et al.: A systematic review of bio-cyber interface technologies and security issues for internet of bio-nano things. IEEE Access 9, 93529–93566 (2021) 4. Ransford, B., et al.: Cybersecurity and medical devices: a practical guide for cardiac electrophysiologists. Pacing Clin. Electrophysiol. 40(8), 913–917 (2017) 5. Halder, S., S¨arest¨oniemi, M., Ahmed, I., Katz, M.: Providing connectivity to implanted electronics devices: experimental results on optical communications over biological tissues with comparisons against UWB. In: Alam, M.M., H¨am¨al¨ainen, M., Mucchi, L., Niazi, I.K., Le Moullec, Y. (eds.) BODYNETS 2020. LNICST, vol. 330, pp. 3–17. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-64991- 3 1 6. Saud, M.S., Ahmed, I., Kumpuniemi, T., Katz, M.: Reconfigurable optical-radio wireless networks: meeting the most stringent requirements of future communica- tion systems. Trans. Emerg. Telecommun. Technol. 30(2), e3562 (2019). https:// doi.org/10.1002/ett.3562 7. Kamalakis, T., Ghassemlooy, Z., Zvanovec, S., Nero Alves, L.: Analysis and simula- tion of a hybrid visible-light/infrared optical wireless network for IoT applications. J. Opt. Commun. Netw. 14(3), 69–78 (2022) ( ) ( ) 8. Bloch, M., Barros, J.: Physical-Layer Security: from Information Theory to Secu- rity Engineering. Cambridge University Press, Cambridge (2011) 9. Soderi, S., Mucchi, L., H¨am¨al¨ainen, M., Piva, A., Iinatti, J.: Physical layer secu- rity based on spread-spectrum watermarking and jamming receiver. Trans. Emerg. Telecommun. Technol. 28(7), e3142 (2017) ( ) ( ) 10. Soderi, S., Dainelli, G., Iinatti, J., H¨am¨al¨ainen, M.: Signal fingerprinting in cogni- tive wireless networks. In: 2014 9th International Conference on Cognitive Radio Oriented Wireless Networks and Communications (CROWNCOM), pp. 266–270 (2014) 11. Soderi, S., De Nicola, R.: 6G networks physical layer security using RGB visible light communications. IEEE Access 10, 5482–5496 (2022) 12. Katz, M., Ahmed, I.: Opportunities and challenges for visible light communications in 6G. In: 2020 2nd 6G Wireless Summit (6G SUMMIT), pp. 1–5 (2020) 13. H¨am¨al¨ainen, M., et al.: Recent Progress in ETSI TC SmartBAN Standardization. In: 2023 IEEE 17th International Symposium on Medical Information and Com- munication Technology (ISMICT), pp. 1–6 (2023) ( ) ( ) 14. 6 Conclusions The research presented in this paper marks a significant stride in enhancing the security of HyWBANs. By integrating semantic communications with jamming receivers, we demonstrate a robust method to protect sensitive health data and biomedical devices within the HyWBANs framework. Our experimental analysis provides valuable insights into the propagation characteristics of hybrid commu- nications in biological tissues, forming the basis for an advanced DL model. This model’s ability to generate and interpret semantic concepts, coupled with a strategic jamming mechanism, ensures the reliable transmission of encrypted data, thereby mitigating potential cybersecurity threats. Notably, our approach outperforms traditional cryptographic methods in energy efficiency, making it a viable solution for the energy-sensitive environment of HyWBANs. The semantic strategy enhances security by transmitting only the necessary and relevant data, reducing the attack surface. The deep learning model con- tributes to this by learning to identify and filter out non-essential information, thus streamlining the communication process and making it more secure. The inherent security advantages of optical communications, such as the line-of-sight requirement for the interception, are exploited in our hybrid system to strengthen overall security further. The findings and methodologies outlined in this study improve the security of current HyWBAN systems and pave the way for their broader adoption in smart healthcare services, aligning with the evolving landscape of 6G technology. Acknowledgments. This research is partially funded by the Research Council of Fin- land Profi6 funded 6G Enabling Sustainable Society (University of Oulu) and Research Council of Finland 6G Flagship (grant 318927). and the European Union’s Horizon 2020 programme under the Marie Sklodowska-Curie grant H2020-MSCA-RISE-2019 agreement No. 872752. S. Soderi et al. 386 References H¨am¨al¨ainen, M., et al.: ETSI SmartBAN architecture: the global vision for smart body area networks. IEEE Access 8, 150611–150625 (2020) 15. IEEE Computer Society: IEEE standard for local and metropolitan area networks - part 15.6: Wireless body area networks. IEEE standard (2012) 16. IEEE Computer Society: IEEE standard for low-rate wireless networks. IEEE stan- dard (2015) 17. ETSI Tc SmartBAN: Smart body area network (SmartBAN) Hub to Hub commu- nication for SmartBAN medium access control (MAC). ETSI standard TS 103, 806 (2023) Securing Hybrid Wireless Body Area Networks (HyWBAN) 387 18. ETSI Tc SmartBAN: Smart body area network (SmartBAN) Relay functionality for SmartBAN medium access control (MAC). ETSI standard TS 103, 805 (2024) 19. Perera, A., Katz, M., Godaliyadda, R., H¨akkinen, J., Str¨ommer, E.: Light-based internet of things: implementation of an optically connected energy-autonomous node. In: 2021 IEEE Wireless Communications and Networking Conference (WCNC), pp. 1–7 (2021) 20. Chowdhury, M.Z., Hasan, M.K., Shahjalal, M., Hossan, M.T., Min Jang, Y.: Opti- cal wireless hybrid networks for 5g and beyond communications. In: 2018 Inter- national Conference on Information and Communication Technology Convergence (ICTC), pp. 709–712 (2018) ( ) ( ) 21. Teli, S.R., et al.: Hybrid optical wireless communication for versatile IoT applica- tions: data rate improvement and analysis. IEEE Access 11, 55107–55116 (2023) 21. Teli, S.R., et al.: Hybrid optical wireless communication for versatile IoT applica- tions: data rate improvement and analysis. IEEE Access 11, 55107–55116 (2023) 22. Saud, M.S., Chowdhury, H., Katz, M.: Heterogeneous software-defined networks: implementation of a hybrid radio-optical wireless network. In: 2017 IEEE Wireless Communications and Networking Conference (WCNC), pp. 1–6 (2017) 22. Saud, M.S., Chowdhury, H., Katz, M.: Heterogeneous software-defined networks: implementation of a hybrid radio-optical wireless network. In: 2017 IEEE Wireless Communications and Networking Conference (WCNC), pp. 1–6 (2017) 23. Otto, C., Milenkovi´c, A., Sanders, C., Jovanov, E.: System architecture of a wireless body area sensor network for ubiquitous health monitoring. J. Mob. Multimed. 1(4), 307–326 (2005) ( ) ( ) 24. Dolev, D., Yao, A.: On the security of public key protocols. IEEE Trans. Inf. Theory 29(2), 198–208 (1983) 25. Calvanese Strinati, E., Barbarossa, S.: 6G networks: beyond Shannon towards semantic and goal-oriented communications. Comput. Netw. 190, 107930 (2021). https://www.sciencedirect.com/science/article/pii/S1389128621000773 26. Du, H., Wang, J., Niyato, D., Kang, J., Xiong, Z., Guizani, M., Kim, D.I.: Rethink- ing wireless communication security in semantic internet of things. IEEE Wirel. Commun. 30(3), 36–43 (2023) 27. References Zafar, S., et al.: A systematic review of bio-cyber interface technologies and security issues for internet of bio-nano things. IEEE Access 9, 93529–93566 (2021) 28. Dutta, D.L., Bharali, S.: TinyML meets IoT: a comprehensive survey. Inter- net Things. 16, 100461 (2021). https://www.sciencedirect.com/science/article/pii/ S2542660521001025 29. Stallings, W.: Cryptography and Network Security: Principles and Practice, 7th edn. Pearson Education Ltd., London (2017) Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
https://openalex.org/W4288414871
https://zenodo.org/records/2590007/files/MMJ-2019-62-1_31-34.pdf
English
null
Informativity of perinatal medical card in fetal growth assessment
DOAJ (DOAJ: Directory of Open Access Journals)
2,019
cc-by
3,509
Abstract Background: Failure to provide antenatal information for pregnant women as well as to complete the perinatal medical records at each visit may have negative effects on the child development. g p Material and methods: A retrospective documentary review study was carried out, which included 100 women of fertile age. The questionnaire included 80 questions within the following rubrics: personal information about the patient, obstetrical-gynecological medical history, and perinatal medical card. Results: The patients were divided into the following age groups <20 years – 10 patients (10.0%), 20-30 years – 67 patients (67.0%), 31-40 years – 21 patients (21.0%) and > – 2 patients (2.0%). The recommended weight gain range is ≤12 kg, whereas 59 patients (59%) reported the highest weight gain ≥ 20 kg and others showed a mean range of 12-20 kg. The assessment of parity’s effects on fetal weight proved that multiparous women give birth to heavyweight newborns. Women in their second pregnancies make up 62%. Vaginal and cesarean births were registered in 88% and 12% of cases, respectively. Spontaneous abortions were reported in 20%, abortions on demand – 21%, and premature births in 5% of cases. The perinatal medical records were fully completed in 49.0% and partially – in 51.0% of cases. p y unt of perinatal medical card fulfillment has reached the lowest level, including the ”gravidogram” that refers to the fetal growth chart m, fetal biometrics, perinatal medical card. p p y Conclusions: The amount of perinatal medical card fulfillment has reached the lowest level, including the ”gravidogram” that refe Key words: gravidogram, fetal biometrics, perinatal medical card. uterine growth restriction (IUGR), based on both FHM and gravidogram. Thus, the ultrasound parameters for diagnosis of IUGR are as follows: bi-parietal diameter (BPD), cranial circumference (CC), abdominal circumference (AC), and femoral length (LF). AC is the most advanced parameter in detecting IUGR of the fetus with a sensitivity of 61% and a specificity of 95%. FL is a prognostic indicator in severe cases of intrauterine growth restriction of vascular origin [5, 6, 7]. Each pregnant woman will undergo three ultrasound scans: at 12-14 weeks, 18-22 weeks, and 30-32 weeks [8]. Due to these examinations, the necessary ultrasound pa- rameters for fetal development can be assessed within the given gestational term [5, 6, 7, 8]. Abstract A dynamic image ultra- sonography of the fetal biometric parameters will be carried out additionally, in case if deviation from normal physical growth is recorded throughout the pregnancy and during the mandatory visits [10, 11]. This may precisely determine failure of a fetus to reach its pre-determined growth poten- tial due to insufficient kinetics of intrauterine growth or ab- normality resulting from the maternal-fetal disorders [12]. Informativity of perinatal medical card in fetal growth assessment *Eugenia Bogonovschi, MD, Undergraduate Student; Ion Bologan, MD, PhD, Associate Professor Department of Obstetrics and Gynecology No 1, Nicolae Testemitsanu State University of Medicine and Pharmacy Chisinau, the Republic of Moldova *Corresponding author: bogonovschi.eugenia@gmail.com Manuscript received January 21, 2019; revised manuscript March 04, 2019 *Corresponding author: bogonovschi.eugenia@gmail.com Manuscript received January 21, 2019; revised manuscript March 04, 2019 E. Bogonovsch et al. Moldovan Medical Journal. March 2019;62(1):31-34 E. Bogonovsch et al. Moldovan Medical Journal. March 2019;62(1):31-34 ORIGINAL ARTICLE DOI: 10.5281/zenodo.2590007 UDC: 618.291-07 DOI: 10.5281/zenodo.2590007 UDC: 618.291-07 Introduction The weight gain of the patients included in the study during the current pregnancy (%). Material and methods Fig. 2. The most common obstetrical complications in patients (%). Fig. 2. The most common obstetrical complications in patients (%). A retrospective documentary study was designed and conducted on 100 women of childbearing age who were re- cruited from the patients in the postpartum period (2-12 days after birth) and admitted within the Department of Obstetrics No 1 and No 2 in the Municipal Clinical Hos- pital No 1, from March to July 2018. The questionnaire was impersonal and did not include any rubrics of personal in- formation.h The questionnaire included 80 questions and was struc- tured according to the following rubrics: patient personal data, obstetrical-gynecological anamnesis, perinatal medi- cal cards. Simultaneously, 12 patients with ultrasound re- sults were interviewed, of which only 5 patients presented the fetal biometric parameters assessed during the three antenatal mandatory visits: (BPD, CC, AC, FL) and the esti- mated fetal weight (EFW) at birth. Fig. 2. The most common obstetrical complications in patients (%). Fig. 2. The most common obstetrical complications in patients (%). Fig. 3. The weight gain of the patients included in the study during the current pregnancy (%). Primary data have been processed via Excel (from Mi- crosoft Office 2010). Introduction Pregnancy is a normal, natural and healthy phenom- enon in a woman’s life. Both the pregnant woman and her family are responsible for her physical and emotional con- dition [1, 2, 3, 4]. The routine prenatal care may include a series of interventions designed to ensure an optimal fetal development. Pregnant women are provided with screen- ing, prophylaxis and counseling. The medical care provided to the pregnant woman and her fetus throughout pregnancy is essential to prevent the early occurrence of any circum- stances that might affect pregnancy outcomes so that they can be treated and monitored. A mandatory standard for antenatal care is the Peri- natal Medical Card (Form 113 / e), approved by Order of the Ministry of Health of the Republic of Moldova No 828, dated of 2011 October 31, that refers to “the approval of the Primary Medical Record Forms” [3, 4, 5]. The perina- tal medical card is an evidence-based record book that is provided free of charge in the first prenatal visit [2, 3, 4]. The perinatal medical card is fulfilled by the family doc- tor or an obstetrician-gynecologist. The pregnant woman keeps the perinatal medical card that will be completed at each medical check-up throughout the entire gestational period [5, 6, 7, 8]. Therefore, the proper assessment of the fetal intrauterine growth is crucial for antenatal care. Fun- dal height measurement (FHM) has a medium diagnostic value in specifying fetuses of a small gestational age (SGA). The steady recording of FHM on the gravidogram increases the sensitivity and specificity of the method [5, 6, 9]. Ul- trasound biometry is indicated in case of a suspected intra- y g Preconception care is of great importance since it deter- mines both the pregnancy outcomes and the health of the future child [10, 11]. It has been recognized that constant monitoring of physiological changes during pregnancy helps to prevent complications by early detection and emer- gency treatment, which are essential for maintenance of the pregnancy as a normal physiological process [13].h The purpose of this study is to assess fetal growth by using standardized fetal growth charts related to the gesta- tional age, as well as the study results via FHM and fetal biometric parameters from ultrasound data. 31 E. Bogonovsch et al. Moldovan Medical Journal. March 2019;62(1):31-34 ORIGINAL ARTICLE Fig. 2. The most common obstetrical complications in patients (%). Fig. 3. Results and discussion According to their age, patients were divided into the following age groups: <20 years – 10 patients (10.0%), 20- 30 years – 67 patients (67.0%), 31-40 years – 21 patients (21.0%) and > 40 years – 2 patients (2.0%), (fig. 1). Fig. 1. Distribution of patients into age groups (%). Fig. 3. The weight gain of the patients included in the study during the current pregnancy (%). Most of the clinical studies that assessed the impact of parity on fetal weight have concluded that multiparous women give birth to heavier weight fetuses. Based on the research, we determined that 34 women were at their first birth (34.0%), 62 women were in their secondary pregnancy (62.0%), and the lowest rate was registered in multiparous women – 4 cases (4.0%), (fig. 4). Fig. 4. Influence of parity on fetal weight (%). Fig. 4. Influence of parity on fetal weight (%). Fig. 1. Distribution of patients into age groups (%). The distribution of interviewees according to the place of residence is as follows: 76 patients (76.0%) came from ur- ban areas and 24 patients (24.0%) – from rural areas. Twenty patients (20.0%) were reported to have a previ- ous early miscarriage in anamnesis, 4 patients (4.0%) – late spontaneous abortion, and 5 cases (5.0%) resulted in pre- mature death. Previous abortions on demand were also considered within the study. It should be noted that 21.0% of patients presented a history of abortions on demand (fig. 2).h Fig. 4. Influence of parity on fetal weight (%). According to the study, 88 cases resulted in natural birth and 12 women (12.0%) underwent caesarean section. As- sessment of pregnancy term deliveries showed 89 (89.0%) full-term pregnancies, 6 cases (6.0%) resulted in premature childbirth and 5 cases (5.0% ) reported birth at the gesta- tional age of 41-42 weeks (fig. 5). The present study reported a weight gain within the recommended range (≤12 kg) in 14 patients (14.0%), an increased weight gain between 12 -20 kg was found in 27 patients (27.0%) and an overweight ≥ 20 kg – in 59 patients (59.0%), (fig. 3). 32 E. Bogonovsch et al. Moldovan Medical Journal. March 2019;62(1):31-34 ORIGINAL ARTICLE Fig. 5. The delivery term and types of childbirth (%). Results and discussion Table 1 The fulfillment of the gravidogram based on the perinatal medical cards used in the study Gravidogram Fulfillment of the gravidogram based on the perinatal medical cards Absolute value Percentage Complete 85 85.0% Incomplete 15 15.0% The Ministry of Health recommends to carry out 6 ante- natal care visits during pregnancy, of which 2 are standard visits to the obstetrician-gynecologist. It should be noted that WHO recommends at least 4 antenatal care visits for a normal pregnancy [2, 3, 15]. Table 2 shows the number of visits during pregnancy. It has been determined that most patients – 62 cases (62.0%) reported to have 6 antenatal visits, 26 patients (26.0%) – 5 visits, 10 patients (10.0%) – 4 visits and 2 patients (2.0%) – 3 visits (tab. 2). Fig. 5. The delivery term and types of childbirth (%). Table 2 Number of antenatal care visits Number of antenatal care visits Number of antenatal care visits Absolute value Percentage 1visit - - 2 visits - - 3 visits 2 2.0% 4 visits 10 10.0% 5 visits 26 26.0% 6 visits 62 62.0% Fig. 5. The delivery term and types of childbirth (%). Proper assessment of the fetal intrauterine growth is the key task of the antenatal care [16, 17]. UFH measurement has a diagnostic value in predicting the fetal weight. It is essential to determine the gestational age and identify the abnormal growth rates in pregnant women since these may lead to a re- duced infant mortality rate. The gestational age is assessed by using clinical criteria such as uterus size measurement, data regarding the last menstrual period, or ultrasound criteria. The last menstrual period data exhibits a rather high degree of errors since some pregnant women do not remember ex- actly the time of the last menstrual period or it did not last for 28 days in all cases. UFH shows a low net value since it can erroneously influence the height of the pregnant woman, some abdominal tumors or uterine fibromas [9]. Therefore, the ultrasound indices are still the most relevant ones [13, 16, 17]. The estimated fetal weight in relation to pregnancy term and actual birth weight was difficult to assess due to the lack / incomplete recording of all ultrasound data (BPD, CC, AC, FL) in the perinatal medical book.h Proper assessment of the fetal intrauterine growth is the key task of the antenatal care [16, 17]. UFH measurement has a diagnostic value in predicting the fetal weight. It is essential to determine the gestational age and identify the abnormal growth rates in pregnant women since these may lead to a re- duced infant mortality rate. The gestational age is assessed by using clinical criteria such as uterus size measurement, data regarding the last menstrual period, or ultrasound criteria. h The increase of the gestational age may increase both the mass index (MI) and the birth weight [14]. Studies of the fetal birth weight showed the highest rate in 84 (84.0%) new- borns with the weight ranging from 2800 g – 3999 g, whereas 9 cases (9.0%) reported a fetal weight greater than 4000 g and 7 cases (7.0%) had less than 2800g at birth (fig. 6). g regarding the last menstrual period, or ultrasound criteria. References 1. Centrul National de Studii pentru Medicina Familiei. Ingrijiri prenatale in sarcina cu risc scazut. Ghid de practica pentru medicul de familie [Routine prenatal care in low-risk pregnancies. Family Practice guide- lines]. Bucharest: InfoMedica; 2005. p. 18. Romanian. 1. Centrul National de Studii pentru Medicina Familiei. Ingrijiri prenatale in sarcina cu risc scazut. Ghid de practica pentru medicul de familie [Routine prenatal care in low-risk pregnancies. Family Practice guide- lines]. Bucharest: InfoMedica; 2005. p. 18. Romanian. p 2. Stratulat P, Curteanu A. Fortificarea sistemului asistenţei serviciilor perinatale de calitate [Enhancement of the quality perinatal-assisted care system]. Buletinul Academiei de Ştiinţe a Moldovei. Ştiinţe Medicale [Bulletin of the Academy of Sciences of Moldova. Medical Sciences]. 2007;(2/11):244-255. Romanian. 2. Stratulat P, Curteanu A. Fortificarea sistemului asistenţei serviciilor perinatale de calitate [Enhancement of the quality perinatal-assisted care system]. Buletinul Academiei de Ştiinţe a Moldovei. Ştiinţe Medicale [Bulletin of the Academy of Sciences of Moldova. Medical Sciences]. 2007;(2/11):244-255. Romanian. 3. Principii de organizare şi acordare a asistenţei perinatale: Ghidul A Naţional de Perinatologie [Principles of organization and perinatal care assistance: National Guidelines for Perinatal Care]. 2nd ed. Chisinau; 2006. 165 p. Romanian. p 4. [Ministry of Health of the Republic of Moldova]. [Order no. 828 of 2011 October 31, on the approval of the Primary Medical Forms]. [cited 2018 Aug 9]. Available from: http://old2.ms.gov.md/sites/default/files/ legislatie/ordinul_nr._828_din_31.10.2011.pdf. Romanian. 4. [Ministry of Health of the Republic of Moldova]. [Order no. 828 of 2011 October 31, on the approval of the Primary Medical Forms]. [cited 2018 Aug 9]. Available from: http://old2.ms.gov.md/sites/default/files/ legislatie/ordinul_nr._828_din_31.10.2011.pdf. Romanian. 5. [National Center for Health Management of the Republic of Moldova]. [Record forms] [cited 2018 Aug 9]. Available from: http://www.cnms. md. Romanian. 6. Rusu C, Stretean A. Ultrasound diagnosis of fetal macrosomia. Acta Medica Transilvanica. 2012;2(2):225-228. 7. Stratulat P, Curteanu A, Carauş T. Experienţa utilizării curbelor individualizate de creştere intrauterină a fătului pentru diagnosticul problemelor de creştere fetală [The experience of using individualized intrauterine growth charts for the diagnosis of fetal growth disor- ders]. Buletinul Academiei de Ştiinţe a Moldovei. Ştiinţe Medicale [Bulletin of the Academy of Sciences of Moldova. Medical Sciences]. 2009;(1/20):235-240. Romanian. Table 2 The last menstrual period data exhibits a rather high degree of errors since some pregnant women do not remember ex- actly the time of the last menstrual period or it did not last for 28 days in all cases. UFH shows a low net value since it can erroneously influence the height of the pregnant woman, some abdominal tumors or uterine fibromas [9]. Therefore, the ultrasound indices are still the most relevant ones [13, 16, 17]. The estimated fetal weight in relation to pregnancy term and actual birth weight was difficult to assess due to the lack / incomplete recording of all ultrasound data (BPD, CC, AC, FL) in the perinatal medical book.h Fig. 6. Fetal birth weight (%). Fig. 6. Fetal birth weight (%). The estimated fetal weight in relation to the clinical pa- rameters of UFH and the abdominal circumference (AC) during 36-38 weeks of gestation according to the Iakubov formula: (UFH + AC) / 4 * 100 has also been considered.h The perinatal medical card is a mechanism for recording information related to antenatal follow-up and deliveries [15]. According to the amount of the perinatal medical card fulfillment, we found that 49.0% were satisfactorily fulfilled and 51.0% of them were partially or unsatisfactorily filled (tab. 1). This calculation formula has shown higher veracity than other formulas within a national study [9]. h The estimated intrauterine weight of the fetus was as- 33 E. Bogonovsch et al. Moldovan Medical Journal. March 2019;62(1):31-34 ORIGINAL ARTICLE sessed only in 5 cases, which included the ultrasound pa- rameters evaluated at mandatory antenatal visits and re- corded in the medical record cards of these patients. Thus, the intrauterine growth charts were carried out individually for each case, whilst fetal growth assessment was related to 10-90 percentiles. In three cases out of five, the intrauterine weight was found within the 10-90th percentile in relation to the newborn’s weight. The ultrasound is considered the method of choice in the diagnosis of fetal IUGR, having a specificity of 80-90%. However, ultrasonography also moni- tors the intrauterine growth process and allows the assess- ment of a growth abnormality and its degree of severity. Fetal IUGR is characterized by decreased parameters that define the process of intrauterine development of the fetus (the weight, waist, skull circumference, abdominal, thorac- ic, subcutaneous tissue and muscular mass). References The ultrasonic fetal cephalometry that is consecutively performed by assessing BPD (sensitivity – 89% and positive prognosis – 68%) and CC (sensitivity – 63% and positive prognosis-75%) is useful not only for detecting the risk of developing IUGR in fetuses in relation to individual intra- uterine growth potential, but it also helps to differentiate between the symmetrical forms of the disease and asym- metrical ones (the method is sensitive in 94% of cases of symmetrical forms and 42% for the asymmetrical ones). Thus, the assessment was not properly performed due to the lack of ultrasound data, which should have been included within the specific rubric of the perinatal medical card. 8. National Institute for Clinical Excellence. Antenatal care: routine care for the healthy pregnant woman. London: NICE; 2003. (NICE Clinical Guideline, no. 6). 9. Corolcova N, Alaeldin M, Rusu C, Raetcaea V. Caracteristica compara- tiva a metodelor de calcul a masei probabile a fatului [Comparative characteristics of methods of calculating the estimated fetal weight]. Scientific Annals of the Nicolae Testemitsanu State University of Medi- cine and Pharmacy (Chisinau). 2008;9(5):37-40. Romanian. 10. [Ministry of Health of Romania, Romanian Association of Neonatolo- gy]. Determinarea vârstei de gestaţie la nou-născut [Determining the Age of Gestation in the Newborn]. Bucharest; 2011. 16 p. (Clinical Guide Collection for Neonatology, 01). Romanian. Table 2 The estimated ultrasound parameters which provide the diagnosis IUGR of the fetus are as following: BPD CC, AC, and FL. Refer- ring to fetal IUGR, the optimal assessment of the individual intrauterine growth rate is being considered, as well as both the mean fetal weight (MFW) at birth and the maternal factor. Therefore, two successive ultrasound examinations within about 14 days apart should be performed for the purpose of a reliable assessment of fetal growth dynamics [13, 16, 17]. h trasound examination should be integrated into the clinical context and for each case apart, as well as the population- specific growth curves should be thoroughly considered. Conclusions 11. Copilul.ro [Internet]. Bucharest; c2004-2019 [cited 2018 Aug 10]. [The length and weight of the fetus every week of pregnancy]. Available from: https://www.copilul.ro/sarcina/despre-sarcina/Lungimea-si-greutatea- fatului-in-fiecare-saptamana-de-sarcina-a11386.html. Romanian. 1. The present study proved that the perinatal medical cards were satisfactorily completed in 49.0% of cases, and partially or unsatisfactorily fulfilled in 51.0% of cases. 1. The present study proved that the perinatal medical cards were satisfactorily completed in 49.0% of cases, and partially or unsatisfactorily fulfilled in 51.0% of cases. 12. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. United States national reference for fetal growth. Obstet Gynecol. 1996;87(2):163-8. i 2. Based on the study, we found that the pregnancy was completed at 37-40 weeks of gestation in 89 cases (89.0%), whereas the fetal weight ranged from 2800 g – 3999g in 84 cases (84.0%), greater than 4000 g in 9 cases (9.0%) and less than 2800 g in 7 cases (7.0%).hi 13. Kiserud T, Piaggio G, Carroli G, Widmer M, Carvalho J, et al. WHO Fetal growth charts: a multinational longitudinal study of ultrasound biometric measurements and estimated fetal weight. PLOS Medicine. 2017;14(3): e1002284. doi:10.1371/journal.pmed.1002220. j p 14. Goldstein RF, et al. Association of gestational weight with maternal and infant outcomes: a systematic review and meta-analysis. JAMA. 2017;317(21):2207-2225. 3. The gravidogram was satisfactorily fulfilled in 85.0% of cases within the current study. y 4. It has been determined that the accuracy of the fetal weight according to the Iakubov formula made up 49.93%, thus allowing to identify indices under the 10th percentile and the 90th percentile for the gestational age in 30.61% of cases. 15. Bologan I, et al. The impact of the perinatal medical card within the perinatal care service. In: [Proceedings of the 5th Congress of Obstet- rics and Gynecology, with international participation; 2010 Oct 7-8; Chisinau, Republic of Moldova]. Chisinau: Ericon; 2010. p. 8-11. 16. Tucker J, Parry G, Fowlie PW, McGuire W. Organisation and delivery of perinatal services. BMJ. 2004;329(7468):730-732. 5. The study results proved the necessity of interpreting the perinatal fetal weight by using the ultrasound indices and their recording within individualized growth charts. The ul- 17. Gardosi J, Francis A. Controlled trial of fundal height measurement plotted on customised antenatal growth charts. Br J Obstet Gynaecol. 1999;106(4):309-17. 34
https://openalex.org/W2118637764
https://www.arkat-usa.org/get-file/32735/
English
null
Acid-catalyzed intramolecular oxa-Michael addition reactions under solvent-free and microwave irradiation conditions
ARKIVOC
2,010
cc-by
4,798
Abstract The acid-catalyzed intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methyl- -pent-1-en-3-ones under solvent-free and microwave irradiation conditions has been investigated. The results showed that Bronsted acids are more efficient than Lewis acids in this reaction. Up to 90% conversion and 81% yield were obtained using trifluoromethanesulfonic acid (triflic acid) as the catalyst, with short reaction times and an environmentally benign procedure. Keywords: Microwave, oxa-Michael addition, solvent-free, trifluoromethanesulfonic acid, (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones ARKIVOC 2009 (xiv) 146-155 ARKIVOC 2009 (xiv) 146-155 General Papers Yi-Ming Hong, Zhen-Lu Shen,* Xin-Quan Hu, Wei-Min Mo, Xiao-Fei He, Bao-Xiang Hu, and Nan Sun College of Chemical Engineering and Materials Science, State Key Laboratory Breeding Base of Green Chemistry-Synthesis Technology, Zhejiang University of Technology, Hangzhou, Zhejiang 310014, China College of Chemical Engineering and Materials Science, State Key Laboratory Breeding Base of Green Chemistry-Synthesis Technology, Zhejiang University of Technology, Hangzhou, Zhejiang 310014, China E-mail: zhenlushen@zjut.edu.cn E-mail: zhenlushen@zjut.edu.cn Acid-catalyzed intramolecular oxa-Michael addition reactions under solvent-free and microwave irradiation conditions Yi-Ming Hong, Zhen-Lu Shen,* Xin-Quan Hu, Wei-Min Mo, Xiao-Fei He, Bao-Xiang Hu, and Nan Sun Introduction Organic chemists are constantly challenged to develop environmentally benign routes for preparing desired target compounds. Microwave irradiation has emerged as a convenient and powerful method for promoting organic reactions, owing to its uniform heating effect.1-3 Another green chemistry principle – solvent-free organic synthesis – has also received much attention from organic chemists. The combination of solvent-free and microwave irradiation can lead to large reductions in reaction times, or enhancements in conversions, and will make synthesis more environmentally benign.4-6 Michael and hetero-Michael additions are effective for the formation of carbon-carbon and carbon- hetero-atom bonds, and have been utilized widely in inter- and intra- molecular reactions to obtain various products, which include important building blocks for many biologically active molecules.7,8 Recently, the conjugate addition of oxygen nucleophiles to α,β-unsaturated carbonyl compounds (the oxa-Michael addition) has attracted much attention.9 Various catalysts have been reported for intermolecular oxa-Michael additions, such as pro-azaphosphatranes ISSN 1551-7012 ©ARKAT USA, Inc. ©ARKAT USA, Inc. Page 146 Page 146 ARKIVOC 2009 (xiv) 146-155 General Papers (P(RNCH2CH2)3N),10 trimethylphosphine, tributylphosphine,11 1,8-diazabicyclo[5.4.0]undec-7- ene (DBU),12 [Rh(COD)(OMe)]2,13 (CH3CN)2PdCl2,14 (1,3-bis-(2,6-di-isopropylphenyl)imidazol- 2-yliden)Cu(OEt), (1,3-bis-(2,6-di-isopropylphenyl)imidazol-2-yliden)Cu(OPh), (1,3-bis-(2,6- di-isopropylphenyl)imidazolin-2-ylidene)Cu(OEt),15 KF/Al2O3,16 (CF3SO2)2NH,17 BF3·Et2O,18 La(NO3)3·6H2O,19 pyrrolidine/CH3SO3H20 and biphenyldiamine-based organocatalyst.21 The use of ionic liquid as solvent for toluenesulfonic acid-catalyzed oxa-Michael addition has also been reported.22 For intramolecular oxa-Michael addition, cinchona alkaloids,23 thiourea derivatives,24 quinine,25 guanidine,26 [Pd(MeCN4)](BF4)2,27 N,N-dioxide nickel(II) complex,28 CF3COOH29 and toluenesulfonic acid (TsOH)30,31 have been utilized as the catalysts. It was reported that the intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones in the presence of TsOH should be carried out by refluxing in dichloroethane for 24 h, while the products were hard to purify due to the impurities.31 In continuation of our work on “green” chemistry,32 we report herein an efficient acid-catalyzed intramolecular oxa-Michael addition reaction of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones 1 under solvent-free and microwave irradiation conditions. Results and Discussion Effect of catalysts. In order to find the most efficient catalyst, we studied first the intramolecular oxa-Michael addition of (E)-4-hydroxy-4-methyl-1-phenylpent-1-en-3-one 1a under solvent-free and microwave irradiation conditions. The results are given in Table 1. Table 1. Effect of catalysts in the intramolecular oxa-Michael addition of 1aa Table 1. Effect of catalysts in the intramolecular oxa-Michael addition of 1aa Entry Catalyst Tfinal (oC) Conversion (%)b Selectivity (%)b Yield (%)c 1 Zn(OTf)2 141 32 83 20 2 Cu(OTf)2 143 59 87 46 3 TsOH·H2O 118 69 91 57 4 CH3SO3H 115 64 97 58 5 TfOH 117 88 94 78 6 DBU 116 0 - - 7 DABCO 116 0 - - 8 (CH3CN)2PdCl2 165 96 16 - a R ti diti 1 (0 5 l) t l t (0 05 l) 650 W i 9 i t t l a Reaction conditions: 1a (0.5 mmol), catalyst (0.05 mmol), 650 W microwave power, 9 min total reaction time, with alternation between 9 s of irradiating time and 21 s of cooling time. a Reaction conditions: 1a (0.5 mmol), catalyst (0.05 mmol), 650 W microwave power, 9 min total reaction time, with alternation between 9 s of irradiating time and 21 s of cooling time. b Conversion of 1a and selectivity to 2a were measured on the reaction mixture by gas chromatography. c Isolated yield. b Conversion of 1a and selectivity to 2a were measured on the reaction mixture by gas chromatography. c Isolated yield. Page 147 ©ARKAT USA, Inc. ISSN 1551-7012 ©ARKAT USA, Inc. Page 147 ARKIVOC 2009 (xiv) 146-155 General Papers General Papers When the Lewis acid zinc trifluoromethanesulfonate (Zn(OTf)2) was used as catalyst, the conversion of 1a and the selectivity to 2,2-dimethyl-5-phenyl-dihydrofuran-3(2H)-one 2a were 32 and 83%, respectively, and the isolated yield of 2a was only 20% (entry 1). The other Lewis acid, copper trifluoromethanesulfonate (Cu(OTf)2) showed higher activity, and a 59% conversion with 87% selectivity was obtained (entry 2). However, it was found that Bronsted acids are more efficient than Lewis acids in this reaction (entries 3-5). Among the tested Bronsted acids trifluoromethanesulfonic acid (triflic acid, TfOH), TsOH, and CH3SO3H, the TfOH gave the best result, with up to 88% conversion and 94% selectivity (entry 5). Results and Discussion Organic bases, such as DBU and 1,4-diazabicyclo[2.2.2]octane (DABCO), were also examined but showed no activity in this intramolecular oxa-Michael addition (entries 6 and 7), although they have been employed as catalysts in some intermolecular oxa-Michael additions.12,33 The highest conversion (96%) of 1a was observed in the presence of (CH3CN)2PdCl2, but many unidentified by-products were observed and the selectivity to 2a was only 16% (entry 8). Comparison of plausible reaction mechanism in the presence of Bronsted acids and Lewis acid A plausible reaction mechanism for the intramolecular oxa-Michael addition of 1a, promoted with Bronsted acid, as given by Johnson’s group, is outlined in Scheme 1.31,34 The carbonyl oxygen of 1a accepts one proton from the Bronsted acid to give the protonated form I and canonical form II, and undergoes ring closure to give the enol form III of 2a. OH OH OH OH O OH H+ OH OH OH OH O OH O O H 1a I II III 2a H Scheme 1. Plausible reaction mechanism in the presence of Bronsted acids. OH OH OH OH O OH H+ 1a OH O OH H 1a III H O O I 2a O O I II 2a I 1a II III 2a Scheme 1. Plausible reaction mechanism in the presence of Bronsted acids. ISSN 1551-7012 Page 148 Page 148 ©ARKAT USA, Inc. ISSN 1551-7012 Page 148 Page 148 ©ARKAT USA, Inc. ISSN 1551-7012 ARKIVOC 2009 (xiv) 146-155 General Papers General Papers It is presumed that in the presence of Bronsted acid, there exists an equilibrium between 1a and 2a.31 When a mixture of 2a (0.5 mmol) and TfOH (0.05 mmol) was irradiated at 650 W for 9 min, it was found that 13% of 2a did convert into 1a. It is presumed that in the presence of Bronsted acid, there exists an equilibrium between 1a and 2a.31 When a mixture of 2a (0.5 mmol) and TfOH (0.05 mmol) was irradiated at 650 W for 9 min, it was found that 13% of 2a did convert into 1a. Spencer and colleagues have reported that protons generated through in situ hydrolysis of metal salts were the active catalysts in some Lewis acid mediated hetero-Michael additions.35 In order to ascertain whether protons are the actual catalysts in this intramolecular oxa-Michael addition in the presence of Lewis acids, Zn(OTf)2 and Cu(OTf)2, the above experiment was repeated using Zn(OTf)2 in place of TfOH. The experiments showed that no 2a was converted into 1a after irradiating for 9 min, while our previous experiments showed both Lewis acids and Bronsted acids could catalyze this intramolecular oxa-Michael reaction (Table 1, entries 1 and 2 versus 3-5). These results suggest that protons were not the active catalysts in this Lewis acid-catalyzed oxa-Michael addition. Based on our experimental results we propose a plausible reaction mechanism shown in Scheme 2. Comparison of plausible reaction mechanism in the presence of Bronsted acids and Lewis acid Zn(OTf)2 and Cu(OTf)2 are highly oxophilic, and could form a carbonyl– metal-ion complex IV with 1a, which would initiate the formation of a C-O bond to generate the intermediate V. This could regenerate the catalyst and give the intermediate VI. After the intramolecular proton transfer step, 2a was obtained. O M(OTf)2 O H M(OTf)2 O OH M(OTf)2 O O O O H 1a IV VI 2a Scheme 2. Plausible reaction mechanism in the presence of M(OTf)2. O M(OTf)2 O H O M( HO O OH M(OTf)2 1a IV V O M(OTf)2 O H Tf)2 IV O M(OTf)2 HO V O OH M(OTf 1a V IV 1a O O 2a M(OTf)2 O O H VI VI 2a Scheme 2. Plausible reaction mechanism in the presence of M(OTf)2. Intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones Intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones To evaluate the scope of the intramolecular oxa-Michael addition, substrates with different steric property and electron property have been investigated using TfOH as the catalyst under the microwave-assisted process. As shown in Table 2, most of the reactions proceeded efficiently, high conversions of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones 1 and selectivities to the ©ARKAT USA, Inc. ISSN 1551-7012 Page 149 ARKIVOC 2009 (xiv) 146-155 General Papers desired products 5-aryl-2,2-dimethyl-dihydrofuran-3(2H)-ones 2 were obtained within several minutes. O O 2a O O 2b O O 2c O O 2d O O Cl 2e O O Cl 2f O O Cl 2g O O OMe 2h O O NO2 2i O O 2j O O O 2k O O N 2l O O 2d O O 2b O O 2a 2d 2a 2a O O Cl 2e 2c O O Cl 2g O O OMe 2h 2b O O Cl 2b O O OMe 2h Cl 2h 2h O O N 2l Cl 2e Cl 2f O O NO2 2i O O 2j 2g O O O 2k 2f O O 2j 2f 2e 2j 2i 2l 2k It was observed that the position of substituents on the benzene ring had little effect on the conversions and selectivities (entries 2-4, methyl as substituent; entries 5-6, chlorine as substituent). Also, the electronic properties of substituents on the benzene ring showed marginal effects on the conversions and selectivities (entry 4, p-Me; entry 7, p-Cl; entry 8, p-OCH3), even with a strongly electron-withdrawing group (entry 9, p-NO2). A similar result was obtained when a naphthalene ring replaced a benzene ring, the conversion of 1j and selectivity to 2j were 84 and 95%, respectively (entry 10). When (E)-1-(furan-2-yl)-4-hydroxy-4-methylpent-1-en-3-one 1k was used as the substrate, the irradiating period and reaction time must be shortened. Otherwise, the selectivity to 2k would drastically decrease, which may be due to the polymerization of 2k. Under our conditions, 17% conversion of 1k and 8% isolated yield of 2k were obtained (entry 11). The intramolecular oxa-Michael addition of (E)-4-hydroxy-4-methyl-1-(pyridin-3-yl)- -pent-1-en-3-one 1l could not occur under microwave irradiation or refluxing conditions because the proton of Bronsted acid would be captured by the N- atom of the pyridinyl group. ISSN 1551-7012 ©ARKAT USA, Inc. Page 150 ARKIVOC 2009 (xiv) 146-155 General Papers General Papers ARK Table 2. Intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones under solvent-free and microwave conditionsa Table 2. Intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones d Total reaction time 3 min, with an alternation between 3 s of irradiating time and 27 s of cooling time. Intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones Intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-ones under solvent-free and microwave conditionsa O O Ar Ar O HO TfOH (10 mol%) MW 1a-l 2a-l Entry Product Tfinal (oC) Conversion (%) Selectivity (%) Yield (%) b 1 2a 117 88 94 78 2 2b 116 85 94 62 3 2c 116 83 93 60 4 2d 117 81 95 63 5 2e 115 90 97 80 6 2f 116 90 98 81 7 2g 116 90 96 80 8c 2h 97 80 92 65 9 2i 117 81 85 55 10 2j 116 84 95 71 11d 2k 30 17 94 8 12 2l - 0 - - a Reaction conditions: 1 (0.5 mmol), TfOH (0.05 mmol), 650 W (microwave power), 9 min (total reaction time, with an alternation between 9 s of irradiating time and 21 s of cooling time). b Isolated yield. c With an alternation between 6 s of irradiating time and 24 s of cooling time. d Total reaction time 3 min, with an alternation between 3 s of irradiating time and 27 s of cooling time. Entry Product Tfinal (oC) Conversion (%) Selectivity (%) Yield (%) b 1 2a 117 88 94 78 2 2b 116 85 94 62 3 2c 116 83 93 60 4 2d 117 81 95 63 5 2e 115 90 97 80 6 2f 116 90 98 81 7 2g 116 90 96 80 8c 2h 97 80 92 65 9 2i 117 81 85 55 10 2j 116 84 95 71 11d 2k 30 17 94 8 12 2l - 0 - - a Reaction conditions: 1 (0.5 mmol), TfOH (0.05 mmol), 650 W (microwave power), 9 min (total reaction time, with an alternation between 9 s of irradiating time and 21 s of cooling time). b Isolated yield. c With an alternation between 6 s of irradiating time and 24 s of cooling time. d Total reaction time 3 min, with an alternation between 3 s of irradiating time and 27 s of cooling time. a Reaction conditions: 1 (0.5 mmol), TfOH (0.05 mmol), 650 W (microwave power), 9 min (total reaction time, with an alternation between 9 s of irradiating time and 21 s of cooling time). b Isolated yield. c With an alternation between 6 s of irradiating time and 24 s of cooling time. Conclusions The acid-catalyzed intramolecular oxa-Michael addition of (E)-1-aryl-4-hydroxy-4-methylpent- -1-en-3-ones could be performed under solvent-free and microwave irradiation conditions, providing an environmentally benign method for preparing 5-aryl-2,2-dimethyl-dihydrofuran- 3(2H)-ones. It was found that Bronsted acids are more efficient than Lewis acids in this reaction. Unlike traditional reactions, the solvent-free and microwave-assisted reactions could be finished within several minutes to give high conversions and selectivities. ISSN 1551-7012 Page 151 ©ARKAT USA, Inc. ARKIVOC 2009 (xiv) 146-155 General Papers Experimental Section General. 1H- and 13C- NMR spectra were obtained on a Bruker Avance III (500 MHz) spectrometer. CDCl3 was used as the solvent with tetramethylsilane (TMS) as the internal standard. Low and high resolution mass spectra were recorded in the EI mode on a Waters GCT Premier mass spectrometer. Melting points were measured using CRC-1 melting point instrument and are uncorrected. Microwave experiments were performed at a LWMC-201 microwave reactor. The reaction temperature was determined by an IR thermometer (SUN-GUN SG-20). Reactions were monitored by gas chromatography (GC-6890) with an HP-5 capillary column (30m x 0.25mm). All reagents were obtained from commercial sources and used as received. (E)-1-Aryl-4-hydroxy-4-methylpent-1-en-3-ones were prepared as previously described.30,36 General procedure for the synthesis of 5-aryl-2,2-dimethyl-dihydrofuran-3(2H)-ones (2a-k) To an open glass tube, were added (E)-1-aryl-4-hydroxy-4-methylpent-1-en-3-one (0.5 mmol) and triflic acid (0.05 mmol). The tube was positioned in the centre of the microwave cavity, and irradiated (650 W) for 9 min (total reaction time, with alternation of 9s irradiating time and 21s of cooling time). After the last period of irradiating, the temperature (on the surface of the tube) was determined by an IR thermometer. After cooling, the reaction mixture was diluted with CH2Cl2 and purified through a column of silica gel to obtain the pure products 2a-k. 2,2-Dimethyl-5-phenyl-dihydrofuran-3(2H)-one (2a). Yellow oil. 1H NMR (CDCl3) δ 1.33 (s, 3H, -CH3), 1.41 (s, 3H, -CH3), 2.56 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.87 (dd, J = 6 Hz; J = 18.5 Hz, 1H, H-4b), 5.21 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.30-7.34 (m, 1H, phenyl-H), 7.37-7.42 (m, 4H, phenyl-H). 13C NMR (CDCl3) δ 21.70 (-CH3), 24.47 (-CH3), 44.24 (C-4), 74.51 (C-5), 81.51 (C-2), 126.13 (phenyl-C), 128.32 (phenyl-C), 128.78 (phenyl-C), 140.69 (phenyl-C), 217.35 (O=C). MS (EI), m/z (%) 190.1 (M+, 2), 104.1 (100). ( ) ( ) ( ) ( ) ( ) 2,2-Dimethyl-5-o-tolyl-dihydrofuran-3(2H)-one (2b). Yellow oil. 1H NMR (CDCl3) δ 1.35 (s, 3H, -CH3), 1.43 (s, 3H, -CH3), 2.34 (s, 3H, phenyl-CH3), 2.46 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.90 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.38 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.17 (d, J = 7.5 Hz, 1H, phenyl-H), 7.20-7.26 (m, 2H, phenyl-H), 7.56 (d, 1H, phenyl-H). Experimental Section 13C NMR (CDCl3) δ 21.18 (phenyl-CH3), 21.62 (-CH3), 24.41 (-CH3), 44.15 (C-4), 74.33 (C-5), 81.39 (C-2), 126.03 (phenyl-C), 129.34 (phenyl-C), 137.54 (phenyl-C), 138.02 (phenyl-C), 217.54 (O=C). MS (EI), m/z (%) 204.1 (M+, 3), 118.1 (100). 5-(2-Chlorophenyl)-2,2-dimethyl-dihydrofuran-3(2H)-one (2e). Yellow oil. 1H NMR (CDCl3) δ 1.35 (s, 3H, -CH3), 1.44 (s, 3H, -CH3), 2.33 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 3.13 (dd, J = 6 Hz; J = 18.5 Hz, 1H, H-4b), 5.19 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.23-7.27 (m, 3H, phenyl-H), 7.32-7.37 (m, 1H, phenyl-H), 7.67-7.67 (m, 1H, phenyl-H). 13C NMR (CDCl3) δ 21.71 (-CH3), 24.21 (-CH3), 42.67 (C-4), 71.35 (C-5), 81.27 (C-2), 126.37 (phenyl-C), 127.33 (phenyl-C), 128.91 (phenyl-C), 129.36 (phenyl-C), 131.90 (phenyl-C), 138.97 (phenyl-C), 216.74 (O=C). MS (EI), m/z (%) 223.9 (M+, 2), 137.6 (100). HRMS Calcd. for C12H13O2Cl 224.0604, found 224.0610. 5-(3-Chlorophenyl)-2,2-dimethyl-dihydrofuran-3(2H)-one (2f). Yellow oil. 1H NMR (CDCl3) δ 1.32 (s, 3H, -CH3), 1.41 (s, 3H, -CH3), 2.50 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.80 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.19 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.26-7.33 (m, 3H, phenyl-H), 7.42 (s, 1H, phenyl-H). 13C NMR (CDCl3) δ 21.59 (-CH3), 24.32 (-CH3), 44.00 (C-4), 73.71 (C-5), 81.59 (C-2), 124.12 (phenyl-C), 126.09 (phenyl-C), 128.31 (phenyl-C), 129.98 (phenyl-C), 134.64 (phenyl-C), 142.79 (phenyl-C), 216.53 (O=C). MS (EI), m/z (%) 224.1 (M+, 2), 138.0 (100). 5-(4-Chlorophenyl)-2,2-dimethyl-dihydrofuran-3(2H)-one (2g). Yellow oil. 1H NMR (CDCl3) δ 1.32 (s, 3H, -CH3), 1.40 (s, 3H, -CH3), 2.49 (dd, J = 10 Hz; J = 18 Hz, 1H, H-4a), 2.87 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.19 (dd, J = 10 Hz; J = 6 Hz, 1H, -CH), 7.33-7.37 (m, 4H, phenyl-H). 13C NMR (CDCl3) δ 21.71 (-CH3), 24.46 (-CH3), 44.19 (C-4), 73.86 (C-5), 81.67 (C-2), 127.49 (phenyl-C), 128.97 (phenyl-C), 134.02 (phenyl-C), 139.33 (phenyl-C), 216.84 (O=C). MS (EI), m/z (%) 224.1 (M+, 2), 138.0 (100). 5-(4-Methoxyphenyl)-2,2-dimethyl-dihydrofuran-3(2H)-one (2h). Yellow oil. 1H NMR (CDCl3) δ 1.32 (s, 3H, -CH3), 1.39 (s, 3H, -CH3), 2.56 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.83 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 3.81 (s, 3H, -OCH3), 5.16 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 6.90-6.93 (m, 2H, phenyl-H), 7.33-7.35 (m, 2H, phenyl-H). Experimental Section 13C NMR (CDCl3) δ 19.19 (phenyl-CH3), 21.60 (-CH3), 24.31 (-CH3), 42.82 (C-4), 71.44 (C-5), 81.11 (C-2), 124.77 (phenyl-C), 126.52 (phenyl-C), 127.80 (phenyl-C), 130.41 (phenyl-C), 134.80 (phenyl-C), 138.79 (phenyl-C), 217.46 (O=C). MS (EI), m/z (%) 204.1 (M+, 3), 118.1(100). HRMS Calcd. for C13H16O2 204.1150, found 204.1158. 2,2-Dimethyl-5-m-tolyl-dihydrofuran-3(2H)-one (2c). Yellow oil. 1H NMR (CDCl3) δ 1.32 (s, 3H, -CH3), 1.41 (s, 3H, -CH3), 2.37 (s, 3H, phenyl-CH3), 2.57 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.86 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.18 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.14 (d, J = 7.5 Hz, 1H, phenyl-H), 7.20 (d, J = 8 Hz, 1H, phenyl-H), 7.23 (s, 1H, phenyl-H), 7.27 (t, J = 8 Hz, 1H, phenyl-H). 13C NMR (CDCl3) δ 21.47 (phenyl-CH3), 21.60 (-CH3), 24.39 (-CH3), 44.13 (C-4), 74.47 (C-5), 81.43 (C-2), 123.16 (phenyl-C), 126.66 (phenyl-C), 128.58 (phenyl-C), 129.00 Page 152 Page 152 ISSN 1551-7012 ©ARKAT USA, Inc. ARKIVOC 2009 (xiv) 146-155 General Papers (phenyl-C), 138.44 (phenyl-C), 140.44 (phenyl-C), 217.42 (O=C). MS (EI), m/z (%) 204.1 (M+, 3), 118.1 (100). (phenyl-C), 138.44 (phenyl-C), 140.44 (phenyl-C), 217.42 (O=C). MS (EI), m/z (%) 204.1 (M+, 3), 118.1 (100). 2,2-Dimethyl-5-p-tolyl-dihydrofuran-3(2H)-one (2d). Yellow oil. 1H NMR (CDCl3) δ 1.32 (s, 3H, -CH3), 1.39 (s, 3H, -CH3), 2.36 (s, 3H, phenyl-CH3), 2.56 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.85 (dd, J = 6 Hz; J = 18.5 Hz, 1H, H-4b), 5.18 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.19 (d, J = 8 Hz, 2H, phenyl-H), 7.30 (d, J = 8 Hz, 2H, phenyl-H). 13C NMR (CDCl3) δ 21.18 (phenyl-CH3), 21.62 (-CH3), 24.41 (-CH3), 44.15 (C-4), 74.33 (C-5), 81.39 (C-2), 126.03 (phenyl-C), 129.34 (phenyl-C), 137.54 (phenyl-C), 138.02 (phenyl-C), 217.54 (O=C). MS (EI), m/z (%) 204.1 (M+, 3), 118.1 (100). 2,2-Dimethyl-5-p-tolyl-dihydrofuran-3(2H)-one (2d). Yellow oil. 1H NMR (CDCl3) δ 1.32 (s, 3H, -CH3), 1.39 (s, 3H, -CH3), 2.36 (s, 3H, phenyl-CH3), 2.56 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.85 (dd, J = 6 Hz; J = 18.5 Hz, 1H, H-4b), 5.18 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.19 (d, J = 8 Hz, 2H, phenyl-H), 7.30 (d, J = 8 Hz, 2H, phenyl-H). Experimental Section 13C NMR (CDCl3) δ 21.73 (-CH3), 24.44 (-CH3), 44.11 (C-4), 74.56 (C-5), 81.55 (C-2), 123.68 (naphthyl-C), 125.03 (naphthyl-C), 126.19 (naphthyl-C), 126.38 (naphthyl-C), 127.76 (naphthyl-C), 128.00 (naphthyl-C), 128.68 (naphthyl-C), 133.22 (naphthyl-C), 137.97 (naphthyl-C), 217.18 (O=C). MS (EI), m/z (%) 240.0 (M+, 5), 153.9 (100). HRMS Calcd. for C16H16O2 240.1150, found 240.1163. 5-(Furan-2-yl)-2,2-dimethyl-dihydrofuran-3(2H)-one (2k). Yellow oil. 1H NMR (CDCl3) δ 1.30 (s, 3H, -CH3), 1.33 (s, 3H, -CH3), 2.80 (dd, J = 7 Hz; J = 18 Hz, 1H, H-4a), 2.93 (dd, J = 9 Hz; J = 18 Hz, 1H, H-4b), 5.25 (dd, J = 9 Hz; J = 7 Hz, 1H, -CH), 6.37-6.40 (m, 2H, furyl-H), 7.44 (m, 1H, furyl-H). 13C NMR (CDCl3) δ 22.28 (-CH3), 24.29 (-CH3), 39.71 (C-4), 67.74 (C-5), 80.97 (C-2), 108.50 (furyl-C), 110.39 (furyl-C), 143.10 (furyl-C), 152.37 (furyl-C), 216.35 (O=C). MS (EI), m/z (%) 180.1 (M+, 5), 94.0 (100). HRMS calcd. for C10H12O3 180.0786, found 180.0795. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (NSFC, 20876149), and the Natural Science Foundation of Zhejiang Province (Y406380). Experimental Section 13C NMR (CDCl3) δ 21.59 (-CH3), 24.42 (-CH3), 44.13 (C-4), 55.34 (-OCH3), 74.17 (C-5), 81.39 (C-2), 114.06 (phenyl-C), 127.47 (phenyl-C), 132.49 (phenyl-C), 159.58 (phenyl-C), 217.56 (O=C). MS (EI), m/z (%) 220.0 (M+, 3), 133.8 (100). 2,2-Dimethyl-5-(4-nitrophenyl)-dihydrofuran-3(2H)-one (2i). Yellow oil. 1H NMR (CDCl3) δ 1.35 (s, 3H, -CH3), 1.43 (s, 3H, -CH3), 2.49 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.97 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.33 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.58-7.61 (m, 2H, phenyl-H), 2,2-Dimethyl-5-(4-nitrophenyl)-dihydrofuran-3(2H)-one (2i). Yellow oil. 1H NMR (CDCl3) δ 1.35 (s, 3H, -CH3), 1.43 (s, 3H, -CH3), 2.49 (dd, J = 10.5 Hz; J = 18 Hz, 1H, H-4a), 2.97 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.33 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.58-7.61 (m, 2H, phenyl-H), ISSN 1551-7012 Page 153 ©ARKAT USA, Inc. ARKIVOC 2009 (xiv) 146-155 General Papers 8.24-8.27 (m, 2H, phenyl-H). 13C NMR (CDCl3) δ 21.61 (-CH3), 24.28 (-CH3), 43.78 (C-4), 73.34 (C-5), 81.79 (C-2), 123.96 (phenyl-C), 126.62 (phenyl-C), 147.65 (phenyl-C), 148.16 (phenyl-C), 215.58 (O=C). MS (EI), m/z (%) 235.1 (M+, 1), 149.0 (100). HRMS Calcd. for C12H13NO2 235.0845, found 235.0851. 2,2-Dimethyl-5-(naphthalen-2-yl)-dihydrofuran-3(2H)-one (2j). Yellow solid, mp 63-64 oC. 1H NMR (CDCl3) δ 1.38 (s, 3H, -CH3), 1.46 (s, 3H, -CH3), 2.66 (dd, J = 10.5 Hz; J = 18.5 Hz, 1H, H-4a), 2.95 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.39 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.47-7.52 (m, 3H, phenyl-H), 7.84-8.89 (m, 4H, phenyl-H). 13C NMR (CDCl3) δ 21.73 (-CH3), 24.44 (-CH3), 44.11 (C-4), 74.56 (C-5), 81.55 (C-2), 123.68 (naphthyl-C), 125.03 (naphthyl-C), 126.19 (naphthyl-C), 126.38 (naphthyl-C), 127.76 (naphthyl-C), 128.00 (naphthyl-C), 128.68 (naphthyl-C), 133.22 (naphthyl-C), 137.97 (naphthyl-C), 217.18 (O=C). MS (EI), m/z (%) 240.0 (M+, 5), 153.9 (100). HRMS Calcd. for C16H16O2 240.1150, found 240.1163. 2,2-Dimethyl-5-(naphthalen-2-yl)-dihydrofuran-3(2H)-one (2j). Yellow solid, mp 63-64 oC. 1H NMR (CDCl3) δ 1.38 (s, 3H, -CH3), 1.46 (s, 3H, -CH3), 2.66 (dd, J = 10.5 Hz; J = 18.5 Hz, 1H, H-4a), 2.95 (dd, J = 6 Hz; J = 18 Hz, 1H, H-4b), 5.39 (dd, J = 10.5 Hz; J = 6 Hz, 1H, -CH), 7.47-7.52 (m, 3H, phenyl-H), 7.84-8.89 (m, 4H, phenyl-H). References 1. Larhed, M.; Moberg, C.; Hallberg, A. Acc. Chem. Res. 2002, 35, 717. 2. Caddick, S.; Fitzmaurice, R. Tetrahedron 2009, 65, 3325. 2. Caddick, S.; Fitzmaurice, R. Tetrahedron 2009, 65, 3325. 3. Lidstrom, P.; Tierney. J.; Wathey, B.; Westman, J. Tetrahedron 2001, 45, 9225. 3. Lidstrom, P.; Tierney. J.; Wathey, B.; Westman, J. Tetrahedron 2001, 45, 9225 4. Zare, A.; Hasaninejad, A.; Beyzavi, M. H.; Zare, A. R. M.; Khalafi-Nezhad, A.; Asadi, F. Baramaki, L.; Jomhori-Angali, S.; Ghaleh-Golabi, R. Synth. Commun. 2009, 39,139. 4. Zare, A.; Hasaninejad, A.; Beyzavi, M. H.; Zare, A. R. M.; Khalafi-Nezhad, A.; Asadi, F Baramaki, L.; Jomhori-Angali, S.; Ghaleh-Golabi, R. Synth. Commun. 2009, 39,139. 4. Zare, A.; Hasaninejad, A.; Beyzavi, M. H.; Zare, A. R. M.; Khalafi Nezhad, A.; Asadi Baramaki, L.; Jomhori-Angali, S.; Ghaleh-Golabi, R. Synth. Commun. 2009, 39,139. 5. Khunt, R. C.; Akbari, J. D.; Manvar, A. T.; Tala, S. D.; Dhaduk, M. F.; Joshi, H. S.; Shah, A. Arkivoc 2008, (xi) 277. 5. Khunt, R. C.; Akbari, J. D.; Manvar, A. T.; Tala, S. D.; Dhaduk, M. F.; Joshi, H. S.; Shah, A. Arkivoc 2008, (xi) 277. 6. Yadav, J. S.; Reddy, B. V. S.; Basak, A. K.; Narsaiah, A. V. J. Mol. Catal. A: Chemical 2007, 274, 105. 7. Jung, M. E. Stabilized Nucleophiles with Electron Deficient Alkenes and Alkynes. In Comprehensive Organic Synthesis; B. M. Trost, I. Fleming, Eds.; Pergamon Press: Oxford, 1991; Vol. 4, pp 1–67. Page 154 ISSN 1551-7012 ©ARKAT USA, Inc. Page 154 ARKIVOC 2009 (xiv) 146-155 General Papers 8. Pelmutter, P. Conjugate Addition Reactions in Organic Synthesis; Pergamon Press: Oxford, 1992. 8. Pelmutter, P. Conjugate Addition Reactions in Organic Synthesis; Pergamon Press: Oxford, 1992. 9. Nising, C. F.; Braese, S. Chem. Soc. Rev. 2008, 37, 1218. B.; Ilankumaran, P.; Fetterly, B. M.; Verkade, J. G. J. Org. Chem. 2002, 67, 3555. 10. Kisanga, P. B.; Ilankumaran, P.; Fetterly, B. M.; Verkade, J. G. J. Org. Chem 11. Stewart, I. C.; Bergman, R. G.; Toste, F. D. J. Am. Chem. Soc. 2003, 125, 8696. 12. Murtagh, J. E.; McCooey, S. H.; Connon, S. J. Chem. Commun. 2005, 227. 13. Farnworth, M. V.; Cross, M. J.; Louie, J. Tetrahedron Lett. 2004, 45, 7441. 14. Miller, K. J.; Kitagawa, T. T.; Abu-Omar, M. M. Organometallics 2001, 20, 4 15. Munro-Leighton, C.; Delp, S. A.; Blue, E. D.; Gunnoe, T. B. Organometallic 16. Yang, L.; Xu, L. W.; Xia, C. References G. Tetrahedron Lett. 2005, 46, 3279. 17. Wabnitz, T. C.; Spencer, J. B. Org. Lett. 2003, 5, 2141. 18. Bernal, P.; Tamariz, J. Tetrahedron Lett. 2006, 47, 2905. 19. Prabhakar, P.; Suryakiran, N.; Narasimhulu, M.; Venkateswarlu, Y. J. Mol. Catal. A: Chemical 2007, 274, 72. 20. Ramachary, D. B.; Mondal, R. Tetrahedron Lett. 2006, 47, 7689. 21. Kano, T.; Tanaka, Y.; Maruoka, K. Tetrahedron Lett. 2006, 47, 3039. 22. Karodia, N.; Liu, X. H.; Ludley, P.; Pletsas, D.; Stevenson, G. Tetrahedron 200 23. Merschaert, A.; Delbeke, P.; Daloze, D.; Dive, G. Tetrahedron Lett. 2004, 45, 4697. 24. Biddle, M. M.; Lin, M.; Scheidt, K. A. J. Am. Chem. Soc. 2007, 129, 3830. 25. Dittmer, C.; Raabe, G..; Hintermann, L. Eur. J. Org. Chem. 2007, 5886. 26. Saito N.; Ryoda A.; Nakanishi W.; Kumamoto T.; Ishikawa T. Eur. J. Org. Chem. 2008, 2759. 26. Saito N.; Ryoda A.; Nakanishi W.; Kumamoto T.; Ishikawa T. Eur. J. Org. C 27. Reiter M.; Turner H.; Gouverneur V. Chem. Eur. J. 2006, 12, 7190. 28. Wang, L.; Liu, X.; Dong, Z.; Fu, X.; Feng, X. Angew. Chem. Int. Ed. 2008, 47, 8670. 29. Brennan, C. M.; Hunt, I.; Jarvis, T. C.; Johnson, C. D.; McDonnell, P. D. Can. J. Chem. 1990, 68, 1780. 30. Baldwin, J. E.; Thomas, R. C.; Kruse, L. I.; Silberman, L. J. Org. Chem. 1977, 42, 3846 31. Ellis, G. W. L.; Johnson, C. D.; Rogers, D. N. J. Am. Chem. Soc. 1983, 105, 50 32. Shen, Z. L.; Jiang, X. Z.; Mo, W. M.; Hu, B. X.; Sun, N. Green. Chem. 2005, 7, 97. 33. Ohnemuller, U. K.; Nising, C. F.; Nieger, M.; Brase, S. Eur. J. Org. Chem. 200 34. Ellis, G. W. L.; Johnson, C. D.; Rogers, D. N. J. Chem. Soc., Chem. Commun. 1982, 36. 35. Wabnitz, T. C.; Yu, J. Q.; Spencer, J. B. Chem. Eur. J. 2004, 10, 484. 36. Nazarov, I. N. Bull. Acad. Sci. URSS, Classe Sci. Chim. 1948, 107. Chem. Abstr. 1948, 42, 7737. ISSN 1551-7012 Page 155 ©ARKAT USA, Inc.
https://openalex.org/W4301394990
https://jurnal-dikpora.jogjaprov.go.id/index.php/jurnalideguru/article/download/436/360
Indonesian
null
Meningkatkan Motivasi Belajar dengan Discovery Learning Berbantuan Media Laboratorium Virtual pada Materi Sistem Ekskresi Manusia
Ideguru
2,022
cc-by
4,913
Meningkatkan Motivasi Belajar dengan Discovery Learning Berbantuan Media Laboratorium Virtual pada Materi Sistem Ekskresi Manusia Dwi Essy Sumaryanti SMA Negeri 5 Yogyakarta, Yogyakarta. Daerah Istimewa Yogyakarta, Indonesia buessy63@gmail.com Abstrak: Tujuan dari penelitian ini yaitu menjelaskan pelaksanaan peningkatan motivasi belajar siswa dengan penerapan model Discovery Learning menggunakan media laboratorium virtual materi sistem ekskresi pada kelas XI MIPA 5 SMA N 5 Yogyakarta. Penelitian ini merupakan Penelitian Tindakan Kelas (PTK) yang mencakup 4 prosedur yaitu perencanaan, tindakan, pengamatan, dan refleksi. Data penelitian diambil dengan angket, observasi, catatan lapangan, dan dokumentasi. Teknik analisis data menggunakan analisis deskriptif kualitatif selama 2 siklus. Tahapan dalam kegiatan pembelajaran menggunakan model Discovery Learning mulai dari pemberian rangsangan, pertanyaan/identifikasi masalah, pengolahan data, pembuktian, menarik kesimpulan/ generalisasi. Pembelajaran model Discovery Learning menggunakan media laboratorium virtual pada materi sistem ekskresi dapat meningkatkan motivasi belajar siswa kelas XI MIPA 5 SMA Negeri 5 Yogyakarta. Hasil penelitian menunjukkan motivasi belajar siswa pada siklus I yaitu 73,87%, yang artinya belum memenuhi indikator keberhasilan. Kenaikan persentase motivasi belajar siswa terjadi di siklus II dimana motivasi belajar siswa mencapai 83,39% sehingga dapat dikatakan telah memenuhi indikator keberhasilan. Kata kunci: discovery learning; laboratorium virtual; motivasi; sistem ekskresi Ideguru : Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 ; Vol.8, No.1, Januari 2023 Journal homepage : https://jurnal-dikpora.jogjaprov.go.id/ DOI : https://doi.org/10.51169/ideguru.v8i1.436 Terakreditasi Kemendikbudristek Nomor: 158/E/KPT/2021 (Peringkat 4) Artikel Penelitian – Naskah dikirim: 05/07/2022 – Selesai revisi: 13/09/2022 – Disetujui: 14/09/2022 – Diterbitkan: 04/10/2022 Ideguru : Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 ; Vol.8, No.1, Januari 2023 Journal homepage : https://jurnal-dikpora.jogjaprov.go.id/ DOI : https://doi.org/10.51169/ideguru.v8i1.436 Terakreditasi Kemendikbudristek Nomor: 158/E/KPT/2021 (Peringkat 4) Artikel Penelitian – Naskah dikirim: 05/07/2022 – Selesai revisi: 13/09/2022 – Disetujui: 14/09/2022 – Diterbitkan: 04/10/2022 Increasing Students' Learning Motivation through Discovery Learning Model Assisted by Virtual Laboratory Media on Human Excretory System Material Abstract: The purpose of this study is to explain the implementation of increasing students' learning motivation by applying the Discovery Learning model using virtual laboratory media for excretory system material in class XI MIPA 5 SMA N 5 Yogyakarta. This research is a Classroom Action Research which includes 4 procedures, namely planning, action, observation, and reflection. Research data were taken by questionnaires, observations, field notes, and documentation. The data analysis technique used descriptive qualitative analysis for 2 cycles. The stages in learning activities using the Discovery Learning model start from providing stimulation, questioning/problem identification, data processing, evidence, drawing conclusions/generalizations. Discovery Learning model learning using virtual laboratory media on excretory system material can increase student motivation in class XI MIPA 5 SMA Negeri 5 Yogyakarta. The results showed that students' learning motivation in the first cycle was 73.87%, which means that they have not met the indicators of success. The increase in the percentage of students' learning motivation occurred in cycle II where students' learning motivation reached 83.39% so that it can be said to have met the indicators of success. f Keywords: discovery learning; virtual laboratory; motivation; excretion system masa pandemi Covid-19 kegiatan praktikum biologi umumnya dilaksanakan di laboratorium. Ketika pemerintah memberlakukan kegiatan belajar dari rumah, sekolah menggunakan metode pembelajaran dalam jaringan atau yang dikenal sebagai pembelajaran daring. Menurut Bilfaqih dan Qomarudin (2015) Hak Cipta ©2023 Dewi Essy Sumaryanti Lisensi: CC BY 4.0 internasional 1. Pendahuluan Pembelajaran biologi lebih dari sekedar mempelajari konsep maupun fakta, namun difokuskan pada inkuiri untuk mendorong proses penemuan dalam biologi. Pentingnya laboratorium perlu ditekankan dalam hal proses pembelajaran biologi. Selama ini sebelum pada - 83 - Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 DOI : https://doi.org/10.51169/ideguru.v8i1.436 dibuat menarik dan menggembirakan meskipun pembelajaran dilakukan secara online. untuk memberikan pengajaran dan pembelajaran kepada audiens yang besar dan luas maka dilaksanakan kegiatan pembelajaran dalam jaringan. Sistem pembelajaran ini dilakukan secara online, tanpa tatap muka antara guru dan siswa (Hapsara, 2020). Walaupun peserta didik tidak berada di sekolah, pembelajaran harus tetap berjalan. Guru bertanggung jawab untuk merancang pembelajaran online yang ringan dan efektif. Guru dapat melakukannya dengan menggunakan perangkat atau media online dan disesuaikan dengan karakteristik materi. untuk memberikan pengajaran dan pembelajaran kepada audiens yang besar dan luas maka dilaksanakan kegiatan pembelajaran dalam jaringan. Sistem pembelajaran ini dilakukan secara online, tanpa tatap muka antara guru dan siswa (Hapsara, 2020). Walaupun peserta didik tidak berada di sekolah, pembelajaran harus tetap berjalan. Guru bertanggung jawab untuk merancang pembelajaran online yang ringan dan efektif. Guru dapat melakukannya dengan menggunakan perangkat atau media online dan disesuaikan dengan karakteristik materi. Uno (2011:23) menjelaskan definisi motivasi belajar yaitu dorongan baik dari dalam maupun luar diri siswa bertujuan untuk mengubah tingkah laku dalam proses belajar ditandai dengan sejumlah indikator yang menunjang. Sardiman (2007:75) menjelaskan definisi motivasi belajar sebagai setiap kekuatan pendorong dalam diri siswa yang memberi arah pada kegiatan belajar untuk mencapai tujuan yang diinginkan oleh subyek tersebut. ciri-ciri yang tampak pada diri seseorang dapat memperlihatkan apakah orang tersebut termotivasi atau tidak. Sardiman (2007: 83) menjelaskan ciri-ciri orang yang mempunyai motivasi antara lain tekun menjalankan tugas, berminat pada berbagai masalah, lebih menyukai bekerja sendiri, mudah bosan mengerjakan pekerjaan rutin, mampu menjaga pendapatnya, sulit melepaskan keyakinannya, gemar mengejar dan menyelesaikan masalah. Orang dengan motivasi yang cukup kuat akan mempunyai ciri- ciri tersebut. Dari pendapat ahli tersebut, peneliti menyimpulkan bahwa motivasi belajar dapat dilihat dari beberapa aspek yaitu (1) tekun menghadapi tugas, (2) ulet menghadapi kesulitan, (3) memperlihatkan minat belajar serta (4) gemar mengikuti pembelajaran. Mengajar Biologi haruslah bisa menghadirkan gambaran peristiwa alam, sebagai konsekuensinya guru perlu menampilkan objek untuk membantu siswa belajar. Sebagai pembelajaran sains sudah barang tentu tak terlepas dari kegiatan praktikum. Melalui praktikum siswa dapat menemukan pengetahuan melalui penelitian. 1. Pendahuluan Dalam kondisi Pandemi Covid-19 tentu praktikum biologi mengalami kendala. Hal ini menuntut guru untuk mendesain pembelajaran virtual, termasuk merancang mendesain praktikum secara virtual atau yang dikenal dengan istilah laboratorium virtual. Guru sebagai fasilitator dan motivator selain dituntut merancang pembelajaran yang asyik dan tidak membosankan, juga dituntut untuk selalu menumbuhkan motivasi siswa. Belajar menjadi asyik merupakan tantangan bagi seorang guru, terlebih pada masa pandemi seperti saat ini. kerjasama guru dan siswa diperlukan agar pembelajaran berlangsung lebih asyik. Dengan adanya kegiatan belajar mengajar secara online siswa diharapkan mampu mengikuti pembelajaran. Pada kenyataanya sebagian besar siswa tidak mengikuti pembelajaran karena mulai timbulnya rasa jenuh dan bosan. Widayanti (2021) menyatakan bahwa media pembelajaran sains yang digunakan secara daring selama pandemi adalah e-modul, Google Classroom, sensor smartphone, dan laboratorium virtual. Penerapan media pembelajaran perlu disesuaikan dengan indikator pencapaian dan jenjang pendidikan. Temuan penelitian ini dapat meningkatkan hasil belajar, motivasi belajar, berpikir kritis, pemecahan masalah, penguasaan konsep, dan keterampilan proses sains. Muawanah (2021) menyatakan bahwa Peningkatan motivasi siswa pada masa pandemi dapat dilakukan melalui strategi dengan beragam cara, salah satunya dapat menggunakan Laboratorium Virtual. E Stark (2017) menyatakan bahwa laboratorium Virtual adalah program komputer dimana siswa dapat melakukan praktikum dengan bantuan komputer. Menurut Fonna (2013) Penggunaan laboratorium virtual pada materi sistem pernapasan manusia menunjukkan perbedaan signifikan ditinjau dari kemampuan berpikir kritis siswa. Menurut Nainggolan (2014) media pendukung praktikum seperti Laboratorium Virtual mampu mengembangkan keaktifan, belajar menjadi lebih menyenangkan serta tidak mudah bosan. Materi dalam media ini disajikan sedemikian rupa sehingga siswa lebih mudah Guru berperan penting dalam mendesain proses Pembelajaran Jarak jauh (PJJ) dengan cara mengubah metode dan model pembelajaran supaya kondisi pembelajaran lebih menyenangkan dan peserta didik tidak bosan. Guru perlu mengendalikan emosi siswa selama pembelajaran agar siswa merasa nyaman di kelas PJJ. Atmosfer belajar perlu dibuat lebih aktif dan bersemangat. Salah satu alternatif membuat pembelajaran menjadi asyik dan menyenangkan dengan mempergunakan media laboratorium virtual yang dapat dikerjakan siswa secara berkelompok maupun secara individual. Upaya ini dilakukan agar motivasi siswa dapat meningkat sehingga tidak beranggapan bahwa belajar itu membosankan. Suasana kelas perlu - 84 - Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 siswa pada materi sistem ekskresi serta apakah pembelajaran model Discovery Learning berbantuan laboratorium virtual ini dapat meningkatkan motivasi belajar siswa pada materi sistem ekskresi. 1. Pendahuluan Selain itu sebagai bahan rujukan dan sumber untuk penelitian berikutnya. Bagi siswa, penelitian ini bermanfaat dalam peningkatan motivasi belajar biologi, dan bagi guru dapat memberikan pengalaman dalam mengimplentasikan pembelajaran saintifik berbantuan laboratorium virtual. 1. Pendahuluan Adapun tujuan penelitian adalah meningkatkan motivasi belajar dengan penerapan Discovery Learning berbantuan media laboratorium virtual serta menjelaskan pelaksanaan peningkatan motivasi belajar dengan penerapan Discovery Learning berbantuan media laboratorium virtual pada materi sistem ekskresi. paham sehingga dapat digunakan untuk belajar mandiri. Kesulitan belajar yang dirasakan baik oleh siswa maupun guru diharapkan dapat diselesaikan dengan pemanfaatan laboratorium virtual (Hendrajanti, 2022). Manfaat penggunaan laboratorium virtual antara lain kegiatan belajar menjadi lebih mengasyikkan, interaktif, efisien waktu, meningkatkan kualitas pembelajaran dan kegiatan belajar bisa terlaksana tanpa terikat tempat dan waktu (Baruno, 2021). paham sehingga dapat digunakan untuk belajar mandiri. Kesulitan belajar yang dirasakan baik oleh siswa maupun guru diharapkan dapat diselesaikan dengan pemanfaatan laboratorium virtual (Hendrajanti, 2022). Manfaat penggunaan laboratorium virtual antara lain kegiatan belajar menjadi lebih mengasyikkan, interaktif, efisien waktu, meningkatkan kualitas pembelajaran dan kegiatan belajar bisa terlaksana tanpa terikat tempat dan waktu (Baruno, 2021). ( , ) Salah satu usaha peningkatan motivasi belajar dapat dilakukan dengan memilih model pembelajaran yang tepat. Direktorat Pembinaan Sekolah Menengah Atas (2017) menyatakan bahwa model pembelajaran merupakan konsep yang menggambarkan tata cara sistematis dalam menyusun pengalaman belajar agar tujuan belajar dapat dicapai. Discovery Learning dianggap sebagai model pembelajaran yang sesuai diterapkan bersama Laboratorium Virtual agar terjadi peningkatan motivasi belajar. Hosnan (2014:282) menyatakan Discovery Learning sebagai model dengan kegiatan menemukan dan menyelidiki secara mandiri. Tahapan model Discovery Learning menumbuhkan cara belajar aktif. Dengan demikian diharapkan siswa mengingat pengetahuan tersebut dalam waktu yang lebih lama. Discovery Learning mengajak siswa belajar menganalisis dan mencari solusi terhadap masalah yang dihadapi secara mandiri. Veerman (2003) menjelaskan bahwa Discovery Learning terdiri dari beberapa tahapan yakni: 1) Orientasi, 2) Menyusun Hipotesis, 3) Menguji Hipotesis, 4) Kesimpulan, 5) Regulasi. Model ini dipilih dalam tindakan karena siswa siswa diajak untuk menyusun pemahaman tentang suatu pengetahuan dengan mandiri. Oleh karena itu penulis menentukan suatu tindakan yang dianggap sesuai dalam upaya peningkatan motivasi belajar pada materi sistem ekskresi dengan model Discovery Learning. Penelitian mengenai Laboratorium Virtual yang sudah ada sebelumnya, menjelaskan adanya peningkatan berpikir kritis dan keaktifan siswa melalui penggunaan laboratorium virtual dalam pembelajaran. Sejauh ini belum ada penelitian tindakan kelas yang mengukur motivasi belajar siswa dengan penerapan Discovery Learning berbantuan laboratorium virtual pada materi sistem ekskresi manusia di SMA N 5 Yogyakarta. l k l h l Manfaat praktis yang dapat diperoleh yaitu menyumbang kontribusi dalam perkembangan ilmu pengetahuan mengenai implementasi Discovery Learning berbantuan laboratorium virtual untuk meningkatkan motivasi belajar. DOI : https://doi.org/10.51169/ideguru.v8i1.436 dari awal hingga pembelajaran berakhir. Wawancara digunakan untuk mendapatkan data yang dijelaskan secara lisan mengenai kesan dan pesan siswa setelah mengikuti kegiatan pembelajaran biologi dengan Discovery Learning menggunakan media laboratorium virtual. Dokumentasi berfungsi melengkapi data dengan merekam peristiwa penting atau mengambil foto yang menggambarkan suasana konkret yang terjadi selama kegiatan pembelajaran. penting yakni perencanaan, pelaksanaan, pengamatan, serta refleksi. Setiap tahapan dilakukan dalam dua siklus yakni siklus I dilanjutkan siklus II. Tahapan penelitian disusun sesuai bagan pada Gambar 1. Gambar 1. Tahapan Penelitian Tindakan Kelas j g p j Analisis data dalam penelitian ini menggunakan teknik analisis deskriptif kualitatif. Peneliti menjelaskan perkembangan kegiatan pembelajaran secara deskriptif. Tahapan teknik analisis data yaitu melakukan reduksi data, menyajikan data dan menarik kesimpulan. Peneliti memilih, menentukan fokus perhatian serta memjadikan data yang diperoleh menjadi lebih sederhana melalui proses reduksi data. Reduksi data ini dilaksanakan dengan menyeleksi data mentah yang berasal dari catatan lapangan agar didapatkan informasi yang lebih bermakna. Setelah data direduksi, langkah selanjutnya adalah penyajian data. Penyajian data dilakukan sebagai usaha untuk menunjukkan data dengan jelas sehingga pembaca dapat lebih mudah paham. Data disajikan dalam berbagai bentuk seperti uraian naratif, tabel, diagram, maupun dalam wujud lainnya. Dengan dilaksanakannya penyajian data diharapkan dapat memberikan gambaran lebih nyata mengenai bagaimana tindakan telah dilakukan serta hasil yang didapat. Langkah terakhir adalah Penarikan kesimpulan dimana ditentukan intisari dari data yang disajikan kemudian disusun dalam suatu kalimat yang mengandung makna yang mudah dipahami. Gambar 1. Tahapan Penelitian Tindakan Kelas Siklus II merupakan serangkaian tahap untuk menyempurnakan kelemahan yang terjadi pada siklus I. Pelaksanaan siklus II diharapkan dapat lebih baik dari siklus I. p Sumber data didapatkan dari siswa serta kolaborator. Pengumpulan data dilakukan dengan beragam teknik yang meliputi kuesioner, angket, pengamatan atau observasi, catatan lapangan, wawancara serta dokumentasi. Data yang akan dicari menggunakan angket adalah data tentang motivasi belajar setelah penerapan model Discovery Learning menggunakan media laboratorium virtual. Angket ini diberikan pada siswa yang menjadi subyek penelitian untuk diisi setelah mereka selesai mengikuti kegiatan pembelajaran. Kegiatan pengamatan atau observasi dilaksanakan ketika proses pembelajaran berlangsung dari awal sampai akhir. Pengamatan dilakukan untuk mengetahui pelaksanaan proses pembelajaran biologi pada materi sistem ekskresi dengan model pembelajaran Discovery Learning menggunakan media laboratorium virtual. Lembar observasi yang telah disiapkan kemudian digunakan dalam proses pengamatan pembelajaran. Catatan lapangan berfungsi menjelaskan suasana dan keadaan ketika pembelajaran berlangsung mulai Indikator keberhasilan pada penelitian tindakan kelas ini, diidentifikasi dari ada atau tidaknya perubahan menuju perbaikan. DOI : https://doi.org/10.51169/ideguru.v8i1.436 Terkait dengan motivasi belajar Biologi, indikator keberhasilan penelitian adalah apabila capaian motivasi belajar siswa rata-rata mencapai 80. 2. Metode Penelitian Penelitian ini merupakan penelitian tindakan kelas (PTK) yang dianalisis secara kualitatif. Penelitian dilakukan selama 2 siklus untuk melihat peningkatan motivasi yang terjadi dari tindakan yang dilakukan. Penelitian bertempat di SMA Negeri 5 Yogyakarta dengan pelaksanaan penelitian selama 5 bulan mulai dari Januari hingga Mei tahun 2021. Subyek penelitian ini adalah semua siswa kelas XI MIPA 5 SMA N 5 Yogyakarta tahun pelajaran 2020/2021 dengan jumlah 36 siswa. Dalam penelitian ini, peneliti didampingi oleh teman sejawat selaku kolaborator dan Observer yang mendampingi selama pelaksanaan siklus. p g p Rencana tindakan yang dilakukan dalam penelitian ini adalah penerapan Discovery Learning sebagai usaha untuk meningkatkan motivasi belajar siswa dalam mata pelajaran Biologi berbantuan media laboratorium virtual di kelas XI MIPA 5 SMA N 5 Yogyakarta tahun pelajaran 2020/2021. Pada kondisi sebelum pandemi, praktikum yang dilakukan di SMA N 5 Yogyakarta dilakukan langsung di Laboratorium. Selama masa pandemi praktikum tidak dapat dilaksanakan karena pembelajaran secara jarak jauh. Penggunaan Laboratorium Virtual dapat menjadi solusi untuk melaksanakan praktikum jarak jauh. Peneliti merumuskan masalah penelitian berdasarkan latar belakang yang disampaikan yaitu bagaimana pelaksanaan pembelajaran Discovery Learning berbantuan laboratorium virtual dapat meningkatkan motivasi belajar Prosedur dalam penelitian mengikuti model PTK oleh Arikunto (2010: 17) yang menjelaskan bahwa setiap siklus PTK mencakup empat tahap - 85 - Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 DOI : https://doi.org/10.51169/ideguru.v8i1.436 belum dirancang agar siswa termotivasi dalam belajar. Banyak siswa yang terlihat pasif, siswa kurang tekun dalam pembelajaran dikarenakan kurang tertarik dengan materi. Siswa juga terlihat kurang ulet dalam menghadapi kesulitan ditandai dengan tidak menyegerakan melakukan praktikum sendiri dirumah serta tidak menyegerakan mengerjakan tugas di Google Classroom padahal tenggat waktu sudah berlalu. Kurangnya minat belajar siswa dalam hal ini terlihat dari belum mempelajari materi serta belum tampak senang selama mengikuti pembelajaran Biologi. Hal ini berdasarkan dari jumlah siswa yang nilai biologinya dibawah KKM masih terhitung cukup banyak. Dari total 36 siswa di kelas XI MIPA 5, hanya 25 siswa yang cukup termotivasi dalam melakukan aktivitas pembelajaran. Dengan demikian persentase siswa yang memiliki motivasi sebesar 69,44%, sedangkan sebanyak 30,56% siswa belum termotivasi dalam belajar. Satu diantara penyebab dari rendahnya motivasi siswa adalah skenario pembelajaran belum optimal untuk membangkitkan dan meningkatkan motivasi belajar seluruh siswa. Materi sistem ekskresi dipilih karena kebanyakan siswa merasa kesulitan mengingat konsep yang ada dalam materi tersebut. Discovery Learning menekankan kegiatan belajar pada proses mengembangkan pengetahuan dan keterampilan siswa melalui proses eksperimen yang terstruktur serta menghasilkan kesimpulan sendiri. Komunikasi guru dengan siswa dilakukan melalui Whatsapp Group. Siswa secara berkelompok dan mandiri berdiskusi dan membuat laporan hasil praktikum. Siswa menggunakan buku paket biologi yang ada maupun sumber terpercaya lain sebagai referensi dalam kegiatan diskusi dan penyusunan laporan. Siswa yang sudah menyelesaikan laporan praktikum melalui perwakilan kelompoknya mengirim ke guru melalui Whatsapp personal chat, kemudian guru memberikan umpan balik. Apabila terdapat kelompok yang menemui kendala dalam eksperimen maupun analisis data maka guru akan melakukan pembimbingan untuk membantu kelompok tersebut. Sebelum pembelajaran berakhir, guru menyampaikan refleksi pembelajaran. Berdasarkan observasi pada siklus I didapatkan persentase keterlaksanaan pembelajaran yaitu 83,33%. Proses pembelajaran pada siklus I secara umum berlangsung dengan baik namun belum semua langkah model pembelajaran Discovery Learning menggunakan Laboratorium Virtual dapat dilaksanakan. Kegiatan guru yang belum dilaksanakan antara lain pada kegiatan pendahuluan belum menyampaikan Kompetensi Dasar, apersepsi dan motivasi pembelajaran serta belum menyampaikan penjelasan tentang model pembelajaran yang hendak dipakai kepada siswa. j Solusi untuk menyelesaikan permasalahan dalam pembelajaran biologi tersebut maka peneliti ingin mengimplementasikan pembelajaran saintifik model Discovery learning menggunakan metode praktikum dengan Laboratorium Virtual. Hal ini dipilih berdasarkan keunggulan metode ini yang dianggap mampu dalam peningkatan motivasi belajar. Penelitian ini dilaksanakan dalam dua siklus yaitu siklus I yang dilanjutkan dengan II. 3. Hasil dan Pembahasan Pembelajaran yang dilakukan selama pandemi di kelas XI MIPA 5 membuat motivasi belajar mereka menjadi turun. Peneliti telah melakulan observasi awal dan menunjukkan bahwa selama pembelajaran berlangsung melalui media Whatsapp Group beberapa siswa ada yang tidak menjawab saat dipanggil, tidak memperhatikan guru, dan lama dalam menjawab pertanyaan. Pada saat guru memberi penugasan beberapa siswa tidak tepat waktu dalam mengumpulkan tugas. Dari hasil observasi awal tersebut diketahui bahwa pembelajaran biologi - 86 - Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 76,56%. Indikator keberhasilan penelitian yaitu apabila motivasi belajar siswa mencapai 80%. Berdasarkan data yang diperoleh pada siklus I, seluruh aspek motivasi belajar belum ada yang mencapai indikator keberhasilan. Sebagai upaya untuk mencapai indikator keberhasilan, dibutuhkan tindakan pada siklus II sebagai perbaikan dari siklus I. pada menu info chat di WhatsApp sehingga diketahui siswa yang tidak menyimak. Guru kemudian akan menyapa siswa yang tidak menyimak melalui personal chat. Hal ini dilakukan untuk memberi motivasi supaya seluruh siswa menyimak apa yang dijelaskan guru melalui Whatsapp Group. Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Siklus I Pada siklus I Guru memberikan petunjuk praktikum Rate respiration eksperimen. Siswa dipersilakan membuka aplikasi berbasis Web secara mandiri. Tampilan virtual laboratorium yang digunakan pada siklus I tampak pada Gambar 2. Pada akhir pertemuan ke dua siklus I, guru membagikan angket motivasi belajar model pembelajaran Discovery Learning menggunakan Laboratorium Virtual yang kemudian diisi oleh siswa. Siswa mengisi angket secara individu sesuai dengan apa yang dialami oleh setiap siswa. Pencapaian motivasi belajar ditinjau dari beberapa aspek. Diagram Motivasi Belajar Siswa untuk setiap Aspek pada Siklus I dapat dilihat pada Gambar 3. Gambar 2. Virtual Laboratorium untuk eksperimen Rate Respiration Motivasi belajar siswa ditinjau dari sejumlah aspek. Aspek tekun menghadapi tugas menunjukkan persentase 72.22%, ulet menghadapi kesulitan 70,83%, memperlihatkan minat belajar mencapai 76,56% serta gemar mengikuti pembelajaran mencapai 75,87%. Semua aspek sudah termasuk dalam kategori Tinggi. Ditinjau dari aspek motivasi belajar, maka aspek menunjukkan minat belajar memperoleh persentase tertinggi yaitu sebesar Gambar 2. Virtual Laboratorium untuk eksperimen Rate Respiration - 87 - Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 Hasil pengamatan observer mengacu pada instrumen pengamatan keterlaksanaan pembelajaran serta catatan lapangan. Pengamatan yang dilakukan oleh observer dilaksanakan pada kegiatan pembelajaran guru yang menggunakan Discovery Learning berbantuan Virtual Laboratorium. Data yang diperoleh menunjukan bahwa guru dan siswa telah melaksanakan pembelajaran sesuai tahapan Discovery learning yang terdapat dalam Rancangan Pelaksanaan Pembelajaran (RPP). Peneliti dibantu oleh satu orang observer dalam mengamati proses pembelajaran. hasil pengamatan keterlaksanaan kegiatan pembelajaran tersebut menunjukkan bahwa kegiatan belajar mengajar berjalan dengan baik yaitu 100% tahapan pembelajaran dengan Discovery Learning berbantuan Laboratorium Virtual sudah terlaksana. seluruh aspek. Semua aspek melebihi persentase pada indikator keberhasilan yaitu sebesar 80% dengan kategori sangat tinggi. Aspek motivasi belajar yang mendapatkan skor tertinggi yaitu pada aspek tekun menghadapi tugas yang mencapai 89,58%. p Refleksi pembelajaran siklus II dilakukan dengan cara diskusi menurut data yang berasal dari instrumen observasi serta catatan lapangan. Berdasarkan observasi keterlaksanaan pembelajaran dapat disimpulkan bahwa 100% tahapan kegiatan pembelajaran model Discovery Learning menggunakan media Virtual Laboratory sudah dilakukan oleh guru. Catatan lapangan menunjukkan mayoritas siswa lebih semangat dalam pembelajaran. Hal ini terbukti dari keterlibatan siswa yang lebih aktif dalam pembelajaran biologi. Hal ini dapat terlihat dari siswa tekun, ulet, minat belajar dan senang mengikuti pelajaran. Diagram perbandingan aspek motivasi belajar siklus I dan siklus II tampak pada Gambar 6. Pada akhir pertemuan ke dua pada siklus II, guru membagikan angket motivasi belajar model Discovery Learning menggunakan Laboratorium Virtual yang kemudian diisi oleh siswa. Siswa mengisi angket secara individu sesuai dengan yang dialami masing-masing siswa. Hasil angket menunjukkan skor motivasi belajar siklus II untuk setiap aspek. Diagram aspek motivasi belajar siklus II tampak pada Gambar 5. Gambar 6. Diagram Aspek Motivasi Belajar Siklus I dan II 72,22 70,83 76,56 75,87 89,58 81,48 81,60 80,90 50 60 70 80 90 100Tekun Menghadapi TugasUlet Menghadapi KesulitanMemperlihatkan minat belajarGemar mengikuti pembelajaran Skor Diagram Aspek Motivasi Belajar Siklus I dan II Siklus I Siklus II Diagram Aspek Motivasi Belajar Siklus I dan II Gambar 5. Diagram Aspek Motivasi Belajar Siklus II 89,58 81,48 81,60 80,90 50,00 60,00 70,00 80,00 90,00 100,00Tekun Menghadapi TugasUlet Menghadapi KesulitanMemperlihatkan minat belajarGemar mengikuti pembelajaran Skor Diagram Aspek Motivasi Belajar Siklus II Gambar 6. Diagram Aspek Motivasi Belajar Siklus I dan II Siklus I Siklus II Gambar 6. Diagram Aspek Motivasi Belajar Siklus I dan II Gambar 5. Siklus II Guru akan melakukan tangkapan layar - 88 - Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Siklus II Gambar 3. Diagram Aspek Motivasi Belajar Siklus I 72,22 70,83 76,56 75,87 50 60 70 80 90 100Tekun Menghadapi TugasUlet Menghadapi KesulitanMemperlihatkan minat belajarGemar mengikuti pembelajaran Skor Desain pembelajaran pada Siklus II diterapkan pada Kompetensi Dasar Menganalisis hubungan antara struktur jaringan penyusun organ pada sistem ekskresi dalam kaitannya dengan bioproses dan gangguan fungsi yang dapat terjadi. Pembelajaran dilakukan pada materi Sistem ekskresi yang diimplementasi dengan pembelajaran saintifik model Discovery learning dengan menggunakan media laboratorium virtual. Guru memberikan kesempatan peserta didik untuk menemukan pengetahuanterkait materi yang dipelajari melalui eksperimen tes urin secara individual. Laboratorium Virtual yang digunakan adalah virtual laboratory biochemical experiment urine test. Tampilan Laboratorium Virtual yang digunakan pada siklus II dapat dilihat pada Gambar 4. Gambar 3. Diagram Aspek Motivasi Belajar Siklus I Dari data catatan lapangan pada siklus I diketahui bahwa belum semua peserta didik pada pelaksanaan pembelajaran segera bergabung dalam kelas daring. Guru sedikit kesusahan untuk mengelola kelas karena masih ada beberapa siswa terlambat dalam menyapa guru dan menangkap penjelasan guru. Media WhatsApp juga memberi keterbatasan guru dalam menjelaskan hal yang belum dipahami oleh beberapa siswa. Beberapa siswa tidak menyimak penjelasan guru terbukti ketika beberapa siswa yang disapa membutuhkan waktu cukup lama dalam merespon bahkan beberapa siswa tidak menjawab. Untuk itu perlu dilakukan pemantauan dan komunikasi yang lebih baik dengan siswa. Pada Siklus I beberapa tahapan pembelajaran belum terlaksana, Oleh karena itu dibutuhkan perbaikan pada siklus II. Gambar 4. Tampilan Virtual Laboratorium untuk eksperimen Urine Test Gambar 4. Tampilan Virtual Laboratorium untuk eksperimen Urine Test Setelah mengumpulkan informasi yang didapat, siswa secara individual mengerjakan LKPD di google classroom (GC). LKPD ini berisi hasil tes urin berdasarkan percobaan virtual dan memperkirakan penyebab serta kerusakan organ pada gangguan yang terjadi. mengenai hasil tes urin, prakiraan penyebab serta kerusakan organ pada gangguan yang terjadi. Siswa menyampaikan hasil uji urin untuk didiskusikan bersama guru, sehingga menghasilkan satu kesimpulan ciri urin orang normal dan sehat, Uji glukosa membuktikan adanya Diabetes Melitus sedangkan uji protein membuktikan adanya albuminuria. Setelah melakukan refleksi, peneliti menyusun rencana perbaikan tindakan yang akan dilakukan di siklus II. Beberapa tindakan perbaikan yang perlu dilaksanakan antara lain menyampaikan kompetensi dasar, tujuan, apersepsi, motivasi, model dan metode pembelajaran yang dipakai. Guru akan berusaha semaksimal mungkin memotivasi siswa dan menyediakan ruang untuk siswa bertanya serta selalu mengingatkan siswa jangan takut untuk bertanya. Daftar Pustaka Arikunto, S. (2010). Penelitian Tindakan Kelas. Jakarta: PT Bumi Aksara Baruno, A. (2021). Peningkatan Kemampuan Berpikir Analisis pada Materi Genetik Melalui Model Pembelajaran Guided Inquiry Terintegrasi Virtual Lab. Ideguru: Jurnal Karya Ilmiah Guru, 6(2), 176-182. https://doi.org/10.51169/ideguru.v6i2.202 Bilfaqih, Y. & Qomarudin, M.N. (2015). Esensi Penyusunan Materi Daring Untuk Pendidikan Dan Pelatihan. Yogyakarta: Deep Publish. DOI : https://doi.org/10.51169/ideguru.v8i1.436 Gambar 7. Diagram Hasil Motivasi Belajar Siklus I dan II 73,87 83,93 50 60 70 80 90 100 Siklus I Siklus II Diagram Hasil Motivasi Belajar 73,87% belum memenuhi indikator keberhasilan. Pada siklus II terjadi peningkatan motivasi belajar siswa menjadi kategori sangat tinggi yaitu sebesar 83,39%. Dengan demikian dapat dikatakan penelitian tindakan kelas ini telah mencapai keberhasilan. Laboratorium Virtual merupakan perangkat lunak komputer berupa multimedia yang berisi fitur interaktif sehingga siswa bisa memperoleh pengalaman praktikum secara virtual. Penggunaan laboratorium virtual ini ternyata dapat melatih kemampuan berpikir kritis, motivasi belajar siswa serta dapat pula digunakan sebagai media pendukung alternatif untuk menyertakan partispasi aktif siswa dalam kegiatan praktikum secara virtual. Diagram tersebut menunjukkan bahwa pada siklus I, motivasi belajar siswa sebesar 73,87% dengan kategori tinggi. Pada siklus II terjadi peningkatan motivasi belajar menjadi 83,39 dengan kategori sangat tinggi. Peningkatan motivasi belajar dengan penggunaan laboratorium sesuai dengan hasil penelitian Nainggolan (2014) yang menjelaskan bahwa Laboratorium Virtual dapat meningkatkan partisipasi siswa secara aktif. Kegiatan belajar terasa lebih mengasyikkan dan tidak mudah merasa jenuh. materi disajikan sedemikian rupa sehingga siswa dengan mudah dapat paham. Selain itu Penggunaan Laboratorium Virtual dapat mengkondisikan siswa untuk belajar mandiri. Hal ini dapat dilihat dari beberapa aspek motivasi dalam penelitian ini, siswa menjadi lebih senang dan menunjukkan minat dalam pembelajaran, serta tekun dan ulet dalam menyelesaikan tugas secara mandiri. Model pembelajaran Discovery Learning berbantuan Laboratorium Virtual bisa menjadi alternatif media pembelajaran siswa dimasa pandemi yang mengharuskan siswa belajar dari rumah (Learning from home). Penelitian lebih lanjut mengenai penggunaan Laboratorium Virtual pada kompetensi dasar yang lain perlu dilakukan. 4. Simpulan dan Saran Kesimpulan yang dapat diambil dari penelitian ini yaitu Pelaksanaan pembelajaran dengan model Discovery Learning menggunakan Laboratorium Virtual mampu meningkatkan motivasi belajar siswa pada materi sistem Ekskresi di kelas XI MIPA 5 SMA Negeri 5 Yogyakarta setelah melaksanakan semua tahapan pembelajaran meliputi dari pemberian rangsangan, pertanyaan atau identifikasi masalah, pengolahan data, pembuktian, menarik kesimpulan atau generalisasi. Pembelajaran dengan model Discovery Learning menggunakan Laboratorium Virtual mampu meningkatkan motivasi belajar siswa pada materi sistem ekskresi siswa kelas XI MIPA 5 SMA Negeri 5 Yogyakarta. Pada siklus I diketahui bahwa Motivasi belajar siswa termasuk dalam kategori tinggi yaitu mencapai 73,87%. Berdasarkan indikator keberhasilan maka pencapaian motivasi belajar siswa pada siklus I sebesar Direktorat pembinaan sekolah Menengah Atas. (2017). Model-model Pembelajaran. Jakarta: Kementerian Pendidikan dan Kebudayaan, Direktorat Jenderal Pendidikan Dasar dan Menengah. E. Stark, P. Bistak, S. Kozak and E. Kucera. (2017). Virtual laboratory based on Node.js technology. 2017 21st International Conference on Process Control (PC), 386-391. Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Diagram Aspek Motivasi Belajar Siklus II Motivasi belajar pada siklus II dapat dilihat dari beberapa aspek. Pada aspek tekun menghadapi tugas persentase mencapai 89.58%, ulet menghadapi kesulitan 81.48%, memperlihatkan minat belajar mencapai 81.60% serta senang mengikuti pelajaran mencapai 80.90%. Berdasarkan hasil tersebut motivasi belajar siswa dalam pembelajaran melalui model Discovery Learning berbantuan Laboratorium Virtual pada materi sistem ekskresi manusia telah mencapai indikator keberhasilan untuk Berdasarkan diagram tersebut pencapaian motivasi siswa dalam pembelajaran model Discovery Learning menggunakan media laboratorium virtual pada siklus satu semua indikator mengalami kenaikan pada siklus dua. Peningkatan pencapaian motivasi belajar tertinggi dari siklus I ke siklus II terdapat pada aspek tekun menghadapi tugas. Diagram hasil motivasi belajar siklus I dan II dapat dilihat pada Gambar 7. - 89 - Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Vol.8, No.1, Januari 2023 Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 https://doi.org/10.1109/PC.2017.7976245 Fonna, T.M., dkk. (2013). Perbedaan Keterampilan Berpikir Kritis Siswa Melalui Penerapan Media Pembelajaran Laboratorium Virtual Pada Konsep Sistem Pernapasan Manusia Di SMA Negeri Unggul Sigli. Biotik: Jurnal Ilmiah Biologi Teknologi dan Kependidikan, 1(2), 67- 136. h //d d i /10 22373/bi ik 1i2 223 Hapsara, A. S. (2020). Peningkatan Partisipasi dan Hasil Belajar Daring Sosiologi melalui - 90 - Vol.8, No.1, Januari 2023 DOI : https://doi.org/10.51169/ideguru.v8i1.436 Vol.8, No.1, Januari 2023 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Pendekatan Problem Posing Berbasis Infografis. Ideguru: Jurnal Karya Ilmiah Guru, 5(2), 9–19. https://doi.org/10.51169/ideguru.v5i2.17 0 Ideguru: Jurnal Karya Ilmiah Guru p-ISSN 2527-5712 ; e-ISSN 2722-2195 Pendekatan Problem Posing Berbasis Infografis. Ideguru: Jurnal Karya Ilmiah Guru, 5(2), 9–19. https://doi.org/10.51169/ideguru.v5i2.17 0 Nainggolan, L. (2014). Pengembangan Media Praktikum Berbasis Laboratorium Virtual (Virtual Laboratory) pada Materi Pembelahan Sel di SMA. Jambi: FKIP Universitas Jambi. Sardiman, A. M. (2007). Interaksi dan Motivasi Belajar Mengajar. Jakarta: PT. Raja Grafindo Persada. Hendrajanti, P. (2022). Discovery Learning Berbantuan Virtual Chemistry Laboratory untuk Meningkatkan Hasil Belajar Siswa. Ideguru: Jurnal Karya Ilmiah Guru, 7(2), 188- 196. https://doi.org/10.51169/ideguru.v7i2.353 Uno, H. (2011). Teori Motivasi dan Pengukurannya. Jakarta: Bumi Aksara. Veerman, K. (2003). Intelligent Support for Discovery Learning. Twente: Twente University Press. Hosnan, M. (2014). Pendekatan Saintifik dan Kontekstual dalam Pembelajaran Abad 21. Bogor: Ghalia Indonesia. Widayanti, W. (2021). Online learning media on science learning during the COVID-19 Pandemic: A literature study in indonesia. Online Learning in Educational Research, 1(1), 55-61. Muawanah, E.I & Muhid, A. (2021). Strategi Meningkatkan Motivasi Belajar Siswa Selama Pandemi Covid–19: Literature Review. JIBK Undiksha, 12(1), 90-98. h d i jjbk i - 91 -
https://openalex.org/W2402420450
https://europepmc.org/articles/pmc4877522?pdf=render
English
null
A Complex Story: Universal Preference vs. Individual Differences Shaping Aesthetic Response to Fractals Patterns
Frontiers in human neuroscience
2,016
cc-by
10,127
ORIGINAL RESEARCH published: 24 May 2016 doi: 10.3389/fnhum.2016.00213 A Complex Story: Universal Preference vs. Individual Differences Shaping Aesthetic Response to Fractals Patterns Nichola Street 1, Alexandra M. Forsythe 2*, Ronan Reilly 3, Richard Taylor 4 and Mai S. Helmy 5 1 School of Psychology, Sport and Exercise Faculty of Health Sciences, Staffordshire University, Stoke-on-Trent, UK, 2 School of Psychology, University of Liverpool, Liverpool, UK, 3 Department of Computer Science, National University of Ireland, Maynooth, Ireland, 4 Department of Physics, University of Oregon, Eugene, OR, USA, 5 Department of Psychology, Menoufia University, Menoufia, Egypt Fractal patterns offer one way to represent the rough complexity of the natural world. Whilst they dominate many of our visual experiences in nature, little large-scale perceptual research has been done to explore how we respond aesthetically to these patterns. Previous research (Taylor et al., 2011) suggests that the fractal patterns with mid-range fractal dimensions (FDs) have universal aesthetic appeal. Perceptual and aesthetic responses to visual complexity have been more varied with findings suggesting both linear (Forsythe et al., 2011) and curvilinear (Berlyne, 1970) relationships. Individual differences have been found to account for many of the differences we see in aesthetic responses but some, such as culture, have received little attention within the fractal and complexity research fields. This two-study article aims to test preference responses to FD and visual complexity, using a large cohort (N = 443) of participants from around the world to allow universality claims to be tested. It explores the extent to which age, culture and gender can predict our preferences for fractally complex patterns. Following exploratory analysis that found strong correlations between FD and visual complexity, a series of linear mixed-effect models were implemented to explore if each of the individual variables could predict preference. The first tested a linear complexity model (likelihood of selecting the more complex image from the pair of images) and the second a mid-range FD model (likelihood of selecting an image within mid-range). Results show that individual differences can reliably predict preferences for complexity across culture, gender and age. However, in fitting with current findings the mid-range models show greater consistency in preference not mediated by gender, age or culture. This article supports the established theory that the mid-range fractal patterns appear to be a universal construct underlying preference but also highlights the fragility of universal claims by demonstrating individual differences in preference for the interrelated concept of visual complexity. This highlights a current stalemate in the field of empirical aesthetics. Keywords: aesthetics, complexity, fractals, preference, culture, gender Edited by: Daniel J. Graham, Hobart and William Smith Colleges, USA Reviewed by: Jay Friedenberg, Manhattan College, USA Andreas Gartus, University of Vienna, Austria *Correspondence: Alexandra M. Forsythe a.forsythe@liv.ac.uk Received: 01 March 2016 Accepted: 25 April 2016 Published: 24 May 2016 Citation: Reviewed by: Jay Friedenberg, Manhattan College, USA Andreas Gartus, University of Vienna, Austria *Correspondence: Alexandra M. Forsythe a.forsythe@liv.ac.uk Received: 01 March 2016 Accepted: 25 April 2016 Published: 24 May 2016 Citation: Street N, Forsythe AM, Reilly R, Taylor R and Helmy MS (2016) A Complex Story: Universal Preference vs. Individual Differences Shaping Aesthetic Response to Fractals Patterns. Front. Hum. Neurosci. 10:213. doi: 10.3389/fnhum.2016.00213 Reviewed by: Jay Friedenberg, Manhattan College, USA Andreas Gartus, University of Vienna, Austria *Correspondence: Alexandra M. Forsythe a.forsythe@liv.ac.uk Received: 01 March 2016 Accepted: 25 April 2016 Published: 24 May 2016 BEAUTY AND COMPLEXITY Exploring the impact of visual complexity in aesthetic experience has stumbled somewhat on the lack of a unified definition of visual complexity. Complexity arises in situations where an increasing number of independent variables begin interacting in unpredictable interdependent ways (Forsythe et al., 2011). One issue the field of visual complexity faces is the inconsistency of stimulus used to explore the relationship between complexity and visual experience. As outlined in the above discussions, some studies use computer or hand generated stimulus, which increases in complexity by the number of objects, amounts of turns, or presence of symmetry. Others use photographs or nature, art or websites to explore responses to complexity in a more ecologically applicable way. There are a wide variety of approaches with which to gather complexity ratings or rankings from stimulus. Whilst some employ human judgments others have attempted to develop advanced quantitative measures to provide this information, as with computational compression techniques. Berlyne’s (1970) curvilinear hypothesis has received the most attention in this regard. Berlyne (1970) argued that preference increases linearly with complexity until an optimum level of visual arousal is reached. When visual stimuli are of low complexity (i.e., simple), or highly complex, aesthetic preference will also be low (Figure 1). FIGURE 1 | Berlyne’s hedonic experience and complexity. string contains mostly numbers to be repeated. A more complex picture will have more image elements and these elements will be less predictable: the file string will be longer and contain an increasing number of alternatives. Using this method, Forsythe et al. (2011) found that when familiarity was controlled for, the relationship between beauty and visual complexity was more linear in character, with preference taking a different pattern to the curvilinear pattern originally proposed by Berlyne (1970). Despite the wealth of current findings, there is still little consensus about what complexity is and how it should be defined and measured (Forsythe, 2009). The lack of consistency in what it is to be complex in a stimulus means that the field cannot move forward in a unified way. Some argue that complexity as one concept does not exist, and instead we need to explore it as a multidimensional construct, that many different sub-sectors of complexity exist instead of one overarching definition (Forsythe et al., 2011). INTRODUCTION FIGURE 1 | Berlyne’s hedonic experience and complexity. The reduction of aesthetic responses to perceptual stimuli to basic formuli has been a goal of perceptual research since its inception. Birkhoff (1933) argued that order was a positive contributor to aesthetic evaluation, that preferences are driven by the need to experience stability in an image and, as a general rule, visual complexity contributed negatively to the aesthetic experience. Gombrich (1984) extended this opinion by stating that aesthetic appeal lay at the mid- point between complete monotony and total unintelligible chaos. Citation: Citation: Street N, Forsythe AM, Reilly R, Taylor R and Helmy MS (2016) A Complex Story: Universal Preference vs. Individual Differences Shaping Aesthetic Response to Fractals Patterns. Front. Hum. Neurosci. 10:213. doi: 10.3389/fnhum.2016.00213 May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org 1 A Complex Story Street et al. Frontiers in Human Neuroscience | www.frontiersin.org CROSS CULTURAL DIFFERENCES Spehar et al. (2003) report the presence of three categories with respect to aesthetic preference for FD (Spehar et al., 2003; Spehar and Taylor, 2013). Humans are consistent in their low preference for patterns with dimensions between (1.1–1.2) and (1.6–1.9), preferring fractal images in the 1.3–1.5 FD range. This effect is constant regardless of whether these fractals were generated by mathematics (Figure 3), artists or natural processes. Forsythe et al. (2008) demonstrated that familiarity influences the extent to which we perceive an object as complex or as simple. This suggests that the prevalent FD value of the environment in which we grow up may also impact upon the extent to which we find images aesthetically pleasing. The anthropologist Lowie (1921) laid the groundwork for exploration of universals in aesthetic responses by examining the decorative artistic and abstract style of parfleches of the Crow and Shoshoni tribes. In comparing the proportions of the geometric shapes to that of the ‘‘golden section’’ proportion (Fechner, 1876), Lowie found neither group demonstrated the measurement as a universally preferred proportion, rather the Shoshoni norm fell above Fechner’s proposals, and Crow ratios fell below. Cross- cultural work continued to grow in the 1950’s with findings demonstrating high correlations in aesthetic responses between Australian Aboriginals and Caucasian participants (McElroy, 1952). However, authors began to raise concerns over the existence of universal aesthetic principles, arguing that beauty could more reasonably be determined by culture. Researchers, including Lawlor (1955), began to credit cultural experience as an overpowering component in understanding aesthetic universals that may exist when looking between two or more very different cultural heritages. Large scale cross-cultural differences have been demonstrated in judgments of geometric illusions. Some included samples from up to 15 societies over a period of 6 years (Segall et al., 1966) and non-western cultures would be differentially susceptible to geometric illusions because they have discovered different visual habits that may produce/inhibit particular illusionary responses. Similar findings were reported between native Arctic and non-native Arctic workers, and those with no Arctic experience (Sonnenfeld, 1967), once again suggesting that landscape preference was a result of culture mediated preference. FIGURE 2 | Fractal landscape. FIGURE 2 | Fractal landscape. FIGURE 3 | Mathematical fractal. FIGURE 3 | Mathematical fractal. FIGURE 3 | Mathematical fractal. perceptual concepts of the natural, stimulate biophilic responses and trigger aesthetic interest and restorative responses (Joye, 2005, 2006). BEAUTY AND COMPLEXITY Fractal dimension (FD) measures the relative amounts of coarse and fine structure in a fractal pattern and has been proposed as a new method of quantifying the complexity of nature and the relationship between the universality and individual differences (Forsythe et al., 2011). The fractal, a self- similar pattern that reoccurs on finer and finer scales, has been demonstrated to characterize many of the visual patterns in the natural world (Figures 2, 3). Fractal analysis has been successfully used in quantifying the complex structure exhibited by many natural patterns and has captured the imagination of scientists and artists alike, being characterized as ‘‘fingerprints of nature’’ and ‘‘the new aesthetics’’. It is thought that fractals tap into specialist cognitive modules that have developed to moderate information about living things (Wilson, 1984) and that such modules are linked with emotional regulation and the reduction of physiological stress (Taylor et al., 2006, 2011). Hagerhall et al. (2008, 2015) report that viewing fractal patterns elicited high alpha in areas of the brain concerned with attention and visuo-spatial processing providing support for the idea that training using fractal shapes could help the development of Berlyne’s theory has received mixed support because it does not predict the point of the preference cusp (Krupinski and Locher, 1988; Martindale et al., 1990). Furthermore, results consistent with the Berlyne hypothesis were often limited by sample size and biased by familiarity with an image (Forsythe et al., 2008). To further explore this relationship with an unbiased sample, Forsythe et al. (2011) applied compression measures of complexity when standardizing 800 artistic images for visual complexity and beauty. Automated measures have also extended our understanding of the relationship between complexity and beauty. In particular, compression measures of visual complexity offer one method to capture the complexity of an image and a number of different compression measures exist that are suitable to different types of image. A compressed image consists of a string of numbers that represent the organization of that picture. This string is a measure of information content (Donderi, 2006). When the image contains few elements or is more homogenous in design, there are few message alternatives and as such the file May 2016 | Volume 10 | Article 213 2 Street et al. A Complex Story in addressing some of the shortcomings of the Berlyne (1970) hypothesis. FIGURE 2 | Fractal landscape. Frontiers in Human Neuroscience | www.frontiersin.org CROSS CULTURAL DIFFERENCES In line with this, The Savannah hypothesis suggests that spontaneous emotional responses are demonstrated towards landscapes that are positive for survival or instincts which may account for the heightened preference for scenes falling within the mid-range of the fractal spectrum (Appleton, 1975; Orians, 1980; Falk and Balling, 2009). Comparisons of art experts in Japan and the United States have demonstrated that whilst there is some cross cultural agreement in expert aesthetic evaluation (Iwao and Child, 1966), as a whole evidence points towards universal aesthetic exceeding the bounds of ‘‘culture’’ as we classify it (Child and Siroto, 1965; Ford et al., 1966; Child and Iwao, 1968; Iwao et al., 1969). Soueif and Eysenck (1971) and Eysenck and Soueif (1972) recruited British and Egyptian art students and lay people (non-art trained participants) to explore aesthetic responses. Participants were asked to rate Birkhoff’s (1933) polygons for pleasantness. Results showed interesting differences between Fractal geometry has established its usefulness in understanding the degeneration of artworks by artists subsequently diagnosed with neurological deterioration (Forsythe et al., 2005) and in determining the authenticity of major works of art. Taylor et al. (1999) examined the work of Jackson Pollock and determined that Pollock was generating paintings with a high FD and that Pollock was able to fine-tune the dimension of his paintings. Taylor’s work may also be useful Frontiers in Human Neuroscience | www.frontiersin.org May 2016 | Volume 10 | Article 213 3 Street et al. A Complex Story FIGURE 4 | Example sample set of a full fractal range (top left through to bottom right D1.1–D1.9 increasing in 0.1 increments). FIGURE 4 | Example sample set of a full fractal range (top left through to bottom right D1.1–D1.9 increasing in 0.1 increments). | Example sample set of a full fractal range (top left through to bottom right D1.1–D1.9 increasing in 0.1 increments). cultural responses to these shapes. British art students showed preference for simple figures, and British lay people preferred complex figures. This trend is reversed within the Egyptian sample with art-educated participants preferring the complex figures and the lay participants demonstrated preference towards the simple images. Despite this curious directional result, no significant differences were found between the British and Egyptian groups as a whole. There were also no significant differences in preference between both art and non-art trained participants, although the trend seemed to suggest reversed trends for complexity preference. CROSS CULTURAL DIFFERENCES Eysenck and Soueif (1972) did not believe that their data support considerably large differences in aesthetic preference between both cultures but instead hint towards more universal preferences over cultural issues. show general consistency between preferences, but marked differences based on demographic details and background. For example, within the Australian sample, preferences of participants from a suburban environment and school differed significantly from participants from an urban and industrialized environment and school. Similarly, Zube and Pitt (1981) and Zube (1984) found differences between Yugoslavian, West Indian and American cultures that suggested that not all cultures shared a preference for natural scenes over manmade. Frontiers in Human Neuroscience | www.frontiersin.org AGING AND GENDER Aesthetic responses to natural environments have been found to show marked age differences (Balling and Falk, 1982). Children showed preferences for Savannah scenes over more familiar environments. Adolescences demonstrated consistently lower preferences for all scenes compared to any of the other groups in the sample. This effect has been reported elsewhere with Studies examining sub-cultural or micro environmental differences in Australia, Pakistan and Thailand have reported variability within sub-cultures (Anderson, 1976). The results May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org Street et al. A Complex Story For example, Fedrizzi (2012) found that paintings that captured behavior evoked more pleasure and attention among female participants over male participants. Fedrizzi (2012) suggests that neuroanatomical studies can enhance the comprehension of why such gender differences appear to exist. Other evidence demonstrating gender differences in cognitive processes (Leder et al., 2004; Marin and Leder, 2013) supports this claim. Cela- Conde et al. (2009) also report gender-related differences in parietal activity during aesthetic appreciation and judgments. While in both sexes activity is focused in the parietal lobe, it appears that males show lateralized right hemisphere activation in the parietal lobe while females show bilateral activity in the same region. These results point towards a difference in the way males and females process aesthetic appreciation, although specifying how the differences manifest in response is challenging and yet to be explored in great depth. adolescent preference for natural scenes distinctively lower when compared to younger or older generations (Herzog et al., 2000). Adolescents show a higher appreciation of developed and urban settings than their different aged counter parts (Lyons, 1983; Medina, 1983, unpublished doctoral dissertation; Kaplan and Kaplan, 2002). Kaplan and Kaplan (1989) argue that preference does not suggest adolescents do not show appreciation for natural settings, rather the pattern of preference related to the adolescent’s relationships with urban spaces. Such spaces facilitate group and social spaces, rather than solo and natural scenes. Elderly samples have been found to display relatively low preference for wild nature over more managed and developed natural landscape (Balling and Falk, 1982; Lyons, 1983; Strumse, 1996; Van den Berg and Vlek, 1998). Some have attributed this difference in elderly samples as relating to evolutionary drives that would make wilderness scenes a larger risk to vulnerability, both physical and psychological. AGING AND GENDER Alternative theories such as increased cultural and experiential shaping have been attributed to this change in preference at older ages (Van den Berg and Koole, 2006). However, studies of non-normal aging appear to show stability in preference for art stimulus despite neurodegenerative diseases such as Alzheimer’s, which suggests that in older age preferences remain stable despite the functioning capacity of working memory (Halpern et al., 2008). Other studies exploring neurological conditions however appear to show aesthetic changing in abnormal ageing such as a marked aesthetic difference in art production and quality following a stroke (Zaimov et al., 1969). The impact of chronological age alone is difficult to interpret as a separate influential factor in aesthetic judgments because individual experience, education and social factors cannot be uncoupled from age. THE CURRENT STUDY Given the visual similarity and theoretical links between FD and visual complexity, this article aims to first measure the relationship between the FD of the stimulus set and visual complexity. Overall patterns of preference were identified and following on from this, linear mixed-effect modeling was used to investigate the extent to which individual differences of culture, gender and age account for variation in preferences. The analysis explored two different models to map and understand patterns of preference. The first model tested a linear complexity model (likelihood of selecting the more complex image from the pair of images) and the second a mid-range FD model (likelihood of selecting an image within the mid-range 1.3–1.5D from the pair of images). Gender could be mediating factor in preference, but the limited research that exists is split with reports of clear aesthetic differences, or no tangible evidence what so ever (Farrell and Rogers, 1982; Limbert and Polzella, 1998). Earlier studies demonstrated that women appeared more attracted to impressionist paintings than men, with men showing preference for modern paintings (Bernard, 1972). Women preferred representational art, which displays soft and curved patterns whereas men preferred more abstract work containing higher numbers of pointed or sharp shapes (Cupchik and Gebotys, 1988). However, given that these studies were reported almost 30 years ago, it is possible that such differences have diminished. More recent studies have reported that women demonstrate an overall higher appreciation of art than males (Frumkin, 1963) and that abstract art was generally rated more highly by females than their male counterparts (Furnham and Walker, 2001). Polzella (2000) looked at gender differences in college students for color reproductions of art stimulus. Results found that impressionist paintings were judged as the most pleasing by females, and also evoked relaxation and alertness. It was concluded that the differences between males and females might be a result of differences in perceptual style and emotional sensitivity between genders (Polzella, 2000). It appears that perceptual styles or affective processing may result in significant differences in aesthetic response between genders. Frontiers in Human Neuroscience | www.frontiersin.org Results To explore the results in depth three stages of analysis were conducted. Initially, overall patterns of preference were explored using SPSS software package (1a) followed by the Complexity Model Linear Mixed Effects analysis (1b) and finally the Mid- Range Model, Linear Mixed Effects Modeling (1c) using the software environment R. Participants Four hundred and forty-three participants, 228 men and 204 women, aged between 17 and 88 (m = 31.03 SD = 14.45) took part in the study. Participants were recruited using a variety of methods including opportunity, online and targeted recruitment. Participants were recruited across four continent locations, Europe (n = 177), North America (n = 24), Central Asia (n = 195) and Africa (n = 97). Each participant was provided with an information sheet ahead of the study and informed consent was recorded for each. Experiment 1a: Exploring Preference Patterns Preference patterns across the scale of the averaged FD were explored for frequency of choice (Figure 5). Overall patterns of preferences found the peak of preference at D1.2 (M = 6.89, SD = 3.89) and preference choices lowest at D1.8 (M = 4.69, SD = 3.89). Mauchly’s test indicated that the assumption of sphericity had been violated, (χ2 (35) = 5518.76, p < 0.001). Therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (ε = 0.18). The results show that there was a significant effect of FD, (F(1.467, 777.46) = 35.06, p < 0.001, η2 p = 0.06). These results suggest that preference ratings differ significantly between each FD supporting an Location To explore the role of individual differences in preference for FD analysis we examined mean preference choices across the fractal scale between continent location groupings. Mauchly’s test indicated that the assumption of sphericity had been violated, (χ2 (35) = 4995.552, p < 0.001). Therefore, degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (ε = 0.184). The results show that there was a significant effect of FD and location grouping (F(4.42,720.6) = 4.00, p < 0.001, η2 p = 0.02). These results suggest that preference ratings differ significantly between FD and location grouping. However, post hoc analysis found no direct differences across the groups (Figure 6). Within Figure 6, error bars demonstrate significant differences between the samples with the North American samples demonstrating significantly greater variance in findings. The North America sample also shows a negative linear preference differing from the suggested positive direction of preference noted in the European, Central Asian and African samples. This finding may be a result of the limited participant numbers in this group. Procedure Participants were asked to provide aesthetic responses to the stimulus set, standardized for complexity. Using a two alternative forced choice analysis, participants made preference choices between 57 stimulus pairs. To run a full forced choice analysis it would have involved running 6480 pair comparisons. The 57 pair selection was based on work by McManus (2009) who developed a regression-based technique for reducing the numbers of pairings required when using a forced choice design in visual aesthetics. In brief, McManus’s method samples an entire range of stimuli but also provides detailed information on closely similar images. This design is further justified by Taylor et al. (2011) who found no significant differences within closely related sets of fractal images. In addition to preference choices, participants also provided demographic details including Age, Gender, and continent of residence to allow exploration of individual differences in preference responses. FIGURE 5 | Line graph of preference choice number (2A-FC) for fractal dimension (FD) for overall sample. Ethical approval for the study was granted by the Ethics Committee in the Institute of Psychology, Health and Society at the University of Liverpool. exploration of preference patterns in a number of directions (Figure 5). Stimulus Measurement The stimuli consisted of 81 abstract monochrome fractal images (9 full sets of 9 iterations of FD) generated using the mid-point displacement technique (Fournier et al., submitted). All stimuli were developed in bitmap format (Figure 4). Initial stages of the article involved obtaining visual complexity scores for 81 abstract fractal patterns by analyzing for visual complexity using GIF compression based on measures by Forsythe et al. (2003). Forsythe et al. (2008, 2011) used compression file size as a measure for visual complexity. This measure works well on highly complex images such as photographs and drawings. Given the nature of the monochrome images in this study, bitmap compression to converted Gif file ratio was applied. The amount of information between the original bitmap file and new compressed file provides the GIF Ratio score. Higher GIF ratios represent lower complexity scores. Higher GIF ratio scores are a result of less difference between the bitmap size and the GIF compression size. The results of this exploration show a strong negative correlation between FD and the GIF ratio complexity measure (r(79) = −0.93, p < 0.01) were found for the 81 image set. In other words, images with small compression ratio scores tend to have a higher FD. This allows assumptions about preferences to both FD and visual complexity to be made in subsequent analyses. May 2016 | Volume 10 | Article 213 5 Street et al. A Complex Story Experiment Procedure FIGURE 5 | Line graph of preference choice number (2A-FC) for fractal dimension (FD) for overall sample. Frontiers in Human Neuroscience | www.frontiersin.org Gender The frequency preference patterns of FD were explored across gender. Male preference peaks at D1.2 (M = 7.32, SD = 3.94) and continues to fall as FD increases from this point. The patterns of preference for male participants point towards a low-to-mid peak in preference for fractal complexity. The female sample shows a different pattern of preference across the fractal scale. Female preference peaks at D1.6 (M = 6.58, SD = 1.88) and analysis shows less variation across each FD than male participants. This lack of May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org 6 Street et al. A Complex Story ar chart of preference choice number (2A-FC) for FD as a function of geographic location (error bars 95% CI). FIGURE 6 | Bar chart of preference choice number (2A-FC) for FD as a function of geographic location (error bars 95% CI). (t(528) = −0.404, p = 0.687) and D1.7 (t(528) = −01.32, p = 0.187) however significant differences were found across D1.1 (t(528) = 2.68, p < 0.01), D1.2 (t(528) = 2.53, p < 0.05), D1.3 (t(528) = 2.11, p < 0.05), D1.6 (t(528) = −4.02 p < 0.001), D1.8 (t(528) = −2.18, p < 0.05), and D1.9 (t(528) = −2.17, p = < 0.05). clear variance across scales means there is no clear directional or curvilinear pattern emerging from the frequency data for the female sample (Figure 7). Additional post hoc analysis explored if gender preference differed as each FD as a function of gender. Results show no significant difference in preference across D1.4 (t(528) = 1.52, p = 0.129), D1.5 Frontiers in Human Neuroscience | www.frontiersin.org Frontiers in Human Neuroscience | www.frontiersin.org May 2016 | Volume 10 | Article 213 7 Street et al. A Complex Story FIGURE 7 | Bar chart of preference choice number (2A-FC) for FD as a function of gender (error bars 95% CI). FIGURE 8 | Bar chart of percentage choice of complex image from a pair across continent. FIGURE 8 | Bar chart of percentage choice of complex image from a pair across continent. FIGURE 7 | Bar chart of preference choice number (2A-FC) for FD as a function of gender (error bars 95% CI). Gender Although the difference between the African and European continents was not significant as a main effect, the model shows a significant interaction between continent (Africa- Europe) and gender (β = 1.446, z = 2.64, p < 0.01). African males show a lower probability of choosing a complex image (7%) than African females (27%), and European males show a lower probability (10%) than European females (35%) of choosing the more complex image from a pair (Figure 9). Interaction effects were also found between Central Asia and Europe and Gender (β = 1.446, z = 2.636, p < 0.01). As discussed, European males show approximately 10% probability with European females, having approximately a 35% probability of choosing the more complex image from a pair. Central Asian males had an approximately 1% probability of choosing the more complex image and females had a 5% probability of choosing the more complex image. A significant interaction was also seen between gender and age (β = 0.028, z = 2.096, p < 0.01). Male and female complexity choices show changes as a function of age. Female participants’ preference for complexity decreases with age, whereas male participants’ preference for complexity increases with age (Figure 10). Experiment 1b: Complexity Model (A) Results Two models of preference were tested using linear mixed effects analysis. The first of two models explored was a Complexity Model, which explored the extent to which the variables, Continent, Gender and Age could predict the choice of more complex (higher FD) fractal images, from a pair. The complexity model accounts for significantly more variance with fixed and random effects (AIC = 17040, df = 21) than the null model with random effects alone (AIC = 17154, df = 3), suggesting that the model is improved with the additional variables (χ2 (18) = 125.14, p < 0.001). Additional goodness of fit analysis found that Continent (χ2 (15) = 99.06, p < 0.001) and Gender (χ2 (7) = 29.29, p < 0.001) significantly improves the overall fit of the model. However, age (χ2 (7) = 6.845, p = 0.445) does not significantly add to the overall prediction of the model. Results of the model identify a significant main effect of continent on preference for complex fractal patterns. European participants show an average 22% choice for the complex image from a pair (Figure 8). This did not differ significantly from the African sample (β = −0.356, z = −0.535, p = 0.592), with an average choice of approximately 17%. Significant differences were however found between European (22%) and North American (8%) samples (β = −1.311, z = −2.601, p < 0.01) and European (22%) and Central Asian (3%; β = −2.261, z = −8.267, p < 0.001) suggesting individual differences in preference across locations. Additional main effects were also found for gender, with females showing a significant increased choice in complex fractal shapes (20%) compared with males (5%; β = 1.585, z = −4.827, p < 0.001). Experiment 1c: Mid-Range Model (B) Results The Mid-Range Model explored the extent to which the variables Continent, Gender and Age could predict the choice of the Mid-Range image from a pair. The Mid-Range Model accounts for significantly more variance with fixed and random effects (AIC = 17040, df = 21) than the null model with random effects alone (AIC = 17154, df = 3), suggesting that the model is improved with the additional variables (χ2 (18) = 125.14, p < 0.001). Additional goodness of fit analysis found that Continent (χ2 (15) = 27.89, p < 0.05) improved the overall fit of the model. However, because the chi-square test based on May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org Frontiers in Human Neuroscience | www.frontiersin.org 8 Street et al. A Complex Story does not significantly add to the overall prediction of the model. FIGURE 9 | Bar chart of percentage choice of complex image from a pair across continent and gender. The model found a main effect of continent on preference for mid-range fractal image. Significant differences in percentage choices were seen between Europe and Central Asia (β = −0.198, z = −2.262, p < 0.05). European samples have an average probability of choosing the mid-range D of 85% and Central Asia samples show an average 83% probability. There are also marginally significant differences in preference choice between European and North American samples (β = −0.339, z = 1.937, p = 0.052). European samples demonstrate a lower probability of choosing mid-range fractal patterns (85%) than North American samples (89%). In addition, the model found a significant main effect of gender on preference for mid-range fractal images with Females (88%) showing an increased probability of choosing mid-range fractal patterns when compared with males (86%; β = −0.215, z = −2.025, p < 0.05). Significant interaction was seen between continent (Central Asia-Europe) and gender (β = 0.299, z = 2.242, p < 0.05), with European males showing a lower preference choice (85%) than European females (87%), and Central Asian males showing a lower choice (82%) than females (85%; see also Figure 11). FIGURE 9 | Bar chart of percentage choice of complex image from a pair across continent and gender. DISCUSSION FIGURE 10 | Interaction of percentage choice of complex image between gender and age in complexity model. Models of complexity and preference can be traced back to the very early field of empirical aesthetics (Birkhoff, 1933). It has been repeatedly noted that complexity plays a role in aesthetic judgments. There have however been inconsistent findings within the field, which may be explained by the associated difficulty in defining visual complexity as one Models of complexity and preference can be traced back to the very early field of empirical aesthetics (Birkhoff, 1933). It has been repeatedly noted that complexity plays a role in aesthetic judgments. There have however been inconsistent findings within the field, which may be explained by the associated difficulty in defining visual complexity as one FIGURE 11 | Bar chart of percentage choice of mid-range image from a pair across continent and gender. FIGURE 10 | Interaction of percentage choice of complex image between gender and age in complexity model. likelihood ratios is anti-conservative and the rule-of-thumb is to halve the p-values (Pinheiro and Bates, 2000), this improvement can only be considered marginally significant; the variable Age marginally improves the model, but not significantly (χ2 (7) = 12.99, p = 0.072) however Gender (χ2 (7) = 11.81, p = 0.107) FIGURE 11 | Bar chart of percentage choice of mid-range image from a pair across continent and gender. Frontiers in Human Neuroscience | www.frontiersin.org Frontiers in Human Neuroscience | www.frontiersin.org Frontiers in Human Neuroscience | www.frontiersin.org May 2016 | Volume 10 | Article 213 9 Street et al. A Complex Story however the variability of preference, as a function of location and gender, does not support the theory of universal preference for mid-range fractal patterns. construct. As such the current article aimed to unpick the multifaceted construct of visual complexity and explore the statistical relationship between computational complexity and FD. As noted by Taylor et al. (2011), most complexity research has focused on the responses to complexity of scenes made up of Euclidean shapes, suggesting that the applicability of previous findings to natural scenes are limited. Here we attempted to further develop this field by first examining the construct of complexity and FD and then exploring the individual factors that can predict preference towards these patterns. DISCUSSION Our results demonstrated the hypothesized relationship between constructs and also found an influence of individual differences across environment (continent) and gender in complex fractal preference. Results from the Linear Mixed effects mid-range model confirm this pattern with high overall preference choice percentages found in the choice of mid-range over non-mid range images (80–90%). One potential rationale could be the survival benefits associated with landscape displaying approximately mid-range FD. Orians (1980) Savannah hypothesis suggested that spontaneous emotional responses to landscapes that are positive for survival or instincts and then preference fall for Savannah type landscape (found to display mid-range fractal patterns) because they are most akin with the environment in which we developed and provided shelter and survival benefits. The Savannah hypothesis has been found to be consistent and preferences for Savannah-type landscape are seen in individuals, even if they have never had visual interactions with this type of environment (Falk and Balling, 2009). This effect is particularly strong in children (Balling and Falk, 1982). Similarly, the refuge theory rationalizes that preference for visual environment is greatest for those that can offer both shelter and ability to move undetected (Appleton, 1975). Although the refuge theory has been criticized for its limited ability to demonstrate why differences exist across culture, Appleton (1996) addressed these criticisms when he asserted that preferences may well be shaped by culture, experience and historical influence. He highlights that the foundations of these trends are not within a vacuum and as such still have links in environmental preferences based upon these evolutionary instincts of aesthetic preference. While it could be suggested that biological and evolutionary foundations play a role in shaping our visual preferences for fractal patterns, they could also be shaped in part by our current visual and cultural experiences as found in some previous cross- cultural research, accounting for significant interactions found here. In models of both visual complexity and fractal aesthetics it appears mid-range fractal images have a powerful aesthetic draw. Our aim was to explore the stability of the mid- range preference and examine the claims that preferences exist universally for this mid-range (Berlyne, 1970; Spehar et al., 2003). This was achieved by examining patterns of preference across Culture, Gender and Age. When examining participant responses as a whole, data reported here corroborate previous studies (Taylor et al., 2011) that demonstrate a preference for mid-range FD. Frontiers in Human Neuroscience | www.frontiersin.org Testing the Mid-Range Hypothesis g g yp The Mid-Range Model found evidence to suggest individual differences account for the variance in preference for mid- range fractal images. Main effects of location were shown with significant differences in preference between European and North American samples. North American participants showed a heightened preference for mid-range images. Females demonstrate a heightened preference choice of the mid- range fractal images. Interaction effects are evident across gender and location (Central Asia—Europe) with females more often choosing the mid range images, than their male counterparts. Aesthetic responses to symmetry can also show how preference for abstract geometric forms can be based on evolutionary survival. Symmetry appears in mate selection as it represents strong genes, or may have been an unintentional by-product of visual shape recognition (Enquist and Arak, 1994). Symmetry is an easily perceived measure of genetic quality (Møller, 1990; Parsons, 1992) and these same conclusions can be drawn about responses to fractal patterns. FD at the mid-range is an effective and easily perceived measure of naturalness and has the most positive psychological and physiological benefit (Taylor et al., 2011). Similarly, preferences for facial attractiveness are found in early infants before learned behavior could be possible, suggesting innate and evolved responses (Langlois et al., 1987). As our experiences with faces grow we begin to understand society specific visual experiences what the ‘‘average’’ face looks like and respectively our preferences for faces become more socially and culturally tied. Using this theory as a guide, we could propose a type of aesthetic development for fractal patterns. Innate responses could (and should) be found toward mid- Initial frequency choice analysis shows the overall patterns of preference demonstrate a peak at D1.2, one point below the previously reported (D1.3–1.5) FD peak (Taylor et al., 2011). The pattern emerging is one of heightened preference for the low to mid-range fractal patterns. This pattern is somewhat consistent with previous findings, however here we find that preference falls incrementally with increases in FD. This suggests that as FD increases preferences decrease. Looking at the overall preference patterns in the data, there is an evident negative relationship, with groups demonstrating higher preference choices towards the lower end of the fractal scale and preference decreasing as FD/complexity increases. Overall preference patterns show a complex story. DISCUSSION However, constituent analysis highlights the potential issues with the current theory of mid- range fractal peak in preference, as culture, and gender of participants appears to play a role in shaping different patterns in preference. CONCLUSION Berlyne (1970) proposed a curvilinear hypothesis of aesthetics and complexity, in which preference increases linearly with complexity until an optimum level of visual arousal is reached, at which point preference begins to fall. Despite its wide spread recognition, Berlyne’s arousal theory has been criticized as it fails to predict the point of preference (Krupinski and Locher, 1988; Martindale et al., 1990). Questions have also been raised about issues such as the samples size and Berlyne (1970) proposed a curvilinear hypothesis of aesthetics and complexity, in which preference increases linearly with complexity until an optimum level of visual arousal is reached, at which point preference begins to fall. Despite its wide spread recognition, Berlyne’s arousal theory has been criticized as it fails to predict the point of preference (Krupinski and Locher, 1988; Martindale et al., 1990). Questions have also been raised about issues such as the samples size and Testing the Complexity Hypothesis In addition to the main effect of continent on fractal complexity preference, continent was also found to interact significantly with gender. Whilst European participants are generally more likely to choose the complex image, there are different directional patterns in probability with males in Central Asia being more likely to pick the complex image and European females being more likely to pick the complex image. The pattern was somewhat different for African/European participants, with female participants being most likely to choose the complex image than males in both continent groups. These results demonstrate that it is possible to predict preference based on continent and gender although the relationship is a complex one. These cross-cultural findings contest many of the previous studies that report a general consistency in aesthetic response across cultures (Child and Iwao, 1968; Iwao et al., 1969; Eysenck and Iwawaki, 1971) and suggest that visual experiences over universal preferences may play a larger role when making aesthetic judgments on complex images. g p y yp Findings discussed above outline the patterns found when exploring the universality of the mid-range hypothesis. However, the study also attempted to explore the links between FD and visual complexity and in addition attempted to shed some light of the patterns of preference for complex stimulus, which is currently unconfirmed. While Berlyne (1970) proposed a curvilinear pattern, Forsythe et al. (2011) proposed a linear relationship between preference and complexity. This stalemate suggests further factors may influence how these preferences emerge. Compared with the likelihood of selection for the mid-range images (approximately 80–90% of choices for the mid-range), there were significantly lower levels of selection of images of higher complexity with selection rates for the complex image from a pair ranging from less than 1% up to 20% selection. Samples demonstrated less preference towards complex images. Compared with the mid-range model, the complexity model found significant individual differences, which appear to shape preference for complexity stimulus. These include location (continent) and gender and point toward a significant role of individual differences when exploring aesthetic response to complexity. Testing the Mid-Range Hypothesis When the entire population is examined, samples predictably show distributions consistent with the mid-range fractal hypothesis, May 2016 | Volume 10 | Article 213 10 Street et al. A Complex Story range, but with repeated exposure to other non-mid or non-fractal environments what we consider most aesthetically pleasing will change with experience. Cultural ties reduce and change a typical scene to which we compare new viewing. We eventually learn the ‘‘norm’’ of the visual environment in which we spend time and these repeated exposure shaped our aesthetic perceptions according to the exposure and fluency hypotheses (Zajonc, 1968; Reber et al., 2004). preference dropping incrementally as FD (and related complexity) increase. The lowest preference choices in this sample are seen in the high end of the fractal scale. The North American sample demonstrates a different pattern of preference, with highest preferences shown for higher FD. Within this sample (the smallest populated sample) preference peaks at D1.4 and is lowest across the lower FDs (D = 1.1–1.3). Analysis of a worldwide sample demonstrated significant differences in the frequency patterns of preference as a result of locational grouping, suggesting that residential location shapes preferences for fractal patterns. These differences were also found in data interrogating both the complexity and mid-range models, suggesting that the location in which we spend time influences the preference choices that we make. North American participants have the highest choice of mid-range fractal patterns. European samples are three times more likely than the other continents to choose complex fractal patterns. Gender and Continent Interactions There are significant interactions between continent and gender in the complexity model, with females more than three times more likely to choose the complex image than their male counterparts. This pattern reverses with aging; with male preference for complexity increasing and female preference decreasing. This raises interesting questions about the stability of preference across age and other individual differences. Gender Despite the consistent patterns emerging that support universal patterns of preference for images falling within the mid-range of the fractal spectrum when exploring between subjects, gender played on a role in the likelihood and patterns of preferences shown toward this range. Males demonstrate a negative linear relationship between preference choice and FD with an incremental decrease in preference as FD increases. Females, however, show a higher peak preference point (D = 1.6) with less variance across the FD scales than the male sample. Linear Mixed Effects modeling suggests that gender is a significant predictor of preference for mid-range fractal patterns. Females have a significantly higher probability of choosing mid-range patterns over males. Frontiers in Human Neuroscience | www.frontiersin.org Location When exploring across location there are emergent differences in preference patterns. The European, Central Asian and African samples show similar patterns of preference with peaks at the lower-to-mid end of the fractal scale and May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org 11 Street et al. A Complex Story the existence of a familiarity bias (Forsythe et al., 2011). Alternative linear patterns of preference have been proposed (Forsythe et al., 2008, 2011) suggesting that further work is required to understand the relationship between beauty and complexity. The Complexity model within this article explored the linear model of complexity, examining if individual differences (Culture, Gender and Age) shaped the likelihood of choosing a complex image from a pair. Results demonstrate significant main effects of gender, with females demonstrating higher preference for complex images than males and a main effect of continent across European and Central Asian, and European and North American samples. European samples show overall a higher preference for fractal complexity. An interaction between gender and age was also discovered within the model and results show an opposite directional effect with male participants’ preference for complexity increasing with age and female participants’ preference for complexity decreasing with age. 1992; Silverman et al., 2007). Could it then be argued that the differences in neural processes are a result of the different visual strategies? Foraging has been proposed to require a greater understanding of the complex visual scenes, therefore this could account for the heightened preference in females for high complexity images over simpler patterns as it indicates signs of vegetation. Females employ categorical spatial relations during aesthetic judgment and this processing relies heavily on long-term memory so associations can be made between previous and current scenes. Suggestions could tentatively be made that if long-term memory plays a greater role in women than men, aesthetic judgment particular for scenes, woman use memory more that could mean that experiences play a larger role in aesthetic processing than it does in males. The current study adopted a dichotomous forced-choice design with pure fractal imagery. Whilst such design permits control of familiarity effects, color and other collative variables it is not ideal for understanding the nature of real world preference. It is further acknowledged that macro visual environment across continent may have some, but limited, differences across the locations used. REFERENCES Donderi, D. C. (2006). Visual complexity: a review. Psychol. Bull. 132, 73–97. doi: 10.1037/0033-2909.132.1.73 Anderson, J. (1976). Language, Memory and Thought. Hillsdale, NJ: Erlbaum Associates. Enquist, M., and Arak, A. (1994). Symmetry, beauty and evolution. Nature 372, 169–172. doi: 10.1038/372169a0 Appleton, J. (1975). The Experience of Landscape. New York, NY: John Wiley. Eysenck, H. J. and Iwawaki, S. (1971). Cultural relativity in aesthetic judgments: an empirical study. Percept. Mot. Skills 32, 817–818. doi: 10.2466/pms.1971. 32.3.817 Appleton, J. (1996). The Experience of Landscape. (Revised edition) New York, NY: John Wiley. Eysenck, H. J. and Soueif, M. (1972). An empirical test of the theory of sexual symbolism. Percept. Mot. Skills 35, 945–946. doi: 10.2466/pms.1972.35.3.945 Balling, J. D., and Falk, J. H. (1982). Development of visual preference for natural environments. Environ. Behav. 14, 5–28. doi: 10.1177/0013916582141001 Falk, J. H., and Balling, J. D. (2009). Evolutionary influence on human landscape preference. Environ. Behav. 42, 479–493. doi: 10.1177/0013916509341244 Falk, J. H., and Balling, J. D. (2009). Evolutionary influence on human landscape Berlyne, D. E. (1970). Novelty, complexity and hedonic value. Percept. Psychophys. 8, 279–286. doi: 10.3758/bf03212593 Berlyne, D. E. (1970). Novelty, complexity 8, 279–286. doi: 10.3758/bf03212593 , J , g, J ( ) y p preference. Environ. Behav. 42, 479–493. doi: 10.1177/0013916509341244 Bernard, Y. (1972). Sex influence in aesthetic behavior. Percept. Mot. Skills 34, 663–666. doi: 10.2466/pms.1972.34.2.663 Farrell, E., and Rogers, P. (1982). Is aesthetic response independent of age, sex and IQ? J. Aesthetic Educ. 16, 96–98. doi: 10.2307/3332528 IQ? J. Aesthetic Educ. 16, 96–98. doi: 10.2307/3332528 Birkhoff, G. (1933). Aesthetic Measure. Cambridge, MA: Harvard University Press. Fechner, G. (1876). Vorschule der Aesthetik [Pre-school of aesthetics]. Leipzig, Germany: Breitkopf and Härtel. Cela-Conde, C. J., Ayala, F. J., Munar, E., Maestú, F., Nadal, M., Capó, M. A., et al. (2009). Sex-related similarities and differences in the neural correlates of beauty. Proc. Natl. Acad. Sci. U S A 106, 3847–3852. doi: 10.1073/pnas. 0900304106 Fedrizzi, L. (2012). Beauty and its perception: historical development of concepts, neuroaesthetics and gender-differences. Rendiconti. Lincei 23, 259–269. doi: 10. 1007/s12210-012-0177-1 Ford, C. S., Prothro, E. T., and Child, I. L. (1966). Some transcultural comparisons of esthetic judgment. J. Soc. Psychol. 68, 19–26. doi: 10.1080/00224545.1966. 9919661 Child, I. L., and Iwao, S. (1968). Personality and esthetic sensitivity: extension of findings to younger age and to different culture. J. Pers. Soc. Psychol. 8, 308–312. doi: 10.1037/h0025599 Child, I. L., and Siroto, L. (1965). AUTHOR CONTRIBUTIONS NS: Work from her PhD. AMF: Supervisor and co-writer. RR: Statistical guidance. RT: Fractal advice. MH: Data collection in Egypt. NS: Work from her PhD. AMF: Supervisor and co-writer. RR: Statistical guidance. RT: Fractal advice. MH: Data collection in Egypt. Location As direct visual environment appears to play a role in shaping preference patterns, a fruitful area to explore would be the effect of micro or sub-cultural environments, for example urban and rural landscapes. Such analysis would allow a further investigation of the role that exposure to fractal over Euclidean geometry has in shaping preference. Speculations could be made based on the results from the study to offer support for the hunter-gather hypothesis that differences in spatial abilities between genders provide a convincing scenario of sex differences based in our evolutionary history, as a result of the division of labor between sexes (Silverman and Eals, 1992; Silverman et al., 2007). Although no neuropsychological measures were used in the current study, differences in neural activity in gender have been noted (Cela- Conde et al., 2009). Viewing scenes has been proposed to differ as a result of the labor division between hunting (a primarily male activity) requiring coordinated spatial relation processing and mental rotation skills and foraging (a mainly female pursuit) requires categorical spatial relations, including recognition and remembering the content of varied objects and spatial relations between the objects (Silverman and Eals, REFERENCES Bakwele and american esthetic evaluations compared. Ethnology 4, 349–360. doi: 10.2307/3772785 Forsythe, A. (2009). Visual complexity: is that all there is? Eng. Psychol. Cogn. Ergon. 158–166 doi: 10.1007/978-3-642-02728-4_17 Forsythe, A., Mulhern, G., and Sawey, M. (2008). Confounds in pictorial sets: the role of complexity and familiarity in basic-level picture processing. Behav. Res. Methods 40, 116–129. doi: 10.3758/brm.40.1.116 Cupchik, G. C., and Gebotys, R. (1988). The experience of time, pleasure and interest during aesthetic episodes. Empirical Stud. Arts 6, 1–12. doi: 10. 2190/5yn3-j3p8-fwhy-udb3 May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org 12 Street et al. A Complex Story Martindale, C., Moore, K., and Borkum, J. (1990). Aesthetic preference: anomalous findings for Berlyne’s psychobiological theory. Am. J. Psychol. 103, 53–80. doi: 10.2307/1423259 Forsythe, A., Nadal, M., Sheehy, N., Cela-Conde, C. J., and Sawey, M. (2011). Predicting beauty: fractal dimension and visual complexity in art. Br. J. Psychol. 102, 49–70. doi: 10.1348/000712610X498958 Forsythe, A., Sheehy, N., and Sawey, M. (2003). Measuring icon complexity: an automated analysis. Behav. Res. Methods Instrum. Comput. 35, 334–342. doi: 10.3758/bf03202562 McElroy, W. A. (1952). Aesthetic appreciation in aborigines of arnhem land: a comparative experimental study. Oceania 23, 81–94. doi: 10.1002/j.1834-4461. 1952.tb00190.x Forsythe, A., Williams, T., and Reilly, R. (2005). What paint can tell us? A fractal analysis of neurological changes in six artists. Neuroscience 238:S13. doi: 10. 1016/s0022-510x(05)80054-1 McManus, I. C. (2009). Rectangle Aesthetics, Research department of clinical and health psychology, University College London. Available online at: http://www.ucl.ac.uk/medical-education/other-studies/aesthetics/resources/ rectangle-aesthetics Frumkin, R. M. (1963). Sex, familiarity and dogmatism as factors in painting preferences. Percept. Mot. Skills 17, 12. doi: 10.2466/pms.1963.17.1.12 Møller, A. P. (1990). Female choice selects for male sexual tail ornaments in the monogamous swallow. Nature 332, 640–642. doi: 10.1038/332640a0 Furnham, A., and Walker, J. (2001). Personality and judgments of abstract, pop art and representational paintings. Eur. J. Pers. 15, 57–72. doi: 10. 1002/per.340 Orians, G. H. (1980). ‘‘Habitat selection: general theory and applications to human behaviour,’’ in The Evolution of Human Social Behavior, ed. J. S. Lockard (New York, NY: Elsevier), 49–77. Gombrich, E. H. (1984). Art and Illusion (Bollingen Series 35). Princeton, NJ: Princeton University Press. Parsons, P. A. (1992). Fluctuating asymmetry: a biological monitor of environmental and genomic stress. Heredity 68, 361–364. doi: 10.1038/hdy. 1992.51 Hagerhall, C. M., Laike, T., Küller, M., Marcheschi, E., Boydston, C., and Taylor, R. P. (2015). REFERENCES doi: 10.1162/leon.2006.39.2.145 Spehar, B., Clifford, C. W. G., Newell, B. R., and Taylor, R. P. (2003). Universal aesthetic of fractals. Comput. Graph. 27, 813–820. doi: 10.1016/s0097- 8493(03)00154-7 Kaplan, R., and Kaplan, S. (1989). The Experience of Nature: A Psychological Perspective. New York, NY: Cambridge University Press. Spehar, B., and Taylor, R. P. (2013). Fractals in art and nature: why do we like them? SPIE Special Edition on Human Vision and Electronic Imaging. Kaplan, R., and Kaplan, S. (2002). ‘‘Adolescents and the natural environment: a time out,’’ in Children and Nature: Psychological, Sociocultural and Evolutionary Investigations, eds P. H. Kahn, Jr. and S. R. Kellert (Cambridge, MA: MIT Press), 227–258. Strumse, E. (1996). The Psychology of Aesthetics: Explaining Visual Preferences for Agrarian Landscapes in Western Norway. Thesis. Bergen: Research Center for Health Promotion, Universiry of Bergen. Krupinski, E., and Locher, P. (1988). Skin conductance and aesthetic evaluative responses to non representational works of art varying in symmetry. Bull. Psychon. Soc. 26, 355–358. doi: 10.3758/bf03337681 Taylor, R., Micolich, A. P., and Jonas, D. (1999). Fractal expressionism. Physics World 12, 25–28. doi: 10.1088/2058-7058/12/10/21 Langlois, J. H., Roggman, L. A., Casey, R. J., Ritter, J. M., Rieser-Danner, L. A., and Jenkins, V. Y. (1987). Infant preferences for attractive faces: rudiments of a stereotype? Dev. Psychol. 23, 363–369. doi: 10.1037/0012-1649. 23.3.363 Taylor, R. P., Micolich, A. P., and Jonas, D. (2006). Fractal analysis: revisiting Pollock’s drip paintings (Reply). Nature 444, E10–E11. doi: 10. 1038/nature05399 Taylor, R. P., Spehar, B., Van Donkelaar, P., and Hagerhall, C. (2011). Perceptual and physiological responses to Jackson Pollock’s fractals. Front. Hum. Neurosci. 5:60. doi: 10.3389/fnhum.2011.00060 Lawlor, M. (1955). Cultural influences on preference for designs. J. Abnormal Soc. Psychol. 51, 690–692. doi: 10.1037/h0047219 Van den Berg, A. E., and Koole, S. L. (2006). New wilderness in the Netherlands: an investigation of visual preferences for nature development landscapes. Landsc. Van den Berg, A. E., and Koole, S. L. (2006). New wilderness in the Netherlands: an investigation of visual preferences for nature development landscapes. Landsc. Urban Plan. 78, 362–372. doi: 10.1016/j.landurbplan.2005.11.006 Leder, H., Belke, B., Oeberst, A., and Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgments. Br. J. Psychol. 95, 489–508. doi: 10. 1348/0007126042369811 Limbert, W. M., and Polzella, D. J. (1998). Effects of music on the perception of paintings. Empirical Stud. Arts 16, 33–39. doi: 10.2190/v8bl-gbjk-tlfp-r321 Van den Berg, A. E., and Vlek, C. A. (1998). REFERENCES Human physiological benefits of viewing nature: EEG responses to exact and statistical fractal patterns. Nonlinear Dynamics Psychol. Life Sci. 19, 1–12. Pinheiro, J. C., and Bates, D. M. (2000). Mixed-Effects Models in S and S-PLUS. New York, NY: Springer-Verlag. Hagerhall, C. M., Laike, T., Taylor, R. P., Küller, M., Marton, T. P., and Küller, R. (2008). Investigation of EEG response to fractal patterns Perception 37, 1488–1494 doi: 10.1068/p5918 Polzella, D. J. (2000). Differences in reactions to painting by male and female college students. Percept. Mot. Skills 91, 251–258. doi: 10.2466/pms.91.5. 251-258 Halpern, A. R., Ly, J., Elkin-Frankston, S., and O’Connor, M. G. (2008). ‘‘I know what I like’’: stability of aesthetic preference in Alzheimer’s patients. Brain Cogn. 66, 65–72. doi: 10.1016/j.bandc.2007.05.008 Reber, R., Schwarz, N., and Winkielman, P. (2004). Processing fluency and aesthetic pleasure: is beauty in the perceiver’s processing experience? Pers. Soc. Psychol. Rev. 8, 364–382. doi: 10.1207/s15327957pspr0804_3 Segall, M. H., Campbell, D. T., and Herskovits, M. J. (1966). The Influence of Culture on Visual Perception. (USA: The Bobbs-Merrill Company, Inc.). Herzog, T. R., Herbert, E. J., Kaplan, R., and Crooks, C. L. (2000). Cultural and developmental comparisons of landscape perceptions and preferences. Environ. Behav. 32, 323–346. Silverman, I., Choi, J., and Peters, M. (2007). The hunter-gatherer theory of sex differences in spatial abilities: data from 40 countries. Arch. Sex. Behav. 36, 261–268 doi: 10.1007/s10508-006-9168-6 Iwao, S., and Child, I. L. (1966). Comparison of esthetic judgments by American experts and by Japanese potters. J. Soc. Psychol. 68, 27–33. doi: 10. 1080/00224545.1966.9919662 Silverman, I., and Eals, M. (1992). ‘‘Sex differences in spatial abilities: evolutionary theory and data,’’ in The Adapted Mind, J. Barkow, L. Cosmides and J. Tooby (New York, NY: Oxford University Press), 487–503. Iwao, S., Child, I. L., and García, M. (1969). Further evidence of agreement between Japanese and American esthetic evaluations. J. Soc. Psychol. 78, 11–15. doi: 10. 1080/00224545.1969.9922334 Sonnenfeld, J. (1967). ‘‘Environmental perception and adaptation-level in the arctic,’’ in Environmental Perception and Behavior, ed. D. Lowenthal (Chicago: Department of Geography, University of Chicago), 42–59. Joye, Y. (2005). Evolutionary and cognitive motivations for fractal art in art and design education. Int. J. Art Design Educ. 24, 175–185. doi: 10.1111/j.1476-8070. 2005.00438.x Soueif, M. I., and Eysenck, H. J. (1971). Cultural differences in aesthetic preferences. Int. J. Psychol. 6, 293–298. doi: 10.1080/00207597108246695 Joye, Y. (2006). Evolutionary and cognitive speculations for biomorphic architecture. Leonardo 39, 145–152. Zube, E. H., and Pitt, D. G. (1981). Cross-cultural perceptions of scenic and heritage landscapes. Landsc. Plan. 8, 69–87. doi: 10.1016/0304-3924(81) 90041-1 Copyright © 2016 Street, Forsythe, Reilly, Taylor and Helmy. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Frontiers in Human Neuroscience | www.frontiersin.org REFERENCES The influence of planned-change context on the evaluation of natural landscapes. Landsc. Urban Plan. 43, 1–10. doi: 10.1016/s0169-2046(98)00102-9 Lowie, R. H. (1921). A note on aesthetics. Am. Anthropol. 23, 170–174. doi: 10. 1525/aa.1921.23.2.02a00040 Wilson, E. O. (1984). Biophilia: The Human Bond with Other Species. Cambridge, MA: Harvard University Press. Lyons, E. (1983). Demographic correlates of landscape preference. Environ. Behav. 15, 487–511. doi: 10.1177/0013916583154005 Zaimov, K., Kitov, D., and Kolev, N. (1969). Aphasie chez un peintre. Envephale 58, 377–417. Marin, M. M., and Leder, H. (2013). Examining complexity across domains: relating subjective and objective measures of affective environmental scenes, paintings and music. PLoS One 8:e72412. doi: 10.1371/journal.pone.00 72412 Zajonc, R. B. (1968). Attitudinal effect of mere exposure. J. Pers. Soc. Psychol. 9, 1–27. doi: 10.1037/h0025848 Zube, E. H. (1984). Themes in landscape assessment theory. Landsc. J. 3, 104–110. May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org 13 Street et al. A Complex Story Copyright © 2016 Street, Forsythe, Reilly, Taylor and Helmy. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. May 2016 | Volume 10 | Article 213 Frontiers in Human Neuroscience | www.frontiersin.org 14
https://openalex.org/W1972976707
https://www.scielo.br/j/pee/a/xxDhrvWsjqF6rYCtZbsXQFp/?lang=es&format=pdf, https://www.redalyc.org/pdf/2823/282321817016.pdf
es
El lecho de Procusto o la convención sobre la competencia humana
Psicologia Escolar e Educacional
2,011
cc-by
8,524
El lecho de Procusto o la convención sobre la competencia humana Gloria Fariñas León Resumen El tema de las competencias ha ganado un gran espacio de debate acerca del desarrollo humano en la actualidad. Este ha reeditado su introducción por D. McClelland en la Psicología de los años 70 con nuevas acepciones y usos que no siempre se ajustan a las exigencias de desarrollo de la Psicología como ciencia y profesión y tampoco a las demandas del perfeccionamiento de la educación, uno de los terrenos de mayor aplicación del concepto. El presente trabajo se plantea un análisis al respecto y persigue esclarecer las implicaciones conceptuales y prácticas del concepto. Se arriba a la conclusión, con argumentos diversos, de que el concepto de competencia, por una parte, tiene un contenido convencional que obedece a fines utilitarios e instrumentales, y por otra, que su empleo retrasa el desarrollo de la Psicología como ciencia y profesión. No obstante, el concepto puede ser utilizado en determinadas circunstancias, que son esclarecidas también. Las fuentes utilizadas para el análisis fueron las observaciones de la autora y la literatura publicada que conducen, a la vez, a un análisis histórico y cultural del término. Palabras clave: Educación, competencias, desarrollo humano. O leito de Procusto ou a convenção da competência humana Resumo A questão das competências ganhou um espaço importante nos debates atuais sobre o desenvolvimento humano. Estes permitiram reeditar o significado dado a este problema por D. McClelland na Psicologia dos anos setenta, com novos significados e usos que nem sempre estão em conformidade com as exigências do desenvolvimento da Psicologia como ciência e profissão, e também não se apresentam suficientes para responder às demandas da educação, como um dos maiores campos de aplicação do conceito. Este trabalho apresenta uma análise sobre o conceito em questão e procura esclarecer suas implicações do ponto de vista crítico. Argumentos diversos permitem concluir que o conceito de competência apresenta, por um lado possui um conteúdo convencional que atende a fins utilitários e pragmáticos e, por outro lado, seu uso atrasa o desenvolvimento da psicologia como ciência e como profissão. Contudo, tal conceito pode ser utilizado em determinadas circunstâncias que são esclarecidas neste artigo. As fontes utilizadas para a análise consistem nas observações da autora e na literatura publicada, o que conduzem por sua vez a uma análise histórica e cultural do conceito. Palavras-chave: Educação, competências, desenvolvimento humano. The Procustean Bed or the Convention on Human Competence Abstract The issue of competences has won today a major forum for discussion on human development. The theme was reissued by D. Mc Clelland in the Psychology of the seventy, with new meanings and uses that not always conform to the requirements of development of psychology as a science and profession. Neither it conforms to the demands of advanced education, one of the largest fields of application of concept.In this paper we analyse the issue, attempting to clarify theoretical and practical implications of the concept, from a critical standpoint. We found different arguments, about the the term competence: on the hand, it has a conventional content due to instrumental and utilitarian purposes, and on the other hand, its use delays the development of psychology as a science and profession. However, the concept can be used in certain circumstances, which are also clarified. The sources we use for analysis were the author´s observations and the published literatura in the area, leading to a historical and cultural study of the concept. Key words: Education, competences, human development. Revista Semestral da Associação Brasileira de Psicologia Escolar e Educacional, SP. Volume 15, Número 2, Julho/Dezembro de 2011: 341-350. 341 Introducción al problema Adentrarse en los problemas del desarrollo humano en estos tiempos obliga, en cierto modo, al análisis sobre las competencias y la competitividad de las personas en las organizaciones, asimismo, a la consideración simultánea del progreso de la sociedad. A juzgar por la diversa literatura que se publica sobre esta cuestión a escala mundial, parece interesar a muchos investigadores y profesionales, especialmente a los profesores y, entre estos, a aquellos que trabajan en la educación superior. Se trata de una temática que exige una visión interdisciplinaria, para alcanzar una comprensión y explicación de profundidad a fin de ser aprovechada con plena conciencia –cosa que será objeto del presente trabajo–, no por obedecer a modas. Su definición en el campo de la Psicología, corresponde a McClelland, autor estadounidense de orientación conductista social. Los trabajos más conocidos de McClelland trataron lo que él llamó la motivación de logro y la sociedad de logro (1961); después el concepto de competencia (1973), muy coherente con ese tipo de motivación y sus atribuciones en el desenvolvimiento del ser humano. El concepto1, elaborado empíricamente partiendo de una amplia investigación acerca de las personas triunfadoras, denota que el éxito depende más de las características específicas (competencias) de estas, que de sus conocimientos, de su currículum o de los resultados que alcanzan en los tests de rendimiento. Las competencias, son consideradas como conductas necesarias para el desempeño laboral exitoso (o sus indicadores observables), y fueron adoptadas por muchos departamentos de recursos humanos, desde hace algún tiempo, a modo de añadir valor a las empresas, y a la vez, como criterio para seleccionar el personal. También, durante los últimos quince años, las competencias se han convertido en patrón de medida de la calidad de la educación superior (igualmente de la educación general). Por su parte, los diccionarios refieren el término competencia en el lenguaje común, con un contenido bastante similar al asumido por las diferentes instituciones2. El Diccionario de la Real Academia Española (DRAE, 2009) plantea en su primera acepción del concepto competencia: la oposición o rivalidad entre dos o más 1 No es justo olvidar en esta historia a autores como Chomsky (1972), quien utilizó el vocablo para referirse al dominio de la lengua; tampoco a Parsons, quien ya en 1949 elaboró un esquema conceptual que permitía estructurar las situaciones sociales, según una serie de variables dicotómicas. Una de ellas era el concepto de logro, orientado a la valoración de las personas por la obtención de resultados concretos; entre otros autores, que de cierto modo anticiparon el concepto. No obstante, en las publicaciones actuales no siempre refieren estas fuentes. 2 El Ministerio del Trabajo y Seguridad Social de Cuba las considera del siguiente modo: un conjunto de conocimientos teóricos, habilidades, destrezas y actitudes que son aplicados por el trabajador en el desempeño de su ocupación o cargo en correspondencia con el principio de idoneidad demostrada y los requerimientos técnicos, productivos y de servicios, así como los de calidad, que se le exigen para el adecuado desenvolvimiento de sus funciones (1999, artículo3). 342 personas que aspiran a la misma cosa, al que agrega: situación entre empresas que rivalizan en un mercado ofreciendo o demandando el mismo producto. En una segunda acepción, considera: pericia, aptitud, idoneidad para hacer algo o para intervenir en un asunto determinado; igualmente señala: atribución legítima a un juez u otra autoridad para el conocimiento o la resolución de un asunto. El diccionario portugués de Buarque de Holanda (2010), plantea las acepciones en orden inverso y señala como significado: capacidad, habilidad, aptitud, idoneidad. Se observa el vínculo entre competencia y competitividad, ambos tienen la misma raíz lingüística y en la práctica suelen considerarse juntos. Estos diccionarios sólo abordan los usos comunes, mientras que WIKIPEDIA (2010), amplía la acepción refiriéndola a distintos campos de trabajo, lo que considera una vulgarización del término inicial y cotidiano. Estos campos son: el Derecho, la Administración, la Contabilidad y la Psicología. En este último iguala las competencias a las habilidades (Fähigkeiten en alemán y abilities en inglés). Canário (1999), refirió la existencia de ciento veintitrés significados del término, lo que destaca, por una parte, el interés en la expansión del concepto, y por otra, la imprecisión en su utilización. La definición del concepto de competencia, contiene, de forma alternativa o conjunta: habilidades, capacidades, actitudes, entre otras formas de la vida espiritual. Con cierta frecuencia aparece igualado al concepto de habilidad o al de capacidad. Es marcado el interés por su uso globalizado, a pesar de su variado empleo. No parece haber imperado hasta hoy, una torre de Babel tan animada, como la provocada por este tema. Cuando se utiliza el vocablo “macroconcepto” –aún con la borrosidad que admite la “Ciencia de la Complejidad” entre los límites de las diferentes construcciones teóricas–, la mayoría de los interesados sabe de qué se habla, en cambio, no ha ocurrido de modo similar con el de competencia. Tiene valor agregar, que el término ha sido asumido por las organizaciones transnacionales encargadas de la evaluación de los servicios educacionales, lo que ha dado lugar a su difusión como requisito de análisis y de prácticas en este campo. Se trata de un concepto entronizado gracias a la globalización (Ramonet, 1998). La agencia europea Tuning (2007), entre otras, asume el concepto para apoyar las metodologías que emplea en la evaluación y acreditación de los planes y programas de estudio de las universidades iberoamericanas (Tuning AL). También existen Tuning EUROPEA y Tuning EEUU. Esta organización defiende las competencias en el sentido de praxis, para lo que articula teoría y práctica, conocimientos científicos y saberes tácitos provenientes de las experiencias laborales y de vida. Tuning considera anacrónico el contenido de la definición como hacer focalizado en la realización de tareas (principalmente en los campos profesionales y académicos), y lo sustituye por la acción informada y orientada de manera reflexiva. También incluye en el concepto de competencia, diversas dimensiones de la personalidad, con independencia de su grado de esencialidad, mediación y jerarquía en el desarrollo Revista Semestral da Associação Brasileira de Psicologia Escolar e Educacional, SP. Volume 15, Número 2, Julho/Dezembro de 2011: 341-350. del educando. Al respecto plantea la necesidad de “combinar dinámicamente conocimientos, habilidades, capacidades y valores (2007, p.5)” con la intención de convertirlo en una idea máximamente abarcadora, dada la complejidad creciente de la cultura y del desenvolvimiento humano. Merece recordarse, que por detrás de estos planteamientos está el proceso de Bolonia, datado en los años noventa, y revitalizado sistemáticamente en el presente siglo, con las miras de crear un espacio europeo de enseñanza superior cohesionado, competitivo y atractivo para docentes y estudiantes de ese continente y de terceros países. Además, capaz de articular, de algún modo, a los actores entre sí (profesores, profesionales y empleadores) y de promover el empleo de los graduados, a la vez que favorecer la movilidad de estos. Debe reconocerse, que Tuning hizo una gran contribución al sistema europeo sobre la acumulación y transferencia de créditos. Sistema que encuentra en las competencias el denominador común, que sustenta las transacciones crediticias esperadas para captar los profesionales más competentes de diferentes países (Carrington & Detragiache, 1999). Dicha organización subraya, que las competencias están asociadas a la competitividad. Entre las competencias genéricas estandarizadas para América Latina, dicha organización planteó las siguientes en el mismo documento referido anteriormente (no se abordan las competencias específicas, por no considerarse necesarias para el análisis): 1. Capacidad de abstracción, análisis y síntesis. 2. Capacidad de aplicar los conocimientos en la práctica. 3. Capacidad para organizar y planificar el tiempo. 4. Conocimientos sobre el área de estudio y la profesión. 5. Responsabilidad social y compromiso ciudadano. 6. Capacidad de comunicación oral y escrita en la propia lengua. 7. Capacidad de comunicación en un segundo idioma. 8. Habilidades en el uso de las tecnologías (información y comunicación). 9. Capacidad de investigación. 10. Capacidad de aprender y actualizarse permanentemente. 11. Habilidades para buscar, procesar y analizar información de fuentes diversas. 12. Capacidad crítica y autocrítica. 13. Capacidad para actuar en nuevas situaciones. 14. Capacidad creativa. 15. Capacidad para identificar, plantear y resolver problemas. 16. Capacidad para tomar decisiones. 17. Capacidad de trabajo en equipo. 18. Habilidades interpersonales. 19. Capacidad de motivar y conducir (liderazgo) hacia metas comunes. 20. Compromiso con la preservación del medio ambiente. 21. Compromiso con su medio socio-cultural. A convenção da competência humana * Gloria Fariñas León 22. Valoración y respeto por la diversidad y multiculturalidad. 23. Habilidad para trabajar en contextos internacionales. 24. Habilidad para trabajar en forma autónoma. 25. Capacidad para formular y gestionar proyectos. 26. Compromiso ético. 27. Compromiso con la calidad. Al comparar los inventarios de las competencias genéricas elaborados por el proyecto europeo y por el latinoamericano, Tuning observó 22 competencias fácilmente comparables en ambos casos. El proyecto latinoamericano incorpora 3 competencias más: responsabilidad social y compromiso ciudadano, compromiso con la preservación del medio ambiente y compromiso con el medio sociocultural. No fueron consideradas 3 competencias del proyecto europeo en la versión latinoamericana: conocimiento de culturas y costumbres de otros países, iniciativa y espíritu emprendedor y motivación de logro. En el ejemplo que expone Tuning, se iguala la competencia con el conocimiento, la habilidad, la capacidad, el compromiso o los valores, sin aclarar el vínculo intrínseco entre estos para designar distintos aspectos del desarrollo humano. Es el caso de las competencias 4, 5, 6, 8, 26, entre otras del inventario que repiten lo mismo. Tuning busca la integración en el sujeto, a través de cuatro factores. El primer factor, por ejemplo, abarca siete formas de capacidad, una de habilidad, una de conocimiento y el proceso de aprendizaje y similarmente en los siguientes factores. A favor y en contra de las competencias: ¿cuál es la historia? No se puede negar, la ocurrencia y desarrollo de las competencias. La permanencia o quiebra de las empresas depende en determinada medida de la competencia y de la competitividad de sus empleados y directivos. Por otra parte, qué sería de un nadador, de un pelotero, de un futbolista en certámenes nacionales o internacionales, sin ser competente y sin saber competir. El interés en estos casos es la lucha por la calidad y la preponderancia, lo que no significa que todo lo que tiene calidad tenga que haber sido motivado por la competitividad. Las competencias operan, luego existen. No obstante, su conceptuación exige un análisis más profundo, básicamente en cuanto a su origen, naturaleza, implicaciones éticas y necesidades de aplicación. Del mismo modo, la incidencia de su significado en el desarrollo de la ciencia, sobre todo en lo tocante a la concepción del desarrollo humano. La aceptación tácita de los conceptos de competenciacompetitividad, ha ayudado a su establecimiento prioritario, como si fuera obvia su necesidad. Algunos autores han hablado del credo de las competencias. El origen del concepto, hay que buscarlo más allá de las definiciones: en el contexto socioeconómico de su aparición; especialmente 343 en el caso de McClelland, ya que este trata de insertarlo en el terreno de la motivación y del desenvolvimiento humano, en una manera más amplia e influyente que Chomsky. El concepto de competencia está intrínsecamente ligado, por una parte, al de motivación de logro, y por otra, al de sociedad de logro –entendida como sociedad capitalista–, en cuyo seno, la competitividad tiene un peso decisivo. Se considera de este modo, que es la motivación de logro la que anima el desenvolvimiento socioeconómico de una nación determinada. Las épocas en que son acuñados estos conceptos en los campos de la psicología social y de la motivación (década de los sesenta y luego de los setenta) y que alcanzan mayor propagación máxime en la educación (desde los años noventa), coinciden con momentos de altas (1944-1973) y posibles peligros en la economía de los Estados Unidos (desde 1973 hasta la fecha). Al hacer historia, debiera tenerse en cuenta que fue homo habilis la denominación dada por la antropología al ser humano que dominó, por primera vez, la elaboración de herramientas y el desarrollo de las habilidades para emplearlas, tanto en un plano material como espiritual (herramientas mentales). Esta denominación del ser humano como sujeto, tiene un valor histórico al mismo tiempo que cultural, principalmente para la ciencia. Su aceptación ha sido ya centenaria, y tiene mucho que ver con la conformación del concepto de sujeto en la historia. El concepto de competencia es propio de una época bastante posterior a la del homo habilis –la era del homo sapiens sapiens– y se utiliza para denotar una finalidad específica del comportamiento, que no necesariamente deviene general, ni en los sistemas capitalistas renovados en tiempos del neoliberalismo3. El término de habilidad por su origen, historia y extensión de contenido, incluye en sí mismo, el término de competencia4, que es particular de una etapa posterior de la evolución de la sociedad, aunque duradera. Si se hablara en un sentido estrictamente histórico y cultural, no pragmático, la competencia sería un caso particular de las habilidades (aquella movida básicamente por la motivación de logro en la búsqueda de la competitividad) y no a la inversa, como proponen sus divulgadores. Sin embargo, la competencia trata de erigirse en calidad de categoría única e universal, capaz de incluir otros conceptos (valores, conocimientos, habilidades, capacidades, etc.), aún sin considerar la naturaleza específica de cada uno ni establecer un orden jerárquico o histórico entre ellos, respecto a la orientación del desenvolvimiento humano en su carácter integral. Es el caso de Tuning, de amplia geopolítica. Allport (1983, p. 35), llamaba a este gusto ecléctico de mezclar ideas de distinta naturaleza y jerarquía, prácticas “de nido de corneja”, porque 3 Es natural que, el Banco Mundial (2002), esté interesado por el tema de las competencias en la educación, lo que por su contenido, merecería un análisis especial que no es posible hacer en esta oportunidad. Cito especialmente la dirigida a la educación superior, pero existen numerosas publicaciones de esta institución referidas a los distintos niveles de enseñanza. 4 Ocurre de manera similar con este concepto de habilidad, respecto al de competencia en la época del homo sapiens sapiens. 344 estas aves suelen llevar a sus nidos, lo mismo una rama de árbol para alimentar a sus pichones, que un zapato. El término, por la variabilidad de su contenido, ajustable a los fines prácticos específicos, ha devenido una suerte de lecho de Procusto, en recordación al personaje de la mitología griega, que estiraba o serruchaba a los huéspedes que pernoctaban en su posada, para adaptarlos maliciosamente al tamaño de un lecho, que era secretamente regulable. Llama la atención que la idea de shopping center sigue la misma lógica, quiere decir, ser un agregado de cosas diversas, cuya razón de estar juntas no obedece a su naturaleza particular, vínculo esencial o jerárquico, sino a una necesidad práctica, en esta oportunidad: la necesidad de atraer los clientes al comercio, por la ventaja que ocasiona tener, en el mismo lugar: tiendas de vestuario, tiendas de alimentos, cine, teatro, cafeterías, restaurantes y otros lugares que servirían a diversos propósitos, lo que ahorraría tiempo a los clientes y conseguiría mayores dividendos para los dueños de los negocios. Raciocinio cómodo para la vida cotidiana, pero aplicado a la ciencia, banaliza la búsqueda de la esencia de las cosas que esta estudia. Por otra parte, un inventario de competencias o de otras dimensiones -que siga la Lógica Formal como es el caso que se trata-, elude la visión profunda del comportamiento humano. Esto ha sido demostrado por diversos enfoques en la historia de la Psicología, que de alguna forma han buscado una concepción del ser humano, no basada en la integración estadística, sino en la integración de sentido o de sentido-significado de la vida como totalidad (Freud, Vygotski, Frankl, Allport, PichonRivière, entre otros, con independencia del nivel de acierto de cada uno de estos autores). El planteamiento de las competencias recuerda, por contraste, a autores como Catell (1972), cuya propuesta tuvo que recurrir al enfoque factorialista en la concepción de la personalidad (test 16 PF de Catell). A través de la combinación de factores, este autor hizo abstracción de la dialéctica del desarrollo humano con su emergencia característica de formaciones psicológicas, de alta complejidad, condicionadas histórica y culturalmente. Si Tuning se refiriera en su conceptuación a los límites entre los perfiles profesionales, a las responsabilidades de cada persona, a lo que compete hacer o saber a cada cual, sería admisible, además, es uno de los usos comunes tradicionales. A juzgar por el tratamiento metodológico y práctico del concepto, los participantes de esta organización parecen pretender una teoría, sobre el funcionamiento del desarrollo humano y de la personalidad. O lo que es peor, eliminar el concepto de personalidad para sustituirlo por una combinatoria de factores. Esta óptica del asunto aleja a la Psicología de las aspiraciones de la ciencia actual, al proponer un regreso al positivismo (Sousa, 1984), al reduccionismo, en fin, a la simplificación del conocimiento sobre el ser humano y a la reificación de las dinámicas personales e interpersonales. En consecuencia, la vuelta a los mitos (Lukács, 1967). El propio McClelland en el prólogo de su libro intitulado The Achieving Society (1961), alude a su “ingenuidad” respecto a los temas Revista Semestral da Associação Brasileira de Psicologia Escolar e Educacional, SP. Volume 15, Número 2, Julho/Dezembro de 2011: 341-350. históricos, y hasta refiere una anécdota acerca del rechazo que recibió de un colega –a quien reconoce como eminente historiador de Harvard–, al saber este que McClelland había tratado de escapar de todos los cursos de historia. Además, hay que preguntarse si el ser humano debe ser competente y competitivo en toda circunstancia. Igualmente, si todo lo bueno por producir tiene que ser resultado de la competitividad. Es importante reconocer el nexo esencial entre competencia y competitividad, ambos constituyen una unidad conceptual, como zona de desarrollo próximo-situación social del desarrollo. Otra pregunta no menos importante sería: ¿Cuál es entonces el ideal de ser humano explícito o implícito en el enfoque de las competencias?. Es difícil imaginarse a pintores y a compositores de música clásica como profesionales preferiblemente competentes y competitivos, lo que no quiere decir que no sean capaces de competir o de ser competentes en determinadas circunstancias. Resulta difícil también imaginarse a un empresario, de personalidad armónicamente integrada, como sujeto competente y competitivo en todas las esferas de la vida y en todo momento. McClelland interpreta un estudio muy interesante realizado por Berlew (1956), sobre la obra de Homero y Hesíodo en la Grecia Antigua, considerados como historiadores exitosos porque consiguen poner en palabras, lo que está en el espíritu de todos (1961, p. 147). El estudio consistió en analizar en la obra de estos autores, el monto del contenido referente al logro. Este se puntuaba en dos niveles: a) siempre que uno de los personajes de la narración analizada esté empeñado en alguna actividad competitiva la cual vencer o actuar de modo igual o mejor que los otros, lo que es declarado explícitamente, como de importancia primordial. b) Cuando este tipo de actividad esté implícito en las narraciones de los personajes. McClelland comete aquí el mismo error de análisis señalado más arriba, porque absolutiza el contenido relacionado con lo competitivo en el texto, al no tener en cuenta el resto de la actividad cotidiana de los personajes y su engranaje con este. En otras palabras, se marginan contenidos que forman la totalidad del empleo del tiempo junto con la actividad competitiva. La competitividad ocupa todo el lugar. Tampoco es un hecho, el dominio absoluto de la motivación de logro. Además, esta forma de motivación no es necesariamente intrínseca. Es memorable la anécdota del poeta que rompía sus poesías una vez terminadas, porque su recreación espiritual radicaba en componerlas. Habría que cuestionarse la competencia de este sujeto, porque su placer provenía del proceso de la creación, no necesariamente del resultado. La subjetividad humana es muy compleja y alcanza dimensiones de plenitud, no reducibles a una tipología, por ejemplo: ganadores versus perdedores, competentes versus incompetentes. Tampoco a la de resultados versus procesos. La Pedagogía, la Sociología y la Psicología tienen cierto retraso, con respecto a otras disciplinas, en la búsqueda de una concepción general del objeto de estudio para su análisis más cabal y profundo. La Psicología, al igual que las Ciencias Sociales en general, A convenção da competência humana * Gloria Fariñas León precisa explicaciones integradoras y profundas sobre el ser humano y su desenvolvimiento. La motivación de logro opera en conjunción con otras motivaciones, razón para no absolutizarla al evaluar una obra prominente. La pertinencia, relevancia y fama de una creación, no obedece obligatoriamente a la tiranía de la motivación de logro de su autor. Cuando se habla de las competencias, se invoca o se actualiza, de alguna forma, el concepto de sujeto racional cartesiano, poniendo en riesgo la aspiración de un ser humano más pleno y movido por el bien común (Hinkelammert, 2006). Siempre cabría la pregunta acerca de cómo el gusto estético –y su refinamiento– pudiera ser considerado una competencia o alguno de sus componentes. Hasta ahora, esto parece no haber interesado mucho para los efectos actuales, aunque el pragmatismo encontraría alguna razón útil para hacerlo. Cuán difícil resulta imaginarse a cantautores tan relevantes para la música universal como Silvio Rodríguez y Chico Buarque de Holanda, preocupados con su competencia profesional y la competitividad de su obra, que va más allá de los intereses comerciales (Britto, 2005). El ideal sobre el “hombre de éxito”, ha sido analizado desde diversas perspectivas, por la caricaturización que representa del “deber ser” en su consideración más humana. Chaplin (1936) expresó dramáticamente en el film Tiempos Modernos, los efectos enajenantes del taylorismo como práctica organizada del capitalismo en el plano laboral, que sirve de alerta ante sus sucesivos regresos, aunque con otros ropajes. El término competencia, se presenta ahora en una visión dulcificada, como si hubiera perdido su naturaleza intrínseca, más allá del taylorismo, del fordismo, en la época de la globalización más acentuada que haya vivido la humanidad. Marcuse se opone a lo abstracto del pensamiento racionalista cartesiano, que entiende al individuo como sujeto ideal – el hombre unidimensional-, descartando el valor de lo corporal y de lo erótico imprescindibles para hablar del ser humano en lo cotidiano. Este autor remarca, al referirse al “extrañamiento bretchiano”, la posibilidad que este ofrece para que el ser humano reflexione sobre el poder alienante de la sociedad capitalista de su época. Decía: La lucha [...] contra su absorción en la unidimensionalidad predominante, se muestra en los esfuerzos de la vanguardia por crear un distanciamiento que haría la verdad artística comunicable otra vez (1968, p. 96). El hombre unidimensional es víctima de su propia impotencia y de la opresión continua de un complejo método de dominación (Borón, 2003). No por gusto resulta absurda la idea del pintor competente o del directivo competente en todos los momentos de su vida cotidiana. Se está hablando de una misma persona que funciona como totalidad, lo que es propio de las personas sanas, no alienadas. Marcuse señala el distanciamiento del ser humano respecto al statu quo, a través del arte. El arte implica inspiración, espontaneidad, recreación espiritual, no obedece estrictamente a la estructura medio-fin programable, lo que es propio de la racionalidad cartesiana, de la racionalidad instrumental. Esta racionalidad es la que 345 sustenta la teoría de las competencias, no es el caso del arte, a menos que responda a fines comerciales. Para él, es el arte, la salida del hombre unidimensional para vivir mejor. La teoría de las competencias se ha tornado un retroceso en la ciencia. Como ya se apuntó, la vanguardia del pensamiento científico actual busca enfoques integradores y dialécticos. El análisis de cualquier inventario de comportamientos –es el caso Tuning–, se realiza a través de técnicas como los clusters, el análisis factorial y la varianza, en otras palabras, la lógica de las mediaciones entre estos comportamientos, no es analizada teniendo en cuenta el funcionamiento real de la vida del sujeto, sino desde una lógica externa, formal, impuesta por el análisis estadístico. Es cuestionable un enfoque que no contribuya al desarrollo teórico de la ciencia y que solo produzca resultados prácticos. Si los epígonos de las competencias pretenden mostrar que el término es mera convención de fines utilitarios, al menos debieran poner al descubierto la trama histórica y cultural en que se inserta su proposición. La presentación actual de sus ideas y prácticas, parece ser “incauta” ante la realidad de ambos condicionamientos. No es ocioso mencionar que, en no pocas ocasiones, los especialistas que se dedican a la difusión de esta noción y a su práctica, desconocen las corrientes fenomenológicas de pensamiento -menos el enfoque materialista dialéctico, a no ser en su versión vulgar- porque siguen el pensamiento instrumentalista. Otro problema que se adiciona es que lo hacen con plena impunidad. El principio que prima es el utilitarista: si funciona, es cierto. La unidad estadística enfrentada a la unidad de sentido, o mejor dicho, la tercera persona enfrentada al sujeto o al yo. Es el sentido de la vida, el drama de la vida, lo que da unidad al desenvolvimiento humano (Frankl, 1991). Si se lleva el análisis al plano reduccionista, es la motivación de logro lo que da unidad al sujeto competente-competitivo o a sus competencias, no su análisis estadístico. Para poder ver el papel del sentido de la vida, hace falta tener una visión verdaderamente dialéctica del desarrollo de la personalidad, lo que no ocurre por una mera declaración de su aceptación conceptual. Vale recordar que no se ha escuchado otra cosa en los debates sobre las competencias, que los cálculos numéricos de una u otra índole. A lo que habría que agregar, que el lenguaje de las competencias se presta bastante bien para el ejercicio de la ingeniería del comportamiento, manera profesional de manipular a las personas o su control. Una educación que pretenda el desarrollo integral del educando, corre el riesgo de desviar su rumbo si sigue esta concepción, ya que más tarde o más temprano acabará sometiéndose al dictado de la fragmentación del desarrollo humano y de los fines utilitaristas. Una educación sigue sin reparo este dictado, cuando en alguna medida es o ha sido instrumentalista. La teoría de las competencias plantea a la educación, una visión abstracta sobre el ser humano y su desarrollo, al presentar una imagen fraccionada de comportamientos al margen de su unidad de sentido –y del sujeto que lo cohesiona–, no calculable por métodos estadísticos sino por análisis basados en una 346 lógica dialéctica. Esa teoría es igualmente abstracta, por la sobrevaloración de comportamientos (competencias), que forman parte de una unidad de análisis mayor, dado su abarque y esencialidad: el drama humano como totalidad (Politzer, s.f.). El burocratismo encuentra en esta concepción un apoyo a su forma de comprender el desenvolvimiento de las personas en la educación u otros ámbitos, gracias a la afinidad, de base mecanicista entre ambas perspectivas. Si se analiza el concepto de competencia, a la luz del cálculo de los factores que presenta Tuning, la competencia es todo y parte al mismo tiempo, cosa que sería difícil de admitir aún en la lógica formal, que es la base del pensamiento de esta transnacional, como lo fue para McClelland, su impulsor más reconocido en el campo profesional. Además ¿por qué asumir un término tan cargado de una determinada ideología? ¿Por qué esta prepondera? ¿Se trata de desideologizar otras ideologías? La designación del ser humano como homo habilis, homo sapiens u homo sapiens sapiens, nació bastante al margen de este debate entre ideologías para referir un hecho histórico y cultural constitutivo de lo que es el ser humano hoy. A la luz de estas designaciones, el concepto de habilidad es más inclusivo o menos discriminador. Hasta los propios usuarios del término competencia igualan ambos con frecuencia. Los conceptos de ser humano, persona, sujeto y personalidad, educando, profesor, entre otros, son igualmente abarcadores y antecedieron también al concepto de competencia, en la historia de las ciencias sociales. Es posible que los autores del término y sus difusores, planteen la necesidad de que la antropología registre la aparición del “homo compĕtentis”. O tal vez, rechacen el concepto de personalidad porque el examen factorialista del ser humano les impide ver su valor. Corresponde decir entonces, que las hojas de los árboles no dejan ver el bosque. Acerca del lenguaje de la ciencia El taylorismo y el conductismo fueron contemporáneos, y no por casualidad tuvieron un mismo fundamento: el escenario socioeconómico de la primera mitad del siglo XX en Estados Unidos. Ambos perseguían la eficiencia del comportamiento humano para el éxito de la empresa capitalista. El taylorismo consiguió un método de organización del trabajo que permitía el aumento de la productividad mediante la máxima división de funciones, la especialización del trabajador y el control estricto del tiempo necesario para cada tarea. El conductismo demostró experimentalmente, lo que pretendía el taylorismo en la práctica social. Los conductistas han sido perspicaces, porque plantearon a través de la historia de la Psicología, diversas formas válidas para el aprovechamiento del potencial pasivo de la actividad humana (la manipulación del comportamiento), y ante la crítica por la desconsideración del potencial humano activo, encontraron pertinazmente formas tan lineales y mecánicas como las ya conseguidas (condicionamiento operante, variables intermediarias, necesidades de logro, motivación de logro, etc.). Las Revista Semestral da Associação Brasileira de Psicologia Escolar e Educacional, SP. Volume 15, Número 2, Julho/Dezembro de 2011: 341-350. aplicaciones del conductismo son muy conocidas en el campo de la Psicología, la Pedagogía y la Sociología. Es el caso de la Learning Village (Hren, 1978) y de Los Horcones tratada en Walden II por Skinner (1973). Al igual que muchos sistemas de pensamiento social experimental, el conductismo autocumplió su profecía, porque realizó sus fines y asumió este hecho como demostración válida de su ideario, sin someterlo a la crítica ética ni contrastarlo con otras ópticas sobre el desenvolvimiento del ser humano. El conductismo ha sido uno de los enfoques más influyentes en el pensamiento de las ciencias sociales y las ciencias económicas del siglo XX, lo que persiste en el presente siglo. Ha sobrevivido, entre otras razones, por su capacidad mimética o de camuflaje. Esta capacidad ha estado a tono con la crítica de sus opositores, sin embargo, no para cambiar su lógica emblemática. Así surgieron nuevas formas de conductismo que preservaron su esencia nata, asumiendo solo formalmente las críticas cruciales de los otros enfoques. Por ejemplo, el Neoconductismo, el Conductismo Social, el Cognitivismo, etc. De manera similar, han aparecido más recientemente diversas variantes específicas de este mismo raciocinio (“miniteorías” en términos allportianos): las inteligencias múltiples y la inteligencia emocional, que intentan disimular su naturaleza originaria, al punto de parecer teorías humanistas. Otra reedición del conductismo ha sido la concepción de las competencias y la competitividad propuesta por Tuning, analizada en este artículo. El conductismo sigue presente, entre otras razones, gracias a que ha asumido el lenguaje de los enfoques que lo han desacreditado. Así existen teorías conductistas y cognitivistas de la personalidad, pues es sabido que los humanistas norteamericanos, que fundamentaron sus concepciones en enfoques fenomenológicos, le dieron una batida importante. Pero no es propósito del presente trabajo echar por tierra, lo que representó el Conductismo para el establecimiento de la Psicología como ciencia positiva a principios del siglo XX, el problema es otro: su ideal de ser humano y la forma en que pretende establecer determinados cánones, sobre todo para la educación. Todas las versiones conductistas, ponen en riesgo la visión integral y dinámica del drama humano o la mecanizan. Por suerte, la obra de tantos escritores como Martí, Balzac, Dostoiévsky, hacen fehaciente la realidad, de que es el sentido de la vida de un ser humano, enclavado en una cultura y en un periodo histórico determinado, lo que propicia su unidad interna como sujeto o su enajenación cuando se trata de su pérdida, que en buena medida es fragmentación, esquizofrenización. Se ha dicho, y es cierto, que los seres humanos se parecen más a los cuentos que a las cuentas, aunque ellos también inventaron estas últimas. Los valores, los intereses, el gusto, las preferencias, la creatividad, entre otros procesos y fenómenos de la personalidad, no obedecen a programas de entrenamiento. Ellos constituyen un tejido de mediaciones bien complejo, no susceptible de inventariarse, por lo que escapan de alguna manera de la manipulación del otro (padre o madre, profesores, jefes, etc.). El cálculo A convenção da competência humana * Gloria Fariñas León estadístico, utilizados para analizar las competencias (las partes) es desigual al sujeto y a la personalidad (el todo) y a su narrativa. En el inventario de las competencias, subyace la creencia de que las dinámicas psicológicas operan de manera lineal por asociación (combinatoria). Es el eterno problema del reduccionismo, que parece no resolverse, sino reeditarse una y otra vez. El conductismo ha adoptado sistemáticamente nuevas formas para sobrevivir, lo que es lógico porque el capitalismo perdura. De esta manera, los conductistas han asimilado el discurso histórico culturalista, para adaptarlo a su esquema racional; manifestación de mimetismo, que significa la deformación del pensamiento de su opositor. Lo hacen de palabra, no de hecho (Fariñas, 2006). Así han recortado la comprensión del significado de ideas tan importantes como las de zona de desarrollo próximo y de situación social del desarrollo, en este último caso reduciéndolo a su mínima consideración, y con esto, han adoptado formalmente el ideario original. Es la lógica del utilitarismo, la que se impone: tomar lo que sirve a los fines prácticos trazados, no cambiar los fines, que es una de las cosas más criticadas. Lógica también presente en el uso del término competencia. El enfoque histórico cultural demostró la necesidad de la integración de todo su aparato conceptual para conseguir validez en cuanto a la comprensión y explicación del desarrollo humano, con un método más cabal, dialéctico y profundo, cosa no pretendida por el modelo conductista. Es además, el gusto por las palabras, la confianza en ellas como sustitutas de la realidad. Porque se dice algo, ya es, ya existe. Porque se dice, todos lo creen. Lo que sucede es que el camuflaje mencionado solo puede lograr cierta efectividad en el campo del discurso, teniendo en cuenta que siempre hay personas dispuestas a creer en las apariencias. El discurso ha sido aquilatado en demasía por el Posmodernismo (Fariñas, 2006). La promesa no siempre conlleva su cumplimiento, además todo vale. Este camuflaje lingüístico opera como si la comunicación del ser humano pudiera sustituir su actividad, como si el hecho de decirlo o escribirlo en un inventario, lo hiciera operar efectivamente en la realidad. Claro está, es posible pretender la educación de la personalidad desde posiciones conductistas, lo que significaría esquematizar su comprensión. El neoliberalismo necesita un fundamento, de otra manera no operaría de manera eficiente a escala social. Estas formas de operar son las competencias, compatibles con el núcleo ideológico del sistema en que se originó esta concepción en todas sus versiones. Las competencias responden al sistema de pensamiento eficientista sobre el comportamiento del ser humano y su desarrollo. Los compromisos en la burocratización de la educación Los directivos de la educación superior, presionados por las transnacionales (Tuning, etc.) se empeñan en la 347 instauración de diversas modalidades de la universidadempresa (unas son más absolutistas, mientras que otras actúan con cierta reserva logrando ser más sutiles o flexibles). El énfasis en estos casos se dirige a la productividad de las universidades y sus centros a cargo (de investigación, etc.). Productividad que se convierte en productivismo al plantearse, por ejemplo: estándares a alcanzar en detrimento de las consideraciones casuísticas o emergentes del proceso de enseñanza; rutinas fijas en la enseñanza que permiten lograr los “resultados esperados”, al mismo tiempo que violentan la dialéctica natural del aprendizaje en el estudiante; controles sistemáticos hasta el último eslabón del proceso, que pudieran afectar el curso de la dinámica creativa de la educación e invadir la organización del tiempo de profesores y estudiantes; ponderación de las categorías evaluativas que buscan un rasero para la comparación entre instituciones o personas, que necesitan reconocer su originalidad. Igualmente destacable es la intensificación del trabajo, por ejemplo, del profesor sometido a los estándares de horas, cantidades de resultados (artículos, patentes, etc.). en fin, toda una industria “fordista”, para la cual, las competencias constituyen su mejor dictado. Las competencias son congruentes con la preponderancia de la motivación de logro en la sociedad de logro, nada más parecido a lo que se señaló sobre la universidad-empresa (de Sousa, 2006); ellas están concebidas en calidad de unidades programables y medibles, también susceptibles de diversos análisis estadísticos acostumbrados en estos casos. Ahora bien, si no es necesario desterrar el término competencia, tampoco hay que hacerlo con el análisis estadístico. Simplemente debe discernirse, cuándo estos son necesarios verdaderamente y cuándo alteran la naturaleza intrínseca de los fenómenos y procesos a los que se refieren, con el objetivo pragmático de informar lo demandado. La universidad, si se precia de ser una institución de pensamiento, debe dar lugar a debates que amplíen la mirada sobre la realidad, y no casarse con un modo de pensamiento único, sobre todo si este tiende a la estandarización de procesos cuya esencia es heterogénea, única e irrepetible (la identidad y el desarrollo, ya sea personal o grupal, disciplinario, sectorial, institucional, etc.). Este tipo de concepción reifica las relaciones humanas, a la vez que da lugar a métodos de dominación (Kohan, 2003). El ideario humanista cubano concibe el desarrollo del estudiante –a través de la educación– como un sujeto culto, con una vida íntima plena, a la vez que comprometido socialmente, crítico y transformador, lo que quedó reflejado en su Reforma Universitaria de 1962 (Consejo de Universidades). El rescate de la competencia a toda costa No ha faltado quien haya tratado de fundamentar el uso de la teoría de las competencias, a partir de otras con pretensiones más elevadas, no solo prácticas. Y para logarlo ha tenido que repetir el mismo procedimiento señalado en el epígrafe anterior: la adopción del discurso 348 opositor deformándolo en su naturaleza. Esto significa, por una parte, el mimetismo con el nuevo discurso, y por otra, el escamoteo de las conquistas de la perspectiva asumida, con su respectiva deformación. Un enfoque teórico que resulta atractivo para esos fines, es el histórico cultural, porque este ofrece soluciones a las insuficiencias tratadas aquí. Una primera variante pudiera ser la asimilación del concepto de competencia, según la definición dada por Tuning (o por cualquier especialista de las competencias), dentro del enfoque histórico culturalista. Entonces habría que preguntar: ¿por qué asumir el término competencia? ¿Para sustituir sus términos tradicionales? ¿A qué responde esta “modernización”? o tal vez es mejor decir posmodernización, dado el valor que se le atribuye al vocablo como convención susceptible de incluir contenido de diversa índole, siempre que sea útil para los fines de sus usuarios (vale todo). El argumento de esta variante es el afán de que el concepto de competencia domine, en la concepción de la educación, de las organizaciones, etc., con un fundamento teórico que históricamente ha tenido prestigio en el abordaje de las problemáticas del desarrollo humano. Muchos educadores miran con buenos ojos al enfoque histórico cultural, por la posibilidad que este ofrece para superar los problemas de la educación actual. De modo que si se coloca el concepto dentro de esta visión teórica, es posible que gane crédito. La segunda variante de solución, con similares intenciones, es incorporar aspectos del enfoque histórico cultural o la totalidad de este, a la concepción de Tuning (o de otros que estudian las competencias); aunque cualquier forma de esta segunda variante es más difícil de llevar a cabo. La cuestión es, que a causa del mimetismo y del escamoteo de visiones ajenas, se puede esperar cualquier proposición, aunque las consideraciones que sigan en el raciocinio den elementos suficientes sobre la incompatibilidad de la pretendida simbiosis, que no llegaría a ser otra cosa que superposición parasitaria. Por consiguiente, es conveniente seguirle la pista a las consecuencias, que tendría esa asociación y sus opciones básicas de camuflaje. No es casual, que los partidarios de estas ideas propongan también el cambio del nombre original del enfoque, al menos, el más reconocido históricamente, por el de Histórico Social y otras combinaciones5. En algunas de las opciones se elimina el peso de lo cultural, quizá por la incomprensión de que: “…sin hombres no hay cultura por cierto, pero igualmente, y esto es más significativo, sin cultura no hay hombres (Geertz, 2003, p. 72). Al eliminarse la denominación “cultural”, sería posible obviar la crítica a la cultura, especialmente a la cultura dominante (Lukács, 1967). Y liberar la cultura de culpa, permitiría depositar la culpa en el ser humano. Es decir, el ser humano es competitivo por naturaleza, todo aquel que no sepa competir es culpable de su fracaso. Para competir y ser competitivo, hay que tener cierta formación cultural. Y esta es precisamente una de las 5 Las expresiones más largas tienden a rechazarse por un problema de economía en el lenguaje o por un concepto estético. Baste recordar el término biopsicosocial, que desconsidera que lo psicológico incluye lo biológico y lo social. Revista Semestral da Associação Brasileira de Psicologia Escolar e Educacional, SP. Volume 15, Número 2, Julho/Dezembro de 2011: 341-350. cosas que no convendría ser tenida en cuenta en el análisis. La división entre personas exitosas y fracasadas caería entonces en el terreno de la oportunidad, del acceso real al conocimiento que hace falta para desarrollarse como ser humano, en este caso competitivo. La cultura…transforma los dotes naturales en correspondencia con los fines del hombre, decía L. S. Vygotski (1987, p. 163). Lo erróneo en estos casos sería concebir instrumentalmente la cultura, cosa propia de su interpretación comercial; igualmente, reafirmar el gusto por la cultura “light”, propia del facilismo de la educación en estos tiempos de las competencias y del Hombre Unidimensional. Se llama la atención al respecto, porque el enfoque histórico cultural ha tenido diversas interpretaciones positivistas a lo largo de su historia, luego es posible su aceptación dentro de la concepción de las competencias, lo que solo de manera formal, descartando su ética –¿y por qué no también su estética?– o asumiéndola en apariencias. Hay que saber distinguir, cuando se habla en un auténtico lenguaje culturalista y cuando se habla en un lenguaje “culturalista” instrumental (Bruner, 1997). El primero es a la vez, histórico, cosa que no interesa mucho a la cultura empresarial, que reduce la trama universal de la cultura a los aspectos utilitarios. El establecimiento de la unidad competenciacompetitividad, como categoría que fundamenta la educación en el contexto actual de la globalización, persigue una cultura parcializada: la cultura empresarial, separada del complejo patrimonio cultural de la Humanidad, en su sentido amplio. Integrar la categoría “competencia-competitividad” a ese panorama cultural, implica de algún modo, el análisis crítico en el sentido que apuntó Lukács, (1967). Lo que se elige en cambio, es decretar su uso. El planteamiento del enfoque vygotskiano, como histórico social, pudiera servir a los fines de posiciones no progresistas. Por ejemplo, aquellas que culpan a los marginales de su situación social, como si en la sociedad actual no hubiera diferencias de oportunidades en el acceso a la cultura. El condicionante en este caso no es la sociabilidad (o la no sociabilidad), sino la cultura o la incultura. “Ser culto es el único modo de ser libre (…)”, decía Martí (1963, p.289). La sociabilidad no se conquista, se nace con ella, es inalienable. Es la oportunidad de acceder a la cultura, a la educación durante toda la vida, la que permite sostenidamente el desarrollo del ser social. La división de los sujetos en competentes e incompetentes, exitosos y fracasados, tiene en su base, aunque de modo tácito, la desesperanza de “llegar a ser alguien”. El análisis histórico y cultural del desarrollo humano muestra el modo en que el sujeto obra su desarrollo y amplía sus potencialidades, en un proceso eminentemente solidario. No obstante, quedan por ver nuevos mimetismos y escamoteos en ese proceso de adaptación del Conductismo, en su versión de las competencias. Ellos responderán a necesidades utilitarias, que dependen de las necesidades particulares de los especialistas que los pongan en práctica. La torre de Babel promete continuar, ahora con nuevas bases. A convenção da competência humana * Gloria Fariñas León Referencias Berlew, D. (1956). The achievement motive and the growth of Greek Civilization. Em McClelland, D. Em The Achieving Society. Princeton: Van Nostrand. Borón, A. (2003). Tras el búho de Minerva (mercado contra democracia en el capitalismo de fin de siglo). La Habana: Ciencias Sociales. Britto, L. (2005). El imperio contracultural: del rock a la posmodernidad. La Habana: Arte y Literatura. Bruner, J. (1997). La educación, puerta de la cultura. Madrid: Visor. Buarque de Holanda, A. (Ed.). (2010). Dicionario Aurélio da Língua Portuguesa (5a ed.). Curitiba: Editora Positivo. Canário, R. (1999). Educação de Adultos. Um Campo e uma Problemática. Lisboa: Educa. Carrington, W., & Detragiache, E. (1999). How Extensive is the Brain Drain?. Finance & Development, (36). Chaplin, C. (Productor/Escritor/Director). (1936). Modern Times [Película]. Estados Unidos: United Artists. Chomsky, N. (1972). Lingüística cartesiana. Madrid: Gredos. Competência. (2010) Wikipédia, a enciclopédia livre. Recuperado: 11 ago 2011. Disponível: http://pt.wikipedia.org/w/index. php?title=Compet% C3%A Ancia&oldid=22952859 Cuba. (1962). Reforma Universitaria del 10 de enero de 1962. La Habana: Universidad de La Habana. Cuba. (1999). Resolución 21. Ministerio del Trabajo y Seguridad Social del 1º de junio de 1999. Disponível: http://www.cubaindustria. cu/contadoronline/ Perfeccionamiento%20 Empresarial/ Legislaci%C3%B3n/RES-MTSS-21-1999.htm#C01 De Sousa Santos, B. (2006). La Universidad en el siglo XXI. La Habana: Casa de las Américas. Real Academia Española. (2008). Diccionario de la lengua española. Microsoft Student con Encarta Premium [DVD]. Washington: Autor. Fariñas, G. (2006). Psicología, educación y sociedad. La Habana: Félix Varela. Frankl, V. (1991). El hombre en busca de sentido. Barcelona: Herder. Geertz, C. (2003). Impacto del concepto de cultura em el concepto de hombre, En Colectivo de autores (Ed.). Selección de Lecturas de Sociología (p. 72). La Habana: Félix Varela. Hinkelammert, F.J. (2006). El sujeto y la ley. El retorno del sujeto reprimido. La Habana: Caminos. 349 Hren, C. (1978). “La escuela primaria de la Learning Village”. En R. Ulrich y cols. (Orgs.), Control de la conducta humana. (Vol. 3, pp. 164-175). México, Trillas. Kohan, N. (2003). Marx en su (Tercer) Mundo. Hacia uma socialismo no colonizado. La Habana: Ciencias Sociales. Lukács, G. (1967). Existencialismo o Marxismo. Sao Paulo: Senzala. Politzer, G. (s/f). Crítica dos fundamentos da Psicologia. Lisboa: Presença. Ramonet, I. (1998). O Pensamento Único e os Regimes Globalitários. Em J. L. Fiori y cols. (Org.), Globalização: o fato e o mito (pp. 5475). Rio de Janeiro: EDUERJ Sousa Patto, M. H. S. (1984). Psicologia e Ideologia: uma introdução crítica à Psicologia escolar. São Paulo: T. A. Queiroz. Marcuse, H. (1968). El hombre unidimensional. Barcelona: Ariel. Martí, J. (1963). Nuestra América (Tomo 8). La Habana: Editorial Nacional de Cuba. McClelland, D. (1961). The Achieving Society. Princeton: Van Nostrand. McClelland, D. (1973). Testing for Competence Rather Than for Intelligence. American Psychologist, (28), 1-14. Parsons, T. (1937). The Structure of Social Action. New York: MacGraw, Skinner, B.F. (1968). Walden II: una sociedad del futuro. Madrid: Orbis, S.A. Tuning (2007). Reflexiones y perspectivas de la Educación Superior en América Latina Informe Final. Bilbao: Universidad de Deusto. Vygotski, L. S. (1987). Historia del desarrollo de las funciones psíquicas superiores. La Habana: Científico Técnica. World Bank (2002). Constructing Knowledge Societies: New Challenges for Tertiary Education. Washington D.C.: The World Bank. Recebido em 12/10/2011 Aprovado 20/10/2011 Dra. Gloria Fariñas León (glofaleon2009@gmail.com, gloria@rect.uh.cu) Facultad de Psicología. Universidad de La Habana Ave. 15 Num. 6403 e/ 64 y 66. Playa. Ciudad de La Habana. C.P. 12000. Cuba Telf: (537) 881 3315 350 Revista Semestral da Associação Brasileira de Psicologia Escolar e Educacional, SP. Volume 15, Número 2, Julho/Dezembro de 2011: 341-350.
https://openalex.org/W4289914397
https://hal.ehesp.fr/hal-03810615/file/12889_2022_Article_13880.pdf
English
null
Incidence of PTSD in the French population a month after the COVID-19 pandemic-related lockdown: evidence from a national longitudinal survey
BMC public health
2,022
cc-by
9,360
To cite this version: Caroline Alleaume, Patrick Peretti-Watel, François Beck, Damien Leger, Guillaume Vaiva, et al.. Incidence of PTSD in the French population a month after the COVID-19 pandemic-related lock- down: evidence from a national longitudinal survey. BMC Public Health, 2022, 22 (1), pp.1500. ￿10.1186/s12889-022-13880-9￿. ￿hal-03810615￿ Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-03810615 https://ehesp.hal.science/hal-03810615v1 Submitted on 11 Oct 2022 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Alleaume et al. BMC Public Health (2022) 22:1500 https://doi.org/10.1186/s12889-022-13880-9 Abstract Background:  In view of experts’ warnings about the potential negative mental health consequences of the sudden nationwide lockdowns implemented in many countries to limit the spread of the COVID-19 pandemic, we sought to study the incidence of posttraumatic stress disorder (PTSD) after traumatic events related to this unprecedented lockdown in the French general population. Methods:  This longitudinal study among adults (aged =18) consisted of two surveys: the first during the last days of the lockdown and the second a month later. We estimated PTSD incidence with the PCL-5 and ran multiple Poisson regression models to identify factors associated with PTSD. Results:  Among the 1736 participants, 30.1% reported at least one traumatic event. PTSD incidence was 17.5% (95% confidence interval CI = 15.7–19.3). It was higher in participants who reported multiple traumatic events, who had high COVID-19-related media use, who had general anxiety disorder (GAD-7) during the lockdown, and who had GAD, depression (PHQ-9), or sleep problems 1 month later. In addition, 43.1% of people with PTSD reported suicidal thoughts. Conclusions:  These results should help clinicians to target people who are at high risk of developing PTSD after a pandemic-related lockdown and could benefit from preventive measures. Collaboration between the media and mental health professionals could be envisioned to inform the population about care resources. Follow-up recom‑ mendations should also be disseminated to general practitioners to facilitate PTSD screening and ensure that they are aware of the appropriate management. Keywords:  Covid-19, PTSD, Lockdown, PCL-5, COCONEL cohort surveys © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Incidence of PTSD in the French population a month after the COVID‑19 pandemic‑related lockdown: evidence from a national longitudinal survey Caroline Alleaume1*, Patrick Peretti‑Watel1,2, François Beck3, Damien Leger4,5, Guillaume Vaiva6, Pierre Verger1 and the COCONEL Group Caroline Alleaume1*, Patrick Peretti‑Watel1,2, François Beck3, Damien Leger4,5, Guillaume Vaiva6, Pierre Verger1 and the COCONEL Group Introductionh The Coronavirus disease-2019 (COVID-19) emerged in late 2019 in China and spread rapidly in early 2020, and continued to circulate actively for the next 2 years, affect- ing million people worldwide with more than 535 million confirmed cases and leading to 6 million deaths by June 2022 [1]. With France among the European countries most strongly affected by this disease in 2020, French *Correspondence: carolinealleaume@hotmail.fr 1 Southeastern Health Regional Observatory (ORS Paca), Faculté des Sciences Médicales et Paramédicales, 27 Bd Jean Moulin 13385, CEDEX 5 Marseille, France Full list of author information is available at the end of the article In France, the prevalence of probable PTSD was found around 19.5% among university stu- dents 1 month after the COVID-19 lockdown [12], and 21.2% among hospital workers at least 3 months after the lockdown [13]. To our knowledge, no studies have been published yet on the PTSD incidence in the French gen- eral population in relation to COVID-19, although recent studies have shown an increase in the prevalence of COVID-19-associated psychological disorders (anxiety and depression) and sleep difficulties [14, 15]. health authorities, like those of many countries, decided in March 2020 to implement a generalized nationwide lockdown. It started on March 17 and ended on May 10. Experts warned early on about the risks of increased psychological disorders, including posttraumatic stress disorders (PTSD) [2, 3]. Previous studies from China showed that PTSD was one of the most prevalent long- term psychiatric disorders among severe acute res- piratory syndrome (SARS) survivors [4]. In view of the unprecedented magnitude of the COVID-19 pandemic, even as early as April, authors recommended investiga- tion of PTSD symptoms in the general population and among health professionals [3]. Since, literature has tended to confirm experts’ fears, with variations by coun- try, epidemic intensity and specific population groups [5–13]. A month after the lockdown began, PTSD preva- lence was 2.7% among home-isolated Chinese university students [6] and 6.1% among the Chinese population aged 17-63, with youth, women, and people with respon- sibilities and concern for others more vulnerable to these symptoms [8]. PTSD estimates in Europe include 28% among adult Italian COVID-19 survivors 1 month after hospital discharge [9], 27.5% in the Italian general popu- lation during the first month of lockdown [10], and 15.8% in the Spanish general population during the equivalent period there [11]. In France, the prevalence of probable PTSD was found around 19.5% among university stu- dents 1 month after the COVID-19 lockdown [12], and 21.2% among hospital workers at least 3 months after the lockdown [13]. To our knowledge, no studies have been published yet on the PTSD incidence in the French gen- eral population in relation to COVID-19, although recent studies have shown an increase in the prevalence of COVID-19-associated psychological disorders (anxiety and depression) and sleep difficulties [14, 15]. France. The sample was selected using the quota method from an online research panel of more than 750,000 households, developed and maintained by IFOP (Poll- ing Institute, Paris, France). A quota sampling method was applied to obtain a sample representative of the adult general population in France for age, gender, occupation, and rural/urban residence. To limit selection bias, pan- elists with low response rates (i.e., aged between 18 and 24 years old, blue-collar workers, and intermediate occu- pations) were oversampled relative to the others. Finally, 2003 panelists participated in the first survey and, among them, 1736 (86.7%) responded to the second (see Fig. 1). Fig. 1  Flowchart for the enrolment and follow-up of subjects in this ­study1. 1The first sample (Mat 7-11, 2020 survey respondents) has been constituted using the quota method fi To expand our knowledge of COVID-19’s conse- quences on mental health in France, we conducted the COCONEL cohort survey to 1) quantify the incidence of PTSD in the general population, 2) characterize the rela- tive frequency of the PTSD symptoms observed, 3) study the sociodemographic, economic, and COVID-19-re- lated factors, including media use to obtain information about the pandemic, associated with PTSD onset, and 4) study PTSD’s comorbidity with generalized anxiety and depression. © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Alleaume et al. BMC Public Health (2022) 22:1500 Alleaume et al. BMC Public Health (2022) 22:1500 Alleaume et al. BMC Public Health (2022) 22:1500 Page 2 of 11 health authorities, like those of many countries, decided in March 2020 to implement a generalized nationwide lockdown. It started on March 17 and ended on May 10. Experts warned early on about the risks of increased psychological disorders, including posttraumatic stress disorders (PTSD) [2, 3]. Previous studies from China showed that PTSD was one of the most prevalent long- term psychiatric disorders among severe acute res- piratory syndrome (SARS) survivors [4]. In view of the unprecedented magnitude of the COVID-19 pandemic, even as early as April, authors recommended investiga- tion of PTSD symptoms in the general population and among health professionals [3]. Since, literature has tended to confirm experts’ fears, with variations by coun- try, epidemic intensity and specific population groups [5–13]. A month after the lockdown began, PTSD preva- lence was 2.7% among home-isolated Chinese university students [6] and 6.1% among the Chinese population aged 17-63, with youth, women, and people with respon- sibilities and concern for others more vulnerable to these symptoms [8]. PTSD estimates in Europe include 28% among adult Italian COVID-19 survivors 1 month after hospital discharge [9], 27.5% in the Italian general popu- lation during the first month of lockdown [10], and 15.8% in the Spanish general population during the equivalent period there [11]. Design and sampleh The COCONEL cohort survey used two successive sur- veys, one taken during the last days of the first French lockdown (May 7-11, 2020) and the other about a month later (June 3-10, 2020) among the same representative sample of adults (18+ years old) residing in mainland Fig. 1  Flowchart for the enrolment and follow-up of subjects in this ­study1. 1The first sample (Mat 7-11, 2020 survey respondents) has been constituted using the quota method Fig. 1  Flowchart for the enrolment and follow-up of subjects in this ­study1. 1The first sample (Mat 7-11, 2020 survey respondents) has been constituted using the quota method Alleaume et al. BMC Public Health (2022) 22:1500 Page 3 of 11 Exposure to a traumatic event during the lockdown f d h l Exposure to a traumatic event during the lockdown Experience of a traumatic event during the lockdown was documented during both surveys: respondents were asked whether they had faced such an event (yes/ no); those who answered yes were asked if the event was associated with some of the government announcements, their work situation, a relative infected by the corona- virus, travel, conflict with a member of the household, and/or the death of a family member not due to Covid- 19 (several answers possible). The repetition of these questions in the second survey allowed construction of a variable measuring the dynamics of the perception of the traumatic event during the lockdown as follows: we defined “acute stress” as reporting a traumatic event dur- ing the lockdown in the first survey only; and “persis- tent stress” as reporting a traumatic event in the second survey, whether participants had or had not reported an event during the first survey. Finally, we collected participants’ gender, age, region of residence, education level, financial (level of income measured as equivalized household income (EHI) quar- tiles) and work situations before and during the lock- down, household composition, overcrowded housing (defined as a living area <  18 square meters per person or < 25 sq.m. for a single person), and the lockdown’s per- ceived impact on the household’s financial situation. Data analysis h d We weighted data so that the structure of the weighted sample matched that of the French population for age, gender, occupation, and population density in the region of residence. All analyses were performed with these weights. Using Student t-tests and proportion compari- son tests, we first calculated PTSD scores (global and by cluster) and estimated PTSD incidence by type of traumatic event experienced. We then compared PTSD incidence according to individual characteristics among two different groups: the entire sample, and those who reported a traumatic event in either survey (prerequi- site for measuring PTSD). We first used Chi-2 tests to explore associations between individual characteristics and PTSD incidence and then tested them in multivariate analyses. Variables significant at p < 0.20 in the univari- ate analyses were eligible for the multivariate models. To obtain relative risks, we performed two multiple modi- fied Poisson regression models: 1/ in the entire popula- tion, and 2/ only among those with a traumatic event, Data collected PTSD instrumenti about to suicidal thoughts, was also analyzed separately to study the association between PTSD and suicidal thoughts, which is frequently highlighted in the literature [20]. We assessed sleep disorders at follow-up with the following item: “Have you had sleep problems during the last 8 days: not at all/yes, a few/yes, a lot” [14, 21]. We identified PTSD with the PTSD Check List Scale for DSM-5 (PCL-5), a 20-item self-report instrument, validated in French [16] and corresponding to DSM-5 symptom criteria for PTSD [17]. Because the first PTSD criterion requires exposure to a traumatic (very stress- ful) situation at least 1 month before the occurrence of symptoms, we used the PCL-5 during the second survey only. The PCL-5 includes four subscales describing the clusters in the DSM-5 corresponding to reexperiencing, avoidance, negative cognition and mood, and arousal. Responses to each item are collected with Likert scales ranging from 0 (not at all) to 4 (extremely), with a total score ranging from 0 to 80. High scores indicate high PTSD levels. We measured PTSD with the following pre- viously published approach [17]: score = 2 on at least one item of the reexperiencing subscale, one of the avoidance subscale, two of the negative cognition and mood sub- scale, and two of the arousal subscale. A cutoff score of 33 was also applied to measure PTSD incidence [17] to ensure comparability with other studies. During the first survey, participants were also asked how much they worried about becoming infected (score from 0, not worried at all, to 10, very worried) and, in the second survey, if they had been diagnosed with COVID-19, and if any relatives had been admitted to an intensive care unit due to the disease. The first sur- vey also addressed participants media consumption for information about COVID-19 during the lockdown, ask- ing them how many hours per day (less than 30 minutes, 30 minutes to 1 hour, 1-2 hours, 2-3, 3-4, 4-5, > 5 hours) they had spent looking at information about it from five different media sources (television, radio, newspaper, online websites, and social media) in the past week; we constructed a media exposure indicator according to a previously published method [15]. As answers to these five items were positively correlated (Cronbach’s alpha: 0.80) we summed them to obtain a score, and we used its fourth quartile as an indicator of high media. Other measures and collected information In addition, we used the Patient Health Questionnaire-9 (PHQ-9, 9 items) and the General Anxiety Disorder-7 (GAD-7, 7 items) to screen for prevalent depression or generalized anxiety disorder (GAD) during the lockdown (first survey) and during the 2 weeks before the second survey [18, 19]. We used a cutoff point of 5 on the PHQ-9 to identify individuals with probable at least mild depres- sion [18], and a cutoff point of 5 on the GAD-7 to identify probable at least mild GAD [19]. The last PHQ-9 item, Alleaume et al. BMC Public Health (2022) 22:1500 Alleaume et al. BMC Public Health (2022) 22:1500 Page 4 of 11 general adult population (Table  2) PTSD in incidence was 17.5% (95% CI = 15.7–19.3) with significantly higher rates among people reporting several types of events (p < 0.001, Fig. 1). The latter group was also more likely to meet each of the symptom subgroup criteria (Table  1). The cutoff score of 33 used in other publications [19] yielded a lower PTSD incidence estimate: 14.3% (95% CI = 12.7–16.0). to estimate the PTSD risks associated with the type of traumatic event and the persistence of stress between the two surveys. We checked the risk of multicollinearity and the potential endogeneity bias with a Heckman selection model. Finally, only significant factors are shown, except for gender, which we used as a basic adjustment variable. All statistical analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC). PTSD incidencei Incidence of PTSD did not differ according to gender (Table 2) but it was significantly higher in young people, those with low EHI, those confined with people other than their partner, and those who had never worked, both in the entire sample and among those exposed to traumatic events. PTSD incidence was higher in people reporting chronic health problems or a history of con- sultation for psychological disorders than among those who did not. During the first and/or second surveys, 1046 participants reported a traumatic experience during the lockdown: 34.5% only during the first survey, 5.5% during the sec- ond, 18.1% during both, 41.9% none, see Additional file 1: Figure  1). Most respondents reported several types of events (55.5%), 14.9% only events associated with some government announcements, and 11.5% only with their work situation (Fig. 2). Among people who reported several types of events, 72.3% included government announcements.i PTSD incidence was also higher in those diagnosed with COVID-19 (50.1%) than in those who were not (16.6%) (Table  2). Similarly, incidence was higher in Among these 1046 participants, 30.1% were classified with PTSD 1 month after the lockdown ended. In the Fig. 2  Types of stressful traumatic events reported by participants during the lockdown and PTSD incidence in each group (COCONEL, N = 1046). Population: Respondents to the May and June COCONEL surveys who reported a traumatic event during the lockdown (N = 1046). Note: Among participants who reported a traumatic event, 14.9% related it to governmental announcements only. Among the latter, the incidence of PTSD was 16.2%. The sum of the categories displayed is 98.3%; the missing 1.7% are people who selected open-ended responses that were too heterogeneous for analysis. Among those reporting several types of events, 72.3% selected at least government announcements, 62.3% their work situation, 52.7% travel, 47.7% conflict with a household member, 32.6% a relative infected with COVID-19, and 27.9% at least the death of a relative not due to COVID-19 Fig. 2  Types of stressful traumatic events reported by participants during the lockdown and PTSD incidence in each group (COCONEL, N = 1046). Population: Respondents to the May and June COCONEL surveys who reported a traumatic event during the lockdown (N = 1046). Note: Among participants who reported a traumatic event, 14.9% related it to governmental announcements only. Among the latter, the incidence of PTSD was 16.2%. PTSD incidencei The sum of the categories displayed is 98.3%; the missing 1.7% are people who selected open-ended responses that were too heterogeneous for analysis. Among those reporting several types of events, 72.3% selected at least government announcements, 62.3% their work situation, 52.7% travel, 47.7% conflict with a household member, 32.6% a relative infected with COVID-19, and 27.9% at least the death of a relative not due to COVID-19 Fig. 2  Types of stressful traumatic events reported by participants during the lockdown and PTSD incidence in each group (COCONEL, N = 1046). Population: Respondents to the May and June COCONEL surveys who reported a traumatic event during the lockdown (N = 1046). Note: Among participants who reported a traumatic event, 14.9% related it to governmental announcements only. Among the latter, the incidence of PTSD was 16.2%. The sum of the categories displayed is 98.3%; the missing 1.7% are people who selected open-ended responses that were too heterogeneous for analysis. Among those reporting several types of events, 72.3% selected at least government announcements, 62.3% their work situation, 52.7% travel, 47.7% conflict with a household member, 32.6% a relative infected with COVID-19, and 27.9% at least the death of a relative not due to COVID-19 Fig. 2  Types of stressful traumatic events reported by participants during the lockdown and PTSD incidence in each group (COCONEL, N = 1046). Population: Respondents to the May and June COCONEL surveys who reported a traumatic event during the lockdown (N = 1046). Note: Among participants who reported a traumatic event, 14.9% related it to governmental announcements only. Among the latter, the incidence of PTSD was 16.2%. The sum of the categories displayed is 98.3%; the missing 1.7% are people who selected open-ended responses that were too heterogeneous for analysis. Among those reporting several types of events, 72.3% selected at least government announcements, 62.3% their work situation, 52.7% travel, 47.7% conflict with a household member, 32.6% a relative infected with COVID-19, and 27.9% at least the death of a relative not due to COVID-19 Alleaume et al. Discussionh people very concerned about developing this infection than in those who were not and increased significantly with media use. Finally, people who lived in an area strongly affected by COVID-19 had PTSD more often than the others. This study aimed to investigate the incidence of PTSD in the French population 1 month after the end of the lockdown period. Six respondents in ten reported trau- matic experiences during this period, and most of them reported several types of events. Among those with at least one traumatic event, 30.1% had PTSD 1 month after the lockdown ended, while the incidence was 17.5% (95% CI = 15.7–19.3) in the general population. The risk of PTSD in the general population was higher in people with high COVID-19-related media use, with at least mild anxiety during the lockdown, and diagnosed with COVID-19. PTSD was also strongly comorbid with anxi- ety, depression, and sleep problems at follow-up. Among people reporting a traumatic event, young age and expo- sure to multiple types of traumatic events were associ- ated with increased PTSD risk. PTSD incidence was significantly higher in partici- pants with anxiety or depression, identified in either survey (Table  2). It was also significantly positively associated with sleep problems and suicidal thoughts at follow-up; only 7.0% of people without PSTD reported suicidal thoughts during the previous 2 weeks versus 43.1% of those with PTSD (p < 0.001). Finally, among people who had experienced a traumatic event, PTSD incidence was significantly higher in respondents reporting persistent stress due to this event than in those reporting acute stress only. PTSD incidencei BMC Public Health (2022) 22:1500 Page 5 of 11 Page 5 of 11 Table 1  Incidence of PTSD and of PTSD Symptom Clusters by Type of Traumatic Event Experienced during Lockdown (PCL-5, COCONEL 2020, N = 1046)a,b a Population: Respondents to the COCONEL May and June surveys who reported a traumatic event during the lockdown (N = 1046) b Government announcements group, work situation group and travel group were separately compared to the multiple-events group c Among people who reported a traumatic event related to government announcements, 43.6% met the reexperiencing group criteria and 16.2% the criteria for PTSD; these rates were lower than among people who reported several traumatic events (40.4% of them met full PTSD criteria, p < 0.001). The “Other” group included people reporting traumatic events related to a relative infected by COVID-19, or to a conflict with a household member, or to the death of a relative not related to COVID-19 All types of events (n=1046) Several events (n = 581)c Government announce-ments (n = 156) Work situation (n = 120) Travel (n = 55) Other (n = 134) % % % p % p % p % p PTSD 30.1 40.4 16.2 < 0.001 23.7 < 0.001 20.7 < 0.001 11.2 < 0.001 Reexperiencing symptom group 49.2 58.7 43.6 < 0.001 41.9 < 0.001 32.0 < 0.001 28.3 < 0.001 Avoidance symptom group 41.7 52.8 29.5 < 0.001 33.4 < 0.001 29.2 < 0.001 20.4 < 0.001 Negative cognition and mood symptom group 47.6 57.7 35.1 < 0.001 43.3 0.004 33.5 < 0.001 27.9 < 0.001 Arousal symptom group 47.4 57.2 35.8 < 0.001 38.9 < 0.001 34.5 < 0.001 31.6 < 0.001 Table 1  Incidence of PTSD and of PTSD Symptom Clusters by Type of Traumatic Event Experienced during Lockdown (PCL-5, COCONEL 2020, N = 1046)a,b Table 1  Incidence of PTSD and of PTSD Symptom Clusters by Type of Traumatic Event Experienced during Lockdown (PCL-5, COCONEL 2020 N 1046)a,b Factors associated with PTSD incidence: multiple regressionshi Several studies have measured PTSD incidence dur- ing this pandemic, with various instruments and defini- tions. Some measured acute stress, e.g., at the epidemic peak [5–8, 11, 22]. Others measured PTSD in specific groups: people aged 18-30 [22], or hospitalized for COVID-19 [9], or healthcare workers [7, 13]. Some of these samples showed marked selection [6, 9, 12, 13]. These methodological variations make comparisons between studies difficult, especially given the differ- ent cultural contexts. These studies report PTSD inci- dence after the lockdown ranging from 4.6 to 31.8%. Our definition of PTSD followed the American Psychi- atric Association (APA) guidelines rather than using the more common global score approach. This choice should have improved our screening, because we used each PTSD symptom cluster to identify PTSD, taking their different specificity values into account [23]. The first multiple regression model (Model 1, Table  3) showed that high media use during the lockdown, at least mild anxiety during the lockdown, and COVID-19 infec- tion were significantly associated with an increased risk of PTSD 1 month later. Conversely, low media use (< 1 hr./day) during lockdown was negatively associated with PTSD. Sleep problems, at least mild symptoms of depres- sion or GAD at follow-up were also positively associated with PTSD. Among people reporting traumatic events during the lockdown, Model 2 found a significantly higher risk of PTSD among respondents younger than 35 years, and for those with multiple traumatic events, or a work-related traumatic event, or traveling. Page 6 of 11 Alleaume et al. Factors associated with PTSD incidence: multiple regressionshi BMC Public Health (2022) 22:1500 Alleaume et al. BMC Public Health (2022) 22:1500 Table 2  (continued) a Population: Respondents to the COCONEL May and June surveys, N = 1736 b Combination of information collected in the first and the second survey Table 2  (continued) Characteristics of the respondents Incidence of PTSD Among people exposed to a traumatic event (n = 1046) Among the entire sample (n = 1736) % p % p No (n = 1030) 27.6 15.5 Diagnosed with Covid-19 < 0.001 < 0.001 Yes (n = 44) 62.0 50.1 No (n = 1692) 28.9 16.6 Relative admitted to an intensive care unit due to Covid-19 0.79 ** 0.005 Yes 31.2 27.0 No 29.9 16.8 Media consumption per day of information about COVID-19 during the lockdown < 0.001 < 0.001 Low (< 1 hr) (n = 415) 17.5 8.5 Intermediate (1- < 4 hrs) (n = 871) 22.4 12.8 High (= 4 hrs) (n = 450) 50.1 34.9 Serious concern about being infected by COVID-19 < 0.001 < 0.001 Yes (n = 179) 42.8 29.5 No (n = 1557) 26.9 15.0 Mental health symptoms Anxiety (mild to severe) during the lockdown < 0.001 < 0.001 Yes (n = 773) 43.3 35.3 No (n = 963) 8.0 3.2 Depression (mild to severe) during the lockdown < 0.001 < 0.001 Yes (n = 776) 42.6 33.5 No (n = 960) 10.9 4.5 Anxiety (mild to severe) at follow-up < 0.001 < 0.001 Yes (n = 745) 47.6 36.8 No (n = 991) 6.7 2.9 Depression (mild to severe) at follow-up < 0.001 < 0.001 Yes (n = 209) 47.0 37.5 No (n = 813) 8.1 3.5 Suicidal thoughts at follow-up < 0.001 < 0.001 Yes (n = 231) 67.0 56.6 No (n = 1505) 21.2 11.5 Sleep problems at follow-up < 0.001 < 0.001 No (n = 654) 15.6 6.3 A few (n = 794) 27.4 17.7 A lot (n = 288) 52.8 42.2 Perception of the traumatic events < 0.001 – Acute stress (n = 632) 22.3 – Persistent stress (n = 414) 41.5 – Incidence of PTSD Acute stress (n = 632) Persistent stress (n = 414) a Population: Respondents to the COCONEL May and June surveys, N = 1736 b Combination of information collected in the first and the second survey imbalance has also been observed for the lockdown’s impact on anxiety, depression, and sleep problems [14, 15]. Factors associated with PTSD incidence: multiple regressionshi BMC Public Health (2022) 22:1500 Table 2  Incidence of PTSD According to Sociodemographic Characteristics, Medical and Psychological History, Types of Stressors, and Psychiatric Comorbidity (COCONEL, N = 1736)a Characteristics of the respondents Incidence of PTSD Among people exposed to a traumatic event (n = 1046) Among the entire sample (n = 1736) % p % p All (%) 30.1 – 17.5 – Sociodemographic characteristics Gender 0.78 0.13 Men (n = 826) 30.5 16.0 Women (n = 910) 29.7 18.8 Age 0.008 < 0.001 < 35 years (n = 446) 37.0 23.2 35–64 years (n = 863) 28.2 16.3 > 64 years (n = 427) 25.7 13.9 Education Level 0.02 0.14 No high school diploma (n = 880) 34.1 18.7 High school diploma (n = 312) 25.7 14.7 Undergraduate degree (n = 244) 30.9 20.1 Postgraduate degree (n = 300) 23.1 14.5 EHI < 0.001 < 0.001 Low (q1) (n = 373) 41.5 23.9 Middle (q2-q3) (n = 823) 28.5 16.6 High (q4) (n = 371) 23.0 12.9 Missing (n = 169) 27.9 17.5 Financial difficulties due to the lockdown < 0.001 < 0.001 Yes (n = 365) 45.4 32.5 No (n = 1371) 24.7 13.5 Confined in an overcrowded dwelling < 0.001 < 0.001 Yes (n = 134) 45.5 31.8 No (n = 1602) 28.5 16.3 Condition of lockdown < 0.001 < 0.001 Alone (n = 375) 28.5 15.6 With one’s partner (n = 1102) 26.9 15.5 With people other than one’s partner (n = 259) 44.0 28.7 Work situation before the ­lockdownb ns 0.49 # 0.07 Work (n = 909) 30.7 18.8 Unemployed (n = 97) 32.1 20.8 Retired (n = 506) 26.2 13.3 Student (n = 102) 31.8 20.1 Other inactive (n = 122) 36.1 19.9 Health conditions reported during the lockdown Chronic health problem or chronic disease < 0.001 < 0.001 Yes (n = 502) 37.5 23.5 No, don’t know (n = 1234) 26.7 15.0 Consulted for psychological issues in the 12 months before lockdown < 0.001 < 0.001 Yes (n = 191) 48.2 41.1 No (n = 1545) 26.6 14.6 Covid-19 exposure Living in an area strongly impacted by Covid-19 0.04 0.007 Y ( 706) 33 4 20 4 Table 2  Incidence of PTSD According to Sociodemographic Characteristics, Medical and Psychological History, Types of Stressors, and Psychiatric Comorbidity (COCONEL, N = 1736)a Characteristics of the respondents Incidence of PTSD Table 2  Incidence of PTSD According to Sociodemographic Characteristics, Medical and Psychological History, Types of Stressors, and Psychiatric Comorbidity (COCONEL, N = 1736)a Incidence of PTSD Among people exposed to a traumatic event (n = 1046) Among the entire sample (n = 1736) % p % p 30.1 – 17.5 – 0.78 0.13 30.5 16.0 29.7 18.8 0.008 < 0.001 37.0 23.2 28.2 16.3 25.7 13.9 0.02 0.14 34.1 18.7 25.7 14.7 30.9 20.1 23.1 14.5 < 0.001 < 0.001 41.5 23.9 28.5 16.6 23.0 12.9 27.9 17.5 < 0.001 < 0.001 45.4 32.5 24.7 13.5 < 0.001 < 0.001 45.5 31.8 28.5 16.3 < 0.001 < 0.001 28.5 15.6 26.9 15.5 44.0 28.7 ns 0.49 # 0.07 30.7 18.8 32.1 20.8 26.2 13.3 31.8 20.1 36.1 19.9 < 0.001 < 0.001 37.5 23.5 26.7 15.0 < 0.001 < 0.001 48.2 41.1 26.6 14.6 0.04 0.007 33.4 20.4 Health conditions reported during the lockdown Chronic health problem or chronic disease Page 7 of 11 Page 7 of 11 Alleaume et al. Factors associated with PTSD incidence: multiple regressionshi This association did not appear in this study of PTSD incidence, most likely because our inclusion of anxiety imbalance has also been observed for the lockdown’s impact on anxiety, depression, and sleep problems [14, 15]. This association did not appear in this study of PTSD incidence, most likely because our inclusion of anxiety Most recent review study have highlighted that health professional was concerned but also COVID-19 patients and general population [24]. Studies have found higher PTSD risks in women than men [6–12]; this gender Alleaume et al. BMC Public Health (2022) 22:1500 Page 8 of 11 Table 3  Factors Associated with PTSD in the General Population (Model 1) and among the Population Exposed to a Traumatic Event (Model 2): Multiple Modified Poisson Regression Models (COCONEL 2020, N = 1736) a Population: Respondents to the COCONEL May and June surveys, N = 1736 b Covariates included in the model were selected in an automatic selection process, Stepwise option (p threshold: 5%) c NI: Covariate not included in the model d Other included: conflict with a relative, relative infected by the coronavirus, death of a relative and other not classified *p < 0.5; ** p < 0.1; ***p < 0.001 Characteristics of the respondents Model 1. PTSD among the general population Model 2. Factors associated with PTSD incidence: multiple regressionshi PTSD among the exposed population RR Adjusted RR 95% CI RR Adjusted RR 95% CI Gendera (ref: Women) Men 0.85 1.06 0.85–1.32 1.03 1.04 0.85–1.27 Agea (ref: 35–64) < 35 1.43* 1.24 0.99–1.55 1.31* 1.22 1.00–1.48 > 64 0.85 0.96 0.70–1.33 0.91 1.05 0.77–1.43 Media consumption of COVID-19-related information (ref: 1- < 4 hrs) Low (< 1 hr) 0.53** 0.67 0.47–0.95 0.58** 0.67 0.49–0.93 High (= 4 hrs) 2.31*** 1.53 1.18–1.97 1.88*** 1.40 1.11–1.78 Anxiety (mild to severe) during the lockdown (ref: No) Yes 11.20** 3.26 1.95–5.46 1.62 1.93 1.22–3.06 Diagnosed with COVID-19 (ref: No) Yes 3.01*** 1.43 1.13–1.80 2.15*** 1.16 0.91–1.47 Anxiety (mild to severe) at follow-up (ref: No) Yes 12.61*** 3.02 1.77– 5.17 7.13*** 2.62 1.61–4.27 Depression (mild to severe) at follow-up (ref: No) Yes 10.74*** 2.44 1.53–3.90 5.79*** 1.94 1.27–2.97 Sleep problems at follow-up (ref: Yes, a few/No) Yes, a lot 3.35*** 1.51 1.22–1.88 2.26*** 1.36 1.12–1.65 Persistent perceived stress (ref: No) Yes NIc – 1.86*** 1.13 0.93–1.38 Types of traumatic event (ref: ­Otherd) Work situation NI – 2.12 2.19 1.14–4.21 Government announcements NI – 1.45 1.31 0.65–2.62 Travel NI – 1.86 2.17 1.08–4.36 Several of the items proposed NI – 3.62*** 2.40 1.36–4.25 Table 3  Factors Associated with PTSD in the General Population (Model 1) and among the Population Exposed to a Traumatic Event (Model 2): Multiple Modified Poisson Regression Models (COCONEL 2020, N = 1736) activities. All these factors could have made them more vulnerable to traumatic events. and depressive disorders in our multivariate analyses captured at least part of the gender effect (Table 3). Our finding that young people exposed to traumatic events during the lockdown are at higher risks of PTSD is con- sistent with previous findings [10, 22, 23]. Young people, especially those with precarious jobs, may suffer more than other population segments from the pandemic’s direct economic consequences, as during earlier health crises  [24, 25]. The health risks might have compro- mised their education, and possibly their entry into the labor market. Factors associated with PTSD incidence: multiple regressionshi This expression might be applied to our findings as we await further studies to clar- ify the traumatic nature of each reported event during COVID-19-related lockdowns. Nevertheless, the daily presentation of the death toll in the media might suffice to meet the criterion of proximity to death required by DSM-5, and forced isolation, deprivation of liberty, the loss of a job and income could be viewed as social vio- lence. Further research, particularly clinical, is needed to confirm these hypotheses and to explore in more detail the nature of the trauma experienced by people with PTSD a month after the COVID-19-related nationwide lockdown. Follow-up of these people for months will be key the media showed during the COVID-19 health crisis (e.g., patients in intensive care and daily recitals of the number of new and total deaths). Our findings support the recommendation by Olagoke et  al. [26] that pub- lic health professionals should work with the media to provide more content about mental health resources in pandemic situations, especially during lockdowns when people are more highly exposed to media coverage than usual. Moreover, our results suggest that probable GAD during the lockdown was predictive of PTSD a month later; evidence indicating that individuals with a history of psychological disorders are at higher risk of PTSD is now supported by several post-COVID-lockdown studies [10, 11]. The media could participate in prevention pro- grams to encourage people with anxiety symptoms dur- ing a lockdown to seek care. Trained health professionals could thus provide individual PTSD prevention care.h p p The collection of data in two waves a month apart allowed us to explore the persistence of the perception of stress related to traumatic events. While acute stress was perceived more frequently than persistent stress, the latter was more highly correlated to incident PTSD, although no longer significantly after comorbidity factors were included in the model. PTSD–depression comor- bidity has frequently been noted in the literature, among military personnel (exposed to combat), victims of sexual assault [30], and even students during the COVID-19 epidemic  [6]. Previous population-based studies have also highlighted PTSD–anxiety comorbidity [28]. Simi- larly, our results about sleep disorders are consistent with earlier findings of serious sleep problems in PTSD patients [10, 29], including in a recent Chinese study during its lockdown [8]. As with comorbid depression, the causal relation between sleep disorders and PTSD is complex and partly reciprocal. Factors associated with PTSD incidence: multiple regressionshi They may also be more highly exposed to stressful information than the rest of the general popula- tion [26] and more vulnerable to aspects of the lockdown, including isolation, social distancing, the closing of places young people gather to socialize, and reduced outdoor As in recent studies [5, 9, 12, 15, 25], PTSD risk was higher in participants with direct exposure to COVID- 19, especially in those diagnosed with it, and remarkably in people indirectly exposed via high media use. Con- versely, low media exposure was associated with a low PTSD risk. This result adds to previous observations of a positive relation between media exposure to informa- tion about this illness and psychological distress, anxi- ety, and depression [15, 26]. Ahern et al. [27] observed a similar relation between PTSD occurrence and media use after the World Trade Center attacks of September 11, 2001, although the nature and intensity of the trauma (e.g., viewing defenestration) differed greatly from what Alleaume et al. BMC Public Health (2022) 22:1500 Page 9 of 11 context. Risk was highest for those reporting several sources of traumatic events. Government announce- ments alone were not associated with PTSD risk in the multivariate analysis, but were rather mostly cited by people exposed to several types of events during the lock- down. Therefore, in addition to other stressful events, these announcements may have played a role in augment- ing the risk. The DSM-5 definition of trauma, however, requires “actual or threatened death, serious injury, or sexual violence”: proximity to death and/or physical vio- lence was infrequent in the traumatic events reported in our study and is reflected in the low prevalence of intru- sion symptoms in our results (Additional file 1: Table 1). Nonetheless, the DSM-5 definition has been controver- sial [31, 32], with several studies reporting that events classically considered nontraumatic (such as losing one’s job) are nevertheless associated with higher rates of PTSD than events considered traumatic [32]. Our find- ings suggest that the combination of several perceived traumatic events — even if they rarely involved proximity to death or violence — could expose people at high risk to PTSD in situations such as a pandemic-related lockdown. North et al. [23] suggested a new PTSD-like syndrome, resulting from a “nontraumatic” stressor, could be named “post-stressor stress disorder”. Funding h The survey was supported by grants from the Agence Nationale de la Recherche (ANR-20-COVI-0035-01). Availability of data and materials Data cannot be shared publicly because they are confidential; they include sensitive data such as health information. Data are available from the Aix Marseille Université, IRD, AP-HM, SSA, VITROME Institutional Data Access after agreement from the University Hospital Méditerranée Infection Ethic Com‑ mitee (contact: IHU – Méditerranée Infection 19-21 Boulevard Jean Moulin 13005 Marseille, or pierre-​edoua​rd.​fourn​ier@​univ-​amu.​fr, with the number #2020-018) for researchers who meet the criteria for access to confidential data. In conclusion, this study is the first to document with a longitudinal design the incidence of PTSD in the French population, a month after the first COVID-19 lockdown ended. It contributes to the demonstration of the psychological impact of the pandemic in the French population and suggests the need for more psychologi- cal support and a nationwide mental health promo- tion program in the general population and in specific groups. PTSD prevention and treatment should focus especially on young people with a history of anxiety, those who spend substantial time following COVID- 19 in the media, and those with comorbidities. Rec- ommendations should be developed for GPs about the detection of PTSD and how to deal with probable cases. Consortium name 2. Boyraz G, Legros DN. Coronavirus disease (COVID-19) and traumatic stress: probable risk factors and correlates of posttraumatic stress disor‑ der. J Loss Trauma. 2020;25(6-7):503–22. https://​doi.​org/​10.​1080/​15325​ 024.​2020.​17635​56. The COCONEL Group includes the following collaborators: Patrick Peretti- Watel1,2 (scientific coordinator), Valérie ­Seror2, Sébastien ­Cortaredona2, Odile ­Launay7,8, Jocelyn ­Raude9, Pierre ­Verger1,5, Caroline ­Alleaume1, Lisa ­Fressard1 (research consortium), Guillaume ­Vaiva6, François ­Beck3, Stéphane ­Legleye3, Damien Léger4,5, Olivier L’Haridon10, Jeremy K. ­Ward2,11 (Steering Committee). 7Inserm, F-CRIN I REIVAC, Toulouse, France. 8Inserm CIC 1417; Univ Paris, Faculté de médecine Paris Descartes; AP-HP, hôpital Cochin; Paris, France. 9EHESP School of Public Health, Rennes, France. 10Univ Rennes, CNRS, CREM UMR 6211, Rennes, France. 11GEMASS, CNRS, Université Paris Sorbonne, Paris, France. 3. Dutheil F, Mondillon L, Navel V. PTSD as the second tsunami of the SARS- Cov-2 pandemic. Psychol Med. 2020:1–2. https://​doi.​org/​10.​1017/​S0033​ 29172​00013​36. 4. Mak I, Chit W, Chu CM, Pan PC, Yiu MGC, Chan VL. Long-term psychiatric morbidities among SARS survivors. Gen Hosp Psychiatry. 2009;31(4):318– 26. https://​doi.​org/​10.​1016/j.​genho​sppsy​ch.​2009.​03.​001. p g j g pp y 5. Gao J, Zheng P, Jia Y, Chen H, Mao Y, Chen S, et al. Mental health problems and social media exposure during COVID-19 outbreak. PLoS One. 2020;15(4):e0231924. https://​doi.​org/​10.​1371/​journ​al.​pone.​02319​24. Factors associated with PTSD incidence: multiple regressionshi Although nightmares of the traumatic event are included in the DSM-5 diagnos- tic criteria for PTSD, sleep disorders may be both a risk factor and an outcome of PTSD [10, 30]. Doctors should be aware that people with anxiety and/or depression symptoms and/or sleep problems, even some time after lockdown, may also suffer from PTSD and should thus routinely screen for it. Clinicians could systematically use tools -e.g. the Short PTSD Rating Interview (SPRINT)- to investigate PTSD in people with these symptoms. More- over, the strong association we observed between PTSD and suicidal thoughts underlines the potential suicidal risk in patients with PTSD. Previous findings have shown that this risk is non-negligible in people with PTSD [20], especially those with psychiatric comorbidity. Assessing the suicide risk in caring for patients with PTSD after COVID-19 lockdowns is essential.i y This study has some limitations. First, because the lockdown obviously affected data collection activities, online surveys were used to administer questionnaires. While effective, online surveys may involve some bias, in view of the risk that segments of the general popula- tion might be missed. The risk is nonetheless limited, given that 89% of French households have internet access, according to a 2018 estimate [33]. Moreover, the invitation email did not mention the study topic, which may have limited potential selection bias due to non- observed factors. While the PCL-5 used to define PTSD is a well-established and often used instrument, it is not a diagnostic instrument; the lack of clinical assessment is a limit of our population-based study. Psychiatric examinations are necessary in patients whose PTSD has More generally, our findings raise the question of the nature of traumatic events in a pandemic containment Alleaume et al. BMC Public Health (2022) 22:1500 Alleaume et al. BMC Public Health (2022) 22:1500 Page 10 of 11 Page 10 of 11 been detected by the PCL-5. This should be organized in cohort studies to follow patients for sufficient time to study the course of PTSD, detect cases occurring some- time after the lockdown, and evaluate the impact of the second wave of COVID-19. References 1. WHO Coronavirus (COVID-19) Dashboard, https://​covid​19.​who.​int/?​mapFi​ lter=​deaths, consulted on 19.06.2022. 1. WHO Coronavirus (COVID-19) Dashboard, https://​covid​19.​who.​int/?​mapFi​ lter=​deaths, consulted on 19.06.2022. 1. WHO Coronavirus (COVID-19) Dashboard, https://​covid​19.​who.​int/?​mapFi​ lter=​deaths, consulted on 19.06.2022. Author details 1 1 Southeastern Health Regional Observatory (ORS Paca), Faculté des Sciences Médicales et Paramédicales, 27 Bd Jean Moulin 13385, CEDEX 5 Marseille, France. 2 Aix-Marseille University, IRD, AP-HM, SSA, VITROME, Marseille, France. 3 CESP, University Paris Sud, Faculté de médecine UVSQ, Inserm, University Paris-Saclay, Villejuif, France. 4 Université de Paris, EA 7330 VIFASOM (Vigilance Fatigue Sommeil et Santé Publique), Paris, France. 5 APHP- Hôtel-Dieu, Centre du sommeil et de la Vigilance, Paris, France. 6 U1172 INSERM Lille Neuro‑ sciences & Cognitions, Centre National de Ressources & Résilience pour les psychotraumatismes (Lille – Paris), Lille, France. Additional file 1: Figure 1. Time that the event was reported, during the lockdown (first survey) or/and one month later (second survey), and incidence of PTSD. Table 1. PTSD score according to population char‑ acteristics among those who experienced a traumatic event during the lockdown (COCONEL 2020, N = 1046). Acknowledgements In the authors, the COCONEL Group includes: Patrick Peretti-Watel (VITROME, Marseille; ORS PACA), Valérie Seror (VITROME, Marseille), Sébastien Cortare‑ dona (VITROME, Marseille), Odile Launay (Inserm, F-CRIN I REIVAC, Toulouse; Inserm CIC 1417, Paris), Jocelyn Raude (EHESP, Rennes), Pierre Verger Watel (VITROME, Marseille; ORS PACA), François Beck (CESP, Villejuif), Stéphane Legleye (CESP, Villejuif), Olivier L’Haridon (CREM UMR6211 Rennes), and Jeremy Ward (VITROME, Marseille; GEMASS, Paris). Finally, many thanks also go to Jo Ann Cahn for revising the English manuscript. Received: 9 May 2021 Accepted: 25 July 2022 Received: 9 May 2021 Accepted: 25 July 2022 Supplementary Information The authors declare that they have no competing interests. The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s12889-​022-​13880-9. The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s12889-​022-​13880-9. The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s12889-​022-​13880-9. Ethics approval and consent to participate The ethics review board of the University Hospital Institute Méditerranée Infection approved the study design (#2020-018). Respondents gave their informed consent to participate to our survey by clicking on the participation link on the invitation mail. This email was sent to adults who already gave their informed consent to be registered on a research panel developed and main‑ tained by IFOP (Polling Institute, Paris, France). They were not compensated to participate. The survey research firm IFOP (Paris, France) is responsible for the consent procedure according to the European regulation, the General Data Protection Regulation (GDPR)). All methods were carried out in accordance with relevant guidelines and regulations. Abbreviations PTSD P Abbreviations PTSD: Post-traumatic Stress Disorder; GAD: Generalized Anxiety Disorder; EHI: Equivalized Household Income. Abbreviations PTSD: Post-traumatic Stress Disorder; GAD: Generalized Anxiety Disorder; EHI: Equivalized Household Income. Authors’ contributionsi p g j p 6. Tang W, Hu T, Hu B, Jin C, Wang G, Xie C, Chen S, Xu J. Prevalence and cor‑ relates of PTSD and depressive symptoms one month after the outbreak of the COVID-19 epidemic in a sample of home-quarantined Chinese CA wrote the first draft and carried out the statistical analyses. PPW, FB, DL, GV and PV critically revised the manuscript. The COCONEL Group designed the surveys. All the authors contributed to and approved the final manuscript. Alleaume et al. BMC Public Health (2022) 22:1500 Page 11 of 11 Page 11 of 11 university students. J Affect Disord. 2020;274:1-7. https://​doi.​org/​10.​ 1016/j.​jad.​2020.​05.​009. university students. J Affect Disord. 2020;274:1-7. https://​doi.​org/​10.​ 1016/j.​jad.​2020.​05.​009. 25. Taylor MR, Agho KE, Stevens GJ, Raphael B. Factors influencing psycholog‑ ical distress during a disease epidemic: data from Australia’s first outbreak of equine influenza. BMC Public Health. 2008;8:347. https://​doi.​org/​10.​ 1186/​1471-​2458-8-​347. 7. Cheng P, Xu LZ, Zheng WH, Ng RMK, Zhang L, Li LJ, Li WH. Psychometric property study of the posttraumatic stress disorder checklist for DSM-5 (PCL-5) in Chinese healthcare workers during the outbreak of corona virus disease 2019. J Affect Disord. 2020;277:368-74. https://​doi.​org/​10.​ 1016/j.​jad.​2020.​08.​038. 26. Olagoke AA, Olagoke OO, Hughes AM. Exposure to coronavirus news on mainstream media: The role of risk perceptions and depression. Br J Health Psychol. 2020;25(4):865-874. https://​doi.​org/​10.​1111/​bjhp.​12427. 8. Jiang Hj, Nan J, Lv Zy, Yang J. Psychological impacts of the COVID-19 epidemic on Chinese people: Exposure, post-traumatic stress symptom, and emotion regulation. Asian Pac J Trop Med. 2020;13:252-9. https://​doi.​ org/​10.​4103/​1995-​7645.​281614. 8. Jiang Hj, Nan J, Lv Zy, Yang J. Psychological impacts of the COVID-19 epidemic on Chinese people: Exposure, post-traumatic stress symptom, and emotion regulation. Asian Pac J Trop Med. 2020;13:252-9. https://​doi.​ org/​10.​4103/​1995-​7645.​281614. 27. Ahern J, Galea S, Resnick H, Kilpatrick D, Bucuvalas M, Gold J, Vlahov D. Television images and psychological symptoms after the September 11 terrorist attacks. Psychiatry. 2002;65(4):289-300. https://​doi.​org/​10.​1521/​ psyc.​65.4.​289.​20240. g 9. Mazza MG, De Lorenzo R, Conte C, Poletti S, Vai B, Bollettini I, Melloni EMT, Furlan R, Ciceri F, Rovere-Querini P; COVID-19 BioB Outpatient Clinic Study group, Benedetti F. Anxiety and depression in COVID-19 survivors: Role of inflammatory and clinical predictors. Brain Behav Immun. 2020;89:594-600. https://​doi.​org/​10.​1016/j.​bbi.​2020.​07.​037. g 9. Mazza MG, De Lorenzo R, Conte C, Poletti S, Vai B, Bollettini I, Melloni EMT, Furlan R, Ciceri F, Rovere-Querini P; COVID-19 BioB Outpatient Clinic Study group, Benedetti F. Authors’ contributionsi Anxiety and depression in COVID-19 survivors: Role of inflammatory and clinical predictors. Brain Behav Immun. 2020;89:594-600. https://​doi.​org/​10.​1016/j.​bbi.​2020.​07.​037. 28. Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB. Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry. 1995;52(12):1048-60. https://​doi.​org/​10.​1001/​archp​syc.​1995.​03950​24006​ 6012. 29. Vaiva G, Jehel L, Cottencin O, Ducrocq F, Duchet C, Omnes C, Genest P, Rouillon F, Roelandt JL. Prévalence des troubles psychotraumatiques en France métropolitaine [Prevalence of trauma-related disorders in the French WHO study: Santé mentale en population générale (SMPG)]. Encephale. 2008;34(6):577-83. French. https://​doi.​org/​10.​1016/j.​encep.​ 2007.​11.​006. 10. Forte G, Favieri F, Tambelli R, Casagrande M. COVID-19 Pandemic in the Italian Population: Validation of a Post-Traumatic Stress Disorder Questionnaire and Prevalence of PTSD Symptomatology. Int J Environ Res Public Health. 2020;17(11):4151. https://​doi.​org/​10.​3390/​ijerp​h1711​4151. 10. Forte G, Favieri F, Tambelli R, Casagrande M. COVID-19 Pandemic in the Italian Population: Validation of a Post-Traumatic Stress Disorder Questionnaire and Prevalence of PTSD Symptomatology. Int J Environ Res Public Health. 2020;17(11):4151. https://​doi.​org/​10.​3390/​ijerp​h1711​4151. 11. González-Sanguino C, Ausín B, Castellanos MÁ, Saiz J, López-Gómez A, Ugidos C, Muñoz M. Mental health consequences during the initial stage of the 2020 Coronavirus pandemic (COVID-19) in Spain. Brain Behav Immun. 2020;87:172-6. https://​doi.​org/​10.​1016/j.​bbi.​2020.​05.​040. 30. Wright KM, Britt TW, Bliese PD, Adler AB, Picchioni D, Moore D. Insomnia as predictor versus outcome of PTSD and depression among Iraq combat veterans. J Clin Psychol. 2011;67(12):1240-58. https://​doi.​org/​10.​1002/​jclp.​ 20845. 12. Wathelet M, Fovet T, Jousset A, et al. Prevalence of and factors associ‑ ated with posttraumatic stress disorder among French university students 1 month after the COVID-19 lockdown. Translational Psychiatry. 2021;11:327. https://​doi.​org/​10.​1038/​s41398-​021-​01438-z. 31. Van Hooff M, McFarlane AC, Baur J, Abraham M, Barnes DJ. The stressor Criterion-A1 and PTSD: a matter of opinion? J Anxiety Disord. 2009;23(1):77-86. https://​doi.​org/​10.​1016/j.​janxd​is.​2008.​04.​001. 32. Pai A, Suris AM, North CS. Posttraumatic Stress Disorder in the DSM-5: Controversy Change and Conceptual Considerations. Behavioral Sci‑ ences. 2017;7(4):7. https://​doi.​org/​10.​3390/​bs701​0007. 13. Fournier A, Laurent A, Lheureux F, Ribeiro-Marthoud MA, Ecarnot F, Binquet C, Quenot JP. Impact of the COVID-19 pandemic on the mental health of professionals in 77 hospitals in France. PLoS One. 2022;17(2):e0263666. https://​doi.​org/​10.​1371/​journ​al.​pone.​02636​66. 33. Statistica. 2020. Available at: https://​fr.​stati​sta.​com/​stati​stiqu​es/​509227/​ menage-​franc​ais-​acces-​inter​net, consulted on 23.09.2020. 14. Beck F, Léger D, Fressard L, Peretti-Watel P, Verger P; Coconel Group. Covid-19 health crisis and lockdown associated with high level of sleep complaints and hypnotic uptake at the population level. J Sleep Res. 2021;30(1):e13119. https://​doi.​org/​10.​1111/​jsr.​13119. Publisher’s Note S i N i Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. 15. Peretti-Watel P, Alleaume C, Léger D, Beck F, Verger P; COCONEL Group. Anxiety, depression and sleep problems: a second wave of COVID-19. Gen Psychiatr. 2020;33(5):e100299. https://​doi.​org/​10.​1136/​ gpsych-​2020-​100299. 16. Ashbaugh AR, Houle-Johnson S, Herbert C, El-Hage W, Brunet A. Psychometric Validation of the English and French Versions of the Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5). PLoS One. 2016;11(10):e0161645. https://​doi.​org/​10.​1371/​journ​al.​pone.​01616​45. 17. Weathers FW, Litz BT, Keane TM, Palmieri PA, Marx BP, Schnurr PP. The PTSD checklist for DSM-5 (PCL-5). 2013. Scale available from the National Center for PTSD at: http://​www.​ptsd.​va.​gov. 17. Weathers FW, Litz BT, Keane TM, Palmieri PA, Marx BP, Schnurr PP. The PTSD checklist for DSM-5 (PCL-5). 2013. Scale available from the National Center for PTSD at: http://​www.​ptsd.​va.​gov. 18. Kroenke K, Spitzer RL. The PHQ-9: A New Depression Diagnostic and Severity Measure. Psychiatric Annals. 2002;32(9):509-15. https://​doi.​org/​ 10.​3928/​0048-​5713-​20020​901-​06. 18. Kroenke K, Spitzer RL. The PHQ-9: A New Depression Diagnostic and Severity Measure. Psychiatric Annals. 2002;32(9):509-15. https://​doi.​org/​ 10.​3928/​0048-​5713-​20020​901-​06. 19. Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assess‑ ing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092-7. https://​doi.​org/​10.​1001/​archi​nte.​166.​10.​1092. 19. Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assess‑ ing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092-7. https://​doi.​org/​10.​1001/​archi​nte.​166.​10.​1092. g 20. Panagioti M, Gooding P, Tarrier N. Post-traumatic stress disorder and suicidal behavior: A narrative review. Clin Psychol Rev. 2009;29(6):471-82. https://​doi.​org/​10.​1016/j.​cpr.​2009.​05.​001. 20. Panagioti M, Gooding P, Tarrier N. Post-traumatic stress disorder and suicidal behavior: A narrative review. Clin Psychol Rev. 2009;29(6):471-82. https://​doi.​org/​10.​1016/j.​cpr.​2009.​05.​001. 21. Parkerson GRJr, Broadhead WE, Tse CK. The Duke Health Profile; A 17-item measure of health and dysfunction. Med Care. 1990;28:1056-72. 21. Parkerson GRJr, Broadhead WE, Tse CK. The Duke Health Profile; A 17-item measure of health and dysfunction. Med Care. 1990;28:1056-72. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Publisher’s Note S i N i Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 22. Liu CH, Zhang E, Wong GTF, Hyun S, Hahm HC. Factors associated with depression, anxiety, and PTSD symptomatology during the COVID-19 pandemic: Clinical implications for U.S. young adult mental health. Psychiatry Res. 2020;290:113172. https://​doi.​org/​10.​1016/j.​psych​res.​2020.​ 113172. 23. North CS, Suris AM, Davis M, Smith RP. Toward validation of the diagnosis of posttraumatic stress disorder. Am J Psychiatry. 2009;166(1):34-41. https://​doi.​org/​10.​1176/​appi.​ajp.​2008.​08050​644. 24. Yunitri N, Chu H, Kang XL, Jen HJ, Pien LC, Tsai HT, Kamil AR, Chou KR. Global prevalence and associated risk factors of posttraumatic stress disorder during COVID-19 pandemic: A meta-analysis. Int J Nurs Stud. 2022;126:104136. https://​doi.​org/​10.​1016/j.​ijnur​stu.​2021.​104136.
https://openalex.org/W4287312631
http://aminbiol.com.ua/images/Journal/2022/1/AB_2022_24_1_5_Khalak.pdf
Ukrainian
null
Selection criteria for highly productive breeding boars and sows by fattening and meat qualities of their offspring using some evaluation indexes
Bìologìâ tvarin/Biologìâ tvarin
2,022
cc-by
4,328
Критерії відбору високопродуктивних кнурів-плідників і свиноматок за відгодівельними і м’ясними якостями їх потомства з використанням деяких оціночних індексів В. І. Халак1, О. М. Жукорський2, О. М. Церенюк3 v16kh91@gmail.com 1Державна установа «Інститут зернових культур НААН України», вул. Володимира Вернадського, 14, м. Дніпро, 49009, Україна 2Національна академія аграрних наук України, вул. Михайла Омеляновича-Павленка, 9, м. Київ, 901010, Україна 3Інститут свинарства і агропромислового виробництва НААН, вул. Шведська Могила, 1, м. Полтава, 36013, Україна Мета роботи — дослідити відгодівельні і м’ясні якості молодняку свиней великої білої породи та визначити критерії відбору високопродуктивних кнурів-плідників і свиноматок з використанням деяких оціночних індексів. Установлено, що молодняк свиней великої білої породи підконтрольної популяції за віком досягнення живої маси 100 кг переважає мінімальні вимоги класу еліта на 6,13%, товщиною шпику на рівні 6–7 грудних хребців — 32,25% і дов­ жиною охолодженої туші — 3,92%. Вірогідна різниця між тваринами різної внутріпородної диференціації за індексом Тайлера (І і ІІІ групи) встановлено за середньодобовим прирос­ том живої маси за період контрольної відгодівлі (91,7 г), віком досягнення живої маси 100 кг (7,3 доби), товщиною шпику на рівні 6–7 грудних хребців (3,4 мм), довжиною охолодженої туші (1,7 см), довжиною беконної половини охолодженої туші (2,7 см). З урахуванням внутрішньопородної диференціації за індексом Сазера-Фредіна встановлено, що молод­ няк свиней ІІ групи переважав ровесників І групи за середньодобовим приростом живої маси за період контрольної відгодівлі на 4,03%, віком досягнення живої маси 100 кг — 3,67% і довжиною охолодженої туші — 0,2%. Коефіцієнт парної кореляції між відгодівельними і м’ясними якостями молодняку свиней та оціночними індексами коливається у межах від –0,710 (tr=10,24) до +0,844 (tr=20,98). Максимальну прибавку додаткової продукції одержано від молодняку свиней І піддослідної групи за індексом Тайлера (+5,03%) та ІІ під­ дослідної групи за індексом Сазера-Фредіна (+1,12%). Вартість додаткової продукції, яку одержано від молодняку свиней зазначених груп, за умови, що ціна реалізації за 1 кг живої маси молодняку свиней на час проведення дослідження дорівнювала 47,5 грн. становить +247,51 і +55,42 грн./гол відповідно. Таким чином, критерієм відбору високопродуктивних кнурів-плідників і свиноматок за відгодівельними і м’ясними якостями є межі індексу Тайлера 214,89–242,85, індексу Сазера-Фредіна — –3,018…–0,102 бала. Ключові слова: молодняк свиней, порода, відгодівельні і м’ясні якості, індекс, економічна ефективність, мінливість, кореляція Результати дослідження відгодівельних і м’ясних якостей молодняку свиней сучасних порід, а також тварин, одержаних на основі промислового схрещу­ вання і гібридизації, свідчать про високий рівень їх фенотипового прояву з урахуванням умов утримання і годівлі та їх відповідності фізіологічній нормі [2, 4, 5, 11, 16]. За даними Храмкової О. М. https://doi.org/10.15407/animbiol24.01.034 UDC 636.082.12.636.4 Received 10.12.2021 ▪ Accepted 13.01.2022 ▪ Published online 01.04.2022 https://doi.org/10.15407/animbiol24.01.034 UDC 636.082.12.636.4 Received 10.12.2021 ▪ Accepted 13.01.2022 ▪ Published online 01.04.2022 https://doi.org/10.15407/animbiol24.01.034 UDC 636.082.12.636.4 Received 10.12.2021 ▪ Accepted 13.01.2022 Ключові слова: молодняк свиней, порода, відгодівельні і м’ясні якості, індекс, економічна ефективність, мінливість, кореляція Результати й обговорення Аналіз результатів свідчить, що середньо­добовий приріст живої маси молодняку свиней піддослідної групи (n=51) за період контрольної відгодівлі стано­ вить 792,4±6,37 г (Сv=5,75%), вік досягнення живої маси 100 кг — 178,0±0,68 доби (Сv=2,76%), товщина шпику на рівні 6–7 грудних хребців — 21,0±0,26 мм (Сv=8,93%), довжина холодженої туші — 96,8±0,22 см (Сv=1,65%), довжина беконної половини охолодженої туші — 86,1±0,37 см (Сv=3,12%). Мета роботи — дослідити відгодівельні і м’ясні якості молодняку свиней великої білої породи різної внутрішньопородної диференціації за індексами Тай­ лера і Сазера-Фредіна, розрахувати рівень кореляцій­ них зв’язків між ознаками та економічну ефективність результатів досліджень. Індекс Тайлера і Сазера-Фредіна коливаються у ме­жах від 178,89 до 242,85 і від –3,018 до +3,427 бала відповідно. Результати дослідження відгодівельних і м’ясних якостей молодняку свиней великої білої породи різної внутрішньопородної диференціації за індексом Тайлера та Сазера-Фредіна наведено в табл. 1 і 2. Критерії відбору високопродуктивних кнурів-плідників і свиноматок за відгодівельними і м’ясними якостями їх потомства з використанням деяких оціночних індексів [6] встановлено, що кращими відгодівельними якостями характеризу­ ється молодняк свиней, одержаний від свиноматок генотипу F1 (Лі×Йі) і кнурів синтетичної лінії MaxGrow. Живої маси 100 кг вони досягали за 171,6 діб, що, порівняно з ровесниками великої білої породи, менше на 14,9 діб або 7,99% (Р≤0,001). Порівняно з ровес­ никами синтетичних ліній MaxTer і OptiMus, різниця становить 1,3–10,5 доби або 0,75–6,12%. Також усі тварини, отримані від кнурів зарубіжної селекції, 34 Біологія тварин, 2022, т. 24, № 1 Selection of breeding boars and sows by fattening and meat qualities of their offspring Khalak V. I., Zhukorsky O. M., Tserenyuk O. M. Khalak V. I., Zhukorsky O. M., Tserenyuk O. M. Selection o мали кращу оплату корму приростами, порівняно з аналогами вітчизняної селекції, на 0,12–0,35 кг або 0,13–0,39 корм. од. σ f — фенотипне стандартне відхилення товщини шпику [9]. σ f — фенотипне стандартне відхилення товщини шпику [9]. Дослідженнями Краснощока О. О. [10] встанов­ лено, що молодняк свиней, одержаний на основі промислового схрещування свиноматок великої білої породи та кнурів породи ландрас, за віком досягнення живої маси 100 кг переважав чистопородних ровес­ ників великої білої породи на 8 діб або на 4,5%. Зазначене свідчить, що в період інтенсифікації селек­ ційного процесу з використанням тварин зарубіжної селекції важливим питанням є дослідження рівня їх адаптації до умов навколишнього середовища, фено­ типового прояву основних господарсько корисних ознак, а також пошук критеріїв відбору високопродук­ тивних тварин за деякими оціночними індексами. Підтвердженням актуальності та практичної цінності вибраного напряму дослідження є роботи вітчизняних та зарубіжних вчених [3, 7, 11, 13, 15]. Силу кореляційних зв’язків між ознаками ви­ значали за шкалою Чеддока (цит. за [14]). Вартість додаткової продукції [12] та біометричну обробку одержаних даних [8] проводили за загальноприйня­ тими методиками. Матеріали і методи Встановлено, що молодняк свиней І групи (ін­ декс Тайлера коливається у межах від 214,89 до 242,85 бала) переважав ровесників ІІІ (індекс Тайле­ ра коливається у межах від 178,89 до 192,72 бала): за середньодобовим приростом живої маси за період контрольної відгодівлі — на 91,7 г (td=9,28; P<0,001), віком досягнення живої маси 100 кг — 7,3 доби (td=4,42; P<0,001), товщиною шпику на рівні 6–7 грудних хреб­ ців — 3,4 мм (td=6,07; P<0,001), довжиною охолодже­ ної туші — 1,7 см (td=3,54; P<0,01), довжиною беконної половини охолодженої туші — 2,7 см (td=3,64; P<0,01). Різниця між тваринами зазначених груп за індексом Тайлера дорівнює 36,2 бала (td=13,35; P<0,001). Дослідження провели в агроформуваннях Дніпро­ петровської обл., м’ясокомбінаті «Джаз» і лабораторії тваринництва Державної установи «Інститут зернових культур НААН». Об’єктом досліджень був молодняк свиней великої білої породи. Оцінку тварин за від­ годівельними і м’ясними якостями проводили з ура­ хуванням наступних показників: середньодобовий приріст живої маси за період контрольної відгодівлі, кг; вік досягнення живої маси 100 кг, діб; товщина шпику на рівні 6–7 грудних хребців, мм; довжина охолодже­ ної туші, cм; довжина беконної половини охолодженої туші, cм. [1]. Індекси Тайлера (1) і Сазера-Фредіна (2) розра­ ховували за формулами: З урахуванням внутрішньопородної диференці­ ації за індексом Сазера-Фредіна встановлено, що молодняк свиней ІІ групи (індекс Сазера-Фредіна у межах від –3,018 до –0,102 бала) переважав ро­ весників І (індекс Сазера-Фредіна у межах від +0,028 до +3,427 бала): за середньодобовим приростом живої маси за період контрольної відгодівлі — на 32,3 г (td=2,92; P<0,01), віком досягнення живої маси 100 кг — 6,7 доби (td=5,58; P<0,001), довжиною охолодженої туші — 0,2 см (td=0,47; P>0,05). Iв = 100 + (242 × K) – (4,13 × L) (1), (1), де Iв — комплексний індекс відгодівельних та м’ясних якостей, балів; де Iв — комплексний індекс відгодівельних та м’ясних якостей, балів; K — середньодобовий приріст живої маси, кг; L — товщина шпику на рівні 6–7 грудних хребців, мм; 242; 4,13 — постійні коефіцієнти [17]; I = 1σg × ΔG1 – 1σf × ΔF1 (2), Протилежну закономірність між групами встанов­ лено за довжиною беконної половини охолодженої туші та товщиною шпику на рівні 6–7 грудних хреб­ ців. Різниця між тваринами І і ІІ груп за довжиною бе­ конної половини охолодженої туші становить 1,1 см (td=1,54; P>0,05), товщиною шпику на рівні 6–7 груд­ них хребців — 2,2 мм (td=5,23; P<0,001). Біологія тварин, 2022, т. 24, № 1 Матеріали і методи (2), де I — індекс Сазера-Фредіна, балів; де I — індекс Сазера-Фредіна, балів; де I індекс Сазера Фредіна, балів; ΔG1 — швидкість росту у відхиленнях від середньої; ΔF1 — товщина шпику у відхиленнях від середньої; σ g — фенотипне стандартне відхилення швидкості росту; ΔG1 — швидкість росту у відхиленнях від середньої; ΔG1 — швидкість росту у відхиленнях від середньої; ΔF1 — товщина шпику у відхиленнях від середньої; σ g — фенотипне стандартне відхилення швидкості росту; σ g — фенотипне стандартне відхилення швидкості росту; 35 The Animal Biology, 2022, vol. 24, no. 1 The Animal Biology, 2022, vol. 24, no. 1 алак В. І., Жукорський О. М., Церенюк О. М. Таблиця 1. Відгодівельні і м’ясні якості молодняку свиней великої білої породи різної внутрішньопородної диференціації за індексом Тайлера Table 1. Fattening and meat qualities of young pigs of large white breed of different intrabreeding differentiation according to the Tyler index Показники, одиниці виміру Indicators, units of measurement Біометричні показники Biometric indicators Градації індексу Тайлера / Tyler index gradations 214,89–242,85 195,52–213,54 178,89–192,72 група / group І (n=13) ІІ (n=25) ІІІ (n=13) Середньодобовий приріст живої маси за період контрольної відгодівлі, г The average daily increase in live weight during the period of control fattening, g X±Sx 834,4±7,80 796,4±7,41 742,7±6,07 σ±Xσ 28,12±5,524 37,08±5,244 21,86±4,294 Сv±Scv, % 3,37±0,662 4,65±0,657 2,94±0,577 Вік досягнення живої маси 100 кг, діб Age of achievement of live weight of 100 kg, days X±Sx 174,4±1,14 177,9±0,85 181,7±1,20 σ±Xσ 4,11±0,807 4,25±0,601 4,34±0,852 Сv±Scv, % 2,35±0,461 2,38±0,336 2,38±0,467 Товщина шпику на рівні 6–7 грудних хребців, мм Thickness of fat at the level of 6–7 thoracic vertebrae, mm X±Sx 19,1±0,37 21,1±0,28 22,5±0,43 σ±Xσ 1,34±0,263 1,43±0,202 1,56±0,309 Сv±Scv, % 7,01±1,377 6,77±0,957 6,93±1,361 Довжина охолодженої туші, cм The length of the cooled carcass, cm X±Sx 97,4±0,35 97,1±0,33 95,7±0,34 σ±Xσ 1,26±2,475 1,69±0,239 1,23±0,241 Сv±Scv, % 1,29±0,253 1,74±0,246 1,28±0,251 Довжина беконної половини охолодженої туші, cм Length of bacon half of chilled carcass, cm X±Sx 87,0±0,58 86,1±0,59 84,3±0,47 σ±Xσ 2,10±0,412 2,98±0,421 1,70±0,333 Сv±Scv, % 2,41±0,473 3,46±0,489 2,01±0,394 за індексом Тайлера Table 1. Fattening and meat qualities of young pigs of large white breed of different intrabreeding differentiation according to the Tyler index Таблиця 2. Відгодівельні і м’ясні якості молодняку свиней великої білої породи різної внутрішньопородної диференціації за індексом Сазера-Фредіна Table 2. Fattening and meat qualities of young pigs of large white breed of different intrabreeding differentiation according to the Sazer-Fredin index Таблиця 2. Матеріали і методи Відгодівельні і м’ясні якості молодняку свиней великої білої породи різної внутрішньопородної диференціації за індексом Сазера-Фредіна Table 2. Fattening and meat qualities of young pigs of large white breed of different intrabreeding differentiation according to the Sazer-Fredin index Таблиця 2. Відгодівельні і м’ясні якості молодняку свиней великої білої породи різної внутрішньопородної диференціації за індексом Сазера-Фредіна Table 2. Fattening and meat qualities of young pigs of large white breed of different intrabreeding differentiation according to the Sazer-Fredin index Показники, одиниці виміру Indicators, units of measurement Біометричні показники bіоmеtics Indicators Градації індексу Сазера-Фредіна / Sazer-Fredin index gradations +0,028…+3,427 –3,018…–0,102 група /group І (n=28) ІІ (n=23) Середньодобовий приріст живої маси за період контрольної відгодівлі, г The average daily increase in live weight during the period of control fattening, g X±Sx 769,1±7,33 801,4±8,25 σ±Xσ 34,37±4,594 39,59±5,839 Сv±Scv, % 4,46±0,596 4,94±0,728 Вік досягнення живої маси 100 кг, діб Age of achievement of live weight of 100 kg, days X±Sx 182,1±0,82 175,4±0,89 σ±Xσ 4,39±0,586 4,28±0,631 Сv±Scv, % 2,41±0,322 2,44±0,359 Товщина шпику на рівні 6–7 грудних хребців, мм Thickness of fat at the level of 6–7 thoracic vertebrae, mm X±Sx 20,0±0,29 22,2±0,31 σ±Xσ 1,53±0,204 1,50±0,221 Сv±Scv, % 7,65±1,022 6,75±0,995 Довжина охолодженої туші, cм. The length of the cooled carcass, cm X±Sx 96,8±0,33 97,0±0,29 σ±Xσ 1,77±0,236 1,39±0,205 Сv±Scv, % 1,82±0,243 1,43±0,210 Довжина беконної половини охолодженої туші, cм Length of bacon half of chilled carcass, cm X±Sx 86,3±0,59 85,2±0,42 σ±Xσ 3,15±0,421 2,01±0,296 Сv±Scv, % 3,65±0,487 2,35±0,346 Біологія тварин, 2022, т. 24, № 1 36 Selection of breeding boars and sows by fattening and meat qualities of their offspring Khalak V. I., Zhukorsky O. M., Tserenyuk O. M. Khalak V. I., Zhukorsky O. M., Tserenyuk O. M. Selection of breeding Результати розрахунків коефіцієнта парної коре­ ляції між відгодівельними і м’ясними якостями мо­ лодняку свиней та оціночними індексами наведено у табл. 3. Коефіцієнт парної кореляції між відгодівель­ ними і м’ясними якостями молодняку свиней та оці­ ночними індексами коливається у межах від –0,710 (tr=10,24) до +0,844 (tr=20,98). ра × товщина шпику на рівні 6–7 грудних хребців, мм (–0,647), індекс Тайлера × довжина охолодженої туші (r = +0,359), індекс Тайлера × довжина беконної по­ ловини охолодженої туші (r = +0,298), індекс Сазера- Фредіна × середньо­добовий приріст живої маси за період кон­трольної відгодівлі (r = –0,371), індекс Сазера-Фредіна × вік досягнення живої маси 100 кг (r = +0,665), індекс Сазера-Фредіна × товщина шпику на рівні 6–7 грудних хребців, мм (–0,710). Матеріали і методи Вірогідні коефіцієнти кореляції встановлено за такими парами ознак: індекс Тайлера × середньо­ добовий приріст живої маси за період контрольної відгодівлі (r=+0,844), індекс Тайлера × вік досяг­ нення живої маси 100 кг (r = –0,546), індекс Тайле­ Результати розрахунку економічної ефективності використання молодняку свиней піддослідних груп наведено в табл. 4. Установлено, що максимальну Таблиця 3. Коефіцієнти парної кореляції між відгодівельними і м’ясними якостями молодняку свиней та оціночними індексами Table 3. Pairwise correlation coefficients between fattening and meat qualities of young pigs and evaluation indices Показник (ознака) / Indicator (attribute) Біометричні показники / Biometric indicators Сила кореляційного зв’язку The correlation strength х у r±Sr tr Індекс Тайлера, балів Tyler index, points 1 0,844±0,0402*** 20,98 Дуже висока / Very high 2 –0,546±0,0982*** 5,56 Помітна / Noticeable 3 –0,647±0,0814*** 7,95 Помітна / Noticeable 4 0,359±0,1220** 2,95 Помірна / Moderate 5 0,298±0,1276* 2,34 Слабка / Weak Індекс Сазера-Фредіна, балів Sazer-Fredin index, points 1 –0,371±0,1207** 3,08 Помірна Moderate 2 0,665±0,0780*** 8,53 Помітна / Noticeable 3 –0,710±0,0694*** 10,24 Висока / High 4 –0,084±0,1391 0,60 — 5 0,010±0,1400 0,07 — Примітка. 1 — середньодобовий приріст живої маси за період контрольної відгодівлі, кг; 2 — вік досягнення живої маси 100 кг, діб; 3 — товщина шпику на рівні 6–7 грудних хребців, мм; 4 — довжина охолодженої туші, cм; 5 — довжина беконної половини охолодженої туші, cм. * — P<0,05; ** — P<0,01; *** — P<0,001. Note. 1 — average daily increase in live weight during the period of control fattening, kg; 2 — age of achievement of live weight of 100 kg, days; 3 — the thickness of the fat at the level of 6–7 thoracic vertebrae, mm; 4 — length of cooled carcass, cm; 5 — length of the bacon half of the cooled carcass, cm. * — P<0.05; ** — P<0.01; *** — P<0.001. Таблиця 3. Коефіцієнти парної кореляції між відгодівельними і м’ясними якостями молодняку свиней та оціночними індексами Table 3. Pairwise correlation coefficients between fattening and meat qualities of young pigs and evaluation indices оефіцієнти парної кореляції між відгодівельними і м’ясними якостями молодняку свиней та оціночними індексами wise correlation coefficients between fattening and meat qualities of young pigs and evaluation indices Примітка. 1 — середньодобовий приріст живої маси за період контрольної відгодівлі, кг; 2 — вік досягнення живої маси 100 кг, діб; 3 — товщина шпику на рівні 6–7 грудних хребців, мм; 4 — довжина охолодженої туші, cм; 5 — довжина беконної половини охолодженої туші, cм. Примітка. 1 — середньодобовий приріст живої маси за період контрольної відгодівлі, кг; 2 — вік досягнення живої маси 100 кг, діб; 3 — товщина шпику на рівні 6–7 грудних хребців, мм; 4 — довжина охолодженої туші, cм; 5 — довжина беконної половини охолодженої туші, cм. * — P<0,05; ** — P<0,01; *** — P<0,001. Note. 1 — average daily increase in live weight during the period of control fattening, kg; 2 — age of achievement of live weight of 100 kg, days; 3 — the thickness of the fat at the level of 6–7 thoracic vertebrae, mm; 4 — length of cooled carcass, cm; 5 — length of the bacon half of the cooled carcass, cm. * — P<0.05; ** — P<0.01; *** — P<0.001. Примітка. * — ціна реалізації молодняку свиней на переробні підприємства регіону на час проведення дослідження дорівнювала 47,5 грн./кг. Note. * — the sale price of young pigs to processing enterprises of the region at the time of the study was 47.5 UAH/kg. The Animal Biology, 2022, vol. 24, no. 1 37 Висновки 4. Khalak VI, Chernyavsky SE, Voloshchuk VM, Pochernyaev KF, Ilchenko MO. Productive and meat traits of young pigs with dif­ ferent SNP c.1426 G>A genotypes of the melanocortin receptor 4 gene (MC4R), provided that they are distributed according to some traits. Pig Breed. Poltava, 2019; 73: 157−165. Available at: https://drive.google.com/file/d/1LBPRQUXycquwpNGzkpn__ Og3VFDj54u7/view (in Ukrainian) 1. Молодняк свиней великої білої породи підкон­ троль­ної популяції за віком досягнення живої маси 100 кг переважає мінімальні вимоги класу еліта на 6,13%, товщиною шпику на рівні 6–7 грудних хреб­ ців — 32,25% і довжиною охолодженої туші — 3,92%. 5. Khalak V, Gutyj B, Bordun O, Ilchenko M, Horchanok A. Effect of blood serum enzymes on meat qualities of piglet productivity. Ukr. J. Ecol. 2020; 10 (1): 158−161. DOI: 10.15421/2020_25. 2. Вірогідну різницю між тваринами різної вну­ трішньопородної диференціації за індексом Тайлера (І і ІІІ групи) встановлено за середньодобовим при­ ростом живої маси за період контрольної відгодівлі (91,7 г; td=9,28; P<0,001), віком досягнення живої маси 100 кг (7,3 доби; td=4,42; P<0,001), товщиною шпику на рівні 6–7 грудних хребців (3,4 мм; td=6,07; P<0,001), довжиною охолодженої туші (1,7 см; td=3,54; P<0,01), довжиною беконної половини охолодженої туші (2,7 см; td=3,64; P<0,01). 6. Khramkova OM. Economic and biological features, adaptive properties of pigs of Irish origin and their use in different breeding methods. Abstract. Ph.D. agr. Science: 06.02.01. Mykolaiv, 2021: 22 p. (in Ukrainian) 7. Khvatova MA. Heterosis effect prognostication according to lin­ ear swine type combinability. Sci. Tech. Bull. Inst. Anim. Husb. NAAS. Kharkiv, 2012; 107: 148–153. (in Ukrainian) 8. Kovalenko VP, Khalak VI, Nezhlukchenko TI, Papakina NS. Biometric analysis of the variability of traits of farm animals and birds. Textbook on genetics of farm animals. Kherson, Oldie, 2010: 160 p. (in Ukrainian) 3. З урахуванням внутріпородної диференціації за індексом Сазера-Фредіна молодняк свиней ІІ групи переважав ровесників І за середньодобовим при­ ростом живої маси за період контрольної відгодівлі на 4,03%, віком досягнення живої маси 100 кг — 3,67% і довжиною охолодженої туші — 0,2%. 9. Kozlovskiy VG. Breeding business in pig breeding. Moscow, Kolos, 1982: 272 p. 10. Krasnoshchok OO. Formation of productivity of pigs depending on methods of cultivation and intensity of growth. Abstract. PhD agr. sci.: 06.02.01. Poltava, 2020: 23 p. (in Ukrainian) 11. Loban NA. System of selection and genetic evaluation methods for fattening and meat trains of pigs. Pig Breed. Poltava, 2014; 65: 69−75. Available at: https://drive.google.com/file/d/1mdWfg3nqWZ- bbKFHnObuSYnxwdsylmqV/view (in Ukrainian) 4. Відбір кнурів і свиноматок за відгодівельними і м’ясними якостями потомства Відбір кнурів і свиноматок за відгодівельними і м’ясними якостями потомства 1. Berezovskyy MD, Khatko IV. Methods for assessing boars and sows on the quality of offspring in breeding plants and breeding breeders. Modern research methods in pig breeding. Poltava, 2005: 32−37. (in Ukrainian) 1. Berezovskyy MD, Khatko IV. Methods for assessing boars and sows on the quality of offspring in breeding plants and breeding breeders. Modern research methods in pig breeding. Poltava, 2005: 32−37. (in Ukrainian) прибавку додаткової продукції одержано від молод­ няку свиней І піддослідної групи за індексом Тайлера (+5,03%) та ІІ піддослідної групи за індексом Сазера- Фредіна (+1,12%). Вартість додаткової продукції, яку одержано від молодняку свиней зазначених груп, за умови, що ціна реалізації за 1 кг живої маси молод­ няку свиней на час проведення дослідження дорів­ нювала 47,5 грн., становить +247,51 і +55,42 грн./гол. відповідно. 2. Gnatyuk SA. The results and prospects of work of “The Zhivprom” corporation. Ukrainian Black Sea Region Agrarian Science. Myko­ laiv, 2015; 2 (84/2): 15–22. Available at: https://visnyk.mnau.edu.ua/ n84v2r2015t2gnatyuk (in Ukrainian) 3. Gryshyna LP, Fesenko OH. The effectiveness of using a spe­ cialized type of pigs at the crossing and hybridization. Ukrainian Black Sea Region Agrarian Science. 2015; 2 (84/2): 40–47. Available at: https://visnyk.mnau.edu.ua/n84v2r2015t2grishina (in Ukrainian) Висновки Коефіцієнт парної кореляції між відгодівельними і м’ясними якостями молодняку свиней та оціночними індексами коливається від –0,710 (tr=10,24) до +0,844 (tr=20,98). 12. Methods for determining the economic efficiency of the use in agriculture of the results of scientific research, new technolo­ gy, inventions and rationalization proposals. Moscow, VAIIPI, 1983: 149 p. 5. Максимальну прибавку додаткової продукції одержано від молодняку свиней І групи за індексом Тайлера (+5,03%) і ІІ за індексом Сазера-Фредіна (+1,12%). Вартість додаткової продукції, отриманої від молодняку свиней зазначених груп, за умови, що ціна реалізації за 1 кг живої маси молодняку свиней на час проведення дослідження 47,5 грн, становить +247,51 і +55,42 грн./гол. відповідно. 13. Povod MH. Influence of technological features on fattening in­ dicators of pigs. Bull. Sumy Nat. Agr. Univ. Livestock Ser. 2014; 2 (2): 30–36. (in Ukrainian) 14. Sidorova AV, Leonova NV, Masich LA. Workshop on the theory of statistics. A textbook. Donetsk, Donetsk National University. 2003: 252 p. (in Ukrainian) 6. Критерієм відбору високопродуктивних кнурів- плідників і свиноматок за відгодівельними і м’яс­ ними якостями їх потомства є межі індексу Тайлера 214,89–242,85, індексу Сазера-Фредіна — від –3,018 до –0,102 балів. 15. Susol RL. Productivity of large white pigs with improved meat qualities, taking into account DNA markers. Sci. Bull. Askaniya- Nova. 2013; 6: 229–235. (in Ukrainian) 16. Tserenyuk OM. Wales and Landrace pigs genetic potential of productivity on feeding qualities. Sci. Tech. Bull. Inst. Anim. Husb. NAAS. Kharkiv, 2018; 120: 160−167. DOI: 10.32900/2312-8402- 2018-120-160-167. (in Ukrainian) 17. Vashchenko PA. Prediction of breeding value of pigs based on linear models of selection indices and DNA markers.Abstr. doct. agr. sci.: 06.02.01 Breeding and selection of animals. Mykolaiv, 2019: 43 p. (in Ukrainian) Матеріали і методи * — P<0,05; ** — P<0,01; *** — P<0,001. Note. 1 — average daily increase in live weight during the period of control fattening, kg; 2 — age of achievement of live weight of 100 kg, days; 3 — the thickness of the fat at the level of 6–7 thoracic vertebrae, mm; 4 — length of cooled carcass, cm; 5 — length of the bacon half of the cooled carcass, cm. * — P<0.05; ** — P<0.01; *** — P<0.001. Таблиця 4. Економічна ефективність результатів досліджень Table 4. Economic efficiency of research results Група Group n Середньодобовий приріст живої маси за період контрольної відгодівлі, г The average daily increase in live weight during the period of control fattening, g Прибавка додаткової продукції, % Increase in additional products, % Вартість додаткової продукції, грн./гол.* Cost of additional products, UAH/animal* Загальна вибірка / Total sample 51 792,4±6,37 — — Внутрішньопородна диференціація за індексом Тайлера / Intrabreed differentiation according to the Tyler index ІІІ 13 742,7±6,07 –6,27 –321,4 ІІ 25 796,4±7,41 +0,50 +25,09 І 13 834,4±7,80 +5,03 +247,51 Внутрішньопородна диференціація за індексом Сазера-Фредіна / Intrabreed differentiation according to the Sazer-Fredin index І 28 769,1±7,33 –2,94 –151,05 ІІ 23 801,4±8,25 +1,12 +55,42 Примітка. * — ціна реалізації молодняку свиней на переробні підприємства регіону на час проведення дослідження дорівнювала 47,5 грн./кг. Note. * — the sale price of young pigs to processing enterprises of the region at the time of the study was 47.5 UAH/kg. Примітка. * — ціна реалізації молодняку свиней на переробні підприємства регіону на час проведення дослідження дорівнювала 47,5 грн./кг. Note. * — the sale price of young pigs to processing enterprises of the region at the time of the study was 47.5 UAH/kg. 37 37 Халак В. І., Жукорський О. М., Церенюк О. М. Перспективи подальших досліджень Подальша робота передбачає провести досліджен­ ня відгодівельних і м’ясних якостей молодняку свиней різних поєднань та генотипів за геном МС4R. Біологія тварин, 2022, т. 24, № 1 38 Khalak V. I., Zhukorsky O. M., Tserenyuk O. M. Khalak VI, Zhukorsky OM, Tserenyuk OM. Selection criteria for highly productive breeding boars and sows by fattening and meat qualities of their offspring using some evaluation indexes. Bìol. Tvarin. 2022; 24 (1): 34–39. DOI: 10.15407/animbiol24.01.034. Key words: young pigs, breed, fattening and meat qualities, index, economic efficiency, variability, correlation Selection criteria for highly productive breeding boars and sows by fattening and meat qualities of their offspring using some evaluation indexes* V. I. Khalak1, O. M. Zhukorsky2, O. M. Tserenyuk3 v16kh91@gmail.com V. I. Khalak1, O. M. Zhukorsky2, O. M. Tserenyuk3 v16kh91@gmail.com 1SE Institute of Grain Crops of National Academy of Agrarian Sciences, 14 Volodymyr Vernadskyi str., Dnipro, 49009, Ukraine 2National Academy of Agrarian Sciences of Ukraine, 9 Mykhailo Omelyanovych-Pavlenko str., Kyiv, 901010, Ukraine 3Institute of Pig Breeding and Agrarian Sciences NAAS, 1 Shvedska Mohyla str., Poltava, 36013, Ukraine The aim of the work is to study the fattening and meat qualities of young white pigs and to determine the criteria for selection of high-yielding breeding boars and sows using some evaluation indices. It was found that young pigs of large white breed of the controlled population at the age of 100 kg exceed the minimum requirements of the elite class by 6.13%, the thickness of the fat at the level of 6–7 thoracic vertebrae — 32.25% and the length of chilled carcass — 3.92%. Significant difference between animals of different intrabreed differentiation according to the Tyler index (groups I and III) was established by the average daily gain of live weight during the control period of fattening (91.7 g), age of 100 kg live weight (7.3 days), fat thickness at the level of 6–7 thoracic vertebrae (3.4 mm), length of chilled carcass (1.7 cm), length of bacon half of chilled carcass (2.7 cm). Taking into account the intrabreed differentiation according to the Sazer-Fredin index, it was found that young pigs of group II outperformed peers of I on average daily live weight gain for the period of control fattening by 4.03%, age of 100 kg — 3.67% and length of chilled carcass — 0.2%. The pairwise correlation coefficient between fattening and meat qualities of young pigs and evaluation indices ranges from –0.710 (tr=10.24) to +0.844 (tr=20.98). The maximum increase in additional products was obtained from young pigs of the first experimental group according to the Tyler index (+5.03%) and the second experimental group according to the Sazer-Fredin index (+1.12%). The cost of additional products obtained from young pigs of these groups, provided that the selling price per 1 kg of live weight of young pigs at the time of the study was 47.5 UAH. is +247.51 and +55.42 UAH/animal respectively. Thus, the criteria for selecting highly productive breeding boars and sows for fattening and meat qualities are the limits of the Tyler index 214.89–242.85, the Sazer-Fredin index — –3.018…–0.102 points.
https://openalex.org/W2157260923
https://www.duo.uio.no/bitstream/10852/34721/2/dravhandling-austvoll-dahlgren.pdf
English
null
What informs parents’ decision‐making about childhood vaccinations?
Journal of advanced nursing
2,010
cc-by
42,985
Improving Health Literacy Improving Health Literacy Improving Health Literacy Identifying barriers and facilitators to people’s ability to obtain health information, and the development and evaluation of an intervention to improve health literacy Identifying barriers and facilitators to people’s ability to obtain health information, and the development and evaluation of an intervention to improve health literacy Astrid Austvoll-Dahlgren Faculty of Medicine, Institute of Health and Society, University of Oslo Faculty of Nursing, Oslo University College Astrid Austvoll-Dahlgren Faculty of Medicine, Institute of Health and Society, University of Oslo Faculty of Nursing, Oslo University College g Faculty of Medicine, Institute of Health and Society, University of Oslo Faculty of Nursing, Oslo University College © Astrid Austvoll-Dahlgren, 2012 Series of dissertations submitted to the Faculty of Medicine, University of Oslo No. 1375 ISBN 978-82-8264-472-3 ........................................................................................................................................................................5 List of papers ............................................................................................................................................................7 Norsk sammendrag ..................................................................................................................................................8 Abstract ....................................................................................................................................................................9 Introduction ........................................................................................................................................................... 11 Models, arenas and differences in participation .................................................................................................... 12 Evidence based information is not readily available ............................................................................................. 14 Health literacy: a public health issue .................................................................................................................... 16 What do health literacy skills really entail? .......................................................................................................... 19 Health literacy and associated outcomes .............................................................................................................. 21 Previous interventions targeting health literacy .................................................................................................... 23 Complex problems may require complex interventions ........................................................................................ 24 Parents’ decision making as a case ....................................................................................................................... 27 Aims ........................................................................................................................................................................ 28 Methods .................................................................................................................................................................. 31 Analysing the problem and identifying barriers and facilitators to change ............................................................ 31 Designing the intervention ................................................................................................................................... 35 Evaluating the intervention .................................................................................................................................. 40 Results .................................................................................................................................................................... 43 Decision making by parents, and barriers and facilitators to obtaining information .............................................. 43 Choice of intervention delivery and content ......................................................................................................... 47 Usability and effectiveness of the web portal ....................................................................................................... 48 Discussion ............................................................................................................................................................... 52 Methodological considerations............................................................................................................................. 54 Capacity, abilities, and the perceived need as barriers .......................................................................................... 59 Parent and public health nurse barriers may be similar ......................................................................................... 63 Attitude and perceived behavioural control are important predictors of intention to search for health information 64 Providing an easy access point to evidence and tools ........................................................................................... 67 The web portal was judged as a relevant tool and may improve attitudes in a pragmatic setting ........................... 70 People may not trust in science, and rational decision making may not result in the ‘recommended’ choice ......... 72 Some have expressed concerns about interventions targeting user participation and health literacy ...................... 73 Conclusions ............................................................................................................................................................ 75 Implications for practice and further research ..................................................................................................... 75 References .............................................................................................................................................................. 77 Table of contents ISBN 978-82-8264-472-3 All rights reserved. No part of this publication may be reproduced or transmitted, in any form or by any means, without permission. Cover: Inger Sandved Anfinsen. Printed in Norway: AIT Oslo AS. Printed in Norway: AIT Oslo AS. Printed in Norway: AIT Oslo AS. Produced in co-operation with Unipub. Produced in co-operation with Unipub. The thesis is produced by Unipub merely in connection with the thesis defence. Kindly direct all inquiries regarding the thesis to the copyright holder or the unit which grants the doctorate. p p The thesis is produced by Unipub merely in connection with the thesis defence. Kindly direct all inquiries regarding the thesis to the copyright holder or the unit which grants the doctorate. holder or the unit which grants the doctorate. 3 Table of contents Takk! ........................................................................................................................................................................5 List of papers ............................................................................................................................................................7 Norsk sammendrag ..................................................................................................................................................8 Abstract ....................................................................................................................................................................9 Introduction ........................................................................................................................................................... 11 Models, arenas and differences in participation .................................................................................................... 12 Evidence based information is not readily available ............................................................................................. 14 Health literacy: a public health issue .................................................................................................................... 16 What do health literacy skills really entail? .......................................................................................................... 19 Health literacy and associated outcomes .............................................................................................................. 21 Previous interventions targeting health literacy .................................................................................................... 23 Complex problems may require complex interventions ........................................................................................ 24 Parents’ decision making as a case ....................................................................................................................... 27 Aims ........................................................................................................................................................................ 28 Methods .................................................................................................................................................................. 31 Analysing the problem and identifying barriers and facilitators to change ............................................................ 31 Designing the intervention ................................................................................................................................... 35 Evaluating the intervention .................................................................................................................................. 40 Results .................................................................................................................................................................... 43 Decision making by parents, and barriers and facilitators to obtaining information .............................................. 43 Choice of intervention delivery and content ......................................................................................................... 47 Usability and effectiveness of the web portal ....................................................................................................... 48 Discussion ............................................................................................................................................................... 52 Methodological considerations............................................................................................................................. 54 Capacity, abilities, and the perceived need as barriers .......................................................................................... 59 Parent and public health nurse barriers may be similar ......................................................................................... 63 Attitude and perceived behavioural control are important predictors of intention to search for health information 64 Providing an easy access point to evidence and tools ........................................................................................... 67 The web portal was judged as a relevant tool and may improve attitudes in a pragmatic setting ........................... 70 People may not trust in science, and rational decision making may not result in the ‘recommended’ choice ......... 72 Some have expressed concerns about interventions targeting user participation and health literacy ...................... 73 Conclusions ............................................................................................................................................................ 75 Implications for practice and further research ..................................................................................................... 75 References .............................................................................................................................................................. 77 3 Table of contents Takk! Takk! Dette prosjektet har vært en fantastisk reise å få være med på, kanskje særlig på grunn av alle de jeg har fått lov til å dele den med. Først og fremst vil jeg gjerne få rette en stor takk til mine veiledere Sølvi Helseth og Arild Bjørndal for deres ro, optimisme og visdom. Jeg har følt meg svært privilegert som har hatt dere som veiledere. Jeg er også svært takknemlig for at Stein Danielsen og Elin Opheim ble en del av dette prosjektet. ”Sunn skepsis” hadde ikke blitt det samme uten dere! Jeg har satt enormt pris på vårt samarbeid, både faglig og sosial. En hjertelig takk rettes også til alle som har deltatt i studiene og utviklingsarbeidet som utgjør denne avhandlingen. Jeg vil rette en spesiell takk til Mona Myrer for sitt bidrag til Sunn skepsis sidene. Jeg vil også takke Helene Stenseth Mølster, Loan Thi Mai Tran og Siri Carlsen Tverli for deres hjelp med datainnsamling. Takk også til Ragnhild Sørum Falk og Jill Francis for å ha bidratt med nyttig ekspertise, og til Simon Goudie for hjelp med språk. Jeg vil gjerne takke mine tidligere ledere og kollegaer ved (tidligere) avdeling for Sykepleierutdanning, Vigdis Granum, Finn Nortvedt, Gry Opsahl, Kari Toverud, Olof Birna Kristjansdottir og Per Nerdrum, samt damene i ”Team for helsefremmende og forebyggende”, Monica Larsen, Kjellaug Kildal Hansen, Pia Kahn og Unni Kjerland. Takk for et godt samarbeid og for at dere hadde troen på at en Samfunnsviter kunne være en nyttig ressurs i sykepleiefaget. Tiden ved avdelingen har gitt meg erfaringer jeg ikke ville vært foruten, og det har vært en sann glede å få lov til å jobbe med studentene. Takk også til stipendiatgruppa for faglig og sosialt samvær! En særskilt takk til ”medstipendiat” Kirsti Riiser for alle de gode fagdiskusjonene, for at du fikk meg opp i en klatrevegg og for ditt vennskap i denne tiden. Det har vært mye god støtte i erfaringsutvekslingene over en kaffekopp på ”La Rustique”. Jeg vil også takke ledelsen og medstipendiater i storbyprogrammet for fantastiske reiser, tverrfaglige diskusjoner og cavakvelder som har gitt bredere horisont. 5 Jeg vil rette en takk til Andy Oxman og Shaun Treweek for at dere valgte å satse på en nyutdannet masterstudent i 2002. Table of contents 3 3 Appendixes ............................................................................................................................................................. 86 Appendix 1: Copy of Interview guides focus groups ............................................................................................ 86 Appendix 2. Copy of questionnaire including TPB, PAM and Honeycomb .......................................................... 91 Appendix 3. Summary of the ratings in usability test ........................................................................................... 97 Appendix 4. Summary of changes after usability test ........................................................................................... 98 Appendix 5. Results of TPB utility testing across the samples and studies ......................................................... 101 Papers I-V ............................................................................................................................................................ 103 4 4 Takk! Takk også for alle ideene, den smittende arbeidsgleden og utfordringene dere har gitt meg igjennom årene, og som har gitt meg mersmak på forskning og ikke minst inspirert tematikken for dettep ph.d - arbeidet. Takk også til andre kollegaer og venner 5 på Kunnskapssenteret som har stilt sin kunnskap og erfaring til rådighet, Signe Flottorp, Liv Merete Reinar, Claire Glenton, Simon Lewin, Jan Odgaard-Jensen, Elin Strømme Nilsen og Marit Johansen. Jeg er veldig heldig som har hatt (og har) gleden av å være en del av et slikt kollegium hvor det å dele kunnskap og fagengasjement alltid står i fokus. Jeg vil også takke Gunn Vist, min seksjonsleder, for støtte og tålmodighet de siste månedene. Til slutt vil jeg takke min kjære Kjetil for alle de gode samtalene, for ditt gode humør og for å minne meg på at livet er mer en jobb. Jeg vil også takke moren min (og helsesøster!), Berith, for at du alltid stiller opp, for at du deler dine erfaringer og innsikt, og for å lære meg gleden over å erverve ny kunnskap. Gode tanker går også til mine svigerforeldre Åse og Kjell, samt gode venner som har vist entusiasme og oppmuntring når dagene har vært travle. Takk også til mine alltid trofaste turkamerater for fantastiske naturopplevelser i all slags vær, mye av det beste tankearbeidet har funnet sted på en skogssti! Oslo 16.02.2012, Astrid Austvoll-Dahlgren 6 6 List of papers Austvoll-Dahlgren, A. and Helseth, S. What informs parents’ decision-making about childhood vaccinations? Journal of Advanced Nursing, 66(11), 2421–2430. Austvoll-Dahlgren, A. and Helseth, S. Public health nurses’ barriers and facilitators to the use of research in consultations about childhood vaccinations. Scandinavian Journal of Caring Sciences 15 DEC 2011. Austvoll-Dahlgren, A. Falk, RS. and Helseth, S. Cognitive factors predicting intentions to search for health information: an application of the Theory of Planned Behaviour. Submitted with revisions 18.01.2012 to Health Information and Libraries Journal. Austvoll-Dahlgren, A. Danielsen, SO. Opheim, E. Bjørndal, A. Reinar, LM.,Flottorp, SA. Oxman, AD. and Helseth, S. A complex intervention: tailoring a web portal to improve user health literacy skills. Submitted 14.02.2012 to Health Information and Libraries Journal. Austvoll-Dahlgren, A. Bjørndal, A.Odgaard-Jensen, J. and Helseth, S. Evaluation of a web portal for improving public access to evidence-based health information and health literacy skills: a pragmatic trial. Submitted with revisions 15.02.2012 to PLoS ONE. 7 7 Objective The overall goal of this project was: to improve people’s health literacy skills related to obtaining health information through an intervention to target key identified barriers and facilitators, and to evaluate this effort. Norsk sammendrag Brukermedvirkning i helsetjenesten et kjerneelement i kunnskapsbasert praksis og en rettighet i følge pasientrettighetsloven. Nyere forskning viser at brukere både kan og vil involveres i beslutninger om helse. Effektiv deltakelse er avhengig av at brukerne har tilgang til pålitelig helseinformasjon samt kunnskaper og ferdigheter for å tilegne seg denne informasjon. Slike kunnskaper og ferdigheter er også definert som helserelatert kompetanse (health literacy). Dessverre er ikke forskningsbasert informasjon lett tilgjengelig, og mange har ikke tilstrekkelig helserelatert kompetanse til å dra nytte av slik informasjon. Det finnes ulike måter å bedre tilgangen til pålitelig informasjon for folk flest, blant annet at forskere oppsummerer forskning på en forståelig måte, nettbaserte søkemotorer som filtrerer ut upålitelig informasjon, at nettsteder og andre former for brukerinformasjon blir kvalitetssikret og merket, samt brukeropplæring i helserelatert kompetanse. Selv om alle disse tiltakene er viktige, er det kun den siste fremgangsmåten som utstyrer brukerne med kunnskap og egenskaper på tvers av kilder, kontekst og tema. Dette området er av stor interesse internasjonalt, og karakteriseres av metodeutvikling og tverrfaglighet. Til nå har det vært lite forskning på tiltak som retter seg mot helserelatert kompetanse knyttet til det å oppsøke og kritisk vurdere helseinformasjon. I denne avhandlingen beskriver jeg hvordan vi gikk frem for å utvikle og evaluere et slikt tiltak, nettportalen ‘Sunn skepsis’ (www.sunnskepsis.no). Prosjektet er en kompleks intervensjon i tre faser: i første fase utforsket vi hva som hindrer og fremmer brukere i å tilegne seg pålitelig informasjon, i den andre fasen utviklet vi nettportalen med et brukerperspektiv og hvor forskningsbaserte strategier ble utformet for å møte disse hindrene, og i den siste fasen ble denne nettportalen evaluert. 8 Background User participation is a core element of evidence based practice and an important contributor to user autonomy and health. However, effective participation is dependent on user access to reliable health information as well as the necessary knowledge and skills to obtain such information. Research has found that evidence based information is not readily available to the public and that many users have poor health literacy skills. Method This project was conducted in three phases. In the first phase, we used qualitative interviews, a questionnaire study and literature searches to identify barriers and facilitators to obtaining health information. In the second phase, we developed a web portal intervention to address the important identified barriers using a choice of evidence based strategies. In the final phase, we evaluated the web portal. 9 9 Results Three main barriers were identified as areas for improvement: the inability of users to understand and critically appraise health information, the inability to exchange information in consultations, and not knowing where to find reliable and relevant health information. The design of the web portal was grounded in the conceptual framework of shared decision making and evidence based practice. It included three main strategies to address the identified barriers: an introduction to research methods and ways to find evidence-based health information efficiently, an introduction to critical appraisal and a checklist, and information about patients’ participation rights in decision making (the section included a decision aid for consultations). When testing the web portal during the pragmatic trial we found improvements in users’ positive beliefs about searching. In addition, the web portal was regarded as a relevant tool by users. Conclusion The web portal was developed in a transparent and systematic way, based on feedback from our target audiences. Open access was provided to the tools included in the web portal: healthcare users could utilise these independently or in consultations with health professionals. Although providing access to such resources may be important in itself, future educational efforts should consider including this web portal in more intensive interventions. These interventions may include, for example, organised courses targeting patients or patient representatives, educational programmes in schools, or the use of the web portal during consultations with health professionals. 10 Introduction The involvement of healthcare users (hereafter referred to as users) is a core element of evidence based practice (1). This has been nationally and internationally affirmed through legislation such as the Norwegian Patients’ Rights Act, and in international declarations such as those issued by the World Health Organization (WHO) (2-4). The rationale for advocating participation is rooted in ethical principles in respect for patient rights and autonomy (2, 5) as well as pragmatic arguments based on research showing that participation may be important for user satisfaction with care, improving the exchange of information, reducing decisional conflict, and improving knowledge (6, 7). Recent arguments have also highlighted the importance of participation and public involvement as a duty and a responsibility, and as necessary for managing care in ways that meet the current challenges in healthcare effectively (5). 11 Effective participation is dependent on users being given the opportunity to be involved in consultations with health professionals, having access to reliable health information (6), and the necessary health literacy skills to obtain, evaluate and act upon such information (8). However, evidence suggests that people are less involved in decision making than they would like to be, and that their communication with health professionals and sharing of information may be sub-optimal (9-11). Moreover, evidence based information is not readily available to the public (12-16), and studies have found that many users have limited health literacy skills (17, 18). Consequently, people are often poorly equipped either to participate in decision-making or to make informed decisions in which they have confidence. There is therefore a need for interventions that facilitate peoples’ access to reliable health information and their ability to use this information, and these efforts should be evaluated in order to inform practice (6, 17, 19, 20). 11 Improving users health literacy skills is by the WHO described as the main outcome of health education (2) and a central priority at the EU level advocated by the European Patients Forum (umbrella organisation of pan-European patient organisations) (21). Based on these arguments and research findings, we decided to develop a resource that could be made freely available to the Norwegian public and that would target users of healthcare with the purpose of improving health literacy skills related to obtaining health information. The aim was to do this through an intervention targeting key identified barriers and facilitators, and to evaluate this effort. Introduction In this thesis, I describe the rationale, development and evaluation of this intervention; a web portal we named ‘Healthy Scepticism’ (Sunn skepsis). Models, arenas and differences in participation Numerous models and terminologies have been used to describe clinical decision making (5, 10, 22). The discourse on participation often relates to consultations and decision making about healthcare, such as treatment or screening options (7). In this context, participation is generally categorised into three main decision making models, namely: provider-led, user-led and shared decision making (5). These models occur across a ‘locus of control’ spectrum and reflect, to a large extent, when different paradigms were preferred at different times (7). 12 The provider-led model is a traditional, paternalistic and prescriptive model in which the provider decides and the user consents or adheres (5). Critics of this model suggest that it does not foster self-care and efficacy, but encourages passivity instead (5). Evidence suggests that this communication style does not facilitate the exchange of information, which may be important for 12 treatment and patient safety (7). It may also foster unrealistic patient demands for healthcare by increasing dependency on health professionals, and by not emphasising self-management. This in turn may lead to frustration and dissatisfaction among patients as well as health professionals (5). In the second model of participation, namely user-led decision making, the user decides and the providers deliver the information (5). This model, based on the libertarian principles of consumerist market mechanisms, was prominent in the early 1980s and developed as a response to the dominant provider-led model of the time (5, 23). A perhaps obvious criticism of this model is that it may generate an unnecessary demand for healthcare. Another shortcoming of this model is that it encourages user demand while failing to emphasise shared responsibilities (5). The third model of participation acknowledges the shortcomings of the first two. Often referred to as the ‘shared decision making’ or ‘partnership’ model, it is now the predominantly accepted and advocated approach (5, 7, 9). In it, both parties share information and responsibility, and decisions are made together (5). It should be noted that the terminology of shared decision- making is often used interchangeably with patient-centred care (5, 7). In addition to emphasising shared roles and responsibilities, this approach adapts the focus of a humanistic biopsychological perspective upon the individual, rather than focusing upon the disease itself (as would be done within a more biomedical perspective) (7). Models, arenas and differences in participation Although this third model is not perfect (if such perfection exists), it is considered preferable to the first two described above because both user and healthcare provider perspectives are respected, and the model acknowledges that both parties have valuable knowledge and preferences to bring to the table (5). 13 It is important to recognise that participation in health is not restricted to consultations. In fact, most health activities take place in everyday life. On the individual level (from a preventive 13 and health promoting perspective), participation includes how we choose to live our life and how we act on health messages (8). The focus on participation outside of healthcare settings is particularly important considering the increased number of people living and coping with incurable or chronic disease (24). These diseases require active engagement from individuals in order for them to master their everyday life, keep up with treatment regimes, manoeuvre their way through the healthcare system, and to manage the enormous volume of health information they are exposed to (8, 24). Furthermore, participation also takes place on a system level in the form of performance assessments of health services and engagement in public health discourse (6, 8, 25). and health promoting perspective), participation includes how we choose to live our life and how we act on health messages (8). The focus on participation outside of healthcare settings is particularly important considering the increased number of people living and coping with incurable or chronic disease (24). These diseases require active engagement from individuals in order for them to master their everyday life, keep up with treatment regimes, manoeuvre their way through the healthcare system, and to manage the enormous volume of health information they are exposed to (8, 24). Furthermore, participation also takes place on a system level in the form of performance assessments of health services and engagement in public health discourse (6, 8, 25). Age is often found to be an important factor shaping differences in participation both inside and outside the consultation setting (9, 26-29). Younger people are typically seen to be more active (26-28) and more likely to prefer a shared decision making model, whereas the elderly may prefer a more traditional provider-led model (9). Models, arenas and differences in participation This variation can be explained by differences in generational culture, but also by educational and practical issues, such as the degree of familiarity with the use of new information technologies (5, 9, 27, 29, 30). The latter factor is associated with what is known as the ‘digital divide’ (6), which is expected to diminish over time as more people of all ages become familiar with such technologies (29). Evidence based information is not readily available Acknowledging users’ rights to participate and their key role in evidence base practice (1, 3, 4) means that user access to reliable health information is a necessity for effective participation 14 and informed decision making (6). As a result of the emergence of new information technologies, health information previously in the exclusive domain of professionals is now also widely available to the public and many people actively search for it (26-29). Mass media and particularly social media are becoming important sources of health information (25, 28). When people use the internet for health research purposes, most search for information about a specific illness, treatment or therapy (28, 29). In a cross-sectional study in Norway, over 90% of the respondents were reported to have read about health or illness issues online (28). Furthermore, the percentage of people using the internet for health-related purposes in Norway has been estimated to be 67% (2007) (28). Despite the importance of new information technologies, many people still rate family and friends and health professionals as their most important sources of health information (27-29), and evidence suggests that other sources are used to complement traditional channels of information (25, 28). Having practice based on the best available evidence is a national and international priority (1, 31, 32). However, the information that people are presented with, or depend on, may not be evidence based. Through mass media, the public is exposed to a wide range of treatment alternatives, claims about health risks, and advice about how to promote good health. Many of these reports are incomplete, biased or not based on research (12, 15, 16). Furthermore, although advice from family and friends may provide support, it is likely to be based on anecdotes and personal experiences rather than systematic research. Health professionals have a vital role in providing users with information, but studies have found that evidence based practice and shared decision making have not always been adopted (11, 33-35), and that the quality of patient information varies (6, 13, 14). 15 Health literacy: a public health issue There are many ways to improve people’s access to reliable health information. Such initiatives include the synthesis and tailoring of evidence based information to individuals by experts (36), automated systems for filtering information online (37), publisher’s codes of conduct and the standardisation of initiatives amongst those who publish information (such as the HONcode) (38); and users’ own development of their skills through education (8, 39). Although all of these types of interventions are important and may contribute to improving access to health information, it is only the latter that prepares people to understand the facts that are presented to them, interpret what these facts are based on, and decide whether they are actually applicable to a range of health issues and settings (8). Contemporary healthcare is complex and demands insight as well as the ability to navigate these systems. For users, participation in decision making is not only dependent on access to valid information, but also on the ability to interpret, understand and act upon such information. These are what are often referred to as health literacy skills (19). Health literacy as a field of clinical and scientific interest is fairly new but is rooted in many disciplines and is of interest to many, including health professionals and health psychologists, health services researchers and health educators, (8). The term ‘health literacy’ was used for the first time in 1974 (40) and since then many attempts have been made to define and refine the concept (8, 19, 40-42). The discourse is still very much alive and evolving, and the aims, content and scope of health literacy are still being sketched out. 16 Health literacy as a field of interest has its roots in two approaches that are somewhat different in their aims and scope (19). The first approach, referred to as a risk model or as 16 ‘medical literacy’, sees limited health literacy as a risk factor that should be addressed in a clinical setting (19, 40). The focus of studies rooted in this tradition is typically on the ability of users to read, understand and act on information relevant to healthcare settings. This includes, for example, being able to read and understand package labels, the ability to self-manage disease, adhere to health advice, share decision making, and seek information (19, 40). Health literacy: a public health issue A commonly used definition of health literacy, according to those who subscribe to this approach is: ‘medical literacy’, sees limited health literacy as a risk factor that should be addressed in a clinical setting (19, 40). The focus of studies rooted in this tradition is typically on the ability of users to read, understand and act on information relevant to healthcare settings. This includes, for example, being able to read and understand package labels, the ability to self-manage disease, adhere to health advice, share decision making, and seek information (19, 40). A commonly used definition of health literacy, according to those who subscribe to this approach is: ‘the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions’ p32, (43). According to this approach, the term ‘health literacy’ describes individual literacy capacities as mediating factors predicting health (19). One criticism of this approach is that it is too narrow in scope and individually-oriented and does not include broader social, ecological and systemic perspectives of health and wellbeing (41). Another concern is that this definition does not necessarily incorporate healthy individuals or those health decisions that are made outside clinical settings (40, 41). A broader definition of health literacy has therefore been proposed in which health literacy is seen as an ‘asset’ and as an end outcome that is critical to empowerment (19). Health literacy according to this approach is not only advocated to optimise health outcomes and cost effectiveness, but also for ethical reasons by facilitating equal access to healthcare, autonomy and the right to citizenship (2, 5, 19, 40, 44-47). This broader understanding of health literacy forms the backdrop of this project and is the position increasingly being adopted by governmental and international organisations. The WHO, for example (2), states: 17 ‘Health literacy represents the cognitive and social skills which determine the motivation and ability of individuals to gain access to, understand and use information in ways which promote and maintain good health’ p10, (2) Further, the WHO states: Further, the WHO states: ‘Health literacy implies the achievement of a level of knowledge, personal skills and confidence to take action to improve personal and community health by changing personal lifestyles and living conditions. Thus, health literacy means more than being able to read pamphlets and make appointments. By improving people’s access to health information, and their capacity to use it effectively, health literacy is critical to empowerment’ p10, (2). These definitions show that health literacy is also seen as a preventive and health promoting initiative, which includes a broader societal perspective. Further, they take into account the importance of system knowledge as well as recognising the implications of health literacy on public health (8, 24, 44). Through health literacy, people are enabled to take control of their own life and health, including the personal, social and environmental determinants of their health (19, 40, 41, 44). This understanding of health literacy is particularly important given the prevalence of preventable chronic diseases, including those affecting young people (24). But it is also an understanding that is important to helping to solve public health issues associated with patients who have already become sick. Chronically ill patients are more likely to have contact with health systems and are also required to manage their illness on daily basis in many ways. Such activities have been found to be dependent on levels of health literacy competency (17). 18 Furthermore, improving peoples’ knowledge and skills is seen as a key element of sustainable health and healthcare, and a facilitator of successful policy implementation through shared information, responsibility and accountability (5, 41, 48). What do health literacy skills really entail? Health literacy has been defined and conceptualised in many different ways. Earlier definitions developed by researchers and policymakers, for example, tend to focus on the basic reading, writing and numeracy skills needed for decision making in healthcare – often referred to as functional or fundamental literacy skills (8, 44). Functional health literacy is important and forms the basis for developing other skills, as well as the ability to participate in individual decision making and within society. But reading and writing alone do not address current challenges in public health and clinical care, and an awareness of this has led to greater emphasis on the importance of more advanced critical and social skills (8, 40-42, 44). It for this reason that the conceptualisation of health literacy formulated by Zarcadoolas and colleagues was adopted in this project (8). According to this perspective, a person who is health literate ‘is able to use health concepts and information generatively—applying information to novel situations and to participate in the ongoing public and private dialogues about health, medicine, scientific knowledge and cultural beliefs’ p196, (49). This model of understanding effectively adapts the public health literacy perspective and contains four central domains: fundamental literacy, science literacy, civic literacy, and cultural literacy (see Figure 1) (8). Importantly, these domains include – and are dependent on – several 19 specific interdependent and overlapping skills and knowledge. For example, functional literacy may be a prerequisite for civic literacy, such as the ability to participate in public debates. On the other hand, the skills that a person has in one domain may also help to compensate for their having inadequate skills in another (8). specific interdependent and overlapping skills and knowledge. For example, functional literacy may be a prerequisite for civic literacy, such as the ability to participate in public debates. On the other hand, the skills that a person has in one domain may also help to compensate for their having inadequate skills in another (8). Figure 1. A multi-dimensional model of health literacy by Zarcadoolas and colleagues ( What do health literacy skills really entail? A multi-dimensional model of health literacy Fundamental literacy Reading, writing, speaking and working with numbers Science literacy Skills and abilities for understanding and using (the process of) science and technology, including:  Knowledge of fundamental scientific concepts  Ability to comprehend technical complexity  An understanding of technology  An understanding of scientific uncertainty and that change in accepted science is possible Civic literacy Skills and abilities that enables citizens to become aware of public issues, participate in critical dialogue about them and become involved in the decision making process, including:  Media literacy  Knowledge about civic and governmental systems and processes  Knowledge of power, inequity and other hierarchical relationships  Knowledge that the behaviour and choices of the individual affect others in larger community and society Cultural literacy Skills and abilities to recognice understand and use collective beliefs, customs, worldview and social identity of diverse individuals.  Knowledge about civic and governmental systems and processes  Knowledge of power, inequity and other hierarchical relationships  Knowledge that the behaviour and choices of the individual affect others in larger community and society Skills and abilities to recognice understand and use collective beliefs, customs, worldview and social identity of diverse individuals. Skills and abilities to recognice understand and use collective beliefs, customs, worldview and social identity of diverse individuals. Figure 1. A multi-dimensional model of health literacy by Zarcadoolas and colleagues (8) 20 Health literacy and associated outcomes The relationship between health literacy and health (although generally only measuring functional literacy skills) has been evaluated in many studies, and the findings indicate that low health literacy levels are associated with poorer health, increased healthcare utilisation, The relationship between health literacy and health (although generally only measuring functional literacy skills) has been evaluated in many studies, and the findings indicate that low health literacy levels are associated with poorer health, increased healthcare utilisation, inappropriate drug use, and a low uptake of disease prevention services (such as vaccination) (17, 18). Health literacy has also been found to be a stronger predictor of health status than age, income, employment status, education level and ethnicity (6). Such outcomes may have serious consequences for individuals, but people’s health literacy skills may also be important on a systemic level. Many health systems face major challenges including increases in healthcare expenses and the need to prioritise issues (5, 8). One such issue is the development of new and costly treatments and screening programmes, which require limited resources to be prioritised in appropriate ways in the face of difficult ethical and socio-economic considerations. Public engagement – and appropriate health literacy skills among the public – can help to facilitate sustainable healthcare systems by that users have insight into the ingredients of such decision making. It is therefore concerning that levels of health literacy skills have often been found to be suboptimal (6, 18). Lower health literacy levels have been found to be more prevalent among lower socioeconomic groups, ethnic minorities, the elderly and those with chronic conditions or disabilities (6, 50, 51). These associations may also be the reasons why the improvement of public health literacy is seen as integral to reducing health inequalities (6). 21 It should be noted that research exploring the prevalence of health literacy skills has been limited in two ways: firstly, most studies have been conducted in the United States of America 21 (USA) and it is difficult therefore to estimate the health literacy level in other contexts such as Norway or Europe (6). Secondly, health literacy has not always been measured consistently and most studies until now have primarily evaluated functional skills (52). The European Health Literacy Survey (HLS-EU) aims to bridge these knowledge gaps by initiating the development of a tool to measure health literacy in Europe and by conducting a cross-sectional Europe-based study (53). Health literacy and associated outcomes The project, funded by the European Union, is hosted by Maastricht University and is a collaborative effort involving eight partner countries (53). The results from this study, following evaluations conducted in Austria, Bulgaria, Germany, Greece, Ireland, the Netherlands, Poland and Spain, are expected to be published in early 2012 (53). (USA) and it is difficult therefore to estimate the health literacy level in other contexts such as Norway or Europe (6). Secondly, health literacy has not always been measured consistently and most studies until now have primarily evaluated functional skills (52). The European Health Literacy Survey (HLS-EU) aims to bridge these knowledge gaps by initiating the development of a tool to measure health literacy in Europe and by conducting a cross-sectional Europe-based study (53). The project, funded by the European Union, is hosted by Maastricht University and is a collaborative effort involving eight partner countries (53). The results from this study, following evaluations conducted in Austria, Bulgaria, Germany, Greece, Ireland, the Netherlands, Poland and Spain, are expected to be published in early 2012 (53). Tools for mapping all relevant health literacy domains may not yet be available but studies have explored important aspects of health literacy. These include: that people rely on sources that are not necessarily evidence based and use sub- optimal critera to evaluate the information they find (13, 27, 54, 55); do not have insight into the ingredients of decision making about treatments (such as screening for breast cancer and childhood vaccination) including principles and processes of medical and health related research (56-60). Furthermore, studies have explored people’s understanding of healthcare policy, such as the reasons for policy switches between using branded or generic drugs, or the rationales for preventive policies. These studies suggest that people may think that information is withheld or that implementation of a drug policy is solely driven by cost-concerns without consideration of safety and efficacy (56, 58, 61, 62). The results of these studies have highlighted the great potential for improving health literacy and point to the fact that people’s health literacy skills are poor in terms of one or more of the four domains defined by Zarcadoolas and colleagues (8). 22 Previous interventions targeting health literacy Presently, most interventions have been developed to improve the functional literacy of patients. Such interventions have adopted a medical literacy approach and have been limited to evaluating the abilities of patients to read, use numbers and the impacts of this on context- specific health behaviour such as adherence to treatment regimes (6, 17, 18). But advice based on reviews of the health literacy literature and studies of conceptual development have extended this focus to include recommendations about the development of initiatives that target critical and interactive skills. This focus goes beyond functional literacy to include a public health perspective (6, 8, 17, 18). Further interventions are needed that target ‘generic’ health literacy skills, and which are not restricted to a specific health decision or illness (63). Only a few studies have been done in this area, targeting skills such as science literacy and critical appraisal of health information (39, 64). The field of research related to user-directed interventions to enable health literacy skills is in rapid development. Such interventions can be broadly divided into three main categories: written information, alternative format resources (such as health informatics), and mass media interventions (6). These main categories are not necessarily exclusive, and interventions can include one, two, or all of them. Evidence suggest that interactive solutions, such as websites, may be more effective educational strategies, and have been associated with improved knowledge, high user satisfaction, and beneficial effects on self efficacy and health behaviours (6, 65-67). 23 Complex problems may require complex interventions The methodological development of research on change and quality improvement is an important and expanding new field of research (68, 69). Such research is often complex given the many potential intervention points and components (68). Methodologists and health services researchers argue that it is unrealistic to expect that one simple improvement measure would be enough to solve all problems, and that research questions are best addressed by what are often referred to as complex interventions (68). These are characterised by several interacting components and use multiple research methods, including both qualitative and quantitative designs (68, 69). Moreover, the complexity of these interventions may also include variations in the number and complexity of behaviours studied, the inclusion of heterogeneous target groups, and the measurement of a range of different outcomes (69, 70). The purpose of complex interventions is usually to optimise patient care and improve health and life quality, but it is also to develop sustainable strategies that work in practice, not just in the laboratory (69, 70). This multi-disciplinary field draws upon knowledge and methodologies within epidemiology, the social sciences, educational sciences, organisation and management science, and economics (68). Several methodological frameworks for complex interventions exist and a single optimal one has yet to be found. However, there are guiding rules that have been advocated by methodologists (69, 70). For instance, an explorative yet systematic and transparent approach is encouraged when designing complex interventions (69, 70). A complex intervention is also generally defined as including several steps or phases (although these may not necessarily follow in a linear sequence) (69, 70). The model by van Boekhoven and colleagues (2004) provides a transparent overview of the phases they included and was used to inform the overall design of 24 this project. These phases are: the identification and analysis of the problem, the development of intervention, piloting/feasibility testing, and implementation and evaluation (see Figure 2) (70). Central to the problem formulation phase is the identification of important barriers and facilitators to change (68-71). This can be done through literature searches, the involvement of target group representatives or other stakeholders, and by conducting qualitative and quantitative explorative studies (68-70, 72). An intervention can then be tailored to address such barriers and facilitators through the choice of the form of intervention delivery and its content. Complex problems may require complex interventions In many cases, more facilitators and barriers are identified than it is possible to address within a single intervention, and choices need to be made about which ones to prioritise (70). These choices depend on their perceived importance, changeability, and a consideration of practical issues such as the resources available (70). Complex interventions may be evaluated using several designs. However, depending on the actual research question, some designs may be better suited than others (73). Whereas qualitative designs and cross-sectional studies may be useful tools for feasibility or process evaluations, experimental designs – and, more specifically, randomised controlled trials – are best suited and introduce less risk of bias when the effects of the interventions are evaluated (69, 73). 25 Figure 2. Model describing phases of complex interventions by van Boekhoven (70) Figure 2. Model describing phases of complex interventions by van Boekhoven (70) Figure 2. Model describing phases of complex interventions by van Boekhoven and colleagues (70) (70) 26 Parents’ decision making as a case Although the target group of the intervention developed in this project was the public in general (and all health decisions), decision making by parents was used as the case for this thesis. There were several reasons for this. At this life stage, parents typically have many questions about treating and preventing health problems. They are also healthcare users with the highest number of health visits per year, both for themselves (a mean number of visits per year of 4.6), and for their children (mean number of visits of 3) (74). Parents, generally being young, are also statistically more likely to search for health information online (28). Furthermore, parents are faced with a range of health issues to consider. One such issue is deciding about child vaccination, and this was chosen as the particular focus in this project. We assumed that when we examined the issue of decision making it would be easier for participants if the discussion related to a concrete decision. Moreover, decision making about vaccination is complex and reflects many of the changes in contemporary healthcare discussed earlier in the Introduction of this thesis. For example, childhood vaccination has been a topic for discussion in the media for many years, and numerous strong and often conflicting opinions about both important benefits and serious side-effects have been expressed in the public debate (15, 58, 75-77). Finally, given that vaccination is voluntary in Norway, this option was seen as a health decision that the majority of the population would need to make. 27 Aims The overall goal of this project was to improve people’s health literacy skills related to obtaining health information through an intervention targeting key identified barriers and facilitators, and to evaluate this effort. This objective was addressed using both quantitative and qualitative methods, and included three phases (see also Figure 3): Phase 1: Analysing the problem and identifying barriers and facilitators to change a. Firstly, we conducted qualitative interviews with parents and public health nurses in order to:  Explore decision making related to a common case (childhood vaccination) and what informed such decisions, and to  Identify perceived barriers and facilitators to obtaining information so that these could inform the development of a questionnaire to further explore them and enable them be targeted by the intervention. These findings are presented in Paper I and II. b. Secondly, we developed and conducted a questionnaire study based on the findings of the previous qualitative study, supplemented by a literature search in order to:  Identify important predictors associated with intention to search for health information so that these could be targeted by the intervention (Phase 2), and to  Provide a tool that we could use to evaluate the effects of the web portal in the randomised controlled trial (Phase 3). These findings are presented in Paper III. 28 Phase 2: Designing the intervention In the second phase of the project, we developed a web portal, integrating the lessons from the web portal pre-studies with findings from a literature search. The purpose of this phase was to:  Tailor the web portal in a transparent and systematic way, so that it addressed key barriers to obtaining reliable health information using evidence based strategies. This development is presented in Paper IV. is presented in Paper IV. Phase 3: Evaluating the intervention Phase 3: Evaluating the intervention The final phase of the project included an evaluation of the web portal in a pragmatic randomised controlled trial. The purpose of this was to:  Evaluate the effects of the web portal intervention compared to no intervention in a real life setting. These findings are presented in Paper V. As Figure 3 shows, we also planned for user testing which would involve potential members of our target groups. Aims Given the size of this project, this work was not lead by me but by a Masters student in Clinical Nursing Science, Stein Ove Danielsen, at the Oslo University College, who was invited to prepare and conduct this user testing work. This work was undertaken as a part of his masters thesis and is therefore not included in this dissertation (78). However, the findings of his study were important for the development of the web portal, and the methods and main results are therefore briefly referenced in this dissertation, with the permission of Stein Ove Danielsen. 29 Figure 3. Project outline showing the three phases Figure 3. Project outline showing the three phases 30 Analysing the problem and identifying barriers and facilitators to change When planning an intervention, the use of qualitative studies is highly recommended for exploring concerns that really matter to users (72, 79). Using a grounded theory approach and semi-structured individual and focus group interviews, we explored the decision making about childhood vaccination made by parents of pre-school children and by public health nurses. In this study we aimed to identify perceived barriers and facilitators to obtaining health information (80- 83). Grounded theory has been found to be particularly well-suited when research is explorative, and is a systematic yet flexible approach for ‘learning about the worlds we study’, and is ‘a method of constructing theory grounded in the data itself’ p10,(80). Parents and public health nurses were recruited from the same maternal and child health centres. Three districts in a major Norwegian city were chosen using strategic sampling to represent diverse socio-economic backgrounds. These included one maternal and child health centre on the west side, one on the east side, and one in the city centre which serves a mixed population. Parents of pre-school children who had made a decision about childhood vaccination within the previous three months were invited to participate, and a total of 10 parents was included. We aimed strategically in our initial recruiting to include those characteristics previously identified as influencing the studied behaviour (i.e. information retrieval), such as age, sex and education (27, 84). The public health nurses were recruited using pragmatic convenience sampling; all the public health nurses who volunteered – a total of 16 – were included. The interviews were led by 31 me, assisted by a nursing student who was the interview secretary, and the sessions were taped and transcribed as part of the analysis process. Data were coded ‘incident by incident’ so that key concepts could be identified, and the analysis was performed continuously (80). In the final stage of the analysis we created a chart based on the principles of axial coding to explore connections between the categories (80). To complement the grounded theory approach, we drew also on framework analysis when managing the data (85). To improve the credibility of the reading and interpretation of the data, this was done independently by me and an additional researcher who was trained as a public health nurse. Analysing the problem and identifying barriers and facilitators to change The interview guide included general topics relating to what had informed the decision to vaccinate, as well as the facilitators and barriers to searching for health information (see Appendix 1 for a copy of the interview guides). me, assisted by a nursing student who was the interview secretary, and the sessions were taped and transcribed as part of the analysis process. Data were coded ‘incident by incident’ so that key concepts could be identified, and the analysis was performed continuously (80). In the final stage of the analysis we created a chart based on the principles of axial coding to explore connections between the categories (80). To complement the grounded theory approach, we drew also on framework analysis when managing the data (85). To improve the credibility of the reading and interpretation of the data, this was done independently by me and an additional researcher who was trained as a public health nurse. The interview guide included general topics relating to what had informed the decision to vaccinate, as well as the facilitators and barriers to searching for health information (see Appendix 1 for a copy of the interview guides). The use of evidence based theory may be useful for tailoring interventions, providing the development process with structure and transparancy (86, 87). We developed and conducted a questionnaire study, based on the theory of planned behaviour (TPB), to explore which cognitive factors predict the independent search for health information outside consultations (79, 88). The TPB is a rigorously tested and widely used social cognition model (79, 89-91) and previous studies have used the TPB model to explore behavioural intentions across a range of health related behaviours and professional practice (92-98). The TPB model has also been applied successfully in the tailoring and evaluation of interventions (79, 92, 98-100). The questionnaire was developed based on an operationalisation of the Theory of Planned Behavior Manual by Francis and colleagues, that provided us with pragmatic and evidence based guidelines throughout the development process (79). According to the TPB model, three variables influence behavioural intentions, namely: attitudes towards the behaviour, subjective 32 norms, and perceived behavioural control (79, 88). These variables are measured using both direct and indirect measures. Whereas direct measures describe respondents’ overall assessment, indirect measures probe the underlying specific beliefs and outcome evaluations associated with these overall assessments (79). Analysing the problem and identifying barriers and facilitators to change Standardised items were used to measure intention to search and direct measures of variables, as recommended by the TPB manual. Items describing indirect measures were created based on beliefs identified in the qualitative study described above and through a literature search (79). The application of the TPB questionnaire was intended to explore, identify and evaluate beliefs associated with the search for information related to an intervention to improve specific domains of health literacy. This aim was also reflected in the choice of the specific beliefs included in the questionnaire. Consequently, the specific beliefs chosen were (changeable) factors related to domains of health literacy (with the exception of subjective norm). The questionnaire was tested twice on two samples. The first sample was taken from the general population. A sample of 100 men and 100 women was drawn from three separate districts of Oslo, the capital city of Norway (these were the same district divisions used in the qualitative study described above). In order to include those who were most likely to have internet access, we decided to exclude people over the age of 65. The questionnaire was administered by post and included a pre-stamped envelope. One reminder letter was sent, and a lottery for a small gift certificate (value 750 NOK) was held as this has been shown to improve response rates (101). 33 The questionnaire was tested twice on two samples. The first sample was taken from the general population. A sample of 100 men and 100 women was drawn from three separate districts of Oslo, the capital city of Norway (these were the same district divisions used in the qualitative study described above). In order to include those who were most likely to have internet access, we decided to exclude people over the age of 65. The questionnaire was administered by post and included a pre-stamped envelope. One reminder letter was sent, and a lottery for a small gift certificate (value 750 NOK) was held as this has been shown to improve response rates (101). The second sample consisted of parents. The questionnaires were administered face-to-face at the local maternal and child health centre and associated shopping centre in a district of Oslo and we aimed to include 80 respondents. All data were entered according to the manual guidelines for the SPSS statistical program, and the items were recoded (79). Analysing the problem and identifying barriers and facilitators to change Datasets were explored to identify 33 The second sample consisted of parents. The questionnaires were administered face-to-face at the local maternal and child health centre and associated shopping centre in a district of Oslo and we aimed to include 80 respondents. All data were entered according to the manual guidelines for the SPSS statistical program, and the items were recoded (79). Datasets were explored to identify 33 missing and illogical values (79). New composite scores were created for intention, direct and indirect measurements and descriptive analysis was done to describe the background characteristics and the distributions of the measures (79). The questionnaire was tested for internal consistency by calculating inter-correlations between theoretically-related constructs, and by using multiple regression analysis. This was done by exploring: a) the internal structure (reliability) of the items measuring the same construct of intention to search, as well as the homogeneity of direct measures (where Cronbach’s Alphas values above 0.6 were considered to be indicators of acceptable consistency) (79), b) the nomological network proposed by the TPB model (by computing simple bivariate correlations using Pearson R between the mean composite scores of the direct, indirect, and intention to search measures) (89), and c) the relationship between indirect and direct measures through a multiple regression analysis in which the mean composite score of the indirect measures for each construct were the independent variables and the corresponding mean composite direct measure were the dependent variables (79). Finally, we tested the overall potential of the questionnaire for predicting intention to search, and identified the predictive strength of each direct measure, by performing a multiple regression analysis in which the three mean composite scores of each direct measure (attitude, subjective norm, and perceived behavioural control) were the independent variables, and the mean composite score of intention to search measures was the dependent variable (79). 34 Designing the intervention Decisions related to the content and main focus of our intervention, namely which specific health literacy skills to improve, were informed by the pre-studies described above as well as a review of existing literature identified through literature searches (102-104). The findings from these suggested several areas of improvement related to all the identified domains of health literacy: functional literacy (numeracy), science literacy, civic literacy (system and roles) and cultural literacy (the understanding of concepts related to decision making about healthcare). These were grouped into three key barriers related to obtaining information. Following this, potential targets for intervention were identified and these are briefly described below. Barrier 2: The inability to exchange information in consultations A second major barrier to obtaining information is the inability of users to exchange information which is important to their decision making during consultations with health professionals (102). Consultations are an important arena for health information exchange. However, studies have found that these exchanges are often sub-optimal (9-11, 20, 102, 104). Often users are unaware of their rights, what to ask their provider and may not be able to remember what they have been told (9, 10, 20, 102). Barrier 1: The inability to understand and critically appraise health information We identified the inability to understand and critically appraise health information as a key barrier to obtaining information (102, 103). Understanding research information and the quality criteria for assessing health information is essential, and studies have found that people may not effectively check the accuracy of health information they find, may overrate the trustworthyness of such information, and have poor understanding of health and medical related research (13, 25, 27, 54, 60, 105). In particular, concepts such as ‘randomisation’, ‘risk’, ‘uncertainty’, ‘causality’ and ‘applicability’ need to be addressed (56, 57, 59, 60, 105-109). 35 Barrier 3: Not knowing where to find reliable and relevant information The third major barrier we identified to obtaining health information is the inability of users to know where and how to find reliable and relevant information (102, 103). Specifically, users feel overwhelmed and frustrated by the vast amount of information available and unsure about who or what they should trust (27, 110). Furthermore, evidence based information is not readily available to the public (12-14, 16). As a result, user searches for health information may be haphazard and rely on general online searching engines (26, 28), or advice from family and friends, news sites or discussion forums that may potentially contain information of poor scientific quality (12-16, 111). Our web portal was therefore tailored to address these three key barriers to obtaining information and below we describe the decisions taken and strategies used to address these barriers. 36 The choice of a web portal as the intervention delivery approach was based on our reviews of related literature which indicated that web-based interventions may be more sustainable and more effective than conventional information strategies (6, 65, 66). The target audience of our web portal was the general public and our aim was to provide an easily accessible resource for lay people searching for health information that they could either use independently or during consultations with health professionals. We chose a design template for the web portal that met recommended accessibility standards (112) and presented the content in compliance with the standards recommended by Health On the Net Foundation’s ‘Code of Conduct for Medical and Health Web Sites (HONcode)’ (38). Furthermore, we sought to root the development of the content within the conceptual frameworks of shared decision making and evidence-based practice (1, 5, 113). This meant that we aimed to: encourage active user involvement; promote the importance of basing decisions on the best available evidence; and to include content that could facilitate skills and the development of knowledge (rather than adherence to any specific regime). The content was presented using mixed media, real-life examples, plain language, and the provision of information in small and ‘digestible’ quantities (6, 20, 114-116). Although the purpose of the web portal was educational, a key objective was also to develop an easy-access point to practical tools and evidence. We therefore organised the content in the form of a set of tools. Barrier 3: Not knowing where to find reliable and relevant information We decided to have three facilitators – or toolsets – to address each of the three main barriers to obtaining information, and these are briefly described below: 37 Toolset 1: Improving critical appraisal skills We provided an introduction to the basic research methods of medical- and health-related research (8, 113, 117-121). We also searched systematically in the literature for checklists that would enable users to critically appraise health information by addressing key quality domains (122, 123). For a checklist to be included in the web portal, it had to be generic and applicable to all settings and preferably evaluated for construct validity and inter-observer reliability. All the references identified were read both by a fellow researcher and by me, and all the potentially relevant tools were retrieved. The final selection was done by me and reviewed by the fellow researcher. Studies that have aimed to improve users understanding of research and critical appraisal skills is scarce, but suggest that such efforts are feasible, may improve knowledge and confidence and are perceived as useful by users (39, 64, 124). Standardised tools such as checklists used for critical appraisal may produce ratings that are similar to that of professionals and assist the user in selecting information (125). Toolset 2: Enabling exchange of health information We decided to include basic information about the rights of patients to participate in decision making (3). In addition, we included information about what decision making related to treatment or screening options entails (113, 117, 118, 126). We also searched systematically in the literature for checklists or decision aids that could be used in consultations. Decision aids have been found to be effective tools for improving peoples’ knowledge of health options, their 38 expectations of possible benefits and harms, their ability to make decisions consistent with their informed values, and increasing participation in decision making (127). expectations of possible benefits and harms, their ability to make decisions consistent with their informed values, and increasing participation in decision making (127). To be included, the tools had to be generic and applicable to all types of consultations. The purpose of the tool also had to be to provide information about the condition of the users and the options available, to help to people clarify their values, and to help them to share their values with their health provider and others (128). The final selection was done by me and reviewed by my fellow researcher. Toolset 3: Improved access to reliable research-based sources of health information We aimed to improve access to reliable sources of health information by providing direct access to medical- and health-related research databases. The key criteria for selecting databases were: that the information should be based on empirical research, that the databases should reflect different study types and levels of synthesis, rely on explicit and systematic criteria (if reviews of summaries of research), and be freely available to the Norwegian public. The final database selection was done jointly through a discussion in the project group and in consideration of the above criteria. Furthermore, we decided to build this toolset section using the guidelines for evidence based practice, adapted to a lay audience by us (1, 113). 39 Evaluating the intervention Evaluation should be a central element in the planning of all interventions (69, 70) Evaluation should be a central element in the planning of all interventions (69, 70). Important foci for such evaluations include feedback about user experiences and about relevance (69, 70). Throughout the development process, the project protocol, the selection of the tools, and the final portal itself were presented to an advisory group for peer review and professional feedback. In addition, usability testing was conducted through a qualitative evaluation in which a mixed group of 17 people (including representatives from patient organisations as well as healthy users), were invited to evaluate and comment upon their user-experiences of the web portal. This evaluation was based Morvilles Honeycomb model (78). This model, a useful instrument frequently applied to measurements of user experiences of websites, encompasses seven domains to assess whether a website is: accessible, usable, credible, valuable, findable, desirable, and useful (78, 112). In specific analysis, the domain ‘findable’ was not tested because the web portal had not yet been made official. Based on feedback from the advisory group and the user representatives, the web portal was updated and finalised before it was evaluated in a randomised controlled trial. In order to evaluate the effects of the web portal in a real life setting, we chose a randomised controlled parallel design with a simple randomisation procedure. We wanted to include typical target-group users for the web portal in the study’s participant sample. In addition, we wished to increase the probability that the portal would be used by participants in association with their visits to health professionals during the trial. Parents with children under the age of 4-years were therefore targeted, as they were highly likely to search for health information and to 40 have contact with the health services (28, 74). Similarly, the issue of vaccination was selected given that it is a very relevant topic for this particular group. Based on a power-analysis, we estimated that a sample size of approximately 140 people was needed. Information about the study was distributed at maternity and child health centres and through online advertisements and social media. Those who were interested and wished to participate were directed to a recruitment web page. Here, people received information about the study and were asked for their consent to participate and to fill in an online questionnaire that enabled inclusion criteria screening. Evaluating the intervention People were excluded if anyone else in their household was already participating in the study (to ensure that the participants were blinded and to protect against potential sample contamination), and if they did not have children aged <4-years. Parents were allocated to receive either access to the portal or no intervention, and assigned three tasks to perform over a three-week period. These tasks were: a searching task in which parents were asked to formulate a question about a health issue and to find information related to this question; a critical appraisal task about online swine flu vaccination information; and reporting on perceptions about participation related to independent search for health information and overall activation. All the information was delivered online and data were collected using an online questionnaire system. The study was single blinded in that participants were not informed about which study group they would be allocated to. We planned to collect data from March to the end of June 2011. The main outcomes were: beliefs about searching for health information and overall activation (participation), use of research-based information, and the development of critical appraisal skills. In addition, we also wanted to obtain feedback from the participants 41 regarding their satisfaction with the web portal. Participation was measured using our previously developed TPB questionnaire and the patient activation measure (PAM) (129). The use of evidence based information was measured by calculating the relative risk between groups in terms of their use of research, and critical skills were measured using the DISCERN tool (125). Satisfaction with the portal was measured using Morvilles Honeycomb criteria (112) . All analyses were performed using available cases, but based on the principle of intention to treat (ITT) (i.e. all participants were analysed in the group to which they were randomly assigned). Given the opportunity that this trial offered in terms of further testing the utility of the TPB questionnaire, we decided therefore to test the internal reliability of the item measuring intention and direct measures. We also tested the relationship between the mean composite scores of these measures (the nomological network) for the responses of the whole sample in the same way as was done in Paper III. Data were treated anonymously and ethical approval was granted by the Norwegian Social Science Data Services (NSD) and the Regional Committees for Medical and Health Research Ethics (REK), Norway. Evaluating the intervention The trial was registered at ClinicalTrials.gov under the identification number NCT01266798. 42 Decision making by parents, and barriers and facilitators to obtaining information In the qualitative study, we found that parental decision making about childhood vaccination was based on both trust and common sense. ‘Being positive towards vaccination’ and ‘being decided’ were found to be main barriers to participation and obtaining information. Other factors were own ‘abilities’ and ‘capacity’. Consequently the parents were found to have had little involvement in the decision making, and in influencing the information obtained during consultation and their own searching behaviour. Parents suggested that their searches for information would be facilitated if public health nurses were able to recommend reliable internet sites to those who wanted information beyond what had been given during consultations. Although the level of parents’ involvement in decision making was low, they had significant expectations of public health nurses and the nurses were seen as their most important source of information. Parents emphasized the public health nurses’ crucial role. However, the public health nurses were found to inform to facilitate vaccinations and not an informed decision in itself. Thus, potentially leaving those who where positive towards vaccinations less informed than those being more critical. The public health nurses were also not necessarily considered by the parents to be balanced in the way they provided information, and were perceived as being biased towards the positive effects of vaccination. There was also confusion about the discrepancy between the information presented by public health nurses and what the parents had 43 learned through media reports or stories from family and friends, particularly with regard to the potential harms associated with vaccination. learned through media reports or stories from family and friends, particularly with regard to the potential harms associated with vaccination. Issues related to the small amount of parental involvement included parents reporting that they had little knowledge about the diseases and about vaccination. Furthermore, some expressed low levels of confidence about the decision they had made and uncertainty about their rights and responsibilities in decision making. The public health nurses only occasionally looked up extra information in their work related to childhood vaccinations and this information rarely included research. Instead, the most important sources of information the nurses consulted about vaccination were the Norwegian national guidelines and other information issued by Norwegian National Institute of Public Health. They also consulted peers, superiors, other health personnel, referred to their own professional experience, textbooks, mass media reports, the internet, and the information from the pharmaceutical industry. Decision making by parents, and barriers and facilitators to obtaining information The public health nurses stressed the importance of providing the best possible quality care and reliable research was considered central to this. Three main facilitators to the use of research were identified during the focus groups: being able to base practice on solid knowledge, the belief that the use of research was important for professional development and necessary, and the need to meet parental demand for extra information. Despite these facilitators, the nurses were reluctant to do so. The main three barriers identified that prevented them from doing so were: a belief that searching for research findings and using them was ‘not part of their role’; and their perceived ‘limited critical appraisal skills’ and ‘capacity’. Thus, research was seen as something that should be managed centrally and quality-approved by a higher authority. 44 In the questionnaire study we quantified some of the findings from the qualitative study. The response rate to our questionnaire was low, with only 16% (n=30) in the mixed population sample and 43% (n=45) in the parent sample. The distribution of age and sex in the mixed population was evenly distributed. In the parent sample, females and people in the age category 31-40 were overrepresented. In both samples, those with higher education were also overrepresented. 45 In the reliability analysis of the items measuring each direct measure respectively, one item (measuring perceived behavioural control) was found to be very poorly inter-item correlated with the other items which described the same construct across both samples, and was therefore subsequently deleted. The Cronbach’s Alpha for the remaining items ranged from satisfactory (0.66) to very good (0.98) across the samples. When examining the theoretical relationship between variables (nomological network), the mean composite scores of direct measures and intention were found to be positively and significantly correlated with intention to search, with the exception of subjective norm which was not found to be significant. Inter-correlations were also found between the direct measures, where subjective norm and perceived behavioural control correlated with attitude. Five single items measuring indirect measures, which were found to be very poorly inter-item correlated with their corresponding direct measure, were deleted. The remaining items and mean composite indirect measures had positive and significant correlations with their corresponding direct measures, with the exception of perceived behavioural control in the parent cohort, which was only borderline significant (p=0.07). Choice of intervention delivery and content The portal was developed based on á priori design principles and published on the site www.sunnskepsis.no. It met the recommended standards for website accessibility and was in compliance with the HONcode (38). Overall, the portal content was illustrated using practical examples that people could relate to. The content was presented in ways that encouraged users to adopt an active role in decision making and encouraged them to see that decisions about healthcare should be informed by the best available, current, valid, and relevant evidence. Central to the design of the structure of the web portal were three sets of tools, each presented in an easily accessible and logically structured way. These are briefly described below: Results Toolset 1: Improving critical appraisal skills General introductions to research methodology, research ethics, and applicability of research findings were provided throughout the portal (8, 113, 117-121). In our search for critical appraisal tools, only one tool met all our inclusion criteria (125) and this – the Norwegian version of DISCERN – was included both as a printable checklist and as an interactive tool (130). Decision making by parents, and barriers and facilitators to obtaining information The association between the mean composite indirect measures and intention to search followed the same pattern as the direct measures. In the regression analysis, the composite indirect measures in the mixed and parent samples explained 53% and 48% of the variance in direct attitude, 25% and 45 12% of the direct subjective norm, and 20% and 8% in the direct perceived behavioural control components respectively. The final questionnaire contained 31 items as well as a set of demographic background questions (see Appendix 2 for a copy of the questionnaire, items 1-20). Overall, the mean intention to search was high in both samples: 4.9 for the mixed population (SD=1.86) and 5.4 (SD=1.09) for the parent sample (range from 1 to 7). In both samples, the intention to search was higher for those with college or university education. The assessments of the mean direct measures for the mixed population and parent samples respectively were: 5.4 (SD 1.5) and 5.6 (SD 1.3) for attitudes, 3.3 (SD 1.3) and 3.6 (SD 1.5) for subjective norm, and finally 4.7 (1.6), and 5.2 (1.2) for perceived behavioural control (ranged from 1 to 7). For indirect measures, the mean assessments of the composite scores for the mixed population and parent samples respectively were: 45.0 (SD 25.6) and 40.3 (SD 26.3) of a possible -84 to + 84 for attitude, 13.9 (SD 18.4) and 14.3 (SD 17.5) of a possible -63 to +63 for subjective norm, and -12.6 (SD 23.6) and -3.0 (SD 19.8) of a possible -63 to +63 for perceived behavioural control. The direct measures of overall prediction of intention to search accounted for 47% of the variance in behavioural intentions in the parent sample and 55% in the mixed population respectively. The strongest positive predictor of intention to search was perceived behavioural control in the mixed population sample (β=.66, p=.00), followed by attitudes β=.38, p=.09. In the parents sample, the strongest predictors were attitudes (β= .51, p=.00), followed by perceived behavioural control .15, p= .25. In both samples, subjective norm had a negative predictive effect, with β=-.08, p=71 in the mixed population sample, and β=-.25, p=.01 in the parent sample. 46 Results Toolset 3: Improved access to reliable sources of health information We developed a toolset which was based on the steps of evidence based practice (1, 113). These steps were adapted for the lay public to facilitate direct access to research-based information and to provide a path through the information retrieval process. This was done by giving an introduction to problem formulation and the corresponding study design(s), a description of each database, and tips on easy search strategies and how to assess the relevance of the information (113). We also described different levels of synthesis (113, 132). Results Toolset 2: Enabling exchange of health information We provided an introduction to decision making about treatment and screening and described the right to participate in decision making (3, 113, 117, 118, 126). In our search for checklists or decision aids, we only identified two tools that met our criteria (118, 131). We 47 included the tool developed by Irwig and colleagues and this was translated into Norwegian and adapted to the web portal with the permission of the authors (118). Because the decision aid did not include an item related to user health conditions/diagnosis, it was further supplemented using IPDASI criteria (128). included the tool developed by Irwig and colleagues and this was translated into Norwegian and adapted to the web portal with the permission of the authors (118). Because the decision aid did not include an item related to user health conditions/diagnosis, it was further supplemented using IPDASI criteria (128). Usability and effectiveness of the web portal 48 The feedback from the advisory group proved valuable for the quality assurance of our project. Furthermore, the results from the usability testing with members of our potential target groups were encouraging: the overall rating of the web portal was in keeping with optimal estimates for a beta-version of a website (approximately 70%) (78). The summary of the ratings per usability category from the interviews is presented in Appendix 3. In general, all the suggestions made led to changes. In cases where the feedback was not taken into account, this was because the proposals conflicted with the intention of the web portal to improve health 48 literacy skills or because the suggestions were not technically possible. These suggestions are explored in more detail in Appendix 4 which provides a summary of the suggested changes and the final solutions chosen. The final web portal was evaluated in a randomised controlled trial. A total of 100 participants signed up for the study. Of these, four did not meet the inclusion criteria. The remaining 96 were included in the study: 47 were randomised to the intervention group and 49 to the control group. The overall response rates for the intervention group and control group were 60% (n=28) and 80% (n=39) respectively. There was a small but borderline statistically significant (p<0.06) difference between the intervention and control group in terms of loss to follow-up: more were seen to drop out of the intervention group. Background characteristics were similar across those who participated in one or more of tasks, but the mean level of education was slightly higher in the control group. An overview of the ingredients of the intervention and the variables hypothesised to influence the study’s outcomes is provided in Table 1. This describes the specific health literacy domain that each of these variables is thought to touch upon. In the table, I also describe which instruments were used to measure each domain. In the searching task, two research-based sources were identified in the intervention group, and one in the control group. The relative risk was 2.8 (CI 0.3-29.2) (p=0.39) favouring the intervention group. In the critical appraisal task, the mean rating of the information was 2.41 (SD 0.80) by the intervention group and 2.44 (SD 1.02) by the control. The mean difference was -0.03 (p=0.904) favouring the intervention group. Usability and effectiveness of the web portal 49 The overall activation (PAM) score was 66.5 in the intervention group and 61.9 in the control group (possible range 1-100 with better indicated by higher), and the mean difference was 4.61 (p=0.20) in favour of the intervention group. In the reporting of respondent beliefs associated with searching, a statistically significant difference of 0.63 was found for direct assessment of attitude: 0.6 (p=0.03) in favour of the intervention group. The mean the differences for the direct assessment of subjective norm (-0.2; p=0.49) and perceived behavioural control (0.41; p=0.15), as well as the indirect measures, were not statistically significant. The direct measurement of attitude was the most important positive predictor of intention (B=0.51; p<0.002), whereas the predictive strength of direct measures of subjective norm and perceived behaviour control were -0.15 (p=0.25) and -0.06 (p=0.72) respectively. The internal consistency (Cronbach’s Alpha) of the TPB questionnaire’s main components (intention and direct measures) for the complete sample ranged from 0.64 to 0.90. The inter-correlations between theoretically-related constructs was consistent with previous testings (see Appendix 5) The mean respondent usefulness rating of the web portal was 4.71 (SD 1.11), the mean usability rating was 4.14 (SD 0.97), and the mean credibility rating was 4.75 (SD 0.93) (possible range 1-7). Table 1. Overview of the intervention components, corresponding health literacy domains hypothesised targeted and measurements to evaluate these Usability and effectiveness of the web portal 50 51 Barriers identified in pre-studies and literature search Facilitators/content of intervention Health literacy domains Evaluated in pragmatic trial All Shared decision making (promoting an active role) and evidence based practice as conceptual framework (promoting evidence based decisions) Civic literacy (system and relationships) Science literacy TPB *(attitude and subjective norms associated with search) PAM** Inability to understand and critically appraise health information Improving critical appraisal skills Introduction to scientific concepts and (checklist for) evaluating trustworthiness of health information Science literacy Examples: Validity, uncertainty, causality Functional literacy (numeracy) Example: Understanding risk Civic literacy (media literacy) Examples: How research and scientific discourse are presented in the media Searching task Critical appraisal task TPB (perceived behavioural control and attitudes towards search) Not knowing where to find reliable and relevant information Improved access to reliable research based sources of health information Introduction to searching for evidence based information (adapted EBP-model) Science literacy Examples: Basic study designs and assessment of relevance Civic literacy (media literacy) Examples: Search strategies, publication types and sources Searching task TPB (perceived behavioural control and attitudes towards search) Inability to exchange information in consultations Enabling exchange of health information Introduction to clinical decision making and checklist for the consultation Science literacy Civic literacy (system and relationships) Cultural literacy (understanding of concepts used in decision making about health care) PAM *Theory of planned behaviour **Patient activation measure Table 1. Overview of the intervention components, corresponding health literacy domains hypothesised targeted and measurements to evaluate these *Theory of planned behaviour **Patient activation measure 51 Discussion This project contributes new knowledge to several areas. Although the included sub- studies are part of a greater whole – the purpose of which was to improve health literacy skills related to obtain health information – these should also be seen as independent studies describing specific areas related to decision making by the users of healthcare. Previous research exploring decision making and participation related to childhood vaccination has generally targeted attitudes towards vaccination or adopted a provider-led decision making perspective. It has also focused on issues of adherence and sub-optimal vaccination coverage. Although these issues may be important, such studies provide little information about those who do decide to vaccinate, the decision making process itself, what informs the decision, and the barriers and facilitators to obtaining information. Furthermore, although several studies have explored the use of evidence based practice by health providers and the barriers associated with this, few – if any – have focused on public health nurses. Public health nurses are key mediators and moderators of health information in the community. Their practice covers a wide range of health topics and is vital for the general public at key life stages. Our study of parents and public health nurses therefore contributes new knowledge to these areas by identifying areas of improvement in order to facilitate informed decision making and potentially identifying targets for intervention. Although health professionals are an important source of health information, independent searches for such information by health users are an important part of participation. Being updated on the latest health information is a precondition for effective participation both in 52 consultations and in public debate, as well as for the management of personal health. Other studies describing searches for health information have generally focused on personal characteristics and sources of health information. To our knowledge few, if any, studies have explored the cognitive factors that predict intention to search. By doing so, our cross-sectional study expands the body of research describing searching behaviour. Moreover, the instruments available for measuring health literacy skills are few and generally focus on reading and numeracy skills (functional literacy). The TPB questionnaire thus also provides a new way of mapping and evaluating specific domains of health literacy related to search. consultations and in public debate, as well as for the management of personal health. Discussion Other studies describing searches for health information have generally focused on personal characteristics and sources of health information. To our knowledge few, if any, studies have explored the cognitive factors that predict intention to search. By doing so, our cross-sectional study expands the body of research describing searching behaviour. Moreover, the instruments available for measuring health literacy skills are few and generally focus on reading and numeracy skills (functional literacy). The TPB questionnaire thus also provides a new way of mapping and evaluating specific domains of health literacy related to search. The development of interventions is often unsystematic and poorly reported. An important part of this project was to identify and integrate user needs and preferences with the evidence from the research literature in the development of our intervention. This was done in a systematic and transparent way within a conceptual and theoretical framework. Although online resources have been created in other international settings, no such resource is available in Norwegian. The web portal is therefore a unique resource for improving not only public access to evidence based health information, but also for providing practical tools that can be used independently by users or in consultations with health professionals. In the randomised controlled trial, we sought to evaluate the extent to which the web portal could improve the use of evidence based information and health literacy skills in a real life setting. The study also provided knowledge about the perceived relevance of the web portal by potential users. This information was valuable given that the purpose of the web portal and the concepts that it introduced would be novel to most users. The methodological considerations of these studies and our findings, and their implications, are discussed below. 53 Methodological considerations In the first phase of this project we conducted qualitative interviews with parents and public health nurses using a grounded theory design. Grounded theory has been criticised for being reductionist and even quantitative in its approach, but the method has both objectivist and constructionist roots (80). Later methodologists, such as Charmaz (80), integrated these perspectives, adopting systematic, explicit and rigorous objectivist principles while acknowledging the subjectivity involved in the data collection process and analysis (80). In grounded theory, as in much of qualitative research, the researcher is not seen as independent to the research process. Instead, a researcher’s discipline, background, and guiding empirical interest form the backdrop of the research and serve as points of departure (80). Consequently, findings (and theory developed from data) are not deterministic explanations offered to us as exact pictures of the world, but are seen as interpretative and explorative (80). In our study, the interdisciplinarity of the research team added different perspectives and viewpoints to the study about, for example, theoretical knowledge and clinical experience. The scope of the study and design of the interview guides and analysis were further influenced by the fact that the study aimed to inform the development of the TPB questionnaire and the web portal, targeting health literacy skills within the theoretical and conceptual framework of both shared decision making and evidence based practice. The data were collected through individual and focus group interviews all of which were held at maternity and child health centres. Using these sites for the interviews may have had potential benefits (such as providing a familiar environment for users), but may also have had disadvantages. It was important that is was made clear to the parents that the moderators of the 54 study were not associated with the centres in any way. We were also explicit in explaining that the public health nurses would not have access to the raw data, so that this would allow parents to speak freely. For the public health nurses, the fact that the interview group included representatives from the management may potentially have moderated their responses. Although, our specific goal was to facilitate and balance the discussion in such a way that no one would be allowed to dominate the discussion, some participants may have adopted more conservative points of view. study were not associated with the centres in any way. Methodological considerations We were also explicit in explaining that the public health nurses would not have access to the raw data, so that this would allow parents to speak freely. For the public health nurses, the fact that the interview group included study were not associated with the centres in any way. We were also explicit in explaining that the public health nurses would not have access to the raw data, so that this would allow parents to speak freely. For the public health nurses, the fact that the interview group included representatives from the management may potentially have moderated their responses. Although, our specific goal was to facilitate and balance the discussion in such a way that no one would be allowed to dominate the discussion, some participants may have adopted more conservative points of view. The theory of planned behaviour informed the development and testing of the questionnaire. (79). An issue related to the measurement properties of the questionnaire was that the TPB is social cognition model, and therefore assumes that behaviour can be explained by primarily cognitive processes (89). In other words, it is what could be termed a ‘complete’ model of behaviour – all other possible factors are theorised to influence behaviour indirectly through these TPB components (89). A criticism which has been raised towards the TPB and which may be a limitation to our study is that the model assumes that all behaviour is rational (89). Thus, it may overlook other relevant factors such as for example affective or emotional variables which in the TPB are considered only as background variables mediating the TPB components (89). This assumption may therefore be a limitation to our study (89). It should be noted that we did not compare intended behaviour (to search) with actual searching behaviour in the questionnaire study (or in the randomised controlled trial), and we therefore do not know the strength of this relationship. Although the measurement of the relationship between intention and actual participation – or the intervention effects of the web portal on potential increased searching activity – may be of interest in future studies, these was 55 not the main study objective of this project. Our main objectives and final end points were instead to facilitate skills (which may enable participation) but not necessarily to increase participation in itself. Methodological considerations This being said, evidence across a range of TPB studies shows intention to be a valid proxy measurement and the relationship to be strong (90, 91) The low response rate we experienced was a limitation of this study. Unfortunately, this is a challenge common to most questionnaire studies, and a response rate above 50% is usually not achieved (79). The extent to which this may have influenced the results of the questionnaire is unclear. Age and education are the factors most likely to influence searching behaviour (26-29), and the data set of those who chose to respond showed the distribution of age to be even across the age groups. However, people with higher education were overrepresented and this may indicate that people with lower education may have been less likely to respond. In addition, the overrepresentation of people with higher education in the sample may potentially have resulted in an overestimation of the overall score on intention (producing a higher level of intention) to search compared to a more mixed population. Unfortunately, we do not have data to enable us to explore these hypotheses further. Another potential limitation to our study was that we did not perform a test-retest procedure, which is used as a psychometric method to test for reliability (79). We predicted that recruiting and achieving an adequate response rate would be challenging and therefore decided against asking people to do a retest. However by administrating the questionnaire to two different samples we were able to compare the questionnaire’s consistency across them. 56 In the second phase of the project we developed the intervention. Complex interventions that are tailored to target important barriers to change are considered more likely to be effective (69, 70). There is, however, no ‘gold’ standard’ for how to develop effective complex interventions. A Cochrane review found that tailored interventions to improve professional practice improved processes of care and patient outcomes, but that there was insufficient evidence to determine the most effective approaches to tailoring, including how to identify barriers and how to select interventions to address identified barriers (133). Our decisions on what barriers to target and the choice of facilitators to address these barriers were based on our judgement of what we considered to be the most important barriers, if there were evidence based strategies available to address those barriers, and what was feasible given the resources available (69, 70). Methodological considerations This meant that not all the identified barriers could be addressed, and a selection of barriers was thus made. Consequently, the intervention may not have included all the important factors, but it did include changeable and relevant variables that could help to improve public access to evidence based information and health literacy skills. Despite our efforts, important barriers and facilitators may have been missed or not appropriately addressed. (69, 70). There is, however, no ‘gold’ standard’ for how to develop effective complex interventions. A Cochrane review found that tailored interventions to improve professional practice improved processes of care and patient outcomes, but that there was insufficient evidence to determine the most effective approaches to tailoring, including how to identify barriers and how to select interventions to address identified barriers (133). Our decisions on what barriers to target and the choice of facilitators to address these barriers were based on our judgement of what we considered to be the most important barriers, if there were evidence based strategies available to address those barriers, and what was feasible given the resources available (69, 70). This meant that not all the identified barriers could be addressed, and a selection of barriers was thus made. Consequently, the intervention may not have included all the important factors, but it did include changeable and relevant variables that could help to improve public access to evidence based information and health literacy skills. Despite our efforts, important barriers and facilitators may have been missed or not appropriately addressed. 57 In the final phase of the project we evaluated the web portal. Its effectiveness was evaluated in a randomised controlled trial of a group of typical users in a real life setting, in what is referred to as a pragmatic trial (134, 135). A distinction is often made between explanatory and pragmatic trials. Explanatory trials aim to determine the effects (efficacy) of a precisely-defined intervention on a specific group of people under optimal conditions (134). Although such studies may hold high internal validity and provide important knowledge, the results may have applicability issues (134). Healthcare interventions are often complex in nature and must be addressed accordingly. Pragmatic trials are recommended as a way of achieving this through their measurement of the effects of an intervention in the same setting that the intervention will be implemented and under real-life circumstances (134, 135). Methodological considerations It is difficult to speculate about the reasons for the loss to follow-up: the majority of the dropouts gave no reasons for leaving. The fact that the loss to follow-up happened between the time of the first screening questionnaire and the time of the first task may indicate two things: firstly, that the tasks were considered too extensive and, secondly, that people joined the study primarily in order to receive access to the portal. Once they had been access, they may have dropped out before the first task was given. education, which was somewhat higher in the control group. This may mean that we underestimated the improvements in the intervention group given that education has been found to be partly correlated with health literacy skills (17). It is difficult to speculate about the reasons for the loss to follow-up: the majority of the dropouts gave no reasons for leaving. The fact that the loss to follow-up happened between the time of the first screening questionnaire and the time of the first task may indicate two things: firstly, that the tasks were considered too extensive and, secondly, that people joined the study primarily in order to receive access to the portal. Once they had been access, they may have dropped out before the first task was given. Methodological considerations Other important criteria for 57 generalisability are a minimum of exclusion criteria and ensuring that a large degree of freedom is associated with the intervention (136). It should be noted that although a clear distinction is often made in the literature between the approaches used in explanatory and pragmatic trials, in practice they are not dichotomous and should be seen as part of a research continuum (134). Pragmatic trials are used to inform practice and the methodological choices that are made will reflect this (134). By using a pragmatic approach, the intention of this study was to maximise the external validity and thus to ensure generalisability. These advantages must be weighed against threats against the internal validity introduced by the real life setting (136). These criteria and considerations were embedded into the design and execution of our trial. All parents with children under the age of four years were included and participant exposure to the intervention was flexible: parents were free to respond to the tasks in their own time and at leisure within a familiar environment. The degree to which we were successful in blinding the parents to the intervention is unclear, but several measures were introduced to ascertain this making it unlikely that this influenced the outcomes to a large degree. All participants were informed that they would receive the portal but at different time points, and standardised and automated communication and data collection methods were also important measures used. Such systems also reduced the risk of potential errors as data could be exported directly into Excel and SPSS for analysis after the trial. This was also logistically beneficial as it provided us with a time-saving and cost-efficient strategy for managing the study and made it easier to keep track of response rates and sending out tasks and reminders. 58 The trial was also limited by the fact that we were not able to recruit as many participants as intended and because the study also suffered some loss to follow-up. The background characteristics between participant groups were very similar, except for the level of participant 58 education, which was somewhat higher in the control group. This may mean that we underestimated the improvements in the intervention group given that education has been found to be partly correlated with health literacy skills (17). Capacity, abilities, and the perceived need as barriers 59 The interviews with parents and public health nurses provided complex data. When we explored the decision making it was evident that the participation of parents as well as public health nurses took the form typical of a traditional provider-led decision making model, but that there was also confusion about roles and expectations. While parents expressed a desire to be involved they were passive at the same time and demanded much of the public health nurses. The public health nurses stressed the right of parents to make a free choice, but were also reported to have given parents the information in a biased way that facilitated vaccination. In other words, their method of informing was characterised by persuasive information delivery in which the main goal was to achieve adherence. In Norway, this form of health communication can be seen in the discourse of public health. For example, an editorial about vaccination against swine flu in the Journal of the Norwegian Medical Association (Tidsskrift for den Norske Legeforening), stated that physicians should not inform patients about ‘everything’ (137) – in other words, that 59 59 they should not make people aware of all the possible risks. A further example is found in the Childhood Vaccination Handbook for Health Professionals issued by the Norwegian Public Health Institute, which states that ‘it must be assumed that [parents] would like vaccination’ (138). A major weakness of such an authoritative and prescriptive decision making model is that it assumes that it correctly knows users preferences and abilities, and also assumes that professionals are competent enough to decide what is appropriate on behalf of users (5, 10, 11). Furthermore, it does not facilitate informed decision making given that information is withheld or tailored to facilitate a specific choice. A systematic review of barriers to shared decision making concluded that one of the main reasons why people are sub-optimally involved in health decisions is that health professionals assume that people are either not willing to participate or not capable of doing so (11). However, the willingness or capacity of people should not be underestimated (5). A large cross-sectional study of peoples perceptions on the responsiveness of health systems and own participation conducted in Germany, Italy, Poland, Slovenia, Spain, Sweden, Switzerland and the United Kingdom (UK), with 8,119 people aged 16-years or over, concluded that patients wished to have a more autonomous role in healthcare (9). Capacity, abilities, and the perceived need as barriers When asked which decision making model they preferred, the shared decision making model was the most popular by far, with 51% of the total sample opting for it (9). 60 The barrier categories we identified to the ability of parents to participate and obtain information were: own capacity, abilities, and being positive towards vaccination and having already decided. The first two categories relate to parents having the opportunity and necessary skills, while the latter category relates to the perceived need of the parents for information associated with the specific health decision. Own capacity such as being overwhelmed and being able to find and assess reliable health information have been identified as common barriers to 60 searching for health information in general (27, 29). The health decision in itself, according to other studies, may also influence the degree to which people want to be involved (139, 140). We know that many people decide to vaccinate because this is a well-known and accepted treatment in our society, recommended by governmental bodies, and seen as ‘the normal thing to do’ (also known as the ‘bandwagon’ effect) (140). Thus, those people who are positive in attitude towards vaccination may perceive less of a need for information than those who are going against ‘the establishment’; leaving adherers less informed. Consequently, the fact that people adhere to guidelines does not necessarily imply that they are well-informed. These identified barriers are not unrelated. Having the necessary abilities may save people time and improve capacity: parents, for example, could more easily navigate the information flow of society and also know what to ask public health nurses. A perceived need for information may also be dependent on the ability to evaluate the information available and consequently on making a judgement about whether more information is still needed to make an informed decision. For example, a person with high science literacy may not be satisfied with health advice that is based on personal anecdotes. Furthermore, as was found in this study, knowledge about people’s own roles and responsibilities in decision making may also influence their perceived need. It can thus be argued that these barriers are interrelated and that an improvement in one area may lead to improvements in the other domains. The parents in our study reported that they had little knowledge about vaccination. Capacity, abilities, and the perceived need as barriers This lack of knowledge about different vaccines, diseases, how immunisation works, or uncertainty about the effects of vaccination, is evident in other developed countries with high vaccination rates (58, 141, 142). Furthermore, our study showed that decision making is often based on 61 common sense, trust and personal experiences (140, 141). Although trust and personal experience are important in healthcare, they are an incomplete foundation for informed decision making about treatments. When trust forms the basis of decisions rather than knowledge, people are encouraged to rely solely on health professionals instead of being empowered to think for themselves (5). Decisions based on trust are also easily weakened and may not apply to new treatments (108). Recent examples of controversies related to vaccination, for example, include public debates about vaccination against the human papilloma virus (HPV) and swine flu (H1N1 influenza virus) (137, 143). A study of responses to governmental recommendations related to swine flu in the UK found a general scepticism toward public messages and that people questioned the credibility of such information (144). To a large degree, decision making was found to be informed by people’s own beliefs about the disease and how to prevent it and by personal anecdotes from people they knew (144). Without adequate knowledge and (health literacy) skills, it is nearly impossible for a layperson to critically appraise the information exchanged in such discourses. Instead, the choice about whether to vaccinate or not may be determined by deciding who to trust instead of weighing the benefits and harms based on the best available evidence. Interestingly, in vaccination behavioural research, who people trust has been found to be one of the characteristics that differentiates vaccinators from non-vaccinators (139, 141). Parents who decides to vaccinate are more likely to trust the government than those who decide not to vaccinate, and who are instead more likely to use alternative providers of health care such as homeopaths (139, 141). These examples and other research suggest that there are important areas for improvements related to health literacy. Firstly, while evidence based practice and user 62 involvement may be stressed in policy documents and by researchers, it may still remain unknown to both users and providers of healthcare, and may not be an integrated part of all practice (11, 33-35). Capacity, abilities, and the perceived need as barriers There is also a need for people to be provided with insight into the elements of clinical decision making to enable them to find and evaluate health information. Parent and public health nurse barriers may be similar 63 Our study identified potential areas for improvement relevant to the practice of public health nurses. Public health nurses were seen by the parents as their most important source of health information and parents expressed great confidence in them, emphasising their roles as counsellors and moderators of health information. Health professionals have a crucial role in health education and the dissemination of health information, a role that has become perhaps even more important and challenging in recent years due to the rapid increase of health information available to the public (7). To keep pace with both the latest developments, and with knowledge which is important for patient outcomes and good quality healthcare, public health nurses must be able to integrate the best available research into practice (1, 31, 145). Moreover, insight into how knowledge is produced and how to evaluate the reliability and applicability of such information is important to address the information needs of parents. Enabling users to search for reliable information and being able to discuss such information is an important activity for health professionals and has been emphasised in studies exploring user information needs (146, 147). Despite acknowledging the importance of using research, the public health nurses that participated in our study were reluctant to search independently for such information. The main reasons for this, in relation to vaccination, were that they did not consider doing so to be part of Our study identified potential areas for improvement relevant to the practice of public health nurses. Public health nurses were seen by the parents as their most important source of health information and parents expressed great confidence in them, emphasising their roles as counsellors and moderators of health information. Health professionals have a crucial role in health education and the dissemination of health information, a role that has become perhaps even more important and challenging in recent years due to the rapid increase of health 63 information available to the public (7). To keep pace with both the latest developments, and with knowledge which is important for patient outcomes and good quality healthcare, public health nurses must be able to integrate the best available research into practice (1, 31, 145). Moreover, insight into how knowledge is produced and how to evaluate the reliability and applicability of such information is important to address the information needs of parents. Parent and public health nurse barriers may be similar Enabling users to search for reliable information and being able to discuss such information is an important activity for health professionals and has been emphasised in studies exploring user information needs (146, 147). Despite acknowledging the importance of using research, the public health nurses that participated in our study were reluctant to search independently for such information. The main reasons for this, in relation to vaccination, were that they did not consider doing so to be part of 63 their role and thought that such searches could potentially conflict with their mandate. Other reasons were critical appraisal skills and capacity. This indicates that the barriers related to obtaining health information (and the health literacy skills required) may not be so different from the skill and capacity issues of parents which were common barriers to obtaining research. Furthermore, this finding underlines the importance of quality-improvement initiatives targeting public health nurses who should be empowered to use and integrate research in practice, and the need for such initiatives to target the barriers and facilitators to change identified in our study. their role and thought that such searches could potentially conflict with their mandate. Other reasons were critical appraisal skills and capacity. This indicates that the barriers related to obtaining health information (and the health literacy skills required) may not be so different from the skill and capacity issues of parents which were common barriers to obtaining research. their role and thought that such searches could potentially conflict with their mandate. Other reasons were critical appraisal skills and capacity. This indicates that the barriers related to obtaining health information (and the health literacy skills required) may not be so different from the skill and capacity issues of parents which were common barriers to obtaining research. Furthermore, this finding underlines the importance of quality-improvement initiatives targeting public health nurses who should be empowered to use and integrate research in practice, and the need for such initiatives to target the barriers and facilitators to change identified in our study. Attitude and perceived behavioural control are important predictors of intention to search for health information However, barriers to searching for health information, as the data from our questionnaire revealed, include people being unsure if the information is understood properly or feeling overwhelmed, frustrated and confused by the information they find (27, 149). In contrast, the finding from our study in addition to what we know from other studies, social pressure or expectations to search may not be a variable positively predicting search (9, 11, 27, 149-152). The operationalisation of TPB and the theory itself have received much attention and literature. The findings of the questionnaire supported the relevance of these identified issues. According to the TPB, by changing the most important predictor(s) ‘we can increase the chance that the person will intend to do a desired action and thus increase the chance of the person actually doing it’ p7,(79) . Thus, when developing an intervention, these predictors should be targeted (79). The participants in both samples had high assessments of their direct attitude towards searching, including the indirect measures. Their assessments of direct perceived behavioural control were moderate to high, but the indirect measures they reported showed that certain areas needed improvement. The score for the direct subjective norm for both samples was moderate, and the assessment of indirect measures associated with social pressure was modest. Attitudes and perceived behavioural control were identified as important positive predictors of intention to search. These results are supported by findings from other descriptive cross-sectional studies which describe searches for information by patients and healthy citizens. In these studies, people were reported to have had positive attitudes associated with searching because it provided them with support with their decision making and in consultations with health professionals. It also provided them with comfort and additional knowledge (27, 148, 149). However, barriers to searching for health information, as the data from our questionnaire revealed, include people being unsure if the information is understood properly or feeling overwhelmed, frustrated and confused by the information they find (27, 149). In contrast, the finding from our study in addition to what we know from other studies, social pressure or expectations to search may not be a variable positively predicting search (9, 11, 27, 149-152). Attitude and perceived behavioural control are important predictors of intention to search for health information To our knowledge this is the first questionnaire to explore the cognitive factors that influence behavioural intentions to search. Based on our analysis, we concluded that the internal structure (reliability) of items measuring intention to search and direct measures were satisfactory, and that the relationships between these variables generally supported the theoretical predispositions of the TPB and previous evidence (79). We also found that the questionnaire’s overall predictive strength (of intention to search) was very good and consistent with what has been found in other studies using the TPB (90). Comparing these values with the same reliability and validity tests performed in the RCT study, the utility of the questionnaire was confirmed by the consistency of findings. This indicates that this questionnaire is a reliable tool for mapping and evaluating intention to search and underlying variables. 64 This questionnaire enabled us to explore further and quantify the findings of the qualitative study of parents and public health nurses, supplemented by findings in the published 64 literature. The findings of the questionnaire supported the relevance of these identified issues. According to the TPB, by changing the most important predictor(s) ‘we can increase the chance that the person will intend to do a desired action and thus increase the chance of the person actually doing it’ p7,(79) . Thus, when developing an intervention, these predictors should be targeted (79). The participants in both samples had high assessments of their direct attitude towards searching, including the indirect measures. Their assessments of direct perceived behavioural control were moderate to high, but the indirect measures they reported showed that certain areas needed improvement. The score for the direct subjective norm for both samples was moderate, and the assessment of indirect measures associated with social pressure was modest. Attitudes and perceived behavioural control were identified as important positive predictors of intention to search. These results are supported by findings from other descriptive cross-sectional studies which describe searches for information by patients and healthy citizens. In these studies, people were reported to have had positive attitudes associated with searching because it provided them with support with their decision making and in consultations with health professionals. It also provided them with comfort and additional knowledge (27, 148, 149). Attitude and perceived behavioural control are important predictors of intention to search for health information However, when evaluating the effects of the intervention – and acknowledging subjective norm as a significant variable found to influence behaviour – the construct provides valuable knowledge about the degree to which the participants perceived the intervention to increase social pressure to search. Attitude and perceived behavioural control are important predictors of intention to search for health information The operationalisation of TPB and the theory itself have received much attention and 65 The operationalisation of TPB and the theory itself have received much attention and scientific interest, and are subject to extensive methodological research and development (89, 90, 153). This research is predominantly in two areas: an exploration of additional predictors to 65 justifications are made (89). Research on the conceptualisations of the existing TPB components is, to a large extent, focused on the component of the subjective norm. The main reason for this is that it has been found to be a weaker predictor of certain behaviours (90). In part, this has been attributed to measurement issues in previous studies (in which only one item has been used), but also because the construct may not adequately capture the complexity of social influence (90). Our study contributes to this discussion and challenges the TPB model’s classical assumption that increased social pressure creates greater behavioural intentions. In our study the subjective norm was found to be poorly correlated with intentions and a non-statistically significant or even negative predictor of intentions. This finding was also reproduced in the randomised controlled trial. 66 The TPB questionnaire in our study was also developed in order to be used to evaluate the effects of the web portal in the randomised controlled trial. The specific beliefs that were included in the questionnaire were all chosen based on the fact that they were (changeable) factors related to specific health literacy skills (with the exception of subjective norm). Attitudes and perceived behavioural control towards search for health information are constructs that can be directly related to individual, personal health literacy skills. Whereas the perceived behavioural control items (direct and indirect) describes actual perceived skills and feelings of mastering the exercise of searching, the items describing a person’s attitude towards searching relate to a person’s belief in their own participation and their knowledge about the benefits of doing so. In other words, more health literate people would hypothetically be more likely to have 66 positive attitudes towards searching, have a higher perceived behavioural control, and thus be more likely to have higher intention to search. In contrast, the subjective norm variable does not relate to an individual’s own health literacy – instead, it relates to social pressure from others. Providing an easy access point to evidence and tools 67 As mentioned earlier in the introduction to this research, many areas for improvement related to health literacy skills have been identified. Our findings and other research have led us to conclude that certain health literacy skills should be targeted specifically in an intervention. In this study, these skills were grouped and organised into three main barrier categories: the inability to understand and critically appraise health information, the inability to exchange information in consultations, and not knowing where to find reliable and relevant information. To address the identified barriers, the content of the intervention was developed within the conceptual frameworks of shared decision making and evidence-based practice (1, 5, 113). To achieve this, three strategies were chosen: providing an introduction to critical appraisal (and an associated checklist); providing information about what decision making about treatment or screening entails (and a decision aid for consultations); and improving access to reliable research-based sources of health information through the provision of an introduction to research methods and providing help with finding evidence-based health information efficiently, based on the principles 67 of evidence-based practice. These strategies touch upon specific domains of health literacy (8) (see Figure 1 and Table 1), including: the understanding of risk (functional literacy), knowledge of fundamental scientific concepts and processes (science literacy), how and where information is presented (civic literacy), roles and rights (civic literacy), and the understanding of concepts used in decision making about healthcare (cultural literacy). Although other efforts have been made to facilitate Norwegian public access to evidence based information, we do not know of any existing online tool that provides insight into these domains of health literacy. Internationally, there is a longer tradition of research on how to improve health literacy and several tools can be found that address related elements (see www.sunnskepsis.no . Other examples of web based resources include: ‘Testing treatments’ (www.testingtreatments.org), which focuses on science literacy and the evaluation of the effects of treatments, and ‘What are your chances’ (www.whatareyourchances.com), which focuses on numeracy and how to understand risk. www.whatareyourchances.com), which focuses on numeracy and how to understand risk. Our web portal was developed to target the public in general and to be used independently or in consultations with health professionals. Providing an easy access point to evidence and tools An important question raised in the project group during this process was: ‘What separates an intervention targeting the health literacy skills of lay people from an intervention targeting those of health professionals?’ As previously discussed, both health professionals and users may be unaccustomed to the principles of evidence based practice and may share the same barriers to obtaining reliable health information. Much of the content included in the web portal may therefore be of equal relevance to users and health professionals alike. However, two main differences were taken into account when developing the web portal. Firstly, not all users may be familiar with medical terminology (and knowledge about medicine). Secondly, users – in contrast to health professionals – are not obliged to adopt 68 evidence based practice in their lives and have the freedom to choose not to be involved in decision making. evidence based practice in their lives and have the freedom to choose not to be involved in decision making. Consequently, we made an effort to structure the web portal in a way that would provide easy access to the included resources, but we aimed as well to develop the content in a way that would be relevant to the contexts of users. We achieved this by using as much plain language as possible, except when we used medical- or research methodology-related jargon deliberately with an educational purpose in mind. Real-life examples, such as scenarios or news stories, were used when presenting text, and we only included tools that had been designed with a user-perspective in mind. Issues of access and the targeting of a lay public were also taken into account when developing the section in the web portal that facilitates direct access to research-based information. This section built on the steps included within circle model of evidence based practice, taking the problem formulation phase as our point of departure (113). In this model, problems are defined into six main categories: prevalence, aetiology, diagnostics, effectiveness of treatments, prognosis, and patient or provider attitudes or experiences (113). However, when creating the problem formulation section, we decided to omit the diagnostics category. Although this may be debated, the field of diagnostics is complicated and we decided that this should be done in consultations with a health professional. Providing an easy access point to evidence and tools A more accessible checklist which could be used to appraise the relevance of research was included instead of the critical appraisal checklists typically used by researchers and professionals. However, to accommodate (and encourage) users who might want to learn more, we added a link to these checklists as well. typically used by researchers and professionals. However, to accommodate (and encourage) users who might want to learn more, we added a link to these checklists as well. 69 The web portal was judged as a relevant tool and may improve attitudes in a pragmatic setting The effectiveness of the web portal was evaluated in a pragmatic trial. When this study was conducted, there was no single available instrument suitable to capture the complete model of health literacy (52) which included not only functional literacy but also science literacy, civic literacy and cultural literacy (8). The outcomes of our study were measured instead using a selection of instruments specifically aimed to evaluate domains of health literacy that our intervention targeted. Only minor and non-statistically significant differences were found in the searching task and the critical appraisal task. The participants in both groups adopted a sceptical point of view when evaluating the information about swine flu. However, when searching for health Only minor and non-statistically significant differences were found in the searching task and the critical appraisal task. The participants in both groups adopted a sceptical point of view when evaluating the information about swine flu. However, when searching for health information, nearly all parents chose to use information that was not evidence based and without any explicit statement on what the information had been based on. This suggests that the parents may not have had enough time to explore the web portal, and may have preferred to rely on sources with which they were familiar. It also signifies that explicit statements about what health information is based on are not important criteria for parents when validating the sources of information they find. Studies exploring which criteria people base their evaluations on, conclude that people largely assess online information on other criteria such as: the source (the publisher), URL (whether it is commercial or not), funding, the date, and how it is presented (29, 55, 105). The high respondent score on beliefs related to both search and overall participation illustrates that the parents were generally positive towards participation. This finding is congruent with other research in Europe which has mapped people’s willingness to participate (9). We 70 found an improvement in favour of the intervention group for overall intention and overall activation. Improvements were also found in favour of the intervention group for the variables predicting intention-perceived behavioural control and attitude. However, only the difference in attitude was significant between the two groups, a variable which was also identified as the most important predictor of intention to search in both samples. The web portal was judged as a relevant tool and may improve attitudes in a pragmatic setting Overall, the course was regarded by the user representatives as very useful and relevant to their role, by giving them attitudes and skills that were important to their work in organisation and representative participation (154). This was specifically because they felt that they were being equipped with a tool – a tool which they saw as already being available to health professionals and official authorities (154). By gaining access to such a resource, the participants felt this would enable them to gain influence, justification and trustworthiness for their views (154). The acknowledgement of their key role and responsibility as representatives for knowledge translation, the provision of access to research, and the ability to evaluate such information were considered by them to be essential as means of quality assurance in meetings they had with members of their organisation (154). The web portal was judged as a relevant tool and may improve attitudes in a pragmatic setting This may indicate that the web portal may have played a role in raising parent awareness about taking on an active role in health decisions. Subjective norm was rated as moderate and had a weak relationship with intention in both groups. This finding adds weight to our previous research showing that social expectations or pressure did not appear to be an important factor associated with people’s intention to search for health information. found an improvement in favour of the intervention group for overall intention and overall activation. Improvements were also found in favour of the intervention group for the variables predicting intention-perceived behavioural control and attitude. However, only the difference in attitude was significant between the two groups, a variable which was also identified as the most important predictor of intention to search in both samples. This may indicate that the web portal may have played a role in raising parent awareness about taking on an active role in health decisions. Subjective norm was rated as moderate and had a weak relationship with intention in both groups. This finding adds weight to our previous research showing that social expectations or pressure did not appear to be an important factor associated with people’s intention to search for health information. The web portal was considered by the participants to have good usability, usefulness and credibility, and this supported the findings from the usability test conducted as part of our research development process (78). Considering that the purpose of the web portal and the concepts introduced were probably novel to most of the parents, this finding is encouraging and we hope will inspire further studies in this area. Parallel to the randomised controlled trial, the web portal was also tested in an independent study to explore how user representatives evaluated the usefulness of a course in evidence-based practice and critical appraisal (154). I was not personally involved in this study conducted by Elin Opheim, a student from the Masters Degree Programme in Evidence Based Practice at Bergen University College, Norway. At the time the research was undertaken, it was the first study of its kind conducted in Norway (154). The course had three steps: an introduction to evidence based practice, searching for health information (based on the web portal), and 71 critical appraisal (also based on the web portal) (154). People may not trust in science, and rational decision making may not result in the ‘recommended’ choice The conceptual and theoretical perspectives adopted in this project are typical of epidemiology and health services research (86) in that they primarily address cognitive factors such as attitudes, knowledge and skills. The underlying assumption of this study is that people balance advantages and disadvantages in a rational decision making process. From this viewpoint, access to adequate information (and the ability to understand and act upon this information) is seen as crucial to change (86). However, the paradox facing all initiatives built on this assumption is that people may reject science in its entirety and choose to base their decisions on other belief systems about health. This has been found to be the case with regard to the topic of vaccination. Some anti-vaccination groups rely on different bases of knowledge rooted in, for 72 72 example, personal anecdotes, philosophy or religion, and these sources of knowledge are often linked to social networks or communities (15, 155, 156). Such reliance on other knowledge paradigms has also been associated with distrust in the healthcare system in general (15, 139). example, personal anecdotes, philosophy or religion, and these sources of knowledge are often linked to social networks or communities (15, 155, 156). Such reliance on other knowledge paradigms has also been associated with distrust in the healthcare system in general (15, 139). The role of new information technologies may add strength to these kinds of networks, demonstrating how knowledge is negotiated and shared not only between professionals and users, but also between users (155). Improving the access people have to evidence based information and educational initiatives therefore may not be adequate as such information may not be accepted by its receiver. demonstrating how knowledge is negotiated and shared not only between professionals and users, but also between users (155). Improving the access people have to evidence based information and educational initiatives therefore may not be adequate as such information may not be accepted by its receiver. Another paradox from a rational decision making perspective is that despite the fact that people are appropriately informed and health literate, their final decision may not be what is considered to be the’ best option’ by health professionals. This was demonstrated in a systematic review of decision aids (127) which showed that although decision aids increased people’s involvement and improved their knowledge and realistic perception of outcomes, the effects on adherence were inconclusive (127). People may not trust in science, and rational decision making may not result in the ‘recommended’ choice Despite this, the improved exchange of health information and clarification of values may make health behaviour and cognitive processes more explicit and be a better starting point for the process of care (7). Conclusions In this thesis we examined a tailored intervention informed by qualitative and quantitative pre-studies as well as literature searches. The studies provided us with rich data that enabled us to identify and target barriers and facilitators that were considered important by members of our key audiences. Through a consideration of these barriers, we developed a web portal designed using the conceptual frameworks of shared decision making and evidence based practice. The purpose of doing this was to improve specific domains of health literacy related to obtaining health information. The effects of the web portal were evaluated in a pragmatic setting on a group of parents. The findings of this study suggest that the web portal may improve positive attitudes towards searching for health information, a variable identified as the most important predictor of intention to search. Furthermore, the relevance of the web portal to users was confirmed. Some have expressed concerns about interventions targeting user participation and health literacy 73 Criticism has been raised about interventions that aim to improve participation and health literacy skills. Firstly, a concern may be that such interventions can create expectations that cannot be met, considering that many health professionals may also be unfamiliar with finding, 73 evaluating and applying research (33-35). Although these concerns may be valid, they are not valid enough reasons to leave users uninformed. Aside from the ethical and legal dimensions of user involvement and education, more-informed patients can also actually play a role in promoting evidence based practice by professionals (157, 158). A second major concern is that better-informed users may increase the demand for healthcare by, for example, through increased consultation length (20, 127, 159). However, evidence regarding such effects is inconclusive (20, 127, 159). Furthermore, the potential costs associated with factors such as potential increases in consultation length may be outweighed by improved user outcomes and more appropriate treatments (20, 159). Findings indicate, for example, that informed users are more likely to choose less risky treatments and adopt a wait-and-see attitude (127). In contrast, misconceptions about generic substitution, the underuse of generic medications, and the increased use of healthcare services have been associated with low literacy levels (17, 62). Thus it can be argued that educating and involving users may in fact result in resources being spent more wisely. The two concerns noted above, it can therefore be argued, are built upon a traditional and outdated view of how information about medical and health related information is transferred, namely that it moves between the professional and user. Today, health information is widely distributed through various communication channels, including between users themselves (25). Acknowledging this, the public must be recognised as key actors in knowledge transfer. Acknowledging this, the public must be recognised as key actors in knowledge transfer. Improving health literacy skills is not only important to ensure informed decision making for the individual, but is also vital for knowledge transfer in society as a whole. Improving health literacy skills is not only important to ensure informed decision making for the individual, but is also vital for knowledge transfer in society as a whole. 74 Implications for practice and further research Research describing decision making and health behaviour related to vaccination has generally been concerned with reasons for non-adherence. This has largely focused on the attitudes and beliefs associated with vaccination and on satisfaction with the associated services. Recognising health literacy as a public health goal, future studies should instead complement our study by exploring further what informs decision making related to vaccination. Furthermore, 75 there is very little evidence available which map the knowledge of parents about immunisation in Norway and about health literacy skills. Future research efforts should include large cross-sectional studies to explore this and inform interventions targeting specific information needs. In this project, emphasis was placed on the identification of the barriers and facilitators associated with obtaining health information. Providing insight into the process, components and intended intervention mechanisms is important for reproducibility and as a way to inform the efforts of other researchers and methodologists within this field (69, 133). Although tailored interventions have been found to be effective, there is no clear evidence on which particular method of tailoring is most appropriate (133). Adopting an explicit conceptual and methodological approach is of particular importance when developing interventions (68, 69), but more evidence is needed on the development of interventions and the potential benefits of tailoring (133). Overall, the findings in the pragmatic trial were modest and indicated that online resources alone – such as the web portal – may be insufficient to effectively improve health literacy skills. As an element of acknowledging the rights and central role of users in evidence based practice, it is important that resources such as the web portal are made available to enable access to evidence based information and to facilitate health literacy skills through practical tools. In order to ensure a more intensive form of intervention, future research efforts should consider interventions in which the web portal is included in consultations with health providers, as part of evidence-based practice as an integrated part of patient education or, for example, in educational programmes in schools. 76 76 References 1. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, Hopayian K, et al. Sicily statement on evidence-based practice. BMC medical education [Internet]. 2005 544887]; 5(1):[1 p.]. Available from: www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=1563435 9. 2. Nutbeam D. Health promotion glossary1998:[36 p.]. Available from: www.who.int/hpr/NPH/docs/hp_glossary_en.pdf. 3. Ministry of Health and Care Services. The Norwegian Bill of Patients’ Rights. LOV-1999-07-02-63 [Internet]. 1999. Available from: www.lovdata.no/cgi-wift/ldles?doc=/all/nl-20060630-045.html. 4. Ministry of Health and Care Services. St.meld. nr. 47 (2008-2009).[The Interaction Reform: The right treatment - at the right place - at the right time]2009:[150 p.]. Available from: www.regjeringen.no/nb/dep/hod/dok/regpubl/stmeld/2008-2009/stmeld-nr-47-2008-2009- .html?id=567201. 5. Coulter A. The Autonomous patient: ending paternalism in medical care. London: TSO; 2002. 121 p. 6. Coulter A, Ellins J. Patient-focused interventions: a review of the evidence2006:[277 p.]. Available from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. 7. Taylor K. Paternalism, participation and partnership - the evolution of patient centeredness in the consultation. Patient education and counseling. 2009;74(2):150-5. 8. Zarcadoolas C, Pleasant A, Greer D. Advancing health literacy: a framework for understanding and action. 1st ed. San Francisco, CA: Jossey-Bass; 2006. 368 p. 9. Coulter A, Jenkinson C. European patients' views on the responsiveness of health systems and healthcare providers. European Journal of Public Health. 2005;15(4):355-60. 10. Bugge C, Entwistle V, Watt I. The significance for decision-making of information that is not exchanged by patients and health professionals during consultations. Social science & medicine. 2006;63(8):2065-78. 11. Legare F, Ratte S, Gravel K, Graham I. Barriers and facilitators to implementing shared decision- making in clinical practice: update of a systematic review of health professionals' perceptions. Patient education and counseling. 2008;73(3):526-35. 12. Glenton C, Paulsen E, Oxman A. Portals to Wonderland? Health portals lead confusing information about the effects of health care. BMC medical informatics and decision making [Internet]. 2005; 5:7:[8 p.]. Available from: www.biomedcentral.com/content/pdf/1472-6947-5-7.pdf. 13. Glenton C, Oxman A. The use of evidence by health care user organizations. Health Expectations. 1998;1(1):14-22. 14. Johansen L, Bjørndal A, Flottorp S, Grøtting T, Oxman A. [Evaluation of health information in newspapers and brochures. What can we believe?]. Tidsskrift for Den norske Legeforening. 1996;116(2):260-4. 15. Wolfe R, Sharp L, Lipsky M. Content and design attributes of antivaccination web sites. Journal of American Medical Association. 2002;287(24):3245-8. 16. Hoye S, Hjortdahl P. ["New wonder pill!"--what do Norwegian newspapers write]. Tidsskrift for Den norske Legeforening. 2002;Jun 30;122(17):1671-6. 17. Berkman N, Sheridan S, Donahue K, Halpern D, Crotty K. References Low Health Literacy and Health Outcomes: An Updated Systematic Review Annals of Internal Medicine 2011;155(2):97-101 1. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, Hopayian K, et al. Sicily statement on evidence-based practice. BMC medical education [Internet]. 2005 544887]; 5(1):[1 p.]. Available from: 1. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, Hopayian K, et al. Sicily statement on evidence-based practice. BMC medical education [Internet]. 2005 544887]; 5(1):[1 p.]. Available from: 1. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, Hopayian K, et al. Sicily statement on evidence-based practice. BMC medical education [Internet]. 2005 544887]; 5(1):[1 p.]. Available from: www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=1563435 9. 2. Nutbeam D. Health promotion glossary1998:[36 p.]. Available from: www.who.int/hpr/NPH/docs/hp_glossary_en.pdf. 3. Ministry of Health and Care Services. The Norwegian Bill of Patients’ Rights. LOV-1999-07-02-63 [Internet]. 1999. Available from: www.lovdata.no/cgi-wift/ldles?doc=/all/nl-20060630-045.html. 4. Ministry of Health and Care Services. St.meld. nr. 47 (2008-2009).[The Interaction Reform: The right treatment - at the right place - at the right time]2009:[150 p.]. Available from: www.regjeringen.no/nb/dep/hod/dok/regpubl/stmeld/2008-2009/stmeld-nr-47-2008-2009- html?id=567201 4. Ministry of Health and Care Services. St.meld. nr. 47 (2008-2009).[The Interaction Reform: The right treatment - at the right place - at the right time]2009:[150 p.]. Available from: www.regjeringen.no/nb/dep/hod/dok/regpubl/stmeld/2008-2009/stmeld-nr-47-2008-2009- .html?id=567201. 5. Coulter A. The Autonomous patient: ending paternalism in medical care. London: TSO; 2 p. Coulter A, Ellins J. Patient-focused interventions: a review of the evidence2006:[277 p.]. 6. Coulter A, Ellins J. Patient-focused interventions: a review of the evidence2006:[277 p.]. Available from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. 6. Coulter A, Ellins J. Patient-focused interventions: a review of the evidence2006:[277 p.]. Available from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. 7. Taylor K. Paternalism, participation and partnership - the evolution of patient centeredness in the consultation. Patient education and counseling. 2009;74(2):150-5. 7. Taylor K. Paternalism, participation and partnership - the evolution of patient centeredness in the consultation. Patient education and counseling. 2009;74(2):150-5. 8. Zarcadoolas C, Pleasant A, Greer D. Advancing health literacy: a framework for understanding and action. 1st ed. San Francisco, CA: Jossey-Bass; 2006. 368 p. 8. Zarcadoolas C, Pleasant A, Greer D. Advancing health literacy: a framework for understanding and action. 1st ed. San Francisco, CA: Jossey-Bass; 2006. 368 p. 9. Coulter A, Jenkinson C. European patients' views on the responsiveness of health systems and healthcare providers. European Journal of Public Health. 2005;15(4):355-60. 10. Bugge C, Entwistle V, Watt I. The significance for decision-making of information that is not exchanged by patients and health professionals during consultations. Social science & medicine. 2006;63(8):2065-78. 11. Legare F, Ratte S, Gravel K, Graham I. Barriers and facilitators to implementing shared decision- making in clinical practice: update of a systematic review of health professionals' perceptions. Patient education and counseling. 2008;73(3):526-35. 12. Glenton C, Paulsen E, Oxman A. Portals to Wonderland? Health portals lead confusing information about the effects of health care. BMC medical informatics and decision making [Internet]. 2005; 5:7:[8 p.]. Available from: www.biomedcentral.com/content/pdf/1472-6947-5-7.pdf. 13. Glenton C, Oxman A. The use of evidence by health care user organizations. Health Expectations. 1998;1(1):14-22. 14. Johansen L, Bjørndal A, Flottorp S, Grøtting T, Oxman A. [Evaluation of health information in newspapers and brochures. What can we believe?]. Tidsskrift for Den norske Legeforening. 1996;116(2):260-4. 15. Wolfe R, Sharp L, Lipsky M. Content and design attributes of antivaccination web sites. Journal of American Medical Association. 2002;287(24):3245-8. 16 Hoye S Hjortdahl P ["New wonder pill!" what do Norwegian newspapers write] Tidsskrift for 15. Wolfe R, Sharp L, Lipsky M. Content and design attributes of antivaccination web sites. Journal of American Medical Association. 2002;287(24):3245-8. 16. Hoye S, Hjortdahl P. ["New wonder pill!"--what do Norwegian newspapers write]. Tidsskrift for Den norske Legeforening. 2002;Jun 30;122(17):1671-6. 17. Berkman N, Sheridan S, Donahue K, Halpern D, Crotty K. Low Health Literacy and Health Outcomes: An Updated Systematic Review. Annals of Internal Medicine. 2011;155(2):97-101. 77 18. Sanders L, Federico S, Klass P, Abrams M, Dreyer B. Literacy and child health: a systematic review. Archives of Pediatric & Adolescent Medicine. 2009;163(2):131-40. 19. Nutbeam D. The evolving concept of health literacy. Social Science and Medicine. 2008;67(12):2072-8. 36. Glenton C, Santesso N, Rosenbaum S, Nilsen E, Rader T, Ciapponi A, et al. Presenting the results of Cochrane Systematic Reviews to a consumer audience: a qualitative study. Medical Decision Making. 2010;30(5):566-77. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. . Detmer D, Singleton P, MacLeod S, Wait S, Taylor M, Ridgwell J. The informed patient: Study t. Cambridge: University of Cambridge; 2003 March 2003. 53 p. Report. Cambridge: University of Cambridge; 2003 March 2003. 53 p. 21. European Patients Forum. European Patients Forum Annual Report 2010. 2010:[44 p.]. A from: www.eu-patient.eu/Documents/Publications/AnnualReports/epf-annual-report-2010.pdf. 22. Moumjid N, Gafni A, Bremond A, Carrere MO. Shared decision making in the medical encounter: Are we all talking about the same thing? Medical Decision Making. 2007;27(5):539-46. 23. Coulter A. Paternalism or partnership? Patients have grown up-and there's no going back. The British Medical Journal. 1999;319(7212):719-20. 24. Manganello J. Health literacy and adolescents: a framework and agenda for future research. Health Education Research. 2008;23(5):840-7. 25. Fox S, Jones S. The Social Life of Health Information. . PEW Internet and American Life Project [Internet]. 2009:[88 p.]. Available from: www.pewinternet.org/Reports/2009/8-The-Social-Life-of- l h f i 25. Fox S, Jones S. The Social Life of Health Information. . PEW Internet and American Life Project [Internet] 2009:[88 p ] Available from: www pewinternet org/Reports/2009/8 The Social Life of 25. Fox S, Jones S. The Social Life of Health Information. . PEW Internet and American Life Project [Internet]. 2009:[88 p.]. Available from: www.pewinternet.org/Reports/2009/8-The-Social-Life-of- Health-Information.aspx. 26. Spadaro R. European union citizens and sources of information about health: Eurobarometer 58.0. 2003:[16 p.]. Available from: www.europa.eu.int/comm/health/ph_information/documents/eb_58_en.pdf. 27. Fox S. Online health search. PEW Internet and American Life Project [Internet]. 2006:[22 p.]. Available from: www.pewinternet.org/PPF/r/190/report_display.asp. 27. Fox S. Online health search. PEW Internet and American Life Project [Internet]. 2006:[22 p.]. Available from www pewinternet org/PPF/r/190/report display asp 27. Fox S. Online health search. PEW Internet and American Life Proje Available from: www.pewinternet.org/PPF/r/190/report_display.asp. 28. Wangberg S, Andreassen H, Kummervold P, Wynn R, Sorensen T. Use of the internet for health purposes: trends in Norway 2000-2010. Scandinavian Journal of Caring Sciences. 2009;23(4):691-6. 29. Pletneva N, Vargas A, Boyer C. Requirements for the general public health search. 2011. Available from: www.hon.ch/Global/pdf/Khresmoi/KHRESMOI_general_public_survey_report.pdf. 30. Statistics Norway. Use of the Internet, frequency and location of use last 3 months. Percentage of population, by sex, age, education and employment situation. 2nd quarter 2010 Statistics Norway; 2011 [cited 2011 19.05]; Available from: www.ssb.no/english/subjects/10/03/ikthus_en/tab-2010-09-24- 04-en.html. 31. Brownson R, Fielding J, Maylahn C. Evidence-Based Public Health: A Fundamental Conce 31. Brownson R, Fielding J, Maylahn C. Evidence-Based Public Health: A Fundamental Concept for Public Health Practice. Annual Review of Public Health. 2009;30:175-201. 32. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. 2010 [cited 2010 30.11.2010]; Available from: www.hon.ch/HONcode/Webmas the net intiative. 2010 [cited 2010 30.11.2010]; Available from: www.hon.ch/HONcode/Webmasters/. 39 Berger B Steckelberg A Meyer G Kasper J Muhlhauser I Training of patient and consumer 39. Berger B, Steckelberg A, Meyer G, Kasper J, Muhlhauser I. Training of patient and consumer representatives in the basic competencies of evidence-based medicine: a feasibility study. BMC medical education. 2010;10:1-9. 40. Peerson A, Saunders M. Health literacy revisited: what do we mean and why does it matter? Health promotion international. 2009;24(3):285-96. 41. Freedman DA, Bess KD, Tucker HA, Boyd DL, Tuchman AM, Wallston KA. Public Health Literacy Defined. American Journal of Preventive Medicine. 2009;36(5):446-51. 41. Freedman DA, Bess KD, Tucker HA, Boyd DL, Tuchman AM, Wallston KA. Public Health Literacy Defined. American Journal of Preventive Medicine. 2009;36(5):446-51. 42. Ishikawa H, Yano E. Patient health literacy and participation in the health-care process. Health Expectations. 2008;11(2):113-22. 43. Nielsen-Bohlman L, Institute of Medicine (U.S.). Committee on Health Literacy. Health literacy: a prescription to end confusion. Washington, D.C.: National Academies Press; 2004. xix, 345 p. 44. Nutbeam D. Health literacy as a public health goal: A challenge for contemporary health education and communication strategies into the 21st century. Health promotion international. 2000;15(3):259-67. 45. The Lancet. Taking health literacy seriously. The Lancet. 2005;366(9480):95. 46. Pirisi A. Low health literacy prevents equal access to care. The Lancet. 2000;356(9244 47. Hironaka L, Paasche-Orlow M. The implications of health literacy on patient-provider communication. Archives of Disease in Childhood. 2008;93(5):428-32. . Coulter A. Editorial: Promoting realistic expectations. Health Expectations. 2000;3(3):159-60 48. Coulter A. Editorial: Promoting realistic expectations. Health Expectations. 2000;3(3):159-60. 49. Zarcadoolas C, Pleasant A, Greer D. Understanding health literacy: an expanded model. Health promotion international. 2005;20(2):195-203. 49. Zarcadoolas C, Pleasant A, Greer D. Understanding health literacy: an expanded model. Health promotion international. 2005;20(2):195-203. promotion international. 2005;20(2):195-203. 50. Muir K, Santiago-Turla C, Stinnett S, Herndon L, Allingham R, Challa P, et al. Health literacy and adherence to glaucoma therapy. American journal of ophthalmology. 2006;142(2):223-6. 50. Muir K, Santiago-Turla C, Stinnett S, Herndon L, Allingham R, Challa P, et al. Health literacy and adherence to glaucoma therapy. American journal of ophthalmology. 2006;142(2):223-6. 51. Morrow D, Clark D, Tu W, Wu J, Weiner M, Steinley D, et al. Correlates of health literacy in patients with chronic heart failure. The Gerontologist. 2006;46(5):669-76. 52. Jordan J, Osborne R, Buchbinder R. 57. Schwartz L, Woloshin S, Black W, Welch H. The role of numeracy in understanding the benefit of screening mammography. Annals of Internal Medicine. 1997;127(11):966-72. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. The Norwegian Directorate for Health and Social Affairs. ...and it's going to get better! National 31. Brownson R, Fielding J, Maylahn C. Evidence-Based Public Health: A Fundamental Concept for Public Health Practice. Annual Review of Public Health. 2009;30:175-201. Public Health Practice. Annual Review of Public Health. 2009;30:175-201. The Norwegian Directorate for Health and Social Affairs. ...and it's going to get better! National for Quality Improvement in health and Social Services (2005-2015). 2005. 32. The Norwegian Directorate for Health and Social Affairs. ...and it's going to get better! N 32. The Norwegian Directorate for Health and Social Affairs. ...and it's going to get better! National Strategy for Quality Improvement in health and Social Services (2005-2015). 2005. g g g g Strategy for Quality Improvement in health and Social Services (2005-2015). 2005. 33. Forsetlund L, Bjorndal A. Identifying barriers to the use of research faced by public health physicians in Norway and developing an intervention to reduce them. Journal of health services research & policy. 2002;7(1):10-8. 34. Treweek S, Flottorp S, Fretheim A, Havelsrud K, Kristoffersen D, Oxman A, et al. [What do general practitioners do to keep themselves up to date?]. Tidsskrift for Den norske Legeforening. 2005;125(3):304-6. 35. Havelsrud K, Flottorp S, Fretheim A, Kristoffersen D, Oxman A, Treweek S, et al. [How do general practice assistants keep themselves up to date?]. Tidsskrift for Den norske Legeforening. 2005;125(3):307-9. 36. Glenton C, Santesso N, Rosenbaum S, Nilsen E, Rader T, Ciapponi A, et al. Presenting the results of Cochrane Systematic Reviews to a consumer audience: a qualitative study. Medical Decision Making. 2010;30(5):566-77. 78 37. Eysenbach G, Diepgen T. Towards quality management of medical information on the internet: evaluation, labelling, and filtering of information. The British Medical Journal. 1998;317(7171):1496-500. 38. Health On The Net Foundation (HON). Trustworthy health and medical information:the 38. Health On The Net Foundation (HON). Trustworthy health and medical information:the health on the net intiative. 2010 [cited 2010 30.11.2010]; Available from: www.hon.ch/HONcode/Webmasters/. 39. Berger B, Steckelberg A, Meyer G, Kasper J, Muhlhauser I. Training of patient and consumer representatives in the basic competencies of evidence-based medicine: a feasibility study. BMC medical education. 2010;10:1-9. the net intiative. 2010 [cited 2010 30.11.2010]; Available from: www.hon.ch/HONcode/Webmasters/. 39. Berger B, Steckelberg A, Meyer G, Kasper J, Muhlhauser I. Training of patient and consumer representatives in the basic competencies of evidence-based medicine: a feasibility study. BMC medical the net intiative. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. How can we help people make sense of medical data? Effective clinical practice. 1999 2(4) 176 83 adults in hospital-based primary care practices. Patient education and counseling. 2008;73(2):377-83. 63. Woloshin S. How can we help people make sense of medical data? Effective clinical practice. 1999;2(4):176-83. 63. Woloshin S. How can we help people make sense of medical data? Effective clinical practic 1999;2(4):176-83. 64. Dickersin K, Braun L, Mead M, Millikan R, Wu AM, Pietenpol J, et al. Development and implementation of a science training course for breast cancer activists: Project LEAD (leadership, education and advocacy development). Health Expectations. 2001;4(4):213-20. 64. Dickersin K, Braun L, Mead M, Millikan R, Wu AM, Pietenpol J, et al. Development and implementation of a science training course for breast cancer activists: Project LEAD (leadership, education and advocacy development). Health Expectations. 2001;4(4):213-20. 65. Murray E, Burns J, See T, Lai R, Nazareth I. Interactive Health Communication Applications for people with chronic disease. Cochrane database of systematic reviews. 2005(4):CD004274. 66. Wantland D, Portillo C, Holzemer W, Slaughter R, McGhee E. The effectiveness of Web-based vs. non-Web-based interventions: a meta-analysis of behavioral change outcomes. Journal of Medical Internet Research. 2004;6(4):40. 67. Lewis D. Computer-based approaches to patient education: a review of the literature. Journal of the American Medical Informatics Association. 1999;6(4):272-82. 68. Grol R, Baker R, Moss F. Quality improvement research: the science of change in health care. In: Grol R, Baker R, Moss F, editors. Quality improvement research: understanding change in health care. London: BMJ Publishing Group; 2004. p. 1-5. 69. Craig P, Dieppe P, Macintyre S, Mitchie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: new guidance: Medical Research Council; 2008. Available from: www.mrc.ac.uk/complexinterventionsguidance. 70. Van Bokhoven M, Kok G, Van der Weijden T. Designing a quality improvement intervention: a systematic approach. In: Grol R, Baker R, Moss F, editors. Quality improvement research. London: BMJ Publishing Group; 2004. p. 147-64. 71. Campbell M, Fitzpatrick R, Haines A, Kinmonth A, Sandercock P, Spiegelhalter D, et al. Framework for design and evaluation of complex interventions to improve health. The British Medical Journal. 2000;321(7262):694-6. 72. Pope C, Van Royen P, Baker R. Qualitative methods in research on healthcare quality. In: Grol R, Baker R, Moss F, editors. Quality improvement research. London: BMJ Publishing Group; 2004. p. 51-63. 73. Eccles M, Grimshaw J, Campbell M, Ramsay C. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Critical appraisal of health literacy indices revealed variable underlying constructs, narrow content and psychometric weaknesses. Journal of clinical epidemiology. 2011;64(4):366-79. 53. The Department of International Health. Measuring health literacy in Europe. Maastricht University. 2011 [22.12.2011]; Available from: http://inthealth.eu/research/health-literacy-hls- eu/measuring health literacy in europe/ 54. Schwartz K, Roe T, Northrup J, Meza J, Seifeldin R, Neale A. Family medicine patients' use of the Internet for health information: a MetroNet study. Journal of the American Board of Family Medicine. 2006;19(1):39-45. 55. Eysenbach G, Kohler C. How do consumers search for and appraise health information on the world wide web? Qualitative study using focus groups, usability tests, and in-depth interviews. The British Medical Journal. 2002;324(7337):573-7. 56. Woloshin S, Schwartz L, Byram S, Sox H, Fischhoff B, Welch H. Women's understanding of the mammography screening debate. Archives of internal medicine. 2000;160:1434-40. 57. Schwartz L, Woloshin S, Black W, Welch H. The role of numeracy in understanding the benefit of screening mammography. Annals of Internal Medicine. 1997;127(11):966-72. 79 58. Mills E, Jadad AR, Ross C, Wilson K. Systematic review of qualitative studies exploring parental beliefs and attitudes toward childhood vaccination identifies common barriers to vaccination. Journal of clinical epidemiology. 2005;58(11):1081-8. 59. Lloyd A. The extent of patients' understanding of the risk of treatments. Quality in Health Care. 2001;10 Suppl 1:i14-8. 60. Robinson E, Kerr C, Stevens A, Lilford R, Braunholtz D, Edwards S, et al. Lay public's understanding of equipoise and randomisation in randomised controlled trials. Research Suppo 60. Robinson E, Kerr C, Stevens A, Lilford R, Braunholtz D, Edwards S, et al. Lay public's understanding of equipoise and randomisation in randomised controlled trials. Research Support, Non- U.S. Gov't. NHS R&D HTA Programme, 2005 Mar. Report No.: Contract No.: 8. U.S. Gov't. NHS R&D HTA Programme, 2005 Mar. Report No.: Contract No.: 8. 61. Ganther-Urmie J, Nair K, Valuck R, McCollum M, Lewis S, Turpin R. Consumer attitudes and factors related to prescription switching decisions in multitier copayment drug benefit plans. American Journal of Managed Care. 2004;10(3):201-8. 62. Iosifescu A, Halm E, McGinn T, Siu A, Federman A. Beliefs about generic drugs among elderly adults in hospital-based primary care practices. Patient education and counseling. 2008;73(2):377-83. 62. Iosifescu A, Halm E, McGinn T, Siu A, Federman A. Beliefs about generic drugs among elderly adults in hospital-based primary care practices. Patient education and counseling. 2008;73(2):377-83. 63. Woloshin S. 94. Renz A, Ide M, Newton T, Robinson P, Smith D. Psychological interventions to improve adherence to oral hygiene instructions in adults with periodontal diseases. Cochrane database of systematic reviews. 2007(2):No.: CD005097. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Research designs for studies evaluating the effectiveness of change and improvement strategies. In: Grol R, Baker R, Moss F, editors. Quality improvement research: understanding change in health care. London: BMJ Publishing Group; 2004. p. 97-114. 74. Statistics Norway. [Use of health services by sex and age]. Statistics Norway,: SSB; 2009 [updated 26.6.2009; cited 2010 01.12.2010]; Available from: 74. Statistics Norway. [Use of health services by sex and age]. Statistics Norway,: SSB; 2009 [updated 26.6.2009; cited 2010 01.12.2010]; Available from: http://statbank.ssb.no/statistikkbanken/Default_FR.asp?PXSid=0&nvl=true&PLanguage=0&tilside=select varval/define.asp&Tabellid=04217. 80 75. Westin S. [Post-modern anti-vaccination movement]. Tidsskrift for Den norske Legeforening. 2002;20(122(8)):772. 76. Horton R. A statement by the editors of The Lancet. The Lancet. 2004;363(9411):820-1. 77. Goldacre B. Medicine and the media: MMR: the scare stories are back. The British Medical Journal. 2007;335(7611):126-7. 78. Danielsen S. Pasientmedvirkning fra pasientens perspektiv -partnerskap eller paternalisme? En kvalitativ studie [Masteroppgave i Klinisk Sykepleievitenskap]: Høgskolen i Oslo; 2011. 78. Danielsen S. Pasientmedvirkning fra pasientens perspektiv -partnerskap eller paternalisme? En kvalitativ studie [Masteroppgave i Klinisk Sykepleievitenskap]: Høgskolen i Oslo; 2011. 79. Francis J, Eccles M, Johnston M, Walker A, Grimshaw J, Foy R, et al. Constructing questionnaires based on the theory of planned behaviour: A manual for health services researchers. 2004:[42 p.]. Available from: www.rebeqi.org/ViewFile.aspx?itemID=212. 79. Francis J, Eccles M, Johnston M, Walker A, Grimshaw J, Foy R, et al. Constructing questionnaires based on the theory of planned behaviour: A manual for health services researchers. 2004:[42 p.]. Available from: www.rebeqi.org/ViewFile.aspx?itemID=212. 80. Charmaz K. Constructing grounded theory: a practical guide through qualitative analysis. London: Sage; 2006. XIII, 208 p. 80. Charmaz K. Constructing grounded theory: a practical guide through qualitative analysis. London: Sage; 2006. XIII, 208 p. 81. Billson JM. The power of focus groups: a training manual for social, policy and market research. 4th ed Barrington Rhode Island: Skywood Press; 2004 81. Billson JM. The power of focus groups: a training manual for social, policy and market research. 4th ed. Barrington, Rhode Island: Skywood Press; 2004. 82. Malterud K, editor. [Qualitative methods in medical research]. Oslo: Universitetsforlaget; 20 82. Malterud K, editor. [Qualitative methods in medical research]. Oslo: Universitetsforlaget; 2004. 83. Kitzinger J. Focus groups with users and providers of health care. In: Pope C, Qualitative Research in Health Care. London: BMJ Publishing Group; 2001. p. 20-9. 84. Kummervold P, Chronaki C, Lausen B, Prokosch H, Rasmussen J, Santana S, et al. eHealth Trends in Europe 2005-2007: A Population-Based Survey. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Medical Internet Research. 2008;Oct-Dec; 10(4): e42. 85. Ritchie J, Spencer L. Qualitative data analysis for applied policy research. In: Bryman A, Burgess R, 84. Kummervold P, Chronaki C, Lausen B, Prokosch H, Rasmussen J, Santana S, et al. eHealth Trends in Europe 2005-2007: A Population-Based Survey. Medical Internet Research. 2008;Oct-Dec; 10(4): e42. 85. Ritchie J, Spencer L. Qualitative data analysis for applied policy research. In: Bryman A, Burgess R, editors. Analyzing qualitative data. London: Routledge; 1994. p. 173-94. 86. Grol R, Wensing M, Hulscher M, Eccles M. Theories on implementation of change in healthcare. In: Grol R, Wensing M, Eccles M, editors. Improving patient care The implementation of change in general practice. London: Elsevier limited; 2005. p. 15- 40. 87. Eccles M, Grimshaw J, Johnston M, Steen N, Pitts NB, Thomas R, et al. Applying psychological theories to evidence-based clinical practice: identifying factors predictive of managing upper respiratory tract infections without antibiotics. Implement Science. 2007;2:26:1-14. 88. Ajzen I. The Theory of Planned Behavior. Organizational Behavior and Human Decision Processes. 1991;50(2):179-211. 89. Conner M, Norman P. Theory of planned behaviour and health behaviour. In: Conner M, Norman P, editors. Predicting health behaviour : research and practice with social cognition models. 2nd ed. Maidenhead: Open University Press; 2005. p. 170-222. 90. Armitage C, Conner M. Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology. 2001;40:471-99. 91. Eccles M, Hrisos S, Francis J, Kaner E, Dickinson H, Beyer F, et al. Do self- reported intentions predict clinicians' behaviour: a systematic review. Implementation Science [Internet]. 2006 1664582]; 1: 28. Available from: www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids= 92. Stead M, Tagg S, MacKintosh A, Eadie D. Development and evaluation of a mass media Theory of Planned Behaviour intervention to reduce speeding. Health Education Research. 2005;20(1):36-50. 93. Krones T, Keller H, Becker A, Sonnichsen A, Baum E, Donner-Banzhoff N. The theory of planned behaviour in a randomized trial of a decision aid on cardiovascular risk prevention. Patient education and counseling. 2010;78(2):169-76. 94. Renz A, Ide M, Newton T, Robinson P, Smith D. Psychological interventions to improve adherence to oral hygiene instructions in adults with periodontal diseases. Cochrane database of systematic reviews. 2007(2):No.: CD005097. 81 95. Gatt S, Sammut R. An exploratory study of predictors of self-care behaviour in persons with type 2 diabetes. International Journal of Nursing Studies. 2008;45(10):1525-33. 95. Gatt S, Sammut R. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. An exploratory study of predictors of self-care behaviour in persons with type 2 diabetes. International Journal of Nursing Studies. 2008;45(10):1525-33. 96. Luzzi L, Spencer AJ. Factors influencing the use of public dental service Theory of Planned Behaviour. BMC health services research. 2008;8:93. 96. Luzzi L, Spencer AJ. Factors influencing the use of public dental services: an application of the Theory of Planned Behaviour. BMC health services research. 2008;8:93. 97. Ramsay C, Thomas R, Croal B, Grimshaw J, Eccles M. Using the theory of planned behaviour as a process evaluation tool in randomised trials of knowledge translation strategies: A case study from UK primary care. Implementation Science. 2010;5(71):9. 97. Ramsay C, Thomas R, Croal B, Grimshaw J, Eccles M. Using the theory of planned behaviour as a process evaluation tool in randomised trials of knowledge translation strategies: A case study from UK primary care. Implementation Science. 2010;5(71):9. 98. Tickner S, Leman PJ, Woodcock A. The Immunisation Beliefs and Intentions Measure (IBIM): Predicting parents' intentions to immunise preschool children. Vaccine. 2010;28(19):3350-62. 99. Webb T, Joseph J, Yardley L, Michie S. Using the Internet to Promote Health Behavior Change: A Systematic Review and Meta-analysis of the Impact of Theoretical Basis, Use of Behavior Change Techniques, and Mode of Delivery on Efficacy. Journal of Medical Internet Research. 2010;12(1):e4. 100. Moan IS, Rise J. Quitting smoking: Applying an extended version of the theory of planned behavior to predict intention and behavior. Journal of Applied Biobehavioral Research. 2005;10(1):39-68. 101. Edwards P, Roberts I, Clarke M, Diguiseppi C, Wentz R, Kwan I, et al. Methods to increase response to postal and electronic questionnaires. Cochrane database of systematic reviews. 2009(3):MR000008. 102. Austvoll-Dahlgren A, Helseth S. What informs parents' decision-making about childhood vaccinations? Journal of Advanced Nursing. 2010;66(11):2421-30. 103. Austvoll-Dahlgren A, Sørum R, Helseth S. Cognitive factors predicting intentions to search 103. Austvoll-Dahlgren A, Sørum R, Helseth S. Cognitive factors pred 103. Austvoll-Dahlgren A, Sørum R, Helseth S. Cognitive factors predicting intentions to search for health information: an application of the Theory of Planned Behaviour. 2012. 103. Austvoll Dahlgren A, Sørum R, Helseth S. Cognitive factors predicting intentions to health information: an application of the Theory of Planned Behaviour. 2012. health information: an application of the Theory of Planned Behaviour. 2012. 104. Austvoll-Dahlgren A, Helseth S. Public health nurses’ barriers and facilitators to the use of research in consultations about childhood vaccinations. Scandinavian Journal of Caring Sciences [Internet]. 2011. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. 105. Sillence E, Briggs P, Harris PR, Fishwick L. How do patients evaluate and make use of online health information? Social science & medicine. 2007;64(9):1853-62. 106. Peterson G, Aslani P, Williams KA. How do consumers search for and appraise information on medicines on the Internet? A qualitative study using focus groups. Journal of Medical Internet Research. 2003;5(4):e33. 107. Weinstein N. What does it mean to understand a risk? Evaluating risk comprehension. Journal of the National Cancer Institute. 1999(25):15-20. 108. Tickner S, Leman P, Woodcock A. Factors underlying suboptimal childhood immunisation. Vaccine. 2006;24(49-50):7030-6. 109. Stead M, Eadie D, Gordon D, Angus K. "Hello, hello - it's English I speak"! a qualitative exploration of patients' understanding of the science of clinical trials. J Med Ethics. 2005;31(11):664-9. 110. Andreassen H, Sandaune A, Gammon D, Hjortdahl P. [Norwegian use of Internet health services]. Tidsskrift for Den norske Legeforening 2002;122(17):1640-4. 111. Quintana Y, Feightner J, Wathen C, Sangster L, Marshall J. Preventive health information on the Internet. Qualitative study of consumers' perspectives. Canada Family Physician. 2001;47:1759-65. 112. Morville P. User Experience Design; honeycomb model. 2004 [cited 2010 19.05]; Available from: http://semanticstudios.com/publications/semantics/000029.php. 113. Bjørndal A, Flottorp S, Klovning A. Knowledge management in medicine and health sciences [Kunnskapshåndtering i medisin og helsefag] Oslo: Gyldendal akademisk; 2007. 195 p. 114. Trevena L, Davey H, Barratt A, Butow P, Caldwell P. A systematic review on communicating with patients about evidence. Journal of evaluation in clinical practice. 2006;12(1):13-23. 82 115. Keselman A, Logan R, Smith CA, Leroy G, Zeng-Treitler Q. Developing Informatics tools and strategies for consumer-centered health communication. Journal of the American Medical Informatics Association. 2008;15(4):473-83. 116. Epstein R, Alper B, Quill T. Communicating Evidence for Participatory Decision Making. Journal of American Medical Association. 2004;291(19):2359-66. 117. Woloshin S, Schwartz L, Welch H. Know Your Chances: Understanding Health Statistics. London: University of California Press; 2008. 158 p. 118. Irwig L, Irwig J, Trevena L, Sweet M. Smart health choices: making sense of health advice. London: Hammersmith; 2008. xi, 242 p. 119. Benestad H, Laake P. [Research Methodology in the Medical and Biological Sciences]. In: Benestad H, Laake P, editors.: Gyldendal akademisk; 2004. 119. Benestad H, Laake P. [Research Methodology in the Medical and Biological Sciences]. In: Benestad H, Laake P, editors.: Gyldendal akademisk; 2004. 120 Johannessen A Tufte P Kristoffersen L [Introduction to social science methodology] 120. Johannessen A, Tufte P, Kristoffersen L. [Introduction to social science methodology]. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Johannessen A, Tufte P, Kristoffersen L, editors. Oslo: Abstrakt forlag as; 2007. 121. Regionale forskningskomiteer. [Research ethics]. 2012 [24.01.2012]; Available from: http://www.etikkom.no/no/Forskningsetikk/. 122. Kim P, Eng T, Deering M, Maxfield A. Published criteria for evaluating health related web sites: review. The British Medical Journal. 1999;318(7184):647-9. 123. Gagliardi A, Jadad A. Examination of instruments used to rate quality of health information on the internet: chronicle of a voyage with an unclear destination. The British Medical Journal. 2002;324(7337):569-73. ( ) 124. Milne R, Oliver S. Evidence-based Consumer Health Information: Developing Teaching in Critical Appraisal Skills. International Journal for Quality in Health Care. 1996;8(5):439-45. 125. Charnock D, Shepperd S, Needham G, Gann R. DISCERN: an instrument for judging the quality of written consumer health information on treatment choices. Journal of epidemiology and community health. 1999;53(2):105-11. 126. Raffe A, Gray J. Screening: Evidence and practice. Oxford: Oxford University Press; 2007. 127. Stacey D, Bennett C, Barry M, Col N, Eden K, Holmes-Rovner M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane database of systematic reviews. 2011(10):CD001431. 128. IPDASI. Criteria for judging the quality of patient decision aids. 2005 [cited 2010 18. 05]; Available from: http://www.ipdas.ohri.ca/what.html. 129. Steinsbekk A. [Patient Activation Measure]. Tidsskrift for den Norske Legeforening. 2008;128(20):2316-8. 130. Charnock D. Håndboken DISCERN. Kvalitetskriterier for helseinformasjon til forbrukere om hvilke behandlingsalternativer som finnes1998:[65 p.]. Available from: // / / / http://www.nifab.no/pasientveiledning/hvordan_vurdere_informasjon/discern_verktoey_for_evaluerin g_av_helseinformasjon. 131. O'Connor A, Stacey D, Jacobsen M. Ottawa personal decision guide. 2007 [24.01.12]; Available from: http://decisionaid.ohri.ca/docs/das/OPDG.pdf. 132. DiCenso A, Bayley L, Haynes R. Accessing preappraised evidence: fine-tuning the 5S model into a 6S model. Annals of Internal Medicine. 2009;151(6):99-101. 132. DiCenso A, Bayley L, Haynes R. Accessing preappraised evidence: fine-tuning the 5S model into a 6S model. Annals of Internal Medicine. 2009;151(6):99-101. 133. Baker R, Camosso-Stefinovic J, Gillies C, Shaw E, Cheater F, Flottorp S, et al. Tailored interventions to overcome identified barriers to change: effects on professional practice and health care outcomes. Cochrane database of systematic reviews. 2010(3):CD005470. 133. Baker R, Camosso-Stefinovic J, Gillies C, Shaw E, Cheater F, Flottorp S, et al. Tailored interventions to overcome identified barriers to change: effects on professional practice and health care outcomes. Cochrane database of systematic reviews. 2010(3):CD005470. y 134. Treweek S, Zwarenstein M. Making trials matter: pragmatic and explanatory trials and the problem of applicability. Trials. 2009;10(37):9. 134. Treweek S, Zwarenstein M. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Making trials matter: pragmatic and explanatory trials and the problem of applicability. Trials. 2009;10(37):9. 83 135. Roland M, Torgerson D. What are pragmatic trials? The British Medical Journal. 135. Roland M, Torgerson D. What are pragmatic trials? The British Medical Journal. 1998;316(7127):285. 135. Roland M, Torgerson D. What are pragmatic trials? The British Medical Journal. 1998;316(7127):285. 136. Godwin M, Ruhland L, Casson I, MacDonald S, Delva D, Birtwhistle R, et al. Pragmatic controlled clinical trials in primary care: the struggle between external and internal validity. BMC medical research methodology. 2003;3(28):7. gy ( ) 137. Haug C. [Panic pandemic?]. Tidsskrift for Den norske Legeforening. 2009;129(10):979. Epub 2009/05/19. Panikkpandemi? 138. The Norwegian Institute of Public Health. The vaccination handbook e-book [Vaksinasjonshåndboka e-bok]: The Norwegian Institute of Public Health; 2007. Available from: (www.fhi.no/eway/). g [Vaksinasjonshåndboka e-bok]: The Norwegian Institute of Public Health; 2007. Available from: (www.fhi.no/eway/). 139. Cassell J, Leach M, Poltorak M, Mercer C, Iversen A, Fairhead J. Is the cultural context of MMR rejection a key to an effective public health discourse? Public Health. 2006;120(9):783-94. 140. Tickner S, Leman PJ, Woodcock A. 'It's just the normal thing to do': exploring parental decision- making about the 'five-in-one' vaccine. Vaccine. 2007;25(42):7399-409. 141. Benin A, Wisler-Scher D, Colson E, Shapiro E, Holmboe E. Qualitative analysis of mothers' decision-making about vaccines for infants: the importance of trust. Pediatrics. 2006;117(5):1532-41. 142. Gellin BG, Maibach E, Marcuse E. Do parents understand immunizations? A national telephone survey. Pediatrics. 2000;106(5):1097-102. decision-making about vaccines for infants: the importance of trust. Pediatrics. 2006;117(5):1532-41. 142. Gellin BG, Maibach E, Marcuse E. Do parents understand immunizations? A national telephone y ; ( ) 143. Haug C. [Prioritization on the paper--and in practice]. Tidsskrift for Den norske Legeforening. 2008;128(8):917. Prioritering pa papiret--og i praksis. 144. Teasdale E, Yardley L. Understanding responses to government health recommendations: public perceptions of government advice for managing the H1N1 (swine flu) influenza pandemic. Patient education and counseling. 2011;85(3):413-8. 145. Rygg M. [Skepticism against vaccination--paradox of preventive work]. Tidsskrift for Den Norske Laegeforening. 2006;126(22). 146. de Boer M, Versteegen G, van Wijhe M. Patients' use of the Internet for pain-related medical information. Patient education and counseling. 2007;68(1):86-97. 147. Diaz J, Griffith R, Ng J, Reinert S, Friedmann P, Moulton A. Patients' use of the Internet for medical information. Journal of general internal medicine. 2002;17(3):180-5. 148. Sabel M, Strecher V, Schwartz J, Wang T, Karimipour D, Orringer J, et al. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Patterns of Internet use and impact on patients with melanoma. Journal of the American Academy of Dermatology. 2005;52(5):779-85. 149. Wainstein B, Sterling-Levis K, Baker S, Taitz J, Brydon M. Use of the Internet by parents of paediatric patients. Journal of paediatrics and child health. 2006;42(9):528-32. 150. Bylund C, Gueguen J, Sabee C, Imes R, Li Y, Sanford A. Provider-patient dialogue about internet health information: An exploration of strategies to improve the provider-patient relationship. Patient education and counseling. 2007;66(3):346-52. 151. Tuffrey C, Finlay F. Use of the internet by parents of paediatric outpatients. Archives of Disease in Childhood. 2002;87(6):534-6. 152. Gordon MM, Capell HA, Madhok R. The use of the Internet as a resource for health information among patients attending a rheumatology clinic. Rheumatology. 2002;41(12):1402-5. 153. Bozionelos G, Bennett P. The theory of planned behaviour as predictor of exercise: the moderating influence of beliefs and personality variables. Journal of health psychology. 1999;4(4):517- 29. 154. Opheim E. Brukermedvirkning på systemnivå: en kvalitativ studie av nytteverdi av kurs i kunnskapsbasert praksis for brukerrepresentanter. [Mastergradsoppgave]: Høgskolen i Bergen; 2011. 84 155. Poland G, Jacobson R. Understanding those who do not understand: a brief review of the anti- vaccine movement. Vaccine. 2001;19(17-19):2440-5. 155. Poland G, Jacobson R. Understanding those who do not understand: a brief review of the anti- vaccine movement. Vaccine. 2001;19(17-19):2440-5. 156. Allen A. Vaccine: the controversial story of medicine's greatest lifesaver. New York: Norton; 2007. 523 p. 156. Allen A. Vaccine: the controversial story of medicine's greatest lifesaver. New York: Norton; 2007. 523 p. 157. Steinman M. Using patients to promote evidence-based prescribing: comment on "Communicating uncertainties about prescription drugs to the public". Archives of internal medicine. 2011;171(16):1468-9. 157. Steinman M. Using patients to promote evidence-based prescribing: comment on "Communicating uncertainties about prescription drugs to the public". Archives of internal medicine. 2011;171(16):1468-9. 158. Pilling S, Williams M, Brackett R, Gourley R, Weg M, Christensen A, et al. Part I, patient perspective: activating patients to engage their providers in the use of evidence-based medicine: a qualitative evaluation of the VA Project to Implement Diuretics (VAPID). Implementation Science. 2010;5:23. 158. Pilling S, Williams M, Brackett R, Gourley R, Weg M, Christensen A, et al. Part I, patient perspective: activating patients to engage their providers in the use of evidence-based medicine: a qualitative evaluation of the VA Project to Implement Diuretics (VAPID). Implementation Science. 2010;5:23. 159. ailable from: www.pickereurope.org/Filestore/PIE_reports/QQUIP/QEI_Review_AB.pdf. Kinnersley P, Edwards A, Hood K, Cadbury N, Ryan R, Prout H, et al. Interventions before consultations for helping patients address their information needs. Cochrane database of systematic reviews. 2007(3):CD004565. 159. Kinnersley P, Edwards A, Hood K, Cadbury N, Ryan R, Prout H, et al. Interventions before consultations for helping patients address their information needs. Cochrane database of systematic reviews. 2007(3):CD004565. 85 86 Appendixes Appendix 1: Copy of Interview guides focus groups Intervjuguide foreldre Introduksjon Dere har nå alle i løpet av de siste tre månedene vært til konsultasjon med helsesøster om vaksinasjon. Fokuset for gruppediskusjonen vil være deres opplevelse av konsultasjonen med helsesøster og om beslutningsprosessen rundt vaksinasjon. Avslutningsvis vil jeg gjerne høre deres meninger om det å oppsøke og ta i bruk forskning om helse som støtte i beslutninger. - Introduksjon av moderator og sekretær - Avklare uavhengighet - Hva er en fokusgruppe - Oppfordre til diskusjon- målet er ikke å oppnå enighet men meningsutveksling - Regler underveis (la en og en snakke ut, la alle slippe til, evt pause etc) - Introduksjon av deltakere (hvorfor/ hvem) Beslutningsprosessen 1. Hvordan opplevde dere generelt konsultasjonen med helsesøster om vaksinasjon? Probe: - Hva var positivt? Hva var negativt? 2. Hvordan opplevde dere informasjonen som ble gitt i konsultasjonen med helsesøster om vaksinasjon? Probe: - Hvordan ble informasjonen gitt? - Var noe av denne informasjonen forskningsresultater som beskrev fordeler og ulemper ved å la barnet vaksineres eller ikke vaksineres? 3. Hva var det dere baserte beslutningen om vaksinasjon på? Probe: - Helsesøsters råd, egen kunnskap forut for konsultasjon, annet? - Oppsøkte dere noen informasjon om vaksinasjon på egen hånd? a. Hvis, ja i så fall hvor? Appendixes Appendix 1: Copy of Interview guides focus groups Appendix 1: Copy of Interview guides focus groups - Introduksjon av moderator og sekretær - Avklare uavhengighet - Hva er en fokusgruppe - Oppfordre til diskusjon- målet er ikke å oppnå enighet men meningsutveksling - Regler underveis (la en og en snakke ut, la alle slippe til, evt pause etc) - Introduksjon av deltakere (hvorfor/ hvem) b. Var noe av denne informasjonen forskningsresultater? 4. Hvordan opplevde dere beslutningsprosessen om vaksinasjon? P b 5. Hvem var det som til syvende og sist tok beslutningen om å vaksinere eller ikke vaksinere? Nå skal jeg spørre dere noen spørsmål om medisinsk og helserelatert forskning (Holdninger) 6. Hva mener dere om det å oppsøke og ta i bruk forskning som støtte i beslutninger om vaksinasjon? Probe: - Hva er fordeler? Hva er ulemper? - Er det noe som særlig gjelder i forhold til vaksinasjon sammenliknet med andre 7. Kjenner dere til noen som ville synes at det var positivt eller negativt om dere oppsøkte forskning som støtte i beslutninger om vaksinasjon? Probe: - Helsestasjon? Familie? Andre individer eller grupper? - Helsestasjon? Familie? Andre individer eller grupper? (Opplevd kontroll med atferden) 8. Hvilke faktorer eller omstendigheter påvirker om dere oppsøker og tar i bruk forskning som støtte i beslutninger om vaksinasjon? Probe: - Hva må ligge til rette? Hva gjør det vanskelig? Beslutningsprosessen Beslutningsprosessen 3. Hva var det dere baserte beslutningen om vaksinasjon på? Probe: - Helsesøsters råd, egen kunnskap forut for konsultasjon, annet? - Oppsøkte dere noen informasjon om vaksinasjon på egen hånd? a. Hvis, ja i så fall hvor? - Helsesøsters råd, egen kunnskap forut for konsultasjon, annet? - Oppsøkte dere noen informasjon om vaksinasjon på egen hånd? - Oppsøkte dere noen informasjon om vaksinasjon på egen hånd? a. Hvis, ja i så fall hvor? 86 87 b. Var noe av denne informasjonen forskningsresultater? 4. Hvordan opplevde dere beslutningsprosessen om vaksinasjon? Probe: - Hva var til hjelp? Hva var til hinder? 5. Hvem var det som til syvende og sist tok beslutningen om å vaksinere eller ikke vaksinere? Nå skal jeg spørre dere noen spørsmål om medisinsk og helserelatert forskning (Holdninger) 6. Hva mener dere om det å oppsøke og ta i bruk forskning som støtte i beslutninger om vaksinasjon? Probe: - Hva er fordeler? Hva er ulemper? - Er det noe som særlig gjelder i forhold til vaksinasjon sammenliknet med andre helsebeslutninger? (Subjektive normer) 7. Kjenner dere til noen som ville synes at det var positivt eller negativt om dere oppsøkte forskning som støtte i beslutninger om vaksinasjon? Probe: - Helsestasjon? Familie? Andre individer eller grupper? (Opplevd kontroll med atferden) 8. Hvilke faktorer eller omstendigheter påvirker om dere oppsøker og tar i bruk forskning som støtte i beslutninger om vaksinasjon? Probe: - Hva må ligge til rette? Hva gjør det vanskelig? Avrunding (signalisere at gruppen nærmer seg slutten) 9. Er det andre emner som dere kommer på når dere tenker på det å ta i bruk forskning i beslutninger om vaksinasjon? b. Var noe av denne informasjonen forskningsresultater? Avrunding (signalisere at gruppen nærmer seg slutten) 9. Er det andre emner som dere kommer på når dere tenker på det å ta i bruk forskning i beslutninger om vaksinasjon? 9. Er det andre emner som dere kommer på når dere tenker på det å ta i bruk forskning i beslutninger om vaksinasjon? 87 Oppsummering - Dobbeltsjekke konklusjoner med mulighet for mer debatt - Identifisere forskjeller, kontrasterende meninger - Gjøre en siste runde for lufting av meninger mulig - Takke! Oppsummering Intervjuguide helsesøstre Intervjuguide helsesøstre Introduksjon Fokuset for denne gruppediskusjonen vil være deres opplevelse av møtet med foreldre i konsultasjoner og beslutningsprosessen om vaksinasjon. Avslutningsvis vil jeg gjerne høre deres meninger om det å oppsøke og ta i bruk forskning som støtte i konsultasjoner. - Introduksjon av moderator og sekretær - Hva er en fokusgruppe - Oppfordre til diskusjon- målet er ikke å oppnå enighet men meningsutveksling - Regler underveis (la en og en snakke ut, la alle slippe til, evt pause etc) - Introduksjon av deltakere (hvorfor/ hvem) - Introduksjon av moderator og sekretær - Hva er en fokusgruppe - Oppfordre til diskusjon- målet er ikke å oppnå enighet men meningsutveksling - Regler underveis (la en og en snakke ut, la alle slippe til, evt pause etc) - Introduksjon av deltakere (hvorfor/ hvem) Beslutningsprosessen 10. Hvordan opplever dere generelt konsultasjonen med foreldrene om vaksinasjon? Probe: - Hva er positivt? Hva er negativt? - Hva er positivt? Hva er negativt? - Kan dere gi eksempler på hvordan dere informerer? l d k å i ? - Hva legger dere vekt på i møtet? 11. Hvordan opplever dere informasjonen/ressursene dere har tilgjengelig i rådgivningsarbeidet når det gjelder vaksinasjon? Probe: - Er noe av denne informasjonen forskningsresultater som beskriver fordeler og ulemper ved å la barnet vaksineres eller ikke vaksineres? - Oppsøker dere informasjon om vaksinasjon andre steder enn retningslinjene som støtte i rådgivningsarbeidet? a. Hvis, ja i så fall hvor? b. Er noe av denne informasjonen forskningsresultater? b. Er noe av denne informasjonen forskningsresultater? b. Er noe av denne informasjonen forskningsresultater? 12. Hvordan opplever dere beslutningsprosessen om vaksinasjon? Probe: 12. Hvordan opplever dere beslutningsprosessen om vaksinasjon? Probe: - Hva er til hjelp? Hva er til hinder? 13. Hva opplever dere at foreldrene baserer beslutningen sin om vaksinasjon på? Probe: - Helsesøsters råd, brukernes egen kunnskap forut for konsultasjon, annet? 14. Hvem er det som til syvende og sist tar beslutningen om å vaksinere eller ikke vaksinere? Nå skal jeg spørre dere noen spørsmål om medisinsk og helserelatert forskning (Holdninger) 15. Hva mener dere om det å oppsøke og ta i bruk forskning som støtte i konsultasjonen med foreldre om vaksinasjon? Probe: - Hva er fordelene? Hva er ulempene? Er det noe som særlig gjelder i forhold til vaksinasjon sammenliknet med andre - Hva er til hjelp? Hva er til hinder? (Subjektive normer) 16. Kjenner dere til noen som ville synes at det var positivt eller negativt om dere oppsøkte og tok i bruk forskning som støtte i konsultasjonen med foreldre om vaksinasjon? Probe: 16. Kjenner dere til noen som ville synes at det var positivt eller negativt om dere oppsøkte og tok i bruk forskning som støtte i konsultasjonen med foreldre om vaksinasjon? Probe: - Kollegaer? Brukerne? Andre individer eller grupper? (Opplevd kontroll med atferden) 17. Hvilke faktorer eller omstendigheter påvirker om dere oppsøker og tar i bruk forskning som støtte i konsultasjonen med foreldre om vaksinasjon? Probe: 17. Hvilke faktorer eller omstendigheter påvirker om dere oppsøker og tar i bruk forskning som støtte i konsultasjonen med foreldre om vaksinasjon? Probe: - Hva må ligge til rette? Hva gjør det vanskelig? 89 Avrunding (signalisere at gruppen nærmer seg slutten) 18. Er det andre emner som dere kommer på når dere tenker på det å ta i bruk forskning i konsultasjonen med foreldre om vaksinasjon? Oppsummering - Dobbeltsjekke konklusjoner med mulighet for mer debatt - Identifisere forskjeller, kontrasterende meninger - Gjøre en siste runde for lufting av meninger mulig - Takke! Avrunding (signalisere at gruppen nærmer seg slutten) 18. Er det andre emner som dere kommer på når dere tenker på det å ta i bruk forskning i konsultasjonen med foreldre om vaksinasjon? Oppsummering - Dobbeltsjekke konklusjoner med mulighet for mer debatt - Identifisere forskjeller, kontrasterende meninger - Gjøre en siste runde for lufting av meninger mulig - Takke! Avrunding (signalisere at gruppen nærmer seg slutten) 90 Appendix 2. b. Er noe av denne informasjonen forskningsresultater? Copy of questionnaire including TPB, PAM and Honeycomb 91 1) Svaralternativene til hvert spørsmål består av en skala. Husk å lese hvert spørsmål grundig før du krysser av svaret ditt da skalaene varierer noe fra spørsmål til spørsmål* 2) Jeg ønsker å oppsøke informasjon på egen hånd til bruk i fremtidige helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 3) Jeg forventer at jeg kommer til å oppsøke informasjon på egen hånd til bruk i fremtidige helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 4) Jeg har intensjoner om å oppsøke informasjon på egen hånd til bruk i fremtidige helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 5) Angi hvor usannsynlig eller sannsynlig de følgende påstander er for deg. Hvis jeg oppsøker informasjon på egen hånd til bruk i helsespørsmål: Usannsynlig 1 2 3 4 5 6 Sannsynlig 7 Vil det gi meg mer innsikt Vil det gi meg nyttig bakgrunnskunnskap i konsultasjon med helsepersonell Vil det være til hjelp hvis jeg er usikker på hva jeg skal gjøre i en helsebeslutning Vil det gi meg mer kunnskap hvis jeg opplever at informasjonen jeg får fra helsepersonell er ufullstendig 5) Angi hvor usannsynlig eller sannsynlig de følgende påstander er for deg. b. Er noe av denne informasjonen forskningsresultater? Hvis jeg oppsøker informasjon på egen hånd til bruk i helsespørsmål: 91 6) Angi hvor uønskelig eller ønskelig de følgende utfall er for deg: Svært uønskelig Uønskelig Noe uønskelig Verken uønskelig eller ønskelig Noe ønskelig Ønskelig Svært ønskelig Å få mer innsikt er Nyttig bakgrunnskunnskap i konsultasjon med helsepersonell er Å ha informasjon til hjelp hvis jeg er usikker på hva jeg skal gjøre i et helsespørsmål er Å få mer kunnskap hvis jeg opplever at jeg får ufullstendig informasjon fra helsepersonell er 7) Det å oppsøke informasjon på egen hånd til bruk i helsespørsmål er: 1 Verdiløst 2 3 4 5 6 7 Nyttig 8) Det å oppsøke informasjon på egen hånd til bruk i helsespørsmål er: 1 Positivt 2 3 4 5 6 7 Negativt 9) Det å oppsøke informasjon på egen hånd til bruk i helsespørsmål er: 1 Skadelig 2 3 4 5 6 7 Fordelaktig 10) Angi hvor negative eller positive de følgende grupper vil være til at du oppsøker informasjon på egen hånd til bruk i helsespørsmål: Svært negative Negative Noe negative Verken negative eller positive Noe positive Positive Svært positive Familie og venner Helsepersonell Andre sosiale grupper eller foreninger du deltar i (f.eks kollegaer, pasientforeninger etc) 6) Angi hvor uønskelig eller ønskelig de følgende utfall er for deg: 92 93 11) Hva familie og venner mener jeg skal gjøre er viktig for meg: 1 I svært liten grad 2 3 4 5 6 7 I svært stor grad 12) Hva helsepersonell mener jeg skal gjøre er viktig for meg: 1 I svært liten grad 2 3 4 5 6 7 I svært stor grad 13) Hva andre sosiale grupper eller foreninger som jeg deltar i mener jeg skal gjøre er viktig for meg: 1 I svært liten grad 2 3 4 5 6 7 I svært stor grad 14) Angi hvor usannsynlig eller sannsynlig de følgende påstander er for deg. b. Er noe av denne informasjonen forskningsresultater? Hvis jeg oppsøker informasjon på egen hånd til bruk i helsespørsmål: Usannsynlig 1 2 3 4 5 6 Sannsynlig 1 Føler jeg at det er vanskelig å få oversikt over all informasjonen Føler jeg at jeg ikke har kunnskap nok til å vurdere kvaliteten på informasjonen jeg finner Opplever jeg at det er tidkrevende å finne informasjonen jeg leter etter 15) Angi til i hvilken grad de følgende faktorene påvirker deg når det gjelder sannsynligheten for at du oppsøker informasjon på egen hånd til bruk i helsespørsmål: Mindre sannsynlig at jeg oppsøker informasjon 1 2 3 4 5 6 Mer sannsynlig at jeg oppsøker informasjon 7 Opplevelsen av at det er vanskelig å få oversikt over all informasjonen gjør det: Det å ikke ha kunnskap nok til å vurdere kvaliteten på informasjonen gjør det: At det å søke etter informasjon kan være tidkrevende gjør det: 11) Hva familie og venner mener jeg skal gjøre er viktig for meg: p Mindre sannsynlig at jeg oppsøker informasjon 1 2 3 4 5 6 Mer sannsynlig at jeg oppsøker informasjon 7 Opplevelsen av at det er vanskelig å få oversikt over all informasjonen gjør det: Det å ikke ha kunnskap nok til å vurdere kvaliteten på informasjonen gjør det: At det å søke etter informasjon kan være tidkrevende gjør det: 93 94 16) Generelt sett opplever jeg at personer som er viktige for meg synes at jeg skal oppsøke informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 17) Generelt sett føler jeg sosialt press til å søke etter informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 18) Generelt sett opplever jeg at det er forventet av meg at jeg oppsøker informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 19) Jeg føler at jeg mestrer det å oppsøke informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 20) Jeg har tillit til at jeg kan klare å oppsøke informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 21) Under står noen utsagn som folk av og til bruker når de snakker om helsen sin. b. Er noe av denne informasjonen forskningsresultater? Angi i hvor stor grad du er enig eller uenig med hvert utsagn. I utsagnene brukes ordet ‘behandling’, tenk på at forebyggende tiltak og livsstil også er behandlinger. Helt uenig Nokså uenig Nokså enig Helt enig Ikke aktuelt Når alt kommer til alt er jeg selv ansvarlig for å ta hånd om min egen helse Det aller viktigste for min egen helse og funksjonsevne er at jeg tar aktiv del i behandlingen Jeg er sikker på at jeg kan gjøre det som er nødvendig for å forebygge eller redusere symptomer eller problemer som skyldes min helsetilstand Jeg vet hvordan de forskjellige medisinene jeg har fått foreskrevet skal virke Jeg vet når jeg trenger medisinsk hjelp for et helseproblem og når jeg kan ta hånd om det selv 16) Generelt sett opplever jeg at personer som er viktige for meg synes at jeg skal oppsøke informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 17) Generelt sett føler jeg sosialt press til å søke etter informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 18) Generelt sett opplever jeg at det er forventet av meg at jeg oppsøker informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 19) Jeg føler at jeg mestrer det å oppsøke informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 20) Jeg har tillit til at jeg kan klare å oppsøke informasjon på egen hånd til bruk i helsespørsmål: 1 Svært uenig 2 3 4 5 6 7 Svært enig 21) Under står noen utsagn som folk av og til bruker når de snakker om helsen sin. Angi i hvor stor grad du er enig eller uenig med hvert utsagn. I utsagnene brukes ordet ‘behandling’, tenk på at forebyggende tiltak og livsstil også er behandlinger. b. Er noe av denne informasjonen forskningsresultater? Helt uenig Nokså uenig Nokså enig Helt enig Ikke aktuelt Når alt kommer til alt er jeg selv ansvarlig for å ta hånd om min egen helse Det aller viktigste for min egen helse og funksjonsevne er at jeg tar aktiv del i behandlingen Jeg er sikker på at jeg kan gjøre det som er nødvendig for å forebygge eller redusere symptomer eller problemer som skyldes min helsetilstand Jeg vet hvordan de forskjellige medisinene jeg har fått foreskrevet skal virke Jeg vet når jeg trenger medisinsk hjelp for et helseproblem og når jeg kan ta hånd om det selv 20) Jeg har tillit til at jeg kan klare å oppsøke informasjon på egen hånd til bruk i helsespørsmål: 94 95 22) Fortsatt fra forrige side. Angi i hvor stor grad du er enig eller uenig med hvert utsagn. Helt uenig Nokså uenig Nokså enig Helt enig Ikke aktuelt Jeg er trygg nok til å kunne ta opp det jeg ønsker, selv om helsepersonell ikke spør Jeg er sikker på at jeg kan gjennomføre den foreskrevne medisinske behandlingen hjemme Jeg forstår både hva helseproblemene mine dreier seg om og årsaken til dem Jeg vet om de ulike behandlingsmuligheter for min helsetilstand Jeg har opprettholdt de endringer i livsstil som jeg har gjort for helsens skyld Jeg vet hvordan jeg skal forebygge forverring av min helsetilstand Jeg kan finne løsninger når det oppstår nye situasjoner eller problemer med min helsetilstand Jeg kan opprettholde endringer i livsstil, for eksempel kosthold og trening, også i perioder med stress Til slutt vil vi spørre deg noen spørsmål om hvordan du opplevde nettportalen Sunn skepsis. 23) Angi hvor nyttig du opplevde Sunn skepsis: Svært unyttig Unyttig Noe unyttig Verken eller Noe nyttig Nyttig Svært nyttig 95 22) Fortsatt fra forrige side. Angi i hvor stor grad du er enig eller uenig med hvert utsagn. b. Er noe av denne informasjonen forskningsresultater? Helt uenig Nokså uenig Nokså enig Helt enig Ikke aktuelt Jeg er trygg nok til å kunne ta opp det jeg ønsker, selv om helsepersonell ikke spør Jeg er sikker på at jeg kan gjennomføre den foreskrevne medisinske behandlingen hjemme Jeg forstår både hva helseproblemene mine dreier seg om og årsaken til dem Jeg vet om de ulike behandlingsmuligheter for min helsetilstand Jeg har opprettholdt de endringer i livsstil som jeg har gjort for helsens skyld Jeg vet hvordan jeg skal forebygge forverring av min helsetilstand Jeg kan finne løsninger når det oppstår nye situasjoner eller problemer med min helsetilstand Jeg kan opprettholde endringer i livsstil, for eksempel kosthold og trening, også i perioder med stress Til slutt vil vi spørre deg noen spørsmål om hvordan du opplevde nettportalen Sunn skepsis. 23) Angi hvor nyttig du opplevde Sunn skepsis: Svært unyttig Unyttig Noe unyttig Verken eller Noe nyttig Nyttig Svært nyttig 95 24) Angi hvor brukervennlig du opplevde Sunn skepsis: Svært lite brukervennlig Noe lite brukervennlig Verken eller Noe brukervennlig Brukervennlig Svært brukervennlig 25) Angi hvor troverdig du opplevde Sunn skepsis: Svært lite troverdig Noe lite troverdig Verken eller Noe troverdig Troverdig Svært troverdig 2 to 20= TPB, 21 to 22=PAM and 23 to 25= Honeycomb 24) Angi hvor brukervennlig du opplevde Sunn skepsis: * 2 to 20= TPB, 21 to 22=PAM and 23 to 25= Honeycomb 96 Appendix 3. Summary of the ratings in usability test * Summary of the ratings per usability category across interviews based on the Honeycomb model by Moreville (38), courtesy of Danielsen (78). + = Positive feedback - = Negative feedback Usability Focus group 1 Focus group 2 Focus group 3 Focus group 4 Focus group 5 1. Usable + + - + - + - + - 2. Useful + + - + + + 3. Valuable + + - + + + 4. Accessible + - + - + - + - + 5. Desirable + + + + + 6. Credible + + + + + Appendix 3. Summary of the ratings in usability test * Summary of the ratings per usability category across interviews based on the Honeycomb model by Moreville (38), courtesy of Danielsen (78). + = Positive feedback - = Negative feedback Usability Focus group 1 Focus group 2 Focus group 3 Focus group 4 Focus group 5 1. b. Er noe av denne informasjonen forskningsresultater? Usable + + - + - + - + - 2. Useful + + - + + + 3. Valuable + + - + + + 4. Accessible + - + - + - + - + 5. Desirable + + + + + 6. Credible + + + + + Appendix 3. Summary of the ratings in usability test * Summary of the ratings per usability category across interviews based on the Honeycomb model by Moreville (38), courtesy of Danielsen (78). + = Positive feedback - = Negative feedback 97 Appendix 4. Summary of changes after usability test Problemområde Forslag løsning Utført Forsiden Titler endres i knapper foran Utføres Ja Logo må på forsiden Logo legges til Ja Usikkerhet rundt hva som ‘samler’ siden, hva er helheten som samler delene? Utvikle et samlende konsept for forsiden. Ja Tag-cloud oppleves forvirrende Gjøre om til lenkesamling til utvalgte artikler i stedet for å lenke til indexside Ja Kritisk vurdering Discern for lang Vurdere å bruke kortversjonen Ja Skåringsverktøy for Discern Legges til som tilleggsfunksjon Ja Søk etter forskning Vanskelig å navigere søkesidene Omorganisere søk etter forskning Ja For mange steder å søke i presise søk Vurdere om vi skal kutte noen Ja Lenkene til ‘hurtigsøk’ og ‘presise søk’ gjøres penere W3spor fikser Ja Finner ikke forskning (det er ikke alt det er forskning på) Legge til en tekst: finner du ikke det du leter etter? …. Ja Finner ikke ut av hvor man kan lese om sykdom Legge til at man får info om dette under omtaler Ja Sjekkliste til konsultasjonen Språk i lista er litt vanskelig Skrive om samt kutte ned på noen spørsmål Ja Generelt Appendix 4. Summary of changes after usability test 98 99 Kommentar om at ‘les mer’ lenker ikke gir nok informasjon til blinde Disse lenkene har en annen tittel allerede som de blinde leser Ja Hvorfor kan man ikke bare søke i google- hva bidrar sunn skepsis med? b. Er noe av denne informasjonen forskningsresultater? Legge inn tekst på dette Ja Kommentar om at man burde kunne søke i alle databaser samtidig gjennom SS Det er ikke gjennomførbart teknisk ei heller ønskelig pga av hensikten med å øke brukernes egne kunnskaper og ferdigheter innen søk og forskning Nei Tale i film må legges til som tekst Utføres Påventes Grønne epler i småbokser blir oppfattet som lenker Epler fjernes Ja Småbokser vanskelig å få øye på Skifte farge på småboksene Ja Menypunkter annen farge enn resten av designet Skifte farge slik at det passer med resten av fargeskjemaet Ja Søke i sunn skepsis forvirrende Designbyrået legger til en beskrivende tekst. I tillegg vil søkeboksene i ‘presise’ søk bli lagt til Ja Det må være mulig å printe sjekklistene Printikon Ja Uthevet fet tekst for å øke synlighet av stikkord Utføres Ja Andre revisjoner etter brukertest HON-code verifisering Utføres Ja Lenker til databaser gjøres om til knapper Utføres Ja Om risiko Utføres Ja Om validitet, kritisk vurdering + lenke til kbp.no Utføres Ja Legge inn caser under alle tre hovedkapitler Utføres Ja Bi-effekt versus feilbehandling Droppe i denne omgang- vil være relevant ved seinere utvidelse av portal Nei Skille på nivåer i diagnostikkteksten Utføres Ja Legge til at discernkalkulatoren også kan brukes for de som utvikler Utføres Ja 99 pasientinformasjon Ingen forskning er ikke det samme som ingen effekt (eller bi-effekt) Utføres Ja Legge inn lenke på skolemedisin/ alternativmedisin til søkesiden Utføres Ja Legge inn tekst med omtaler, oversikter og enkeltstudier Utføres Ja Se over språk ny side om beskrivelser av kilder Utføres Ja Endre tittel på filmer Utføres Ja Bilde på side om forskning Utføres Ja Legg inn snutt om forskningsetikk + deltakelse i forskning under om forskning Utføres Ja 100 Appendix 5. b. Er noe av denne informasjonen forskningsresultater? Results of TPB utility testing across the samples and studies Questionnaire sample 1: Mixed population (n=30) Questionnaire sample 2: Parents (n=45) Randomised controlled trial sample: Parents (n=66) Correlation coefficients (Pearson R) Correlation coefficients (Pearson R) Correlation coefficients (Pearson R) Direct measures Direct measures Direct measures Intentions ATT SN PBC Intentions ATT SN PBC Intentions ATT SN PBC Direct measures Attitudes (ATT) 0,58** - 0,58** - 0.56** - Subjective Norms (SN) 0,24 0,49** - -0,16 0,22 - -0.01 0.18 - Perceived Behavioural Control (PBC) 0,70** 0,55** 0,28 - 0,44** 0,47** 0,07 - 0.41** 0.59** 0.69 - * Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significant at the 0.01 level (2- tailed) ppendix 5. Results of TPB utility testing across the samples and studies * Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significant at the 0.01 level (2- tailed) 101 102 Papers I-V 103 II III Evaluation of a web portal for improving public access to evidence- based health information and health literacy skills: a pragmatic trial       ! "        ! "  # $ % &'  ()) !**+* & (" !**#  / 0& #/ 01 0 '( " ! *# / 0&# #5  '( '  *; " # / 0&# # $ % &'  ()) !**+* & (" !**#  / 0& # $ % &'  ()) !**+* & (" !**#  / 0& #/ 01 0 '( " ! *# / 0&# #/ 01 0 '( " ! *# / 0&# #5  '( '  *; " # / 0&# #<  (" ! *&% & ( # / 0&# ' * = # <(>/'=?@@DEF 5 ><G) **  * H ( G   *&$  5 ><G) **  * H ( G   *&$   1 Abstract Background:%* *)  (G0 $ (* G$ * H )**  0H )   0   )       *    )*(*      G)  &) H *J  *&$ # Objective:=   ((* (0H)  * G)      ( # Methods:  )G*  G *   )      G)   G  ( E@)0 * L &#M0 * * **  )      $ )( G  0$) #= *   * $  **  ))  $  )   )*)  H  )*) #0* *( G;*    ?# Results:  % (  0H )   0 (    G)     0 * (    ( G #= H 0(G G) )*  (  * H  G) #  M*) *   0H )         H & (   *H &# =     )  0   *    (  * H ( G  **  ))  $   )*)  * G)     ) *      H ((*0 * &(*# Conclusion:)(* &0 ) 00( 0H)   G& )  ((*   0*(  ( G #+ G  )*) *   0H )     H      # <  G)    *   Q)G 0 0H H  *     *  )*)     * *  G$# 2 Introduction Introduction =*  G ( * (( R J  )* (         G&   **  * G)  (  H      * VW*   * H)**VW#" 0 ((* )*) ) ** * H( G $ 0 H    H      * )   ( G    H  VW# ! Introduction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functional  fundamental literacy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how *  *  *  why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arriers identified in pre-studies and literature search Facilitators/content of intervention Health literacy domains* Evaluated in pragmatic trial All Shared decision making (promoting an active role) and evidence based practice as conceptual framework (promoting evidence based decisions) Civic literacy (system and relationships) TPB **(attitude and subjective norms associated with search)/ PAM*** Science literacy Inability to understand and critically appraise health information Improving critical appraisal skills Introduction to scientific concepts and (checklist for) evaluating trustworthiness of health information Science literacy Examples: Validity, uncertainty, causality Searching task/ critical appraisal task/ TPB (perceived behavioural control and attitudes towards search) Functional literacy (numeracy) Example: Understanding risk Civic literacy (media literacy) Examples: How research and scientific discourse are presented in the media Not knowing where to find reliable and relevant information Improved access to reliable research based sources of health information Introduction to searching for evidence based information (adapted EBP-model) Science literacy Examples: Basic study designs and assessment of relevance Searching task/ TPB (perceived behavioural control and attitudes towards search) Civic literacy (media literacy) Examples: Search strategies, publication types and sources Inability to exchange information in consultations Enabling exchange of health information Introduction to clinical decision making and checklist for the consultation Science literacy PAM Civic literacy (system and relationships) Cultural literacy (understanding of concepts used in decision making about health care) `"  *& GH G  H&\* *  VW``= & ( ) H  ```M*  G  Table 1. Participants and recruitment p    0 * &)*  ( 0H)  )*)G) #<  00 *) HH &)  0 H H&)*)  * 0   ) (    #M0*   ( & (0( # ()&)* & G& ]  H )  ) H G#=&   * 0  GH (  )&H ( G G GH ( )& ( #@( * G GH (  (V?W#! *)  * & G  $ & *(  ( G   VW# Design = &0)G*)   0*  )* ** )       )0   )*    *    )# Introduction Overview of the intervention components, corresponding hypothesised health literacy domains targeted based and measurements to evaluate these Table 1. Overview of the intervention components, corresponding hypothesised health literacy domains targeted based and measurements to evaluate these Table 1. Overview of the intervention components, corresponding hypothesised health literacy domains targeted based and measurements to evaluate these 7 Sources and methods of recruitment  <( G H  &0H G&*  *    G  * G0 $ < *_  *  (  G ( )#= 0 00 )*)0* * G 0H)# Sample size justification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nformed consent and inclusion of participants  M ) 0 Q))*)* ( G H  & 0$  0*  )*) *   ]  (  *  **#M*)0Q* (&    0  &)*) &  )*)0H  ) * )  G) * G (&  * L &#<()*)  G*  * 0  G(H*$*H  0&&0 H  )*)# 9 9 Figure 1. Consort 2010 flow diagram Randomisation methods and allocation concealment  = 0 G*  *0 G G)  G  ) *   )H&!M!!#= &0 H )*)0  Figure 1. Consort 2010 flow diagram Randomisation methods and allocation concealment Randomisation methods and allocation concealment  = 0 G*  *0 G G)  G  ) *   )H&!M!!#= &0 H )*)0  10 ( GH  & ) 0*&0 *# )*)0 & 0 H)*)00H)   *H &0 *   ** )  ((G# )*)0 G( G  ]  &   (  G  &GG # Intervention delivery   ( G 0    0* *   ]   &G#M*)0$H&G + ((G#=    )0 0** 0H)  GG &( G   & Q) *  H( * ($#[* ($ * )  0H)  JG* * G &>G)   ( *H( G G) ** ))  ( ( G G)  H (H )*) **  #=$0( 0> 1. Searching task: Using research-based information 1. Searching task: Using research-based information = ) )  (  ( $ 0          0* )*)  *H *  H *( G #M*)0$ ( G  ]   0H&*(   ( G #=   )0  $)*(* &  0H)  ( $0 ( *( ( G  & *&( 0 H ( * 0H)  # 11 11 Figure 2. Overview of the study design Figure 2. Overview of the study design 2. Critical appraisal task 2. Critical appraisal task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eliefs about search for health information and activation Participant retention *     H*$  ( G   G   0 $)  (     0( *)  & (  ( 0 )#  G*G)$0  )*)(&(  ) 0Q&# 3. Beliefs about search for health information and activation  =$0 Q) )  ((*H0   *   )G (H (H  ( G *  *  Q)* )* H   *    *#<    Q)  ((*      (*  #<( $   )0    (   ]      (* 00H)  #  [* ($0 )*)0$ &  #=G   0* ( 0  >( &)*)0) 0 ((*G  * G) $ 0)*#!*  & )     H0$ * ) H & )  & * ] * )   )*) ( # ( &( H* **   ) 13 0   ** 0H)  #=G (G( * * 0) )*( H;*    ?# Missing data  / G)0G G) G#  &0)( G   H  *H H )*)  (  <==## )*)0 &  ) 0*&0 G &# The searching task: using research-based information The searching task: using research-based information The searching task: using research-based information = * G0  H&* **<G (H& &) $( G   HH *#=( G  0*  HR* HJ( $( G () H   *## )G& &  GG*HH Q) *&G** ##&G* 0 *  )) #=( G  0 Q* ( (* 0)  ((0 Q) *&G**   0 0& & *  (*  H * # = G  ( H& &) $ 0 *  H& 0  )  H  * 0    (  (  * H)** *&#=* *  0* (  0 *#<0Q)**H( G 0 H( H&G  )*)       )   *    )# = &)  0  H& * *   $* ) * (*  # Outcome assessment and analysis  +0   H (   G) G  *&$ G G)  &  G*&$  0   &( *   *&VEW#    0G  &   **  * $ 0(     H  G H  *   #= * G ( &0 (     *  ( G0*0* GH0*  ]    G  (G)   *&$ H&   # =G * G* )  G*H(  H 0   H * H *  H  0 H   H (# 0 14   G* &Q)  !M!!   * * ) G#   G* &Q)  !M!!   * * ) G# Critical appraisal of health information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eliefs about intention to search for health information and activation = * G 0 G   0   G#  ]   0  )*(* &( ) )  (  H (H *(  ( G H  =MVW#=M*  ; M;0 0)G ( G< "      *  )*)J ( GGH VZW# Measuring beliefs about search for health information = =M    *  *   G      &   0 &    )* H      VZZZW#**   G   *   H  *)*H    + VZZZW#=> #  0H  H (H * ] * (H  )     GH * ] * # ! Critical appraisal of health information H*   G H   H    ) J 0 )*)  (  *  )   16 16 Q)* G    * G) & Q)* G    * G) & #M* H   *   0G *) *   H   0 * () ( H HH  )( G  )( GH   Figure 3. Model of the Theory of planned behaviour by Ajzen 1991 [55] Figure 3. Model of the Theory of planned behaviour by Ajzen 1991 [55] 17 =  H   G   *  * G  VZW# X   * G  *H    H (  * 0 H         & (   *(  ( G *G ))*(*H (   &    G  0   Q)  & (* # =      ( * (       *  * =M * G)  0 * *   ((*G  * )*(*H ( 0  #=GG G GQG GG* G) * ( *G 0 D  j( )*(** G  @ @(   * G) ** (  H*  G)*  17 H   *   F jF(  0  *G ( H      *  )    )*  H    *  # =     )0Q)*   H (*   )# =  Q)   ((* (      )*     *  )*&H0R J* G) *G ( G=M]   0     G )    G    0* R J 0  ) H #_ )G0H)   #   * G) * ( G =MG  G &R  0 H  JR! Critical appraisal of health information H*   GJ RM*  H   *  J0) H #< 0G((* (   )*&H0* G) *  H& * GH0 )G* (* G) *   ) H    G  # + G   =M G    * ]  &  ]   *  (    )    (  *  *  * 0 *  G) H   )H0+ #=  )0Q)  H& * G)  G)  H  *     M J # = * =M G   &)    H )   & *   0   H   * G  G&    H   * ) (* VZZW# Measuring overall activation Satisfaction with the web portal Satisfaction with the web portal f p X H (* (H*$( G   )H 0H)  #      (0H " &* GHG  VZFW (  G))   G G (< Q)*VZEW#=G  * G)  G  00H**H  H *H   H (H H  (  VZFW# <   & ]   0      (  G  (  (*  *H & (   H & * G    (*  *  ( G  D    *  (* # = ( 0  0   GG (  *  G>G    G]  # Measuring overall activation 18   *  0G  M*  ; M;    G)) *H  H ) &) ) VZZ@W#= G* (  $& GH  ) G)  * (*$ 0 *&   $ *  *  & $ *    G)  J    &  *   0*G H&   (GVZ@ZDW#  *  *   ( G ? ?? 0 ?? *  *    ? *   *   VZ@ZDW# X *)*        )0 H*      ( *   )#=    * G0  0  G  ((*G  * # Safety monitoring and adverse events f y g   * )*) * *)  & *&)   # = G0*)  =M]   H*  G# ((* 0G  $ &# 19 Ethical aspects 0 &G  &* ))  0H&/ 0! *  !**! */!5  ' GG( ;* " 5*[* 5[1#  =   0    ' * = #  (*   GH /'=?@@DEF# Description of study participants + )*)0Q* ( G &H* & G*   *  ( * L &#) ) * G) * * # <  E@)*)0*  (0*D0 G    ) E *   )#=  ) (    )*    )0@?kxFF?kxE)*  &=H #= 0 *  * G)  ($ &*    0  &# Intervention Control Response rates total 60 % (n=28) 80 % (n=39) % Men 20 % 22 % % Females 80 % 78 % % Primary school 9 % 0 % % High school 16 % 12 % % 1-3 years of college/ University education 22 % 22 % % 3+ years of college/ University education 53 % 66 % Table 2. Description of participant characteristics 20 Improved use of research =0  * H  * 0 (       )     *   )#=  $0#F'<?# E#)x?#E# Critical appraisal =G (( G 0#!?#F?H&   )# !#?H&*  #=G((*0 ?#? )x?#E? # =((*H0Q) (   )0 #?( *   )0#((*x ?#? ^)x?#E?  =H   (* 0  )0 (* &((_)^ M ' !] x#@?Z)x?#F =H   (* 0  )0 (* &((_)^ M ' !] x#@?Z)x?#F Graph 1: Rating of quality across study groups Graph 1: Rating of quality across study groups Graph 1: Rating of quality across study groups 21 Beliefs about searches for health information and activation Beliefs about searches for health information and activation * &(*((* (?#@0( (     0*> ?#@ )x?#?   (  (      )  =H   (  G * # = G ((* (      H*   G  ?#^ )x?#E  )*  H    *   ?#^)x?#Z0 )*(*H ( )  z?#Z  *0  * &(*# =)*&H0** G) G ( G=M   * 0 (      (( (* & *    ) )z?# (     * G)  * #==M*  *J  )*  (  ** * G) G)  0)) QG &Dk ( H       *#  0 G G) )  )*  ( Hx?#Z^)L?#??0)*  (  H*   G  )*  H   *   0 ?#Z )x?#Z  ?#?@ )x?#D )*  &# * )* H   *  * (*)   *   0 M Jx?#Z@?#)*  &0  (  H  *   M J x?#ZE   )L?#??# ! Beliefs about searches for health information and activation H*   G     * & (*   *    0   M J x ?#^ )x?#E    * & (*)  *   0 M Jx?#F^)x?#Z)* H    *  M Jx?#@E^)x?#ZF# =  *  M;* 0@@#Z   )@#E *   )G((*0#@)x?#?# 22 Variable Mean Intervention (SD) Mean Control (SD) Mean difference (95 % CI) P- value Intention* 6.1 (1.1) 5.8 (1.1) 0.3 (-0.2 to 0.9) 0.20 Direct attitude* 5.8 (1.1) 5.2 (1.2) 0.6 (0.1 to1.2) 0.03 Direct subjective norm* 3.4 (1.3) 3.6 (1.2) -0.2 (-0.8 to 0.4) 0.49 Direct perceived behavioural control* 5.6 (1.1) 5.3 (1.1) 0.4 (-0.2 to 1.0) 0.15 Overall indirect attitude** 53.7 (24.2) 50.8 (24.8) 2.9 (-9.3 to 15) 0.64 #1 Provides insight*** 14.3 (6.7) 14.0 (6.5) 0.3 (-2.9 to 3.6) 0.83 #2 Useful as background in consultations*** 12.8 (7.8) 11.4 (7.7) 1.4 (-2.5 to 5.3) 0.48 #3 Helpful if unsure in health decision*** 12.4 (6.8) 12.4 (7.0) 0 (-3.4 to 3.4) 0.99 #4 Provides additional information If incomplete information from health*** 14.1 (6.8) 13.0 (7.4) 1.1 (-2.4 to 4.7) 0.52 Overall indirect subjective norm** 13.9 (14.5) 10.3 (12.5) 3.6 (-3.0 to 10.3) 0.28 #1 Family and friends*** 8.1 (6.5) 6.1 (4.9) 2.0 (-0.8 to 4.8) 0.15 # 2 Health professionals*** 0.7 (6.9) 1.0 (7.4) -0.3 (-3.9 to 3.3) 0.89 #3 Other social groups (colleagues, patient organisations)*** 5.1 (5.1) 3.2 (3.5) 1.8 (-0.3 to 4.0) 0.09 Overall indirect perceived behavioural control** -4.4 (13.4) -3.1 (17.3) -1.3 (-9.1 to 6.6) 0.74 #1 Difficult to attain an overview*** -0.3 (6.5) -0.3 (7.2) 0 (-3.4 to 3.5) 0.99 #2 Not possessing knowledge*** -0.8 (5.4) -1.2 (6.6) 0.4 (-2.6 to 3.4) 0.79 #3 Time consuming*** -3.3 (5.9) -1.6 (7.0) -1.7 (-5.0 to 1.5) 0.30 *Mean minimum and maximum score possible is 1 to 7 (stronger beliefs indicated by higher) **Mean minimum and maximum score possible is -63 to 63 for subjective norm and perceived behavioural control, and -84 to +84 for attitude. ***Mean minimum and maximum score possible is -21 to 21. Table 3. Distribution of means and differences between groups *Mean minimum and maximum score possible is 1 to 7 (stronger beliefs indicated by higher) **Mean minimum and maximum score possible is -63 to 63 for subjective norm and perceived behavioural control, and -84 to +84 for attitude. Table 3. Distribution of means and differences between groups Beliefs about searches for health information and activation ***Mean minimum and maximum score possible is -21 to 21. Table 3. Distribution of means and differences between groups 23 Satisfaction with the web portal Satisfaction with the web portal =G (  (0H)  0#D!#G H &# !?#EDG*H &#DZ!?#E=H # Usefullness Usability Credibility Mean 4.71 4.14 4.75 Median 5.00 4.00 5.00 Standard deviation 1.117 .970 .928 Percentiles 25 4.00 3.00 4.00 50 5.00 4.00 5.00 75 6.00 5.00 5.00 *Mean minimum and maximum score possible is 1 to7 (stronger satisfaction indicated by 7) Table 4. Satisfaction with the web-portal Study limitations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ntervention effects on critical skills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eliefs about participation and intervention effects  ( H  )*)  0      =M ]         *   & (*    M; ]   *H   *  #=  * ( H (H )*) H H =MM; 0H  )*)*) 0$G *  ( H &# X* G)0  )0( G * &  (*G) G (  (   )0H  G*  * *  # <=M]  H  )0  (    ( H    0*(  ( G #XH  ))   )*  H   *  )  )*(*H ( (**)* ( *0)* H  &((* #=   * )  *0 G # ( )*   0  (*  ) Q)* &   )   Q)* ( G ) (  0 G H*)  *   # (G   ( G   0)*(   ( G  &* (*    **    V@?@@@W# 27 = G ((* ( * G  *    )  0  * & = G ((* ( * G  *    )  0  * & 27 = G ((* ( * G  *    )  0  * & 27 27 (*((*(  (   )G (   ?#@)x?#? # = 0H)  G&H (  ( G)   0* (    ( G      (  0 (  (*    0    (  HG G) )*  (  *H  )#=0(  ( H   HG) (*  *0*(  ( G   )) H&  V@?@W*   &0*G]   0 VW#M* H   *  0  &0$ * &(* )*  (  *H 0(    *   0H    #=    (  (H*   (*  * &)*  G& HG)   & H #! Beliefs about participation and intervention effects H*  G0  (    0$  )0 ( ))  * 0  * Q)*  )   )) HG)  )  )*  (   *VW# X( * &(* ((*H0   ) G (  ((* (   )*  (  *#==M* G) J   )*  (  *0(* & &* 0   0*=MH  )*H    *  (  )* VZW# Satisfaction with the portal =(* 00H)  0 #=*  ) )  (0H)  * *)  *0)  &   G ) # 28 * ) *0  00 G)  **  * H ( G 0  0 G) $ 0 $ #X  )0H)   * G) G 0 (  * H0 H &  0 &     **  H    ( G  *   ( *( QG) H& 0) )    0H)#' ]  &0H)  G&)  &GG  (        * ( ( G #)  ( &* 0G& * G 0& 0* H ** * &# <Q) ( *0H)  0 HG) H * &  H 0  H * )G# = 0  G  0 ) )  * (    ( G     *0H)  0    (0&0  * H  G$ * H ( G G   H   ) H *#= 0H)    *    )GG ( 0 H* ( *  # <( Q) * *  (  GH ***H  *    G*   ( G ) H *H# Implications for practice <   )*** G$ HH H  H  *         *# 5*      ) &  *      * H )** ) )    H *    $   *   # ;    * *0H)   HG  H   (* **  **   (* H( G # +  ((    G  Q)GG 0 0H H *    *   G *#{G)  (&G  H   #+ G    * G& H ((* G)   *&$  ((*  &# ;       *  *    (  0H )     *   0 )  ) ( * H)*** (   * )) 0  )*  * # Conclusion =  & 0  )G*   *  *     ( # = * G ( )*)0* ) 0 G0 0  ((* G   Q)*#          & H ) 0 0 (  G) G   0*(  ( G  H ( G G)  )*  (   *# =  * (0H )      0 * (G H&  (*   )*) *   0H )         H & 29 ( *H &# Acknowledgements g X(  &QG(  * )) )  ( &# X0    $ $' _  { ;5 (1  15( ) Q)Q  * G  & ( # References Coulter A (2002) The Autonomous patient: ending paternalism in medical care. London: TSO. 121 2. Nutbeam D (2000) Health literacy as a public health goal: A challenge for contemporary health education and communication strategies into the 21st century. Health Promotion International 15: 259-267. 3. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, et al. (2005) Sicily statement on evidence-based practice. BMC Medical Education. pp. 1. 3. Dawes M, Summerskill W, Glasziou P, Cartabellotta A, Martin J, et al. (2005) Sicily statement on evidence-based practice. BMC Medical Education. pp. 1. 4. Zarcadoolas C, Pleasant A, Greer D (2006) Advancing health literacy: a framework for understanding and action San Francisco CA: Jossey-Bass 368 p 4. Zarcadoolas C, Pleasant A, Greer D (2006) Advancing health literacy: a framework for understa and action. San Francisco, CA: Jossey-Bass. 368 p. 5. Nutbeam D (1998) Health promotion glossary. World Health Organization. pp. 36. 6. Freedman DA, Bess KD, Tucker HA, Boyd DL, Tuchman AM, et al. (2009) Public Health Literacy Defined. American Journal of Preventive Medicine 36: 446-451. 7. Peerson A, Saunders M (2009) Health literacy revisited: what do we mean and why does it matter? Health Promotion International 24: 285-296. 8. Berkman N, Sheridan S, Donahue K, Halpern D, Crotty K (2011) Low Health Literacy and Health Outcomes: An Updated Systematic Review. Annals of Internal Medicine 155: 97-101. 9. Glenton C, Paulsen E, Oxman A (2005) Portals to Wonderland? Health portals lead confusing information about the effects of health care. BMC Medical Informatics and Decision Making. 15 March ed. pp. 8. pp 10. Hoye S, Hjortdahl P (??) ["New wonder pill!"--what do Norwegian newspapers write]. Tidsskrift for Den norske Legeforening Jun 30;122(17): @D @D@. 11. Johansen L, Bjørndal A, Flottorp S, Grøtting T, Oxman A (1996) [Evaluation of health information in newspapers and brochures. What can we believe?]. Tidsskrift for Den norske Legeforening 116: @? @. 12. Andreassen H, Sandaune A, Gammon D, Hjortdahl P (2002) [Norwegian use of Internet health services]. Tidsskrift for Den norske Legeforening 122: 1640-1644. 13. Fox S (2006) Online health search. PEW Internet and American Life Project: PEW Internet and American Life Project. pp. 22. 14. Schwartz K, Roe T, Northrup J, Meza J, Seifeldin R, et al. (2006) Family medicine patients' use of the Internet for health information: a MetroNet study. Journal of the American Board of Family Medicine 19: 39-45. 15. Conflict of interests 30 Conflict of interests =) *0( H& $ % &'  ()) !** Conflict of interests =) *0( H& $ % &'  ()) !** 30 30 ( G & % &' #=  * &  * ( * (# ( G & % &' #=  * &  * ( * (# ( G & % &' #=  * &  * ( * (# References Fox S, Jones S (2009) The Social Life of Health Information. . PEW Internet and American Life Project. Washington. pp. 88. g pp 16. Sillence E, Briggs P, Harris PR, Fishwick L (2007) How do patients evaluate and make use of online health information? Social Science & Medicine 64: 1853-1862. 17. Robinson E, Kerr C, Stevens A, Lilford R, Braunholtz D, et al. (2005) Lay public's understanding of 16. Sillence E, Briggs P, Harris PR, Fishwick L (2007) How do patients evaluate and make use of online health information? Social Science & Medicine 64: 1853-1862. 17. Robinson E, Kerr C, Stevens A, Lilford R, Braunholtz D, et al. (2005) Lay public's understanding of equipoise and randomisation in randomised controlled trials. NHS R&D HTA Programme. 192 p. 18. Glenton C, Oxman A (1998) The use of evidence by health care user organizations. Health E i 1 14 22 17. Robinson E, Kerr C, Stevens A, Lilford R, Braunholtz D, et al. (2005) Lay public's understanding of equipoise and randomisation in randomised controlled trials. NHS R&D HTA Programme. 192 p. 18. Glenton C, Oxman A (1998) The use of evidence by health care user organizations. Health Expectations 1: 14-22. 19. Woloshin S, Schwartz L, Byram S, Sox H, Fischhoff B, et al. (2000) Women's understanding of the mammography screening debate. Archives of Internal Medicine 160: 1434-1440. 20. Weinstein N (1999) What does it mean to understand a risk? Evaluating risk comprehension. Journal of the National Cancer Institute: 15-20. 21. Lloyd A (2001) The extent of patients' understanding of the risk of treatments. Quality in Health Care 31 21. Lloyd A (2001) The extent of patients' understanding of the risk of treatments. Quality in Health Care 31 10 Suppl 1: i14-18. 22. Schwartz L, Woloshin S, Black W, Welch H (1997) The role of numeracy in understanding the benefit of screening mammography. Annals of Internal Medicine 127: 966-972. 23. Tickner S, Leman P, Woodcock A (2006) Factors underlying suboptimal childhood immunisation. Vaccine 24: 7030-7036. 24. Austvoll-Dahlgren A, Helseth S (2010) What informs parents' decision-making about childhood vaccinations? Journal of Advanced Nursing 66: 2421-2430. 25. Legare F, Ratte S, Gravel K, Graham I (2008) Barriers and facilitators to implementing shared decision- making in clinical practice: update of a systematic review of health professionals' perceptions. Patient Education and Counseling 73: 526-535. 26. References Bugge C, Entwistle V, Watt I (2006) The significance for decision-making of information that is not exchanged by patients and health professionals during consultations. Social Science & Medicine 63: 2065-2078. 27. Detmer D, Singleton P, MacLeod S, Wait S, Taylor M, et al. (2003) The informed patient: Study Report. Cambridge: University of Cambridge. 53 p. 28. Coulter A, Jenkinson C (2005) European patients' views on the responsiveness of health systems and healthcare providers. European Journal of Public Health 15: 355-360. 29. Nutbeam D (2008) The evolving concept of health literacy. Social Science and Medicine 67: 2072- 2078. 30. Coulter A, Ellins J (2006) Patient-focused interventions: a review of the evidence. London: Pic Institute Europe. pp. 277. 31. Wantland D, Portillo C, Holzemer W, Slaughter R, McGhee E (2004) The effectiveness of Web-based vs. non-Web-based interventions: a meta-analysis of behavioral change outcomes. Journal of Medical Internet Research 6: 40. 32. Murray E, Burns J, See T, Lai R, Nazareth I (2005) Interactive Health Communication Applications for people with chronic disease. Cochrane Database of Systematic Reviews: CD004274. 33. Austvoll-Dahlgren A, Helseth S (2011) Public health nurses’ barriers and facilitators to the use of research in consultations about childhood vaccinations. Scandinavian Journal of Caring Sciences. 33. Austvoll-Dahlgren A, Helseth S (2011) Public health nurses’ barriers and facilitators to the use of research in consultations about childhood vaccinations. Scandinavian Journal of Caring Sciences. 34. Austvoll-Dahlgren A, Sørum R, Helseth S (2012) Cognitive factors predicting intentions to search for 34. Austvoll-Dahlgren A, Sørum R, Helseth S (2012) Cognitive factors predicting intentions to search for health information: an application of the Theory of Planned Behaviour. 34. Austvoll ahlgren A, Sørum R, Helseth S ( 0 ) Cognitive factors predicting intentions to search for health information: an application of the Theory of Planned Behaviour. 35. Austvoll-Dahlgren A, Danielsen S, Opheim E, Bjørndal A, Reinar L, et al. (2012 ) A complex intervention: tailoring a web portal to improve user health literacy skills. g , , p , j , , ( ) p intervention: tailoring a web portal to improve user health literacy skills. 36. Bjørndal A, Flottorp S, Klovning A (2007) Knowledge management in medicine and health scienc [Kunnskapshåndtering i medisin og helsefag] Oslo: Gyldendal akademisk. 195 p. 37. DiCenso A, Bayley L, Haynes R (2009) Accessing preappraised evidence: fine-tuning the 5S mo a 6S model. Annals of Internal Medicine 151: 99-101. 38. References The British Medical J l 335 126 127 of American Medical Association 287: 3245-3248. 46. Goldacre B (2007) Medicine and the media: MMR: the scare stories are back. The British Medical Journal 335: 126-127. 46. Goldacre B (2007) Medicine and the media: MMR: the scare stories are back. The British Med Journal 335: 126-127. 47. Haug C (2009) [Fear and expectations]. Tidsskrift for Den norske Legeforening 129: 1619. 48. Teasdale E, Yardley L (2011) Understanding responses to government health recommendations: public perceptions of government advice for managing the H1N1 (swine flu) influenza pandemic. Patient Education and Counseling 85: 413-418. 49. Ishikawa H, Yano E (2008) Patient health literacy and participation in the health-care process. Health Expectations 11: 113-122. 50. Eysenbach G, Powell J, Kuss O, Eun-Ryoung S (2002) Empirical studies assessing the quality of health information for consumers on the world wide web. Journal of American Medical Association 287: 2691-2700. 51. Steinsbekk A (2008) [Patient Activation Measure]. Tidsskrift for den Norske Legeforening 128: 2316- 2318. 52. Francis J, Eccles M, Johnston M, Walker A, Grimshaw J, et al. (2004) Constructing questionnaires based on the theory of planned behaviour: A manual for health services researchers. . UK: Centre for Health Services Research, University of Newcastle. pp. 42. 53. Armitage C, Conner M (2001) Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology 40: 471-499. 54. Eccles M, Hrisos S, Francis J, Kaner E, Dickinson H, et al. (2006) Do self- reported intentions predict clinicians' behaviour: a systematic review. Implementation Science. 55. Ajzen I (1991) The Theory of Planned Behavior. Organizational Behavior and Human Decision Processes 50: 179-211. 56. Hibbard JH, Mahoney ER, Stockard J, Tusler M (2005) Development and testing of a short form of the patient activation measure. Health Services Research 40: 1918-1930. 57. Hibbard JH, Stockard J, Mahoney ER, Tusler M (2004) Development of the patient activation measure (PAM): Conceptualizing and measuring activation in patients and consumers. Health Services Research 39: 1005-1026. 58. Morville P (2004 ) User Experience Design; honeycomb model. 58. Morville P (2004 ) User Experience Design; honeycomb model. 59. Rosenbaum S, Glenton C, Cracknell J (2008) User experiences of evidence-based online resources for health professionals: User testing of The Cochrane Library. BMC Medical Informatics and Decision Making. 60. Wainstein B, Sterling-Levis K, Baker S, Taitz J, Brydon M (2006) Use of the Internet by parents of paediatric patients. References Charnock D, Shepperd S, Needham G, Gann R (1999) DISCERN: an instrument for judging the quality of written consumer health information on treatment choices. Journal of Epidemiology and Community Health 53: 105-111. 38. Charnock D, Shepperd S, Needham G, Gann R (1999) DISCERN: an instrument for judging the quality of written consumer health information on treatment choices. Journal of Epidemiology and Community Health 53: 105-111. 39. Irwig L, Irwig J, Trevena L, Sweet M (2008) Smart health choices: making sense of health advice. London: Hammersmith. xi, 242 p. 39. Irwig L, Irwig J, Trevena L, Sweet M (2008) Smart health choices: making sense of health advice. London: Hammersmith. xi, 242 p. p 40. Statistics Norway (2009) [Use of health services by sex and age]. Statistics Norway,: SSB. 41. Wangberg S, Andreassen H, Kummervold P, Wynn R, Sorensen T (2009) Use of the internet for health purposes: trends in Norway 2000-2010. Scandinavian Journal of Caring Sciences 23: 691-696. d h ( ) h d d d d h l h 41. Wangberg S, Andreassen H, Kummervold P, Wynn R, Sorensen T (2009) Use of the internet for health purposes: trends in Norway 2000-2010. Scandinavian Journal of Caring Sciences 23: 691-696. 42 R d i h D (2006) D i i h d I K R di U d di h l h 41. Wangberg S, Andreassen H, Kummervold P, Wynn R, Sorensen T (2009) Use of the internet for health purposes: trends in Norway 2000-2010. Scandinavian Journal of Caring Sciences 23: 691-696. 42. Radosevich D (2006) Designing an outcomes research study. In: Kane R, editor. Understanding health care outcomes research. 2nd ed. Sudbury, Mass.: Jones and Bartlett. pp. 23-58. 43. Charnock D (1998) The DISCERN handbook: Quality criteria for consumer health information on 32 treatment choices. Oxon: Radcliffe Medical Press Ltd. 44. Mills E, Jadad AR, Ross C, Wilson K (2005) Systematic review of qualitative studies exploring parental beliefs and attitudes toward childhood vaccination identifies common barriers to vaccination. Journal of Clinical Epidemiology 58: 1081-1088. olfe R, Sharp L, Lipsky M (2002) Content and design attributes of antivaccination web sites. Journ of American Medical Association 287: 3245-3248. 45. Wolfe R, Sharp L, Lipsky M (2002) Content and design attributes of antivaccination web sites. Journal of American Medical Association 287: 3245-3248. 46. Goldacre B (2007) Medicine and the media: MMR: the scare stories are back. impact on patients with melanoma. Journal of the American Academy of Dermatology 52: 779- 785. References Journal of Paediatrics and Child Health 42: 528-532. 61. Bylund C, Gueguen J, Sabee C, Imes R, Li Y, et al. (2007) Provider-patient dialogue about internet health information: An exploration of strategies to improve the provider-patient relationship. Patient Education and Counseling 66: 346-352. 61. Bylund C, Gueguen J, Sabee C, Imes R, Li Y, et al. (2007) Provider-patient dialogue about internet health information: An exploration of strategies to improve the provider-patient relationship. Patient Education and Counseling 66: 346-352. 62. Tuffrey C, Finlay F (2002) Use of the internet by parents of paediatric outpatients. Archives of Disease in Childhood 87: 534-536. 62. Tuffrey C, Finlay F (2002) Use of the internet by parents of paediatric outpatients. Archives of Disease in Childhood 87: 534-536. 63. Gordon MM, Capell HA, Madhok R (2002) The use of the Internet as a resource for health information among patients attending a rheumatology clinic. Rheumatology 41: 1402-1405. 63. Gordon MM, Capell HA, Madhok R (2002) The use of the Internet as a resource for health information among patients attending a rheumatology clinic Rheumatology 41: 1402 1405 information among patients attending a rheumatology clinic. Rheumatology 41: 1402-1405. 64. Sabel M, Strecher V, Schwartz J, Wang T, Karimipour D, et al. (2005) Patterns of Internet use and g p g gy gy 64. Sabel M, Strecher V, Schwartz J, Wang T, Karimipour D, et al. (2005) Patterns of Internet use and 64. Sabel M, Strecher V, Schwartz J, Wang T, Karimipour D, et al. (2005) Patterns of Internet use and 33 34 Errata Feil i referanseliste ble rettet i artikkel V. Samme referanse var oppført to ganger i referanselisten, og er nå erstattet med referanse 34. Feil i referanseliste ble rettet i artikkel V. Samme referanse var oppført to ganger i referanselisten, og er nå erstattet med referanse 34.
https://openalex.org/W4385570887
https://aclanthology.org/2023.findings-acl.492.pdf
English
null
Scaling Laws for BERT in Low-Resource Settings
null
2,023
cc-by
12,247
Abstract et al., 2022a) and the size of the corpora on which the pre-training is performed increase (Zhang et al., 2021). For this reason, there is now a tendency to build ever-larger models trained on ever-growing corpora. This trend has resulted in a never-ending increase of the computational requirements to per- form model pre-training, but also for the subse- quent fine-tuning and inference processes at pro- duction time. Moreover, building very large models require huge training corpora, which is only avail- able for a handful of rich-resource languages. Large language models are very resource in- tensive, both financially and environmentally, and require an amount of training data which is simply unobtainable for the majority of NLP practitioners. Previous work has researched the scaling laws of such models, but optimal ratios of model parameters, dataset size, and com- putation costs focused on the large scale. In contrast, we analyze the effect those variables have on the performance of language mod- els in constrained settings, by building three lightweight BERT models (16M/51M/124M parameters) trained over a set of small corpora (5M/25M/125M words). We experiment on four languages of different linguistic character- istics (Basque, Spanish, Swahili and Finnish), and evaluate the models on MLM and sev- eral NLU tasks. We conclude that the power laws for parameters, data and compute for low- resource settings differ from the optimal scaling laws previously inferred, and data requirements should be higher. Our insights are consistent across all the languages we study, as well as across the MLM and downstream tasks. Fur- thermore, we experimentally establish when the cost of using a Transformer-based approach is worth taking, instead of favouring other com- putationally lighter solutions. Kaplan et al. (2020) and Hoffmann et al. (2022) propose power-law formulas that relate model size, corpora size and computation power, and help find the optimal settings in advance given a fixed bud- get. However, their analysis is focused on autore- gressive models of relatively big sizes, that require large corpora to train. In this paper, we analyze whether the conclusions drawn in these works also apply to discriminative (encoder-only) models in low-resource settings, where both the data size and budget are constrained. We analyze the perfor- mance of several combinations of model and data sizes using a simulated low-resource scenario in four linguistically diverse languages from different families (Basque, Spanish, Swahili and Finnish). Rodrigo Agerri2 Aitor Soroa2 1Orai NLP Technologies [g.urbizu|i.sanvicente|x.saralegi]@orai.eus 1Orai NLP Technologies [g.urbizu|i.sanvicente|x.saralegi]@orai.eus Abstract Our study reveals that the data size and model size power law values provided by Kaplan et al. (2020) and Hoffmann et al. (2022) are not optimal in these scenarios. Instead, our experiments show that data size should be relatively bigger than what those scaling laws estimated when training small models (16-124M parameters) for optimal perfor- mance. Furthermore, given a fixed computational budget, it is better to train big models instead of Findings of the Association for Computational Linguistics: ACL 2023, pages 7771–7789 July 9-14, 2023 ©2023 Association for Computational Linguistics Scaling Laws for BERT in Low-Resource Settings Iñaki San Vicente1 Xabier Saralegi1 Gorka Urbizu1 Rodrigo Agerri2 2 Related Work Since the emergence of the attention-based Trans- former (Vaswani et al., 2017) architecture and the masking pre-training strategies introduced since BERT (Devlin et al., 2019), different pre-training strategies have been published. But aside from the improvements to the architecture or training procedures, the qualitative improvement in results is mainly the result of increasing the model size, alongside the amount of text corpora used to train them: Chinchilla (70B parameters) (Hoffmann et al., 2022), LaMDA (137B) (Thoppilan et al., 2022), GPT-3 (175B) (Brown et al., 2020), Go- pher (280B) (Rae et al., 2021) and PaLM (540B) (Chowdhery et al., 2022). This fast-growing in- crease in model sizes and data is proven to surface new abilities in larger models, but not present in smaller models (Wei et al., 2022a). Hoffmann et al. (2022) find the optimal model size and the number of training tokens for given a fixed FLOPs budget. For this purpose, they draw their own scaling laws, based on the losses of over 400 models, ranging from 70M to 16B parame- ters, and trained on 5B to 400B tokens. They state that model size and the number of training tokens should scale equally, based on three alternative ap- proaches, while Kaplan et al. (2020) extrapolates that every time the model size is increased by 8, the data only needs to be increased by 5. Thus, after concluding that the performance of most of the current large language models is bottlenecked by the undersized corpora, they train Chinchilla. The relationship between the size of the pre- training corpus and the performance of the lan- guage model in NLU tasks has been addressed in the literature before. The performance improves when the amount of data is increased (Zhang et al., 2021; Hu et al., 2020; Micheli et al., 2020; Raffel et al., 2020), although, at a certain point, the in- crease in performance slows down when the model size is kept fixed (Inoue et al., 2021; Martin et al., 2020; Micheli et al., 2020; Raffel et al., 2020; Liu et al., 2021). Furthermore, it is more convenient to improve the diversity of training datasets, than to add more text from the same domains (Inoue et al., 2021; Martin et al., 2020; Liu et al., 2021). However, the scaling laws of Kaplan et al. (2020) and Hoffmann et al. (2022) are not useful for the low-resource settings we want to focus on. 1 Introduction Pre-trained neural language models based on the Transformer architecture have shown impressive results on many NLP tasks to the point that their use has become standard practice. The capabil- ities of these models improve as the complexity (in terms of parameters) of their architecture (Wei 7771 computing more model updates in smaller models. computing more model updates in smaller models. Additionally, we establish the minimally re- quired combinations of compute, model and dataset sizes of Transformer-based approaches that outper- form other lighter neural baselines, taking CO2 emissions into consideration. et al. (2022), whose aim is aligned with this work, empirically study the optimal ratios of the training tokens, model parameters, and computation to train dense language models and infer scaling laws. computing more model updates in smaller models. Additionally, we establish the minimally re- quired combinations of compute, model and dataset sizes of Transformer-based approaches that outper- form other lighter neural baselines, taking CO2 emissions into consideration. Kaplan et al. (2020) train models of a size rang- ing from 768 to 1.5 billion parameters with datasets ranging from 22 million to 23 billion tokens and conclude that LM performance improves smoothly as we increase the model size, dataset size, and amount of computation. They show that all three factors must be scaled up in tandem, to avoid bot- tleneck issues. Furthermore, they note that larger models are more sample-efficient, and that conver- gence is inefficient, suggesting that it’s better to under-train a bigger model than converge a smaller one on the same computing budget. 2 Related Work Accord- ing to Kaplan et al. (2020) we need very small train- ing corpora (e.g. 744K tokens for a BERTBASE, which is clearly not enough or optimal). Hoffmann et al. (2022) infers significantly bigger training corpora: e.g. 3M tokens for a BERTMINI or 86M tokens for a BERTBASE. Current models for low-resource languages are trained with corpora around that range (Joshi et al., 2020): 161M for Irish (Barry et al., 2021), 130M for Luxembourgish (Lothritz et al., 2022), 45M for Galician (Vilares et al., 2021), 16M for Swahili (Martin et al., 2022b) and 4.4M for Quechua (Zevallos et al., 2022). How- ever, increasing pre-training data several orders of magnitude has been proved beneficial for base size models (Liu et al., 2019). The correlation between model size and model performance on NLU tasks has also been analyzed. The performance of the model improves when scal- ing the model size (and FLOPs) (Turc et al., 2019; Raffel et al., 2020; Xia et al., 2022; Clark et al., 2022). However, all these works used very large pre-training datasets. They do not analyze if the increase in performance slows down bottlenecked by the pre-training dataset size and thus, the con- clusion of scaling being always beneficial cannot be extended to low-data scenarios. Finally, those optimal scaling laws have been deduced from models trained over one epoch, while in low/medium-resource settings models are often The works of Kaplan et al. (2020) and Hoffmann 7772 L HH INT H NEP Param. BERT124M 12 768 3072 12 86M 124M BERT51M 8 512 2048 8 25M 51M BERT16M 4 256 1024 4 3M 16M Table 1: Model sizes in our experiments. L: layers. HH: hidden dimensions. INT: intermediate layer dimensions. H: attention heads. NEP: non-embedding parameters. L HH INT H NEP Param. BERT124M 12 768 3072 12 86M 124M BERT51M 8 512 2048 8 25M 51M BERT16M 4 256 1024 4 3M 16M trained over several epochs (Martin et al., 2020; Lothritz et al., 2022; Zevallos et al., 2022). Nonetheless, for certain NLU tasks (e.g. NeQA and Quote Repetition) scaling language models is detrimental (Perez and McKenzie, 2022), creating inverse scaling laws. However, Wei et al. (2022b) implies U-shaped scaling laws where even larger models might be able to solve those tasks that com- prise a true and a distractor task. trained over several epochs (Martin et al., 2020; Lothritz et al., 2022; Zevallos et al., 2022). 2 Related Work Nonetheless, for certain NLU tasks (e.g. NeQA and Quote Repetition) scaling language models is detrimental (Perez and McKenzie, 2022), creating inverse scaling laws. However, Wei et al. (2022b) implies U-shaped scaling laws where even larger models might be able to solve those tasks that com- prise a true and a distractor task. Table 1: Model sizes in our experiments. L: layers. HH: hidden dimensions. INT: intermediate layer dimensions. H: attention heads. NEP: non-embedding parameters. Table 1: Model sizes in our experiments. L: layers. HH: hidden dimensions. INT: intermediate layer dimensions. H: attention heads. NEP: non-embedding parameters. of syntax and semantics. Thus, we set the other two corpora sizes at 25M and 125M words, keeping a constant increase rate among them. 3 Experimental Setup We aim to find the optimal combination of model- size, dataset-size and computing in low-resource environments and assess whether they follow the scaling laws established in the literature. In addi- tion, we seek to find the minimum requirements to overcome computationally lighter neural baselines. Therefore, we carry out experiments for 3 corpus sizes and 3 model sizes, in 4 languages, training a total of 36 different models. Regarding the nature of the texts, corpora for Basque and Spanish are a mix of 75% news and 25% text from Wikipedia. We selected the news- paper Berria1 for Basque, and El Pais2 for Span- ish. Corpora for Swahili and Finnish were built by randomly selecting documents (longer than 10 sen- tences) from the web-crawled cc100 corpus (Con- neau et al., 2020; Wenzek et al., 2020). 3.2 Corpora For each language, we created three corpora com- prising 125M, 25M and 5M words, respectively. We limited the number of corpora sizes to three in order to control the number of experiments, and thus the computational resources needed. 3.1 Language Selection In a similar fashion to (Turc et al., 2019), we em- ploy three different variants of the BERT model, dubbed BERT124M 3, BERT51M and BERT16M. These models have 12, 8 and 4 layers respectively, also shrinking other parameters proportionally (hid- den dimension, number of attention heads, etc.), since model shape does not affect performance sig- nificantly (Kaplan et al., 2020). Table 1 shows a detailed view of the parameters in each model. We also increased the vocabulary sizes from the original 30K subword tokens to 50K because it is beneficial for agglutinative languages (Agerri et al., 2020). We trained each model up to 500K steps with a batch of 256 and a sequence length of 512. For more pre-training details see Appendix B. We conduct the experiments in four languages from different language families, selected among those that have enough monolingual data to train LMs, as well as enough available evaluation datasets for NLU tasks. Hence, the low-resource setting has been simulated in some cases. Among other languages that fulfil those criteria, we opted for Basque (eu), Spanish (es), Swahili (sw) and Finnish (fi). In addition to being part of disjoint language families, these languages are linguistically diverse with different complexities across morphology, syn- tax, verb system and vocabulary (Coloma, 2015) (see Appendix A). 1https://www.berria.eus 2https://elpais.com 3This corresponds to BERTbase in Devlin et al. (2019) 4 Evaluation Settings We evaluate all models intrinsically and extrin- sically. For the intrinsic evaluation, we tested the models on masked language modeling; for the extrinsic evaluation, we selected four NLU downstream tasks with available datasets in all the selected languages: Name Entity Recognition and Classification (NERC), Topic Classification (Topic), Sentiment Analysis (SA) and Question- answering NLI (QNLI). Our selection of tasks in- Preliminary experiments showed a big fall in the results when reducing pre-training data to just 1M words, in consistency with Zhang et al. (2021). Since obtaining corpora of about 5M words is achievable by most languages that have annotated datasets (Joshi et al., 2020), we set the lower bound at 5M words. Zhang et al. (2021) shows that 10M to 100M words of pretraining data are enough for a language model to acquire the linguistic capacities 7773 7773 EU ES SW FI Task train dev test train dev test train dev test train dev test M. NERC 52K 13K 36K 265K 53K 52K 175K 25K 51K 180K 14K 46K F1 Topic 9K 2K 2K 9K 1K 4K 10K 3K 7K 10K 10K 10K F1 SA 6K 1K 1K 5K 2K 1K 6K 782 1K 4K 633 1K F1 QNLI 2K 230 238 30K 4K 4K 4K 624 1K 7K 1K 1K acc MLM 1M 1M 1M 1M acc Table 2: Datasets used in the evaluation. The size for NERC and MLM are not reported in examples but in tokens. F1 refers to the micro-average F1-score, while acc refers to accuracy. Table 2: Datasets used in the evaluation. The size for NERC and MLM are not reported in examples but in tokens. F1 refers to the micro-average F1-score, while acc refers to accuracy. BHTCv2 dataset including 12 thematic classes (Ur- bizu et al., 2022). The Spanish counterpart is ML- doc (Schwenk and Li, 2018) which has 4 classes. For Swahili, we employed Swahili: News Classi- fication Dataset (David, 2020) which has 4 the- matic classes. Since a development dataset split was missing, we randomly selected the 20% of the training split to create it. Furthermore, since the fine-tuning dataset is bigger than the smallest of the pre-training dataset (5M), we downsampled this training to 10K examples. And for Finnish, we selected the 10% version of the Yle corpus6, which contains 10 thematic classes. Performance is measured with the F-score score. 4.4 SA Sentiment Analysis (SA) is a sequence classifica- tion task. For Basque, we employed the dataset BEC2016eu (Urbizu et al., 2022), which has pos- itive, negative and neutral classes. InterTass2020 (Cumbreras et al., 2016) is the Spanish dataset se- lected for SA, which also has positive, negative and neutral classes. For Swahili, we utilized the dataset presented by Martin et al. (2022b). The dataset was mapped to polarity annotation following guidelines from the article: joy ([1]) = positive, disgust ([4]) = negative, neutral ([0]) and surprise ([5]) = neutral. Only examples with a single label were mapped. Original train/dev/test splits were maintained. And lastly, for Finnish, we chose Finnish sentiment7 which only contains positive and negative labels. We use F-score as the performance metric. 4.5 QNLI Question-answering NLI (QNLI) is a sequence classification task. For Basque, we employed QNLIeu (Urbizu et al., 2022). And for Spanish, Swahili and Finnish, we adapted already available 4 Evaluation Settings cludes one sequence labeling task and three se- quence classification tasks including sentiment analysis and QNLI, which are tasks that require a deeper NLU than the shallow linguistic tasks of NERC and Topic Classification (Zhang et al., 2021). Table 2 shows the details of each dataset. 4.1 MLM Masked Language Modeling (MLM) is one of the default pre-training objective functions of BERT. We report both the loss and accuracy of MLM. For this purpose, we created test datasets, from news sources not used for pre-training the models. For Basque we gathered texts from Argia4 news magazine. For Spanish, we opted for texts from the newspaper El Mundo 5. For Swahili, as a data source not used in the pre-training, we randomly selected a sub-corpus from the pre-train data for SwahBERT model (Martin et al., 2022a), which is mostly made up of news (%80). For Finnish we opted for a subset of cc100 not used in the pre- training, due to the lack of document-level news corpora available with an open license. 6www.github.com/spyysalo/yle-corpus 7www.huggingface.co/datasets/sepidmnorozy/ Finnish_sentiment 4www.argia.eus 5www.elmundo.es 4.2 NERC Named Entity Recognition and Classification (NERC) is a token classification task. For Basque, we used the in-domain NERC dataset from the BasqueGLUE benchmark (Urbizu et al., 2022). For Spanish, we opted for the Conll2002 dataset (Sang, 2002). For Swahili, we selected Masakhaner (Ade- lani et al., 2021). And lastly, for Finnish, we used FiNER (Ruokolainen et al., 2019). Each dataset has 4 categories, and we use the F-score as the performance metric. 5www.elmundo.es 4www.argia.eus 5 8In the case of Basque and Spanish, those sequences are a single sentence, while the sequences for Finnish and Swahili are short paragraphs, because the QA datasets were annotated following different methodologies. 4.3 Topic Classification Topic classification is a sequence classification multi-class task. For Basque, we chose the 7774 odels for Basque. dels for Spanish. Figure 3: MLM loss and FLOPs for models for Swahili. Figure 4: MLM loss and FLOPs for models for Finnish. Figure 1: MLM loss and FLOPs for models for Basque. Figure 2: MLM loss and FLOPs for models for Spanish. Figure 3: MLM loss and FLOPs for models for Swahili. Figure 4: MLM loss and FLOPs for models for Finnish. Figure 3: MLM loss and FLOPs for models for Swahili. Figure 1: MLM loss and FLOPs for models for Basque. Figure 3: MLM loss and FLOPs for models for Swahili. Figure 1: MLM loss and FLOPs for models for Basque. Figure 3: MLM loss and FLOPs for models for Swahili. Figure 1: MLM loss and FLOPs for models for Basque. Figure 4: MLM loss and FLOPs for models for Finnish. Figure 4: MLM loss and FLOPs for models for Finnish. Figure 2: MLM loss and FLOPs for models for Spanish. Figure 4: MLM loss and FLOPs for models for Finnish. Figure 2: MLM loss and FLOPs for models for Spanish. epochs9, which are considered default values (De- vlin et al., 2019; Mosbach et al., 2020). For each task and language, we report an average of 5 runs. In order to compare the performance of our mod- els to other lighter approaches, we implement a competitive neural baseline based on contextual embeddings using Flair (Akbik et al., 2019). For sequence labeling tasks, embeddings are passed into a BiLSTM-CRF system based on the archi- tecture proposed by (Huang et al., 2015). For text classification tasks, the computed Flair embeddings are fed into a BILSTM to produce a document-level embedding which is then used in a linear layer to make the class prediction. epochs9, which are considered default values (De- vlin et al., 2019; Mosbach et al., 2020). For each task and language, we report an average of 5 runs. conversational Question Answering (QA) datasets, into a sequence-pair binary classification task fol- lowing the design of QNLI for English (Wang et al., 2019). The QA dataset selected were SQADes (Car- rino et al., 2020), Tydiqasw and Tydiqafi (Clark et al., 2020). Each QNLI dataset has a Question- Sequence pair for entailment8. Tydiqa only pro- vides splits for train and development. 9Selecting the best-performing epoch on the development set. 10Each model took 80h to train on an Nvidia TitanV GPU. 4.3 Topic Classification Thus, we used that development split as our test split, and randomly select some examples from the training set to create our development set. We follow the English QNLI design and use accuracy as the eval- uation metric. In order to compare the performance of our mod- els to other lighter approaches, we implement a competitive neural baseline based on contextual embeddings using Flair (Akbik et al., 2019). For sequence labeling tasks, embeddings are passed into a BiLSTM-CRF system based on the archi- tecture proposed by (Huang et al., 2015). For text classification tasks, the computed Flair embeddings are fed into a BILSTM to produce a document-level embedding which is then used in a linear layer to make the class prediction. We pre-train our own contextual Flair embed- dings using the 125M corpora for each language with the following hyperparameters: Hidden size of 2048, sequence length of 250, a mini-batch size of 100 and 10 epochs. The rest of the training parameters are left in their default setting10. 5 Results Hoffmann et al. (2022) estimates that 3M, 25M and 86M tokens are optimal to train BERT16M BERT51 and BERT124 respectively, while Kaplan et al. (2020) estimates much lower values. Our results show that the amount of data needed to train an LM optimally is no less than 25M words for BERT16M and 125M for BERT51 and BERT124. We carry out an in-depth comparison of these re- sults with additional data in Appendix H. 5.1 Down-Scaling Laws Figures 1-411 show the relation between the FLOPs and the MLM loss in the development dataset for the models in the four languages. Different colours stand for different model sizes (16M/51M/125M), and different symbols represent the pre-training data (5M/25M/125M). Each line is formed by 5 checkpoints (every 100K steps). We can appreciate how the lowest loss is achieved by the biggest model with the most pre-training data, as expected from scaling laws, which dictate an improvement in performance when model-size, dataset-size and compute bud- get are increased simultaneously. Furthermore, the plot shows that increasing pre-training data is more beneficial than increasing the model size or amount of compute in this low-resource scenario; the gap between the loss obtained from increas- ing pre-training data is much bigger than the im- provements obtained when increasing model-size or training steps. 4.6 Systems and Baselines For the extrinsic evaluation, we fine-tuned each of the 36 BERT models making use of the Transform- ers library (Wolf et al., 2020), with a lr of 3e−5, an effective batch size of 32 and training up to 10 In addition to LSTM-based neural baselines, we also include in the comparison a multilingual BERT 7775 model, namely mBERTbase (Devlin et al., 2019). We assume that this kind of multilingual language models will always be available and that they can somehow be an alternative in some low-resource settings, namely when limited resources refer to the computational capacity for pre-training and to the availability of enough text in the target language. In that line, we perform this comparison only for the Basque and Swahili languages which approx- imately include a training corpus size in mBERT no larger than those of the corpora used in our ex- periments (roughly 35M for eu and 11M for sw). Nevertheless, over-fitting issues during pre- training, do not have a direct impact neither on MLM accuracy nor on downstream tasks (see Ap- pendix D). Thus, for the evaluation of the models regarding MLM accuracy (analysis available in Ap- pendix C) and NLU downstream tasks (Section 5.2), we employed the final checkpoints at 500K steps. model, namely mBERTbase (Devlin et al., 2019). We assume that this kind of multilingual language models will always be available and that they can somehow be an alternative in some low-resource settings, namely when limited resources refer to the computational capacity for pre-training and to the availability of enough text in the target language. In that line, we perform this comparison only for the Basque and Swahili languages which approx- imately include a training corpus size in mBERT no larger than those of the corpora used in our ex- periments (roughly 35M for eu and 11M for sw). Regarding languages, a comparison of the four figures (Figures 1-4) shows that the correlation be- tween MLM loss and combinations of model-size, dataset-size and FLOPs is consistent across lan- guages. MLM accuracies are also consistent across languages (Appendix C). 11Zoomed in for BERT16M in the Appendix G. 12outliers marked in red 5.2 Evaluation on NLU tasks This section analyses the performance of our mod- els on the NLU tasks listed in section 4, to measure the effect of model-size and pre-training data-size once finetuned on the downstream tasks. The results for the NERC task are shown in Table 3. As expected, there is a clear positive correlation between the evaluation metric and the model and corpora size, but corpora size has slightly more impact on the performance. The results for topic classification at Table 4 follow the same trends, al- beit with smoother differences. Table 5 presents the results for SA, again repeating the trends, but with a few outliers. Lastly, for QNLI (see Table 6), we observe there is a general trend of improving results while increasing dataset and model sizes. However, many results present large standard devi- ations, leading to several outliers12 that stand out from the general trend. The figures also show that models trained with small datasets yield larger MLM losses in devel- opment with further training (× curves), which we attribute to overfitting, as the training MLM loss does shrink as training advances. Big mod- els with medium datasets (red and black ⋄curves) also show the same tendency, although to a lesser extent. According to the figures, unless we use a dataset of at least 125M in the case of BERT51M and BERT124M, or a dataset of at least 25M in the case of the smallest BERT16, we should consider applying early stopping to our models to avoid over- fitting, which corresponds with the first checkpoint of 100K steps we plotted in most cases. The models trained obtain competitive results, as shown by the results for NERC, topic classification and SA for Swahili, which are new SotA for those datasets to the best of our knowledge. Besides, some of the results obtained with the BERT124M and 125M words are comparable with SotA models trained over huge datasets (See appendix F). 7776 NERCeu 5M 25M 125M BERT16M 63.90±0.5 72.23±0.6 74.12±0.3 BERT51M 70.14±0.4 79.07±0.4 82.98±0.1 BERT124M 73.14±0.5 79.09±0.8 84.58±0.2 NERCes 5M 25M 125M BERT16M 76.57±0.3 81.56±0.5 81.70±0.5 BERT51M 80.43±0.4 85.11±0.8 86.34±0.7 BERT124M 81.75±0.4 84.99±0.8 87.28±0.3 NERCsw 5M 25M 125M BERT16M 86.36±0.2 88.62±0.2 88.63±0.4 BERT51M 88.74±0.2 90.68±0.2 91.63±0.1 BERT124M 88.93±0.4 90.97±0.2 92.09±0.2 NERCfi 5M 25M 125M BERT16M 76.82±0.3 81.48±0.4 81.83±0.3 BERT51M 79.73±0.2 85.27±0.4 87.02±0.2 BERT124M 80.56±0.7 85.77±0.3 88.99±0.2 Table 3: Results for the 9 models on NERC (F1) for Basque, Spanish, Swahili and Finnish. 5.2 Evaluation on NLU tasks SAeu 5M 25M 125M BERT16M 67.80±0.5 68.63±1.0 67.59±0.5 BERT51M 67.00±1.0 68.54±0.5 69.40±0.9 BERT124M 67.22±0.7 68.79±0.7 68.91±0.5 SAes 5M 25M 125M BERT16M 37.67±1.5 37.62±0.7 37.51±1.4 BERT51M 36.05±2.1 37.57±0.5 39.89±0.0 BERT124M 36.37±2.2 37.17±1.1 43.27±1.1 SAsw 5M 25M 125M BERT16M 71.52±0.6 75.56±0.3 74.84±0.6 BERT51M 70.49±0.7 75.39±1.4 77.07±0.0 BERT124M 69.60±1.3 75.54±0.9 79.04±0.7 SAfi 5M 25M 125M BERT16M 89.69±0.2 90.96±0.2 91.14±0.4 BERT51M 89.61±0.6 91.86±0.3 92.58±0.0 BERT124M 90.32±0.5 91.55±0.3 94.38±0.3 Table 5: Results for the 9 models on sentiment analysis (F1) for Basque, Spanish, Swahili and Finnish. Table 3: Results for the 9 models on NERC (F1) for Basque, Spanish, Swahili and Finnish. Table 5: Results for the 9 models on sentiment analysis (F1) for Basque, Spanish, Swahili and Finnish. Topiceu 5M 25M 125M BERT16M 68.00±0.6 71.81±0.3 72.49±0.4 BERT51M 69.98±0.6 73.16±0.6 74.87±0.4 BERT124M 71.70±0.7 74.61±0.3 76.06±0.4 Topices 5M 25M 125M BERT16M 94.54±0.3 95.86±0.3 95.42±0.4 BERT51M 94.89±0.3 95.45±0.2 95.91±0.4 BERT124M 95.32±0.4 95.82±0.3 96.27±0.3 Topicsw 5M 25M 125M BERT16M 91.64±0.3 91.96±0.3 92.45±0.2 BERT51M 92.12±0.2 92.39±0.1 92.88±0.2 BERT124M 91.95±0.4 92.69±0.1 93.07±0.2 Topicfi 5M 25M 125M BERT16M 88.15±0.1 88.94±0.2 89.16±0.2 BERT51M 88.53±0.3 89.40±0.3 89.61±0.3 BERT124M 88.41±0.2 89.72±0.2 90.14±0.1 Table 4: Results for the 9 models on topic classification (F1) for Basque, Spanish, Swahili and Finnish. QNLIeu 5M 25M 125M BERT16M 68.19±2.3 68.95±2.0 71.22±5.0 BERT51M 65.06±0.8 76.37±2.5 74.18±1.6 BERT124M 67.43±2.7 72.66±2.0 74.09±1.7 QNLIes 5M 25M 125M BERT16M 65.01±0.6 70.89±1.4 72.72±0.7 BERT51M 67.00±1.8 74.11±1.1 78.00±0.0 BERT124M 67.39±0.5 73.07±1.3 81.10±0.7 QNLIsw 5M 25M 125M BERT16M 62.80±1.1 62.45±1.2 63.42±1.0 BERT51M 62.27±1.7 63.83±1.4 63.87±1.5 BERT124M 64.08±1.1 62.68±1.2 63.34±1.4 QNLIfi 5M 25M 125M BERT16M 51.49±0.9 50.96±0.7 58.89±3.7 BERT51M 54.28±2.5 54.58±3.2 57.30±4.3 BERT124M 54.07±2.6 59.89±4.5 58.56±1.1 Table 6: Results for the 9 models on QNLI (acc) for Basque, Spanish, Swahili and Finnish. QNLIeu 5M 25M 125M BERT16M 68.19±2.3 68.95±2.0 71.22±5.0 BERT51M 65.06±0.8 76.37±2.5 74.18±1.6 BERT124M 67.43±2.7 72.66±2.0 74.09±1.7 QNLIes 5M 25M 125M BERT16M 65.01±0.6 70.89±1.4 72.72±0.7 BERT51M 67.00±1.8 74.11±1.1 78.00±0.0 BERT124M 67.39±0.5 73.07±1.3 81.10±0.7 QNLIsw 5M 25M 125M BERT16M 62.80±1.1 62.45±1.2 63.42±1.0 BERT51M 62.27±1.7 63.83±1.4 63.87±1.5 BERT124M 64.08±1.1 62.68±1.2 63.34±1.4 QNLIfi 5M 25M 125M BERT16M 51.49±0.9 50.96±0.7 58.89±3.7 BERT51M 54.28±2.5 54.58±3.2 57.30±4.3 BERT124M 54.07±2.6 59.89±4.5 58.56±1.1 Table 6: Results for the 9 models on QNLI (acc) for Basque, Spanish, Swahili and Finnish. Table 4: Results for the 9 models on topic classification (F1) for Basque, Spanish, Swahili and Finnish. Table 6: Results for the 9 models on QNLI (acc) for Basque, Spanish, Swahili and Finnish. 5.2.1 Evaluation vs Baseline Systems BERT16M models trained with the smallest cor- pus (5M), not included in this table, outperform the Flair baseline (trained over a corpus of 125M words). Finally, computational costs (analysed in Section 5.3) ought to be a factor to consider and de- cide if the gain in performance is worth the increase in computational requirements. Table 7 contains the results for the models trained with the corpora of 125M words, compared to the BiLSTM-CRF Flair baseline (trained with the same 125M corpora) and mBERT (for the languages with a comparable target-language pre-training corpus size). BERT models outperform the Flair neural baseline, but, depending on the evaluation dataset (task and language), the baseline is outperformed only by the BASE124M model or by all three model sizes. Furthermore, for some datasets, even the The boost in performance when increasing the model size is larger in downstream tasks than in the MLM intrinsic task, particularly when shifting from the smallest BERT16M to the intermediate 7777 BERT16M BERT51M BERT124M Flair mBERT eu NERC 74.12±0.3 82.98±0.1 84.58±0.2 82.13±0.4 79.39±1.0 Topic 72.49±0.4 74.87±0.4 76.06±0.4 67.89±0.3 70.57±0.5 SA 67.59±0.5 69.40±0.9 68.91±0.5 68.17±0.3 67.34±0.7 QNLI 71.22±5.0 74.18±1.6 74.09±1.7 48.66±5.2 78.48±1.9 es NERC 81.70±0.5 86.34±0.7 87.28±0.3 87.09±0.3 — Topic 95.42±0.4 95.91±0.4 96.27±0.3 94.08±0.4 — SA 37.51±1.4 39.89±0.0 43.27±1.1 34.73±3.0 — QNLI 72.72±0.7 78.00±0.0 81.10±0.7 56.42±0.6 — sw NERC 88.63±0.4 91.63±0.1 92.09±0.2 92.04±0.1 91.17±0.1 Topic 92.45±0.2 92.88±0.2 93.07±0.2 91.83±0.2 91.52±0.2 SA 74.84±0.6 77.07±0.0 79.04±0.7 73.60±0.5 69.17±1.2 QNLI 63.42±1.0 63.87±1.5 63.34±1.4 52.82±2.1 63.48±1.1 fi NERC 81.83±0.3 87.02±0.2 88.99±0.2 84.76±0.4 — Topic 89.16±0.2 89.61±0.3 90.14±0.1 86.58±0.7 — SA 91.14±0.4 92.58±0.0 94.38±0.3 89.74±0.5 — QNLI 58.89±3.7 57.30±4.3 58.56±1.1 51.54±1.2 — Table 7: Results for BERT and Flair models pre-trained with the biggest dataset (125M words) and mBERT (for the languages pre-trained with a comparable corpus). Table 7: Results for BERT and Flair models pre-trained with the biggest dataset (125M words) and mBERT (for the languages pre-trained with a comparable corpus). Table 7: Results for BERT and Flair models pre-trained with the biggest dataset (125M words) and mBERT (for the languages pre-trained with a comparable corpus). (5M token) available, the results obtained with a small model (BERT16M) are on par with its bigger siblings at MLM, Topic, SA and QNLI, but not in NERC, where increasing the model size (up to BERT51M) is needed to get competent results. 13with an average standard deviation of 1.8 14Finetuning values are computed for a single run at Spanish topic classification. 15https://mlco2.github.io/impact#compute 5.2.1 Evaluation vs Baseline Systems In a scenario with a small dataset (25M), BERT16M would only obtain comparable results at topic clas- sification and SA, but a BERT51M model obtains results as good as, or even better than BERT124M. Thus, we can opt for the BERT51M and use only half of the compute. However, if we are work- ing with a pre-training dataset bigger than 125M, BERT124M obtains the best results by far, indicat- ing that it is worth investing the compute needed to train such a model. BERT51M. This indicates that a larger model is better suited for fine-tuning, as the number of train- able parameters is also higher. The results and scaling trends across languages are very consistent. The results and the trends we obtained are also consistent across different tasks, with the exception of QNLI, where results are volatile13 and have many outliers. 5.3 FLOPs and CO2 Emissions Table 8 shows the computational costs and CO2 emissions for each system for training, finetuning14 and inference. We calculated the FLOPs following the same method as Hoffmann et al. (2022). For non-transformer baselines, FLOPs were computed following (Zhang et al., 2018). CO2 emissions were estimated with Machine-Learning Impact cal- culator15 (Lacoste et al., 2019). The neural base- line based on Bi-LSTMs is lighter FLOP-wise, on pre-training, fine-tuning and inference time, even against the smallest BERT16M model. Still, the Flair baseline has higher CO2 emissions for fine- tuning, due to its inability to parallelize from the recurrent nature of the LSTMs. However, here we are comparing models of different sizes, trained for the same amount of steps. What would happen if we want the best model for a fixed computational budget? We an- swer that in Appendix E, where we compare the BERT51M and BERT124M, pre-trained on a compa- rable amount of computation. In line with (Kaplan et al., 2020), we conclude that it is better to under- train a BERT124M than overtraining a BERT51M with the same amount of computation. If we revisit the results on MLM and NLU tasks (Sections 5.1 and 5.2) with computational costs in mind, we can say that if we only have a tiny corpus 14Finetuning values are computed for a single run at Spanish topic classification. 15 15https://mlco2.github.io/impact#compute 13with an average standard deviation of 1.8 16https://github.com/orai-nlp/low-scaling-laws Limitations First of all, our study is limited to languages that use the Latin script. Still, the 4 languages are from different language families and are typologically diverse. Secondly, the low-resource scenario is simulated. As mentioned in 3, in order to carry out the experi- ments the languages involved were required to have enough monolingual data to train LMs, as well as available evaluation datasets for NLU tasks. 6 Conclusions We present a study of the performance of language models in constrained settings, to analyze if the same scaling laws studied for large-language mod- 14Finetuning values are computed for a single run at Spanish topic classification. 15 7778 Pre-training Fine-tuning Inference Model FLOPs CO2eq FLOPs CO2eq FLOPs CO2eq BERT124M 4.9e+19 98 kg 3.4e+16 47 g 1.3e+11 0.18 mg BERT51M 2.0e+19 41 kg 1.4e+16 23 g 5.3e+10 0.07 mg BERT16M 6.3e+18 13 kg 4.4e+15 11 g 1.6e+10 0.02 mg Flair 1.4e+17 4 kg 5.3e+14 334 g 5.3e+09 0.01 mg Table 8: Computational costs in FLOPs for pre-training, fine-tuning and inference and their estimated CO2 emissions. Table 8: Computational costs in FLOPs for pre-training, fine-tuning and inference and their estimated CO2 e Table 8: Computational costs in FLOPs for pre-training, fine-tuning and inference and their estimated CO2 emissions. els apply to low-resource scenarios. We find out that the estimated values for optimal balance of model size and corpora size do not hold in these scenarios, and that pre-training tokens should be higher than the amount of model parameters. Our study is limited to 3 language model sizes and 3 pre-training corpora sizes. Including other model sizes like a BERT-Large or a model between 51M and 16M (where there is a big gap in results), and adding more pre-training corpora sizes (let’s say 625M and 1M words) were out of the scope of this work. From our experiments, we conclude that it is preferable to train big models on as much data as possible rather than using the computational power to further train smaller models. We see a clear trend where bigger models tend to quickly over- fit when pre-trained for many epochs with small corpora. Still, even when they overfit in the pre- training stage, bigger models consistently outper- form smaller models in downstream applications which require fine-tuning. In addition, we use the default hyperparam- eters that are commonly used for BERT-base (BERT124M) for the pre-training and fine-tuning of the BERT51M and BERT16M models without any hyperparameter tuning. Acknowledgements This work has been partially funded by the Basque Government (ICL4LANG project, grant no. KK-2023/00094). It has also received funding from the following MCIN/AEI/10.13039/501100011033 projects: (i) DeepKnowledge (PID2021-127777OB-C21) and ERDF A way of making Europe and, (ii) DeepR3 (TED2021-130295B-C31) and European Union NextGeneration EU/PRTR. Rodrigo Agerri currently holds the RYC-2017-23647 fellowship (MCIN/AEI/10.13039/501100011033 and 63.34ESF Investing in your future). We also acknowledge the support of Google’s TFRC program. The experimental results are consistent among languages. Additionally, we empirically estab- lish when the computational cost of using a Transformer-based approach is worth taking. All the pre-training corpora, models and datasets created in this work are publicly available16. References 2020. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901. Asier Gutiérrez Fandiño, Jordi Armengol Estapé, Marc Pàmies, Joan Llop Palao, Joaquin Silveira Ocampo, Casimiro Pio Carrino, Carme Armentano Oller, Car- los Rodriguez Penagos, Aitor Gonzalez Agirre, and Marta Villegas. 2022. Maria: Spanish language mod- els. Procesamiento del Lenguaje Natural, 68. Casimiro Pio Carrino, Marta R Costa-jussà, and José AR Fonollosa. 2020. Automatic spanish translation of squad dataset for multi-lingual question answering. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 5515–5523. Jordan Hoffmann, Sebastian Borgeaud, Arthur Men- sch, Elena Buchatskaya, Trevor Cai, Eliza Ruther- ford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, et al. 2022. Train- ing compute-optimal large language models. arXiv preprint arXiv:2203.15556. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. 2022. Palm: Scaling language modeling with pathways. arXiv preprint arXiv:2204.02311. Jennifer Hu, Jon Gauthier, Peng Qian, Ethan Wilcox, and Roger Levy. 2020. A systematic assessment of syntactic generalization in neural language mod- els. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1725–1744. Aidan Clark, Diego de Las Casas, Aurelia Guy, Arthur Mensch, Michela Paganini, Jordan Hoffmann, Bog- dan Damoc, Blake Hechtman, Trevor Cai, Sebastian Borgeaud, et al. 2022. Unified scaling laws for routed language models. In International Conference on Machine Learning, pages 4057–4086. PMLR. Zhiheng Huang, Wei Xu, and Kai Yu. 2015. Bidirec- tional lstm-crf models for sequence tagging. arXiv preprint arXiv:1508.01991. Jonathan H. Clark, Eunsol Choi, Michael Collins, Dan Garrette, Tom Kwiatkowski, Vitaly Nikolaev, and Jennimaria Palomaki. 2020. Tydi qa: A benchmark for information-seeking question answering in ty- pologically diverse languages. Transactions of the Association for Computational Linguistics. Go Inoue, Bashar Alhafni, Nurpeiis Baimukan, Houda Bouamor, and Nizar Habash. 2021. The interplay of variant, size, and task type in arabic pre-trained language models. In Proceedings of the Sixth Arabic Natural Language Processing Workshop, pages 92– 104. Germán Coloma. 2015. Efectos de compensación entre indicadores de la complejidad de los idiomas. Tech- nical report, Serie Documentos de Trabajo. Pratik Joshi, Sebastin Santy, Amar Budhiraja, Kalika Bali, and Monojit Choudhury. 2020. The state and fate of linguistic diversity and inclusion in the nlp world. arXiv preprint arXiv:2004.09095. References David Ifeoluwa Adelani, Jade Abbott, Graham Neubig, Daniel D’souza, Julia Kreutzer, Constantine Lignos, Chester Palen-Michel, Happy Buzaaba, Shruti Rijh- wani, Sebastian Ruder, et al. 2021. Masakhaner: Named entity recognition for african languages. Transactions of the Association for Computational Linguistics, 9:1116–1131. The source of the pre-training corpora for Swahili and Finnish (cc100) is not completely com- parable with the corpora used for Basque and Span- ish (75% news, 25% Wikipedia), due to the un- availability of a large curated corpus for Swahili, and the lack of big news corpora for Finnish with an open license that allowed us to share freely the pre-training data. Rodrigo Agerri, Iñaki San Vicente, Jon Ander Campos, Ander Barrena, Xabier Saralegi, Aitor Soroa, and Eneko Agirre. 2020. Give your text representation models some love: the case for basque. In Proceed- ings of the 12th Language Resources and Evaluation Conference, pages 4781–4788. Rodrigo Agerri, Iñaki San Vicente, Jon Ander Campos, Ander Barrena, Xabier Saralegi, Aitor Soroa, and Eneko Agirre. 2020. Give your text representation models some love: the case for basque. In Proceed- ings of the 12th Language Resources and Evaluation Conference, pages 4781–4788. 7779 7779 Alan Akbik, Tanja Bergmann, and Roland Vollgraf. 2019. Pooled contextualized embeddings for named entity recognition. In NAACL 2019, 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics, page 724–728. Miguel Ángel García Cumbreras, Eugenio Martínez Cámara, Julio Villena Román, and Janine García Morera. 2016. Tass 2015–the evolution of the span- ish opinion mining systems. Procesamiento del Lenguaje Natural, I(56):33–40. Davis David. 2020. Swahili : News classification dataset. The news version contains both train and test sets. Mikel Artetxe, Itziar Aldabe, Rodrigo Agerri, Olatz Perez-de Viñaspre, and Aitor Soroa. 2022. Does cor- pus quality really matter for low-resource languages? arXiv preprint arXiv:2203.08111. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language under- standing. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Tech- nologies, Volume 1 (Long and Short Papers), pages 4171–4186, Minneapolis, Minnesota. Association for Computational Linguistics. James Barry, Joachim Wagner, Lauren Cassidy, Alan Cowap, Teresa Lynn, Abigail Walsh, Mícheál J Ó Meachair, and Jennifer Foster. 2021. gabert–an irish language model. arXiv preprint arXiv:2107.12930. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. References Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettle- moyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Pro- ceedings of the 58th Annual Meeting of the Asso- ciation for Computational Linguistics, pages 8440– 8451, Online. Association for Computational Lin- guistics. Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. 2020. Scaling laws for neural language models. arXiv preprint arXiv:2001.08361. Taku Kudo. 2018. Subword regularization: Improv- ing neural network translation models with multiple 7780 Ethan Perez and Ian McKenzie. 2022. Inverse scaling prize: Round 1 winners. subword candidates. In Proceedings of the 56th An- nual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 66–75. Jack W Rae, Sebastian Borgeaud, Trevor Cai, Katie Millican, Jordan Hoffmann, Francis Song, John Aslanides, Sarah Henderson, Roman Ring, Susan- nah Young, et al. 2021. Scaling language models: Methods, analysis & insights from training gopher. arXiv preprint arXiv:2112.11446. Alexandre Lacoste, Alexandra Luccioni, Victor Schmidt, and Thomas Dandres. 2019. Quantifying the carbon emissions of machine learning. arXiv preprint arXiv:1910.09700. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Man- dar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. RoBERTa: A robustly optimized bert pretraining ap- proach. arXiv preprint arXiv:1907.11692. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J Liu, et al. 2020. Exploring the limits of transfer learning with a unified text-to-text trans- former. J. Mach. Learn. Res., 21(140):1–67. Zeyu Liu, Yizhong Wang, Jungo Kasai, Hannaneh Ha- jishirzi, and Noah A Smith. 2021. Probing across time: What does roberta know and when? In Find- ings of the Association for Computational Linguistics: EMNLP 2021, pages 820–842. Teemu Ruokolainen, Pekka Kauppinen, Miikka Silfver- berg, and Krister Lindén. 2019. A finnish news corpus for named entity recognition. Language Re- sources and Evaluation, pages 1–26. Cedric Lothritz, Bertrand Lebichot, Kevin Allix, Lisa Veiber, Tegawendé François D Assise Bissyande, Jacques Klein, Andrey Boytsov, Anne Goujon, and Clément Lefebvre. 2022. Luxembert: Simple and practical data augmentation in language model pre- training for luxembourgish. In Proceedings of the Language Resources and Evaluation Conference, 2022, pages 5080–5089. Erik F Sang. 2002. Tjong kim (2002).“introduction to the conll-2002 shared task: Language-independent named entity recognition”. In COLING-02: The 6th Conference on Natural Language Learning. References 7781 David Vilares, Marcos Garcia, and Carlos Gómez- Rodríguez. 2021. Bertinho: Galician bert representa- tions. arXiv preprint arXiv:2103.13799. Minjia Zhang, Wenhan Wang, Xiaodong Liu, Jianfeng Gao, and Yuxiong He. 2018. Navigating with graph representations for fast and scalable decoding of neu- ral language models. In Advances in Neural Infor- mation Processing Systems, volume 31. Curran Asso- ciates, Inc. Antti Virtanen, Jenna Kanerva, Rami Ilo, Jouni Luoma, Juhani Luotolahti, Tapio Salakoski, Filip Ginter, and Sampo Pyysalo. 2019. Multilingual is not enough: Bert for finnish. arXiv preprint arXiv:1912.07076. Yian Zhang, Alex Warstadt, Xiaocheng Li, and Samuel Bowman. 2021. When do you need billions of words of pretraining data? In Proceedings of the 59th An- nual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer- ence on Natural Language Processing (Volume 1: Long Papers), pages 1112–1125. Alex Wang, Yada Pruksachatkun, Nikita Nangia, Aman- preet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel Bowman. 2019. Superglue: A stick- ier benchmark for general-purpose language under- standing systems. Advances in neural information processing systems, 32. Jason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, Sebastian Borgeaud, Dani Yogatama, Maarten Bosma, Denny Zhou, Donald Metzler, Ed H. Chi, Tatsunori Hashimoto, Oriol Vinyals, Percy Liang, Jeff Dean, and William Fedus. 2022a. Emer- gent abilities of large language models. Transactions on Machine Learning Research. Survey Certifica- tion. Jason Wei, Yi Tay, and Quoc V Le. 2022b. In- verse scaling can become u-shaped. arXiv preprint arXiv:2211.02011. Guillaume Wenzek, Marie-Anne Lachaux, Alexis Con- neau, Vishrav Chaudhary, Francisco Guzmán, Ar- mand Joulin, and Edouard Grave. 2020. CCNet: Extracting high quality monolingual datasets from web crawl data. In Proceedings of the Twelfth Lan- guage Resources and Evaluation Conference, pages 4003–4012, Marseille, France. European Language Resources Association. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier- ric Cistac, Tim Rault, Remi Louf, Morgan Funtow- icz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mariama Drame, Quentin Lhoest, and Alexander Rush. 2020. Trans- formers: State-of-the-art natural language processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 38–45, Online. Association for Computational Linguistics. Mengzhou Xia, Mikel Artetxe, Chunting Zhou, Xi Vic- toria Lin, Ramakanth Pasunuru, Danqi Chen, Luke Zettlemoyer, and Ves Stoyanov. 2022. Training tra- jectories of language models across scales. References Holger Schwenk and Xian Li. 2018. A corpus for mul- tilingual document classification in eight languages. In Proceedings of the Eleventh International Confer- ence on Language Resources and Evaluation (LREC 2018). Gati Martin, Medard Edmund Mswahili, Young-Seob Jeong, and Jiyoung Woo. 2022a. SwahBERT: Lan- guage model of Swahili. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Hu- man Language Technologies, pages 303–313, Seattle, United States. Association for Computational Lin- guistics. Romal Thoppilan, Daniel De Freitas, Jamie Hall, Noam Shazeer, Apoorv Kulshreshtha, Heng-Tze Cheng, Alicia Jin, Taylor Bos, Leslie Baker, Yu Du, et al. 2022. Lamda: Language models for dialog applica- tions. arXiv preprint arXiv:2201.08239. Gati Martin, Medard Edmund Mswahili, Young-Seob Jeong, and Jeong Young-Seob. 2022b. Swahbert: Language model of swahili. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Hu- man Language Technologies, pages 303–313. Iulia Turc, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Well-read students learn better: On the importance of pre-training compact models. arXiv preprint arXiv:1908.08962. Gorka Urbizu, Iñaki San Vicente, Xabier Saralegi, Ro- drigo Agerri, and Aitor Soroa. 2022. BasqueGLUE: A Natural Language Understanding Benchmark for Basque. In Proceedings of the Language Resources and Evaluation Conference, pages 1603–1612, Mar- seille, France. European Language Resources Asso- ciation. Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suárez, Yoann Dupont, Laurent Romary, Éric Ville- monte De La Clergerie, Djamé Seddah, and Benoît Sagot. 2020. Camembert: a tasty french language model. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7203–7219. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, 30. Vincent Micheli, Martin d’Hoffschmidt, and François Fleuret. 2020. On the importance of pre-training data volume for compact language models. In Proceed- ings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 7853–7858. Manuel García Vega, Manuel Carlos Díaz- Galiano, Miguel Ángel García Cumbreras, Flor Miriam Plaza del Arco, Arturo Montejo-Ráez, Salud María Jiménez Zafra, Eugenio Martínez Cámara, César Antonio Aguilar, Marco Antonio Sobrevilla Cabezudo, Luis Chiruzzo, et al. 2020. In IberLEF@ SEPLN. Marius Mosbach, Maksym Andriushchenko, and Diet- rich Klakow. 2020. On the stability of fine-tuning bert: Misconceptions, explanations, and strong base- lines. In International Conference on Learning Rep- resentations. B Pre-Training Details We use a cased sub-word vocabulary containing 50K tokens trained with the unigram language model based sub-word segmentation algorithm proposed by Kudo (2018). The vocabularies are learned from each training corpus with a charac- ter coverage of 99.95%, to ignore rare characters. Thus, we obtain 3 vocabularies for each language, one for each size of the pre-training corpora (5M, 25M, 125M), which are shared among LMs of dif- ferent sizes throughout our experiments. On the other hand, increasing the model size only helps once we reach a certain amount of pre-training data. Increasing model size from BERT16M to BERT51M does not improve MLM accuracy for a 5M corpus, suggesting that 3M non- embedding parameters are enough to absorb the knowledge of such a small dataset. However, in- creasing model size from BERT51M to BERT124M for the same 5M corpus does improve the overall performance for all languages except for Finnish, This might be due to larger language models being more sample-efficient (Kaplan et al., 2020). We apply several Maskings to the same sen- tences, to create different examples from the same text17, which is a common practice during the pre- processing of the pre-training data. We applied 10 different random maskings to each text and we em- ployed whole-word masking, where whole words are masked instead of the sub-word units. All mod- els were trained on TPUv3-8 machines using the same set of default hyperparameters (Devlin et al., 2019) in all model sizes: a learning rate 1e−4, β1 = 0.9, β2 = 0.999, L2 weight decay of 0.01, a learning rate warmup of 10K steps, and training the mod- els for a total of 500K steps with a batch size of 256 and a sequence length of 512. This means that we will be doing many epochs (500/2500/12500K) over the same corpus, a common practice, when there is not an enormous pre-training corpus avail- able, for instance, in the original publication of BERT (Devlin et al., 2019), the model is pre-trained for 40 epochs. Surprisingly, BERT51M outperforms BERT124M consistently across all languages when pre-trained with a 25M corpus; this goes against the intuition of larger models being more sample-efficient. Further- more, the table shows that a slightly smaller model with more data can outperform a larger model with smaller corpora; every BERT51M model trained with 125M token corpora outperforms BERT124M model trained with 25M tokens. References Rodolfo Zevallos, John Ortega, William Chen, Richard Castro, Nuria Bel, Cesar Toshio, Renzo Venturas, Hilario Aradiel, and Nelsi Melgarejo. 2022. Intro- ducing qubert: A large monolingual corpus and bert model for southern quechua. In Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing, pages 1–13. 7782 C MLM Evaluation Table 10 shows the accuracies obtained in the MLM task for each language (Basque, Spanish, Swahili and Finnish), for each model size and corpus size combination. As expected, larger models trained with the biggest corpora yield the best results, and a positive correlation exists between model/corpora size and accuracy in every language we compare. Moreover, results show that in overall, in these low-resource settings, it is preferable to increase the pre-training data over model size. Increasing pre-training data improve results on MLM for all languages, with the exception of the BERT16M model trained with the 125M token dataset. The gain obtained with the smallest BERT16M models as we keep adding training data diminishes, which suggests that performance is reaching a plateau in these models. Table 9 shows the linguistic characteristics of the languages we selected for our experiments, which are Basque (eu), Spanish (es), Swahili (sw) and Finnish (fi). On one hand, we have the language families they belong to, and on the other hand, their complexity in morphology, syntax, verb system and vocabulary according to Coloma (2015). 17I love cats: I love [MASK]; I [MASK] cats; [MASK] love cats E Optimizing for a Fixed Budget We have shown that increasing the amount of pre- training data and model size improves their per- formance. Thus, the conclusion regarding data in low-resourced settings is to use all the data there is available, independently of the model size. With respect to the model size, however, even if the available corpus size suggests that increasing it improves the performance, there is usually a lim- ited computational budget constraining this. Thus, we need to choose the best model size within our budget. Kaplan et al. (2020) concludes that con- vergence is inefficient, which means that we obtain optimal performance by training larger models and stopping significantly short of convergence when working with a fixed compute budget. For this purpose, we compare the BERT51 and BERT124, pre-trained on a comparable amount of compute. We employed the pre-training corpus of 125M words, and pre-trained BERT51 for 500K steps, and BERT124 for 200K steps. Table 10: Accuracies on MLM for the 4 languages and averages for each model-dataset size. Rows correspond to different Language Model sizes; columns correspond to different corpus sizes employed during pre-training. The results obtained are shown in Table 11. BERT124M, the model with the most parameters outperforms BERT51M in most of the tasks: 4/4 for Spanish, 3/4 for Swahili, 2/4 for Basque and 2/4 for Finnish. These results agree with the claim of Kaplan et al. (2020) that convergence being in- efficient. However, since there is not a big gap in the results, other factors might be also considered. For example, an undertrained BERT124M model has more room for improvement with further pre- training, while BERT51M is cheaper and faster to finetune and deploy. tasks once finetuned, here we compare the check- points of the models at 100K steps, with the last checkpoint of our models at 500K steps. We did this comparison with 2 models, BERT16 and BERT51, both of them trained on the smallest cor- pora (5M words), which are among those with the most pronounced increasing loss curves. The results for each checkpoint for both models after finetuning on the tasks are shown in Table 12. All in all, the 500K step checkpoints are on a par with the 100K step counterparts, without a clear winner, but definitely equalizing the gap that there is at MLM loss. D Does Overfitting at Pre-training Propagate to Finetuning at Downstream Tasks? MLMeu 5M 25M 125M BERT16M 32.08 38.68 41.56 BERT51M 32.42 44.29 50.07 BERT124M 34.50 43.46 53.19 MLMes 5M 25M 125M BERT16M 39.09 49.06 48.31 BERT51M 39.24 53.49 59.04 BERT124M 42.45 52.58 62.00 MLMsw 5M 25M 125M BERT16M 38.03 45.71 44.98 BERT51M 38.08 50.12 55.27 BERT124M 40.43 49.06 58.82 MLMfi 5M 25M 125M BERT16M 29.43 37.03 37.73 BERT51M 28.18 42.07 45.86 BERT124M 29.30 41.75 49.88 D Does Overfitting at Pre-training Propagate to Finetuning at Downstream Tasks? Although the models are trained for the same amount of steps and batch size, the time needed for training each of them is different, with larger models taking more time. We trained all our mod- els using TPUv3-8 machines; in which we trained BERT124M models to 500K steps in 76 hours, BERT51M models in 32 hours and BERT16M mod- els in 10 hours. The loss curves in Section 5.1 suggested that some model-dataset size ratios, which have the least data and more model parameters, have been trained for too long, to a degree in which the loss starts to increase significantly. To analyze if those overfitting issues from when we keep pre-training over and over again on the same training data propagates to the downstream 7783 Language Language family Morphology Syntax Verb System Vocabulary Basque Language isolate 0.73 0.58 0.77 0.62 Spanish Romance (Indo-European) 0.64 0.42 0.62 0.69 Swahili Bantu (Niger-Congo) 0.64 0.42 0.54 0.31 Finnish Uralic 0.82 0.42 0.46 0.31 Table 9: The four selected languages and their complexity in morphology, syntax, verb system and vocabulary. Language family Morphology Syntax Verb System Vocabulary ly Morphology Syntax Verb System Vocabulary Spanish Romance (Indo-European) 0.64 0.42 0.62 0.69 Swahili Bantu (Niger-Congo) 0.64 0.42 0.54 0.31 Finnish Uralic 0.82 0.42 0.46 0.31 Table 9: The four selected languages and their complexity in morphology, syntax, verb system and vocabulary. Table 9: The four selected languages and their complexity in morphology, syntax, verb system and vocabulary. MLMeu 5M 25M 125M BERT16M 32.08 38.68 41.56 BERT51M 32.42 44.29 50.07 BERT124M 34.50 43.46 53.19 MLMes 5M 25M 125M BERT16M 39.09 49.06 48.31 BERT51M 39.24 53.49 59.04 BERT124M 42.45 52.58 62.00 MLMsw 5M 25M 125M BERT16M 38.03 45.71 44.98 BERT51M 38.08 50.12 55.27 BERT124M 40.43 49.06 58.82 MLMfi 5M 25M 125M BERT16M 29.43 37.03 37.73 BERT51M 28.18 42.07 45.86 BERT124M 29.30 41.75 49.88 Table 10: Accuracies on MLM for the 4 languages and averages for each model-dataset size. Rows correspond to different Language Model sizes; columns correspond to different corpus sizes employed during pre-training. F Comparison with SotA on Downstream Tasks Thus, since the decline in loss when kept pre- training does not spread to the downstream tasks, we decide to employ the last checkpoints (500K steps), to evaluate and compare the models at Ap- pendix C and Section 5.2, to avoid adding another variable to the evaluation. In Table 13 we compare the results of our BERT124M trained over the corpora of 125M words, the baselines of Flair, and mBERT with the current state-of-the-art results on each language and task. We improve SotA results for NERC, topic 7784 Figure 6: MLM loss and FLOPs for BERT16M models for Spanish, zoomed in from Figure 2. BERT51M BERT124M Steps 500K 200K FLOPs 1.002E+19 1.347E+19 eu NERC 82.98±0.1 83.67±0.3 Topic 74.87±0.4 75.49±0.3 SA 69.40±0.9 68.43±0.8 QNLI 74.18±1.6 72.49±2.9 es NERC 86.34±0.7 86.47±0.7 Topic 95.91±0.4 95.95±0.1 SA 39.89±0.0 42.79±1.1 QNLI 78.00±0.6 79.65±0.4 sw NERC 91.63±0.1 91.66±0.2 Topic 92.88±0.2 93.07±0.1 SA 77.07±0.0 77.60±1.1 QNLI 63.87±1.5 63.63±0.9 fi NERC 87.02±0.2 86.86±0.7 Topic 89.61±0.3 89.63±0.2 SA 92.58±0.0 92.63±0.5 QNLI 57.30±4.3 56.35±7.6 Table 11: Results of BERT51M and BERT124M models trained with comparable computational budget (1E+19 FLOPs), for 500K and 200K steps respectively. Figure 6: MLM loss and FLOPs for BERT16M models for Spanish, zoomed in from Figure 2. Figure 7: MLM loss and FLOPs for BERT16M models for Swahili, zoomed in from Figure 3. Table 11: Results of BERT51M and BERT124M models trained with comparable computational budget (1E+19 FLOPs), for 500K and 200K steps respectively. Table 11: Results of BERT51M and BERT124M models trained with comparable computational budget (1E+19 FLOPs), for 500K and 200K steps respectively. Figure 7: MLM loss and FLOPs for BERT16M models for Swahili, zoomed in from Figure 3. classification18 and sentiment analysis for Swahili, and obtain similar results for topic classification for Finnish. Figure 8: MLM loss and FLOPs for BERT16M models for Finnish, zoomed in from Figure 4. G MLM Loss Plots Zoomed in for BERT16M Figure 5: MLM loss and FLOPs for BERT16M models for Basque, zoomed in from Figure 1. G MLM Loss Plots Zoomed in for BERT16M G MLM Loss Plots Zoomed in for Figure 8: MLM loss and FLOPs for BERT16M models for Finnish, zoomed in from Figure 4. 18Our model is finetuned with a subset of the dataset 19a = 0.73 and b = 0.27 20a = 0.5 and b = 0.5 H Takeaways from Scaling Laws for Low-Resource Settings Figure 5: MLM loss and FLOPs for BERT16M models for Basque, zoomed in from Figure 1. In Tables 14-17 we compare our results to the pre- dictions of previous scaling laws from Kaplan et al. (2020)19 and Hoffmann et al. (2022)20. Since the loss curve lines for BERT16M for the corpora of 25M and 125M tokens are hard to see in Figures 1, 2, 3 and 4, we zoomed in on them in the figures 5, 6, 7 and 8 respectively. Tables 14-17 show the estimates of Kaplan et al. (2020) do not hold in this low-resource setting, by several magnitudes of order. Table 13: Results of BERT124 model pre-trained with the biggest dataset (125M words), the Flair baseline pre- trained with the same corpora, mBERT and current SOTA. H Takeaways from Scaling Laws for Low-Resource Settings Table 15 shows how 19a = 0.73 and b = 0.27 20a = 0.5 and b = 0.5 19a = 0.73 and b = 0.27 20a = 0.5 and b = 0.5 7785 BERT16M BERT51M Lang Task 100K step 500K step 100K step 500K step eu MLM loss 4.6724 5.0323 5.9511 8.4830 MLM acc 31.56 32.08 33.60 32.42 NERC 64.75±0.6 63.90±0.5 70.13±0.4 70.14±0.4 Topic 68.56±0.4 68.00±0.6 70.18±0.6 69.98±0.6 SA 67.16±0.5 67.80±0.5 67.32±0.4 67.00±1.0 QNLI 68.20±1.2 68.10±2.3 64.64±1.6 65.06±0.8 es MLM loss 3.9658 4.3367 5.8144 8.2425 MLM acc 40.15 39.91 39.27 38.93 NERC 75.84±0.4 76.57±0.3 81.11±0.3 80.43±0.4 Topic 94.69±0.4 94.54±0.3 94.89±0.3 94.89±0.3 SA 37.26±0.8 37.67±1.5 35.68±3.9 36.05±2.1 QNLI 65.38±1.0 65.01±0.6 66.34±1.6 67.00±1.8 sw MLM loss 4.2363 4.5952 5.9836 8.4719 MLM acc 37.64 38.03 38.70 38.08 NERC 85.92±0.5 86.36±0.2 89.05±0.2 88.74±0.2 Topic 91.28±0.3 91.64±0.3 91.85±0.4 92.12±0.2 SA 71.31±0.5 71.52±0.6 71.01±0.6 70.49±0.7 QNLI 60.29±0.9 62.80±1.1 62.88±0.9 62.27±1.7 fi MLM loss 5.2866 5.5322 6.5559 8.9016 MLM acc 28.76 29.43 29.52 28.18 NERC 76.47±0.3 76.82±0.3 80.19±0.2 79.73±0.2 Topic 88.53±0.1 88.15±0.1 88.45±0.1 88.53±0.3 SA 89.95±0.1 89.69±0.2 90.50±0.2 89.61±0.6 QNLI 51.51±1.1 51.49±0.9 52.37±0.7 54.28±2.5 parison on downstream tasks of different checkpoints (100-500K step) of the mode BERT16M BERT51M Lang Task 100K step 500K step 100K step 500K step eu MLM loss 4.6724 5.0323 5.9511 8.4830 MLM acc 31.56 32.08 33.60 32.42 NERC 64.75±0.6 63.90±0.5 70.13±0.4 70.14±0.4 Topic 68.56±0.4 68.00±0.6 70.18±0.6 69.98±0.6 SA 67.16±0.5 67.80±0.5 67.32±0.4 67.00±1.0 QNLI 68.20±1.2 68.10±2.3 64.64±1.6 65.06±0.8 es MLM loss 3.9658 4.3367 5.8144 8.2425 MLM acc 40.15 39.91 39.27 38.93 NERC 75.84±0.4 76.57±0.3 81.11±0.3 80.43±0.4 Topic 94.69±0.4 94.54±0.3 94.89±0.3 94.89±0.3 SA 37.26±0.8 37.67±1.5 35.68±3.9 36.05±2.1 QNLI 65.38±1.0 65.01±0.6 66.34±1.6 67.00±1.8 sw MLM loss 4.2363 4.5952 5.9836 8.4719 MLM acc 37.64 38.03 38.70 38.08 NERC 85.92±0.5 86.36±0.2 89.05±0.2 88.74±0.2 Topic 91.28±0.3 91.64±0.3 91.85±0.4 92.12±0.2 SA 71.31±0.5 71.52±0.6 71.01±0.6 70.49±0.7 QNLI 60.29±0.9 62.80±1.1 62.88±0.9 62.27±1.7 fi MLM loss 5.2866 5.5322 6.5559 8.9016 MLM acc 28.76 29.43 29.52 28.18 NERC 76.47±0.3 76.82±0.3 80.19±0.2 79.73±0.2 Topic 88.53±0.1 88.15±0.1 88.45±0.1 88.53±0.3 SA 89.95±0.1 89.69±0.2 90.50±0.2 89.61±0.6 QNLI 51.51±1.1 51.49±0.9 52.37±0.7 54.28±2.5 Table 12: Comparison on downstream tasks of different checkpoints (100-500K step) of the models that showed over-fitting issues during pre-training due to over-parametrizing. Table 12: Comparison on downstream tasks of different checkpoints (100-500K step) of the models that showed over-fitting issues during pre-training due to over-parametrizing. H Takeaways from Scaling Laws for Low-Resource Settings Furthermore, Tables 14 and 16 show that the optimal FLOPs needed for those models are a few orders of magnitude higger than predicted by Optimal FLOPs Dataset Kaplan Hoffman Ours 125M 9.74E+29 1.56E+16 3.37E+19 25M 2.51E+27 6.25E+14 5.39E+18 5M 6.47E+24 2.50E+13 1.35E+18 Table 14: Optimal FLOPs for each dataset size. Optimal model size Datasets Kaplan Hoffman Ours 125M 7.79E+21 1.25E+08 >8.56E+07 25M 1.00E+20 2.50E+07 >8.56E+07 5M 1.29E+18 5.00E+06 8.56E+07 Table 15: Optimal model size in parameters for each dataset. Optimal FLOPs Dataset Kaplan Hoffman Ours 125M 9.74E+29 1.56E+16 3.37E+19 25M 2.51E+27 6.25E+14 5.39E+18 5M 6.47E+24 2.50E+13 1.35E+18 Table 14: Optimal FLOPs for each dataset size. • Given a fixed computational budget, it is bet- ter to train big models instead of using the computational power to compute more model updates in smaller models. • For a dataset of 125M words: BERT124M, trained for at least 3.37E+19 FLOPs21 is rec- ommended. Table 14: Optimal FLOPs for each dataset size. Table 14: Optimal FLOPs for each dataset size. Optimal model size Datasets Kaplan Hoffman Ours 125M 7.79E+21 1.25E+08 >8.56E+07 25M 1.00E+20 2.50E+07 >8.56E+07 5M 1.29E+18 5.00E+06 8.56E+07 Table 15: Optimal model size in parameters for each dataset. • For a 25M dataset: BERT124M, trained for 5.39E+18 FLOPs22 is recommended, but a BERT51M model trained for 3.01E+18 FLOPs23, obtains similar results, and it is lighter for finetuning and inference. Table 15: Optimal model size in parameters for each dataset. • For a 5M dataset: BERT124M, trained for 1.35E+18 FLOPs24 or less is recommended, but a BERT51M model trained for 7.01E+17 FLOPs25, obtains similar results, which is lighter for finetuning and inference. Optimal FLOPs Model Kaplan Hoffman Ours BERT124M 7.36E+10 7.33E+15 3.37E+19 BERT51M 1.40E+10 6.49E+14 1.00E+19 BERT16M 8.47E+08 1.08E+13 1.29E+18 Table 16: Optimal FLOPs for each model size. Table 16: Optimal FLOPs for each model size. Table 16: Optimal FLOPs for each model size. Optimal dataset size Model Kaplan Hoffman Ours BERT124M 8.59E+02 8.56E+07 >1.25E+08 BERT51M 5.48E+02 2.55E+07 >1.25E+08 BERT16M 2.57E+02 3.29E+06 1.25E+08 Table 17: Optimal dataset size in tokens for each model size. Table 17: Optimal dataset size in tokens for each model size. Hoffmann estimations for optimal model size are not that far from our results, but Table 17 suggests that the optimal dataset size required to train small language models is around an order of magnitude higger than what the scaling laws of Hoffmann et al. predict. 21500K steps with batch=256 and sequence-length=512 2280K steps with batch=256 and sequence-length=512 23150K steps with batch=256 and sequence-length=512 2420K steps with batch=256 and sequence-length=512 2535K steps with batch=256 and sequence-length=512 H Takeaways from Scaling Laws for Low-Resource Settings Table 12: Comparison on downstream tasks of different checkpoints (100-500K step) of the models that showed over-fitting issues during pre-training due to over-parametrizing. BERT124M Flair mBERT SotA eu NERC 84.58±0.2 82.13±0.4 79.39±1.0 86.98±0.4 roberta-euscrawl-l(Artetxe et al., 2022) Topic 76.06±0.4 67.89±0.3 70.57±0.5 86.51±0.4 ElhBERTeu (Urbizu et al., 2022) SA 68.91±0.5 68.17±0.3 67.34±0.7 70.87±0.5 Berteus (Agerri et al., 2020) QNLI 74.09±1.7 48.66±5.2 78.48±1.9 76.04±1.5 ElhBERTeu (Urbizu et al., 2022) es NERC 87.28±0.3 87.09±0.3 87.21±0.4 88.51 roBerta-b(Gutiérrez Fandiño et al., 2022) Topic 96.27±0.3 94.08±0.4 95.92±0.6 97.14 BETO(Gutiérrez Fandiño et al., 2022) SA 43.27±1.1 34.73±3.0 39.21±1.8 49.80 Vega et al. (2020) QNLI 81.10±0.7 56.42±0.6 83.92±0.2 82.02 roberta-l(Gutiérrez Fandiño et al., 2022) sw NERC 92.09±0.2 92.04±0.1 91.17±0.1 88.60 swahBERT (Martin et al., 2022b) Topic 93.07±0.2 91.83±0.2 91.52±0.2 90.90 swahBERT (Martin et al., 2022b) SA 79.04±0.7 73.60±0.5 69.17±1.2 71.12 swahBERT (Martin et al., 2022b) QNLI 63.34±1.4 52.82±2.1 63.48±1.1 64.72±0.4 swahBERT(our evaluation) fi NERC 88.99±0.2 84.76±0.4 88.87±0.4 92.40±0.1 finBERT(Virtanen et al., 2019) Topic 90.14±0.1 86.58±0.7 88.16±0.3 90.57±0.2 finBERT(Virtanen et al., 2019) SA 94.38±0.3 89.74±0.5 88.05±0.7 95.61±0.3 finBERT(our evaluation) QNLI 58.56±1.1 51.54±1.2 52.79±3.0 57.18±2.6 finBERT(our evaluation) Table 13: Results of BERT124 model pre-trained with the biggest dataset (125M words), the Flair baseline pre- 7786 Optimal FLOPs Dataset Kaplan Hoffman Ours 125M 9.74E+29 1.56E+16 3.37E+19 25M 2.51E+27 6.25E+14 5.39E+18 5M 6.47E+24 2.50E+13 1.35E+18 Table 14: Optimal FLOPs for each dataset size. Optimal model size Datasets Kaplan Hoffman Ours 125M 7.79E+21 1.25E+08 >8.56E+07 25M 1.00E+20 2.50E+07 >8.56E+07 5M 1.29E+18 5.00E+06 8.56E+07 Table 15: Optimal model size in parameters for each dataset. Optimal FLOPs Model Kaplan Hoffman Ours BERT124M 7.36E+10 7.33E+15 3.37E+19 BERT51M 1.40E+10 6.49E+14 1.00E+19 BERT16M 8.47E+08 1.08E+13 1.29E+18 Table 16: Optimal FLOPs for each model size. Optimal dataset size Model Kaplan Hoffman Ours BERT124M 8.59E+02 8.56E+07 >1.25E+08 BERT51M 5.48E+02 2.55E+07 >1.25E+08 BERT16M 2.57E+02 3.29E+06 1.25E+08 Table 17: Optimal dataset size in tokens for each model size. Hoffmann estimations for optimal model size are not that far from our results, but Table 17 suggests that the optimal dataset size required to train small language models is around an order of magnitude higger than what the scaling laws of Hoffmann et al. predict. 2420K steps with batch=256 and sequence-length=512 25 H Takeaways from Scaling Laws for Low-Resource Settings Furthermore, Tables 14 and 16 show that the optimal FLOPs needed for those models are a few orders of magnitude higger than predicted by Hoffmann et al., where models are trained for a single epoch, which is clearly not optimal in low- resource settings. All in all, we can underline the following take- aways for NLP practitioners working on LMs in low-resource settings: • Use as much text as available. • Pretraining for several (100s) epochs is clearly beneficial. 7787 ACL 2023 Responsible NLP Checklist □ A4. Have you used AI writing assistants when working on this paper? Left blank. B □ Did you use or create scientific artifacts? 3 Experimental Setup □B1. Did you cite the creators of artifacts you used? No response. □B3. Did you discuss if your use of existing artifact(s) was consistent with their intended use, provided that it was specified? For the artifacts you create, do you specify intended use and whether that is compatible with the original access conditions (in particular, derivatives of data accessed for research purposes should not be used outside of research contexts)? Not applicable. Left blank. □B4. Did you discuss the steps taken to check whether the data that was collected / used contains any information that names or uniquely identifies individual people or offensive content, and the steps taken to protect / anonymize it? Not applicable. Left blank. □ B5. Did you provide documentation of the artifacts, e.g., coverage of domains, languages, and linguistic phenomena, demographic groups represented, etc.? 3 Experimental Setup and 4 Evaluation Settings □B6. Did you report relevant statistics like the number of examples, details of train / test / dev splits, etc. for the data that you used / created? Even for commonly-used benchmark datasets, include the number of examples in train / validation / test splits, as these provide necessary context for a reader to understand experimental results. For example, small differences in accuracy on large test sets may be significant, while on small test sets they may not be. No response. □B6. Did you report relevant statistics like the number of examples, details of train / test / dev splits, etc. for the data that you used / created? Even for commonly-used benchmark datasets, include the number of examples in train / validation / test splits, as these provide necessary context for a reader to understand experimental results. For example, small differences in accuracy on large test sets may be significant, while on small test sets they may not be. No response. 5 Results and Appendix □ C1. Did you report the number of parameters in the models used, the total computational budget (e.g., GPU hours), and computing infrastructure used? 3 Experimental Setup, 5 Results and Appendix The Responsible NLP Checklist used at ACL 2023 is adopted from NAACL 2022, with the addition of a question on AI writing assistance. □ C1. Did you report the number of parameters in the models used, the total computational budget (e.g., GPU hours), and computing infrastructure used? 3 Experimental Setup, 5 Results and Appendix The Responsible NLP Checklist used at ACL 2023 is adopted from NAACL 2022, with the addition of a question on AI writing i t □ C1. Did you report the number of parameters in the models used, the total computational budget (e.g., GPU hours), and computing infrastructure used? 3 Experimental Setup, 5 Results and Appendix 7788 □ C2. Did you discuss the experimental setup, including hyperparameter search and best-found hyperparameter values? 3 Experimental Setup and Appendix □ C2. Did you discuss the experimental setup, including hyperparameter search and best-found hyperparameter values? 3 Experimental Setup and Appendix □ C3. Did you report descriptive statistics about your results (e.g., error bars around results, summary statistics from sets of experiments), and is it transparent whether you are reporting the max, mean, etc. or just a single run? 3 Experimental Setup and 5 Results □ C3. Did you report descriptive statistics about your results (e.g., error bars around results, summary statistics from sets of experiments), and is it transparent whether you are reporting the max, mean, etc. or just a single run? 3 Experimental Setup and 5 Results □ C4. If you used existing packages (e.g., for preprocessing, for normalization, or for evaluation), did you report the implementation, model, and parameter settings used (e.g., NLTK, Spacy, ROUGE, etc.)? 3 Experimental Setup, 4.6 Systems and Baselines and 5 Results D □ Did you use human annotators (e.g., crowdworkers) or research with human participants? Left blank. D □ Did you use human annotators (e.g., crowdworkers) or research with human participants? Left blank. □D1. Did you report the full text of instructions given to participants, including e.g., screenshots, disclaimers of any risks to participants or annotators, etc.? No response. □D2. Did you report information about how you recruited (e.g., crowdsourcing platform, students) and paid participants, and discuss if such payment is adequate given the participants’ demographic (e.g., country of residence)? No response. 5 Results and Appendix □D3. Did you discuss whether and how consent was obtained from people whose data you’re using/curating? For example, if you collected data via crowdsourcing, did your instructions to crowdworkers explain how the data would be used? No response. □D4. Was the data collection protocol approved (or determined exempt) by an ethics review board? No response. □D5. Did you report the basic demographic and geographic characteristics of the annotator population that is the source of the data? No response. 7789
https://openalex.org/W4361967430
https://aacr.figshare.com/articles/journal_contribution/Figure_S1_from_Low-pass_Whole-genome_Sequencing_of_Circulating_Cell-free_DNA_Demonstrates_Dynamic_Changes_in_Genomic_Copy_Number_in_a_Squamous_Lung_Cancer_Clinical_Cohort/22474496/1/files/39925991.pdf
English
null
Figure S1 from Low-pass Whole-genome Sequencing of Circulating Cell-free DNA Demonstrates Dynamic Changes in Genomic Copy Number in a Squamous Lung Cancer Clinical Cohort
null
2,023
cc-by
588
A C B 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% , 0.173 , 0.078 , 0.016 , 0.109 Cell line, p-value H520 H1781 A549 LXF529 A C B 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% , 0.173 , 0.078 , 0.016 , 0.109 Cell line, p-value H520 H1781 A549 LXF529 A B , 0.173 , 0.078 , 0.016 , 0.109 Cell line, p-value B A , 0.173 , 0.078 , 0.016 , 0.109 Cell line, p-value B A C 0 1% 5% 10% 25% 50% 75% 100% LXF529 C 0 1% 5% 10% 25% 50% 75% 100% H520 A549 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% log2 ratio <-0.5 >0. 0 H520 H1781 A549 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% H520 H1781 A549 0 1% 5% 10% 25% 50% 75% 00% H520 0 1% 5% 10% 25% 50% 75% 100% log2 ratio <-0.5 >0.5 0 H1781 Figure S1. Tumor cell line DNA spike-in. Tumor DNA from 4 lung cancer cell lines was mixed with normal DNA at 0, 1%, 5%, 10%, 25%, 50%, 75% or 100%. A, Comparison of estimated tumor fraction by LP-WGS to real spike-in tumor fraction. Wilcoxon signed ranked test p-values are listed. B, Ploidy of the four cell lines and normal DNA estimated by LP-WGS C, CNVs across the whole genome detected at different spike-in ratios. Heatmap density represents CNV log2 ratios between cfDNA sample and normal reference, ranging between between <-0.5 to >0.5. Blue, copy number gain; red, copy number loss. D, CNV fold change of lung cancer related genes normalized to the healthy reference detected at D 0% 20% 40% 60% 80% 0 0.1 0.2 0.3 0.4 0.5 0.6 LXFL529 A549 H520 H1781 % Genome amplified () Median copy number gain log2 FC E D D D D 0% 20% 40% 60% 80% 0 0.1 0.2 0.3 0.4 0.5 0.6 LXFL529 A549 H520 H1781 % Genome amplified () Median copy number gain log2 FC E E Figure S1. Tumor cell line DNA spike-in. Tumor DNA from 4 lung cancer cell lines was mixed with normal DNA at 0, 1%, 5%, 10%, 25%, 50%, 75% or 100%. A, Comparison of estimated tumor fraction by LP-WGS to real spike-in tumor fraction. Wilcoxon signed ranked test p-values are listed. A C B 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% 0 1% 5% 10% 25% 50% 75% 100% , 0.173 , 0.078 , 0.016 , 0.109 Cell line, p-value H520 H1781 A549 LXF529 B, Ploidy of the four cell lines and normal DNA estimated by LP-WGS. C, CNVs across the whole genome detected at different spike-in ratios. Heatmap density represents CNV log2 ratios between cfDNA sample and normal reference, ranging between between <-0.5 to >0.5. Blue, copy number gain; red, copy number loss. D, CNV fold change of lung cancer-related genes, normalized to the healthy reference, detected at different spike-in ratios. E, Cell line median copy number gain (bars) and % geneome amplified (diamonds) in the 100% spike in (pure cell line).
https://openalex.org/W2253091129
https://pure.mpg.de/pubman/item/item_2328258_5/component/file_2631842/Rjosk_2016.pdf
English
null
Mirror Visual Feedback-Induced Performance Improvement and the Influence of Hand Dominance
Frontiers in human neuroscience
2,016
cc-by
6,997
ORIGINAL RESEARCH published: 20 January 2016 doi: 10.3389/fnhum.2015.00702 Edited by: Bradley R. King, Centre de Recherche de l’Institut Universitaire de Gériatrie de Montréal, Canada Reviewed by: Steven Andrew Jax, Moss Rehabilitation Research Institute, USA Laura Avanzino, University of Genoa, Italy Ippei Nojima, Rehabilitation, Japan Keywords: mirror visual feedback (MVF), hand dominance, motor learning, motor skill learning, handedness *Correspondence: Patrick Ragert patrick.ragert@uni-leipzig.de *Correspondence: Patrick Ragert patrick.ragert@uni-leipzig.de Mirror Visual Feedback-Induced Performance Improvement and the Influence of Hand Dominance Viola Rjosk 1, Elisabeth Kaminski 1, Maike Hoff 1,2, Bernhard Sehm 1, Christopher J. Steele 1,3, Arno Villringer 1,4 and Patrick Ragert 1,2* 1 Department of Neurology, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2 Institute for General Kinesiology and Exercise Science, University of Leipzig, Leipzig, Germany, 3 Department of Psychiatry, Cerebral Imaging Centre, Douglas Mental Health Institute, McGill University, Montreal, QC, Canada, 4 Mind and Brain Institute, Charité and Humboldt University, Berlin, Germany Mirror visual feedback (MVF) is a promising technique in clinical settings that can be used to augment performance of an untrained limb. Several studies with healthy volunteers and patients using transcranial magnetic stimulation (TMS) or functional magnetic resonance imaging (fMRI) indicate that functional alterations within primary motor cortex (M1) might be one candidate mechanism that could explain MVF-induced changes in behavior. Until now, most studies have used MVF to improve performance of the non-dominant hand (NDH). The question remains if the behavioral effect of MVF differs according to hand dominance. Here, we conducted a study with two groups of young, healthy right-handed volunteers who performed a complex ball-rotation task while receiving MVF of the dominant (n = 16, group 1, MVFDH) or NDH (n = 16, group 2, MVFNDH). We found no significant differences in baseline performance of the untrained hand between groups before MVF was applied. Furthermore, there was no significant difference in the amount of performance improvement between MVFDH and MVFNDH indicating that the outcome of MVF seems not to be influenced by hand dominance. Thus our findings might have important implications in neurorehabilitation suggesting that patients suffering from unilateral motor impairments might benefit from MVF regardless of the dominance of the affected limb. INTRODUCTION Mirror visual feedback (MVF) is a promising technique in the context of neurorehabilitation to induce performance improvements without training. For MVF, a mirror is placed in the subject’s midsagittal plane with one limb behind the mirror. Then the subject performs a motor task with one limb in front of the mirror while watching its reflection giving the illusion of the other limb moving. Importantly, the opposite limb behind the mirror should be at rest throughout the MVF task. MVF was originally used by Ramachandran et al. (1995) to reduce phantom limb pain in amputees. Since then, MVF has been successfully applied to improve motor deficits in stroke patients (Altschuler et al., 1999; Yavuzer et al., 2008; Dohle et al., 2009). Recently, several studies have indicated that MVF is also capable of improving performance of an untrained limb in both young and old volunteers without neurological deficits (Hoff et al., 2015; von Rein et al., 2015). Moreover, it has been shown that MVF leads to functional alterations Received: 30 October 2015 Accepted: 14 December 2015 Published: 20 January 2016 Keywords: mirror visual feedback (MVF), hand dominance, motor learning, motor skill learning, handedness Subjects A total number of 32 right-handed healthy young volunteers (mean age: 26.78 ± 0.78 years; range 20–38 years; 19 females) participated in the present study. All volunteers gave their written informed consent before starting the experiment. The study was performed in accordance with the Declaration of Helsinki, and was approved by the local ethics committee of the University of Leipzig. None of the volunteers had a history of neurological illness, and during the time of the experiment none of the volunteers was taking any central-acting drugs. All volunteers were task naïve and right-handed, as assessed with the Edinburgh Handedness Questionnaire (mean handedness score of 88.19 ± 2.95; Oldfield, 1971). Highly skilled musicians or sportsmen were excluded from the study, even though some of the volunteers were currently doing sports on a regular basis or were experienced in playing a musical instrument. Total hours of sports per week and hours of fine-motor training per week (e.g., playing a musical instrument, knitting, doing handcrafts, playing videogames with keypad or joystick) were assessed with a questionnaire. Sixteen volunteers were enrolled in the first study group, who performed the MVF-task with the DH (MVFDH), 16 volunteers were enrolled in the second study group (MVFNDH), who performed the MVF-task with the NDH (for details, see Table 1 for group demographics). Before and after the experiment, all volunteers rated their levels of attention, fatigue and discomfort on a visual analog scale (VAS). Interestingly, most of the above mentioned studies investigated the effect of MVF on the non-dominant hand (NDH) by performing the MVF-task with the DH. It still remains elusive if similar behavioral effects of MVF can be observed for the DH by training with the NDH. This in turn might be important in the context of neurorehabilitation of sensorimotor function after stroke. Previous studies have shown that the level of impairment is stronger in patients with motor deficits of the NDH as compared to the DH (Harris and Eng, 2006). Hence, the question whether the factor hand dominance influences the effectiveness of MVF is of high clinical interest, but not yet investigated. Citation: Rjosk V, Kaminski E, Hoff M, Sehm B, Steele CJ, Villringer A and Ragert P (2016) Mirror Visual Feedback-Induced Performance Improvement and the Influence of Hand Dominance. Front. Hum. Neurosci. 9:702. doi: 10.3389/fnhum.2015.00702 January 2016 | Volume 9 | Article 702 Frontiers in Human Neuroscience | www.frontiersin.org 1 Mirror Visual Feedback and Hand Dominance Rjosk et al. in the human motor system as assessed by transcranial magnetic stimulation (TMS; Nojima et al., 2012). In support of this, recent studies using continuous theta burst stimulation (cTBS) or transcranial direct current stimulation (tDCS) provide further evidence that functional alterations in motor cortex (M1) contralateral to MVF might play a crucial role in mediating performance improvements of the untrained hand (Nojima et al., 2012; Hoff et al., 2015; von Rein et al., 2015). For example, upregulating excitability within M1 contralateral to MVF by means of anodal tDCS has been shown to induce superior performance improvements in both healthy younger and older adults relative to sham stimulation (Hoff et al., 2015; von Rein et al., 2015). However, the exact underlying mechanisms of MVF- induced performance improvements still remain controversial, since other TMS studies indicated different MVF-induced effects within and between M1s (Garry et al., 2005; Fukumura et al., 2007; Lappchen et al., 2012; Avanzino et al., 2014). Functional magnetic resonance imaging (fMRI) studies of MVF provide further evidence that functional alterations are not limited to M1 but also affect other motor-related brain areas such as dorsal premotor cortex (dPMC), ventral premotor cortex (vPMC) and supplementary motor area (SMA; Hamzei et al., 2012). task (Todor and Kyprie, 1980; Triggs et al., 1997; Brouwer et al., 2001; Garry et al., 2004; Goble and Brown, 2008). Furthermore, we hypothesized that the beneficial effect of MVF will be more pronounced for the NDH as compared to the DH due to a quicker saturation of performance in the DH (Ridding and Flavel, 2006). Subjects To address this question, we conducted a study with two groups of young, healthy right-handed volunteers who performed a complex ball-rotation task where one group performed the task in front of the mirror with the DH and hereby received MVF from the DH (MVFDH) while the other group received MVF from the NDH (MVFNDH). The aim of the present study was to investigate: (a) potential baseline differences of the untrained hand (primary outcome measure, either DH or NDH) and (b) differential effects of MVF on the untrained hand (either DH or NDH). Our assumption was that baseline performance of the NDH is worse than that of the DH in a complex fine-motor Experimental Procedure We used a modified version of the complex fine-motor ball- rotation task introduced by Nojima et al. (2012). All volunteers performed the ball-rotation task with two cork balls (diameter 30 mm; weight 10 g) with their DH and NDH in a specific order and direction as described below. During the task, volunteers were seated in a comfortable chair with their elbows flexed TABLE 1 | Group demographics. Group Age (years) Gender (female/male) LQ Sports/week (hours) Fine-motor training/week (hours) MVFDH n = 16 26.56 ± 0.91 9/7 87.75 ± 4.17 2.84 ± 0.45 0.22 ± 0.19 MVFNDH n = 16 27 ± 1.30 10/6 88.63 ± 4.32 3.09 ± 0.48 0.06 ± 0.06 LQ, Laterality Quotient as assessed with the Edinburgh Handedness Scale [range: −100 (full left-handed) to + 100 (full right-handed)]. Hours of sports per week and hours of fine-motor training per week (e.g., playing a musical instrument, knitting, doing handcrafts, playing videogames with keypad or joystick) were assessed with a questionnaire. All values are depicted as mean ± standard error (SE) of the mean. Statistical analysis revealed no differences in age, gender, LQ, sports/week, fine-motor training/week between groups. TABLE 1 | Group demographics. Group Age (years) Gender (female/male) LQ Sports/week (hours) Fine-motor training/week (hours) MVFDH n = 16 26.56 ± 0.91 9/7 87.75 ± 4.17 2.84 ± 0.45 0.22 ± 0.19 MVFNDH n = 16 27 ± 1.30 10/6 88.63 ± 4.32 3.09 ± 0.48 0.06 ± 0.06 LQ, Laterality Quotient as assessed with the Edinburgh Handedness Scale [range: −100 (full left-handed) to + 100 (full right-handed)]. Hours of sports per week and hours of fine-motor training per week (e.g., playing a musical instrument, knitting, doing handcrafts, playing videogames with keypad or joystick) were assessed with a questionnaire. All values are depicted as mean ± standard error (SE) of the mean. Statistical analysis revealed no differences in age, gender, LQ, sports/week, fine-motor training/week between groups. LQ, Laterality Quotient as assessed with the Edinburgh Handedness Scale [range: −100 (full left-handed) to + 100 (full right-handed)]. Hours of sports per week and hours of fine-motor training per week (e.g., playing a musical instrument, knitting, doing handcrafts, playing videogames with keypad or joystick) were assessed with a questionnaire. All values are depicted as mean ± standard error (SE) of the mean. Statistical analysis revealed no differences in age, gender, LQ, sports/week, fine-motor training/week between groups. Experimental Procedure January 2016 | Volume 9 | Article 702 Frontiers in Human Neuroscience | www.frontiersin.org 2 Mirror Visual Feedback and Hand Dominance Rjosk et al. at 90◦and with their pronated hands resting on a desk in front of them. In both groups, volunteers rotated the balls with the NDH always in a counterclockwise direction and with the DH always in a clockwise direction. The number of ball- rotations/min was counted and used to assess motor dexterity. Motor performance was videotaped throughout the experiment and analyzed (number of ball-rotations/min) offline by an experimenter who was blinded to the study procedures. on the movement in the mirror and to relax the untrained hand behind the mirror in both study groups (Figure 1C). To facilitate the mirror illusion, it was taken care that the mirror image of the training hand got superimposed on the untrained hand behind the mirror and volunteers were instructed to take off any jewelry of their hands prior to the experiment. After this training phase, performance of the untrained hand was retested for 1 min (untrained hand post): volunteers in MVFDH rotated the balls with their NDH in a counterclockwise direction, volunteers in MVFNDH performed the task with the DH in a clockwise direction. Importantly, during performance of the untrained hand (pre-MVF and post-MVF) volunteers in both groups were instructed to watch their moving hand. To assess baseline performance, volunteers in MVFDH were asked to rotate two cork balls with their NDH as fast as possible in a counterclockwise direction for 1 min (untrained hand pre). Subsequently, the training period with MVF was conducted: volunteers in MVFDH were instructed to rotate the balls with the DH (trained hand) in a clockwise direction as quickly as possible while observing the movement in a mirror placed between their arms. The MVF-task was performed for 10 trials (trial length 1 min each), separated by 30 s break to prevent muscle fatigue, adding up to a total of 15 min of MVF- training (Figure 1A). Volunteers in MVFNDH conducted the same task, but switched hands respectively (Figure 1B): baseline performance was assessed by 1 min training with the DH in a clockwise direction (untrained hand pre). Then they were asked to complete the set of 10 trials of training (1 min each with 30 s between the trials) with the NDH (trained hand) in a counterclockwise orientation while MVF was provided. Statistical Analyses Statistical analyses were conducted using the Statistical Software Package for Social Sciences (IBM SPSS Version 22). For all analyses, motor performance was assessed as the number of ball- rotations/min both for the untrained hand before (pre-) MVF and after (post-) MVF and for the trained hand during the training period. According to our research aims we first tested for differences in baseline performance of the untrained hand. Hence, the number of ball-rotations/min with the untrained hand pre-MVF was compared between groups (MVFDH vs. MVFNDH) using an independent samples t-test. Second, in order to assess potential influences of hand dominance on the effect of MVF on performance improvements of the untrained FIGURE 1 | Experimental setup and design. Volunteers performed a complex ball-rotation task with two cork balls. (A) Volunteers in MVFDH rotated the balls with their non-dominant hand (NDH) in a counterclockwise direction for 1 min (untrained hand pre), followed by a 15 min training period with the dominant hand (DH) in a clockwise direction while mirror visual feedback (MVF) was provided (10 trials of 1 min each with 30 s break in between). After the training phase, the performance of the NDH was retested (untrained hand post). (B) Volunteers in MVFNDH received the same instructions and conducted the same task just vice versa: MVF-training was performed with the NDH while the DH was assessed before and after MVF (untrained hand pre and post). In both groups, volunteers rotated the balls with the NDH always in a counterclockwise direction and with the DH always in a clockwise direction. (C) During the task, volunteers in both groups were seated in a comfortable chair with their elbows flexed at 90◦and with their pronated hands rested on a desk in front of them. While MVF was provided, a mirror was placed in the subject’s midsagittalplane and the performing hand was covered by a wooden barrier to prevent direct view. See text for details. FIGURE 1 | Experimental setup and design. Volunteers performed a complex ball-rotation task with two cork balls. (A) Volunteers in MVFDH rotated the balls with their non-dominant hand (NDH) in a counterclockwise direction for 1 min (untrained hand pre), followed by a 15 min training period with the dominant hand (DH) in a clockwise direction while mirror visual feedback (MVF) was provided (10 trials of 1 min each with 30 s break in between). Experimental Procedure During MVF-training, direct view of the training hand was prevented by a wooden barrier and volunteers were instructed to concentrate RESULTS Volunteers in both groups did not differ with regard to age [t(30) = −0.276, P = 0.784] gender [t(30) = −0.349, P = 0.729], laterality quotient [LQ; t(30) = −0.146, P = 0.885], their weekly hours of sports [t(30) = −0.381; P = 0.706] or fine-motor training [t(30) = 0.789; P = 0.437; see also Table 1]. Both groups did not differ regarding their level of fatigue [t(30) = −1.125, P = 0.270] or discomfort [t(30) = −0.338, P = 0.737] prior to the experiment. We found, however, a statistically significant difference on the VAS in attention at baseline between groups [t(30) = −2.590, P = 0.015] as well as a significant increase in attention in both groups: by 1.13 ± 0.26 on the VAS in MVFDH [t(15) = −4.392; P = 0.001] and by 0.38 ± 0.02 on Volunteers in both groups did not differ with regard to age [t(30) = −0.276, P = 0.784] gender [t(30) = −0.349, P = 0.729], laterality quotient [LQ; t(30) = −0.146, P = 0.885], their weekly hours of sports [t(30) = −0.381; P = 0.706] or fine-motor training [t(30) = 0.789; P = 0.437; see also Table 1]. Both groups did not differ regarding their level of fatigue [t(30) = −1.125, P = 0.270] or discomfort [t(30) = −0.338, P = 0.737] prior to the experiment. We found, however, a statistically significant difference on the VAS in attention at baseline between groups [t(30) = −2.590, P = 0.015] as well as a significant increase in attention in both groups: by 1.13 ± 0.26 on the VAS in MVFDH [t(15) = −4.392; P = 0.001] and by 0.38 ± 0.02 on Before and after the MVF-task, attention, fatigue and discomfort were assessed with the VAS questionnaire. Attention scale, 1–10: 1, no attention; 10, highest level of attention. Fatigue scale, 1–10: 1, high level of fatigue; 10, no fatigue. Discomfort scale, 1–10: 1, no discomfort; 10, highest level of discomfort. All values are presented as mean ± standard error (SE) of the mean. Before and after the MVF-task, attention, fatigue and discomfort were assessed with the VAS questionnaire. Attention scale, 1–10: 1, no attention; 10, highest level of attention. Fatigue scale, 1–10: 1, high level of fatigue; 10, no fatigue. Discomfort scale, 1–10: 1, no discomfort; 10, highest level of discomfort. All values are presented as mean ± standard error (SE) of the mean. Statistical Analyses After the training phase, the performance of the NDH was retested (untrained hand post). (B) Volunteers in MVFNDH received the same instructions and conducted the same task just vice versa: MVF-training was performed with the NDH while the DH was assessed before and after MVF (untrained hand pre and post). In both groups, volunteers rotated the balls with the NDH always in a counterclockwise direction and with the DH always in a clockwise direction. (C) During the task, volunteers in both groups were seated in a comfortable chair with their elbows flexed at 90◦and with their pronated hands rested on a desk in front of them. While MVF was provided, a mirror was placed in the subject’s midsagittalplane and the performing hand was covered by a wooden barrier to prevent direct view. See text for details. January 2016 | Volume 9 | Article 702 Frontiers in Human Neuroscience | www.frontiersin.org 3 Mirror Visual Feedback and Hand Dominance Rjosk et al. hand, a repeated-measures ANOVA (ANOVA-RM) with factor TRIAL (untrained hand pre vs. untrained hand post) and GROUP (MVFDH vs. MVFNDH) was performed. This analysis was supported by a second analysis, where absolute performance improvement (untrained hand post—untrained hand pre) of the untrained hand was compared across groups (MVFDH vs. MVFNDH) with an independent samples t-test. Furthermore, we performed a control analysis in order to ensure that volunteers improved with their trained hand during MVF. Hence, the number of ball-rotations/min of the trained hand in the first trial of training (T1) was compared between groups (MVFDH vs. MVFNDH) using an independent samples t-test. Trained hand performance over the whole training period was evaluated using another ANOVA-RM with factor TRIAL (T1–10) and GROUP (MVFDH vs. MVFNDH). If necessary, Greenhouse- Geisser correction was applied, and P-values were corrected for multiple comparisons using Bonferroni correction. A P-value of <0.05 was considered to be significant. As a measure of the effect size, the Eta-squared (η2) is reported for each ANOVA. As proposed by Miles and Shevlin (2001), a η2 of ≥0.02 is considered to be a small, a η2 of ≥0.13 a medium and a η2 of ≥0.26 a large effect. Behavioral data are presented as mean ± standard error (SE). the VAS in MVFNDH [t(15) = −2.423; P = 0.029]. Performance of the Untrained Hand Performance of the Untrained Hand There was no difference in baseline performance of the untrained hand between groups [MVFDH pre-MVF: 31.94 ± 3.19; MVFNDH pre-MVF: 33.50 ± 2.75 ball-rotations/min, t(30) = −0.371; P = 0.713; Figure 2A]. Performance of the untrained hand improved in both groups significantly by 9.44 ± 1.40 ball- rotations/min in MVFDH [t(15) = −6.730; P < 0.001] and by 6.19 ± 2.01 ball-rotations/min in MVFNDH [t(15) = −3.080; P = 0.008; Figure 2A]. There were no significant differences in behavioral improvements of the untrained hand after MVF between groups [ANOVA-RM with factor TRIAL (untrained TABLE 2 | Visual analog scale (VAS). Group Before After MVFDH Attention (1–10) 7.00 ± 0.27 8.13 ± 0.27 Fatigue (1–10) 6.88 ± 0.48 7.88 ± 0.41 Discomfort (1–10) 1.13 ± 0.13 1.00 ± 0.14 MVFNDH Attention (1–10) 8.19 ± 0.37 8.59 ± 0.34 Fatigue (1–10) 7.56 ± 0.38 7.75 ± 0.39 Discomfort (1–10) 1.19 ± 0.0 1.06 ± 0.06 Before and after the MVF-task, attention, fatigue and discomfort were assessed with the VAS questionnaire. Attention scale, 1–10: 1, no attention; 10, highest level of attention. Fatigue scale, 1–10: 1, high level of fatigue; 10, no fatigue. Discomfort scale, 1–10: 1, no discomfort; 10, highest level of discomfort. All values are presented as mean ± standard error (SE) of the mean. Statistical Analyses To exclude a correlation between the attention prior to the experiment and the performance improvement of the untrained hand, we performed a bivariate correlation and did not find a significant interaction [r = 0.10; P = 0.589]. We furthermore did not find a significant correlation between the change in attention and the performance improvement of the untrained hand [r = 0.12; P = 0.502] in a second bivariate correlation. See Table 2 for a complete breakdown of attention, fatigue and discomfort levels. Performance of the Trained Hand Performance of the Trained Hand In the first trial of MVF-training (T1), there was no significant difference in performance of the trained hand between groups [MVFDH: 23.13 ± 4.75; MVFNDH: 17.63 ± 3.41 ball- rotations/min, t(27.208) = 0.942; P = 0.355]. Performing the ball- rotation task with MVF during the training phase (trials 1–10) resulted in significant performance gains of the trained hand in both groups. On average participants improved by 23.19 ± 3.15 ball-rotations/min in MVFDH and by 17.06 ± 2.00 ball- rotations/min in MVFNDH [ANOVA-RM with factor TRIAL (T1–10): F(4.330,129.885) = 41.073; P < 0.001; η2 = 0.578] while there was no significant difference in the learning rate between groups [ANOVA-RM with factor TRIAL (T1–10) × GROUP (MVFDH vs. MVFNDH): F(4.330,129.885) = 1.070; P = 0.376; η2 = 0.034]. See Table 3 for details of group data. RESULTS TABLE 3 | Group data of the untrained hand pre-MVF and post-MVF and of the trained hand during learning phase (T1–T10) in the ball-rotation task. pre-MVF T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 post-MVF MVFDH 31.94 ± 3.19 23.13 ± 4.75 29.38 ± 4.52 34.88 ± 4.20 37.44 ± 3.48 37.88 ± 3.48 39.88 ± 3.36 41.69 ± 3.41 43.63 ± 3.48 44.63 ± 3.24 46.31 ± 3.33 41.38 ± 3.49 MVFNDH 33.50 ± 2.75 17.63 ± 3.41 21.19 ± 3.89 25.00 ± 3.65 26.88 ± 4.02 25.56 ± 3.51 27.56 ± 3.14 31.13 ± 3.50 32.88 ± 3.17 33.88 ± 3.62 34.69 ± 2.87 39.69 ± 2.75 Behavioral data for the trained and untrained hand. Performance of the untrained hand before and after MVF-training (trials 1–10). Performing the ball-rotation task with MVF during learning phase (trials 1–10) resulted in significant performance gains of the trained hand in both groups while there was no difference in the learning rate between groups. For details, see text. Data are presented as mean ± standard error (SE) of the mean. January 2016 | TABLE 3 | Group data of the untrained hand pre-MVF and post-MVF and of the trained hand during learning phase (T1–T10) in the ball-rotation task. pre-MVF T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 post-MVF MVFDH 31.94 ± 3.19 23.13 ± 4.75 29.38 ± 4.52 34.88 ± 4.20 37.44 ± 3.48 37.88 ± 3.48 39.88 ± 3.36 41.69 ± 3.41 43.63 ± 3.48 44.63 ± 3.24 46.31 ± 3.33 41.38 ± 3.49 MVFNDH 33.50 ± 2.75 17.63 ± 3.41 21.19 ± 3.89 25.00 ± 3.65 26.88 ± 4.02 25.56 ± 3.51 27.56 ± 3.14 31.13 ± 3.50 32.88 ± 3.17 33.88 ± 3.62 34.69 ± 2.87 39.69 ± 2.75 Behavioral data for the trained and untrained hand. Performance of the untrained hand before and after MVF-training (trials 1–10). Performing the ball-rotation task with MVF during learning phase (trials 1–10) resulted in significant performance gains of the trained hand in both groups while there was no difference in the learning rate between groups. For details, see text. Data are presented as mean ± standard error (SE) of the mean. hand pre vs. untrained hand post) × GROUP (MVFDH vs. MVFNDH): F(1,30) = 1.760; P = 0.195; η2 = 0.055]. RESULTS A comparison of the absolute amount of performance improvement showed no significant difference between groups [t(30) = 1.327; P = 0.195; Figure 2B]. There was no correlation between the absolute amount of performance improvement of the untrained hand (post-MVF −pre-MVF) and the absolute amount of performance improvement of the trained hand (T10−T1) in neither of the groups [MVFDH: r = 0.20; P = 0.461; MVFNDH: r = −0.28; P = 0.301]. Frontiers in Human Neuroscience | www.frontiersin.org RESULTS FIGURE 2 | Effect of MVF on motor performance of the untrained hand. (A) Number of ball-rotations/min of the untrained hand before and after MVF. Note that baseline performance of the untrained hand did not differ between groups. (B) Absolute performance improvement of the untrained hand. Both groups improved their performance with the untrained hand significantly. There was no significant difference in absolute performance improvement of the untrained hand between groups. The plot shows mean values, and whiskers represent standard error (SE) values. ∗P < 0.05; n/s, not significant. FIGURE 2 | Effect of MVF on motor performance of the untrained hand. (A) Number of ball-rotations/min of the untrained hand before and after MVF. Note that baseline performance of the untrained hand did not differ between groups. (B) Absolute performance improvement of the untrained hand. Both groups improved their performance with the untrained hand significantly. There was no significant difference in absolute performance improvement of the untrained hand between groups. The plot shows mean values, and whiskers represent standard error (SE) values. ∗P < 0.05; n/s, not significant. January 2016 | Volume 9 | Article 702 Frontiers in Human Neuroscience | www.frontiersin.org 4 Mirror Visual Feedback and Hand Dominance Rjosk et al. TABLE 3 | Group data of the untrained hand pre-MVF and post-MVF and of the trained hand during learning phase (T1–T10) in the ball-rotation task. pre-MVF T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 post-MVF MVFDH 31.94 ± 3.19 23.13 ± 4.75 29.38 ± 4.52 34.88 ± 4.20 37.44 ± 3.48 37.88 ± 3.48 39.88 ± 3.36 41.69 ± 3.41 43.63 ± 3.48 44.63 ± 3.24 46.31 ± 3.33 41.38 ± 3.49 MVFNDH 33.50 ± 2.75 17.63 ± 3.41 21.19 ± 3.89 25.00 ± 3.65 26.88 ± 4.02 25.56 ± 3.51 27.56 ± 3.14 31.13 ± 3.50 32.88 ± 3.17 33.88 ± 3.62 34.69 ± 2.87 39.69 ± 2.75 Behavioral data for the trained and untrained hand. Performance of the untrained hand before and after MVF-training (trials 1–10). Performing the ball-rotation task with MVF during learning phase (trials 1–10) resulted in significant performance gains of the trained hand in both groups while there was no difference in the learning rate between groups. For details, see text. Data are presented as mean ± standard error (SE) of the mean. January 2016 | Volume 9 | Article 702 DISCUSSION The aim of the present study was to investigate whether MVF from the DH and NDH during motor skill learning differentially affects performance of the untrained hand. Contrary to our hypothesis, there was no significant difference in baseline performance of the untrained hand between groups before MVF. Both groups improved the dexterity of the untrained hand significantly and there was no significant difference in the amount of performance improvement. Moreover, there was no significant difference in the learning rate of the trained hand during the training phase with MVF. Our results seem to be in contrast with other studies showing worse performance of the NDH in motor-tasks like finger tapping or the pegboard task (Triggs et al., 1997; Brouwer et al., 2001; Garry et al., 2004). One potential explanation for these divergent results might be related to the fact that the motor-task in the present study (ball-rotation task) is a more complex and attentionally demanding motor skill that was completely novel to participants. In line with this, several other studies have argued that the dominant and non-dominant arm have complementary roles during complex motor skill tasks, with the dominant arm specializing in specification and control of arm/joint trajectory and the non-dominant arm preferentially encoding sensory-mediated error correction (Sainburg and Kalakanis, 2000; Sainburg and Wang, 2002; Bagesteiro and Sainburg, 2003). We showed that the untrained hand improved significantly, irrespective if volunteers used their DH or NDH hand for the MVF-task. Hence, MVF-induced performance improvements do Frontiers in Human Neuroscience | www.frontiersin.org January 2016 | Volume 9 | Article 702 5 Mirror Visual Feedback and Hand Dominance Rjosk et al. not seem to be affected by hand dominance, at least in right- handed volunteers. find any alterations in IHI after MFV was applied. Furthermore, they showed that even in callosotomized patients, MVF-induced performance improvements were still observable (Nojima et al., 2013). Hence, MVF as compared to intermanual transfer may rely on different underlying mechanisms. This, however, needs to be further investigated in future studies, since at least one study indicated potential alterations in IHI when MVF was provided for a simple paced finger tapping movement (Avanzino et al., 2014). The question remains whether MVF-induced performance improvements of the untrained hand are due to intermanual transfer or other unknown factors. Future Implications and Summary Future Implications and Summary We here provide novel evidence that MVF from the DH as well as NDH is capable of improving the dexterity of the untrained hand in a complex fine-motor task. In the present study, we investigated whether the behavioral effect of MVF differs according to hand dominance. Future studies should explore potential differences or similarities between the directionality of MVF on a neurophysiological level. Furthermore, future research should investigate the observed behavioral effects of MVF in left-handed volunteers, as well, to see if hand dominance affects their dexterity, differently. Since our volunteers only conducted the task once, we cannot exclude the possibility that hand dominance may induce differences in the amount of MVF-induced performance improvement after several sessions of MVF. Since MVF is successfully used in the context of neurorehabilitation as an adjuvant strategy to augment performance in the paretic arm after focal brain lesion (Altschuler et al., 1999; Yavuzer et al., 2008; Dohle et al., 2009) and to reduce pain in patients with complex regional pain syndrome (Moseley, 2004), our findings might have important implications from a clinical perspective and support the application of MVF in patients regardless of the dominance of the affected limb. FUNDING The work of EK is funded by the Fazit Stiftung GmbH, Frankfurt. AUTHOR CONTRIBUTIONS VR and PR and AV designed the study. VR performed the experiment. PR, VR, CJS analyzed the data. VR, EK, MH, BS, CJS, PR interpreted results of the experiment. PR and VR drafted the manuscript. PR, VR, EK, MH, CJS and BS edited and revised the manuscript. All authors approved the final version of the manuscript. Apart from that, the underlying neural mechanisms of MVF and intermanual transfer seem to be divergent: for example, Perez et al. (2007) and Camus et al. (2009) showed that alterations in intracortical and interhemispheric inhibition (IHI) between homologous M1s predominantly contribute to intermanual transfer. On the other hand, Nojima et al. (2012) could not DISCUSSION Intermanual transfer, a phenomenon where unilateral skill training improves not only the trained, but also the untrained hand (Obayashi, 2004; Perez et al., 2007; Camus et al., 2009) is well described for different motor paradigms. However, studies reported conflicting results concerning the directionality of transfer between the DH and the NDH. For example, studies showed a symmetric transfer to both directions (Imamizu and Shimojo, 1995; van Mier and Petersen, 2006) as well as a greater transfer from the NDH to the DH (Taylor and Heilman, 1980; Parlow and Kinsbourne, 1990; Lavrysen et al., 2003) while other studies showed the reverse phenomenon (Parlow and Dewey, 1991; Halsband, 1992; Thut et al., 1996; Criscimagna-Hemminger et al., 2003; Redding and Wallace, 2008). Moreover, other studies have shown that some aspects of the same visuo-motor task transferred only in one direction while others in the other direction (Sainburg and Wang, 2002). The diversity in the literature seems to reflect the complexity of the phenomenon of intermanual transfer and suggests that there is a dependency between a- and/or symmetry and the task and paradigm used. Concerning intermanual transfer, one could argue that MVF is not the driving mechanism behind the observed performance improvements in the untrained hand. We believe, however, that pure intermanual transfer cannot explain the observed MVF- induced behavioral improvement. In favor of this, Nojima et al. (2012) performed a control experiment with the same complex ball-rotation task and showed no performance improvements of the untrained left hand when motor-training with the right hand was performed without MVF. Interestingly, a recent study by Reissig et al. (2015) showed divergent findings. Here, the authors found no difference between a MVF group and a group that received no MVF during training. Obvious reasons behind these opposing findings need to be addressed in future studies. However, as pointed out by Reissig et al. (2015) one obvious explanation might be related to the fact that the kinesthetic illusion for MVF, which seems to be important for the observed behavioral effects, might have been different between studies. REFERENCES Bagesteiro, L. B., and Sainburg, R. L. (2003). Nondominant arm advantages in load compensation during rapid elbow joint movements. J. Neurophysiol. 90, 1503–1513. doi: 10.1152/jn.00189.2003 Altschuler, E. L., Wisdom, S. B., Stone, L., Foster, C., Galasko, D., Llewellyn, D. M., et al. (1999). Rehabilitation of hemiparesis after stroke with a mirror. Lancet 353, 2035–2036. doi: 10.1016/s0140-6736(99) 00920-4 Brouwer, B., Sale, M. V., and Nordstrom, M. A. (2001). Asymmetry of motor cortex excitability during a simple motor task: relationships with handedness and manual performance. Exp. Brain Res. 138, 467–476. doi: 10. 1007/s002210100730 Avanzino, L., Raffo, A., Pelosin, E., Ogliastro, C., Marchese, R., Ruggeri, P., et al. (2014). Training based on mirror visual feedback influences transcallosal communication. Eur. J. Neurosci. 40, 2581–2588. doi: 10.1111/ejn.12615 Camus, M., Ragert, P., Vandermeeren, Y., and Cohen, L. G. (2009). Mechanisms controlling motor output to a transfer hand after learning a sequential pinch Frontiers in Human Neuroscience | www.frontiersin.org Frontiers in Human Neuroscience | www.frontiersin.org January 2016 | Volume 9 | Article 702 6 Mirror Visual Feedback and Hand Dominance Rjosk et al. force skill with the opposite hand. Clin. Neurophysiol. 120, 1859–1865. doi: 10. 1016/j.clinph.2009.08.013 force skill with the opposite hand. Clin. Neurophysiol. 120, 1859–1865. doi: 10. 1016/j.clinph.2009.08.013 Parlow, S. E., and Dewey, D. (1991). The temporal locus of transfer of training between hands: an interference study. Behav. Brain Res. 46, 1–8. doi: 10. 1016/s0166-4328(05)80091-9 Criscimagna-Hemminger, S. E., Donchin, O., Gazzaniga, M. S., and Shadmehr, R. (2003). Learned dynamics of reaching movements generalize from dominant to nondominant arm. J. Neurophysiol. 89, 168–176. doi: 10.1152/jn.00622.2002 Parlow, S. E., and Kinsbourne, M. (1990). Asymmetrical transfer of braille acquisition between hands. Brain Lang. 39, 319–330. doi: 10.1016/0093- 934x(90)90017-b Dohle, C., Pullen, J., Nakaten, A., Küst, J., Rietz, C., and Karbe, H. (2009). Mirror therapy promotes recovery from severe hemiparesis: a randomized controlled trial. Neurorehabil. Neural Repair 23, 209–217. doi: 10.1177/1545968308324786 Perez, M. A., Tanaka, S., Wise, S. P., Sadato, N., Tanabe, H. C., Willingham, D. T., et al. (2007). Neural substrates of intermanual transfer of a newly acquired motor skill. Curr. Biol. 17, 1896–1902. doi: 10.1016/j.cub.2007.09.058 Fukumura, K., Sugawara, K., Tanabe, S., Ushiba, J., and Tomita, Y. (2007). Influence of mirror therapy on human motor cortex. Int. J. Neurosci. 117, 1039–1048. doi: 10.1080/00207450600936841 Ramachandran, V. S., Rogers-Ramachandran, D., and Cobb, S. (1995). Touching the phantom limb. Nature 377, 489–490. doi: 10.1038/377489a0 Garry, M. I., Kamen, G., and Nordstrom, M. REFERENCES A. (2004). Hemispheric differences in the relationship between corticomotor excitability changes following a fine- motor task and motor learning. J. Neurophysiol. 91, 1570–1578. doi: 10.1152/jn. 00595.2003 Redding, G. M., and Wallace, B. (2008). Intermanual transfer of prism adaptation. J. Mot. Behav. 40, 246–262. doi: 10.3200/JMBR.40.3.246-264 Reissig, P., Puri, R., Garry, M. I., Summers, J. J., and Hinder, M. R. (2015). The influence of mirror-visual feedback on training-induced motor performance gains in the untrained hand. PLoS One 10:e0141828. doi: 10.1371/journal.pone. 0141828 Garry, M. I., Loftus, A., and Summers, J. J. (2005). Mirror, mirror on the wall: viewing a mirror reflection of unilateral hand movements facilitates ipsilateral M1 excitability. Exp. Brain Res. 163, 118–122. doi: 10.1007/s00221-005-2226-9 Ridding, M. C., and Flavel, S. C. (2006). Induction of plasticity in the dominant and non-dominant motor cortices of humans. Exp. Brain Res. 171, 551–557. doi: 10.1007/s00221-005-0309-2 Goble, D. J., and Brown, S. H. (2008). The biological and behavioral basis of upper limb asymmetries in sensorimotor performance. Neurosci. Biobehav. Rev. 32, 598–610. doi: 10.1016/j.neubiorev.2007.10.006 Sainburg, R. L., and Kalakanis, D. (2000). Differences in control of limb dynamics during dominant and nondominant arm reaching. J. Neurophysiol. 83, 2661–2675. Halsband, U. (1992). Left hemisphere preponderance in trajectorial learning. Neuroreport 3, 397–400. doi: 10.1097/00001756-199205000-00005 Hamzei, F., Lappchen, C. H., Glauche, V., Mader, I., Rijntjes, M., and Weiller, C. (2012). Functional plasticity induced by mirror training: the mirror as the element connecting both hands to one hemisphere. Neurorehabil. Neural Repair 26, 484–496. doi: 10.1177/1545968311427917 Sainburg, R. L., and Wang, J. (2002). Interlimb transfer of visuomotor rotations: independence of direction and final position information. Exp. Brain Res. 145, 437–447. doi: 10.1007/s00221-002-1140-7 Taylor, H. G., and Heilman, K. M. (1980). Left-hemisphere motor dominance in righthanders. Cortex 16, 587–603. doi: 10.1016/s0010-9452(80)80006-2 Harris, J. E., and Eng, J. J. (2006). Individuals with the dominant hand affected following stroke demonstrate less impairment than those with the nondominant hand affected. Neurorehabil. Neural Repair 20, 380–389. doi: 10. 1177/1545968305284528 Thut, G., Cook, N. D., Regard, M., Leenders, K. L., Halsband, U., and Landis, T. (1996). Intermanual transfer of proximal and distal motor engrams in humans. Exp. Brain Res. 108, 321–327. doi: 10.1007/bf00228105 Hoff, M., Kaminski, E., Rjosk, V., Sehm, B., Steele, C. J., Villringer, A., et al. (2015). Augmenting mirror visual feedback-induced performance improvements in older adults. Eur. J. Neurosci. 41, 1475–1483. doi: 10.1111/ejn.12899 Todor, J. I., and Kyprie, P. M. (1980). REFERENCES Hand differences in the rate and variability of rapid tapping. J. Mot. Behav. 12, 57–62. doi: 10.1080/00222895.1980.10735205 Triggs, W. J., Calvanio, R., and Levine, M. (1997). Transcranial magnetic stimulation reveals a hemispheric asymmetry correlate of intermanual differences in motor performance. Neuropsychologia 35, 1355–1363. doi: 10. 1016/s0028-3932(97)00077-8 Imamizu, H., and Shimojo, S. (1995). The locus of visual-motor learning at the task or manipulator level: implications from intermanual transfer. J. Exp. Psychol. Hum. Percept. Perform. 21, 719–733. doi: 10.1037/0096-1523.21.4.719 van Mier, H. I., and Petersen, S. E. (2006). Intermanual transfer effects in sequential tactuomotor learning: evidence for effector independent coding. Neuropsychologia 44, 939–949. doi: 10.1016/j.neuropsychologia.2005.08.010 Lappchen, C. H., Ringer, T., Blessin, J., Seidel, G., Grieshammer, S., Lange, R., et al. (2012). Optical illusion alters M1 excitability after mirror therapy: a TMS study. J. Neurophysiol. 108, 2857–2861. doi: 10.1152/jn.00321.2012 von Rein, E., Hoff, M., Kaminski, E., Sehm, B., Steele, C. J., Villringer, A., et al. (2015). Improving motor performance without training: the effect of combining mirror visual feedback with transcranial direct current stimulation. J. Neurophysiol. 113, 2383–2389. doi: 10.1152/jn.00832.2014 Lavrysen, A., Helsen, W. F., Tremblay, L., Elliott, D., Adam, J. J., Feys, P., et al. (2003). The control of sequential aiming movements: the influence of practice and manual asymmetries on the one-target advantage. Cortex 39, 307–325. doi: 10.1016/s0010-9452(08)70111-4 y j Yavuzer, G., Selles, R., Sezer, N., Sütbeyaz, S., Bussmann, J. B., Köseo˘glu, F., et al. (2008). Mirror therapy improves hand function in subacute stroke: a randomized controlled trial. Arch. Phys. Med. Rehabil. 89, 393–398. doi: 10. 1016/j.apmr.2007.08.162 Miles, J., and Shevlin, M. (2001). Applying Regression and Correlation : A Guide for Students and Researchers. Thousand Oaks, CA: Sage Publications. Moseley, G. L. (2004). Graded motor imagery is effective for long-standing complex regional pain syndrome: a randomised controlled trial. Pain 108, 192–198. doi: 10.1016/j.pain.2004.01.006 Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Nojima, I., Mima, T., Koganemaru, S., Thabit, M. N., Fukuyama, H., and Kawamata, T. (2012). Human motor plasticity induced by mirror visual feedback. J. Neurosci. 32, 1293–1300. doi: 10.1523/JNEUROSCI.5364-11.2012 Nojima, I., Oga, T., Fukuyama, H., Kawamata, T., and Mima, T. (2013). Mirror visual feedback can induce motor learning in patients with callosal disconnection. Exp. Brain Res. 227, 79–83. January 2016 | Volume 9 | Article 702 REFERENCES doi: 10.1007/s00221-013-3486-4 Copyright © 2016 Rjosk, Kaminski, Hoff, Sehm, Steele, Villringer and Ragert. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Obayashi, S. (2004). Possible mechanism for transfer of motor skill learning: implication of the cerebellum. Cerebellum 3, 204–211. doi: 10. 1080/14734220410018977 Oldfield, R. C. (1971). The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97–113. doi: 10.1016/0028-3932(71)90067-4 January 2016 | Volume 9 | Article 702 Frontiers in Human Neuroscience | www.frontiersin.org
https://openalex.org/W1903589011
https://ueaeprints.uea.ac.uk/id/eprint/71225/1/Published_Version.pdf
English
null
The effect of genetic structure on molecular dating and tests for temporal signal
Methods in ecology and evolution
2,015
cc-by
10,605
Summary This is not an artefact of genetic structure or tree shape per se, and can arise even when sequences have measurably evolved during the sampling period. More positively, we show that a ‘clustered permutation’ approach introduced by Duch^ene et al. (Molecular Biology and Evolution, 32, 2015, 1895) can successfully cor- rect for this artefact in all cases and introduce techniques for implementing this method with real data sets. 4. The confounding of temporal and genetic structures may be difficult to avoid in practice, particularly for out- breaks of infectious disease, or when using ancient DNA. Therefore, we recommend the use of ‘clustered permu- tation’ for all analyses. The failure of the standard tests may explain why different methods of dating pathogen origins have reached such wildly different conclusions. Key-words: Bayesian dating, dated-tips, pathogen origins, permutation tests, Staphylococcus aureus too short for sufficient evolutionary change to occur or if evo- lutionary rates were too variable (Drummond, Pybus & Ram- baut 2003b; Firth et al. 2010; Duch^ene et al. 2015a). However, evolutionary rates are often unknown, and molecular dating methods will usually converge on an estimate whether or not temporal signal is present (Firth et al. 2010). As such, it is cru- cial to test the molecular data for temporal signal. The effect of genetic structure on molecular dating and tests for temporal signal Gemma G. R. Murray1*, Fang Wang1, Ewan M. Harrison2, Gavin K. Paterson2,3, Alison E. Mather2,4, Simon R. Harris4, Mark A. Holmes2, Andrew Rambaut5 and John J. Welch1 1Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, UK; 2Department of Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES, UK; 3School of Biological, Biomedical and Environmental Sciences, University of Hull, Cottingham Road, Hull HU6 7RX, UK; 4Wellcome Trust Sanger Institute, Hinxton CB10 1SA, UK; and 5Institute of Evolutionary Biology, University of Edinburgh, King’s Buildings, Edinburgh EH9 3FL, UK 1Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, UK; 2Department of Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES, UK; 3School of Biological, Biomedical and Environmental Sciences, University of Hull, Cottingham Road, Hull HU6 7RX, UK; 4Wellcome Trust Sanger Institute, Hinxton CB10 1SA, UK; and 5Institute of Evolutionary Biology, University of Edinburgh, King’s Buildings, Edinburgh EH9 3FL, UK © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. doi: 10.1111/2041-210X.12466 Methods in Ecology and Evolution 2016, 7, 80–89 Summary 1. ‘Dated-tip’ methods of molecular dating use DNA sequences sampled at different times, to estimate the age of their most recent common ancestor. Several tests of ‘temporal signal’ are available to determine whether data sets are suitable for such analysis. However, it remains unclear whether these tests are reliable. 2. We investigate the performance of several tests of temporal signal, including some recently suggested modifi- cations. We use simulated data (where the true evolutionary history is known), and whole genomes of methi- cillin-resistant Staphylococcus aureus (to show how particular problems arise with real-world data sets). 3. We show that all of the standard tests of temporal signal are seriously misleading for data where temporal and genetic structures are confounded (i.e. where closely related sequences are more likely to have been sampled at similar times). This is not an artefact of genetic structure or tree shape per se, and can arise even when sequences have measurably evolved during the sampling period. More positively, we show that a ‘clustered permutation’ approach introduced by Duch^ene et al. (Molecular Biology and Evolution, 32, 2015, 1895) can successfully cor- rect for this artefact in all cases and introduce techniques for implementing this method with real data sets. 4. The confounding of temporal and genetic structures may be difficult to avoid in practice, particularly for out- breaks of infectious disease, or when using ancient DNA. Therefore, we recommend the use of ‘clustered permu- tation’ for all analyses. The failure of the standard tests may explain why different methods of dating pathogen origins have reached such wildly different conclusions. 1. ‘Dated-tip’ methods of molecular dating use DNA sequences sampled at different times, to estimate the age of their most recent common ancestor. Several tests of ‘temporal signal’ are available to determine whether data sets are suitable for such analysis. However, it remains unclear whether these tests are reliable. 2. We investigate the performance of several tests of temporal signal, including some recently suggested modifi- cations. We use simulated data (where the true evolutionary history is known), and whole genomes of methi- cillin-resistant Staphylococcus aureus (to show how particular problems arise with real-world data sets). 3. We show that all of the standard tests of temporal signal are seriously misleading for data where temporal and genetic structures are confounded (i.e. where closely related sequences are more likely to have been sampled at similar times). The effect of genetic structure on molecular dating and tests for temporal signal Gemma G. R. Murray1*, Fang Wang1, Ewan M. Harrison2, Gavin K. Paterson2,3, Alison E. Mather2,4, Simon R. Harris4, Mark A. Holmes2, Andrew Rambaut5 and John J. Welch1 1Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, UK; 2Department of Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES, UK; 3School of Biological, Biomedical and Environmental Sciences, University of Hull, Cottingham Road, Hull HU6 7RX, UK; 4Wellcome Trust Sanger Institute, Hinxton CB10 1SA, UK; and 5Institute of Evolutionary Biology, University of Edinburgh, King’s Buildings, Edinburgh EH9 3FL, UK Methods in Ecology and Evolution 2016, 7, 80–89 doi: 10.1111/2041-210X.12466 The effect of genetic structure on molecular dating and tests for temporal signal Gemma G. R. Murray1*, Fang Wang1, Ewan M. Harrison2, Gavin K. Paterson2,3, Alison E. Mather2,4, Simon R. Harris4, Mark A. Holmes2, Andrew Rambaut5 and John J. Welch1 1Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, UK; 2Department of Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES, UK; 3School of Biological, Biomedical and Environmental Sciences, University of Hull, Cottingham Road, Hull HU6 7RX, UK; 4Wellcome Trust Sanger Institute, Hinxton CB10 1SA, UK; and 5Institute of Evolutionary Biology, University of Edinburgh, King’s Buildings, Edinburgh EH9 3FL, UK Methods in Ecology and Evolution 2016, 7, 80–89 doi: 10.1111/2041-210X.12466 dent, but Navascues, Depaulis & Emerson (2010) suggest *Correspondence author. E-mail: ggrm2@cam.ac.uk © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. *Correspondence author. E-mail: ggrm2@cam.ac.uk BASIC DATING ANALYSES We estimated the tMRCA for all of our data sets using BEAST v1.8 (Drum- mond et al. 2012). In all cases, we used a constant population size coa- lescent prior for the node ages, and (except for Bayes factor calculations) the BEAUti v1.8 default priors for all other parameters (Drummond et al. 2012). After each run, convergence was assessed using TRACER v1.6 (Rambaut et al. 2014) and burn-in removed as required. For the tMRCA, we recorded the maximum a posteriori (MAP) estimate, estimated from the MCMC using the Venter mode estimator from the R package modeest (Venter 1967; Poncet 2012), and the 95% highest posterior density (HPD) interval. Finally, a distinct approach uses model selection and com- pares the fit of models with the sampling dates included or excluded, thereby failing to take special account for any evolu- tion that might have taken place during the sampling period (Rambaut 2000; Drummond, Pybus & Rambaut 2003b; Drummond et al. 2003a; Baele et al. 2012). Temporal signal is confirmed if the inclusion of the sampling dates improves the fit. All of the approaches above are widely used, but it is not clear how well they identify temporal signal, especially if we define temporal signal as the ability of a data set to yield reli- able date estimates. Previous studies have shown that dated-tip methods can be unreliable not only for data with too short a sampling period or too variable an evolutionary rate, but also for data with strong population structure (Navascues & Emer- son 2009) or imbalanced trees (Duch^ene, Duch^ene & Ho 2015b). Furthermore, Duch^ene et al. (2015a) showed that the Bayesian permutation test gave false evidence of temporal sig- nal for simulated data where the sampling period was too short, but where clusters of closely related sequences were sam- pled at the same time, that is where temporal and genetic struc- tures were confounded. To solve this problem, they introduced a ‘clustered permutation approach’ where dates were ran- domly reassigned among clusters of sequences sampled on the same date. For the simulated data sets, we fit the same evolutionary model that was used to simulate the data, namely the HKY+Γ substitution model and a strict molecular clock. For the reanalysis of the S. aureus data, we also used the HKY+Γ substitution model. For the data from Holden et al. Introduction Molecular dating uses evolutionary change between homolo- gous DNA sequences to infer the time since their most recent common ancestor (tMRCA). If the genomes were sampled at similar times, then this inference requires external temporal information, such as a known rate of evolution, to calibrate the molecular clock. But if the genomes were sampled at suffi- ciently different times, then the sampling dates are all the tem- poral information required (Rambaut 2000; Drummond, Pybus & Rambaut 2003b; Drummond et al. 2003a). Such ‘dated-tip’ methods have been particularly useful in the study of viral and bacterial pathogens and have been used to under- stand the origins and spread of diseases, as well as transmission pathways within a single outbreak (e.g. Smith et al. 2009; Didelot et al. 2012; McAdam et al. 2012; Gire et al. 2014). Several approaches have been used to test for temporal sig- nal. The simplest is a linear regression of phylogenetic root- to-tip distance against sampling date (Buonagurio et al. 1986; Shankarappa et al. 1999; Korber et al. 2000; Drummond, Pybus & Rambaut 2003b). If sampling dates are sufficiently different, then more recently sampled sequences will have undergone substantially more evolutionary change than earlier sampled sequences, and this should create a strong positive correlation. This test obviously requires a rooted phylogeny, and when the root is unknown, it is common to estimate the root simultaneously with the regression, so as to maximize the model fit (Drummond et al. 2003a). Significance is not gener- ally calculated, because root-to-tip distances are non-indepen- dent, but Navascues, Depaulis & Emerson (2010) suggest Dated-tip methods are only valid if there is temporal signal in the data. This will not be the case if the sampling period was © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Temporal signal in structured molecular data 8 81 using permutation, asking whether the correlation is stronger than expected if the sampling dates were assigned to sequences at random. confounding can arise naturally from clinical sampling practice, suggesting that the unreliable date estimates may be widespread. Regression test To regress phylogenetic root-to-tip distance against sampling date, we obtained crude root-to-tip distances from a neighbour-joining tree esti- mated using a K80 nucleotide substitution model with the APE package in R (Paradis, Claude & Strimmer 2004). Following the PATH-O-GEN software (Rambaut 2013), the root was fit simultaneously with the regression, so as to minimize the residual mean squares (see also Korber et al. 2000). Following the suggestion of Navascues, Depaulis & Emerson (2010), the significance of the regression was assessed by random permutation of the sampling dates over the sequences, using the correlation coefficient as the test statistic. For all reported results, we generated 1000 replicates of the data, with the sampling dates ran- domly permuted. The P-value is the proportion of replicates with a test statistic greater than or equal to the true value. The null hypothesis is that a negligible amount of evolution took place between the sampling dates, so that the correlation observed can be attributed to stochastic variation in molecular branch length estimates and (when the root is not known independently) to our having rooted the tree to maximize clocklikeness. Here, we investigate the performance of tests of temporal signal on a variety of simulated and real-world structured data sets. We show that while structured data can generate accurate estimates of the tMRCA with dated-tip methods, when temporal and genetic structures are confounded, esti- mates are consistently misleading, regardless of the level of temporal structure in the data. We further show that the standard tests of temporal signal fail to identify data sets that result in unreliable estimates when temporal and genetic structures are confounded. We demonstrate that the clustered permutation approach of Duch^ene et al. (2015a) can be applied to both the regression and Bayesian tests for temporal signal, and that it successfully identifies those data sets that give reliable estimates in the presence of con- founded genetic structure. Finally, through analysis of two sets of whole-genome data from Staphylococcus aureus, with very different sampling periods, we develop methods of applying these tests to real data and show that Introduction Linear regression is a crude method of molecular dating, but analogous tests can be used with more formal methods. Most commonly, the tMRCA or rate estimate from a Bayesian dated- tip analysis is used as the test statistic. If more recently sampled sequences have undergone more molecular evolution, then the true sampling dates should yield a tMRCA that differs substan- tially from the equivalent estimates with the sampling dates randomly permuted over sequences (Ramsden, Holmes & Charleston 2009; e.g. Duffy & Holmes 2009; Firth et al. 2010; Fraile et al. 2011; Pagan & Holguın 2013; Duch^ene, Holmes & Ho 2014b; Duch^ene et al. 2015a). DATA SETS Details of our simulated and real data sets are provided in the Support- ing information. BASIC DATING ANALYSES (2013), we used the uncorrelated log-normal relaxed molecular clock (replicating the published analysis), whereas for the data from Paterson et al. (2015) we used a strict clock due to the small number of variable sites. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 Clustered permutation tests The permutation tests described above assume that the sampling dates are exchangeable under the null. This will not be true if closely related sequences were preferentially sampled at the same date. A heuristic approach to dealing with this artefact was introduced by Duch^ene et al. (2015a). Their approach is to randomize dates over clusters of sequences, rather than individual sequences. Clusters are defined as monophyletic clades, which were sampled at the same time. If we have n clusters, then the maximum number of permutations of these clusters is n!, and if each sampling date is associated in more than one of the clusters, then the total number of unique permutations is n! Qdates i m1 i where mi is the number of clusters associated with sampling date i. When this number is suitably small, it is easiest to generate all possible permutations. For example, in Fig. 1c,d, there were only 3! = 6 possi- ble permutations, which made 1/6 the smallest possible P-value for these extreme cases. For the simulated data, we identified single-date clusters from a neighbour-joining tree, rooted to minimize the residual mean squares of a linear regression of sampling time against root-to-tip distance (see Results for definitions for real data). All tests were implemented in R scripts (R Core Team 2014), which are provided in the Supporting information. We first simulated data with a high degree of temporal struc- ture, by selecting three sampling dates such that an average of 20 nucleotide substitutions per genome occurred between each sampling. We also assumed a ‘balanced’ sampling scheme, such that all three genetic clades were sampled equally thor- oughly on all three dates. With this high temporal structure, and balanced sampling, the dating was a success. When corre- lating root-to-tip distance with sampling dates, all of the 1000 simulated data sets showed the signature of temporal signal (see Fig. S1a for a histogram of r-values). Figure 1a shows a detailed analysis of a single typical replicate, with a permuta- tion test, confirming that the correlation was unlikely to have arisen by chance (Fig. 1a, middle column; see also Table S1); indeed, for these simulated data, variation around the regres- sion line must be attributed to stochastic variation in the sub- stitution process, or to estimation error in the branch lengths. The intercept of the regression was also similar to the true tMRCA used to simulate the data. Tests of model fit A final test of temporal signal is to compare some measure of model fit for phylogenetic analyses with or without sampling dates (Baele et al. 2012). In practice, for the ‘no dates’ model, to keep the two cases as sim- ilar as possible, we set all sequences to the most recent sampling date in the original data set. We compared two model comparison statistics. The AICM is computationally cheap and robust to specification of improper priors. It can also be transformed into a true hypothesis test, using Akaike weights (Burnham & Anderson 2002). To do this, when the ‘with dates’ model was preferred, the relative support for this model, equivalent to the P-value, was calculated as w ¼ 1 þ e 1 2DAICM  1 , where DAICM is the improvement in the fit. AICM was estimated in TRACER v1.6 (Rambaut et al. 2014), which was also used to check convergence. We next simulated data with the same balanced sampling, but little temporal structure, that is with sampling dates that were so close that only 02 substitutions per genome were expected between them (Fig. 1b). In this case, tMRCA estimates were highly inaccurate, but tests of temporal signal correctly indicated that these estimates could not be trusted. In particu- lar, none of the 1000 data sets gave high r-values (Fig. S1b), and tests confirmed that similar results could be obtained after randomly permuting the sampling dates. Therefore, with bal- anced sampling (Fig. 1a,b), tests of temporal signal perform well. To calculate Bayes factors, we used the path sampling approach with 100 steps, as implemented in BEAST v.2 (Baele et al. 2012; Bouckaert et al. 2014). The method relies on the specification of priors that are proper (integrating to unity), and not too diffuse (Baele et al. 2012). © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 DATING ARTEFACTS WITH SIMULATED DATA To illustrate the performance of tests of temporal signal on genetically structured samples, we simulated molecular data sampled on three different dates, from a highly structured pop- ulation, consisting of three distinct and equally related clades, whose most recent common ancestor lived 10 000 years before the present (ybp), evolving at a comparable rate to some bacte- ria and viruses (16 9 106 subs per site per year). We applied the standard tests of temporal signal to these simulated data and estimated their tMRCA (Fig. 1). Clustered permutation tests Bayesian molecular dating with BEAST (Drummond et al. 2012) also performed well (Fig. 1a, right-hand column): the tMRCA estimate (red point) was accurate and precise, and also highly outlying when com- pared with replicate analyses with sampling dates randomly permuted (purple points). Bayesian dating permutation tests To test for temporal signal using Bayesian dating, each analysis was repeated 10 times, after randomly permuting the sampling dates across sequences (e.g. Ramsden, Holmes & Charleston 2009). We then asked © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 82 G. G. R. Murray et al. 82 whether the tMRCA estimate from the true data was outlying when com- pared with the estimates from the randomly permuted data. This is not standard hypothesis test, since each of the 11 estimates is associated with uncertainty, and therefore, P-values were not calculated. We note that the choice of the tMRCA as a test statistic is somewhat arbitrary, and that other alternatives (such as mean rate) could also be used. Unless one or other statistic was of particular interest, statistics might be preferred whose posterior distributions are easier to estimate from the MCMC. Results DATING ARTEFACTS WITH SIMULATED DATA Tests of model fit (This may be difficult for data sets where a priori plausible date or rate estimates span several orders of magnitude.) We set the mean rate prior to a gamma distribution with a shape parameter of 01 and a scale parameter of 1, the standard deviation of the rate prior to an exponential distribution with a mean of 1, the population size prior to an exponential distribution with a mean of 100, the HKY transition– transversion parameter prior to a gamma distribution with a shape parameter of 2 and a scale parameter of 1, and the between-site rate gamma shape prior to an exponential distribution of mean 1. Performance declined substantially when sampling was con- founded with genetic structure, that is when each genetic clade was sampled on a different date (Fig. 1c,d). In these cases, esti- mates of the tMRCA were highly inaccurate, but tests of tempo- ral signal wrongly indicated that the inaccurate dates could be trusted. These artefacts occurred both when there was high temporal structure (Fig. 1c), and when there was low temporal structure (Fig. 1d). Indeed, with low temporal structure, over a third of the simulated data sets showed a high correlation between sampling date and root-to-tip distance (Fig. S1d), and Temporal signal in structured molecular data 83 a typical data set gave strong evidence of temporal signal, despite yielding a wildly inaccurate estimate of the tMRCA: 51 ybp as opposed to the true value of 10 000 ybp To show why confounding misleads molecular dating, Fig. 2 illustrates two sections of phylogeny with the same sam- pling period but different levels of confounding (a) will tend (a) (b) (c) (d) Fig. 1. The left-hand column shows schematic representations of the tree topologies over which evolution was simulated. The grey triangle repre- sents the variable branching patterns of a simulated coalescence process. The middle column shows results of the regression of root-to-tip distance against sampling date. A significant positive correlation is consistent with the presence of temporal signal. P-values were obtained by random permu- tation of sampling dates across sequences (P) or monophyletic clusters of sequences that shared a sampling date (Pclust). The right-hand column shows the maximum a posteriori estimate of the tMRCA with 95% highest posterior density intervals (red) as inferred using BEAST. Tests of model fit These are compared to equivalent estimates from data sets with the sampling dates randomly permuted across sequences (purple), or clusters of sequences (blue). For the model selection approach, we report the increase in AICM values when sampling dates were included in the analysis. (a) and (b) represent a ‘bal- anced’ sampling strategy where each clade was sampled equally thoroughly at each of the sampling times; (c) and (d) represent a confounded sam- pling strategy where each clade was sampled at a different time. For (a) and (c), true temporal structure is high, such that a substantial amount of molecular evolution could occur between the sampling dates, while for (b) and (d), temporal structure is low. Temporal signal in structured molecular data 83 (a) (b) (c) (d) Fig. 1. The left-hand column shows schematic representations of the tree topologies over which evolution was simulated. The grey triangle repre- sents the variable branching patterns of a simulated coalescence process. The middle column shows results of the regression of root-to-tip distance against sampling date. A significant positive correlation is consistent with the presence of temporal signal. P-values were obtained by random permu- tation of sampling dates across sequences (P) or monophyletic clusters of sequences that shared a sampling date (Pclust). The right-hand column shows the maximum a posteriori estimate of the tMRCA with 95% highest posterior density intervals (red) as inferred using BEAST. These are compared to equivalent estimates from data sets with the sampling dates randomly permuted across sequences (purple), or clusters of sequences (blue). For the model selection approach, we report the increase in AICM values when sampling dates were included in the analysis. (a) and (b) represent a ‘bal- anced’ sampling strategy where each clade was sampled equally thoroughly at each of the sampling times; (c) and (d) represent a confounded sam- pling strategy where each clade was sampled at a different time. For (a) and (c), true temporal structure is high, such that a substantial amount of molecular evolution could occur between the sampling dates, while for (b) and (d), temporal structure is low. a typical data set gave strong evidence of temporal signal, despite yielding a wildly inaccurate estimate of the tMRCA: 51 ybp, as opposed to the true value of 10 000 ybp. To show why confounding misleads molecular dating, Fig. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 DATING ARTEFACTS WITH WHOLE GENOMES OF METHICILLIN-RESISTANT S. AUREUS 1a,b) or in clusters sampled in the same year (reproduc- ing the ‘confounded’ sampling of Fig. 1c,d). In all cases, the balanced subsampling provided consistent estimates of the tMRCA (Fig. 3a, purple points), albeit with wider credible inter- vals, reflecting the reduced sample size. However, the con- founded subsamples produced much younger dates (Fig. 3a, blue points). In addition, all six subsamples gave evidence of temporal signal using the standard tests. If we were to trust these standard tests, we might draw quite different conclusions about the evolution of antibiotic resistance in the UK. When Duch^ene et al.’s (2015a) method of clustered permu- tation was applied to our simulated data, performance remained good with balanced sampling (Fig. 1a,b: Pclust for regression, blue points for the Bayesian analysis) and improved dramatically with confounded sampling (Fig. 1c,d). In particu- lar, with confounded sampling, neither test of temporal signal reached significance, indicating – correctly – that both date estimates were unreliable. The same applies when we analysed subsamples collected over a 3-year period (Fig. S4g,h). Given evolutionary rates for these strains, fewer than 7 nucleotide substitutions per genome would be expected during this entire sampling period, and so this produces ‘low temporal structure’ data sets. For both data sets, the estimated tMRCA differed from its true value (as inferred from the complete data set; Fig. 3b, red dashed line). For the balanced subsample, all tests confirmed this lack of temporal signal, but the standard tests failed for the con- founded subsample, resulting in false confidence in an inaccu- rate and deceptively precise estimate of the tMRCA (Fig. 3b). Clustered permutation corrects for the confounding of genetic and temporal structure, but sometimes this confound- ing can arise from the evolutionary process itself, and not from sampling artefacts. Any evolutionary change in the genetic constitution of a population could lead to sequences sampled on the same date being more closely related to each other. A classic example is the ‘ladderized’ genealogy of influenza A, caused by regular selective sweeps (Grenfell et al. 2004), but the same effect could arise from genetic drift (Gray, Pybus & Salemi 2011). In either case, temporal and genetic structures are inherently confounded, and so clustered permutation becomes conservative. To explore the power of the clustered tests in this situation, we simulated ladderized genealogies with high temporal structure (Fig. S2). DATING ARTEFACTS WITH WHOLE GENOMES OF METHICILLIN-RESISTANT S. AUREUS Figure 1 illustrates dating artefacts with extreme cases, but the same artefacts occur with more realistic data. In the Support- ing information, we demonstrate this with simulations (Tables S1–S3, Figs S1 and S3), but it can also be observed with real- world data. To see this, we reanalysed 157 complete genomes of epi- demic methicillin-resistant S. aureus sequence type (ST) 22, sampled over a 17-year period (Holden et al. 2013). In agree- ment with Holden et al. (2013), we estimated the tMRCA of these sequences as 1980, 28 years prior to the youngest sample (Fig. S4). Several lines of evidence suggest that this tMRCA is plausible. First, all tests indicated very strong temporal signal (Fig. 3, Table S4); secondly, the inferred rate of evolution is consistent with previous estimates from S. aureus (Weinert et al. 2012); and finally, this dating places the acquisition of flu- oroquinolone resistance at the time and location where fluoro- quinolone drugs were first tested in UK clinical trials (Holden et al. 2013). Fig. 2. Illustrative phylogenies in which genetic and temporal structure are (a) unconfounded or (b) confounded. Grey arrows describe the dis- tance between pairs of sequences sampled on different dates (t0 and t1). to give better results than (b) for two connected reasons. First, in (b) the sampling period constitutes a much smaller propor- tion of the lineages that connect sequences sampled at different times. Secondly, (a) contains two quasi-independent opportu- nities to measure the evolutionary change between the sampling dates, while in (b), the measurements are clearly non-independent. As such, Fig. 2 suggests that the failure of random permutation can be understood, intuitively, as an inflation of the true sample size, when there is confounding. For this reason, one way to correct for the artefact is to identify genetic clusters (monophyletic groups) in the data that share a sampling date, and then permute dates over these clusters, rather than over the individual sequences (Duch^ene et al. 2015a). With this approach (a) would contain four clusters, but (b) would contain only two, and so a clustered permutation test will be less likely to reach significance. We next re-estimated the tMRCA after subsampling the S. aureus strains. These subsamples were chosen to transect the same root node and to retain the 17-year sampling period (illustrated in Fig. S4a–f). With these constraints, we chose strains either at random (reproducing the ‘balanced’ sampling of Fig. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 Tests of model fit 2 illustrates two sections of phylogeny with the same sam- pling period, but different levels of confounding. (a) will tend © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 84 G. G. R. Murray et al. (a) (b) Fig. 2. Illustrative phylogenies in which genetic and temporal structure are (a) unconfounded or (b) confounded. Grey arrows describe the dis- tance between pairs of sequences sampled on different dates (t0 and t1). (b) (a) DATING ARTEFACTS WITH WHOLE GENOMES OF METHICILLIN-RESISTANT S. AUREUS For BEAST permutation tests: ✓indicates that the true MAP estimate lay outside of the range of the MAP estimates from the randomized data sets, ✓✓indicates that the true MAP estimate is not within the HPD intervals of the estimates from randomized data sets, and ✓✓✓indicates that the HPD interval of the true estimate does not overlap with the HPD intervals of estimates from the randomized data sets. For the model selection approaches, we report the probability that the model without sampling dates is the ‘true’ model (AICM analysis), or the Bayes factor support for the inclusion of sampling dates (Kass & Raftery 1995). Tests indicating temporal signal are in bold; *P < 005; **P < 001. Fig. 3. Dating analyses for Staphylococcus aureus genomes sampled over 17 years. Plots show the maximum a posteriori (MAP) estimates of the tMRCA, with 95% highest posterior density (HPD) intervals. (a) shows the estimate from the complete data set (red), and from random (purple) or confounded (blue) subsamples, all with the same common ancestor and range of sampling dates. (b) shows estimates from subsamples with a nar- rower sampling range, and a different true tMRCA. Red dashed lines and shaded areas describe the best estimate of the tMRCA and its 95% HPD inter- val as inferred from the complete data set. Grey dashed lines show the youngest possible tMRCA, as determined by the oldest sample. Below are the results of tests of temporal signal and confounding. For BEAST permutation tests: ✓indicates that the true MAP estimate lay outside of the range of the MAP estimates from the randomized data sets, ✓✓indicates that the true MAP estimate is not within the HPD intervals of the estimates from randomized data sets, and ✓✓✓indicates that the HPD interval of the true estimate does not overlap with the HPD intervals of estimates from the randomized data sets. For the model selection approaches, we report the probability that the model without sampling dates is the ‘true’ model (AICM analysis), or the Bayes factor support for the inclusion of sampling dates (Kass & Raftery 1995). Tests indicating temporal signal are in bold; *P < 005; **P < 001. Pybus & Rambaut 2003b; Drummond et al. 2003a; Baele et al. 2012). Various measures and estimators of model fit are avail- able (Rambaut 2000; Suchard, Weiss & Sinsheimer 2003; Kitchen, Miyamoto & Mulligan 2008; Baele et al. 2012). DATING ARTEFACTS WITH WHOLE GENOMES OF METHICILLIN-RESISTANT S. AUREUS Results showed that the clus- tered permutation approach was still able to detect the tempo- ral signal (an appreciable rate of false negatives arose only when the basal clade was monophyletic, and fewer than four clusters were simulated). As with the simulated data, these problems can be solved by using the clustered permutation approach of Duch^ene et al. (2015a). If we define clusters as monophyletic groups sampled in same year, regression and Bayesian approaches both cor- rectly identified the data sets that yielded inaccurate estimates of the tMRCA. However, for these data, there is something arbi- trary about the choice to cluster by year (we might also have chosen to cluster by month). This highlights the need for a test of confounding that can be applied to real-world data sets. An obvious choice is a Mantel test of the correlation between Temporal signal in structured molecular data 85 85 (a) (b) Fig. 3. Dating analyses for Staphylococcus aureus genomes sampled over 17 years. Plots show the maximum a posteriori (MAP) estimates of the tMRCA, with 95% highest posterior density (HPD) intervals. (a) shows the estimate from the complete data set (red), and from random (purple) or confounded (blue) subsamples, all with the same common ancestor and range of sampling dates. (b) shows estimates from subsamples with a nar- rower sampling range, and a different true tMRCA. Red dashed lines and shaded areas describe the best estimate of the tMRCA and its 95% HPD inter- val as inferred from the complete data set. Grey dashed lines show the youngest possible tMRCA, as determined by the oldest sample. Below are the results of tests of temporal signal and confounding. For BEAST permutation tests: ✓indicates that the true MAP estimate lay outside of the range of the MAP estimates from the randomized data sets, ✓✓indicates that the true MAP estimate is not within the HPD intervals of the estimates from randomized data sets, and ✓✓✓indicates that the HPD interval of the true estimate does not overlap with the HPD intervals of estimates from the randomized data sets. For the model selection approaches, we report the probability that the model without sampling dates is the ‘true’ model (AICM analysis), or the Bayes factor support for the inclusion of sampling dates (Kass & Raftery 1995). Tests indicating temporal signal are in bold; *P < 005; **P < 001. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 DATING ARTEFACTS WITH WHOLE GENOMES OF METHICILLIN-RESISTANT S. AUREUS We initially tried the AICM, an analogue of the Akaike Informa- tion Criterion, which is estimated from the MCMC (Raftery et al. 2007; Baele et al. 2012). pairwise genetic distances and absolute differences in sampling dates. Applying this test to the S. aureus data successfully iden- tified the confounding in all of the confounded data sets, and in one of the smallest balanced data sets (Fig. 3). We then repeated the Mantel test after clustering the data (using the average pairwise genetic between clusters, and the absolute dif- ference between sampling years). This test confirmed that our choice to cluster sampling dates by year was sufficiently coarse- grained to eliminate the signal of confounding in these data (Fig. 3). On simulated data, the AICM approach performed very well, showing strong support for the ‘with dates’ model when- ever the tMRCA was well estimated, and weak or no support when the tMRCA was poorly estimated (Fig. 1, Table S1). How- ever, for the real S. aureus data, only one subsample gave evi- dence of temporal signal, and this was a confounded subsample where the tMRCA estimate was extremely poor (Fig. 3; Table S4). We next calculated full Bayes factors, using path sampling (Baele et al. 2012; Bouckaert et al. 2014; Leache DATING ARTEFACTS WITH WHOLE GENOMES OF METHICILLIN-RESISTANT S. AUREUS Temporal signal in structured molecular data 85 (a) (b) Fig. 3. Dating analyses for Staphylococcus aureus genomes sampled over 17 years. Plots show the maximum a posteriori (MAP) estimates of the tMRCA, with 95% highest posterior density (HPD) intervals. (a) shows the estimate from the complete data set (red), and from random (purple) or confounded (blue) subsamples, all with the same common ancestor and range of sampling dates. (b) shows estimates from subsamples with a nar- rower sampling range, and a different true tMRCA. Red dashed lines and shaded areas describe the best estimate of the tMRCA and its 95% HPD inter- val as inferred from the complete data set. Grey dashed lines show the youngest possible tMRCA, as determined by the oldest sample. Below are the results of tests of temporal signal and confounding. For BEAST permutation tests: ✓indicates that the true MAP estimate lay outside of the range of the MAP estimates from the randomized data sets, ✓✓indicates that the true MAP estimate is not within the HPD intervals of the estimates from randomized data sets, and ✓✓✓indicates that the HPD interval of the true estimate does not overlap with the HPD intervals of estimates from the randomized data sets. For the model selection approaches, we report the probability that the model without sampling dates is the ‘true’ model (AICM analysis), or the Bayes factor support for the inclusion of sampling dates (Kass & Raftery 1995). Tests indicating temporal signal are in bold; *P < 005; **P < 001. (a) (b) Fig. 3. Dating analyses for Staphylococcus aureus genomes sampled over 17 years. Plots show the maximum a posteriori (MAP) estimates of the tMRCA, with 95% highest posterior density (HPD) intervals. (a) shows the estimate from the complete data set (red), and from random (purple) or confounded (blue) subsamples, all with the same common ancestor and range of sampling dates. (b) shows estimates from subsamples with a nar- rower sampling range, and a different true tMRCA. Red dashed lines and shaded areas describe the best estimate of the tMRCA and its 95% HPD inter- val as inferred from the complete data set. Grey dashed lines show the youngest possible tMRCA, as determined by the oldest sample. Below are the results of tests of temporal signal and confounding. Discussion Molecular dates obtained with ‘dated-tip’ methods are reliable only if the sequence data exhibit temporal signal. As such, we cannot trust dates obtained from these methods unless we can also trust the tests for temporal signal. We have shown that all of the standard tests of temporal sig- nal can be severely misled for data sets where temporal and genetic structures are confounded, that is when closely related sequences are more likely to have been sampled at similar times. Our results show that the reliability of date estimates cannot be determined from the degree of genetic structure per se (data sets in Fig. 1a–d had equally high levels of structure), nor from the number of sequences sampled (Fig. 3a shows that subsamples of any size can yield both inaccurate and accurate estimates) and nor from the overall range of the sampling dates, or level of temporal structure (which was held constant across both Figs 1a,c and 3a). However, we have shown that when confounding is present, the clustered permutation approach of Duch^ene et al. (2015a), can give good results, whether applied to linear regression or Bayesian dating, and to data with or without temporal structure. Standard tests of temporal signal supported this scenario. For the dog samples, Mantel tests yielded no evidence of con- founding of temporal and genetic structures (P = 088), and permutation tests of the Bayesian dates detected temporal sig- nal (Fig. 4 purple points) even, weakly, with clustered permu- tation (Fig. 4, blue points). However, a combined analysis of all 175 isolates shows that this tMRCA estimate – and thus the epidemiological inference – is probably unreliable. In particu- lar, the genealogies of the dog and staffsamples are intermin- gled, implying that they share a most recent common ancestor (Fig. S6; Paterson et al. 2015). What is wrong with the analysis above? The answer is clear from comparing a neighbour-joining tree to the Maximum Clade Consensus (MCC) tree from the BEAST analyses (Fig. 4). The neighbour-joining tree has very little resolution reflecting the low genetic diversity in these data and confirms that the level of confounding is weak. In contrast, the BEAST tree is fully resolved and contains very high levels of confounding (Mantel tests using patristic distances: P < 0001; Fig. 4, and We have also introduced some refinements to the approach of Duch^ene et al. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 APPLICATION TO DATA FROM A SINGLE OUTBREAK When phylogenetic resolution is low, there are two ways to test for temporal signal, which avoid the artefact described above. The first is to use the regression approach, with a phy- logeny that was inferred without making any assumptions about molecular rates. The second is to use the clustered Baye- sian dating permutation approach, but with clusters identified from the MCC tree (Fig. 4, green points). Both approaches found no temporal signal in our S. aureus data (from either the dog, or the staffmember; Fig. S5), confirming that tMRCA esti- mates from these data cannot be trusted. Examples above used data that were subsampled in a contrived way, but the same artefacts can arise with complete data sets. To illustrate this, we analysed whole genomes of S. aureus ST22, from a single disease outbreak. These samples were obtained from a veterinary hospital over approximately 2 months, initially from a dog admitted to the clinic (141 iso- lates), and then from a staffmember (34 isolates) involved in the dog’s treatment (Paterson et al. 2015). Dated-tip analyses of these data placed the tMRCA of the dog strains on the day after the dog’s admission to the hospital, and the closely related strains from the staffmember at c. 12 weeks earlier (Fig. 4, for the dog samples, and Fig. S5, for the staffmember samples, red points). Together, these esti- mates suggest a scenario in which the dog was infected in the hospital, possibly by a staffmember with a long-standing infec- tion, and where transmission was likely associated with a strong bottleneck (since all of the genetic variation in the dog can be traced back to a single feasible transmission event). MODEL SELECTION APPROACH A test for temporal signal not considered so far, is to compare the fit of models with the sampling dates either included (‘with dates’) or ignored (‘no dates’) (Rambaut 2000; Drummond, G. G. R. Murray et al. 86 Table S5). This shows that, in the absence of phylogenetic sig- nal, the dating algorithm has enhanced the confounding, clus- tering the sequences by date to improve the fit of its clock model. (We note that no such difference was found in data sets analysed in earlier sections, where the data contained much higher levels of genetic diversity.) et al. 2014). This had the opposite problem: all but one sub- sample yielded strong support for the ‘with dates’ model. As such, model selection led to false confidence in inaccurate esti- mates of the tMRCA. The failure of this approach is initially surprising, since it makes no explicit assumptions about random sampling or exchangeability. Since the approach worked well on simulated data (which used a strict clock and known substitution pro- cess), this is probably explained by model inadequacy. Evolu- tionary models may be good enough to provide accurate estimates of the tMRCA, and yet sufficiently different from real- ity to render unreliable a comparison of model fit with and without sampling dates. It is also notable that the Bayes factor approach worked well when sampling was random, but not when sampling was confounded (Fig. 3; Table S4). This might be a failure analogous to ‘overfitting’, given the reduction in effective sample sizes in the confounded data sets (Fig. 2). It is important to note that low levels of genetic diversity would not be a problem, were there not also some genuine con- founding of temporal and genetic structures, for in the absence of any confounding, a random permutation approach would succeed. For these S. aureus data, weak confounding – unde- tected by the Mantel test – probably arose from the clinical sampling practice. In particular, different sets of anatomical sites of the dog were sampled on different dates (in part, as a consequence of the progression of the disease), and genetic structure was associated with these sites (Paterson et al. 2015). As a result, we find genetic structure between the earliest dog samples, and those taken on later dates (permutation test of Hudson’s Fst estimator: P < 0001; Hudson, Slatkin & Maddison 1992), although not between the two later dates. Discussion (2015a), which show how clustered permuta- tion can be best applied to real-world data. In particular, we have shown how a Mantel test, comparing genetic distance Temporal signal in structured molecular data 87 Temporal signal in structured molecular data 87 Fig. 4. The Bayesian dating test for Staphylo- coccus aureus strains sampled from a dog dur- ing an outbreak in a veterinary hospital. Differences in the degree of clustering with sampling date are apparent between the phylogenies estimated with (the MCC tree from the Bayesian dated-tip analysis) and without the use of temporal information (a neighbour-joining tree). Colour and symbol shape represent strains sampled on the same date. The plot shows the maximum a posteriori estimates of the tMRCA (on a log scale) with 95% highest posterior density intervals. The true estimate (red) is compared to estimates with the sampling dates randomly permuted across sequences (purple), or across single- date clusters identified from the neighbour- joining tree (blue), or the MCC tree (green). The blue horizontal line indicates the date of admission of the dog into the veterinary hospi- tal. Significance levels are described in the legend of Fig. 3. Finally, we have suggested that confounding is likely to be common when serially sampled-pathogen genomes are used to study the course of a single outbreak. This is partly because confounding can arise naturally from clinical sampling prac- tice. For example, different individuals will often be sampled at different times (Harris et al. 2013; Paterson et al. 2015), and these individuals will generally contain distinct populations of a pathogen, resulting from transmission barriers between indi- viduals, and population bottlenecks during transmission events. The same also applies to different tissues within an indi- vidual (e.g. Sacristan et al. 2003; Lee et al. 2008; Paterson et al. 2015) and to different geographic locations (Holmes 2008). We have also shown that the confounding may be enhanced when little evolutionary change has taken place, which may often be the case during a single outbreak. Consis- tent with this prediction, we have presented data from an out- break of S. aureus where standard tests provide support for date estimates – and thereby transmission scenarios – that are doubtful on other grounds (Paterson et al. 2015). and difference in sampling dates, can identify data sets where confounding is present (Fig. 3). © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 References Leache, A.D., Fujita, M.K., Minin, V.N. & Bouckaert, R.R. (2014) Species delimitation using genome-wide SNP data. Systematic Biology, 63, 534–542. Baele, G., Lemey, P., Bedford, T., Rambaut, A., Suchard, M.A. & Alekseyenko, A.V. (2012) Improving the accuracy of demographic and molecular clock model comparison while accommodating phylogenetic uncertainty. Molecular Biology and Evolution, 29, 2157–2167. Lee, H.Y., Perelson, A.S., Park, S.-C. & Leitner, T. (2008) Dynamic correlation between intrahost HIV-1 quasispecies evolution and disease progression. PLoS Computational Biology, 4, e1000240. Bouckaert, R., Heled, J., K€uhnert, D., Vaughan, T., Wu, C.-H., Xie, D., Suchard, M.A., Rambaut, A. & Drummond, A.J. (2014) BEAST 2: a software platform for bayesian evolutionary analysis. PLoS Computational Biology, 10, e1003537. McAdam, P.R., Templeton, K.E., Edwards, G.F., Holden, M.T.G., Feil, E.J., Aanensen, D.M. et al. (2012) Molecular tracing of the emergence, adaptation, and transmission of hospital-associated methicillin-resistant Staphylococcus aureus. Proceedings of the National Academy of Sciences, 109, 9107–9112. Buonagurio, D.A., Nakada, S., Parvin, J.D., Krystal, M., Palese, P. & Fitch, W.M. (1986) Evolution of human influenza A viruses over 50 years: rapid, uni- form rate of change in NS gene. Science, 232, 980–982. Navascues, M., Depaulis, F. & Emerson, B.C. (2010) Combining contemporary andancient DNA in populationgenetic and phylogeographical studies. Molec- ular Ecology Resources, 10, 760–772. Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Infer- ence: A Practical Information-Theoretic Approach, 2nd edn. Springer, New York, NY. Navascues, M. & Emerson, B.C. (2009) Elevated substitution rate estimates from ancient DNA: model violation and bias of Bayesian methods. Molecular Ecol- ogy, 18, 4390–4397. ence: A Practical Information-Theoretic Approach, 2nd edn. Springer, New York, NY. Didelot, X., Eyre, D.W., Cule, M., Ip, C.L., Ansari, M.A., Griffiths, D. et al. (2012) Microevolutionary analysis of Clostridium difficile genomes to investi- gate transmission. Genome Biology, 13, R118. Pagan, I. & Holguın, A. (2013) Reconstructing the timing and dispersion routes of HIV-1 subtype B epidemics in the Caribbean and Central America: a phylo- genetic story. PLoS ONE, 8, e69218. Drummond, A.J., Pybus, O.G. & Rambaut, A. (2003b) Inference of viral evolu- tionary rates from molecular sequences. Advances in Parasitology, 54, 331– 358. Paradis, E., Claude, J. & Strimmer, K. (2004) APE: analyses of phylogenetics and evolution in R language. Bioinformatics, 20, 289–290. Paterson, G.K., Harrison, E.M., Murray, G.G.R., Welch, J.J., Warland, J.H., Holden, M.T.G. et al. (2015) Capturing the cloud of diversity reveals complex- ity and heterogeneity of MRSA carriage, infection and transmission. References Nature Communications, 6, 6560. Drummond, A.J., Pybus, O.G., Rambaut, A., Forsberg, R. & Rodrigo, A.G. (2003a) Measurably evolving populations. Trends in Ecology & Evolution, 18, 481–488. Drummond, A.J., Suchard, M.A., Xie, D. & Rambaut, A. (2012) Bayesian phylo- genetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29, 1969–1973. Poncet, P. (2012). Modeest: Mode Estimation. Available from: http://cran. r-project.org/web/packages/modeest/index.html. R Core Team (2014) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. http://www. R-project.org. Duch^ene, D., Duch^ene, S. & Ho, S.Y.W. (2015b) Tree imbalance causes a bias in phylogenetic estimation of evolutionary timescales using heterochronous sequences. Molecular Ecology Resources, 15, 785–794. Raftery, A.E., Newton, M.A., Satagopan, J.M. & Krivitsky, P.N. (2007). Esti- mating the integrated likelihood via posterior simulation using the harmonic mean identity. Bayesian Statistics 8 (eds J.M. Bernardo, M.J. Bayarri J.O. Ber- ger, A.P. David, D. Heckerman, A.F.M. Smith & M. West), pp. 371–416. Ox- ford University Press, Oxford, UK. Duch^ene, S., Holmes, E.C. & Ho, S.Y.W. (2014) Analyses of evolutionary dynamics in viruses are hindered by a time-dependent bias in rate estimates. Proceedings of the Royal Society B, 281, 20140732. Duch^ene, S., Duch^ene, D., Holmes, E.C. & Ho, S.Y.W. (2015a) The performance of the date-randomisation test in phylogenetic analyses of time-structured virus data. Molecular Biology and Evolution, 32, 1895–1906. Rambaut, A. (2000) Estimating the rate of molecular evolution: incorporating non-contemporaneous sequences into maximum likelihood phylogenies. Bioinformatics, 16, 395–399. Duffy, S. & Holmes, E.C. (2009) Validation of high rates of nucleotide substitu- tion in geminiviruses: phylogenetic evidence from East African cassava mosaic viruses. Journal of General Virology, 90, 1539–1547. Rambaut, A. (2013) Path-O-Gen v1.4. Available from http://tree.bio.ed.ac.uk/ software/pathogen/. Firth, C., Kitchen, A., Shapiro, B., Suchard, M.A., Holmes, E.C. & Ram- baut, A. (2010) Using time-structured data to estimate evolutionary rates of double-stranded DNA viruses. Molecular Biology and Evolution, 27, 2038–2051. Rambaut, A., Suchard, M.A., Xie, D. & Drummond, A.J. (2014) Tracer v1.6. Available from http://beast.bio.ed.ac.uk/Tracer. Ramsden, C., Holmes, E.C. & Charleston, M.A. (2009) Hantavirus evolution in relation to its rodent and insectivore hosts: no evidence for codivergence. Molecular Biology and Evolution, 26, 143–153. Fraile, A., Pagan, I., Anastasio, G., Saez, E. & Garcıa-Arenal, F. (2011) Rapid genetic diversification and high fitness penalties associated with pathogenicity evolution in a plant virus. Molecular Biology and Evolution, 28, 1425–1437. Sacristan, S., Malpica, J.M., Fraile, A. & Garcıa-Arenal, F. Discussion Harris, S.R., Cartwright, E.J., T€or€ok, M.E., Holden, M.T., Brown, N.M., Ogilvy-Stuart, A.L. et al. (2013) Whole-genome sequencing for analysis of an outbreak of meticillin-resistant Staphylococcus aureus: a descriptive study. The outbreak of meticillin-resistant Staphylococcus aureus: a descriptive study. The Lancet Infectious Diseases, 13, 130–136. p y p y Lancet Infectious Diseases, 13, 130–136. Lancet Infectious Diseases, 13, 130–136 Holden, M.T.G., Hsu, L.-Y., Kurt, K., Weinert, L.A., Mather, A.E., Harris, S.R. l (2013) A i t it f th l ti d l b l d Acknowledgements Holden, M.T.G., Hsu, L.-Y., Kurt, K., Weinert, L.A., Mather, A.E., Harris, S.R. et al. (2013) A genomic portrait of the emergence, evolution, and global spread of a methicillin-resistant Staphylococcus aureus pandemic. Genome Research, 23, 653–664. Holden, M.T.G., Hsu, L.-Y., Kurt, K., Weinert, L.A., Mather, A.E., Harris, S.R. et al. (2013) A genomic portrait of the emergence, evolution, and global spread Holden, M.T.G., Hsu, L. Y., Kurt, K., Weinert, L.A., Mather, A.E., Harris, S.R. et al. (2013) A genomic portrait of the emergence, evolution, and global spread of a methicillin-resistant Staphylococcus aureus pandemic. Genome Research, 23 653 664 et al. (2013) A genomic portrait of the emergence, evolution, and global spread of a methicillin-resistant Staphylococcus aureus pandemic. Genome Research, 23, 653–664. GGRM is supported by a Medical Research Council studentship. JJW was sup- ported by the Issac Newton Trust and Wellcome Trust ISSF. AEM and SRH are supported by Wellcome Trust grant 098051. GGRM is supported by a Medical Research Council studentship. JJW was sup- ported by the Issac Newton Trust and Wellcome Trust ISSF. AEM and SRH are supported by Wellcome Trust grant 098051. of a methicillin-resistant Staphylococcus aureus pandemic. Genome Research, 23, 653–664. Holmes, E.C. (2008) Evolutionary history and phylogeography of human viruses. Annual Review of Microbiology, 62, 307–328. Hudson, R.R., Slatkin, M. & Maddison, W.P. (1992) Estimation of levels of gene flow from DNA sequence data. Genetics, 132, 583–589. Data accessibility Kass, R.E. & Raftery, A.E. (1995) Bayes factors. Journal of the American Statisti- cal Association, 90, 773–795. The two S. aureus data sets discussed are provided in Holden et al. (2013) and Paterson et al. (2015). R scripts of the all tests described have been uploaded as online supporting information. Kitchen, A., Miyamoto, M.M. & Mulligan, C.J. (2008) Utility of DNA viruses for studying human host history: case study of JC virus. Molecular Phylogenet- ics and Evolution, 46, 673–682. Korber, B., Muldoon, M., Theiler, J., Gao, F., Gupta, R., Lapedes, A., Hahn, B.H., Wolinsky, S. & Bhattacharya, T. (2000) Timing the ancestor of the HIV-1 pandemic strains. Science, 288, 1789–1796. Discussion We have also shown how the same test can confirm whether a particular choice of clusters has successfully removed the confounding (this is particularly useful when samples were taken on a very large range of dates, as in the S. aureus data from Holden et al. (2013). Finally, we have shown that an additional problem can arise for data with low levels of phylogenetic resolution, when dating algorithms may enhance the true level of confounding. We have suggested that, to mitigate this problem, clusters should be chosen from the tree estimated in the dating analysis (Fig. 4). The problem of confounding, discussed here, may explain some previously noted failures of the dated-tip approach. For example, Navascues & Emerson (2009) showed that inaccurate estimates of the tMRCA could be obtained in structured popula- tions when ancient and modern sequences came from different genetic clusters. Indeed, confounding is likely to be particularly severe when the temporal information comes from a small number of ancient DNA sequences. Similarly, Duch^ene, Duch^ene & Ho (2015b) showed that inaccurate results could be obtained when trees were highly imbalanced. Again, this might result from confounding, since imbalanced trees contain smaller clades, which are more likely to share a sampling date just by chance (this possibility is supported by simulations showing that unbalanced trees can give reliable results when sampling is balanced; Fig. S3f). If confounding of temporal and genetic structures is com- mon, then many dated-tip analyses may need revisiting. A remarkably common finding in the study of pathogen evolu- tion has been that plausible biogeographic scenarios imply much slower evolutionary rates (and so much older tMRCA), than are obtained from dated-tip analyses of serially sampled genomes; often, these estimates differ by several orders of G. G. R. Murray et al. 88 magnitude (Sharp & Simmonds 2011). We have shown that artefactual evidence of temporal signal often leads to false con- fidence in dates that bear no relation to the true age of diver- gence (see, e.g. Fig. 1d). As such, results reported here may explain some of the wilder disagreements about pathogen ori- gins. Gray, R.R., Pybus, O.G. & Salemi, M. (2011) Measuring the temporal struc- ture in serially-sampled phylogenies. Methods in Ecology and Evolution, 2, 437–445. Grenfell, B.T., Pybus, O.G., Gog, J.R., Wood, J.L.N., Daly, J.M., Mumford, J.A. & Holmes, E.C. (2004) Unifying the epidemiological and evolutionary dynamics of pathogens. Science, 303, 327–332. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 References Tests of temporal signal for two S. aureus data sets from a single outbreak. Received 26 June 2015; accepted 23 August 2015 Handling Editor: M. Gilbert Received 26 June 2015; accepted 23 August 2015 Handling Editor: M. Gilbert Appendix S1. Materials and methods. Appendix S1. Materials and methods. Appendix S2. A description of the R scripts used to implement a test of confounding temporal and genetic structure and two tests of temporal signal, that can be applied to serially-sampled sequence data. References (2003) Estimation of population bottlenecks during systemic movement of Tobacco Mosaic Virus in tobacco plants. Journal of Virology, 77, 9906–9911. Gire, S.K., Goba, A., Andersen, K.G., Sealfon, R.S.G., Park, D.J., Kanneh, L. et al. (2014) Genomic surveillance elucidates Ebola virus origin and transmis- sion during the 2014 outbreak. Science, 345, 1369–1372. Shankarappa, R., Margolick, J.B., Gange, S.J., Rodrigo, A.G., Upchurch, D., Farzadegan, H. et al. (1999) Consistent viral evolutionary changes associated © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 Temporal signal in structured molecular data 8 89 Table S1. Results of tests of temporal signal and dating analyses for simulated data sets with a true tMRCA of 10 000 ybp. Table S1. Results of tests of temporal signal and dating analyses for simulated data sets with a true tMRCA of 10 000 ybp. with the progression of human immunodeficiency virus type 1 infection. Jour- nal of Virology, 73, 10489–10502. Sharp, P.M. & Simmonds, P. (2011) Evaluating the evidence for virus/host co-evolution. Current Opinion in Virology, 1, 436–441. Table S2. The parameters used to simulate data sets. Table S2. The parameters used to simulate data sets. Smith, G.J.D., Vijaykrishna, D., Bahl, J., Lycett, S.J., Worobey, M., Pybus, O.G. et al. (2009) Origins and evolutionary genomics of the 2009 swine-origin H1N1 influenza A epidemic. Nature, 459, 1122–1125. Table S3. Simulation results when ‘high’ and ‘low’ temporal signal were created by adjusting the substitution rate instead of the sampling dates (see Table S2). Suchard, M.A., Weiss, R.E. & Sinsheimer, J.S. (2003) Testing a molecular clock without an outgroup: derivations of induced priors on branch-length restric- tions in a Bayesian framework. Systematic Biology, 52, 48–54. Venter, J.H. (1967) On estimation of the mode. The Annals of Mathematical Statistics, 38, 1446–1455. Table S4. Tests of temporal signal for an S. aureus data set from Hol- den et al. (2013) and subsamples of these data. Weinert, L.A., Welch, J.J., Suchard, M.A., Lemey, P., Rambaut, A. & Fitzgerald, J.R. (2012) Molecular dating of human-to-bovid host jumps by Staphylococcus aureus reveals an association with the spread of domestication. Biology Letters, 8, 829–832. Table S5. © 2015 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society., Methods in Ecology and Evolution, 7, 80–89 Received 26 June 2015; accepted 23 August 2015 Handling Editor: M. Gilbert Supporting Information Additional Supporting Information may be found in the online version of this article. Appendix S3. An R script that provides functions used in in the tests. Fig. S1. The distribution of signed r2 values from regressions of phylo- genetic root-to-tip distance against sampling date, for each of 1000 sim- ulated data sets. Appendix S4. An R script for a regression of phylogenetic root-to-tip distance against sampling date with both standard and clustered per- mutation tests. Fig. S2. Results of tests of temporal signal for data simulated with a high level of temporal signal, but a ladderised genealogy, in which genetic structure arises over time from the evolution of a single pop- ulation. Appendix S5. An R script for both standard and clustered randomisa- tions of dates over sequences in BEAST xml files. Appendix S6. An R script for a Mantel test of confounding of genetic and temporal distances. Fig. S3. The topologies over which data was simulated. Fig. S3. The topologies over which data was simulated. Fig. S4. The MCC tree produced from 157 methicillin-resistant S. aureus genomes from Holden et al. (2013). Appendix S7. An example sequence data input file. Appendix S8. An example tree input file. Fig. S5. The Bayesian dating test for S. aureus strains sampled from a staffmember (34 isolates) during an outbreak in a veterinary hos- pital (Paterson et al. 2015). Appendix S9. An example BEAST xml input file. Fig. S6. A genealogy of the strains sampled from the dog and the staffmember from (Paterson et al. 2015).
https://openalex.org/W2617865703
https://iris.unito.it/bitstream/2318/1642129/1/ncomms15107.pdf
English
null
Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer
Nature communications
2,017
cc-by
19,368
ARTICLE Received 28 Jan 2016 | Accepted 1 Mar 2017 | Published 31 May 2017 Received 28 Jan 2016 | Accepted 1 Mar 2017 | Published 31 May 2017 1 Department of Oncology, University of Torino School of Medicine, 10060 Candiolo Torino, Italy. 2 Candiolo Cancer Institute—FPO IRCCS, 10060 Candiolo Torino, Italy. 3 Department of Control and Computer Engineering, Torino School of Engineering, 10129 Torino, Italy. 4 Department of Medical Sciences, University of Torino School of Medicine, 10060 Candiolo Torino, Italy. 5 Department of Oncology, Johns Hopkins University, Baltimore, 21287 Maryland, USA. 6 National Institute of Biostructures and Biosystems, INBB, 00136 Rome, Italy. * These authors contributed equally to this work. Correspondence and requests for materials should be addressed to L.T. (email: livio.trusolino@unito.it) or to E.M. (email: enzo.medico@unito.it) or to A.B. (email: andrea.bertotti@unito.it). Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer Claudio Isella1,2, Francesco Brundu3, Sara E. Bellomo1,2, Francesco Galimi1,2, Eugenia Zanella2, Roberta Porporato2, Consalvo Petti1,2, Alessandro Fiori2, Francesca Orzan1,2, Rebecca Senetta2,4, Carla Boccaccio1,2, Elisa Ficarra3, Luigi Marchionni5, Livio Trusolino1,2, Enzo Medico1,2,* & Andrea Bertotti1,2,6,* Stromal content heavily impacts the transcriptional classification of colorectal cancer (CRC), with clinical and biological implications. Lineage-dependent stromal transcriptional components could therefore dominate over more subtle expression traits inherent to cancer cells. Since in patient-derived xenografts (PDXs) stromal cells of the human tumour are substituted by murine counterparts, here we deploy human-specific expression profiling of CRC PDXs to assess cancer-cell intrinsic transcriptional features. Through this approach, we identify five CRC intrinsic subtypes (CRIS) endowed with distinctive molecular, functional and phenotypic peculiarities: (i) CRIS-A: mucinous, glycolytic, enriched for microsatellite instability or KRAS mutations; (ii) CRIS-B: TGF-b pathway activity, epithelial–mesenchymal transition, poor prognosis; (iii) CRIS-C: elevated EGFR signalling, sensitivity to EGFR inhibitors; (iv) CRIS-D: WNT activation, IGF2 gene overexpression and amplification; and (v) CRIS-E: Paneth cell-like phenotype, TP53 mutations. CRIS subtypes successfully categorize independent sets of primary and metastatic CRCs, with limited overlap on existing transcriptional classes and unprecedented predictive and prognostic performances. 1 Department of Oncology, University of Torino School of Medicine, 10060 Candiolo Torino, Italy. 2 Candiolo Cancer Institute—FPO IRCCS, 10060 Candiolo Torino, Italy. 3 Department of Control and Computer Engineering, Torino School of Engineering, 10129 Torino, Italy. 4 Department of Medical Sciences, University of Torino School of Medicine, 10060 Candiolo Torino, Italy. 5 Department of Oncology, Johns Hopkins University, Baltimore, 21287 Maryland, USA. 6 National Institute of Biostructures and Biosystems, INBB, 00136 Rome, Italy. * These authors contributed equally to this work. Correspondence and requests for materials should be addressed to L.T. (email: livio.trusolino@unito.it) or to E.M. (email: enzo.medico@unito.it) or to A.B. (email: andrea.bertotti@unito.it). 1 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 A a large fraction of metastatic surgical specimens (range 80–95%, Fig. 1c, Supplementary Data 3). A number of classification systems based on gene expression have been proposed that stratify colorectal cancer (CRC) in subgroups with distinct molecular and clinical features1–7. Comparative analyses in different data sets have revealed substantial classification coherence across the various signatures, particularly in the case of a ‘Stem/Serrated/Mesenchymal’ (SSM) subtype endowed with negative prognosis8–10. Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer g pp y We then analysed the GEPs of 515 PDXs from CRC-LM from 244 patients (GSE76402)12. To take into account intra-tumour heterogeneity, for most of the cases (149/244, 61%) multiple PDXs derived from regionally distinct areas of the original tumour were profiled (see Methods for details). A total of 115 of these patients (corresponding to 240 PDX profiles) were also included in the CRC-LM data set (Supplementary Data 4). In this setting, confident classification by published signatures was strongly reduced (range 50–90%; Fig. 1c, Supplementary Data 5). In particular, we observed a systematic loss of classification rate in the SSM classes, which had been previously reported to be mainly sustained by transcripts of stromal origin8,9 (w2-test assuming as null hypothesis that the fraction of cases classified as SSM in PDXs is not different from that obtained in the CRC-LM data set; Po5  10  10, w2 ¼ 41.084, for CCS3; Po1  10  4, w2 ¼ 15.688, for CRCA5; Po5  10  09, w2 ¼ 41.679, for CCMS4). SSM classes also demonstrated the lowest validation rate when comparing the classification of CRC-LM with that of their corresponding PDXs (Fig. 1d,e, Supplementary Tables 3 and 5). We reasoned that this discre- pancy in classification performance between original and mouse-propagated metastases might be ascribed to obliteration of human stroma in PDXs and substitution by host mouse components, which were not detected by human-specific arrays. 12 17 yp ( ) Interestingly, we and others independently reported that a large portion of the genes sustaining the SSM subtype (CMS4 within the CMS) are of stromal origin, and that the presence of stromal cells, mainly cancer-associated fibroblasts (CAFs), is a strong indicator of tumour aggressiveness8,9. Paradoxically, this could suggest that the non-neoplastic populations and the extrinsic factors of the tumour reactive stroma play the leading role in dictating cancer progression, while the intrinsic features of cancer cells convey less relevant cues. Alternatively, in whole tumour lysates the transcriptional consequences of biologically meaningful traits that are inherent to cancer cells might be obscured by the presence of a dominant, lineage-dependent transcriptional component of stromal origin. Indeed, an abundant tumour stromal content is expected to mask subtle gene expression profiles (GEPs) specifically exhibited by cancer cells. At present, very little is known about how and to what extent cancer cell-specific gene expression traits contribute to classify cancer. Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer At least in principle, distilling variants based on genes that are expressed only by the transformed cells, in a context that is purified of heterologous multicellular complexity, might uncover subtypes that demonstrate higher predictive/ prognostic value when used as classifiers. p y p y In line with previous genetic and phenotypic evidence12,17, the classification incongruence between PDXs and their original counterparts is unlikely to be the consequence of genetic or functional drifts associated with PDX engraftment and propagation. Indeed, correlative analyses confirmed that the PDXs and the corresponding original CRC-LMs were significantly more similar than unmatched PDX/CRC-LM pairs (Wilcoxon rank sum Po5  10  16, Fig. 1f). Moreover, in 88% of the cases, the best correlate for each PDX was its matched original counterpart (Supplementary Data 6). This suggests that PDXs largely maintain the transcriptional identity of their pre-implantation surgical pairs and puts forward the notion that depletion of stroma-derived signals is likely the major source of transcriptional variation between surgical specimens and PDXs. To tackle this issue, we exploited a large collection (n ¼ 515 samples from 244 patients) of patient-derived xenografts (PDXs)12. In PDXs, the stromal components of the original tumour are substituted by their murine counterparts as a consequence of xenotransplantation13,14, so that detection of their transcripts can be avoided by appropriate use of human- specific arrays8. On these premises, here we leverage the unique opportunity afforded by PDXs to selectively explore the cancer cell-specific transcriptome of colorectal tumours. By doing so, we define the colorectal cancer intrinsic subtypes (CRIS) and evaluate their prognostic and predictive potential. Cancer-cell intrinsic transcriptional traits classify CRC. The tumour stroma certainly contributes to tumour biology but at the same time is not a direct expression of the transformed phenotype of neoplastic cells. Furthermore, classical histopathological studies have shown that many solid tumours (including CRC) feature a desmoplastic reaction whereby dense connective tissue, produced by activated fibroblasts, tends to prevail quantitatively over areas occupied by cancer cells18,19. We therefore speculated that stromal content might act as a dominant source of variation that could mask more subtle—albeit biologically relevant—traits inherently related to properties of cancer cells. To explore this hypothesis, we took advantage from PDX-derived GEPs to uncover cancer-cell intrinsic transcriptional subtypes in CRC. 20 Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer These classification efforts have been recently consolidated by a multi-institutional initiative that comprehensively cross compared the different subtype assignments on a common set of samples, leading to the definition of the consensus molecular subtypes11 (CMS). Fig. 1c, Supplementary Data 3). We then analysed the GEPs of 515 PDXs from CRC-LM from 244 patients (GSE76402)12. To take into account intra-tumour heterogeneity, for most of the cases (149/244, 61%) multiple PDXs derived from regionally distinct areas of the original tumour were profiled (see Methods for details). A total of 115 of these patients (corresponding to 240 PDX profiles) were also included in the CRC-LM data set (Supplementary Data 4). In this setting, confident classification by published signatures was strongly reduced (range 50–90%; Fig. 1c, Supplementary Data 5). In particular, we observed a systematic loss of classification rate in the SSM classes, which had been previously reported to be mainly sustained by transcripts of stromal origin8,9 (w2-test assuming as null hypothesis that the fraction of cases classified as SSM in PDXs is not different from that obtained in the CRC-LM data set; Po5  10  10, w2 ¼ 41.084, for CCS3; Po1  10  4, w2 ¼ 15.688, for CRCA5; Po5  10  09, w2 ¼ 41.679, for CCMS4). SSM classes also demonstrated the lowest validation rate when comparing the classification of CRC-LM with that of their corresponding PDXs (Fig. 1d,e, Supplementary Tables 3 and 5). We reasoned that this discre- pancy in classification performance between original and mouse-propagated metastases might be ascribed to obliteration of human stroma in PDXs and substitution by host mouse components, which were not detected by human-specific arrays. In line with previous genetic and phenotypic evidence12,17, the classification incongruence between PDXs and their original counterparts is unlikely to be the consequence of genetic or functional drifts associated with PDX engraftment and propagation. Indeed, correlative analyses confirmed that the PDXs and the corresponding original CRC-LMs were significantly more similar than unmatched PDX/CRC-LM pairs (Wilcoxon rank sum Po5  10  16, Fig. 1f). Moreover, in 88% of the cases, the best correlate for each PDX was its matched original counterpart (Supplementary Data 6). This suggests that PDXs largely maintain the transcriptional identity of their pre-implantation surgical pairs and puts forward the notion that depletion of stroma-derived signals is likely the major source of transcriptional variation between surgical specimens and PDXs. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications Results CRC PDXs fail assignment to public transcriptional subtypes. We and others have recently reported that CRC classification based on published transcriptional signatures is heavily affected by the tumour stromal content8,9. As a further evidence to this notion, we conducted principal component analyses on a large gene expression data set of primary colorectal tumours (n ¼ 450)15, assembled from information downloaded from the TCGA data portal (Supplementary Data 1), and confirmed that stromal genes have a major impact on the transcriptional profiles of such cases (Supplementary Fig. 1, Supplementary Data 2). To test whether this scenario extends beyond the context of primary tumours, we applied three published CRC signatures1–3 to a proprietary data set (n ¼ 185) of CRC liver metastases (CRC-LM) (GSE73255) using the nearest template prediction (NTP) algorithm16. This analysis showed that liver metastases behaved comparably to primary lesions, with a superimposable classification pattern (Fig. 1a, Supplementary Fig. 2a,b) that was sustained by similar transcriptional traits (Fig. 1b, Supplementary Fig. 2c,d). Overall, published signatures confidently classified By applying non-negative matrix factorization (NMF)20 to the 515-PDX expression data set, we identified optimal partitioning into five clusters (Fig. 2a, Supplementary Data 7). As previously described, we employed significance analysis of microarrays (SAM21; FDRo0.005) and prediction analysis for microarrays (PAM22; FDR ¼ 0.035) to identify 903 genes whose expression best discriminates each subtype. Results P values are calculated by Fisher’s exact test using the Submap tool available from Gene Pattern. (c) C er metastases (CRC-LM), or their corresponding PDXs, which were confidently assigned (NTP, FDRo0.2 sifiers. (d) Fraction of PDXs assigned to the same class of their corresponding liver metastasis according do view of correspondences between the CRCA class assignments of CRC-LM samples and those of thei earson’s correlation values obtained by analysing unmatched (grey line) and matched (red line) CRC-LM 0 0.2 0.4 0.6 0.8 1 CCS2 CCS1 CCS3 0 0.2 0.4 0.6 0.8 1 CRCA1 CRCA2 CRCA4 CRCA3 CRCA5 0 0.2 0.4 0.6 0.8 1 CCMS2 CCMS3 CCMS1 CCMS5 CCMS6 CCMS4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX CRC-LM PDX 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX 0 0.1 0.2 0.3 0.4 –1 –0.5 0 0.5 1 CRCA1 CRCA2 CRCA3 CRCA4 CRCA5 e Fraction of samples classified with confidence d c Fraction of classified LMs confirmed in PDXs f Cosine correlation Frequency PDX CRC-LM –1 +1 z score CRC-LM CRCA2 CRCA1 CRCA4 CRCA3 CRCA5 CCMS3 CCMS2 CCMS5 CCMS1 CCMS6 CCMS4 CCS2 CCS1 CCS3 0 1 P value Figure 1 | CRC PDXs fail assignment to public transcriptional subtypes. (a) Heatmap showing NTP classification of the CRC-LM d signatures of the CRCA classification system. (b) Submap analysis of CRCA subtype similarity between CRC liver metastases (CRC tumours (GSE14333). P values are calculated by Fisher’s exact test using the Submap tool available from Gene Pattern. (c) Column fractions of CRC liver metastases (CRC-LM), or their corresponding PDXs, which were confidently assigned (NTP, FDRo0.2) to the different public classifiers. (d) Fraction of PDXs assigned to the same class of their corresponding liver metastasis according to the (as in c). Results (e) Caleydo view of correspondences between the CRCA class assignments of CRC-LM samples and those of their PDX (f) Distribution of Pearson’s correlation values obtained by analysing unmatched (grey line) and matched (red line) CRC-LM/PDX 0 0.2 0.4 0.6 0.8 1 CCS2 CCS1 CCS3 0 0.2 0.4 0.6 0.8 1 CRCA1 CRCA2 CRCA4 CRCA3 CRCA5 0 0.2 0.4 0.6 0.8 1 CCMS2 CCMS3 CCMS1 CCMS5 CCMS6 CCMS4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX CRC-LM PDX 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX 0 0.1 0.2 0.3 0.4 –1 –0.5 0 0.5 1 CRCA1 CRCA2 CRCA3 CRCA4 CRCA5 e Fraction of samples classified with confidence d c Fraction of classified LMs confirmed in PDXs f Cosine correlation Frequency PDX CRC-LM –1 +1 z score CRC-LM CRCA2 CRCA1 CRCA4 CRCA3 CRCA5 CCMS3 CCMS2 CCMS5 CCMS1 CCMS6 CCMS4 CCS2 CCS1 CCS3 0 1 P value DXs fail assignment to public transcriptional subtypes. (a) Heatmap showing NTP classification of the CRC-LM data set based on the CRCA classification system. (b) Submap analysis of CRCA subtype similarity between CRC liver metastases (CRC-LM) and primary 33). P values are calculated by Fisher’s exact test using the Submap tool available from Gene Pattern. (c) Column chart showing the iver metastases (CRC-LM), or their corresponding PDXs, which were confidently assigned (NTP, FDRo0.2) to the subtypes of three assifiers. (d) Fraction of PDXs assigned to the same class of their corresponding liver metastasis according to the three classifiers ydo view of correspondences between the CRCA class assignments of CRC-LM samples and those of their PDX counterparts. Pearson’s correlation values obtained by analysing unmatched (grey line) and matched (red line) CRC-LM/PDX pairs. Results To generate a cancer-cell intrinsic classifier that was not influenced by stroma-derived transcripts in human CRC samples, we excluded from the NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications 2 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 0 0.2 0.4 0.6 0.8 1 CCS2 CCS1 CCS3 0 0.2 0.4 0.6 0.8 1 CRCA1 CRCA2 CRCA4 CRCA3 CRCA5 0 0.2 0.4 0.6 0.8 1 CCMS2 CCMS3 CCMS1 CCMS5 CCMS6 CCMS4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX CRC-LM PDX 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX 0 0.1 0.2 0.3 0.4 b a CRCA1 CRCA2 CRCA3 CRCA4 CRCA e Fraction of samples classified with confidence d c Fraction of classified LMs confirmed in PDXs f Frequency PDX CRC-LM –1 +1 z score CRC-LM CRCA2 CRCA1 CRCA4 CRCA3 CRCA5 CCMS3 CCMS2 CCMS5 CCMS1 CCMS6 CCMS4 CCS2 CCS1 CCS3 1 2 3 4 5 0 1 P value 1 2 3 4 5 CRCA subtypes in CRC-LM (liver metastases) CRCA subtypes in GSE14333 (primary tumours) 1 2 3 4 5 CRCA signatures a –1 +1 z score CRC-LM 1 2 3 4 5 CRCA signatures b 0 1 P value 1 2 3 4 5 CRCA subtypes in CRC-LM (liver metastases) CRCA subtypes in GSE14333 (primary tumours) 1 2 3 4 5 b a 0 0.2 0.4 0.6 0.8 1 CCS2 CCS1 CCS3 0 0.2 0.4 0.6 0.8 1 CRCA1 CRCA2 CRCA4 CRCA3 CRCA5 0 0.2 0.4 0.6 0.8 1 CCMS2 CCMS3 CCMS1 CCMS5 CCMS6 CCMS4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX CRC-LM PDX 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX 0 0.1 0.2 0.3 0.4 –1 –0.5 0 0.5 1 CRCA1 CRCA2 CRCA3 CRCA4 CRCA5 e Fraction of samples classified with confidence d c Fraction of classified LMs confirmed in PDXs f Cosine correlation Frequency PDX CRC-LM –1 +1 z score CRC-LM CRCA2 CRCA1 CRCA4 CRCA3 CRCA5 CCMS3 CCMS2 CCMS5 CCMS1 CCMS6 CCMS4 CCS2 CCS1 CCS3 0 1 P value 1 2 3 4 5 CRCA subtypes in CRC-LM (liver metastases) yp (primary tumours) 1 2 3 4 5 CRCA signatures s fail assignment to public transcriptional subtypes. (a) Heatmap showing NTP classification of the CRC RCA classification system. (b) Submap analysis of CRCA subtype similarity between CRC liver metastase ). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 = 0.9964 CRIS-C CRIS-D CRIS-E 0 0.2 0.4 0.6 0.8 1 Fraction of classified PDXs confirmed in CRC-LM A B C D E CRIS e h GSE14333 A B C D E CRIS signatures –1 +1 z score CRIS signatures h g Cophenetic coefficient Coph. coeff. = 0.9576 2 3 4 5 6 0.96 0.92 1.00 GSE14333 TCGA K A B C D E A B C D E CRIS signatures –1 +1 z score –1 +1 z score CRIS i CRIS signatures h g Cophenetic coefficient Coph. coeff. = 0.9576 2 3 4 5 6 0.96 0.92 0 0.1 0.2 0.3 0.4 1.00 GSE14333 TCGA K A B C D E A B C D E CRIS signatures –1 +1 z score –1 +1 z score i A B C D E CRIS NC Fraction of samples re 2 | CRIS classifies both CRC tumours and PDXs based on cancer-cell intrinsic transcriptional traits. (a) NMF-based consensus c data set and plot of cophenetic coefficients for K ¼ 2–6. (b,c) Heatmap showing NTP-based CRIS classification of the PDX (b) and ata sets. (d) Column chart showing the fractions of CRC-LMs and PDXs that were confidently classified (FDRo0.2) by CRIS; colour esent the relative distribution of subtypes within each data set. (e) Column chart showing, for each CRIS subtype, the fractions of sa h g CRIS signatures g TCGA A B C D E –1 +1 z score g CRIS signatures 0 0.1 0.2 0.3 0.4 GSE14333 TCGA z score z score i A B C D E CRIS NC Fraction of samples i on cancer-cell intrinsic transcriptional traits. (a) NMF-based consensus clustering of the Figure 2 | CRIS classifies both CRC tumours and PDXs based on cancer-cell intrinsic transcriptional traits. (a) NMF-based consensus clustering of the PDX data set and plot of cophenetic coefficients for K ¼ 2–6. (b,c) Heatmap showing NTP-based CRIS classification of the PDX (b) and the CRC-LM (c) data sets. (d) Column chart showing the fractions of CRC-LMs and PDXs that were confidently classified (FDRo0.2) by CRIS; coloured sub-columns represent the relative distribution of subtypes within each data set. (e) Column chart showing, for each CRIS subtype, the fractions of samples for which CRC-LMs and their corresponding PDXs were confidently assigned (FDRo0.2) to the same class. (f) Caleydo view of CRIS correspondences between CRC-LM samples and their PDX counterparts. Results 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX CRC-LM PDX 0 0.2 0.4 0.6 0.8 1 CRC-LM PDX Fraction of samples classified with confidence c CRCA2 CRCA1 CRCA4 CRCA3 CRCA5 CCMS3 CCMS2 CCMS5 CCMS1 CCMS6 CCMS4 CCS2 CCS1 CCS3 0 0.2 0.4 0.6 0.8 1 CCS2 CCS1 CCS3 0 0.2 0.4 0.6 0.8 1 CRCA1 CRCA2 CRCA4 CRCA3 CRCA5 0 0.2 0.4 0.6 0.8 1 CCMS2 CCMS3 CCMS1 CCMS5 CCMS6 CCMS4 d Fraction of classified LMs confirmed in PDXs d c 0 0.1 0.2 0.3 0.4 –1 –0.5 0 0.5 1 f Cosine correlation Frequency CRCA1 CRCA2 CRCA3 CRCA4 CRCA5 e PDX CRC-LM f e Figure 1 | CRC PDXs fail assignment to public transcriptional subtypes. (a) Heatmap showing NTP classification of the CRC-LM data set based on the signatures of the CRCA classification system. (b) Submap analysis of CRCA subtype similarity between CRC liver metastases (CRC-LM) and primary tumours (GSE14333). P values are calculated by Fisher’s exact test using the Submap tool available from Gene Pattern. (c) Column chart showing the fractions of CRC liver metastases (CRC-LM), or their corresponding PDXs, which were confidently assigned (NTP, FDRo0.2) to the subtypes of three different public classifiers. (d) Fraction of PDXs assigned to the same class of their corresponding liver metastasis according to the three classifiers (as in c). (e) Caleydo view of correspondences between the CRCA class assignments of CRC-LM samples and those of their PDX counterparts. (f) Distribution of Pearson’s correlation values obtained by analysing unmatched (grey line) and matched (red line) CRC-LM/PDX pairs. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 = 0.9576 2 3 4 5 6 0.96 0.92 0 0.1 0.2 0.3 0.4 1.00 –1 +1 z score –1 +1 z score CRC-LM PDX GSE14333 TCGA CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E PDX CRC-LM K A B C D E A B C D E CRIS signatures –1 +1 z score –1 +1 z score 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Fraction of samples classified with confidence Fraction of classified PDXs confirmed in CRC-LM A B C D E CRIS A B C D E CRIS CRC-LM PDX e i A B C D E CRIS NC Fraction of samples Figure 2 | CRIS classifies both CRC tumours and PDXs based on cancer-cell intrinsic transcriptional traits. (a) NMF-based consensus clustering of the PDX data set and plot of cophenetic coefficients for K ¼ 2–6. (b,c) Heatmap showing NTP-based CRIS classification of the PDX (b) and the CRC-LM (c) data sets. (d) Column chart showing the fractions of CRC-LMs and PDXs that were confidently classified (FDRo0.2) by CRIS; coloured sub-columns represent the relative distribution of subtypes within each data set. (e) Column chart showing, for each CRIS subtype, the fractions of samples for which CRC-LMs and their corresponding PDXs were confidently assigned (FDRo0.2) to the same class. (f) Caleydo view of CRIS correspondences between CRC-LM samples and their PDX counterparts. (g,h) Heatmaps showing CRIS classification of primary tumours based on RNAseq data from the TCGA (g) and Affymetrix microarray data of the GSE14333 data set (h). (i) Column chart showing the distribution of CRIS subtypes based on NTP classification across 16 independent CRC data sets (Supplementary Data 11); NC (non-classified) indicates samples for which NTP could not confidently assign any of the subtypes (FDR40.2). d 0 0.2 0.4 0.6 0.8 1 Fraction of samples classified with confidence A B C D E CRIS CRC-LM PDX e d f d Coph. coeff. 0.9949 K=4 Coph. coeff. = 0.9693 CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E PDX CRC-LM 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Fraction of samples classified with confiden Fraction of classified PDXs confirmed in CRC-LM A B C D E CRIS A B C D E CRIS CRC-LM PDX e e 0 0.2 0.4 0.6 0.8 1 Fraction of classified PDXs confirmed in CRC-LM A B C D E CRIS e K=5 Coph. coeff. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 CRIS signatures c b a d f h g K=2 Coph. coeff. = 0.9954 Cophenetic coefficient K=3 Coph. coeff. = 0.9949 K=4 Coph. coeff. = 0.9693 K=5 Coph. coeff. = 0.9964 K=6 Coph. coeff. = 0.9576 2 3 4 5 6 0.96 0.92 0 0.1 0.2 0.3 0.4 1.00 CRIS signatures –1 +1 z score –1 +1 z score CRC-LM PDX GSE14333 TCGA CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E PDX CRC-LM K A B C D E A B C D E A B C D E A B C D E CRIS signatures CRIS signatures –1 +1 z score –1 +1 z score 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Fraction of samples classified with confidence Fraction of classified PDXs confirmed in CRC-LM A B C D E CRIS A B C D E CRIS CRC-LM PDX e i A B C D E CRIS NC Fraction of samples S classifies both CRC tumours and PDXs based on cancer-cell intrinsic transcriptional traits. (a) NMF-based consensus cl nd plot of cophenetic coefficients for K ¼ 2–6. (b,c) Heatmap showing NTP-based CRIS classification of the PDX (b) and t d) Column chart showing the fractions of CRC-LMs and PDXs that were confidently classified (FDRo0.2) by CRIS; coloure elative distribution of subtypes within each data set. (e) Column chart showing, for each CRIS subtype, the fractions of sam their corresponding PDXs were confidently assigned (FDRo0.2) to the same class. (f) Caleydo view of CRIS corresponden es and their PDX counterparts. (g,h) Heatmaps showing CRIS classification of primary tumours based on RNAseq data fro trix microarray data of the GSE14333 data set (h) (i) Column chart showing the distribution of CRIS subtypes based on NT E NATURE COMMUNICATIONS | DOI: 10.1038/ a K=2 Coph. coeff. = 0.9954 K=3 Coph. coeff. = 0.9949 b CRIS signatures –1 +1 z score PDX A B C D E c –1 +1 z score CRC-LM A B C D E CRIS signatures b c c a CRIS signatures d f h g Cophenetic coefficient Coph. coeff. = 0.9949 K=4 Coph. coeff. = 0.9693 K=5 Coph. coeff. = 0.9964 K=6 Coph. coeff. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 (g,h) Heatmaps showing CRIS classification of primary tumours based on RNAseq data from the TCGA (g) and Affymetrix microarray data of the GSE14333 data set (h). (i) Column chart showing the distribution of CRIS subtypes based on NTP classification across 16 independent CRC data sets (Supplementary Data 11); NC (non-classified) indicates samples for which NTP could not confidently assign any of the subtypes (FDR40.2). NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications 4 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 0% 10% 20% 30% 40% 50% 60% 70% 0.01 0.1 1 10 100 1,000 Somatic mutation prevalence (per megabase) Fraction of genome affected by CNA Number of samples CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E a b c d RMHC = Right colon, Mucinous, Hypermutator, CIMP R MH C RMH C RMH C R MH C R MH C MSI 25 0 25 50 MSS N/A Negative Positive MSS MSI Figure 3 | Molecular and phenotypic characteristics of CRIS subtypes in the TCGA data set. (a) Clustering of CRIS classes based on similarity of the distances, as calculated by NTP, when classifying the TCGA data set. (b) Frequency of somatic single-nucleotide variations. Individual samples are represented as black dots; the mean of each CRIS subtype is plotted in light blue. (c) Extent of somatic copy number alterations. Samples are represented as red or black dots to denote MSI or MSS tumours, respectively; the mean of each CRIS subtype is plotted in light blue. (d) Distribution of the indicated phenotypic and genetic characteristics of CRC samples across CRIS subtypes. 0% 10% 20% 30% 40% 50% 60% 70% 0.01 0.1 1 10 100 1,000 Somatic mutation prevalence (per megabase) Fraction of genome affected by CNA CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E a b c MSS MSI classifier all the transcripts for which 50% or more of the signal was originated from stromal cells, based on the analysis of PDX RNAseq data (Methods). By applying this filter, we selected 565 genes with unambiguous epithelial expression (Suppleme- ntary Data 8). When applied to the PDX training data set (Fig. 2b), CRIS confidently classified 94% of the samples (Supplementary Data 7). Importantly, similar classification results were obtained in the original CRC-LM data set (Fig. 2c, Supplementary Data 9). All CRIS subtypes maintained a similar fraction of assigned cases in the PDX and CRC-LM data sets (Fig. 2d). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Collectively, these results indicate that the transcriptional traits captured by CRIS reflect stable intrinsic features of cancer cells and are not ostensibly affected by the origin of the samples or the expression analysis platform. Number of samples d RMHC = Right colon, Mucinous, Hypermutator, CIMP R MH C RMH C RMH C R MH C R MH C MSI 25 0 25 50 MSS N/A Negative Positive RMHC = Right colon, Mucinous, Hypermutator, CIMP Figure 3 | Molecular and phenotypic characteristics of CRIS subtypes Figure 3 | Molecular and phenotypic characteristics of CRIS subtypes Figure 3 | Molecular and phenotypic characteristics of CRIS subtypes in the TCGA data set. (a) Clustering of CRIS classes based on similarity of the distances, as calculated by NTP, when classifying the TCGA data set. (b) Frequency of somatic single-nucleotide variations. Individual samples are represented as black dots; the mean of each CRIS subtype is plotted in light blue. (c) Extent of somatic copy number alterations. Samples are represented as red or black dots to denote MSI or MSS tumours, respectively; the mean of each CRIS subtype is plotted in light blue. (d) Distribution of the indicated phenotypic and genetic characteristics of CRC samples across CRIS subtypes. Figure 3 | Molecular and phenotypic characteristics of CRIS subtypes in the TCGA data set. (a) Clustering of CRIS classes based on similarity of the distances, as calculated by NTP, when classifying the TCGA data set. (b) Frequency of somatic single-nucleotide variations. Individual samples are represented as black dots; the mean of each CRIS subtype is plotted in light blue. (c) Extent of somatic copy number alterations. Samples are represented as red or black dots to denote MSI or MSS tumours, respectively; the mean of each CRIS subtype is plotted in light blue. (d) Distribution of the indicated phenotypic and genetic characteristics of CRC samples across CRIS subtypes. odds ratio 17.309, confidence interval (CI)¼ 5.670–71.189; Fig. 3d). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Submap analysis23 confirmed that each subtype was associated with similar underlying transcriptional traits in the PDX and CRC-LM data sets (Supplementary Fig. 3). This indicates that the transcriptional patterns identified in PDXs are represented and detectable in their original counterparts, even in the presence of human stromal cells. This notion is further reinforced by the high level of concordance observed when comparing the classification of matched PDXs versus original tumour pairs (75% and kappa score ¼ 0.62 for CRIS versus 28–46% and kappa scores ranging from 0.2 to 0.3 for existing classifiers) (Fig. 2e,f; compare with Fig. 1d,e). a b 0.01 0.1 1 10 100 1,000 Somatic mutation prevalence (per megabase) b 0% 10% 20% 30% 40% 50% 60% 70% Fraction of genome affected by CNA c MSS MSI ) ( g p g ) Because CRIS was derived from metastatic samples grown as xenografts in mice, we wished to verify whether CRIS signatures could be also applied to CRC gene expression profiles obtained in different conditions and by means of diverse technologies. To this aim, we initially tested the CRIS classifier against two independent gene expression data sets of primary CRCs, the previously mentioned 450-sample TCGA RNAseq data set8,15 (Supplementary Data 1) and GSE14333, a 290-sample Affymetrix data set24 (Fig. 2g,h, Supplementary Data 10). In both instances, more than 90% of the samples were significantly assigned to CRIS classes (Supplementary Data 11). In the case of the TCGA data set, for which microsatellite instability status was available, CRIS classification efficiency was high in both microsatellite-stable (MSS) and microsatellite-unstable (MSI) samples (93% and 95% of confidently classified samples, respectively). This was particularly interesting because the CRIS training data set had very few MSI cases (Supplementary Data 4), as expected for metastatic CRC25. For additional independent CRIS validation, we applied the classifier to 14 other data sets11,26, including one Affymetrix microarray data set of liver metastases and one Illumina microarray data set of CRC cell lines (Supplementary Data 11). Overall, CRIS confidently classified 3,396 out of 3,738 samples (90.8%). Further, the five subtypes maintained a similar fraction of assigned cases, irrespective of technological platforms and experimental conditions (Fig. 2i). Submap analysis confirmed that the transcriptional attributes distinguishing the different subgroups were comparable in CRC-LM and primary tumours (Supplementary Fig. 4). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 HYPOXIA MSI-like IL2-STAT5 IL6-JAK-STAT6 TNFα-NFκB INTERFERONγ INTERFERONα EMT EGFR WNT MYC MTORC1 LGR5 BOTTOM CRYPT ALLOREJECTION TGFβ A B C D E 8q 13q 20 a +1.5 L2R –1.5 7 e PANETH 3.5 3 2.5 2 1.5 1 0.5 –0.5 0 MUCINOUS NEUROENDOCRINE GLYCOLYSIS INFLAMM. RESP. HYPOXIA MSI-like IL2-STAT5 IL6-JAK-STAT6 TNFα-NFκB INTERFERONγ INTERFERONα EMT EGFR WNT MYC MTORC1 LGR5 BOTTOM CRYPT ALLOREJECTION TGFβ a e EPHA4 EPHB4 EPHB3 NOTCH2 TNFRSF10D TNFRSF10B TNFRSF10A IL1R2 IL6ST NOTCH3 TNFRSF11B BMPR2 TNFRSF6B EPHA1 EPHA2 ACVR1 BMPR1A IGF2R FGFR4 FGFR3 ACVR1B EPHA8 ERBB3 EGFR f TCGA 3.5 3 2.5 2 1.5 1 –0.5 0.5 0 f A B C D E 8q.24.21 11p15.5 IGF2 BRAF KRAS TP53 b c d A B C D E +1.5 L2R –1.5 g CMS1 CMS2 CMS3 CMS4 CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E E C L E C L b g d BRAF KRAS TP53 d A B C D E CMS3 CMS4 Figure 4 | Genetic and functional features of CRIS classes in MSS colorectal cancer. (a) H Figure 4 | Genetic and functional features of CRIS classes in MSS colorectal cancer. (a) Heatmap showing the distribution of broad copy number changes displaying the top-five highest average variations in individual CRIS subtypes in MSS samples of the TCGA data set. (b) Heatmap showing selected focal amplifications in MSS samples of the TCGA data set. Significant subtype enrichments are marked by coloured boxes. (c) Distribution of cases overexpressing IGF2 according to the TCGA data set15. (d) Distribution of sequence alterations in BRAF, KRAS and TP53 in MSS samples extracted from the TCGA data set. Significant subtype enrichments are marked by coloured boxes. For enrichment analysis, Po0.05 by Fisher’s exact test was considered significant. (e) Radar plot showing phenotypic and functional characteristics of CRIS subtypes, expressed as normalized enrichment scores (NESs) calculated by GSEA. (f) Radar plot showing autocrine stimulation loops in CRIS subtypes, expressed as NESs and calculated based on gene expression analysis of ligand-receptor pairs. (g) Caleydo view of correspondences between CMS subtypes and CRIS classes in the TCGA data set. Heatmaps show estimates of stromal infiltration derived from gene expression analysis of specific signatures (C, CAFs; E, endothelial cells; L, leukocytes). Po5  10 3, odds ratio ¼ 10.417, CI¼ 2.031–69.611 for CIMP; Fig. 3d). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 These two classes proved to be also enriched for right- colon tumours featuring mucinous histology, CpG island methy- lator phenotype (CIMP) and a hypermutator genotype (Fisher’s exact test, CRIS-A/B against all other samples, Po5  10 3, odds ratio ¼ 2.701, CI¼ 1.459–5.038 for right-colon location; Po5  10 10, odds ratio ¼ 16.207, CI¼ 5.751–56.814 for mucinous histology; Po1  10 11, odds ratio ¼ 27.254, CI¼ 7.811–147.338 for CIMP; Po5  10 7, odds ratio ¼ 16.094, CI¼ 4.388–89.516 for hypermutator phenotype; Fig. 3d). Of note, some of these MSI-associated features were also shared by the MSS members of CRIS-A (‘MSI-like’ samples; Fisher’s exact test, CRIS- A MSS samples against all other MSS samples, Po5  10 6, odds ratio ¼ 23.231, CI¼ 5.318–144.640 for mucinous histology and Major peculiarities of CRIS classes. Using NTP-calculated centroid distances in the TCGA data set (Supplementary Data 11), CRIS classes could be clustered into two major subfamilies composed of two (CRIS-A and CRIS-B) and three (CRIS-C, CRIS-D and CRIS-E) members (Fig. 3a; Methods). As suggested by the asymmetric distribution of mutational and copy number load (Fig. 3b,c), TCGA MSI and MSS samples (Supplementary Data 1) were not equally partitioned across CRIS subfamilies. Specifically, MSI tumours were predominantly assigned to CRIS-A and—to a lower extent—to CRIS-B (Fisher’s exact test, CRIS-A/B against all other samples, Po5  10  10, 5 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 EPHA4 EPHB4 EPHB3 NOTCH2 TNFRSF10D TNFRSF10B TNFRSF10A IL1R2 IL6ST NOTCH3 TNFRSF11B BMPR2 TNFRSF6B EPHA1 EPHA2 ACVR1 BMPR1A IGF2R FGFR4 FGFR3 ACVR1B EPHA8 ERBB3 EGFR f A B C D E A B C D E 8q.24.21 11p15.5 IGF2 BRAF KRAS TP53 8q 13q 20 a b c d e g A B C D E CMS1 CMS2 CMS3 CMS4 CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E TCGA PANETH 3.5 3 2.5 2 1.5 1 0.5 –0.5 0 3.5 3 2.5 2 1.5 1 –0.5 0.5 0 MUCINOUS NEUROENDOCRINE GLYCOLYSIS INFLAMM. RESP. HYPOXIA MSI-like IL2-STAT5 IL6-JAK-STAT6 TNFα-NFκB INTERFERONγ INTERFERONα EMT EGFR WNT MYC MTORC1 LGR5 BOTTOM CRYPT ALLOREJECTION TGFβ +1.5 L2R –1.5 +1.5 L2R –1.5 E C L E C L 7 Figure 4 | Genetic and functional features of CRIS classes in MSS colorectal cancer. (a) Heatmap showing the distribution of broad copy number changes displaying the top-five highest average variations in individual CRIS subtypes in MSS samples of the TCGA data set. (b) Heatmap showing selected focal amplifications in MSS samples of the TCGA data set. Significant subtype enrichments are marked by coloured boxes. (c) Distribution of cases overexpressing IGF2 according to the TCGA data set15. (d) Distribution of sequence alterations in BRAF, KRAS and TP53 in MSS samples extracted from the TCGA data set. Significant subtype enrichments are marked by coloured boxes. For enrichment analysis, Po0.05 by Fisher’s exact test was considered significant. (e) Radar plot showing phenotypic and functional characteristics of CRIS subtypes, expressed as normalized enrichment scores (NESs) calculated by GSEA. (f) Radar plot showing autocrine stimulation loops in CRIS subtypes, expressed as NESs and calculated based on gene expression analysis of ligand-receptor pairs. (g) Caleydo view of correspondences between CMS subtypes and CRIS classes in the TCGA data set. Heatmaps show estimates of stromal infiltration derived from gene expression analysis of specific signatures (C, CAFs; E, endothelial cells; L, leukocytes). EPHA4 EPHB4 EPHB3 NOTCH2 TNFRSF10D TNFRSF10B TNFRSF10A IL1R2 IL6ST NOTCH3 TNFRSF11B BMPR2 TNFRSF6B EPHA1 EPHA2 ACVR1 BMPR1A IGF2R FGFR4 FGFR3 ACVR1B EPHA8 ERBB3 EGFR f e TCGA PANETH 3.5 3 2.5 2 1.5 1 0.5 –0.5 0 3.5 3 2.5 2 1.5 1 –0.5 0.5 0 MUCINOUS NEUROENDOCRINE GLYCOLYSIS INFLAMM. RESP. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 4d), while they were depleted in CRIS-A and D (Fisher’s exact test, CRIS-A or D against all other MSS samples, Po0.05 for both, odds ratio ¼ 0.361 with CI ¼ 0.132–0.942 for CRIS-A and 0.281 with CI ¼ 0.095–0.771 for CRIS-D; Fig. 4d). By leveraging the genetic annotations available in our data set of CRC PDXs (Supplementary Data 4), we were able to confirm several of the molecular correlates found in the TCGA cohort (Supplementary Fig. 5). Together, these findings indicate that CRIS subclasses are endowed with specific genetic traits that presumably drive functional features associated with the corresponding expression signatures. In summary, CRIS identified CRC subclasses with different functional and phenotypic characteristics related to specific molecular alterations: CRIS-A was an MSI-like, BRAF- or KRAS-mutated, and secretory subtype with sustained glycolytic metabolism and inflammatory traits; CRIS-B grouped a subset of poorly differentiated tumours with active TGFb signalling and EMT features; CRIS-C clustered KRAS wild-type CIN tumours exhibiting elevated ERBB/EGFR pathway activity and MYC copy number gains; CRIS-D tumours were characterized by a typical stem phenotype with high WNT signalling, associated with strong enrichment in IGF2 amplification/overexpression and FGFR autocrine stimulation; CRIS-E again displayed WNT-related features, but associated with a Paneth-like phenotype and a TP53-mutated genotype. To better delineate the functional attributes inherent to CRIS subtypes, we exploited both unbiased and supervised approaches. On the one hand, to gather a comprehensive picture of biological traits associated with CRIS classes, we ran gene set enrichment analysis (GSEA) against the ‘Hallmark’ gene sets, a panel of signatures representing an extensive spectrum of key biological functions, from the MSigDB (ref. 28) (Fig. 4e, Suppleme- ntary Data 12). On the other hand, to explore transcriptional patterns specifically relevant in CRC, we selected ten signatures capturing different phenotypic or functional aspects of normal and neoplastic intestinal biology (Supplementary Data 13) and analysed to what extent each sample displayed the phenotype captured by every signature. To do this, for each sample and each signature a transcriptional score was calculated by subtracting the average expression of the genes negatively associated with the phenotype from the average expression of the genes positively associated with the phenotype. Then, to test how the different CRIS subtypes are enriched for samples strongly displaying the selected transcriptional signature, we conceived a procedure named ‘Sample Set Enrichment Analysis’ (SSEA). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Overall, CRIS-C, D and E typically displayed chromosomal instability (CIN), characterized by heavy copy number alteration burden (Wilcoxon rank-sum test, Po5  10  08; Fig. 3c). More specifically, gains of entire chromosome arms were particularly evident for CRIS-C (for example, Chr7, Chr8q and Chr20) and Through an exploratory inspection of broad and focal copy number changes specifically in MSS samples, we identified genomic traits peculiar to individual CRIS classes (Fig. 4a,b). 6 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 ARTICLE CRIS-E (Chr13q) (Fig. 4a). Interestingly, CRIS C–D–E subtypes displayed significant association with specific focal amplifications (Supplementary Table 1). In particular, besides some alterations with unknown functional consequences, CRIS-C displayed focal amplification of 8q.24.21, which contains the MYC proto-oncogene (Fisher’s exact test, CRIS-C samples against all other MSS samples, Po0.01, odds ratio ¼ 2.802, CI ¼ 1.207–7.006; Fig. 4b). CRIS-D was specifically enriched for amplification of Chr11p15.5 (Fisher’s exact test, CRIS-D samples against all other MSS samples, Po1  10  05, odds ratio 6.020, CI ¼ 2.268–17.046; Fig. 4b). This locus contains IGF2, a gene whose strong overexpression has been reported in a subset of CRCs15. Accordingly, we observed that Chr11p15.5 amplification in CRIS-D was paralleled by high IGF2 expression levels (Fig. 4c). GSEA algorithm is employed to test ‘sample sets’ (in our case samples belonging to each CRIS subtype) for enrichment in high- rank samples (Fig. 4e, Supplementary Data 13; see Methods). Using SSEA, we also estimated the enrichment of each subtype in mitogenic/anti-apoptotic autocrine loops, as a proxy of growth factor-dependent oncogenic signalling. In this case, the samples were ranked based on ‘receptor activity scores’, which were calculated by averaging the expression of the receptor itself with that of its ligands (Fig. 4f, Supplementary Data 14; Methods). To avoid the confounding effect due to dilution of cancer-cell specific transcripts by stroma-derived RNA, all these analyses were performed on PDX GEPs. p Both GSEA-based and SSEA-based phenotypic analyses were concordant in attributing secretory and MSI-like features to CRIS-A, in agreement with the mucinous histology of this subgroup detected in the TCGA data set (Fig. 4e). CRIS-A was also associated with inflammatory traits and a glycolytic/hypoxic status (Fig. 4e), which is typical of some KRAS-mutated tumours29 and has been linked to intense secretory activity30,31. Similar to CRIS-A, CRIS-B featured an inflammatory phenotype (Fig. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 4e); but different from CRIS-A, CRIS-B also displayed strong TGFb activity and marked traits of epithelial– mesenchymal transition (EMT) (Fig. 4e). This is consistent with the established role of TGFb as an EMT inducer32. Interestingly, blinded pathological inspection of images from the TCGA cancer digital slide archive33 revealed that CRIS-B members included a large number of poorly differentiated tumours, in which the glandular architecture of the original tissue was completely lost or barely detectable (Supplementary Fig. 6). These preliminary observations deserve future, more extensive exploration. CRIS-C was mainly characterized by a combination of elevated ERBB1/3 autocrine stimulation loops and moderate WNT (Fig. 4e,f). CRIS-D showed traits of high intrinsic IGFR stimulation (in accordance with the strong prevalence of IGF2 amplification observed in the TCGA data set) and FGFR activity (Fig. 4f). Finally, CRIS-D—and, to a lesser extent, CRIS-E—featured high WNT activity and a bottom crypt phenotype. In particular, CRIS-D was positive for an LGR5 signature, which typifies intestinal stem cells34, and CRIS-E displayed a phenotype recalling WNT-producing Paneth cells (Fig. 4e,f). p y g p g To further investigate the genetic correlates of CRIS transcrip- tional classes, we took advantage of the mutational profiles available from the TCGA data portal15. As previously reported27, the prevalence of mutations in the BRAF gene was high in MSI samples (Supplementary Data 1). While the few MSS cases with BRAF alterations did not cluster in any particular CRIS subtype (Fig. 4d), 13 of the 15 MSI samples with mutated BRAF were assigned to CRIS-A (one with high FDR) and two to CRIS-B. Within MSS samples, KRAS was frequently mutated in CRIS-A (Fig. 4d; Fisher’s exact test, CRIS-A MSS against all other MSS samples, Po0.005, odds ratio ¼ 4.320, CI ¼ 1.570–13.274) and, within the CRIS C–D–E subfamily, specifically in CRIS-E (Fisher’s exact test, CRIS-E against CRIS-C/D, Po5  10  4, odds ratio ¼ 5.645, CI ¼ 2.059–16.120; Fig. 4d). CRIS-C was instead strongly enriched in KRAS wild-type samples (Fisher’s exact test, CRIS-C against all other MSS samples, Po1  10  4, odds ratio ¼ 0.161 CI ¼ 0.050–0.441; Fig. 4d). Finally, TP53 mutations prevalently occurred in CRIS-E (Fisher’s exact test, CRIS-E against all other MSS samples, Po0.01, odds ratio ¼ 3.892, CI ¼ 1.307–14.156; Fig. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 As sho CRIS-C Other CRIS 9 10 11 12 a b +1.5 A B C D E Sensit Resistant L2R –1.5 ALL CRIS-C OTHER CRIS KRAS 12/13 mutant Other resistance markers WT Cetuximab-sensitive Cetuximab-resistant c d e GSE5851 (KRAS 12/13 WT) Cetuximab-resistant Cetuximab-sensitive CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E CRIS-C Cetuximab-sensitive Cetuximab-resistant OTHER CRIS CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E PDX EGFR pathway activity geneset (average expression log2) P < 5×10–16 EGFR-pathway geneset Cetuximab sensitivity EGFR HER3 EREG HBEGF IRS2 a a Cetuximab-resistant Cetuximab-sensitive CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E b ALL CRIS-C OTHER CRIS KRAS 12/13 mutant Other resistance markers WT Cetuximab-sensitive Cetuximab-resistant PDX b The sensitivity of CRIS for more granular detection of functionally relevant cancer-cell intrinsic traits is particularly evident for CMS2. This subgroup, the vastest of the consensus, includes CIN tumours that show overall upregulation of WNT targets, and displays consistently low stromal scores. The CMS2 subtype was resolved by CRIS into three distinct groups that displayed class-specific features superimposed on a common background of high WNT signalling: CRIS-C is characterized by high EGFR pathway activity; CRIS-D is enriched for IGF2 overexpressors; CRIS-E contains a high number of samples with KRAS and TP53 mutations. Collectively, these findings further attest to the independent value and higher resolution of CRIS taxonomy. c GSE5851 (KRAS 12/13 WT) CRIS-C Cetuximab-sensitive Cetuximab-resistant OTHER CRIS CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E OTHER CRIS Cetuximab-sensitive Cetuximab-resistant CRIS-A CRIS-B CRIS-C CRIS-D CRIS-E CRIS-C Other CRIS 9 10 11 12 A B C D E d e EGFR pathway activity geneset (average expression log2) P < 5×10–16 d Figure 5 | CRIS-C and EGFR pathway activity predict cetuximab sensitivity. (a) Prevalence of CRIS classes among PDX cases scored as sensitive or resistant to cetuximab. (b) Bar-of-pie charts showing the distribution of validated resistance biomarkers (KRAS mutations in codons 12 and 13) and proposed resistance biomarkers (other resistance markers; for their enumeration, see Supplementary Data 4) in PDXs (pies). The fraction of sensitive or resistant cases specifically in the PDX subpopulation that does not harbour resistance biomarkers is shown in the bars. The analysis encompasses the whole PDX data set, CRIS-C alone, and all other CRIS classes. (c) Bar-of-pie chart showing the distribution of CRIS subtypes in KRAS wild-type cases of GSE5851, a data set of CRC liver metastases annotated for clinical response to cetuximab monotherapy37 (pie). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 orthogonally distributed across all five CRIS classes, with an enrichment for CRIS-B cases. These differences are largely attributable to the influence of stromal infiltration, as evidenced by analysing the distribution of stromal signatures individually expressed by CAFs, leukocytes or endothelial cells8 (C, L and E stromal scores, respectively) in the two classifiers. As shown by the heatmaps in Fig. 4g, asymmetric expression of stromal signatures (presented as the average expression of the component genes of each signature) was clearly detected in the CMS partition but not in the CRIS partition. Indeed, CMS1 samples displayed a high L score, in line with the fact that BRAF-mutant MSI CRC tumours, mostly included in CMS1, typically exhibit an extensive immune cell infiltration11 (Fig. 4g). Conversely, CRIS-A comprises BRAF-mutant MSI tumours but also KRAS-mutant MSS samples with MSI-like features, and overall has distinctive metabolic features. Accordingly, only a fraction of CRIS-A samples had a high L score, and this class received additional samples from CMS3, encompassing tumours with mixed MSI status, enrichment for KRAS mutations, and metabolic deregulation, but devoid of intense leukocyte infiltration. Most CMS4 tumours are characterized by very high levels of all stromal scores, in particular of the C score (Fig. 4g). The absence of overlaps with CRIS classes corroborates the notion that assignment to CMS4 mostly depends on stromal transcript contribution rather than intrinsic cancer-cell features. A fraction of the CMS1 and CMS4 samples converge into the CRIS-B group, featuring EMT and high TGF-b signalling; these pro-invasive traits are indeed assigned by CMS only to the CSM4 class, probably because they are estimated by stroma- derived transcripts. A similar lack of overlap with CRIS classification and asymmetric distribution of stromal scores across subtypes was also observed when previously published transcriptional CRC classifiers were applied to the TCGA data set (Supplementary Fig. 7). expressed by CAFs, leukocytes or endothelial cells8 (C, L stromal scores, respectively) in the two classifiers. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 In SSEA, samples are ranked by the calculated signature scores, then the g y Of note, many of these characteristics were not reported to associate with transcriptional subgroups in previous studies1–7, strongly suggesting that the elimination of stroma-related effects during the class discovery process enhanced sensitivity towards detection of intrinsic cancer cell-specific traits. To further support this notion, we analysed the distribution of CRIS subtypes across CMS classes assigned to TCGA samples11. Overall, partition of the samples by the two classifiers displayed limited overlap (Fig. 4g; see Supplementary Table 2 for cross-tabulation and statistics). Most CMS1 samples were assigned to CRIS-A and B; CMS2 samples were partitioned in CRIS-C, D and E; CMS3 contributed mostly to CRIS-A; and finally, CMS4 was 7 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 These alterations, particularly the clinically validated mutations in KRAS and NRAS genes, were under-represented in CRIS-C (Fisher’s exact test for KRAS and NRAS, Po1  10 13, odds ratio ¼ 0.066, CI ¼ 0.022–0.163; Fig. 5b). In principle, this could suggest that the higher rate of cetuximab-sensitive CRIS-C tumours is due to depletion in cases harbouring resistance-conferring alterations. However, CRIS-C maintained strong predictive power also when such cases were not factored in the analysis (Fisher’s exact test, Po1  10  5, odds ratio¼ 3.971, CI ¼ 2.08–7.08; Fig. 5b). As an external validation, we tested CRIS classification on a published data set of CRC-LM annotated for clinical response to cetuximab40 (GSE5851; Supplementary Data 15). Again, CRIS-C predicted cetuximab sensitivity both in the whole cohort (Fisher’s exact test, Po0.005, odds ratio¼ 6.780, CI ¼ 1.809–29.62; Supplementary Fig. 9) and, notably, also when KRAS-mutated samples were excluded from the analysis (Fisher’s exact test, Po0.05, odds ratio¼ 6.61, CI ¼ 1.07–74.094; Fig. 5c). py y To evaluate possible prognostic interactions between the CRIS-B subtype and high levels of CAF infiltration, we stratified tumours with high or low CAF content, as previously estimated in this data set8 (Supplementary Data 10). By this approach, CRIS-B was found to be informative specifically in low-CAF tumours (log-rank w2, Po5  10  3, HR ¼ 2.707, CI ¼ 1.384–5.295; Fig. 6d and Supplementary Fig. 13a). Symmetrically, high CAF content predicted poor prognosis only in non-CRIS-B tumours (log-rank w2, Po5  10  6, HR ¼ 5.216, CI ¼ 2.486–10.95; Supplementary Fig. 13b,c). These results suggest that CRIS-B membership and high CAF infiltration identify alternative means to acquire analogous traits of cancer aggressiveness, whose negative prognostic impact is not further exacerbated by the coexistence of the two (Supplementary Fig. 14). This observation prompted us to test a prognostic indicator that integrates CAF score and CRIS-B assignment, whereby all patients categorized as CRIS-B or high CAF, or assigned to both groups, were assumed to have a poor prognosis. The combined classifier performed better than either CAF score or CRIS-B alone (log-rank w2, Po5  10  5, HR ¼ 3.248, CI ¼ 1.843–5.724; Supplementary Fig. 15). As previously reported8, the contribution of the CAF score was negligible for treated patients (Supplementary Fig. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Bars denote the fraction of tumours responsive or resistant to cetuximab within CRIS-C or in all other CRIS classes. (d) Average expression of the gene set indicative of EGFR pathway activity in samples assigned to CRIS-C versus samples assigned to all other CRIS classes; error bars represent s.d. from the mean; statistical significance was calculated by Wilcoxon rank-sum test. (e) Heatmap showing response to cetuximab, average expression of the gene set indicative of EGFR pathway activity, and expression of individual genes of the EGFR pathway activity gene set across CRIS classes. The analysis includes PDXs that do not harbour known genetic markers of resistance. +1.5 A B C D E Sensitive Resistant L2R –1.5 e EGFR-pathway geneset Cetuximab sensitivity EGFR HER3 EREG HBEGF IRS2 e e 8 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 predictor of DFS not only in untreated cases (log-rank w2, Po0.05, HR ¼ 2.674, CI ¼ 1.045–6.845; Fig. 6b), as previously reported for high stromal content8,9, but also in cases that underwent adjuvant treatment (log-rank w2, Po0.005, HR ¼ 3.042, CI ¼ 1.366–6.777; Fig. 6c). This indicates that the negative prognostic value of CRIS-B is not biased by chemotherapy sensitivity. predictor of DFS not only in untreated cases (log-rank w2, Po0.05, HR ¼ 2.674, CI ¼ 1.045–6.845; Fig. 6b), as previously reported for high stromal content8,9, but also in cases that underwent adjuvant treatment (log-rank w2, Po0.005, HR ¼ 3.042, CI ¼ 1.366–6.777; Fig. 6c). This indicates that the negative prognostic value of CRIS-B is not biased by chemotherapy sensitivity. CRIS predicts response to anti-EGFR antibodies. To test whe- ther CRIS partitioning could predict drug sensitivity, we deployed expression profiles from 241 PDXs annotated for their response to the anti-EGFR antibody cetuximab12,35,36 (Supplementary Data 4, Supplementary Fig. 8). Interestingly, CRIS-C was significantly over-represented among cetuximab-sensitive tumours and depleted from resistant cases (Fisher’s exact test, Po5  10  10, odds ratio¼ 8.23, CI ¼ 4.017–17.545; Fig. 5a). The opposite was true for CRIS-A, which correlated with lack of response (Fisher’s exact test, Po1  10 5, odds ratio ¼ 0.104, CI ¼ 0.020–0.354; Fig. 5a). We and others have identified a number of genetic alterations that associate with resistance to cetuximab12,36–39. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 16) and, accordingly, the efficacy of the combined CRIS-CAF predictor was even more pronounced in untreated cases (log-rank w2, Po5  10  5, HR ¼ 5.736, CI ¼ 2.334–14.1; Fig. 6e). Complementary to genetic markers of resistance, the expression of a set of genes indicative of EGFR pathway activity has been found to correlate with cetuximab sensitivity35,36,40. This gene set displayed consistently higher expression in CRIS-C (Wilcoxon rank-sum test, CRIS-C samples against all other samples, Po1  10  8; Fig. 5d,e; Supplementary Fig. 10; Supplementary Tables 4 and 9). Moreover, expression of the gene set was significantly associated with cetuximab sensitivity across the whole data set, even after exclusion of all genetic markers of resistance (Wilcoxon rank-sum test, Po1  10  4; Supplementary Fig. 11a,b) and—of note—also when CRIS-C cases were excluded (Wilcoxon rank-sum test, Po5  10  4; Fig. 5e; Supplementary Fig 11c). Multiple logistic regression analysis demonstrated that CRIS classes, genetic markers and EGFR pathway activity transcripts have independent predictive value (Supplementary Table 3). This suggests that, although partially overlapping, these layers of information convey integrative knowledge, which could be combined to obtain more accurate prediction of cetuximab sensitivity. Among those individuals who did not receive adjuvant treatment based on clinical and pathological parameters, the integrative deployment of CRIS classes and CAF score was able to discriminate a relatively large subset (30%) of poor-prognosis patients who relapsed in 5 years in more than 40% of cases (sensitivity of 0.68 and specificity of 0.75 for 5-year DFS). The identification of patients with tumours that, in spite of displaying ‘benign’ clinical and pathological features, appear to have an aggressive biological behaviour, may prove useful to inform a more proactive follow up in such high-risk subpopulation. Importantly, the association between CRIS-B and poor prognosis was confirmed in an independent cohort of 1,261 samples, which was assembled by combining data from five independent data sets (Supplementary Fig. 17a; Supplementary Data 16). Also in this case, both CRIS-B and high CAF content predicted poor prognosis (Supplementary Fig. 17), but the combination of CRIS-B and high CAF outperformed either of the individual indicators (log-rank w2, Po1  10  7, HR ¼ 1.777, CI ¼ 1.433–2.204; Fig. 6f). CRIS is an independent predictor of CRC prognosis. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 CRIS-B vs others, treated cases GSE14333 CRIS-B vs others, untreated cases GSE14333 CRIS-B vs others, low CAF-score GSE14333 CRIS subtypes, all samples GSE14333 e b c d a f g 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving 1 DFS (years) N N N N 0 0.2 0.4 0.6 0.8 Fraction surviving 1 0 1 2 3 4 5 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving CRIS-B or high CAF-score vs others, untreated cases of GSE14333 P < 5×10–4 P < 5×10–5 P < 0.05 P < 0.005 P < 5×10–5 A B C D E CRIS Others CRIS-B Others CRIS-B Others CRIS-B Others CRIS-B/high-CAF Others CRIS-B/high-CAF CRIS-B or high CAF-score vs others, in 1,261 cases from 5 datasets P < 1×10–7 Clinical indication to adjuvant treatment High CAF score CRIS-B CRIS-B No adjuvant treatment Standard treatment Consider alternative Adjuvant treatment Y Y Y Y es predict CRC prognosis independent of clinical stage and stromal infiltration. (a) Kaplan–Meier pl ally annotated CRC gene expression data set. (b–e) Kaplan–Meier plot comparing the DFS of CRIS-B pa d not (b) or did receive (c) adjuvant chemotherapy and in cases with low (d) CAF infiltration. (e,f) Kap CAF infiltration or CRIS-B membership versus that of all other patients in the untreated population (n¼ extracted from six independent data sets (f). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 (g) Flowchart summarizing a possible decision tree taking NATURE COMMUNICATIONS | CRIS subtypes, all samples GSE14333 a 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving P < 5×10–4 A B C D E CRIS CRIS-B vs others, low CAF-score GSE14333 d 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving P < 5×10–5 Others CRIS-B d a CRIS-B vs others, untreated cases GSE14333 b 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving P < 0.05 Others CRIS-B b e 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving CRIS-B or high CAF-score vs others, untreated cases of GSE14333 P < 5×10–5 Others CRIS-B/high-CAF e e CRIS-B or high CAF-score vs others, untreated cases of GSE14333 CRIS-B vs others, treated cases GSE14333 c f g 1 DFS (years) N N N N 0 0.2 0.4 0.6 0.8 Fraction surviving 1 0 1 2 3 4 5 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving P < 0.005 Others CRIS-B Others CRIS-B/high-CAF CRIS-B or high CAF-score vs others, in 1,261 cases from 5 datasets P < 1×10–7 Clinical indication to adjuvant treatment High CAF score CRIS-B CRIS-B No adjuvant treatment Standard treatment Consider alternative Adjuvant treatment Y Y Y Y Figure 6 | CRIS subtypes predict CRC prognosis independent of clinical stage and stromal infiltration. (a) Kaplan–Meier plot o CRIS-B vs others, treated cases GSE14333 c 1 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving 0 1 2 3 4 5 P < 0.005 Others CRIS-B f f 1 0 1 2 3 4 5 DFS (years) 0 0.2 0.4 0.6 0.8 Fraction surviving Others CRIS-B/high-CAF CRIS-B or high CAF-score vs others, in 1,261 cases from 5 datasets P < 1×10–7 c CRIS-B or high CAF-score vs others, in 1,261 cases from 5 datasets Fraction surviving g N N N N Clinical indication to adjuvant treatment High CAF score CRIS-B CRIS-B No adjuvant treatment Standard treatment Consider alternative Adjuvant treatment Y Y Y Y g Figure 6 | CRIS subtypes predict CRC prognosis independent of clinical stage and stromal infiltration. (a) Kaplan–Meier plot of DFS of CRIS subtypes in GSE14333, a prognostically annotated CRC gene expression data set. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 To explore CRIS prognostic impact we examined the association of CRIS subtypes with disease-free survival (DFS) of 290 patients included in a publicly available and clinically annotated CRC data set24 (GSE14333; Supplementary Tables 10 and 11). The analysis revealed a trend towards better prognosis for CRIS-D tumours and a significant association with worse prognosis for CRIS-B tumours (log-rank w2, Po5  10  4, hazard ratio (HR) ¼ 2.961, CI ¼ 1.615–5.43; Fig. 6a). The prognostic value of CRIS-B was independent of tumour stage (Cox regression multivariate analysis, Po5  10  4; Supplementary Table 4) and was maintained in both Duke’s B and C patients (Supplementary Fig. 12). This suggests that CRIS- based stratification could be exploited in combination with clinical and pathological parameters for a superior prognostic assessment of CRC. Notably, CRIS-B membership was a negative g As an initial attempt to translate the CRIS taxonomy into a diagnostic tool amenable to clinical applications, we developed a single-sample classifier based on the top scoring pair algorithm (TSP)41 and its multiclass extension k-TSP42–44. A TSP is a binary predictor based on the relative ranking of two measurements (for example, the expression of a pair of transcripts), which switch order between two subclasses of samples. This approach can be extended to multiclass problems by identifying the TSPs associated with each pair-wise subclass comparison and then aggregating the votes across all gene pairs. We took advantage of such method to derive an algorithm for assignment of CRIS subtypes, which we named CRIS-TSP. To ensure cross-platform portability of the classifier, candidate TSP genes were challenged against a training data set of 624 gene NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Altogether, these data show that reducing the size of the CRIS classifier to 80 genes preserves most of its classifying capability across different gene expression platforms, and indicate the feasibility of deploying such a reduced gene set for a single-sample classification based on a TSP approach. To attempt unambiguous exploration of cancer-cell gene expression attributes we took advantage of a large collection of PDXs, in which transcripts of manifest cancer-cell origin could be extracted by the deployment of human-specific probes. The ensuing transcriptional profiles were then leveraged for a class discovery effort. This led to the identification of CRIS, an original classifier that categorizes CRC in five novel transcriptional classes. Of note, CRIS subclasses only barely overlap with the reported CRC transcriptional classification systems, which empowers a higher dimension of analytical resolution and refines biological insight into CRC heterogeneity. In particular, removal of stromal signals in the class discovery process resulted in remarkable orthogonality between CRIS and the recently published CMS signatures, with lack of classification for CMS subtypes enriched for mesenchymal phenotypes (CMS1 and CMS4) and detection of genetic and functional peculiarities with a potential to instruct novel diagnostic and therapeutic approaches. CRIS-A is highly enriched for BRAF-mutated MSI tumours and KRAS-mutated MSS tumours, for which no targeted treatment options are currently available. It is worth noting that such tumours typically exhibit strong glycolytic/hypoxic signatures. Although further studies based on preclinical experimentation and prospective trials in patients are needed to support this assumption, CRIS-A might pinpoint tumour subgroups potentially responsive to anti-metabolic therapies29. CRIS-B identifies a previously neglected subset of invasive tumours with poor prognosis and high TGF-b signalling. Since these characteristics are cancer cell specific, CRIS-B is unrelated to the CMS4 mesenchymal subtype, which also includes aggressive tumours and features transcriptional traits of TGF-b pathway activation, but of stromal origin. Novel therapeutic approaches targeting TGFb-mediated signals in the stroma have been recently proposed9; CRIS-B tumours could represent additional candidates for investigational testing of these drugs. CRIS-C categorizes tumours that are strictly dependent on EGFR signals and are sensitive to treatment with anti-EGFR antibodies. The positive predictive value of CRIS-C is of particular importance because it proved to be independent of all known (genetic) biomarkers of response or resistance. CRIS-D groups a subset of tumours enriched for IGF2 overexpression, which has been recently implicated in desensitization to EGFR blockade in patients with KRAS wild-type tumours35. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 CRIS-NTP80 also confirmed the prognostic significance of CRIS-B (log-rank w2, Po5  10  4, HR ¼ 1.581, CI ¼ 1.246–2.006; Supplementary Fig. 18). Altogether, these data show that reducing the size of the CRIS classifier to 80 genes preserves most of its classifying capability across different gene expression platforms, and indicate the feasibility of deploying such a reduced gene set for a single-sample classification based on a TSP approach. obscures biologically relevant transcriptional features inherently displayed by cancer cells. expression profiles from both PDXs and original tumours, obtained using multiple technological platforms (Supplementary Data 17). This process resulted in the selection of 40 gene pairs (Methods; Supplementary Data 18). When applied to the training set, CRIS-TSP demonstrated good concordance with the original CRIS classifier (K ¼ 0.7384). CRIS-TSP was then applied to eight independent data sets, for a total of 2,024 samples (Supplementary Data 17). When challenged against six gene expression data sets annotated for clinical outcome (total samples ¼ 1,487), CRIS-TSP assignments confirmed the poor prognosis of CRIS-B patients (log-rank w2, Po5  10  5, HR ¼ 1.6613, CI ¼ 1.307–2.112; Supplementary Fig. 18, Supplementary Data 17). The classification concordance between CRIS and CRIS-TSP was, however, suboptimal in these data sets (K ¼ 0.6459). We also found limited concordance when compar- ing CRIS-TSP classification of surgical resections of liver metastases versus their corresponding PDXs (K ¼ 0.4053), overall suggesting that the performance of the CRIS-TSP classifier was less robust when applied outside the training set. To assess whether such reduced classification coherence was due to low classification efficacy of the selected TSP genes, we reconstituted a NTP-based classifier using the same TSP transcripts (CRIS-NTP80, Supplementary Tables 21 and 22). When exploiting the CRIS-NTP80 algorithm for classification of the clinically annotated CRC data sets, the overall classification concordance with respect to the original full-size CRIS classifier was improved over CRIS-TSP (K ¼ 0.7149). Consistently, the classification coherence between original tumours and PDXs was also increased (K ¼ 0.5577) and similar to that obtained with the full-size NTP. CRIS-NTP80 also confirmed the prognostic significance of CRIS-B (log-rank w2, Po5  10  4, HR ¼ 1.581, CI ¼ 1.246–2.006; Supplementary Fig. 18). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 The finding that high IGF2 levels attenuate dependency on the EGFR pathway underscores the functional relevance of this alteration, which is also a candidate target for alternative treatment protocols35. Finally, CRIS-E aggregates KRAS-mutated, Paneth cell-like CIN tumours that are refractory to treatment with anti-EGFR antibodies. High WNT pathway activity was more generally observed in the CRIS-C–D–E subfamily, thus defining a subset of tumours for which pharmacologic inhibitors of this pathway48,49 may have therapeutic potential. As a further layer of relevant information for translational purposes, not only does CRIS introduce a new partitioning of known molecular traits, but it also puts forward a number of autocrine signalling loops that are selectively enriched in distinct classes. If validated through functional studies, these signals could constitute an entirely new population of candidate druggable targets for specific CRC subtypes. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 (b–e) Kaplan–Meier plot comparing the DFS of CRIS-B patients versus that of all other patients in cases who did not (b) or did receive (c) adjuvant chemotherapy and in cases with low (d) CAF infiltration. (e,f) Kaplan–Meier plot comparing the DFS of cases with high CAF infiltration or CRIS-B membership versus that of all other patients in the untreated population (n¼ 138) of GSE14333 (e) and in 1,261 unselected cases extracted from six independent data sets (f). (g) Flowchart summarizing a possible decision tree taking into account CRIS-B membership and high CAF infiltration, together with the clinical indication to adjuvant treatment, to refine the therapeutic decisional process after surgery. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications 10 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 expression profiles from both PDXs and original tumours, obtained using multiple technological platforms (Supplementary Data 17). This process resulted in the selection of 40 gene pairs (Methods; Supplementary Data 18). When applied to the training set, CRIS-TSP demonstrated good concordance with the original CRIS classifier (K ¼ 0.7384). CRIS-TSP was then applied to eight independent data sets, for a total of 2,024 samples (Supplementary Data 17). When challenged against six gene expression data sets annotated for clinical outcome (total samples ¼ 1,487), CRIS-TSP assignments confirmed the poor prognosis of CRIS-B patients (log-rank w2, Po5  10  5, HR ¼ 1.6613, CI ¼ 1.307–2.112; Supplementary Fig. 18, Supplementary Data 17). The classification concordance between CRIS and CRIS-TSP was, however, suboptimal in these data sets (K ¼ 0.6459). We also found limited concordance when compar- ing CRIS-TSP classification of surgical resections of liver metastases versus their corresponding PDXs (K ¼ 0.4053), overall suggesting that the performance of the CRIS-TSP classifier was less robust when applied outside the training set. To assess whether such reduced classification coherence was due to low classification efficacy of the selected TSP genes, we reconstituted a NTP-based classifier using the same TSP transcripts (CRIS-NTP80, Supplementary Tables 21 and 22). When exploiting the CRIS-NTP80 algorithm for classification of the clinically annotated CRC data sets, the overall classification concordance with respect to the original full-size CRIS classifier was improved over CRIS-TSP (K ¼ 0.7149). Consistently, the classification coherence between original tumours and PDXs was also increased (K ¼ 0.5577) and similar to that obtained with the full-size NTP. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications Methods A way to conclusively cope with this issue would be to exploit alternative methods to gather pure cancer cell transcriptional profiles from patient tumours, and test whether the key CRIS features remain valid. Such kind of approaches— mainly based on cell sorting of dissociated tumours or microdissection of histological specimens—have been already applied in small-scale efforts50–53, but need to be broadened to larger data sets for reliable validation54. For the remaining 515 samples—derived from 244 unique patients—probes were filtered to select those that showed detectable signal (detection P value ¼ 0) in at least 10% of the samples. For each of such genes, only the probe with the highest variance of signal was selected. Similar analyses were carried out for the panel of liver metastatic CRCs composing GSE73255. The panel included a total of 185 samples corresponding to 167 unique patients. Clinical and molecular annotations from the cancer genome atlas (TCGA). The TCGA data set was obtained from the TCGA data portal, as described previously8. Molecular/pathological annotations and IGF2 expression data were downloaded from the Supplementary Tables of the original work15, while mutational alteration and DFS data were retrieved from the cBioPortal (TGCA, Colorectal Adenocarcinoma Provisional data set; http://www.cbioportal.org/ index.do, see Supplementary Tables 1 and 20). The mutational load was calculated based on exome sequencing (Illumina) data available from the TCGA data portal. In particular the READ and COAD data sets were mined. All the somatic mutations were included in the calculation and normalized assuming an approximate exome size of 30 megabases. The analysis was carried out with the SomaticSignatures R package59. In conclusion, our data advocate the integrative exploitation of independently assessed cancer-cell intrinsic and stromal CRC transcriptional components as a key opportunity to improve patients’ management in the context of precision medicine approaches. Similar findings are beginning to emerge also in other tumour types, using different methodologies46. This suggests that the same basic concepts introduced here for CRC can be generalized, with wide impact on cancer diagnosis and treatment. To calculate the copy number variation load we took advantage of the segmented data provided by the Broad Institute and available from the TCGA data portal (Genome-Wide Human SNP Array 6.0). Only data unambiguously referred to unique tumour samples were considered. For each sample, all the regions with an absolute segmented value greater than 0.3 were categorized as altered. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 that the capture of cancer-cell intrinsic traits by CRIS can be efficiently integrated with stromal signatures to obtain even superior prognostic and predictive power. In particular, high CAF score and assignment to CRIS-B independently predict poor prognosis for almost one third of tumours whose clinical and pathological features would not dictate adjuvant treatment. Therapeutically, CAF score and CRIS-B can be integratively employed to stratify CRC patients in three groups: (i) patients with low CAF, non-CRIS-B tumours have good prognosis and may be spared adjuvant treatment; (ii) patients with high CAF-score tumours have poor prognosis but are likely responsive to adjuvant treatment with standard chemotherapy; (iii) patients with CRIS-B, low CAF-score tumours have poor prognosis and are expected to be mostly refractory to conventional adjuvant regimens, but they might benefit from investigational therapies with agents targeting the TGF-b pathway (Fig. 6g). These results call for prospective validation in larger cohorts for drawing definitive conclusions and, if confirmed, could have major clinical implications. The translation of the CRIS taxonomy into a clinically useful companion diagnostic would require the development of a tool for effective classification of individual patients. Here we show that a set of 40 gene pairs amenable to TSP-based single-sample classification retains the classification power of the original 565 gene classifier. However, the performance of CRIS-TSP was negatively affected by retrospective application to existing data sets, likely because of the diversity of procedures adopted and technological platforms used for data generation. Therefore, whenever the goal is to classify already available gene expression data sets obtained by diverse technological platforms (hybridization-based or sequencing-based), NTP-based CRIS categorization remains the option of choice. At the same time, we found that the TSP genes perform well when rechallenged for classification using the NTP approach. This suggests that implementation of this signature into a clinically applicable TSP-based single-sample classifier is feasible for prospective classification of new samples, for which dedicated and standardized data-generation procedures can be adopted. Methods To minimize the noise due to cross-species hybridization of transcripts deriving from murine infiltrates in PDX tissues, two pure murine samples were hybridized on human arrays8 in a pilot experiment, and all probes that generated detectable signals in this assay were removed from further analyses. Moreover, following an initial evaluation of signal quality and distribution, 14 samples were found to be contaminated by murine or human lymphomas and were not further considered in the analysis (9967501046_G, 9967501046_H, 6898368007_A, 6898368007_B, 9235792014_K, 9235792014_L, 8981245030_C, 8803828041_C, 9031292068_C, 9031292069_B, 5688887009_I, 8803828041_D, 8981245029_G, 8981245029_H). For the remaining 515 samples—derived from 244 unique patients—probes were filtered to select those that showed detectable signal (detection P value ¼ 0) in at least 10% of the samples. For each of such genes, only the probe with the highest variance of signal was selected. Similar analyses were carried out for the panel of liver metastatic CRCs composing GSE73255. The panel included a total of 185 samples corresponding to 167 unique patients. Microarray data preprocessing. For GSE76402, probe intensity data were extracted using the Illumina Genome Studio software (Genome Studio V2011.1) and subjected to Loess normalization using the Lumi R package57,58. To minimize the noise due to cross-species hybridization of transcripts deriving from murine infiltrates in PDX tissues, two pure murine samples were hybridized on human arrays8 in a pilot experiment, and all probes that generated detectable signals in this assay were removed from further analyses. Moreover, following an initial evaluation of signal quality and distribution, 14 samples were found to be contaminated by murine or human lymphomas and were not further considered in the analysis (9967501046_G, 9967501046_H, 6898368007_A, 6898368007_B, 9235792014_K, 9235792014_L, 8981245030_C, 8803828041_C, 9031292068_C, 9031292069_B, 5688887009_I, 8803828041_D, 8981245029_G, 8981245029_H). h l d d f b One potential limitation of our study is that some CRIS features could in fact emerge as a consequence of tumour xenotransplantation. In principle, the PDX approach might exert a number of distortive effects on the transcriptome of cancer cells, including selection drifts related to engraftment and propagation, limited cross-species reactivity between human and mouse cytokines with consequent perturbation of paracrine signals, and lack of proper immune components in recipient animals. However, the likelihood of a strong impact of such biases on the CRIS taxonomy is reduced by the observation that CRIS efficiently classified several data sets from bulk CRC patient tumours, regardless of their source of origin (primary or metastatic). NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications Methods Specimen collection and annotation. A total of 244 tumour samples and matched normal samples were obtained from patients who had undergone surgical resection of liver metastases at the Candiolo Cancer Institute, the Mauriziano Umberto I Hospital and the San Giovanni Battista Hospital (Torino, Italy). All patients provided informed consent and study approval was obtained from the review boards of the three institutions (the ‘Comitato Etico Istituto di Candiolo—FPO IRCCS’, the ‘Comitato Etico Azienda Ospedaliera Mauriziano Umberto I’ and the ‘Comitato Etico Azienda Ospedaliero-Universitaria Citta` della Salute e della Scienza’). PDX generation and annotation. Each collected sample was fragmented and either frozen or prepared for implantation subcutis as previously described12,55. Non-obese diabetic/severe combined immunodeficinet mice (4–6 weeks old, both males and females) were used for tumour implantation. At passage two, multiple samples were subjected to gene expression profiling: two samples for 225 tumours, three samples for 13 tumours and four samples for 10 tumours. Whenever possible (149/244 cases, 61%) the different samples analysed originated from independent propagation of regionally distinct areas of the same original tumour, with the aim to take into account intra-tumour heterogeneity. Genetic data and annotation of sensitivity to cetuximab were obtained as described previously12,36. In vivo experiments and related biobanking data were stored in the Laboratory Assistant Suite, a web-based, in-house developed data management system for automated data tracking56. All animal procedures were approved by the Animal Care Committee of the Candiolo Cancer Institute, in accordance with Italian legislation on animal experimentation. Microarray data generation. RNA was extracted using the miRNeasy Mini Kit (Qiagen), according to the manufacturer’s protocol. Synthesis of cDNA and biotinylated cRNA (from 500 ng total RNA) was performed using the IlluminaTotalPrep RNA Amplification Kit (Ambion), according to the manufacturer’s protocol. Quality assessment and quantitation of total RNA and cRNAs were performed with Agilent RNA kits on a Bioanalyzer 2100 (Agilent). Hybridization of cRNAs (750 ng) was carried out using Illumina Human 48 k gene chips (Human HT-12 V4 BeadChip). Array washing was performed by Illumina High Temp Wash Buffer for 10 min at 55 C, followed by staining using streptavidin-Cy3 dyes (Amersham Biosciences). Hybridized arrays were stained and scanned in a Beadstation 500 (Illumina). Microarray data preprocessing. For GSE76402, probe intensity data were extracted using the Illumina Genome Studio software (Genome Studio V2011.1) and subjected to Loess normalization using the Lumi R package57,58. Discussion G Gene expression analysis based on total RNA of bulk cancer tissues provides an aggregate portrait of the main components that make up the whole tumour ecosystem, including cancer cells, vessels, fibroblasts and immune cells. Although global differences in gene expression patterns have proved useful to distinguish cancer subtypes for effective disease stratification45, separating the molecular signatures of tissue compartments from measurements of total tumour samples is expected to provide higher resolution of biologically and clinically pertinent parameters8,46. p The contribution of individual tumour constituents to better capturing some cancer characteristics has been mainly documen- ted for stromal cells in several tumour types. For example, a signature reflecting response of human fibroblasts to serum, suggestive of active wounds, was found in a subgroup of cases at early stages, persisted during treatment, and predicted increased risk of metastasis and death in breast, lung and gastric carcinomas47. In pancreatic ductal adenocarcinomas, the integration of tumour- and stroma-specific gene expression profiles resulted in improved prognostic power over traditional signatures46. In CRC, stromal traits have been shown to critically impact cancer prognosis and response to therapy8,9. On the contrary, how cancer-cell intrinsic gene expression patterns influence subtype classification remains elusive, likely because the proportion of normal tissue lineages present in whole tumour transcriptomes acts as a dominant source of variation that Although the analysis of cancer-cell intrinsic traits can provide relevant information for CRC management, the contribution of the stromal compartment should not be overlooked. We and others have reported that the extent of stromal infiltration predicts poor outcome, resistance to radiotherapy and—possibly—sensitivity to chemotherapy8,9. Here we show 11 ARTICLE Methods This approach allowed us to dichotomize the ICOR values and classify the 526 intrinsic genes based on their reproducibility across platforms. Supplementary Figure 21b shows the distribution of the total ICOR along with the thresholds identified by the expectation- maximization algorithm. In our analyses we retained only the 268 most reproducible genes with mean ICOR40.488. Generation of the CRIS classifier. To identify gene signatures able to discriminate CRIS subtypes, we first focused on the samples that best represented each subtype, using silhouette width—as previously published1,62—to remove the samples with negative score values (n ¼ 90; Supplementary Fig. 19c) from each cluster. By applying significance analysis of microarrays21 on the remaining 425 samples, 1,083 genes differentially expressed across subtypes were identified. This was obtained by applying an FDR threshold of 0.005, which was selected to maximize the number of genes to be further selected with PAM4, while minimizing the overall error rate in cross-validation analyses. Through PAM, we then generated the shrunken centroids of each class by selecting the configuration that minimized the overall error rate in leave-one-out cross-validation analyses. This led to further prioritization of the discriminating transcripts to a total of 903 genes, with an overall error rate of 0.035 (Supplementary Fig. 19d). pp y g For implementation of an NTP-based classifier, we selected genes positively and specifically associated to each of the subtypes. Indeed, the PAM score represents the extent and sign of association of each gene to each class. Starting from the 903 genes selected as specified above, 102 genes did not have a positive PAM score for any of the classes (as a consequence of the centroid shrinkage procedure) and could not be used for NTP. Then, we deployed our published methodology8 to remove from the classifier those genes for which the major component of signal was defined as having stromal origin. To do so, we calculated the fraction of stromal (mouse) transcripts contributing to the overall signal of each gene using RNAseq data from CRC PDXs, in which mouse stroma substitutes the human stroma8. This estimate was exploited to exclude from further analyses 84 genes characterized by a stromal contribution above 50% (ref. 8), which brought to 717 the number of genes to be processed in subsequent steps. The NTP algorithm does not allow redundancy between the signatures used to assign membership to different classes. Methods NMF was performed with the predetermined number of clusters (K) varying from 2 to 6. Quality of the clustering was evaluated by the cophenetic coefficient, with K ¼ 5 gathering the highest value and K ¼ 2, K ¼ 3 displaying similarly high coefficients (Fig. 2a). Among these, K ¼ 5 was selected because it (i) demonstrated higher subtyping resolution, splitting K ¼ 2 and K ¼ 3 in smaller and more homogeneous subclusters (Supplementary Fig. 19a) and (ii) maintained, despite the smaller cluster size, very high cophenetic coefficient also when different s.d. thresholds were used to select the variable genes (Supplementary Fig. 19b). g y g transcriptional profiles from PDXs were exploited for class discovery. To take into account tumour heterogeneity in the process of subtype identification, PDXs g The TSP algorithm assigns a sample to a specific phenotype if gene A is larger than gene B, or to the other phenotype otherwise. Since there are five CRIS classes, there are ten possible pair-wise comparisons among the CRIS classes: (1) A versus B; (2) A versus C; (3) A versus D; (4) A versus E; (5) B versus C; (6) B versus D; (7) B versus E; (8) C versus D; (9) C versus E; and (10) D versus E. ( ) ( ) ( ) ( ) There are 138,075 possible TSPs that can be formed using all combinations of 526 genes. To avoid over-fitting, however, we limited the search space in the training phase by filtering out all genes that proved to be irreproducible across the three analytical platforms considered (Illumina, RNA-seq, and Affymetrix). To this end, we used the MergeMaid R-package to calculate a gene reproducibility index called ICOR65,66, which allows to identify genes that are reproducible across distinct data sets without relying on any phenotypic information. We calculated within each separate study, and for each pair of genes, the correlation coefficient of expression value ranks across subjects, and then retained only the genes for which such correlations agreed across studies. Supplementary Figure 21a shows the histograms, the observed and the null distributions (as obtained from 1,000 permutations) for the three pairwise integrative correlations across the three data sets. To select the most reproducible genes we analysed the total integrative correlation obtained by averaging the pairwise integrative correlations using the expectation-maximization (EM) algorithm67. Methods Affymetrix gene expression profiles of 80 metastatic40 and 290 primary CRCs24 were downloaded from GEO (GSE5851 and GSE14333, respectively). The data set provided by the Colorectal Cancer Subtyping Consortium, corresponding to GSE39582, GSE2109, GSE17536, GSE13294, GSE20916, GSE37892, GSE33113, GSE13067, GSE35896, GSE23878, GSE5851, PETACC3 and KFSYSCC, was downloaded from the synapse repository. For genes with multiple probe sets, those with the highest average levels were selected. p To this end we first identified candidate genes for classifier development starting from 526 CRIS genes (out of 565) in common across three distinct data sets obtained from different platforms: 254 PDXs analysed on Illumina microarrays, 104 RNA-seq samples from TCGA and 266 samples analysed on Affymetrix microarrays available from the public domain (gse13067, gse13294 and gse35896), for a total of 624 samples (Supplementary Data 17). We then developed our ranked-based classifier using the k-TSP algorithm and compared its agreement with the CRIS classification obtained using the original NTP. None of the 624 samples used to develop our k-TSP-based classifier was included in subsequent analyses investigating the clinical relevance of the CRIS classification. To implement our k-TSP classifier we performed the following analytical steps: (i) identification of the most robust and reproducible set of genes across the three analytical platforms using integrative correlation (ICOR)65,66; (ii) splitting of the entire combined data set of 624 samples into training (2/3 of samples) and internal test (1/3 of samples) sets balanced in terms of platform and CRIS class composition; (iii) development of the classification rule using the k-TSP algorithm on the training set; and (iv) assessment of the agreement between k-TSP and NTP using the internal test set. Identification of cancer-cell intrinsic transcriptional subclasses. The identifi- cation of cancer-cell intrinsic subtypes was performed by applying unsupervised clustering analysis, following consolidated methods1. All the available transcriptional profiles from PDXs were exploited for class discovery. To take into account tumour heterogeneity in the process of subtype identification, PDXs derived from the same original tumour were treated as independent samples. To maximize the portability of the results across multiple platforms, we restricted our analyses only to transcripts that were also explored in the RNAseq data set available from the TCGA data portal. We applied consensus-based NMF20 to the 1,084 most variable genes. In accordance with Sadanandam et al.1,62, these were selected by applying a threshold of s.d. ¼ 0.8 followed by sample data normalization N (0,1). Methods Thus, all genes featuring a positive PAM score for more than one class had to be non-redundantly assigned to one class only. To do so, we used our previously published procedure8 and assigned genes that were positively associated to more than one class to the best PAM scoring class only when the second highest value for assignment to another PAM class was at least 0.2 points lower. In all other cases (corresponding to 152 transcripts), the genes were excluded from the analysis. This procedure yielded an NTP-compatible classifier composed of 565 genes, with the following partition: 173 genes for CRIS-A; 73 genes for CRIS-B; 149 genes for CRIS-C; 86 genes for CRIS-D; 84 genes for CRIS-E (Supplementary Data 8). The whole analytical pipeline, from class discovery to the 565-gene NTP classifier, is shown in Supplementary Fig. 20. There are still 35,778 possible TSPs that can be formed using all combinations of the 268 most reproducible genes. Hence, to further avoid over-fitting, we limited the search space in the training phase by selecting only the top 50 most differentially expressed genes for each class (if available) using a Wilcoxon rank- sum test, and then by constraining pairing only between genes belonging to the CRIS classes being considered (that is, searching only pairs between CRIS-A and CRIS-B genes when training TSPs to distinguish CRIS-A from CRIS-B, and so on). We then identified k-TSPs for each of the ten pair-wise comparison between the CRIS classes. To select disjoint TSPs for each class comparison, the genes used to form pairs were omitted from the search in subsequent comparisons. In selecting the most discriminative TSPs we started from the comparisons between CRIS-B and the other classes, since this class showed prognostic value in our previous analyses. We then proceeded to compare CRIS-C to the other classes, since this class showed association with EGFR signalling and cetuximab sensitivity in our previous analyses. We then proceeded with the remaining class comparisons according to the total number of available genes to form the pairs, in increasing order. The order of the search was therefore as follows: CRIS-B versus E, versus D, versus C and versus A; CRIS-C versus E, versus D and versus A; CRIS-E versus D and versus A; CRIS-D versus A. Methods This threshold was chosen based on previous work, in which standard methods for calling a copy number alteration for a segment in GISTIC analysis of a single sample were defined60. The copy number load was calculated as the number of nucleotides included in such altered regions, relative to the sum of all nucleotides in all the 12 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 segments identified in the genome of the patient under consideration. The analysis was performed using pandas and scipy packages (http://pandas.pydata.org/; http://docs.scipy.org/doc/scipy/reference/). segments identified in the genome of the patient under consideration. The analysis was performed using pandas and scipy packages (http://pandas.pydata.org/; http://docs.scipy.org/doc/scipy/reference/). Nearest template prediction. NTP-based classification16 was performed using scripts from the GenePattern Bioportal63 (see URLs). The threshold chosen for significant classification of a sample was Benjamini–Hochberg-corrected false discovery rate (BH.FDR)o0.2, as previously reported1. When referring to published classifications (that is, Fig. 1b, Supplementary Figs 2b,c and 4g), we employed the class assignments provided in the original works by Sadanandam et al.1 and Guinney et al.11, respectively. p py g py To represent gains and losses of whole chromosome arms, we employed the GISTIC61 results provided by the Broad Institute for the TCGA data set (broad_values_by_arm.txt). Focal amplifications were obtained from the GISTIC analysis48 provided by the Broad Institute in the Firehose portal (http://gdac.broadinstitute.org) (Supplementary Data 1). Focal amplifications were obtained from the GISTIC analysis48 provided by the Broad Institute in the Firehose portal (http://gdac.broadinstitute.org) (Supplementary Data 1). Development of a simplified single-sample CRIS classifier. To develop a simplified classification system for CRIS, we used the TSP approach, a rank-based, parameter-free binary predictor relying on the relative ordering of two features (for example, the order of expression of two genes), and its extension, the k-TSP classifier, which aggregates the votes of multiple TSPs and can be used for multiclass problems41–43,64 as detailed below. Public metastatic and primary colorectal cancer gene expression profiles. Affymetrix gene expression profiles of 80 metastatic40 and 290 primary CRCs24 were downloaded from GEO (GSE5851 and GSE14333, respectively). The data set provided by the Colorectal Cancer Subtyping Consortium, corresponding to GSE39582, GSE2109, GSE17536, GSE13294, GSE20916, GSE37892, GSE33113, GSE13067, GSE35896, GSE23878, GSE5851, PETACC3 and KFSYSCC, was downloaded from the synapse repository. For genes with multiple probe sets, those Public metastatic and primary colorectal cancer gene expression profiles. NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications Signature scorej ¼ ModuleUPj  ModuleDOWNj; where ModuleUP is the average expression value of the genes positively correlated with the phenotype and ModuleDOWN is the average expression value of the transcripts anti-correlated with the phenotype in sample j. 11. Guinney, J. et al. The consensus molecular subtypes of colorectal cancer. Nat. Med. 21, 1350–1356 (2015). 12. Bertotti, A. et al. A molecularly annotated platform of patient-derived xenografts (‘xenopatients’) identifies HER2 as an effective therapeutic target in cetuximab-resistant colorectal cancer. Cancer Discov. 1, 508–523 (2011). y j We then evaluated the enrichment in class assignment for each CRIS class by performing GSEA preranked analysis using as ranked lists the samples ordered by the score of interest, and as sets the lists of sample membership to the different CRIS subtypes. Calculations were done with 1,000 permutations. 13. Chou, J. et al. Phenotypic and transcriptional fidelity of patient-derived colon cancer xenografts in immune-deficient mice. PLoS ONE 8, e79874 ð2013Þ: For SSEA of autocrine loops, a total of 472 receptor/ligand interactions were downloaded from the Database of Ligand Receptor Partners (http://dip.doe- mbi.ucla.edu/dip/DLRP.cgi). Then, each set of genes encoding the receptor and its ligands was used to rank samples, based on median-centered Log2 ratios of gene expression values, as follows: 14. Julien, S. et al. Characterization of a large panel of patient-derived tumor xenografts representing the clinical heterogeneity of human colorectal cancer. Clin. Cancer Res. 18, 5314–5328 (2012). 15. Network, C. G. A. Comprehensive molecular characterization of human colon and rectal cancer. Nature 487, 330–337 (2012). Receptor Activation Scorej ¼ Receptorj þ Ligandsj 2 ; 16. Hoshida, Y. Nearest template prediction: a single-sample-based flexible class prediction with confidence assessment. PLoS ONE 5, e15543 (2010). where Receptor is the mRNA expression of the receptor of interest and Ligands stands for the average expression of its corresponding ligands in sample j. 17. Gao, H. et al. High-throughput screening using patient-derived tumor xenografts to predict clinical trial drug response. Nat. Med. 21, 1318–1325 (2015). 18. Kang, N., Gores, G. J. & Shah, V. H. Hepatic stellate cells: partners in crime for liver metastases? Hepatology 54, 707–713 (2011). Statistical analyses. Statistical analyses were performed by two-tailed Student’s t-test, Wilcoxon rank test, Fisher’s exact test and w2-test using the GraphPad Prism software or statistical R packages. For survival analyses, Cox regression hazard model and Kaplan–Meier analyses were conducted using the R-Bioconductor ‘survival’ package7,22. For all tests, the level of statistical significance was set at Po0.05. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 k-TSP classifier and available for analysis (that is, discard TSPs for which the genes were not measured); (ii) count the proportion of votes for each CRIS class (that is, the number of votes for a specific class divided by the total number of comparisons involving that class); (iii) assign each sample to the class with the highest proportion of votes; (iv) in case of ties between three classes or more, do not classify sample (‘Unassigned’); (v) in case of ties between two classes, assign samples to the class with the highest proportion of votes in the pair-wise comparisons between these two classes; do not classify sample otherwise (‘Unassigned’). Code availability. Essential scripts to reproduce our classification on independent CRC data sets can be found in R package CRISclassifier, provided as Supplementary Software to the manuscript. The CRISclassifier package was designed to provide easy and guided access to the full CRIS classification criteria and the TSP-based implementation. Updated versions of this package will be available on Bioconductor (package name: CRISclassifier). Contact claudio.isella@ircc.it for more information. g In the training phase on 2/3 of the data, we evaluated the concordance between the classification obtained using the k-TSP classifier and the reference NTP classification based on the complete 526 ‘intrinsic’ gene set. We assessed the agreement starting from the minimum set of ten TSPs (k ¼ 1), then incremented the number of TSPs until reaching the maximum agreement with NTP. Hence, we developed our kTSP classifier using 10, 20, 30, 40 and 50 non-overlapping TSPs for a total of 20, 40, 60, 80 and 100 genes, respectively. We then compared the agreement of the aggregated k-TSP classifier with NTP classification in the training set and internal testing set and selected the one showing the highest agreement in the internal testing set (k ¼ 4 for each pair-wise comparison, for a total of 40 TSPs and 80 genes), as shown in Supplementary Table 5. The 80 genes were also used for NTP-based classification (CRIS-NTP80), by assigning to each CRIS class the 16 TSP genes calling for that class in the pairwise comparisons. Data availability. Gene expression microarray data generated in the course of this study have been deposited in the GEO database with accession number GSE76402 (PDX data, 529 profiles from 244 patients) and GSE73255 (liver metastases data, 185 profiles from 167 patients). References 1. Sadanandam, A. et al. A colorectal cancer classification system that associates cellular phenotype and responses to therapy. Nat. Med. 19, 619–625 (2013). SubMap. The similarity between transcriptional traits underlying subtypes across different data sets was evaluated using the SubMap R package23 available from gene pattern (http://www.broadinstitute.org/cancer/software/genepattern/). 2. De Sousa E Melo, F. et al. Poor-prognosis colon cancer is defined by a molecularly distinct subtype and develops from serrated precursor lesions. Nat. Med. 19, 614–618 (2013). Principal component analysis. Principal component analysis (PCA) was employed to identify the main sources of variation in the TCGA gene expression data set. In particular, we exploited the PCA class from Scikit-learn python library (http://scikit-learn.org/stable/). Enrichment analyses were conducted on the first principal component (PCA0; Supplementary Data 2). 3. Marisa, L. et al. Gene expression classification of colon cancer into molecular subtypes: characterization, validation, and prognostic value. PLoS Med. 10, e1001453 (2013). 4. Roepman, P. et al. Colorectal cancer intrinsic subtypes predict chemotherapy benefit, deficient mismatch repair and epithelial-to-mesenchymal transition. Int. J. Cancer 134, 552–562 (2014). Comparison of CRC-LM and PDX gene expression profiles. Genes with high variance (0.8) in the PDX data set were used to calculate Pearson’s correlations for any possible CRC-LM/PDX permutation. Correlations of matched versus unmatched CRC-LM/PDX pairs were compared by two-tailed Student’s t-test. Comparison of CRC-LM and PDX gene expression profiles. Genes with high variance (0.8) in the PDX data set were used to calculate Pearson’s correlations for any possible CRC-LM/PDX permutation. Correlations of matched versus unmatched CRC-LM/PDX pairs were compared by two-tailed Student’s t-test. 5. Budinska, E. et al. Gene expression patterns unveil a new level of molecular heterogeneity in colorectal cancer. J. Pathol. 231, 63–76 (2013). 6. Schlicker, A. et al. Subtypes of primary colorectal tumors correlate with response to targeted treatment in colorectal cell lines. BMC Med. Genomics 5, 66 (2012). Gene set enrichment analyses. The GSEA software was downloaded from the http://www.broadinstitute.org/gsea/index.jsp. The significance of enrichment was estimated using default settings and 1,000 gene permutations28,68. 7. Perez-Villamil, B. et al. Colon cancer molecular subtypes identified by expression profiling and associated to stroma, mucinous type and different clinical behavior. BMC Cancer 12, 260 (2012). 8. Isella, C. et al. Stromal contribution to the colorectal cancer transcriptome. Nat. Genet. 47, 312–319 (2015). Sample set enrichment analysis. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Other gene-expression data that support the findings of this study are available through the TCGA data portal (TCGAcrcmRNA; URL http://bioconductor.org/ packages/release/data/experiment/html/TCGAcrcmRNA.html); from the GEO database (GSE5851, GSE14333; URL https://www.ncbi.nlm.nih.gov/geo/) and from the Synapse data portal from the Colorectal Cancer Molecular Subtyping Consortium (GSE39582, GSE2109, GSE17536, GSE13294, GSE20916, GSE37892, GSE33113, GSE13067, GSE35896, GSE23878, GSE5851, PETACC3 and KFSYSCC; URL http://sagebase.org/research-projects/colorectal-cancer-subtyping- consortium-crcsc/). g g p p Essential scripts required to apply the CRIS classifiers (NTP and TSP) to independent tumour data sets are provided with the CRISclassifier R package, available as a Supplementary Software. References For SSEA of curated functional signatures, a score was calculated for each signature and each sample using median-centered Log2 ratios of gene expression values, as follows: 9. Calon, A. et al. Stromal gene expression defines poor-prognosis subtypes in colorectal cancer. Nat. Genet. 47, 320–329 (2015). 10. Sadanandam, A. et al. Reconciliation of classification systems defining molecular subtypes of colorectal cancer: interrelationships and clinical implications. Cell Cycle 13, 353–357 (2014). Methods For each of the ten pair-wise comparisons we selected from 1 to 5 TSPs, for a total of 10, 20, 30, 40 and 50 disjoint TSPs, using a total of 20, 40, 60, 80 and 100 non-overlapping genes, respectively. The classification rule to assign a new sample to a CRIS class was constructed as There are still 35,778 possible TSPs that can be formed using all combinations of the 268 most reproducible genes. Hence, to further avoid over-fitting, we limited the search space in the training phase by selecting only the top 50 most differentially expressed genes for each class (if available) using a Wilcoxon rank- sum test, and then by constraining pairing only between genes belonging to the CRIS classes being considered (that is, searching only pairs between CRIS-A and CRIS-B genes when training TSPs to distinguish CRIS-A from CRIS-B, and so on). We then identified k-TSPs for each of the ten pair-wise comparison between the CRIS classes. To select disjoint TSPs for each class comparison, the genes used to form pairs were omitted from the search in subsequent comparisons. In selecting the most discriminative TSPs we started from the comparisons between CRIS-B and the other classes, since this class showed prognostic value in our previous analyses. We then proceeded to compare CRIS-C to the other classes, since this class showed association with EGFR signalling and cetuximab sensitivity in our previous analyses. We then proceeded with the remaining class comparisons according to the total number of available genes to form the pairs, in increasing order. The order of the search was therefore as follows: CRIS-B versus E, versus D, versus C and versus A; CRIS-C versus E, versus D and versus A; CRIS-E versus D and versus A; CRIS-D versus A. For each of the ten pair-wise comparisons we selected from 1 to 5 TSPs, for a total of 10, 20, 30, 40 and 50 disjoint TSPs, using a total of 20, 40, 60, 80 and 100 non-overlapping genes, respectively. Th l ifi i l i l CRIS l d The classification rule to assign a new sample to a CRIS class was constructed as follows: (i) classify each sample according to each individual TSP included in the 13 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications Signature scorej ¼ ModuleUPj  ModuleDOWNj; In case of multiple testing, the results were considered significant when the Benjamini–Hochberg FDR was below 0.2. 19. Gre´goire, M. & Lieubeau, B. The role of fibroblasts in tumor behavior. Cancer Metastasis Rev. 14, 339–350 (1995). 20. Brunet, J. P., Tamayo, P., Golub, T. R. & Mesirov, J. P. Metagenes and molecular pattern discovery using matrix factorization. Proc. Natl Acad. Sci. USA 101, 4164–4169 (2004). 14 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 21. Tusher, V. G., Tibshirani, R. & Chu, G. Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl Acad. Sci. USA 98, 5116–5121 (2001). 49. Lau, T. et al. A novel tankyrase small-molecule inhibitor suppresses APC mutation-driven colorectal tumor growth. Cancer Res. 73, 3132–3144 (2013). 22. Tibshirani, R., Hastie, T., Narasimhan, B. & Chu, G. Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc. Natl Acad. Sci. USA 99, 6567–6572 (2002). 50. Calon, A. et al. Dependency of colorectal cancer on a TGF-b-driven program in stromal cells for metastasis initiation. Cancer Cell 22, 571–584 (2012). 51. Kagawa, Y. et al. Cell cycle-dependent Rho GTPase activity dynamically regulates cancer cell motility and invasion in vivo. PLoS ONE 8, e83629 ð2013Þ: 23. Hoshida, Y., Brunet, J. P., Tamayo, P., Golub, T. R. & Mesirov, J. P. Subclass mapping: identifying common subtypes in independent disease data sets. PLoS ONE 2, e1195 (2007). ð Þ 52. Tsukamoto, S. et al. Clinical significance of osteoprotegerin expression in human colorectal cancer. Clin. Cancer Res. 17, 2444–2450 (2011). 24. Jorissen, R. N. et al. Metastasis-associated gene expression changes predict poor outcomes in patients with dukes stage B and C colorectal cancer. Clin. Cancer Res. 15, 7642–7651 (2009). 53. Skrzypczak, M. et al. Modeling oncogenic signaling in colon tumors by multidirectional analyses of microarray data directed for maximization of analytical reliability. PLoS ONE 5, e13091 (2010). 25. Haddad, R. et al. Microsatellite instability as a prognostic factor in resected colorectal cancer liver metastases. Ann. Surg. Oncol. 11, 977–982 (2004). 54. Dunne, P. D. et al. Challenging the cancer molecular stratification dogma: intratumoral heterogeneity undermines consensus molecular subtypes and potential diagnostic value in colorectal cancer. Clin. Cancer Res. 22, 4095–4104 (2016). 26. Medico, E. et al. The molecular landscape of colorectal cancer cell lines unveils clinically actionable kinase targets. Nat. Commun. 6, 7002 (2015). 55. Galimi, F. et al. Signature scorej ¼ ModuleUPj  ModuleDOWNj; Genetic and expression analysis of MET, MACC1, and HGF in metastatic colorectal cancer: response to met inhibition in patient xenografts and pathologic correlations. Clin. Cancer Res. 17, 3146–3156 (2011). 27. Lochhead, P. et al. Microsatellite instability and BRAF mutation testing in colorectal cancer prognostication. J. Natl Cancer Inst. 105, 1151–1156 (2013). 56. Baralis, E., Bertotti, A., Fiori, A. & Grand, A. LAS: a software platform to support oncological data management. J. Med. Syst. 36, S81–S90 (2012). 28. Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545–15550 (2005). 57. Gentleman, R. C. et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5, R80 (2004). 29. Kimmelman, A. C. Metabolic dependencies in RAS-driven cancers. Clin. Cancer Res. 21, 1828–1834 (2015). 58. Du, P., Kibbe, W. A. & Lin, S. M. lumi: a pipeline for processing Illumina microarray. Bioinformatics 24, 1547–1548 (2008). 30. Chaika, N. V. et al. MUC1 mucin stabilizes and activates hypoxia-inducible factor 1 alpha to regulate metabolism in pancreatic cancer. Proc. Natl Acad. Sci. USA 109, 13787–13792 (2012). y f 59. Gehring, J. S., Fischer, B., Lawrence, M. & Huber, W. SomaticSignatures: inferring mutational signatures from single-nucleotide variants. Bioinformatics 31, 3673–3675 (2015). g, J , , , , , g inferring mutational signatures from single-nucleotide variants. Bioinformatics 31, 3673–3675 (2015). 31. Shukla, S. K. et al. MUC16-mediated activation of mTOR and c-Myc reprograms pancreatic cancer metabolism. Oncotarget 6, 19118–19131 ð2015Þ: 60. Zack, T. I. et al. Pan-cancer patterns of somatic copy number alteration. Nat. Genet. 45, 1134–1140 (2013). 61. Mermel, C. H. et al. GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol. 12, R41 (2011). ð Þ 32. Thiery, J., Acloque, H., Huang, R. & Nieto, M. Epithelial-mesenchymal transitions in development and disease. Cell 139, 871–961 (2009). 33. Gutman, D. A. et al. Cancer digital slide archive: an informatics resource to support integrated in silico analysis of TCGA pathology data. J. Am. Med. Inform. Assoc. 20, 1091–1098 (2013). 62. Verhaak, R. G. et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 17, 98–110 (2010). f 34. Merlos-Suarez, A. et al. The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse. Acknowledgements We thank our friends and colleagues of the Laboratories of Translational Cancer Medicine and Oncogenomics for suggestions and discussion; Alessandro Ferrero, Francesco Giraldi, Alfredo Mellano, Andrea Muratore, Mauro Papotti, Gianluca Paraluppi and Anna Sapino for sample acquisition; and Antonella Cignetto, Daniela Gramaglia and Francesca Natale for secretarial assistance. Work in the authors’ laboratory was supported by AIRC, Associazione Italiana per la Ricerca sul Cancro, Investigator Grants 15571 (A.B.), 12944 (E.M.), and 14205 (L.T.). AIRC 2010 Special Program Molecular Clinical Oncology 5  1,000, project 9970 (E.M. and L.T.). 2012 Fight Colorectal Cancer—AACR, American Association for Cancer Research, Career Development Award in memory of Lisa Dubow, Grant Number 12-20-16-BERT (A.B.). MIUR FIRB, Ministero dell’Universita` e della Ricerca, Fondo per gli Investimenti della Ricerca di Base—Futuro in Ricerca (A.B.). Fondazione Piemontese per la Ricerca sul Cancro-ONLUS, 5  1,000 Ministero della Salute 2010 (E.M.) and 2011 (E.M. and L.T.). Transcan, TACTIC (L.T.). NIH grant P30CA006973 (L.M.). A.B., E.M. and L.T. are members of the EurOPDX Consortium. We thank our friends and colleagues of the Laboratories of Translational Cancer Medicine and Oncogenomics for suggestions and discussion; Alessandro Ferrero, Francesco Giraldi, Alfredo Mellano, Andrea Muratore, Mauro Papotti, Gianluca Paraluppi and Anna Sapino for sample acquisition; and Antonella Cignetto, Daniela Gramaglia and Francesca Natale for secretarial assistance. Work in the authors’ laboratory was supported by AIRC, Associazione Italiana per la Ricerca sul Cancro, Investigator Grants 15571 (A.B.), 12944 (E.M.), and 14205 (L.T.). AIRC 2010 Special Program Molecular Clinical Oncology 5  1,000, project 9970 (E.M. and L.T.). 2012 Fight Colorectal Cancer—AACR, American Association for Cancer Research, Career Development Award in memory of Lisa Dubow, Grant Number 12-20-16-BERT (A.B.). MIUR FIRB, Ministero dell’Universita` e della Ricerca, Fondo per gli Investimenti della Ricerca di Base—Futuro in Ricerca (A.B.). Fondazione Piemontese per la Ricerca sul Cancro-ONLUS, 5  1,000 Ministero della Salute 2010 (E.M.) and 2011 (E.M. and L.T.). Transcan, TACTIC (L.T.). NIH grant P30CA006973 (L.M.). A.B., E.M. and L.T. are members of the EurOPDX Consortium. 41. Geman, D., d’Avignon, C., Naiman, D. Q. & Winslow, R. L. Classifying gene expression profiles from pairwise mRNA comparisons. Stat. Appl. Genet. Mol. Biol. 3, 19 (2004). 42. Tan, A. C., Naiman, D. Q., Xu, L., Winslow, R. L. & Geman, D. Simple decision rules for classifying human cancers from gene expression profiles. Bioinformatics 21, 3896–3904 (2005). f 43. Marchionni, L., Afsari, B., Geman, D. & Leek, J. T. Signature scorej ¼ ModuleUPj  ModuleDOWNj; Cell Stem Cell 8, 511–524 (2011). 3. Reich, M. et al. GenePattern 2.0. Nat. Genet. 38, 500–501 (200 64. Afsari, B., Fertig, E. J., Geman, D. & Marchionni, L. switchBox: an R package for k-Top Scoring Pairs classifier development. Bioinformatics 31, 273–274 (2015). 35. Zanella, E. R. et al. IGF2 is an actionable target that identifies a distinct subpopulation of colorectal cancer patients with marginal response to anti- EGFR therapies. Sci. Transl. Med. 7, 272ra12 (2015). 65. Cope, L., Zhong, X., Garrett, E. & Parmigiani, G. MergeMaid: R tools for merging and cross-study validation of gene expression data. Stat. Appl. Genet. Mol. Biol. 3, 29 (2004). 36. Bertotti, A. et al. The genomic landscape of response to EGFR blockade in colorectal cancer. Nature 526, 263–267 (2015). 66. Parmigiani, G., Garrett-Mayer, E. S., Anbazhagan, R. & Gabrielson, E. A cross- study comparison of gene expression studies for the molecular classification of lung cancer. Clin. Cancer Res. 10, 2922–2927 (2004). 37. De Roock, W. et al. Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. Lancet Oncol. 11, 753–762 (2010). 67. Dempster, A. P., Laird, N. M. & Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B 39, 1–38 (1977). 38. Douillard, J. Y. et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. N. Engl. J. Med. 369, 1023–1034 (2013). 68. Mootha, V. K. et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat. Genet. 34, 267–273 (2003). g 39. Di Nicolantonio, F. et al. Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer. J. Clin. Oncol. 26, 5705–5712 (2008). 40. Khambata-Ford, S. et al. Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J. Clin. Oncol. 25, 3230–3237 (2007). Acknowledgements A simple and reproducible breast cancer prognostic test. BMC Genomics 14, 336 (2013). 44. Sung, J. et al. Multi-study integration of brain cancer transcriptomes reveals organ-level molecular signatures. PLoS Comput. Biol. 9, e1003148 (2013). 45. Sørlie, T. Molecular portraits of breast cancer: tumour subtypes as distinct disease entities. Eur. J. Cancer 40, 2667–2675 (2004). 46. Moffitt, R. A. et al. Virtual microdissection identifies distinct tumor- and stroma-specific subtypes of pancreatic ductal adenocarcinoma. Nat. Genet. 47, 1168–1178 (2015). NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15107 Additional information Additional information Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary Information accompanies this paper at http://www.nature.com/ naturecommunications This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ Competing interests: The authors declare no competing financial interests. Author contributions 47. Chang, H. Y. et al. Gene expression signature of fibroblast serum response predicts human cancer progression: similarities between tumors and wounds. PLoS Biol. 2, E7 (2004). C.I. contributed study design, data analysis and manuscript writing. R.P., F.G., C.P. and E.Z. collected samples and generated microarray data. F.G., E.Z. and F.O. generated and curated PDX molecular annotations. F.B., S.E.B., L.M. and A.F. contributed bioinformatics analyses. R.S. interpreted histological analyses. C.B., E.F. and L.T. con- tributed study design and manuscript writing. A.B. and E.M. contributed study design, data analysis, bioinformatics, manuscript writing and project oversight. 48. Arques, O. et al. Tankyrase inhibition blocks Wnt/b-catenin pathway and reverts resistance to PI3K and AKT inhibitors in the treatment of colorectal cancer. Clin. Cancer Res. 22, 644–656 (2015). 15 NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ How to cite this article: Isella, C. et al. Selective analysis of cancer-cell intrinsic transcriptional traits defines novel clinically relevant subtypes of colorectal cancer. Nat. Commun. 8, 15107 doi: 10.1038/ncomms15107 (2017). r The Author(s) 2017 p y yp Nat. Commun. 8, 15107 doi: 10.1038/ncomms15107 (2017). NATURE COMMUNICATIONS | 8:15107 | DOI: 10.1038/ncomms15107 | www.nature.com/naturecommunications 16
https://openalex.org/W4237052708
https://zenodo.org/record/1950694/files/article.pdf
English
null
The Origin of the Major and Minor Modes
˜The œmusical times/Musical times
1,917
public-domain
11,434
The Origin of the Major and Minor Modes Author(s): Kathleen Schlesinger and A. H. Fox Strangways Source: The Musical Times, Vol. 58, No. 896 (Oct. 1, 1917), pp. 452-457 Published by: Musical Times Publications Ltd. Stable URL: http://www.jstor.org/stable/908207 Accessed: 06-11-2015 02:12 UTC The Origin of the Major and Minor Modes Author(s): Kathleen Schlesinger and A. H. Fox Strangways Source: The Musical Times, Vol. 58, No. 896 (Oct. 1, 1917), pp. 452-457 Published by: Musical Times Publications Ltd. Stable URL: http://www.jstor.org/stable/908207 Accessed: 06-11-2015 02:12 UTC Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Musical Times Publications Ltd. is collaborating with JSTOR to digitize, preserve and extend access to The Musical Times. http://www.jstor.org This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions TO THE EDITOR OF 'TTHE MUSICAL TIMES.' After assuming that ' the inborn feeling of the eye for harmonious proportion and symmetry possessed by primitive and untutored man guided him in placing the holes at equal distances along his reed pipe,' she says that 'this may be considered as the firs cause, quite unrelated in the mind to its effect as sound; thus a system of scales came into being quite naturally without preconceived musical notions.' In other words, primitive man, when piercing his flate, had no though of the harmonic series reversed. But later on Miss Schlesinger says: ' Wherever a reed-pipe, or a flute (with certain reservations due to the incidence of the law of diameters) (6), was bored in the way described above, one or other of the pipe-scales in octave or gapped [sic] form inevitably resulted and still results.' That is to say, primitive man adjusted the position of the holes in accordance with the varying proportion which the diameter of the pipe bore to its speaking length. Now, if primitiv (or any other) man placed the holes exactly equidistant (7), the sounds issuing from his pipe would not be in tune with those of the harmonic series. On the other hand, if he adjusted the position of the holes to accord with the law of diameters, what becomes of the above-quoted statemen regarding'preconceivedmusicalnotions'? g Corporal W. G. SWEETMAN, who died after being wounded in France. He was the son of Mr. Sweetman, the well-known Liverpool concert agent. For four years he was a chorister at St. Peter's Church, and was one of the original members of the first Liverpool Cathedral (Lady Chapel) Choir. p CALDER O'BEIRNE, aged sixty-nine. He was a teacher of voice-production and elocution, and in earlier days had toured the country with an opera company in which he took the leading tenor r6les. Referring to an announcement of the death in Berlin of ROBERT VON MENDELSSOHN, senior partner of Mendelssohn & Co., bankers, a correspondent writes to Tke Times as follows : 'In less stormy days the death of Robert von Mendelssohn would give rise to more than a mere record He was, like many of his family from the days of his father' cousin Felix, a frequent visitor to England and had many friends in London musical circles. TO THE EDITOR OF 'TTHE MUSICAL TIMES.' SIR,--(I) The details of the manufacture of facsimiles of ancient pipes, given by 'Miss Schlesinger in her recent paper under the above title, are of course extremely interesting; but their practical bearing upon present musical art appears to be simplynil. d ss t simply In the first place, one may reasonably ask, What is the 'fundamentally new fact in the history of music' which Miss Schlesinger professes to announce? (2) It cannot be the fact that the division of a string into any number of aliqu )t parts will give a portion of the harmonic series reversed because this has been known for many centuries. (3) She cannot intend to imply that the holes with which ancient pipes were pierced invariably divided the latter into a certain number of aliquot parts, because the contrary is notoriousl the case. And if it be asserted (as apparently it is) that the major and minor modes have their origin in the harmoni series, it is not difficult to demonstrate the fallacy of such an idea (4). h p g Second-Lieut. A. G. MILLARD. Killed in action in lrance. He was educated at Eton Choir School, and while there acted as assistant-organist to Dr. C. H. Lloyd, the then Percentor. At an early age he was appointed organist at St. John's, Farnham Common, afterwards holding a similar appointment at the Parish Church of Farnham. Later he became a student at the Royal College, and while there gained the Palmer Scholarship, and studied the organ under Sir Walter Parratt. iHe was a Fellow of the Royal College of Organists and Associate of the Royal College of Music. After finishing his musical training, he was appointed assistant music-master at Marlborough College, remainin there four years. From there Lieut. Millard went to Felstead School as music-master. He was twenty-six years of age, and was the only son of Mr. and Mrs. Millard, of LangdaleHouse,Thornton Heath. t g (4). It cannot escape observation that two of Miss Schlesinger' statements contradict each other (5). Obituar We regret to announce the followings deaths : This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions THE MUSICAL TIMES.-OCTOBER I, 1917. 452 THE ORIGIN OF THE MAJOR AND MINOR MODES. g g Captain WILLIAM HERBERT BAMBRIDGE, the only son of Mr. William Samuel Bambridge, Mus. Bac., of Marlborough, who was killed on August I9, in the twenty eighth year of his age. He was educated at Marlboroug College, where he was prominent as a cricketer and an all-round athlete. After leaving school he went to Canada, where he engaged in farming for about two-and-a-half years Possessing a good baritone voice, he returned to England and went to the Royal Academy of Music. For some time afterwards he was a member of Mr. George Edwardes's company at the Adelphi. At the outbreak of war he enlisted in the University and Public Schools Battalion of the Royal Fusiliers, and afterwards was given a commission. He went to the front about a year ago, was appointed bombing officer to his battalion, was promote lieutenant in March last and captain in the following June The major of his battalion writes : 'His splendid fearlessne may have contributed to his gallant end. His contempt for danger was only equalled by his eagerness to court it, and his cheerfulness under all circumstances never failed him. We are proud that he belonged to the battalion.' TO THE EDITOR OF 'TTHE MUSICAL TIMES.' An amateur 'cellist, so accomplished as to rank with professionals and to take his part on equal terms with the Joachim Quartet, he always gave ready help to English music when he was our guest, and showed a reciprocating kindliness towards English visitors to his famous house in Berlin. He was a devoted friend of Joachim, and intimate with Piatti, of whose famou Stradivarius he became the possessor after Piatti's death. Although he was the leading banker in Berlin, he oddly enough had no trace of German blood; for his father was pure Hebrew, and his mother pure French (of the old noblesse type). He himself was markedly Parisian in appearance and in manner. He was an early friend of Lord Leighton and Tadema, and his house was a centre for painters as well as for musicians. It is only right, even at this juncture when feeling is keen, and everyone across the Rhine under the same dark cloud, to recall a genial, kindly and artistic personality, who was the friend of so many of Germany's best-Clara Schumann, Joachim, and Brahms-- none of whom, happily, lived to see the wreck of their ideals' (see page 472). regarding preconceived notions ? (8) In 189o Mr. Flinders Petrie brought some ancient flutes from Egypt. They were fully described in the Musical Times for October and December of that year. Two of these were found in a rock-cellar tomb at Kahun, in the Fayum, about sixty miles south of Cairo. One was I71 inches in length, and had three holes. The notes heard on blowingacross were :B h s B w p--y--- - - -- -:--- -_ ---Y- But when played with a straw reed of the clarinet type, the notes were : s The other flute was r7 inches in length, and had four holes. The notes on blowing across were: The other flute was r7 inches in length, and had four holes. The notes on blowing across were: The other flute was r7 inches in length, and had four holes. The notes on blowing across were: The other flute was r7 inches in length, and had four holes. TO THE EDITOR OF 'TTHE MUSICAL TIMES.' The whole-tone scale and its harmonies are transition stages on the way back to natura intonation.' ,' g l. To this I would in the first place reply that the longest way round is, I suppose, the shortest way home. The so-called whole-tone scale depends upon equal-temperame for its very existence; given true intonation, the octave cannot be divided into six tones, whether the latter be major or minor. I would humbly advise any composer who wishes to tread the path to natural intonation, to avoid the whole-tone scale as he would the devil. compel (15) Under the heading of 'harmonic potentialities Miss Schlesinger makes an allusion to Eastern music, saying that it can only be harmonized from its own scale-materia But she can hardly be supposed to imply that the music of Asiatic nations, with its third-tones and quarter-tones, is derived, consciously or unconsciously, from the harmoni series,direct or inverted. c . h , d But is music, after all, travelling in a circle? Emil Naumann believed that it was ; and that the centuries which culminated in the genius of Bach and Beethoven would inevitably be followed by an equal period of decadence. And if we are going to travel by way of the whole-tone scale back to the melody of the primitive pipe, I am inclined to agree with him. Meanwhile, some of us may find comfor in the words of Stainer, who, perhaps, was not altogethe ignorant of the subject. In his ' Treatise on Harmony' he says: ' It is, perhaps, hardly necessary to remind the reader that music has made the scales, not the scales music.' There we have the history of the art summed up in a single sentence.-Yours faithfully, series, Leaving her pipes, Miss Schlesinger turns to the monochord, and naively inserts her '6A' among the notes which she derives from it. She fails to perceive that the monochord gives no authority whatever for this innovation If she wishes to form her scale from the sixth to the twelft harmonic, she must be content with seven notes: in the same way, if she were to elect to form it from the eighth harmonic she would be compelled to include nine notes. You cannot playfast and loose with the harmonic series. TO THE EDITOR OF 'TTHE MUSICAL TIMES.' (Stainer: The Music of the Bible, pp. 89, 90o.) However, let us consider 'the harmonic series reversed a little more closely. Miss Schlesinger gives us what she calls an octave of the harmonic series reversed, and endeavours to identify it with the Phrygian Tropos (12). Why she gives it in the wrong order I do not know, especially as the ancient Greeks always thought of their scales as descending. The proper sequence is as follows : y p y y g 6 6L 7 8 9 10 11 12 ( p 0- (See page 300, Fig. 4.) (See page 300, Fig. 4.) r) of a series of undertones which are the converse of the overtones. His views were rejected by Helmholtz and Ellis; and I am not aware that they have been accepted by any scientific authority. The overtones are obtained by merely touching a string at a node, or by overblowing a pipe; there is no known method by means of which the supposedundertones can be rendered audible. ' The starting-point (not 'generator,' pace Miss Schlesinger of this series is of course g2. But what is B flat (6J) doing here? The term o:tave as applied to the harmonic series (13 & 14) has no real significance, for in this series the only octave of sounds occurs between the seventh and fourteenth harmonics: between any other two notes of the same name there lie either more or less than six notes There is no 'gap' between the sixth and seventh harmonics and the B flat which Miss Schlesinger has inserted above is a purely arbitrary addition, affording one proof among many that the harmonic series, whether direct or reversed, is not the origin of the major and minor modes. The above scale can be tuned to agree with the ancient Phrygian mode ; but in that case, even with the omission of the intruding B flat it will not correspond with the supposed portion of the harmonic series. All attempts to derive any of our scales from the harmonic series compel us to 'doctor' the latter. ' supposed In conclusion, Miss Schlesinger asks, ' What is the exact value to us in the 20th century of all these discoverie concerning the music of antiquity and the established realities of natural intonation? Simply this: it is the dawn of the music of the future. . .. TO THE EDITOR OF 'TTHE MUSICAL TIMES.' The notes on blowing across were: - _ -- _-_ - z 0 -- F--- - This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions THE MUSICAL TIMES.-OCTOBER I, 1917. 453 With the reed, these became: produce certain notes in tune. We must learn from this, not to place much confidence in conclusions drawn from actual experiments on old pipes. Suppose, for instance, it were attempted to discover the series of scale-sounds of such an instrument by placing it in the hands of a modern performer, it would be impossible to say whether any noticeable variations from known forms of the scale ought to be attributed to the intentional design of the instrumen itself (i i) or to our loss of those traditions which influence itsuse ' (Stainer: 'The MusicoftheBibl ' 89 90 ) produce certain notes in tune. We must learn from this, not to place much confidence in conclusions drawn from actual experiments on old pipes. Suppose, for instance, it were attempted to discover the series of scale-sounds of such an instrument by placing it in the hands of a modern performer, it would be impossible to say whether any noticeable variations from known forms of the scale ought to be attributed to the intentional design of the instrumen itself (i i) or to our loss of those traditions which influence itsuse ' (St i 'The MusicoftheBibl ' 89 90 ) ---~ ---p--~--- - __-__ --K -- It should be added that Mr. D. J. Blaikley assisted in the examination of these pipes, pipes, (9) There would appear to be nothing in these scales to suggest the harmonic series reversed. But 'with regard to the open pipes, or flutes, of the ancients, it should be borne in mind that it must have been most difficult to produce a series of sounds, either similar in timbre or perfectly true in pitch, without the aid of keys (io). Up to the last century, certain holes in the then existing flute had to be only partially covered by the fingers in order to s its use. TO THE EDITOR OF 'TTHE MUSICAL TIMES.' play It is curious that after asserting that 'a system of scales came into being quite naturally, without preconceive musical notions or arbitrary interference, by purely mechanical means, and as a consequence of embodyin natural laws' (I6), Miss Schlesinger, towards the con- clusion of her paper, mentions ' three distinct stages in the development of this natural system of music,' and adds: 'The second stage is one of transition, during which the desire to make the pipe-scales approximate to tne sequences of the harmonic series becomes evident.' So that, alter all, preconceived musical notions did lead to an alteration in the position of the holes. But these notions were not the result of obedience to natural laws. ARTHUR T. FROGGATT. g [Numerals are inserted in the above letter and Miss Schlesinger's reply to facilitate reference.-ED., .3 fT.] y g [Numerals are inserted in the above letter and Miss Schlesinger's reply to facilitate reference.-ED., .3 fT.] This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions g p y TO THE EDITOR OF 'THE MUSICAL TIMES.' t c SIR,-Innovations and revolutionary ideas inevitably meet with a certain amount of blind opposition and wilfu misconception. When one has devoted one's life to origina research in the archaological or any other field, one accepts such attacks as the above philosophically, knowing that the basic principles and natural laws which underlie all evolution are unassailable, and can take care of themselves when they have been discovered. (17) The construction of a musical instrument with a scale consisting of a portion of the harmonic series reversed is not in accordance with any known natural law. The harmoni scale (the series of upper partials, the scale of nature), and a similar succession of sounds inverted, stand upon a totally different footing. Dr. Hugo Riemann, indeed, is so obsessed by the theory of the polar opposition of the major and minor modes, that he believes himself to be tile discovere y r The only thing, therefore, that is open to question is whether I have expressed these with sufficient clarity. Let us examine the points at issue, which have been numbered in order to avoid useless repetitions. (Miss Schlesinger's reply to the above.) g p y TO THE EDITOR OF 'THE MUSICAL TIMES.' This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions Precise instruction are added as to the use of the first hole a; a vent, and to the partial covering of holes to produce the ?irutis; reference is likewise made to an extensive system of cross-fingering. (See also (Io) above. ) Experiments The Phrygian Tropos is the ANCIENT Greek Phrygian mode, and the later Greek Phrygian to which Dr. Froggatt refers was derived from it. (14) My pianoforte has thirteen notes within the compass of an octave; not eight! The Greeks knew all about octave scales, with a varying number of notes between any two sounds in octave relation, and they called their octave scale diapason= 'through all' and not di'octo (Aristotle Problem XXXII., Sect. 19). The notes ruled out by the tetrachordal nature of the extended Tropoi scales, were used as alternative notes for the Chroai or Shades. s ns s (15) Most certainly I do assert 'that the music of Asiatic nations with its so-called third-tones and quarter-tones' is derivzed froo the za;r;zonic se;ies bothz direct and inverted, and I can prove this. There is, however, such a thing as evolution in music and musical systems, and it is putting the cart before the horse to talk of 'doctoring' the harmonic series in order to explain that which has evolved from it. If Dr. Froggatt wants evidence of the evolution of natural intonation before the advent of the tempered system, he has only to study the I6th and I7th century writers: Virdung, Agricola, Pretorius, and Mersenne. He will find abundant evidence of the prevalence of natural intonation and of its gradual modification under the influence of the keyboard instruments, which culminated in the extinction of natural intonation at the dawn of equal temperament. (7) In the reed-blown pipes used by the ancients this practice of using the first hole as a vent is also sometime observed, but it is not due to difficulties connected with the law of diameters, whose influence on cylindrical pipes played by means of a reed mouthpiece is negligible. In these favourite pipes of small calibre there is, moreover, seldom any appreciable difference in the diameters of pipe and holes, whereas in the flutes discrepancies in the diameters of both finger-holes and embouchure as compared with that of the pipe are more pronounced and have to be compensated. equal temperament. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions p p p p (8) The Maket pipes are, of course, included in my investigations. They belong to Stage II. (August, page 354). This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions It was certainly not suggested in my article that the 13th was used by the earliest pipe- makers in the 12 Tropos (Phrygian). (.See fig. 4, where the I3th was first introduced in the division of the string under 'Experiments with a Monochord.') n h s s July, page 298b) string (i3) The offending B flat inserted as 13th or 61 is quite rightly placed where it is, and does not by any means constitute an arbitrary addition, since every harmonic is potentially present in every octave, and the expedient of halving the distance in order to obtain a note belonging to the next octave of the Tropos was as frequently resorted to by the ancients as was the device of half-covering the holes, which produced the same result. The Phrygian or any other Tropos can be doubled (or even quadrupled) in order to obtain an extended range, as already explained (August page 352 ; see also page 354, under fig. I I). In fact, that was the method used in order to obtain the chromatic and enharmonic genera in the Greater Complete System. The rules for the tetrachordal structure of the scales amongst the ancient Greeks, and for the loci of the movable notes, preclude any imputation of arbitrariness in handling the harmonic series in this way. It was certainly not suggested in my article that the 13th was used by the earliest pipe- makers in the 12 Tropos (Phrygian). (.See fig. 4, where the I3th was first introduced in the division of the string under 'Experimentswith a Monochord.') g (5) & (6) It seems hardly necessary to point out that any explanation suggested concerning first stages in evolution does not necessarily apply also to all subsequent stages! Because, therefore, man in his first attempts to bore holes in pipes fell in with a natural law which gave him a system of scales without his possessing any previous knowledge of their existence, it does not follow that when he has learnt to manipulate the pipe it will not, in his hands, pass through other well-defined transitions owing to his efforts to make the pipe render some preconceived sequence derived, for instance, from the Harmonic Series. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions Modern musicians have too readily assumed that the notation familiar to them has always represented sounds consisting of aggregates of one unit alone, i.e , the semitone. p p p p (8) The Maket pipes are, of course, included in my investigations. They belong to Stage II. (August, page 354). This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions Constant readers of the Mi3usical Times will be familiar with my conception of the harmonic basis of all music (lMusical Times, February 1914, page 95 ; June, 1914, page 375 ; June, 1915, page 338), and will doubtless understand how this can be. This conception concerning the hearing of harmonics has found support all along its line of development, and receive something very near proof from stages numbered 2, 3 and 4, on page354,Augustnumber. page 354, August The law of diameters proves a stumbling block on 1fute only. The ancients, however, were equal to the occasion, and deliberately used the first hole of flutes as a vent measuring the length from the centre of the vent to the centre of the embouchure, which disposes of the main difficulty. Incidentally it proves also that by that time various Tropos sequences obtained on wind instruments had become recognized, and that the flute-maker did not rest until he had succeeded in reproducing them correctly (see August, page 355 b). Bharata, the writer of the 'Nityasistra, a Sanskrit treatise on the Drama, ascribed to the 5th or 7th century A D., and Sarfigadeva, the 13th century author of the 'ratnakara,' both give lists of flutes of various length with holes bored at equal distances. Precise instruction are added as to the use of the first hole a; a vent, and to the partial covering of holes to produce the ?irutis; reference is likewise made to an extensive system of cross-fingering. (See also (Io) above. ) p g ,August The law of diameters proves a stumbling block on 1fute only. The ancients, however, were equal to the occasion, and deliberately used the first hole of flutes as a vent, measuring the length from the centre of the vent to the centre of the embouchure, which disposes of the main difficulty. Incidentally it proves also that by that time various Tropos sequences obtained on wind instruments had become recognized, and that the flute-maker did not rest until he had succeeded in reproducing them correctly (see August, page 355 b). Bharata, the writer of the 'Nityasistra,' a Sanskrit treatise on the Drama, ascribed to the 5th or 7th century A D., and Sarfigadeva, the 13th century author of the 'ratnakara,' both give lists of flutes of various length with holes bored at equal distances. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions THE MUSICAL TIMES.-OCTOBER I, 1917. 454 , . nt (9) If Dr. Froggatt had given consideration to the ratios of vibration given by Mr. D. J. Blaikley in the Musical Times (December, 1890, page 715), he might possibly have read the riddle. The four-holed pipe gives the 15-Tropos and the three-holed the 14-Tropos, both slightly modified. The notation of the sequences quoted above under (8) does not convey the true intonation of the sounds obtained from the pipe by Mr. Blaikley ; for instance, the interval E? to G suggests a major third, whereas the ratio 6 : 5 is nearer the mark, 6 192 v.p.s. (194 D.J.B.) 1 5 200 v.p.s. 5 4 n s , . nt h (9) If Dr. Froggatt had given consideration to the ratios of vibration given by Mr. D. J. Blaikley in the Musical Times (December, 1890, page 715), he might possibly have read the riddle. The four-holed pipe gives the 15-Tropos and the three-holed the 14-Tropos, both slightly modified. The notation of the sequences quoted above under (8) does not convey the true intonation of the sounds obtained from the pipe by Mr. Blaikley ; for instance, the interval E? to G suggests a major third, whereas the ratio 6 : 5 is nearer the mark, 6 192 v.p.s. (194 D.J.B.) 1 5 200 v.p.s. 5 4 ( ) I am not concerned with matters of personalopinion (x) I am not concerned with matters of personal opinion -as for instance in paragraphI-but onlywithpoint (x) I am not concerned with matters of personal opinion -as, for instance, in paragraph I-but only with points that seem to requirefurther elucidation. (x) I am not concerned with matters of personal opinion -as, for instance, in paragraph I-but only with points that seem to requirefurther elucidation. (x) p p -as, for instance, in paragraph I-but only with point that seem to requirefurther elucidation. require (2) Certainly not ! That would be a new fact in the science of acoustics, but not in the ' History of Music.' The fundamentally new fact is that the boring of pipes at equal distances, which are also aliquot parts of the total length produces a definite system of scales (some of which are gapped or transilient) based upon the harmonic series reversed Diversity of mode is the result, primarily of the length of pipe conditioned by convenience in fingering. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions The importan fact is that we have in the reed pipe an instrument which preserves intact the exact intonation of the music it was capable of rendering thousands of years ago, providing the underlying law is understood and that the pipe is played with the right kind of mouthpiece. I have identified the simple sequences of this system with the seven ancient octave Tropoi given by Plato and other classical writers, and the Mese in each case falls correctly upon an octave of the keynote or generator. The fuller sequences of the system (which were developed by Pythagoras on the strings correspond exactly with the fifteen ancient Tropoi of the Greater Complete System given in the Fragment of Alypius I have explained the process of doubling the number of the Tropos (August, page 352b and 354a, section 6). Of require (2) Certainly not ! That would be a new fact in the science of acoustics, but not in the ' History of Music.' The fundamentally new fact is that the boring of pipes at equal distances, which are also aliquot parts of the total length produces a definite system of scales (some of which are gapped or transilient) based upon the harmonic series reversed Diversity of mode is the result, primarily of the length of pipe conditioned by convenience in fingering. The importan fact is that we have in the reed pipe an instrument which preserves intact the exact intonation of the music it was capable of rendering thousands of years ago, providing the underlying law is understood and that the pipe is played with the right kind of mouthpiece. I have identified the simple sequences of this system with the seven ancient octave Tropoi given by Plato and other classical writers, and the Mese in each case falls correctly upon an octave of the keynote or generator. The fuller sequences of the system (which were developed by Pythagoras on the strings correspond exactly with the fifteen ancient Tropoi of the Greater Complete System given in the Fragment of Alypius I have explained the process of doubling the number of the Tropos (August, page 352b and 354a, section 6). m ) . p (I I) The Tropos with its precise boring disposes of these ambiguities, and provides a criterion quite independent of the performer's renderingof the scale. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions r d ble p g (12) Some few modern writers, on the strength of the Greek vocal notation, have concluded that the Greeks were in the habit of using their scales in descending motion, but this is disproved entirely by the order of the notes named in the tables of Alypius (Meibomius), and of the Greek theorists who invariably begin with Proslambanomenos and rise up to Nete. (See also Gaudentius, Nicomachus, page 17, Aristide Quintilianus, Philolaos (edit. Boeckh, page 66), Aristoxenus and others.) What Dr. Froggatt gives under (12) above as the proper sequence is the harmonic series reversed, which is not quite the same thing as the Tropos or mode founded upon it. Let him examine figs. I, 3 and 4, July, page 297. He will see that we are quite agreed on this point; but he does not seem to have grasped the significance of the Tropos, or the reason why it begins with the note of the whole pipe (see July,page298b) or with the stringnote. p ( g ,p g , (3) Of course not-see August, page 354-where four different stages of transition in the boring of pipes are described. (4) I should welcome such a demonstration. It would be interesting to compare it with the incontrovertib evidence given in the article. n h s s July, page 298b) or with the string note. (i3) The offending B flat inserted as 13th or 61 is quite rightly placed where it is, and does not by any means constitute an arbitrary addition, since every harmonic is potentially present in every octave, and the expedient of halving the distance in order to obtain a note belonging to the next octave of the Tropos was as frequently resorted to by the ancients as was the device of half-covering the holes, which produced the same result. The Phrygian or any other Tropos can be doubled (or even quadrupled) in order to obtain an extended range, as already explained (August page 352 ; see also page 354, under fig. I I). In fact, that was the method used in order to obtain the chromatic and enharmonic genera in the Greater Complete System. The rules for the tetrachordal structure of the scales amongst the ancient Greeks, and for the loci of the movable notes, preclude any imputation of arbitrariness in handling the harmonic series in this way. TO THE EDITOR OF 'THE MUSICAL TIMES.' The whole history of the scale is ignored,-the Greek, Ecclesias- tical, Byzantine influence, and the successive modification introduced by harmony-and the pipe-scale is represented as being the sufficient explanation of its present form. Wherever the facts do not fit the theory, the intonation is described vaguelyas havingbeen altered ' later on' : d s The place where the theory appears to go wrong is in extending the pipe-intervals to the whole octave, instead of limiting them to a few notes, which all the evidence shows were originally four. Provision is made for this on page 354 under the heading 'A fortuitous discovery,' where it is asserted that the piper knew how to get the overblown Fifth, and so to produce tetrachords. These tetrachord must, however, have been similar: unless it is assumed, though not stated, that there were not three holes but four which would have admitted of alternative stopping. Yet all our modern scales as here explained postulate dissimilar tetrachords, so that the upper half of them could not have been got by overblowing the lower, but must have been produced on a pipe with seven holes. The division of the scale into tetrachords for the purpose of analysis is therefor illusory. g 'Un cheval vient d'equus sans doute; Mais il faut avouer Que, pour venir ici, Il a bien dfichanger en route.' g 'Un cheval vient d'equussans doute; The attempt is one more of many that have been made to show that the intervals which form the vocabulary of musical language are not selections of art, but dictations of science It proposes to show through the pipe-scale that our scales are based on the harmonics of nature, either straightforward or upside down. But as Mr. Clutsam remarks in this month' issue, where he is discussing a book which takes a much saner view, 'the attempt to take nature as a guide has always failed, and always.will fail,' and 'nobody in his senses would conceive of music as an exact science.' Our scales are, however, not based on the pipe-scale because neither the major nor two out of the three minor show any trace of the principle that successive ascending intervals increase in size, while in the third of them the characteristic augmented interval is derived from the pipe- scale in quite another way than that described in the article. TO THE EDITOR OF 'THE MUSICAL TIMES.' SIR,-The article in your July and August numbers on the 'Origin of the Major and Minor Modes' rightly draws attention to the existence of reed-pipe scales pierced at equal distances. The ruling principle of them all, which the article does not however mention, is that the intervals between successive notes increase in size as the scale ascends ; and that is the principle on which the whole discussion of the genera by Aristoxenus turns. He is evidently concerned mainly with the tabulation of a large number of scales, known to Greek ears and coming to them from very differen sources, several of which betray their pipe origin in this peculiarity of the rising interval. The actual measuremen therefore of a small or large number of ancient or modern pipes will not establish the fact of the existence in ancient times of 'equidistant' scales any further than it is already established by Greek theory. s equidistance and only -it could not be more than accidental s d nt nt n ns r d The ascending minor is too hard a nut for the reverse series to crack, and the natural series has to be called in. The higher numbers of this series, which are alone employed, require long practice or else mechanical aid to hear. Many of us were unable to follow a certain modern work on harmony which derived the minor third from the nineteenth harmonic; not because the interval is then out of tune-for it is only what we hear on the pianoforte every day-but because such an altitude was beyond our power of estimate; and the author abandoned the view in his second edition. But the explanation of the ascending minor given in this article assumes that we can hear not only the 19th but the 29th, and then there will be three notes to be ' raised later on.' With a little more practice we might have heard the 32nd, and then there would be only two faulty notes; and if we persevered till we could hear the 240th there would be none-but then, perhaps, there would not be much time left to make music in. y theory. But the superstructure which the article raises upon this fact seems to me to pass the bounds of common sense. faithfully, KATHLEEN SCHLESINGER. Fellow Institute of Arckreoloy, University of Liverpool. Fellow Institute of Arckreoloy, University of Li Fellow Institute of Arckreoloy, University of Liverpool. [The following correction should be made in Miss Schlesinger's article on page 353 (August), first paragraph column 2, line 4 : for ' first' read 'just.'] , s y g get There is no one who would be more surprised than the reed-piper, primitive or contemporary, if he were told that the scale he had made by measuring off inches on his pipe, and had played in all his life, satisfied his ear only because it was derived from a series he had never heard of and identical with a series he could not hear. He would feel like M. Jourdain : 'Je n'ai point dtudie, et j'ai fait cela du premier coup. Je vous remercie de tout mon cceur.' So that if there were any musical connection between harmonic and equidistance-and there is only a numerical connection -it could not be more than accidental. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions THE MUSICAL TIMES.-OCTOBER I, 1917. THE MUSICAL TIMES.-OCTOBER I, 1917. 455 (16) & (17) I have shown in my reports addressed to the Institute of Archaeology of the University of Liverpool that it is a fallacy to regard the harmonic series (ascending as the natural scale, and the Tropos as an artificial or arbitrary succession of sounds. In fact the harmonic series or so-called natural scale cannot be conceived, explained, or played upon any instrument without first presuming the Tropoi. This will be fully demonstrated before long, in my book now in preparation. Finally, without wishing to be discourteous, it seems necessary to remind Dr. Froggatt that he is using scissors and paste, so to speak, in order to endeavour to disprove the results of original research ; that the investigations upon which the article is based are not merely theoretical, they have been carried into practice, and instruments have been made to test the results obtained -Yours faithfully, But even if our scales were in fact based on the pipe-scale that would not show any connection between them and the harmonic series, taken either forward or backward. Let us consider this reversed series first. It is inaudible. There is no question here of 'resultant tones,' which do actually sound, and upon which v. Oettingen, taking his cue from Rameau, based his theory of the minor chord; for a 'resultant' is caused by the simultaneous sounding of two higher tones, and neither two nor one are sounded here. The downward series is a matter for the eye, not the ear ; the ear can neither demand this series of sounds nor be satisfied by it, for there is in it no sound which contains and produces all the others, as there is in the upward series. It is true that the fractions which represent these sounds are the reciprocals of the fractions which measure the pipe ; but the connection between the harmonic and the equidistan series resides wholly in number, as it does in any other measurement of heterogeneous things; as when, for instance we buy I-lb. of sugar and take J-hour to getit. KATHLEEN SCHLESINGER. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions TO THE EDITOR OF 'THE MUSICAL TIMES.' Moreover, the reed-piper who is supposed to have been responsible for our scales is a more sophisticated musician than the piper who inspired the Phrygian scale on page 298, since he is able to select his notes; in other words, equal distance is with them abandoned for unequal, and with that goes the whole assumption on which the theory stands. It is not true, therefore, to say, as on page 352, that our four scales are 'based upon the law of the aliquot division of strings,' for an unexplained principle of selection has modified that law out of all recognition, and even the notes so obtained are subject to alteration 'later on.' s g illusory. On the same page we find that 'it is not suggested that the ancients adopted the pipe-scale'-i.e., the equidistan scale-' by choice, but of necessity' ; for the alternative to equidistance-to place the holes so as to give this or that segment of the harmonic series-presents, as the article quite truly says, practical difficulties. But of the millions of pipes which have been made since the days of Jubal or Trhamyris there can hardly have been a large number that were pierced at exactly equal distances so as to give the reversed harmonics exactly. For, in the first place, a very slight error of measurement makes a great difference in the pitch of the note. In the example on page 298, for instance, in order to make the A, which is a whole quarter of a tone sharp, into a true A , the first hole would have to be bored lower by f of an inch only-say, the thickness of the finest sewing needle. And in the second place, there is a presump- tion that such pipes were made rather at haphazard, as This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions THE MUSICAL TIMES.-OCTOBER 1, 1917. 456 n apparently the sole property of the Pythagoreans Nicomachus attempts a very slight exposition of the Tropos principle. (Meibomius, lib. I, page 19 ) n apparently the sole property of the Pythagoreans Nicomachus attempts a very slight exposition of the Tropos principle. (Meibomius, lib. I, page 19 ) n apparently the sole property of the Pythagoreans Nicomachus attempts a very slight exposition of the Tropos principle. (Meibomius, lib. I, page 19 ) the following little experience indicates. Some tribesme were to sing and dance for me, but as they had to march in over a long distance they arrived a day late, so that there was only three-quarters of an hour before my train went They sang and danced, and would have played the bamboo flute, but 'there was not time to make one.' The implica tion-that a flute was a thing you just made out of the nearest tree, when you wanted a little music-does not suggest any very careful measurement. In fact, there has in all ages been, we may imagine, a good deal more of choice than of necessityin the piercingof these holes. the following little experience indicates. Some tribesme were to sing and dance for me, but as they had to march in over a long distance they arrived a day late, so that there was only three-quarters of an hour before my train went They sang and danced, and would have played the bamboo flute, but 'there was not time to make one.' The implica tion-that a flute was a thing you just made out of the nearest tree, when you wanted a little music-does not suggest any very careful measurement. In fact, there has in all ages been, we may imagine, a good deal more of choice than of necessityinthe piercingof these holes. - d p p p g It was, however, precisely the measuring and making of facsimiles of a number of ancient pipes which did establish the idenity, of the Greek Tropoi as pipe-scales. They had until then remained unsuspected by countless writers on Greek music for two or more centuries. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions It is not possible in a short article to give numerous proofs of all statements advanced, but it is evidence of a positive and constructive kind consistent with archzeological methods in original research, and very different from the negative and destructive statements, without basis or principle to support them, with which Mr. Fox Strangways endeavours to refute my work. (See also 15above.) A. 1H. Fox STRANGWAYS. (Miss Schlesinger's reply to Mr. Fox Strangways's letter (Miss Schlesinger's reply to Mr. Fox Strangways's letter [The numerals refer to the paragraphs in the letter.- ED., M. T.] (Miss Schlesinger's reply to Mr. Fox Strangways's letter. [The numerals refer to the paragraphs in the letter.- ED., M. T.] [The numerals refer to the paragraphs in the letter.- ED., M. T.] [The numerals refer to the paragraphs in the letter.- ED., M. T.] As since the proof of the above letter was submitted to me there is a very limited time in which to reply to Mr. Fox Strangways's strictures, I shall be unable, to my regret, to touch upon more than one or two points, although all would have been interesting. interesting. (I) The principle of the arithmetical progression downward proper to the reversed Harmonic Series was, I considered sufficiently obvious throughoutthe article. sufficiently throughout The whole of this first paragraph seems likely to give rise to some misconception which must at the outset be removed The not too felicitous or clear wording of the paragraph would give the impression that my discovery and identificatio of the pipe-scales with the scales and theories of ancient Greece was no discovery after all. It seems to imply that the explanation was there all the time, patent and obvious, and that, above all, Mr. Fox Strangways himself was well aware of it. Now I know Mr. Fox Strangways too well not to feel certain that this is the last thing he would consciously do or tolerate, and therefore in justice to both, some explanationof this pointis due. explanation point Aristoaxenus knew nothing about the principle underlyin the ptee-scales, and was frankly puzzled by them, as he acknowledges many times in his famous treatise. For instance (page 92, Macran's Translation, 37) : 'The fifth part of our science deals with the keys [r6vovc=the tonal use of a Tropos as a mere key-K.S.] in which the scales are placed for the purposes of melody. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions Readers of the Ailisical Times would scarcely imagine from the statemen to which reference is here made that I had the honour and pleasure of introducing my discovery of these so-called 'equidistant ' scales and their identity with the Greek modes to Mr. Fox Strangways's notice exactly two years ago, and of receiving some very helpful criticism from him, too ! (See also 2 and 13 in the reply to Dr. Froggatt.) necessity piercing The truth surely is that the scale at any given moment is, like all institutions of man and most operations of Nature, never the single effect of one cause : in fact, it might say of itself, as a certain famous little girl said, ' I specs I growed. In its infancy what seems to happen is this : The untutore voice and the unsophisticated pipe start at different points: the former 'thinks' intervals (increasingly consonant) round but chiefly downwards from, a high tonic, the latter builds up empirical intervals (tending to increase in size as they ascend) from a low tonic. The subsequent career of the scale is a reconciliation of these impulses, modified through the centuries by a series of purely musical consideration So that the primitive reed-piper can be called the father of our modern scales only in the sense in which Tubal Cain was the 'instructor' of those who planned and built the Forth Bridge. In that sense, certainly, we may agree with the article.-Yours, &c., d h s. s h on y ?r e 13 reply Froggatt.) (i & 9) Before going further, I would point out that I did not undertake in this article to trace the ez'olution of our major and minor modes, but only to make known their origin-the natural scales upon which they are based. I have pointed out (page 354 and in Figs. 11 and 12) that our tempered major scale, consisting of two tetrachords exactly equal as to ratios (and differing only to the extent of a comma in just intonation), is derived from the 20 Tropos- Hypolydian-and that the idea of the two equal tetrachord arose through the manipulation of the reed mouthpiece. I have nowhere suggested that our minor scales were also due to the same device as stated by Mr. Fox Strangways in (8). This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions Moreover, the device is not overblowzi,z, it is a shift of the lips upon the vibrating tongue of the mouthpiece, shortenin its effective length ; overblowing is possible also, but both the miodus operandi and the resulting timbre are quite differen (2) Since, therefore, the major scale has existed through the ages without modification, when and wherever the boring of a three-holed pipe with holes at distances of - of the total length was found, it seems to me that the method t loong dltonr of deriving chevalfrom eqztus-to which Mr. Fox Strangways playfully refers in suggesting that I have passed over the history of the evolution of the scale-is hardly to be commended, since the derivation from Ki/3aXXij (Caballes), vid the Latin Cabal!zts, which I have chosen, has the double merit of being correct and direct. And so with the scale. I have, in fact, spoken only of the steps which can be proved, i.e., those preserved by the borings of pipes (see page 355a). How can Mr. Fox Strangways or any other writer trace or prove any shades of intonation through the early Church music, which was based upon the organ? The spacing of the keys on the keyboard unfortunately tells no tales and gives no clues. If he feels so inclined, let Mr. Fox Strangways turn rather to Agricola (ed. 1545), where he will find the Tropos principle rampant; it is used in order to determine the position of the frets on the necks of lutes and viols; for marking out the ratios on the monochord which was still used in the churches to train singers for a cappella singing, in which the voices were not forced into line with the organ. Above all, one finds the living Tropos in the schemes for the fingering of pipes upon the great staff ; the holes opened to produce the notes of the natural scale of the pipe are given, in most cases, as well as the cross-fingeri in use for such modifications of the natural scales as were usual. The II, 12, 15, and 20 Tropos predominate. One of the flute schemes (page 102, folio 3I) shows plainly that the law of diameters had not even been empirically solved on the flute, as the first hole is used as vent, although the note of the whole pipe-all holes closed-is given as well. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions be9, o, II, 12, 13, 15, 16, which means that the services of the 9th, zoth, IIth, 12th, I3th,isth and I6th BY N. DE V. HART. BY N. DE V. HART. y r My subject requires no apology. It is quite true that if every Englishman knew French, German, Italian, Russian, Spanish, and the Scandinavian languages, there would be no necessity for the writing of this article ; but as few Englishmen know, at the most, more than one other language beside their own, it seems clear that it is, and for a considerabl time will be, needful to have songs and operas translate fromforeignlanguagesinto English. egg. (9) The point urged by Mr. Fox Strangways is rather damaged by the fact that the :th part of an inch which he claims to represent all the difference between the position of a hole bored at 1.1 from the extremity of the pipe, or at {? (in order to substitute for a certain flattened A, a correctly-tuned A7), applies only to a pipe one inch in length ! This one-inch pipe, moreover, would give sounds which the human ear, as at present constituted, would fail fail to apprehend. This is a clear case of Nemesis pursuing Mr. Fox Strangways for pouring contempt upon the Tropos principle and allowing it only a numerical existence and an accidental birth. On a real, not a Lilliputian pipe, measuring '353 metre: g foreign languages g Assuming, then, that we must have translations, the question arises: Are our present methods of translatin satisfactory ? Ask the singers : they will tell you that for the most part translations are unsingable. Ask the poets: they will tell you that they cannot go to a concert or an opera without feeling that their muse has been insulted. Ask musicians : they will tell you that translators have no respect for musical phrasing, accent, or rhythm. Ask audiences and critics : they will tell you that most translations sound either bald or incongruous, and almost invariably render the task of the singer so difficult that his discomfort is inevitabl communicated to the listener. g "153 X II '353 = 3235 position of hole I from mouthpiece 12 g "153 X II '353 = 3235 position of hole I from mouthpiece 12 *353 x 15 16 = '331 ,, ,, ,, , g "153 X II '353 = 3235 position of hole I from mouthpiece 12 The reason for this state of things is not far to seek. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions No exaplanation has yet "eei qfered oJ the manner in which those keys [rpd7ro7] are to Ae fbtnd, or of the principle by which one must be guided in enunciating their number.' On page 193 he mentions the Tropoi as the scales obtained by boring holes in the auloi or clarinets (see also pp. 196 and 197,&c ) ( pp. 196 197, &c.). The discussion on the genera by Aristoxenus can hardly be said to turn on the principle of successive increase in the size of intervals, since Aristoxenus does not recognis that progression as such, nor yet as a principle or law, and only casually describes two scales (not several) characterised by this successive increase, which he certainly does not anywhere connect with an instrumen In fact he denies emphatically, and with unnecessar violence, the possibility ' that the essence or order of harmony [i e., natural law-K. S.] depend upon the properties of instruments, and more especially of the aulos or clarinet [page 196] 'there is no error so fatal and preposterous,' 'it is sheer folly,' &c. What little was understood by the Greek theoristsconcerningtheTroposand the underlyinglaw was t. y y t y ( 15 ) (5) & (6) No, Mr. Fox Strangways ! The connection is not on a numerical but a proportional basis. and it is most certainly not accidenLtal. Where would the Harmonic Series be without the Tropos principle may I ask ? What happens when a segment of this series is to be used on a pipe; for instance, from 8 to I6 ? The ratios of length will t. y y y ( 15 ) (5) & (6) No, Mr. Fox Strangways ! The connection is not on a numerical but a proportional basis. and it is most certainly not accidenLtal. Where would the Harmonic Series be without the Tropos principle may I ask ? What happens when a segment of this series is to be used on a pipe; for instance, from 8 to I6 ? The ratios of length will be9, o, II, 12, 13, 15, 16, which means that the services of the 9th, zoth, IIth, 12th, I3th, isth and I6th This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions THE MUSICAL TIMES.-OCTOBER I, I9i7 457 TO THE EDITOR OF ' THE MUSICAL TIMES.' SIR,-Would Mr. H. Davey be kind enough to state to what 'anti-Handel outbreak' of mine he refers? I canno call one to mind.-Yours, &c., EDWARDC BAIRSTOW SIR,-Would Mr. H. Davey be kind enough to state to what 'anti-Handel outbreak' of mine he refers? I cannot call one to mind.-Yours, &c., EDWARDC BAIRSTOW SIR,-Would Mr. H. Davey be kind enough to state to what 'anti-Handel outbreak' of mine he refers? I cannot call one to mind.-Yours, &c., C BAIRSTOW provided. Rule 9.-The distribution of syllables must be such as to permitof flowing, easy pronunciation. provided. Rule 9.-The distribution of syllables must be such as to permitof flowing, easy pronunciation. permit g, yp Rule Io.-The construction of the sentences must be simple, and free from complicated parentheses and inversions. permit g, yp Rule Io.-The construction of the sentences must be simple, and free from complicated parentheses and inversions. I, Minster Court, York, EDWARD C. BAIRSTOW September 12, 1917. I, Minster Court, York, September 12, 1917. Finally, there are the literary requirements Finally,there are the literary requirements Finally, y q Rule xI.-The translation must reproduce the metre of the originalpoem. y, y Rule xI.-The translation must reproduce the metre of the original poem. TO THE EDITOR OF 'THE MUSICAL TIMES.' h hb h unity p Rule 3.-The important note or notes which bear the important word or words in the original, must bear also the important word or words in the translation The climax of the words and the climax of the music must synchronise y p Rule 3.-The important note or notes which bear the important word or words in the original, must bear also the important word or words in the translation The climax of the words and the climax of the music must synchronise synchronise. Rule 4.-Accented syllables must fall on accented notes R l th i synchronise. Rule 4.-Accented syllables must fall on accented notes R l th i y Rule 5.-Where in the original one syllable is spread over two or more notes, the same thing must happen in the translation. y Rule 5.-Where in the original one syllable is spread over two or more notes, the same thing must happen in the translation. f h ndon, W. r ; fai (Signed)A. H. MANN 19, Street, London, W. r ; faithfull (Signed)A. H. MANN, on, ; fait (Signed)A. H. MANN, gned) , Hon. Treasurer of gned) , Hon. Treasurer of King s Field, of Cambridge. The Incorporated Society of i usicians. King s Field, of Cambridge. The Incorporated Societyof i usicians Thirdly, there are the requirements of the singer : Rule 6.-The vowels of the original poem must be reproducedby similar vowels in the translation. h b e. The Incorporated Societyof i usicians Thirdly, there are the requirements of the singer : Rule 6.-The vowels of the original poem must be reproduced by similar vowels in the translation. f h tb The Incorporated Society of i usicians p KATHLEEN SCHLESINGER. p KATHLEEN SCHLESINGER. p KATHLEEN SCHLESINGER. TO THE EDITOR OF 'THE MUSICAL TIMES.' TO THE EDITOR OF 'THE MUSICAL TIMES.' proceed. An ideal and prfect translation would be one that obeyed in complete detail the following rules, which I would classify thus : proceed. An ideal and prfect translation would be one that obeyed in complete detail the following rules, which I would classify thus : SIR,-May I beg for a little of your space to ask my brother musicians for some help. As a result of some sonata recitals that Mr. Alexander Cohen, of HIarrogate, has been recently giving-(all honour to him, I say)-he has sent me a cheque for /23 and hopes to send another donation after the end of the approaching season. This money is to be devoted entirely to helping the children, or distressed widow, of some musician who has been killed in the War. The I.S. M. has g'adly undertaken to distribute Mr. Cohen's splendid gift in accordance with his expressed wishes, and would indeed rejoice if any other musician-or body of musicians-would endeavour to do as he has done. We should be most grateful if anybody would kindly send information of any cases which would accord with Mr. Cohen's desires, to Mr. Hugo Chadfield, the general secretary of the I.S. M., 19, Berners Street, London, W.-r ; or to Yours faithfully, A H y Firstly, there is the poetic, or spiritual, requirement : Rule I.-The translation must reproduce, in suitable language, the spirit and meaning of the original poem. h l y Firstly, there is the poetic, or spiritual, requirement : Rule I.-The translation must reproduce, in suitable language, the spirit and meaning of the original poem th th i l Firstly, there is the poetic, or spiritual, requirement : Rule I.-The translation must reproduce, in suitable language, the spirit and meaning of the original poem. Secondly therearethemusicalrequirements : language, p g g p Secondly, there are the musical requirements : Rule 2.-The translation must preserve the rhythm of themusicandthe it ofthemusical h language, p g g p Secondly, there are the musical requirements : Rule 2 The translationmust the h th of Secondly, there are the musical requirements : Rule 2.-The translation must preserve the rhythm of the music and the unity of the musical phrase. hi hb th Secondly, there are the musical requirements Rule 2.-The translation must preserve the rhythm o the music and the unity of the musical phrase. TO THE EDITOR OF 'THE MUSICAL TIMES.' original poem. Rule Iz.-The translation must reproduce the rhymin scheme of the original poem. original poem. Rule Iz.-The translation must reproduce the rhymin scheme of the original poem. SIR,-My recollection is that Dr. Bairstow some time since was reported to have contrasted Handel with Bach; and not only to have pronounced Bach the greater composer but to have expressed, with considerable emphasis, his opinion that Handel was far inferior. That Bach was on the whole the greater composer is also my opinion, as I have repeatedly stated; but I do not consider the gap between them nearly so great as Dr. Bairstow was reported to have affirmed, if my recollection is accurate. If I misrepresented his views, I apologise &c., H DAVEY g p Rule 3.--The translation must reproduce the rhythm of the original poem. g p Rule 3.--The translation must reproduce the rhythm of the original poem. g p Rule i4.-The translation must reproduce all the special poetic devices of the original poem, such as allitera- tion, assonance, grouping of vowels and consonants. g p Rule i4.-The translation must reproduce all the special poetic devices of the original poem, such as allitera tion, assonance, grouping of vowels and consonants. g p g Now it is obvious that no translation can satisfy all these requirements at once. In this lies the peculiar difficulty of translating : at every point a large number of these require- ments have to be set aside, and it rests with the tact, g p g Now it is obvious that no translation can satisfy all these requirements at once. In this lies the peculiar difficulty of translating : at every point a large number of these require- ments have to be set aside, and it rests with the tact, H. DAVEY. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions 'THtE DEPRECIATION AND APPRECIATION OF HANDEL.' reproduced by Rule 7.-The consonants of the original poem must be reproducedbysimilar consonants in the translation. reproduced y Rule 7.-The consonants of the original poem must be reproduced by similar consonants in the translation. f b reproduced by Rule 8.-Suitable moments for breathing must be provided. f tb h reproduced by Rule 8.-Suitable moments for breathing must be provided. d b f tb h THE TRANSLATION OF SONGS AND OPERAS INTO ENGLISH. Tropos will be requisitioned, and that from each, one member, the 8th will be selected for use. Tropos will be requisitioned, and that from each, one member, the 8th will be selected for use. , This sequence, which represents a segment of the Harmonic Series, is in reality seen to consist of the keynotes of seven different minor keys borrowed from so many Tropoi. A similar explanation applies to the time element in the pitch of the notes and to the vibrational law when the true harmonics of a clang are considered from the point of view of the physical basis of sound, but there is no opportunit or time to go into the subject here. It is the old question of the hen and the egg. BY N. DE V. HART. The average translation is bad from many points of view, but all its faults arise from one single, fundamental error: the translator has a false conception of the aim of a translation The aim of the average translation is to reproduce in English the effect of the poem as it is when written on paper and read : the translation appeals to the eve of a reader. The true aim should be to reproduce the effect of the poem as it is sung: it should appeal to the ear of the person who listens to the translation set to music. This is the root of the matter stated broadly; its full significance will become clear as we proceed. *353 x 15 16 = '331 ,, ,, ,, , *353 x 15 16 = '331 ,, ,, ,, , *353 x 15 16 = '331 ,, ,, ,, , ,, The difference = 72 millimetres, or about 14 times more than Mr. Fox Strangways'scomputation. This content downloaded from 131.91.169.193 on Fri, 06 Nov 2015 02:12:03 UTC All use subject to JSTOR Terms and Conditions
https://openalex.org/W3120019220
https://amu.hal.science/hal-04146387/document
English
null
Oligodendroglial Epigenetics, from Lineage Specification to Activity-Dependent Myelination
Life
2,021
cc-by
20,213
To cite this version: Mathilde Pruvost, Sarah Moyon. Oligodendroglial Epigenetics, from Lineage Specification to Activity- Dependent Myelination. Life, 2021, 11 (1), pp.62. ￿10.3390/life11010062￿. ￿hal-04146387￿ Distributed under a Creative Commons Attribution 4.0 International License Citation: Pruvost, M.; Moyon, S. Oligodendroglial Epigenetics, from Lineage Specification to Activity- Dependent Myelination. Life 2021, 11, 62. https://doi.org/10.3390/ life11010062 HAL Id: hal-04146387 https://amu.hal.science/hal-04146387v1 Submitted on 30 Jun 2023 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Mathilde Pruvost and Sarah Moyon * Mathilde Pruvost and Sarah Moyon * Neuroscience Initiative Advanced Science Research Center, CUNY, 85 St Nicholas Terrace, New York, NY 10031, USA; mpruvost@gc.cuny.edu * Correspondence: smoyon@gc.cuny.edu; Tel.: +1-212-413-3187 Abstract: Oligodendroglial cells are the myelinating cells of the central nervous system. While my- elination is crucial to axonal activity and conduction, oligodendrocyte progenitor cells and oli- godendrocytes have also been shown to be essential for neuronal support and metabolism. Thus, a tight regulation of oligodendroglial cell specification, proliferation, and myelination is required for correct neuronal connectivity and function. Here, we review the role of epigenetic modifications in oligodendroglial lineage cells. First, we briefly describe the epigenetic modalities of gene regulation, which are known to have a role in oligodendroglial cells. We then address how epigenetic enzymes and/or marks have been associated with oligodendrocyte progenitor specification, survival and pro- liferation, differentiation, and finally, myelination. We finally mention how environmental cues, in particular, neuronal signals, are translated into epigenetic modifications, which can directly influ- ence oligodendroglial biology. Keywords: oligodendrocyte; oligodendrocyte progenitor cell; epigenetics; DNA methylation; his- tone; chromatin; long non-coding RNA; lamina 1. Introduction Oligodendrocytes (OLs) are glial cells of the central nervous system (CNS) that rep- resent 20% of cells in an adult mouse brain and even up to 40% of neural cells in human neocortical regions [1,2]. While long considered simply as myelin-producing cells, many studies have since shown that they also interact with and respond to their environment, especially to neuronal signals [3]. Oligodendroglial cells in the forebrain originate from neural stem cells (NSCs) in three successive developmental waves, namely, ventral, dor- sal, and eventually cortical, with the latest generating 50% of the total oligodendrocyte progenitor cells (OPCs) [4–6]. OPC specification in the spinal cord also follows two waves, from ventral to dorsal, resulting in approximately 80% of ventrally derived OPCs [7–9]. Once generated, OPCs can proliferate, migrate, and differentiate into mature myelinating OLs [10–13]. A proportion of these OPCs are also maintained as undifferentiated progen- itors in the adult CNS, which directly interact with neurons and contribute to adult mye- lination and remyelination [14–16]. Received: 19 December 2020 Accepted: 10 January 2021 Published: 15 January 2021 Received: 19 December 2020 Accepted: 10 January 2021 Published: 15 January 2021 Publisher’s Note: MDPI stays neu- tral with regard to jurisdictional claims in published maps and insti- tutional affiliations. y Oligodendroglial specification and then differentiation are regulated by the dynamic expression of transcription factors, which control the expression of lineage-specific genes (e.g., Ascl1, Olig1, Sox10) or myelinating genes (e.g., Yy1, Myrf) [17–21]. While these tran- scriptional events are tightly orchestrated, environmental cues are also critical in this pro- cess. For instance, OPCs have been shown to be highly responsive to neuronal activity, which can induce their proliferation or differentiation to mature oligodendroglial cells [22]. The response of OPCs toward external stimuli, and in particular, neuronal activity, is crucial for neuronal network organization and adaptivity. For example, myelin plastic- ity and remodeling in response to neuronal activity is known to be essential for learning Copyright: © 2021 by the authors. Li- censee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and con- ditions of the Creative Commons At- tribution (CC BY) license (http://crea- tivecommons.org/licenses/by/4.0/). www.mdpi.com/journal/life Life 2021, 11, 62. https://doi.org/10.3390/life11010062 Life 2021, 11, 62 2 of 30 2 of 30 and cognitive function of the CNS [23–25]. 1. Introduction It is also crucial in response to injury and dis- ease, when rapid OPC proliferation and differentiation form new myelin to support re- covery and protect axons and neuronal function [26]. The integration of external cues, such as neuronal activity, into intrinsic signals is mediated by epigenetic modifications, which are known to control chromatin organization and, in turn, regulate gene expres- sion. In particular, chromatin condensation and accessibility are regulated by DNA meth- ylation, histone modifications, and chromatin remodelers, which interact with long non- coding RNA (lncRNA) and microRNA (miRNA), as well as nuclear organization via lam- ins [27]. Recently, the methylation of mRNA has also been described as an epigenetic mod- ification, resulting in gene expression regulation at the translational level [28] (Figure 1). In this review, we summarize how external signals communicate to the oligoden- droglial cell lineage through epigenetic “messages.” In the first part, we focus on how each step of oligodendrocyte development, from lineage choice to myelination, is regulated by epigenetic modifications. We then focus on how environmental factors, and especially neu- ronal activity, can influence OPC gene expression by triggering epigenetic changes. 2.2. Histone Post-Translational Modifications A second epigenetic modification refers to the post-translational changes of amino acids in histones, which are the fundamental units of the nucleosome and are composed of an octamer of the four core histone proteins (H2A, H2B, H3, and H4). The tail of each histone is a site of post-translational modification, including methylation, acetylation, cit- rullination, SUMOylation, phosphorylation, ubiquitination, proline isomerization, and ADP-ribosylation [38]. The main epigenetic histone modifications described in oligoden- drocytes occur on H3 or H4, at specific lysine (K) or arginine (R) sites, and consist of acet- ylation (ac), methylation (me1, me2, me3), or citrullination. The histone methylation of lysine residues is catalyzed by histone lysine methyltransferases (HMTs) (e.g., KMT2/MLL, SETDB1, and SV39H1/2, as well as SUZ12, EZH2, and EED, which are parts of the polycomb repressive complex 2, PRC2, in oligodendroglial cells) at lysine residues and by protein arginine methyltransferases (PRMTs) (e.g., PRMT1 and PRMT5 in oli- godendroglial cells) at arginine residues [39–44]. Histone acetylation is catalyzed by his- tone acetyltransferases (HATs) (e.g., CREBBP and EP300 in oligodendroglial cells), and reversed by histone deacetylases (HDACs) (e.g., HDAC1, HDAC2, HDAC3, SIRT1, and SIRT2 in oligodendroglial cells) [45,46]. Histone modifications can either lead to gene ac- tivation (e.g., H3K9ac, H3K14ac, H3K27ac, H3K4me1, H3K4me3) or gene repression (e.g., H3K9me3, H3K27me3, H4R3me2s) [47–53]. Of special interest, H3K27ac and H3K4me1, as well as H3K4me3, have been strongly associated with gene activation at enhancers and transcription start sites, respectively, and are experimentally used to identify these regions [53,54]. Histone arginine citrullination, which is mainly catalyzed by peptidylarginine dei- minase 2 (PADI2) in oligodendroglial cells, has a dual role in gene regulation. Citrullina- tion is directly associated with gene activation, as it reduces proteins’ charges and histone- DNA histone interaction, favoring chromatin decondensation, while indirectly, it also blocks repressive methylation on arginine residues [55]. 2.1. DNA Methylation and Hydroxymethylation DNA methylation consists of the addition of a methyl group, mostly occurring at the fifth carbon of cytosines in CpG dinucleotide sites in eukaryotic genomes [29]. In mam- malian cells, the three major DNA methyltransferases (DNMTs) are DNMT1, DNMT3A, and DNMT3B [30,31]. Their catalytic activity is associated with the methylation of cyto- sines, which could prevent the access of transcription factors to their binding sequence and the associated recruitment of cofactors that modulate chromatin conformation. This modification results in transcriptional repression occurring in the gene promoter regions [32,33]. DNA hydroxymethylation, in contrast, is catalyzed by the ten-eleven translocation (TET) enzymes, mainly TET1, TET2, and TET3, and refers to the oxidation of 5-methylcy- tosine to 5-hydroxymethylcytosine [34,35]. DNA hydroxymethylation is mostly associ- ated with transcriptional activation and is often detected at gene bodies and transcrip- tional starting sites [36,37]. 2. Epigenetic Modifications Involved in Chromatin Accessibility and the Regulation of Gene Expression As previously mentioned, epigenetic modifications, in the current acceptation of the term, encompass the various mechanisms that modulate chromatin organization and gene expression. We discuss almost exclusively those modifications that have been shown to regulate cells within the oligodendroglial cell lineage (Figure 1). Figure 1. Several epigenetic layers are known to regulate gene expression, from DNA modification to post-translational histone modifications, chromatin remodeling, association with the nuclear lamina, and post-transcriptional regulation by RNA methylation and long non-coding and mi- croRNA. Figure 1. Several epigenetic layers are known to regulate gene expression, from DNA modification to post-translational histone modifications, chromatin remodeling, association with the nuclear lamina, and post-transcriptional regulation by RNA methylation and long non-coding and mi- croRNA. Life 2021, 11, 62 3 of 30 3 of 30 2.1. DNA Methylation and Hydroxymethylation 2.4. 3D Chromatin Organization with Nuclear Lamina Higher-order chromatin organization includes further genomic interaction with the nuclear lamina, which maintains nuclear structural integrity [66]. Lamina-associated do- mains are usually enriched for the repressive histones (e.g., H3K27me3 and H3K9me3) characteristic of heterochromatin [67]. Two protein families have been shown to form the nuclear lamina: A-type lamins (lamin A and C) and B-type lamins (lamin B1 and B2) (e.g., LMNB1 in particular in oligodendroglial cells) [66,68]. 2.5. Post-Transcriptional Modifications Long non-coding RNA (lncRNA) and microRNA (miRNA) are transcribed but non- translated RNAs that regulate gene expression at transcriptional and post-transcriptional levels, in particular, during CNS development [69–72]. The lncRNA are more than 200 nucleotides long, while the miRNA, which are processed by RNA polymerase II and DICER, are typically 20–25 nucleotides long. In addition to the ability to bind and regulate chromatin accessibility, lncRNA and miRNA can affect and regulate the expression of nascent mRNA transcripts that are encoded at distant genome loci due to post-transcrip- tional mechanisms [73]. Recently, additional chemical modifications have been identified at the messenger RNA (mRNA) level [74,75]. The most common in mammalian cells, namely, methylation of N6-methyladenosine (m6A) mRNA, is involved in post-transcriptional regulation of gene expression by influencing mRNA stability, translation, localization, and splicing [76– 78]. In oligodendroglial cells, METTL14 has been described as the core component of the m6A methyltransferase complex, while FTO is the main mRNA demethylase [79,80]. An accumulation of evidence has revealed that all layers of epigenetic modifications described in this section are involved in the oligodendroglial cell lineage, from specifica- tion choice to survival and differentiation. 2.3. Chromatin Remodeling Chromatin organization can further be rearranged by chromatin remodeling en- zymes, which use ATP as an energy source to control chromatin accessibility [56]. Chro- matin remodelers are classified into four different families, including the switch/sucrose- non-fermenting (SWI/SNF) complex (e.g., BRG1 in oligodendroglial cells), the inositol re- quiring 80 (INO80)/SWI-related (SWR) complex (e.g., EP400 in oligodendroglial cells), and the chromodomain-helicase DNA-binding (CHD) complex (e.g., CHD7 and CHD8 in oli- godendroglial cells) [57–62]. Chromatin remodelers have the ability to regulate the acti- vation or repression of gene expression by opening or closing the chromatin architecture [63]. Among the chromatin architectural proteins, the high-mobility group (HMG) super- family comprises three families of proteins, namely, HMGA, HMGB, and HMGN, which Life 2021, 11, 62 4 of 30 dynamically bind to chromatin, promoting its decompaction and binding to nuclear reg- ulatory factors [64,65]. dynamically bind to chromatin, promoting its decompaction and binding to nuclear reg- ulatory factors [64,65]. 2.4. 3D Chromatin Organization with Nuclear Lamina 3. Epigenetic Marks with Roles in Oligodendroglial Cell Lineage NSCs in the developing and adult mammalian brain harbor the ability to self-renew and to generate neurons, astrocytes, and oligodendrocytes [81]. The differentiation of NSCs occurs in response to extracellular signals, along with the interplay between dy- namic epigenetic modifications and lineage gene expression regulation [82,83]. Lineage specification requires both the activation of lineage genes and the repression of alternative lineage genes. In this section, we will focus on how the transition from NSCs to OLs is regulated by successive waves of DNA methylation and hydroxymethylation, as well as by histone modifications, chromatin remodelers, microRNAs, and lncRNA (Figure 2 and Table 1). Life 2021, 11, 62 5 of 30 Figure 2. A combination of epigenetic modifications is involved in oligodendroglial cell specification, survival and proliferation, differentiation, and myelin remodeling. Red arrows represent activation and green arrows represent repression. Modifications and/or enzymes in faded font represent data that are still being discussed. Figure 2. A combination of epigenetic modifications is involved in oligodendroglial cell specification, survival and proliferation, differentiation, and myelin remodeling. Red arrows represent activation and green arrows represent repression. Modifications and/or enzymes in faded font represent data that are still being discussed. Life 2021, 11, 62 6 of 30 Table 1. List of epigenetic marks and their roles during OPC specification. DNA Modification Enzyme/Mark Role Targeted Genes or Functions Model Methods References DNA methylation Negative role in NSCs differentia- tion to glia Repression of astro- gliogenesis (Stat3 binding element in Gfap promoter) In vitro neuroepithelial cells BS PCR on neuroepi- thelial cells Takizawa et al. [84] DNMT1 Negative role in NPCs differentia- tion to glia Control of the timing of astrogliogenesis by repression of astro- cyte-specific genes (Gfap, Stat1) Nestin-cre;Dnmt1flox NPCs BS PCR on Nestin- cre;Dnmt1flox NPCs Fan et al. [85] Positive role in NSCs differentia- tion to glia Repression of neu- ron-specific genes (Dlx1, Dlx2, Trb1) Sox2-EGFP mice WGBS on NS/PCs from Sox2-EGFP transgenic mice Sanosaka et al. [86] DNMT1 Positive role in NSCs specification to OPCs Repression of neu- ron-specific (Ndrg4, Camk1, Ephb2) and as- trocyte-specific (Aldh1l1, Pax6, Rfx4) genes Olig1cre;Dnmt1flox mice RNA-Seq and ERRBS on sorted neonatal OPCs and OLs, RNA- Seq on sorted Olig1cre;Dnmt1flox OPCs Moyon et al. [87] DNA demethyla- tion Positive role in ESCs transition to NSCs Activation of tran- scription factors (Sox2, Sox21, Ascl1) Sox2-EGFP mice WGBS on NS/PCs from Sox2-EGFP transgenic mice Sanosaka et al. 3. Epigenetic Marks with Roles in Oligodendroglial Cell Lineage In vitro NSCs, Olineu, OPCs + siRNA/inhibi- tors Life 2021, 11, 62 7 of 30 7 of 30 HDAC3 Positive role in as- trocyte–OL fate switch Activation of enhanc- ers of OPC genes (Olig2, Ng2) and re- pression of astrogli- ogeneis genes (Stat3) and neuronal genes (Bdnf) In vitro astrocytes and OPCs + inhibitors/ex- pression vectors; Olig1Cre;Hdac3flox, PDGFRaCre- ERT2;Hdac3flox; Syn1Cre;Hdac3flox mice ChIP-Seq on in vitro OPCs and OL; RNA- Seq on optic nerves from Olig1cre;Hdac3flox mice Zhang et al. [99] SIRT1 (H3K9deac) Negative role in NSCs differentia- tion to OPCs Repression of differ- entiation to OPCs (deacetylation at the Pdgfrα promoter) In vitro OPCs, NS/PCs + inhibitors; Nestin- Cre;Sirt1flox mice ChIP-qPCR on in vitro NS/PCs Rafalski et al. [100] Chromatin Organi- zation Enzyme/Mark Role Targeted Genes or Functions Model Methods References Chromatin remod- elers HMGA1, 2 Negative role in NPCs differentia- tion to glia Repression of astro- genic transition In vitro and in vivo NPCs + shRNA/over- expression Kishi et al. [101] HMGB1, 2, 3, 4 Dynamically ex- pressed in NSCs In vitro NSCs, Nestin- GFP mice, HMGB2−/− mice IHC, qPCR, shotgun proteomics on in vitro NSCs Abraham et al. [102] HMGB2 Possible role in NSCs proliferation and maintenance In vitro NSCs, Nestin- GFP mice, HMGB2−/− mice IHC, qPCR, shotgun proteomics on in vitro NSCs Abraham et al. [102] Role in neuron– glia fate switch In vitro NS/PCs, HMGB2−/− mice Bronstein et al. [103] HMGB4 Negative role in NSCs differentia- tion to OPCs Regulation of neu- ronal, astrocyte, oli- godendrocyte genes (Fabp7, NeuroD1, Gfap, Ppp1r14a) In vitro neurons, neuro- spheres and various cell lines + lentivirus; HMGB4 Vivo-Morpho- linos Microarray on HMGB4-EGFP over- expressing HEK 293T cells Rouhiainen et al. [104] HMGN family Positive role in neuron–glia fate switch Modulation of the re- sponse to gliogenic signals In vitro NPCs + shRNA/overexpres- sion qPCR, IHC, BS and microarray on in vitro NPCs Nagao et al. [105] BRG1 Positive role in NSCs maintenance and gliogenesis Repression of neu- ronal differentiation in NSCs In vitro NSCs, Nestin- Cre;Brg1flox mice IHC; DNA microar- ray on CNS tissues of NestinCre;Brg1flox mice Matsumoto et al. [106] Positive role in NSCs differentia- tion to neurons Activation of neu- ronal genes (Ngn and NeuroD) In vitro pluriopotent P19 cells + plasmids, Xenopus + Brg1 mor- pholino IHC Seo at al. [107] Limited role in NSCs zebrafish + Brg1 mor- pholino IHC, ISH Gregg et al. 3. Epigenetic Marks with Roles in Oligodendroglial Cell Lineage [86] Positive role in NSCs differentia- tion to glia Activation of glio- genic promoters (NFI, Tcf3, Gfap, Kcnj10, Sox8) Activation of glio- genic promoters (Stat3 binding site in Gfap promoter, Aldoc) In vitro NPCs and as- trocytes from mice MIAMI and BS on NPCs and astrocytes Hatada et al. [88] DNA hy- droxymethylation TET1 Positive role in NSCs differentia- tion to OPCs Activation of OL genes (Olig1, Sox10, Id2/4) Olig1cre;Tet1flox mice hMeDIP-Seq on in vitro neonatal NSCs and OPCs, RNA-Seq on in vitro Olig1cre;Tet1flox OPCs Zhang et al. [89] Histone Modifica- tion Enzyme/Mark Role Targeted Genes or Functions Model Methods References Histone methyla- tion PRC2 (EZH2) (H3K27me3) Positive role in NSCs differentia- tion to OPCs In vitro primary cul- tures + Olineu + Ezh2 expression vec- tor/shRNA Sher et al. [90] Repression of neu- ronal (NeuroD2, Tlx3), astrocytic (Tal1), OL (Olig2, Pdgfra, Nkx2.2) genes In vitro NSCs and OPCs + shRNA ChIP-Seq on in vitro NSCs and OLs Sher et al. [91] PRC2 (EED) (H3K27me3) Positive role in as- trocyte–OPC fate switch Olig1cre;Eedflox and PdgfracreRT;Eedflox mice Wang et al. [92] H3K27me3 Positive role in NSCs differentia- tion to OPCs Repression of global lineage alternative choice In vitro OPCs; Cnp- EGFP mice ChIP-Seq on in vitro OPCs and OLs (H3K27me3) Liu et al. [93] PRMT1 Positive role in NSCs differentia- tion to OPCs Nestin-cre;Prmt1flox mice Hashimoto et al. [94] Histone (de)acety- lation CBP (H3K9/K14ac) Positive role in NSCs differentia- tion Sequential activation of promoters of neu- ronal (Tuba1a), astro- cytes (Gfap), OL (Mbp) In vitro cortical precur- sors + siRNA/inhibi- tors; cbp+/− mice ChIP-qPCR on cbp+/− cortices Wang et al. [95] HDACs Positive role in oli- godendrogenesis In vivo HDAC inhibi- tion in rats Liu et al. [96] HDAC1/HDAC2 Positive role in Shh-induced oli- godendrogenesis Repression of genes associated with Notch signaling (Hey1, Hey2) and Wnt signaling (Tbx3) In vitro OPCs, Olineu cells + shRNA/inhibi- tors ChIP on in vitro Olineu and GeneChip on in vitro OPCs Wu et al. [97] HDAC2 (H3K9deac) Negative role in NSCs differentia- tion to OPCs Repression of oli- godendroglial differ- entiation genes (Sox10) in the pres- ence of thyroid hor- mone In vitro NSCs, Olineu, OPCs + siRNA/inhibi- tors ChIP-Seq on in vitro NSCs Castelo-Branco et al. [98] Table 1. List of epigenetic marks and their roles during OPC specification. 3.2.2. Histone Acetylation and Deacetylation Regulate the Lineage Specification The normal differentiation of cortical precursors in vitro and in vivo in mice is ac- companied by sequential histone acetylation and the subsequent activation of promoters of neuronal and astrocyte genes, then postnatally in oligodendrocyte genes, such as Mbp and Plp. The key histone acetyltransferase in this process is the CREB binding protein (CBP), which regulates H3K9/K14 acetylation [95]. Indeed, blocking histone deacetylation in rats reduces oligodendrogenesis and favors the differentiation of cells along alternative lineage choices [96]. However, histone deacetylation has also been shown to be involved in the regulation of neural progenitors [120]. In particular, HDAC2 and HDAC3 associate with key genes regulating the differentiation of NSCs [98,121]. In NSCs, HDAC3 acts first as a repressor of neuronal differentiation in cortical NSCs [98], then antagonizes astrogli- ogenesis by inhibiting the acetylation levels of Stat3, which is a core effector in the Janus kinase (JAK)–STAT pathway [99]. Astrocytic genes are also repressed by HDAC1 and HDAC2 in vitro during Sonic Hedgehog (Shh)-induced oligodendrogenesis [97]. Finally, the cooperation of HDAC3 with the histone acetyltransferase p300 targets and activates enhancers of the OPC genes, such as Olig2 [99]. Inversely, the differentiation to OPC is prevented by HDAC2 and sirtuin 1 (SIRT1), which, at least in part, can inhibit specific OL differentiation genes, such as the key transcription factor Sox10 [98,100]. 3. Epigenetic Marks with Roles in Oligodendroglial Cell Lineage [108] Post-Transcrip- tional Modifica- tion Enzyme/Mark Role Targeted Genes or Functions Model Methods References Long non-coding RNA lnc-158 Positive role in NSCs differentia- tion to OPCs Activation of OL genes (Cnp, Mbp, Mag, Osp) In vitro NSCs + lnc-158 overexpression and siRNA Li et al. [109] lnc-OPC Positive role in NSCs differentia- tion to OPCs Activation of OL genes (Mbp, Plp1, Cnp) NSCs + shRNA ChIP-Seq and RNA- Seq on in vitro NSCs Dong et al. [110] Neat1 Positive role in NSCs differentia- tion to OPCs Activation of OL genes (Olig1, Olig2, Gpr17, Sox8) Neat1−/− mice ChIP-Seq on human tissues; RNA-Seq on Neat1−/− mouse brains Katsel et al. [111] Sox8OT Possible role in NSCs differentia- tion to OPCs Via Sox8 activation Descriptive study Mercer et al. [112] MicroRNA miR-124 Positive role in NSCs differentia- tion to neurons Repression of Ezh2 expression In vitro NSCs + N2a neuroblastoma + P19 cells + overexpression Microarray on N2a cells Neo et al. [113] miR-153 Negative role in NSCs differentia- tion to glia Repression of glio- genic genes (Nfia/b) In vitro ESCs + shRNA + artifical miRNAs Tsuyama et al. [114] miR-17/106 Positive role in NSCs differentia- tion to glia Activation of gliogen- esis (p38) in vitro ESCs and mouse embryos + lenti- virus Naka-Kaneda et al. [115] BS: bisulfite sequencing, ChIP: chromatin immunoprecipitation, ERRBS: enhanced reduced representation bisulfite se- quencing, hMeDIP-Seq: hMeDIP: hydroxyMethylated DNA immunoprecipitation; IHC: immunohistochemistry; ISH: in situ hybridization; NPCs: neural precursor or progenitor cells, NS/PCs: neural stem/progenitor cells; MIAMI: microarray- based integrated analysis of methylation by isoschizomers, qPCR: quantitative polymerase chain reaction; Seq: sequenc- i WGBS h l bi lfit i IHC, qPCR, shotgun proteomics on in vitro NSCs IHC, qPCR, shotgun proteomics on in vitro NSCs BS: bisulfite sequencing, ChIP: chromatin immunoprecipitation, ERRBS: enhanced reduced representation bisulfite se- quencing, hMeDIP-Seq: hMeDIP: hydroxyMethylated DNA immunoprecipitation; IHC: immunohistochemistry; ISH: in situ hybridization; NPCs: neural precursor or progenitor cells, NS/PCs: neural stem/progenitor cells; MIAMI: microarray- based integrated analysis of methylation by isoschizomers, qPCR: quantitative polymerase chain reaction; Seq: sequenc- ing, WGBS: whole-genome bisulfite sequencing. BS: bisulfite sequencing, ChIP: chromatin immunoprecipitation, ERRBS: enhanced reduced representation bisulfite se- quencing, hMeDIP-Seq: hMeDIP: hydroxyMethylated DNA immunoprecipitation; IHC: immunohistochemistry; ISH: in situ hybridization; NPCs: neural precursor or progenitor cells, NS/PCs: neural stem/progenitor cells; MIAMI: microarray- based integrated analysis of methylation by isoschizomers, qPCR: quantitative polymerase chain reaction; Seq: sequenc- ing, WGBS: whole-genome bisulfite sequencing. 3.1. DNA Methylation Waves in NSC Differentiation to OPCs The specification of NSCs from embryonic stem cells (ESCs) toward neurogenesis and gliogenesis occurs through specific methylation patterns. A study from Sanosaka et al. suggests that the demethylation of neuron-specific genes first occurs as ESCs transition to NSCs [86], while methylation on glial promoters, such as glial fibrillary acidic protein (GFAP), is maintained to suppress its expression [84]. Gliogenic transition is then favored and occurs via the demethylation of glial gene promoters and genes involved in gliogenic pathways, as well as the de novo methylation of neuronal genes [84–86,88,116–118]. Sub- sequently, the de novo methylation of astrocytic genes occurs during the specification of oligodendrocytes [85,87], along with the hydroxymethylation of oligodendrocyte genes, such as Olig1, Sox10, and Id2/4 [89,119]. 3. Epigenetic Marks with Roles in Oligodendroglial Cell Lineage Life 2021, 11, 62 8 of 30 8 of 30 3.1. DNA Methylation Waves in NSC Differentiation to OPCs 3.2.1. Repression of Lineage-Specific Genes in NSCs by Histone Methylation Lineage determination is also regulated by histone modifications, such as H3K27me3, which is catalyzed by the polycomb repressive complex 2 (PRC2) and is con- stituted of the enhancer of the zeste homolog 2 (EZH2) and embryonic ectoderm develop- ment (EED). EZH2 and EED are highly expressed in proliferating NSCs and target genes related to developmental processes and neurogenesis [91,92]. As NSCs differentiate into OPCs, EZH2 and EED expression remain high, while their levels decrease in cells transi- tioning to neuronal and astrocyte lineages [90,92]. Moreover, the ChIP-sequencing of H3K27me3 in postnatal-day-1 rat OPCs highlights its genomic regulation of genes in- volved in the “global alternative lineage choice” [93]. Indeed, conditional ablation of Eed favors the differentiation into astroglial cells, to the detriment of the oligodendroglial lin- eage [92]. OPC lineage progression also depends on arginine methylation, as the condi- tional knockdown of Prmt1 in NSCs drastically reduces the number of OLs in mice [94]. 3.2.2. Histone Acetylation and Deacetylation Regulate the Lineage Specification 3.2. Histone Modifications 3.2.1. Repression of Lineage-Specific Genes in NSCs by Histone Methylation 3.4. Post-Transcriptional Modifications Several miRNAs play key roles in neural lineage development. For instance, miR-124 is necessary for fate specification into neurons by reducing the expression of EZH2 [113]. Other miRNAs are necessary for the neurogenic-to-astrogenic transition, such as miR-153, which regulates the acquisition of gliogenic competence [114], or miR-17/106 [115]. To our knowledge, no miRNA has been described as being specifically required for the differen- tiation from an NSC to an OL. Dong and collaborators have investigated whether lncRNAs are involved in the reg- ulation of NSC differentiation into OPCs [110]. They identified lnc-OPC, a specific and highly expressed lncRNA in OPCs, which is critical for cell fate determination. In vitro loss- and gain-of-functions experiments that targeted lnc-OPC, as well as Sox8OT, Neat1 and lnc-158, have highlighted the positive prominent role of lncRNAs for oligodendroglial specification [109–112]. 3.3. Chromatin Reorganization during the Differentiation from NSCs to OPCs Chromatin structure and organization in NSCs changes during their specification throughout brain development to generate the major cell types of the CNS. These dynamic chromatin states are modulated by chromatin architectural proteins, such as the HMG proteins, and chromatin modifiers, such as BRG1. In the early stages of NSCs character- ized by a neurogenic potential, HMGA1 and 2 proteins are highly expressed and mediate Life 2021, 11, 62 9 of 30 9 of 30 the global chromatin opening. As the levels of HMGA proteins decrease, chromatin be- comes more condensed in a stage-dependent manner that allows for an astrogenic transi- tion [101]. All four mammalian forms of HMGB (HMGB1, 2, 3, and 4) are expressed in proliferating NSCs [102] with specific roles described for HMGB2 and HMGB4 in the neu- rogenic-to-gliogenic fate transition [103,104]. Members of the HMGN family (HMGN1, 2 and 3) also positively regulate the neuron–glia fate switch [105]. BRG1, within the SWI/SNF-related chromatin remodeling complex, is a critical regu- lator of NSC specification by repressing neuronal differentiation, while favoring gliogen- esis and differentiation in mammalian neural development. In mice, the ablation of Brg1 specifically in NSCs does not impact the initial neuronal differentiation but abolishes glial generation and differentiation, as seen by a dramatic decrease in astrocyte, oligodendro- cyte progenitor, and myelin protein markers in late embryonic stages [106]. However, the function of BRG1 appears different in lower vertebrates, as the inhibition of BRG1 blocks neuronal differentiation in Xenopus [107] and has restricted effects in retinal ganglion cells in zebrafish [108]. 4. Epigenetic Marks Maintain an Oligodendroglial Progenitor Cell Pool In the CNS, oligodendroglial cells are highly sensitive to their environment, which influences their survival and proliferation rate as immature OPCs and/or their differenti- ation capacities into myelinating OLs. Overall, this tight regulation maintains a relative homeostatic pool of OPCs throughout time and space in both the developmental and adult CNS [122]. In particular, OPCs proliferate in response to chemical cues, such as mitogens, growth factors, and cytokines [10,123–127], and to neuronal cues, such as electrical activity [22,128–131]. OPCs are now thought to translate these specific extracellular cues into in- trinsic signals that affect survival and proliferation via the regulation of epigenetic modi- fications (Figure 2 and Table 2). Life 2021, 11, 62 10 of 30 Table 2. List of epigenetic marks and their roles during OPC survival and proliferation. DNA Modification Enzyme/Mark Role Targeted Genes or Functions Model Methods References DNA methylation DNMT1/DNMT3A Positive role in OPC sur- vival and proliferation In vitro OPCs + siRNA Egawa et al. [132] No effect on OPC sur- vival Olig1cre;Dnmt1fl ox and Olig1cre;Dnmt3fl ox mice Moyon et al. [133] DNMT1 Positive role in OPC pro- liferation Increased DNA methyl- ation and downregula- tion of cell cycle and cell proliferation genes (Cdc6, Meis2) Olig1cre;Dnmt1fl ox mice RNA-Seq and ERRBS on sorted neonatal OPCs and OLs, RNA- Seq on sorted Olig1cre;Tet1flox OPCs Moyon et al. [87] DNA demethylation 5cac mark 5cac enriched in glial cells during NSCs dif- ferentiation (Gfap, Olig1/2) Descriptive study in vivo Wheldon et al. [118] DNA hydroxymeth- ylation TET1 Slight positive role dur- ing developmental OPC proliferation Olig1cre;Tet1flox mice Zhang et al. [89] No role in adult OPC proliferation Olig1cre;Tet1flox and Pdgfra- creRT;Tet1flox mice Moyon et al. [134] Histone Modifica- tion Enzyme/Mark Role Targeted Genes or Functions Model Methods References Histone methylation PRC2 (EED) (H3K27me3) Positive role in OPC pro- liferation (in brain, not spinal cord) Olig1cre;Eedflox and Pdgfra- creRT;Eedflox mice Wang et al. [92] PRMT5 Positive role in OPC sur- vival Inhibition of p53 path- way Olig1cre;Prmt5flo x mice and in vitro OPCs + siRNA RNA-Seq on in vitro PRMT5- CRISPRKO OPCs Scaglione et al. [135] Histone (de)acetyla- tion H3K9ac/H3K14ac marks Activation of cell cycle genes In vitro OPCs + cMyc silencing Magri et al. [136] HDACs Positive role in OPC pro- liferation In vitro O4+ cells + HDAC inhibitors Conway et al. 4.2. Cell Cycle and Proliferation The nuclei of proliferative OPCs are mostly euchromatic and characterized by a re- laxed and transcriptionally competent chromatin structure that is enriched for permissive marks, such as histone acetylation (e.g., H3K9ac and H3K14ac) [100,136,143]. The OPC cell cycle is also regulated by the oncogene transcription factor cMyc in response to platelet- derived growth factor (PDGF) and fibroblast growth factor (FGF) mitogens [136,144]. cMyc recruits histone acetyltransferases and the ablation of cMyc in vitro can directly de- crease H3K9ac and H3K14ac in OPCs [136,145–147]. This suggests that, on top of acting as a DNA binding transcription factor, cMyc is capable of translating mitogenic extracel- lular stimuli into epigenetic signals in proliferative OPCs. p g g p Evidence suggests that the epigenetic regulation of OPC proliferation might differ depending on the CNS region. For example, differences in chromatin remodeling have been noted in brain and spinal cord tissues. The ablation of Chd7 or Chd8 in OPCs does not lead to an increased number of OPCs, since, despite an increase in cell cycle and pro- liferative genes, this also leads to an increase in apoptosis, at least in brain tissues [59,61,139]. Surprisingly, following Chd7 or Chd8 ablation, OPC proliferation is not in- creased but reduced in spinal cord tissues, where apoptosis is not perturbed [138,139]. p p p p OPC proliferation and cell cycle exits appear to be regulated by several epigenetic marks, including repressive ones, such as DNA and histone methylation. For example, the ablation of Dnmt1 in neonatal OPCs revealed reduced DNA methylation and defective gene repression, in particular, at cell cycle genes [87]. This result was not replicated in vivo in adult OPCs, suggesting once again that there is an age-dependent epigenetic regulation of transcriptome in OPCs [133]. However, the presence of DNA modification alone is not sufficient to induce precocious proliferation. Ablating DNMT1- or DNMT3A-mediated DNA methylation or EED-mediated histone methylation in OPCs does not result in ec- topic proliferation in vitro or in vivo, and even slightly negatively perturbates OPC pro- liferation during development [87,92,133]. Inversely, TET1-mediated DNA hydroxymeth- ylation does not seem to affect neonatal or adult OPC proliferation in vitro or in vivo [89,134]. 4.1. Apoptosis vs. Survival 4.1. Apoptosis vs. Survival Increasingly, evidence has revealed the heterogeneity of the OPC properties between the developmental and adult OPC pools, but also according to their CNS localization. OPC survival has been shown to be variably associated with DNA and histone modifica- tions and chromatin remodeling, depending on time and space. During development, the ablation of Dnmt1 or Dnmt3a in neonatal OPCs tends to induce DNA damage and decreases OPC survival in vitro [87,132]. However, these effects were not observed after the ablation of Dnmt1 and/or Dnmt3a in adult OPCs during re- myelination experiments [133]. OPC survival during development has also been associ- ated with the PRMT5-dependent H4R3me2s mark, as the ablation of PRMT5 in OPCs could activate p53 pathways and increase apoptosis [135]. In addition, neonatal non-pro- liferative OPCs are protected from apoptosis by the chromatin remodelers EP400 and CHD7, the latest known remodelers to control chromatin closing and p53 transcriptional repression [61,140]. In adult OPCs, cell survival is partially regulated by the CHD8, which shares many common binding sites with CHD7 [61]. Indeed, the global or oligoden- droglial-specific ablation of Chd8 in OPCs results in increased apoptosis, in particular, in adult spinal cord tissues, but not during development or in the brain [139]. 4. Epigenetic Marks Maintain an Oligodendroglial Progenitor Cell Pool [137] Chromatin Organi- zation Enzyme/Mark Role Targeted Genes or Functions Model Methods References Chromatin remodel- ers CHD7 Positive role in OPC sur- vival Regulation of cell cycle genes (Ccnd1, Cdk4 and Cdk6) and of cell sur- vival/apoptosis genes (e.g., p53/Trp53, Bax, Apaf1) Pdgfra- creRT;Chd7flox mice RNA-Seq on in vitro Pdgfra- creRT;Chd7flox OPCs Marie et al. [61] Positive role in OPC pro- liferation Pdgfra- creRT;Chd7flox mice Doi et al. [138] No role in OPC survival or proliferation Olig1cre;Chd7flox mice He et al. [59] CHD8 Positive role in adult OPC survival and prolif- eration in spinal cord Olig1cre;Chd8flox and Pdgfra- creRT;Chd8flox mice Zhao et al. [139] EP400 Positive role in OPC sur- vival Cnpcre;Ep400flox mice Elsesser et al. [140] Post-Transcrip- tional Modification Enzyme/Mark Role Targeted Genes or Functions Model Methods References MicroRNA DICER Positive role in OPC pro- liferation Olig1cre;Dicerflox mice Zhao et al. [141] m6A RNA methyla- tion PPRC2A Positive role in OPC pro- liferation Olig2cre;Pprc2afl ox and Nes- cre;Pprc2aflox mice Wu et al. [142] ERRBS: enhanced reduced representation bisulfite sequencing, Seq: sequencing. Table 2. List of epigenetic marks and their roles during OPC survival and proliferation. Table 2. List of epigenetic marks and their roles during OPC survival and proliferation. Life 2021, 11, 62 11 of 30 11 of 30 5. Epigenetic Marks Regulate the Oligodendroglial Progenitor Cell Differentiation into Mature Oligodendrocytes One of the main and most studied functions of neonatal and adult OPCs is their abil- ity to differentiate into mature OLs. OPC differentiation is in part an intrinsic propensity but is also regulated by environmental and neuronal cues that can influence epigenetic Life 2021, 11, 62 12 of 30 12 of 30 modifications (Figure 2 and Table 3). Recent studies highlighted the inter-neuronal (e.g., myelinated or non-myelinated axons) and intra-neuronal (e.g., variable size of internodes) heterogeneity of myelination in the CNS, which appeared to be essential for global brain connectivity and function, and which could reflect the heterogeneity of the OPC popula- tion and/or their environment [148–153]. tion and/or their environment [148 153]. Table 3. List of epigenetic marks and their roles during OPC differentiation. DNA Modification Enzyme/Mark Role Targeted Genes or Functions Model Methods References DNA methylation DNMT1 Positive role in OPC differentiation In vitro OPCs + siRNA Egawa et al. [132] Negative role in OPC differentiation and myelination Decreased DNA meth- ylation and upregula- tion of lipid synthesis and myelin formation genes (Mog, Mag, Gpr37) Olig1cre;Dnmt1flox mice (no phenotype in Cnpcre;Dnmt1flox mice) RNA-Seq and ERRBS on sorted neonatal OPCs and OLs, RNA-Seq on sorted Olig1cre;Dnmt1flox OPCs Moyon et al. [87] DNMT1/DNMT3A Positive role in OPC differentiation and re- myelination PlpcreRT;Dnmt1flox; Dnmt3flox mice Moyon et al. [133] DNA hydroxymeth- ylation TET1/TET2/TET3 Positive role in OPC differentiation In vitro OPCs + siR- NAs Zhao et al. [35] TET1 Positive role in OPC differentiation and (re)myelination (no role for TET3) Activation of cell differ- entiation genes (Mag) Olig1cre;Tet1flox mice hMeDIP-Seq on in vitro neonatal NSCs and OPCs, RNA-Seq on in vitro Olig1cre;Tet1flox OPCs Zhang et al. [89] Positive role in OPC differentiation and adult remyelination only (no role for TET2) Activation of cell differ- entiation, myelination, biosynthesis and neu- roglial communication genes (Slc family mem- bers, Pcdh family mem- bers) Olig1cre;Tet1flox and PdgfracreRT;Tet1flox mice RNA-Seq and RRHP on sorted adult OPCs and OLs, RNA-Seq on oligo-enriched Olig1cre;Tet1flox lesioned spinal cord Moyon et al. [134] Histone Modifica- tion Enzyme/Mark Role Targeted Genes or Functions Model Methods References Histone methylation PRC2 (EED) (H3K27me3) Negative role in OPC differentiation and (re)myelination Chromatin silencing of differentiation genes (Bmp, Wnt) Olig1cre;Eedflox, Plpcre-Eedflox and PdgfracreRT;Eedflox mice RNA-Seq and ChIP-Seq on in vitro Olig1cre;Eed- flox OPCs Wang et al. 5. Epigenetic Marks Regulate the Oligodendroglial Progenitor Cell Differentiation into Mature Oligodendrocytes Life 2021, 11, 62 13 of 30 Life 2021, 11, 62 HDAC2 Negative role in OPC differentiation HDAC6 Positive role in OPC differentiation and OL morphology Deacetylation of alpha- tubulin In v siR HDAC11 Positive role in OPC differentiation In v siR SIRT2 Negative role in OPC differentiation Deacetylation of alpha- tubulin In v siR Histone citrullina- tion PADI2 Positive role in OPC differentiation Activation of cell differ- entiation genes (Septins, Mbp, Sox9/10, Tcf7l2) (+ role on non-histone tar- gets, e.g., myelin pro- teins) Pdg cre in v siR Oli sion Chromatin Organi- zation Enzyme/Mark Role Targeted Genes or Functions Chromatin remodel- ers CHD7 Positive role in OPC differentiation and (re)myelination Transcriptional activa- tion of differentiation genes (Sox10, Gpr17, Sirt2, Nkx2.2) Pdg cre Pdg cre Transcriptional activa- tion of differentiation genes (Myrf, Sox10) Oli mic CHD8 Positive role in OPC differentiation and (re)myelination Transcriptional activa- tion of differentiation genes (Tcf7l2, Myrf, Zfp488, Lnc-OL1) Oli and cre BRG1 Positive role in OPC differentiation and myelination Transcriptional activa- tion of differentiation genes (Myrf, Sox10) Oli mic Cn EP400 Positive role in OPC differentiation Transcriptional activa- tion of differentiation genes (Plp1, Mbp, Mog, Myrf, Sox10) Cn mic Nuclear lamina LMNB1 Negative role in OPC differentiation Repression of differen- tiation genes (choles- terol synthesis, Lss) In v LM sion Post-Transcrip- tional Modification Enzyme/Mark Role Targeted Genes or Functions Long non-coding RNA lnc-OL1 Positive role in OPC differentiation and (re)myelination Silencing of OPC pro- gram during their dif- ferentiation (interaction with SUZ12, within PRC2) In v OL and Oli mic Neat1 Positive role in OPC differentiation Over-representation of oligodendroglial path- ways among differen- tially expressed genes Ne lnc-158 Positive role in OPC differentiation In v 158 and MicroRNA DICER Positive role in OPC differentiation and myelination Oli and mic Positive role in mye- lination Oli mic miR-219 Positive role in OPC differentiation Repression of inhibitors of differentiation (Zfp238, Foxj3) In v 219 Repression of inhibitors of differentiation (Sox6, Hes5) In v 219 in v por and miR-338 Positive role in OPC differentiation miR-23 Positive role in OPC differentiation and myelination Upregulation of myelin genes (Cnp, Plp1, Mag, Mog) and downregula- tion of OPC and nu- clear lamina genes (Pdgfra, Lmnb1) In v exp cell miR-32 Positive role in OPC differentiation Promotes expression of myelin proteins and In v 32 and HDAC2 Negative role in OPC differentiation HDAC6 Positive role in OPC differentiation and OL morphology Deacetylation of alpha- tubulin In vitro OPCs + siRNA Noack et al. 5. Epigenetic Marks Regulate the Oligodendroglial Progenitor Cell Differentiation into Mature Oligodendrocytes [158] HDAC11 Positive role in OPC differentiation In vitro Olineu cells + siRNA Liu et al. [159] SIRT2 Negative role in OPC differentiation Deacetylation of alpha- tubulin In vitro Olineu cells + siRNA Li et al. [160] Histone citrullina- tion PADI2 Positive role in OPC differentiation Activation of cell differ- entiation genes (Septins, Mbp, Sox9/10, Tcf7l2) (+ role on non-histone tar- gets, e.g., myelin pro- teins) Pdgfra- creRT;Padi2flox mice, in vitro OPCs + siRNA, in vitro Olineu + overexpres- sion Proteomic and ATAC-Seq on in vitro OPCs + siRNA Falcão et al. [55] Chromatin Organi- zation Enzyme/Mark Role Targeted Genes or Functions Model Methods References Chromatin remodel- ers CHD7 Positive role in OPC differentiation and (re)myelination Transcriptional activa- tion of differentiation genes (Sox10, Gpr17, Sirt2, Nkx2.2) Pdgfra- creRT;Chd7flox mice RNA-Seq on in vitro Pdgfra- creRT;Chd7flox OPCs Marie et al. [61] Pdgfra- creRT;Chd7flox mice Doi et al. [138] Transcriptional activa- tion of differentiation genes (Myrf, Sox10) Olig1cre;Chd7flox mice RNA-Seq on Olig1cre;Chd7flox spinal cord He et al. [59] CHD8 Positive role in OPC differentiation and (re)myelination Transcriptional activa- tion of differentiation genes (Tcf7l2, Myrf, Zfp488, Lnc-OL1) Olig1cre;Chd8flox and Pdgfra- creRT;Chd8flox mice ChIP-Seq and ATAC-Seq on in vitro Olig1cre;Chd8flox OPCs Zhao et al. [139] BRG1 Positive role in OPC differentiation and myelination Transcriptional activa- tion of differentiation genes (Myrf, Sox10) Olig1cre;Brg1flox mice RNA-Seq and ChIP-Seq on Olig1cre;Brg1flox optic nerve Yu et al. [62] Cnpcre;Brg1flox mice Bischof et al. [57] EP400 Positive role in OPC differentiation Transcriptional activa- tion of differentiation genes (Plp1, Mbp, Mog, Myrf, Sox10) Cnpcre;Ep400flox mice ChIP-Seq on sorted Cnpcre;Ep400flox OPCs Elsesser et al. [140] Nuclear lamina LMNB1 Negative role in OPC differentiation Repression of differen- tiation genes (choles- terol synthesis, Lss) In vitro OPCs + LMNB1 overexpres- sion DamID on in vitro OPCs, combining with LMNB1 main- tained expression Yattah et al. [68] Post-Transcrip- tional Modification Enzyme/Mark Role Targeted Genes or Functions Model Methods References Long non-coding RNA lnc-OL1 Positive role in OPC differentiation and (re)myelination Silencing of OPC pro- gram during their dif- ferentiation (interaction with SUZ12, within PRC2) In vitro OPCs + lnc- OL1 overexpression and Ezh2 siRNA, Olig1cre;Ezh2flox mice He et al. [161] Neat1 Positive role in OPC differentiation Over-representation of oligodendroglial path- ways among differen- tially expressed genes Neat1−/− mice RNA-Seq of Neat1−/− mouse brains Katsel et al. 5. Epigenetic Marks Regulate the Oligodendroglial Progenitor Cell Differentiation into Mature Oligodendrocytes [92] H3K9me3 Positive role in OPC differentiation (no role for H3K27me3) In vitro OPCs and OLs ChIP-Seq on in vitro OPCs and Ols Liu et al. [93] KMT2/MLL (H3K4me3) Positive role in OPC differentiation Recruited by CHD8, mark accumulation at promoters of differenti- ation genes (Tcf7l2, Myrf, Zfp488, Lnc-OL1) Olig1cre;Chd8flox mice + KDM5 deme- thylase inhibitors ChIP-Seq on in vitro Olig1cre;Chd8flox OPCs Zhao et al. [139] PRMT5 Positive role in OPC differentiation and myelination Regulation of OPC dif- ferentiation genes Olig1cre;Prmt5flox mice and in vitro OPCs + siRNA RNA-Seq on in vitro PRMT5-CRIS- PRKO OPCs Scaglione et al. [135] Positive role in OPC differentiation and myelination Regulation of OPC dif- ferentiation genes Olig2cre;Prmt5flox mice RNA-Seq on Olig2cre;Prmt5flox brain Calabretta et al. [154] Positive role in OPC differentiation Repression of inhibitors of differentiation (Id2/4) In vitro OPCs and gli- oma cells + siRNA Huang et al. [155] Histone deacetyla- tion HDACs Positive role in OPC differentiation In vitro O4+ cells + HDAC inhibitors Conway et al. [137] In vitro OPCs + HDAC inhibitors Marin-Husstege et al. [143] Positive role in OPC differentiation and myelination In vivo HDAC inhibi- tors Shen et al. [156] Positive role in OPC differentiation and re- myelination Shen et al. [157] HDAC1/HDAC2 Positive role in OPC differentiation and myelination Olig1cre;Hdac1flox;H dac2flox mice Ye et al. [121] HDAC1 Positive role in OPC differentiation In vitro OPCs + siRNA Egawa et al. [132] Table 3. List of epigenetic marks and their roles during OPC differentiation. Table 3. List of epigenetic marks and their roles during OPC differentiation. 5.2. Chromatin Modifications Are Involved in OPC Differentiation 5. Epigenetic Marks Regulate the Oligodendroglial Progenitor Cell Differentiation into Mature Oligodendrocytes [111] lnc-158 Positive role in OPC differentiation In vitro NSCs + lnc- 158 overexpression and siRNA Li et al. [109] MicroRNA DICER Positive role in OPC differentiation and myelination Olig2cre;Dicerflox and Cnpcre;Dicerflox mice Dugas et al. [162] Positive role in mye- lination Olig1cre;Dicerflox mice Zhao et al. [141] miR-219 Positive role in OPC differentiation Repression of inhibitors of differentiation (Zfp238, Foxj3) In vitro OPCs + miR- 219 overexpression Dugas et al. [162] Repression of inhibitors of differentiation (Sox6, Hes5) In vitro OPCs + miR- 219 overexpression, in vivo electro- poration in chicks and zebrafish Zhao et al. [141] miR-338 Positive role in OPC differentiation miR-23 Positive role in OPC differentiation and myelination Upregulation of myelin genes (Cnp, Plp1, Mag, Mog) and downregula- tion of OPC and nu- clear lamina genes (Pdgfra, Lmnb1) In vivo miR-23 over- expression in CNP+ cells (mouse model) RNA-Seq Lin et al. [163] I it o OPC + iR Life 2021, 11, 62 14 of 30 14 of 30 glucose/lipi metabolism (MBP, SLC45A3) miR-27a Negative role in OPC differentiation and (re)myelination Targets mature OL-spe- cific genes (Mbp) In vitro OPCs + miR- 27a overexpression, ex vivo inhibition in cerebellum slices, in vivo intranasal inhibi- tion in mice RNA-Seq Tripathi et al. [165] miR-212 Negative role in OPC differentiation Inhibition of oligoden- droglial genes (Mbp, Olig1/2, Sox10) In vitro OPCs + miR- 212 overexpression and siRNA Validation by rtqPCR Wang et al. [166] miR-125-3p Negative role in OPC differentiation Inhibition of differenti- ation genes (Fyn, Smad4, Nrg1) In vitro OPCs + miR- 125-3p overexpres- sion and siRNA Validation by rtqPCR Lecca et al. [167] m6A RNA methyla- tion METTL14 Positive role in OPC differentiation Necessary for mature myelin gene expression (Mbp, Mog, Mag, Plp1, Cnp) and regulation of histone modification enzymes (HATs, HMTs, HDACs, KDMs) Olig2cre;Mettl14flox and Cnpcre;Mettl14flox mice RNA-Seq on in vitro Olig2cre;Mettl14flo x OPCs Xu et al. [168] FTO Positive role in OPC differentiation and myelination Promotes Olig2 degra- dation Fto-TG mice RNA-Seq on Nes- cre;Prrc2aflox brain and RIP-Seq on Olig2cre;Prrc2aflox brain Wu et al. [142] ATAC: assay for transposase-accessible chromatin, ChIP: chromatin immunoprecipitation, DamID: DNA adenine methyl- transferase identification, ERRBS: enhanced reduced representation bisulfite sequencing, hMeDIP: hydroxyMethylated DNA immunoPrecipitation, RIP: RNA immunoprecipitation, RRHP: reduced representation 5-hydroxymethylcytosine profiling, rtqPCR: reverse transcription quantitative polymerase chain reaction, Seq: sequencing. 5.1. Global DNA Demethylation Is Associated with OPC Differentiation For example, the chromatin remod- elers BRG1 and EP400 have been shown to be essential for OPC differentiation and mye- lination, at least during development when targeting early lineage [57,62,140]. However, they seem to be dispensable for later differentiation stages and during myelin mainte- nance [57,140]. Both BRG1 and EP400, in association with OLIG2, directly bind to differ- entiation genes, such as Myrf and Sox10, especially at enhancers and transcription start sites, which are characterized by the permissive histone marks H3K27ac and H3K4me3, respectively [57,59,62,139,140]. They also share similar chromatin occupancies with CHD7 and CHD8, which are essential for OPC differentiation, during developmental and adult (re)myelination [59,61,139]. In particular, the ablation of Chd8 in neonatal OPCs results in massive hypomyelination of the CNS, leading to seizures and eventually death of the mice at postnatal day 21 [139]. Interestingly, CHD8 can itself directly recruit KMT2/MLL, a his- tone lysine methyltransferase that is responsible for the addition of H3K4me3 and the subsequent activation of oligodendroglial genes (i.e., Olig1/Olig2, Sox10, Myrf) [139]. 5.2.2. Repressive Marks Regulate the Transition from OPC Proliferation to Differentia- tion 5.2.2. Repressive Marks Regulate the Transition from OPC Proliferation to Differentia- tion 5.2.2. Repressive Marks Regulate the Transition from OPC Proliferation to Differentia- OPC differentiation is also dependent on marks that are generally associated with gene repression, such as histone methylation, deacetylation, and citrullination. Indeed, the ablation or inhibition of Prmt5, Ezh2, Eed, Hdac1/2, or Padi2 in OPCs results in defective differentiation and myelination [55,92,93,121,132,135,137,143,155–157,159]. These marks mainly regulate the downregulation of OPC-specific inhibitors of differentiation (such as Id2/Id4) or cell cycle (such as Cdk4/6, Cxcl2/5/10/14) genes, and therefore, are often essential for OPC cell cycle exits and early OL differentiation, but less involved in myelin mainte- nance [55,92,93,155]. These enzymes, such as HDACs, PRMT5, and PADI2, can also mod- ify non-histone targets, altering the function or localization of oligodendroglial proteins or transcription factors (i.e., OLIG1, alpha-tubulin, PDGFR, or myelin proteins) [55,154,158,160,170]. Because of steric constraints, histone modifications can also be de- pendent on or exclusive of each other. For example, the ablation of Prmt5 in OPC results in decreased symmetric H4R3me2s, allowing for H4K5 acetylation and preventing differ- entiation and myelination. The addition of histone acetylation inhibitors in vitro is suffi- cient to rescue OPC differentiation, even without PRMT5 [135]. 5.1. Global DNA Demethylation Is Associated with OPC Differentiation 5.1. Global DNA Demethylation Is Associated with OPC Differentiation Early evidence showed that OPC differentiation is associated with demethylation of a specific myelin gene, Mag, during rat development [169]. A more global demethylation during OPC differentiation that is associated with permissive gene expression was further confirmed by DNA methylation whole-genome sequencing. During development, OPC differentiation into OL is correlated with the decreased DNA methylation and increased DNA hydroxymethylation of genes involved in lipid synthesis and myelin formation [87,89]. However, because of its role in OPC proliferation and survival, the ablation of DNA methyltransferases in neonatal OPCs does not induce early differentiation, but in contrast, results in global hypomyelination of the CNS [87,132]. Indeed, the ablation of DNA methyltransferases in developmental and adult post-mitotic OPCs has a slight effect on myelination and remyelination [87,133]. Recent studies also highlight an age-depend- ent role of demethylation on OPC differentiation, suggesting a more important role of DNA hydroxymethylation in adult oligodendroglial cells [89,134]. In particular, TET1- mediated DNA hydroxymethylation targets genes involved in the late stages of OPC dif- ferentiation, such as biosynthesis and neuroglial communication [89,119,134]. The abla- tion of Tet1 in adult OPCs is sufficient to reduce DNA hydroxymethylation at these spe- cific genomic sites, downregulating their expression and blocking OL late differentiation [134]. In old mice, lower TET1 expression and decreased DNA hydroxymethylation levels could be directly associated with the delayed remyelination observed in aging [134]. While the downregulation of TET enzymes in vitro has been shown to affect neonatal OPC differentiation, this effect is still being examined in developmental studies in vivo [78,89,134]. Overall, these studies suggest a dual role for DNA modifications in oligoden- droglial cells, balancing both the methylation and demethylation of specific genomic re- gions at different ages and different stages of the differentiation process. 5.2. Chromatin Modifications Are Involved in OPC Differentiation 5.2. Chromatin Modifications Are Involved in OPC Differentiation 5.2.1. Chromatin Reorganization Allows for the Accessibility of Differentiation Gene in OPCs At the chromatin level, OPC differentiation is mainly associated with open confor- mation at discrete loci, which is reflected by chromatin remodeling, histone modifications, Life 2021, 11, 62 15 of 30 15 of 30 and nuclear lamin reorganization, allowing for access of transcription factors to specific genes that are characteristic of the differentiated state. 5.1. Global DNA Demethylation Is Associated with OPC Differentiation Eventually, repressive marks are also associated with the nuclear lamina, which maintains mainly repressive heterochromatin at the nuclear periphery. Recently, LMNB1 has been associated with oligodendroglial maturation genes (i.e., myelin genes and cho- lesterol synthesis pathways), which appeared to be sufficient to block OPC differentiation in vitro [68,171]. 6. Epigenetic Marks Modulate Myelination and Myelin Remodeling Finally, in addition to timely OPC differentiation into mature OLs, recent studies have highlighted the importance of correct myelination for neuronal connectivity. This includes myelin ensheathment, compaction around the axon, and internode sizes. This is not limited to developmental myelination as myelin is constantly remodeled in adult CNS, especially during learning and when facing depressive or social isolation experi- ences [24,25,176–181] (Table 4). Table 4. List of epigenetic marks and their roles during OPC myelination. DNA Modification Enzyme/Mark Role Targeted Genes or Func- tions Model Methods References DNA hydroxymethyla- tion TET1 Positive role in myelination and neuroglial com- munication Activation of cell differenti- ation and cell communica- tion genes (Ca2+ homeosta- sis) Olig1cre;Tet1 flox mice hMeDIP-Seq on in vitro ne- onatal NSCs and OPCs, RNA-Seq on in vitro Olig1cre;Tet1flox OPCs Zhang et al. [89] Activation of cell differenti- ation, myelination, biosyn- thesis, and neuroglial com- munication genes (Slc fam- ily members, Pcdh family members) Olig1cre;Tet1 flox mice RNA-Seq and RRHP on sorted adult OPCs and OLs, RNA-Seq on oligo-enriched Olig1cre;Tet1flox lesioned spinal cord Moyon et al. [134] Chromatin Organiza- tion Enzyme/Mark Role Targeted Genes or Func- tions Model Methods References Chromatin remodelers CHD7 Positive role in myelination Preferentially binding and increased accessibility of myelinogenesis (Mbp, Plp1 and Cnp) and lipid metabo- lism genes (Enpp2, Nfya, and Elovl7) Olig1cre;Chd 7flox mice RNA-Seq on in vitro Olig1cre;Chd7flox spinal cord He et al. [59] Post-Transcriptional Modification Enzyme/Mark Role Targeted Genes or Func- tions Model Methods References MicroRNA DICER Positive role in myelin mainte- nance Induces expression of mye- lin proteins PlpcreRT;Dic erflox mice Shin et al. [173] m6A RNA methylation METTL14 Positive role in myelination and regulation of in- ternodes lenghts Alternative splicing of glial paranodal genes (Nfasc155) Olig2cre;Mett l14flox mice RNA-Seq on in vitro Olig2cre;Mettl14flox OPCs and OLs Xu et al. [168] hMeDIP: hydroxymethylated DNA immunoprecipitation, RRHP: reduced representation 5-hydroxymethylcytosine pro- filing, Seq: sequencing. Table 4. List of epigenetic marks and their roles during OPC myelination. hMeDIP: hydroxymethylated DNA immunoprecipitation, RRHP: reduced representation 5-hydroxymethylcytosine pro- filing Seq: sequencing hMeDIP: hydroxymethylated DNA immunoprecipitation, RRHP: reduced representation 5-hydroxymethylcytosine pro- filing, Seq: sequencing. A few epigenetic marks have been associated with late myelination processes. Com- pared to other chromatin remodelers, CHD7 binds preferentially to myelinogenesis (i.e., Mbp, Plp1, Cnp) and lipid metabolism genes (i.e., Enpp2, Nfya, Elovl7), which would sug- gest involvement in late OPC differentiation and myelination [59]. 5.3. Post-Transcriptional Modifications That Are Essential for OPC Differentiation In addition to chromatin condensation and conformation, the OPC differentiation program also depends on post-transcriptional modifications. Recent models using the ab- lation of Dicer (required for the generation of functional microRNA) or the de novo anal- ysis of oligodendroglial transcriptomic datasets identified microRNAs and lncRNAs that are essential for OPC differentiation. For example, miR-219, miR-338, miR-23, and miR-32, as well as lnc-OL1, lnc-158, and Neat1, promote differentiation, while miR-27a, miR-212, and miR-125-3p inhibit differentiation [109,111,141,161–167,171–174]. Interestingly, these post-transcriptional signals directly regulate some chromatin-modifying genes, suggest- ing feedback from miRNA and lncRNA regarding chromatin conformation. For example, miR-23 has been shown to suppress LMNB1 expression, thus rescuing OPC differentiation in vitro [171]. Similarly, lnc-OL1 can directly interact with SUZ12, a part of the histone methylation complex PRC2, with both being required for OPC differentiation during de- velopment and repair [161,175]. Life 2021, 11, 62 16 of 30 16 of 30 Recently, studies have identified the essential role of RNA methylation on OPC dif- ferentiation. The ablation of the methyltransferase Mettl14 (“writer”), the demethylase Fto (“eraser”), or the m6A “reader” Prrc2a in the oligodendroglial cell lineage all resulted in a lower number of mature OLs and global hypomyelination of the CNS [142,168]. In addi- tion to the downregulation of several myelin genes (i.e., Mbp, Mog, Mag), many histone post-translational modification readers and writers (i.e., HMTs, HDACs) are also dysreg- ulated when METTL14 is lacking, suggesting here again the potential feedback from m6A- modified mRNA to histone modifications [168]. 6. Epigenetic Marks Modulate Myelination and Myelin Remodeling The genetic ablation of Tet1 in OPC induces defective remyelination, which is characterized by swellings in adult CNS after injury. TET1-mediated hydroxymethylation tends to target genes related to late myelination, as well as neuroglia communication genes involved in ion exchange and the maintenance of a tight space between the axon body and the myelin membrane (i.e., Slc12a2) [89,134]. METTL14 and m6A RNA affects the alternative splicing of some par- anodal genes, such as glial neurofascin 155 [168]. In Mettl14 mutants, Xu et al. noticed Life 2021, 11, 62 17 of 30 17 of 30 increased nodal/paranodal spaces, which were associated with lower numbers of nodes, suggesting a role for mRNA methylation on internode length [168]. increased nodal/paranodal spaces, which were associated with lower numbers of nodes, suggesting a role for mRNA methylation on internode length [168]. Because the myelination and myelin remodeling processes have not been analyzed in detail yet, additional epigenetic modifications are to be identified in this later process to finely tune the ensheathment, internodal length, and myelin compaction. 7.1. Chemical and Physical Cues The OPC-to-OL switch is modulated by pro-proliferation or pro-differentiation chemical cues. For example, growth factors (e.g., PDGF and FGF) can induce cMyc ex- pression in OPC, which can directly recruit histone acetyltransferases, upregulating genes involved in OPC proliferation [136,144–147]. Inversely, the addition of T3, known to trig- ger OPC differentiation in vitro, has been shown to induce Brg1 expression, which would favor the accessibility of differentiating and myelinating genes [62]. Physical constraints in vitro and tissue stiffness in vivo have also been shown to influence OPC proliferation and differentiation [184–186]. Hernandez and colleagues showed that, in vitro, these me- chanical stimuli are transduced via cytoskeleton and nucleoskeleton complexes into nu- clear cues, resulting in histone methylation changes [185]. 7. Epigenetic Changes Translate Environmental Cues into Intrinsic Signals in Oli- godendroglial Cells OPC differentiation is an intrinsic program that is capable of myelinating dead axons or nanofibers in vitro [182–184]. Meanwhile, OPCs and OLs are also highly sensitive to their environment, which can modulate their physiology, in part by epigenetic mecha- nisms. Perhaps most interestingly, OPC proliferation and differentiation have been shown to be partially regulated by neuronal signals, including neurotransmitters and axonal ac- tivity, some of which could translate into epigenetic changes in oligodendroglial cells. 7.2.1. Oligodendrocytes Respond to Neuronal Activity Neuronal activity plays a central role in brain function by shaping developmental circuits and providing effective transmission of the information necessary for behavioral and cognitive functions. OLs are key contributors to this process as they ensheath axons with their processes, forming the multilamellar membrane, myelin, which allows for en- ergy-efficient saltatory conduction [187]. Several lines of evidence now demonstrate a re- ciprocal relationship between neurons and OLs, with OLs providing metabolic support to axons, which is crucial to their survival [3,188,189], and neurons being able to regulate OPC proliferation and differentiation [14,22,23,128,131,190–192], as well as myelin for- mation, repair, and remodeling [149,193–195]. A variety of in vitro and in vivo ap- proaches, combined with experimental modulation of neuronal activity, have been used to characterize this relationship, such as the use of myelinating co-cultures [129,196–198], electrophysiology [199], enriched or deprived environments [149,177,178,200,201], motor and learning tasks [23,25,202,203], electrical stimulation [131], pharmacogenetic and chemogenetic tools [191,204], and optogenetics [22,194,205]. While neuronal activity can undeniably control OL biology, myelination can yet occur in the absence of such signals [206,207]. Interestingly, the pattern of neuronal activity is also thought to differentially impact OL and myelination [14,208]. Over the years, various signaling mechanisms have been proposed as the means for OLs to respond to neurons, with some requiring direct contact and others thought to occur via intermediate messengers to OL membrane receptors and ion channels. The identified pathways involve contact-mediated factors, such as N-cadherin [209,210]; trophic and growth factors, such as Neuroligin-3 and BDNF [211]; calcium signaling [212]; vesicular and non-vesicular release of neurotransmitters [196,213–216]. Life 2021, 11, 62 18 of 30 18 of 30 Single-cell studies have begun to identify transcriptional changes induced in the CNS in response to neuronal activity, including in OLs. In the mouse visual cortex, dynamic transcriptional changes, including in OL gene networks, were revealed after exposure to light for 4 h following 7 days in the dark. Among the genes that were differentially ex- pressed by the presence of neuronal activity were several transcription factors (Egr1, Pou3f1, and Erf) that were specifically induced in OPCs. In addition, multiple OL popula- tions displayed induction of the serum/glucocorticoid-regulated kinase, Sgk1, which was shown to be involved in oligodendrocyte process arborization [217]. Similarly, transcrip- tional changes in OPCs were also identified in the mouse cortex following seizure induc- tion [218]. 7.2.1. Oligodendrocytes Respond to Neuronal Activity Therefore, neuronal activity appears to be a critical regulator of OPC proliferation, differentiation into OLs, and myelin formation through various messengers, receptors, and pathways. However, the extent to which neuronal activity directly influences epige- netic responses in OL remains understudied. 7.2.2. Activity-Dependent Epigenetic Modifications in the CNS Neurological disorders and behaviors are closely related to epigenetic changes. Ac- tivity-dependent regulation of the genome occurs in response to different environmental factors, such as learning, memory, and visual experience, and the epigenome is known to be altered in several neurological disorders, such as schizophrenia, Alzheimer’s disease, and multiple sclerosis [219–223]. While certain intracellular pathways and transcriptional changes have been correlated with neuronal activity, the mechanisms that link neuronal activation and these changes remain to be determined. Interestingly, it has been suggested that neuronal activity could modify the epigenetic landscape of neural cells, including oligodendroglial cells, as described in two studies. Further characterization of how these epigenetic changes may occur and impact oligodendroglial gene expression will need to be addressed by future experiments (Figure 3). Figure 3. Neuronal activity modifies the epigenetic landscape of neural cells. Social isolation per- turbs neuronal activity and reduces myelination. In this model, fewer heterochromatic nuclei are detected, along with increased histone acetylation and decreased histone methylation in OLs [177]; The stimulation of neuronal activity enhances myelination and is associated with a repressive chro- matin state. OPC nuclei show decreased histone acetylation and increased histone methylation lev- els [22]; neuronal activity also modifies DNA methylation and histone acetylation and methylation marks in neurons. Figure 3. Neuronal activity modifies the epigenetic landscape of neural cells. Social isolation per- turbs neuronal activity and reduces myelination. In this model, fewer heterochromatic nuclei are detected, along with increased histone acetylation and decreased histone methylation in OLs [177]; The stimulation of neuronal activity enhances myelination and is associated with a repressive chro- matin state. OPC nuclei show decreased histone acetylation and increased histone methylation lev- els [22]; neuronal activity also modifies DNA methylation and histone acetylation and methylation marks in neurons. Life 2021, 11, 62 19 of 30 19 of 30 7.2.3. Activity-Dependent Epigenetic Modifications in Oligodendrocytes 7.2.3. Activity-Dependent Epigenetic Modifications in Oligodendrocytes In 2014, Gibson et al. investigated how neuronal activity may regulate myelin-form- ing cells in vivo [22]. In this study, neuronal activity was induced by optogenetic stimula- tions in the premotor cortex of Thy1:channelrhodopsin 2 (ChR2) mice, in which ChR2- expressing neurons are activated by light [224] and elicited complex motor behavior in mice. The increased proliferation of OPCs was detected as early as 3 h after stimulation and resulted in a higher number of mature OLs and increased myelination 4 weeks after a 7-day stimulation paradigm. 7.2.2. Activity-Dependent Epigenetic Modifications in the CNS Notably, a concomitant increase in H3K9me3 and decrease in H3 acetylation was detected in dividing OPCs between 3 and 24 h after stimulation, suggesting that neuronal activity could enhance differentiation by inducing a repressive chromatin state that is necessary for OPC differentiation. Furthermore, administration of the histone deacetylase inhibitor TSA during the week of stimulation prevented mature oligodendrogenesis [22]. Chromatin in OLs can also be modified in response to social iso- lation, which has been associated with altered neuronal activity [225]. Indeed, in a mouse model of social isolation showing impaired myelination of the prefrontal cortex, immature chromatin and a lower level of heterochromatin formation was detected in the nucleus of an OL in the vicinity of axons with thinner myelin [177]. In agreement with Gibson’s study, the opposite changes in epigenetic marks were associated with chronic social iso- lation, namely, a decrease in repressive histone methylation and an increase in H3 acety- lation. In parallel, changes in the level of transcripts for enzymes regulating these marks were detected, with significantly lower levels of Hdac1 and Hdac2 mRNAs [177]. Therefore, these two studies advocate for the role of neuronal activity regulation of the epigenome in OLs. One should consider that OPCs could differentially respond to neuronal activity, depending on their location, cell state, or age. For instance, region-spe- cific changes in myelination have been detected in response to chronic stress [226]. While myelination deficits are observed in the nucleus accumbens of both susceptible and resil- ient mice, only the motor prefrontal cortex of susceptible mice show similar deficits, high- lighting the distinct responses of oligodendroglial cells to chronic stress, depending on their spatial location. Moreover, recent single-cell studies have highlighted heterogeneity among the OL lineage in the healthy or diseased CNS, according to their location, but also their age. For example, the enrichment of oligodendroglial cells at mature stages, to the detriment of progenitor and immature stages, is detected as the brain ages [15,227–229]. This could be explained by different intrinsic properties in each oligodendroglial popula- tion, but also by the accumulation of distinct external and neuronal inputs throughout time. Of note, in zebrafish, functionally diverse subgroups of OPCs were identified in the spinal cord and were shown to differentially respond to neuronal activity [228]. 7.2.4. Activity-Dependent Epigenetic Modifications in Other Cell Types 7.2.4. Activity-Dependent Epigenetic Modifications in Other Cell Types Epigenetic modifications in response to neuronal activity have been identified in other cell types of the CNS, particularly in neurons. For instance, histone and DNA methylome modifications have both been detected in vivo following long-term potentia- tion in rats [230,231]. Neuronal activity can modulate the subcellular localization of his- tone deacetylases in neurons in vitro, following calcium influx through calcium channels and NMDA receptors, which could result in histone mark changes [232]. Indeed, mem- brane depolarization, which induced neural precursor cell differentiation, is associated with changes in histone methylation and acetylation marks [233]. In parallel, DNA de- methylation in neurons in vitro has also been detected shortly after membrane depolari- zation and was associated with decreased H3K9me2 and increased histone acetylation in the regulatory region for brain-derived neurotrophic factor (BDNF), suggesting that DNA methylation changes may be associated with methylation-induced chromatin remodeling, which induces adaptive neuronal responses [234]. Furthermore, lower mRNA levels of DNMT1 and DNMT3a and higher levels of BDNF were detected in cortical neurons after depolarization by KCl and a sodium channel agonist [235]. DNA methylation changes Life 2021, 11, 62 20 of 30 20 of 30 have also been observed in epilepsy, which are characterized by abnormal patterns of neuronal activity, but also following electroconvulsive stimulation in mice [236,237]. Among DNMTs, only DNMT3a was higher after stimulation, making it a candidate for activity-dependent de novo methylation [237]. However, methylation changes observed in neurons might not reflect similar changes in glial cells. One study comparing hippocampal neuronal NeuN+ and glial NeuN− sorted cells following kainic acid injection, which elicits epilepsy, showed that there was almost no overlap in differentially methylated CpGs between neurons and non- neuronal cells [236]. Further investigation will be necessary to determine whether neurons and glial cells share common activity-dependent epigenetic responses. 8. Conclusions As put forward by this review and others [83,236–238], the epigenome is critical in the regulation of the OL lineage, including DNA methylation, histone modifications, chro- matin remodeling, lncRNA, and miRNAs. Further studies analyzing how distinct epige- netic layers might contribute together to produce gene regulation would be essential to more precisely characterize the role of epigenetics on the oligodendroglial cell lineage. Importantly, these mechanisms allow oligodendroglial cells to respond to their environ- ment and, therefore, facilitate the appropriate response to maintain CNS function. Neu- ronal activity appears as an important modulator of OL biology, and we highlight here the few studies identifying activity-dependent epigenetic changes that can further regu- late myelination. A deeper understanding of the link between neuronal activity and the epigenetic regulation of OLs would serve not only as a valuable resource to better under- stand neuroglia interaction but would also provide insights that could help in the devel- opment of therapies for neurodegenerative diseases, such as multiple sclerosis or Alz- heimer’s disease, where the dysregulation of neuronal activity and myelin are often closely related. Author Contributions: M.P. and S.M. both designed the figures and wrote and edited the paper. Data used for this review have been previously peer-reviewed, except for pre-prints, and are refer- enced at the end of the manuscript. Both authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the National Multiple Sclerosis Society (FG-1907-34565, M.P.). Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data sharing not applicable. Data Availability Statement: Data sharing not applicable. Acknowledgments: We thank the Casaccia lab members for their comments on the manuscript. We apologize to colleagues whose work and papers we could not mention. Acknowledgments: We thank the Casaccia lab members for their comments on the manuscript. We apologize to colleagues whose work and papers we could not mention. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. 5. Kessaris, N.; Fogarty, M.; Iannarelli, P.; Grist, M.; Wegner, M.; Richardson, W.D. Competing Waves of Oligodendrocytes in the Forebrain and Postnatal Elimination of an Embryonic Lineage. Nat. Neurosci. 2006, 9, 173–179, doi:10.1038/nn1620. j 4. Davies, J.E.; Miller, R.H. Local Sonic Hedgehog Signaling Regulates Oligodendrocyte Precursor Appearance in Multiple Ven- tricular Zone Domains in the Chick Metencephalon. Dev. Biol. 2001, 233, 513–525, doi:10.1006/dbio.2001.0224. 3. Saab, A.S.; Tzvetavona, I.D.; Trevisiol, A.; Baltan, S.; Dibaj, P.; Kusch, K.; Möbius, W.; Goetze, B.; Jahn, H.M.; Huang, W.; et al. Oligodendroglial NMDA Receptors Regulate Glucose Import and Axonal Energy Metabolism. Neuron 2016, 91, 119–132, doi:10.1016/j.neuron.2016.05.016. 1. Valério-Gomes, B.; Guimarães, D.M.; Szczupak, D.; Lent, R. The Absolute Number of Oligodendrocytes in the Adult Mouse Brain. Front. Neuroanat. 2018, 12, doi:10.3389/fnana.2018.00090. 2. Pelvig, D.P.; Pakkenberg, H.; Stark, A.K.; Pakkenberg, B. Neocortical Glial Cell Numbers in Human 2008, 29, 1754–1762, doi:10.1016/j.neurobiolaging.2007.04.013. References 1. Valério-Gomes, B.; Guimarães, D.M.; Szczupak, D.; Lent, R. The Absolute Number of Oligodendrocytes in the Adult Mouse Brain. Front. Neuroanat. 2018, 12, doi:10.3389/fnana.2018.00090. 1. Valério-Gomes, B.; Guimarães, D.M.; Szczupak, D.; Lent, R. The Absolute Number of Oligodendrocytes in the Adult Mouse Brain. Front. Neuroanat. 2018, 12, doi:10.3389/fnana.2018.00090. Life 2021, 11, 62 21 of 30 21 of 30 6. Pringle, N.P.; Yu, W.P.; Howell, M.; Colvin, J.S.; Ornitz, D.M.; Richardson, W.D. Fgfr3 Expression by Astrocytes and Their Precursors: Evidence That Astrocytes and Oligodendrocytes Originate in Distinct Neuroepithelial Domains. Development 2003, 130, 93–102, doi:10.1242/dev.00184. 7. Cameron-Curry, P.; Le Douarin, N.M. Oligodendrocyte Precursors Originate from Both the Dorsal an Spinal Cord. Neuron 1995, 15, 1299–1310, doi:10.1016/0896-6273(95)90009-8. 8. Fogarty, M.; Richardson, W.D.; Kessaris, N. A Subset of Oligodendrocytes Generated from Radial Glia in the Dorsal Spinal Cord. Development 2005, 132, 1951–1959, doi:10.1242/dev.01777. p 9. Pringle, N.P.; Richardson, W.D. A Singularity of PDGF Alpha-Receptor Expression in the Dorsoventral Axis of the Neural Tube May Define the Origin of the Oligodendrocyte Lineage. Development 1993, 117, 525–533. 10. Barres, B.A.; Lazar, M.A.; Raff, M.C. A Novel Role for Thyroid Hormone, Glucocorticoids and Retinoic Acid in Timing Oli- godendrocyte Development. Development 1994, 120, 1097–1108. 11. Nishiyama, A.; Chang, A.; Trapp, B.D. NG2+ Glial Cells: A Novel Glial Cell Population in the Adult Brain. J. Neuropathol. Exp. Neurol. 1999, 58, 1113–1124, doi:10.1097/00005072-199911000-00001. 12. Pringle, N.P.; Mudhar, H.S.; Collarini, E.J.; Richardson, W.D. PDGF Receptors in the Rat CNS: During Late Neurogenesis, PDGF Alpha-Receptor Expression Appears to Be Restricted to Glial Cells of the Oligodendrocyte Lineage. Development 1992, 115, 535– 551. 13. Raff, M.C.; Abney, E.R.; Cohen, J.; Lindsay, R.; Noble, M. Two Types of Astrocytes in Cultures of Developing Rat White Matter: Differences in Morphology, Surface Gangliosides, and Growth Characteristics. J. Neurosci. 1983, 3, 1289–1300, doi:10.1523/JNEU- ROSCI.03-06-01289.1983. 14. Nagy, B.; Hovhannisyan, A.; Barzan, R.; Chen, T.-J.; Kukley, M. Different Patterns of Neuronal Activity Trigger Distinct Re- sponses of Oligodendrocyte Precursor Cells in the Corpus Callosum. PLoS Biol. 2017, 15, e2001993, doi:10.1371/jour- nal.pbio.2001993. p 15. Spitzer, S.O.; Sitnikov, S.; Kamen, Y.; Evans, K.A.; Kronenberg-Versteeg, D.; Dietmann, S.; de Faria, O.; Agathou, S.; Káradóttir, R.T. Oligodendrocyte Progenitor Cells Become Regionally Diverse and Heterogeneous with Age. Neuron 2019, doi:10.1016/j.neuron.2018.12.020. j 16. Zawadzka, M.; Rivers, L.E.; Fancy, S.P.; Zhao, C.; Tripathi, R.; Jamen, F.; Young, K.; Goncharevich, A.; Pohl, H.; Rizzi, M.; et al. References CNS-Resident Glial Progenitor/Stem Cells Produce Schwann Cells as Well as Oligodendrocytes during Repair of CNS Demye- lination. Cell Stem Cell 2010, 6, 578–590, doi:10.1016/j.stem.2010.04.002. , , , j 17. Emery, B.; Agalliu, D.; Cahoy, J.D.; Watkins, T.A.; Dugas, J.C.; Mulinyawe, S.B.; Ibrahim, A.; Ligon, K.L.; Rowitch, D.H.; Barres, B.A. Myelin Gene Regulatory Factor Is a Critical Transcriptional Regulator Required for CNS Myelination. Cell 2009, 138, 172– 185, doi:10.1016/j.cell.2009.04.031. , j 18. He, Y.; Dupree, J.; Wang, J.; Sandoval, J.; Li, J.; Liu, H.; Shi, Y.; Nave, K.A.; Casaccia-Bonnefil, P. The Transcription Factor Yin Yang 1 Is Essential for Oligodendrocyte Progenitor Differentiation. Neuron 2007, 55, 217–230, doi:10.1016/j.neuron.2007.06.029. 19. Küspert, M.; Wegner, M. SomethiNG 2 Talk about—Transcriptional Regulation in Embryonic and Adult Oligodendrocyte Pre- cursors. Brain Res. 2016, 1638, 167–182, doi:10.1016/j.brainres.2015.07.024. 20. Lu, Q.R.; Sun, T.; Zhu, Z.; Ma, N.; Garcia, M.; Stiles, C.D.; Rowitch, D.H. Common Developmental Requirement for Olig Func- tion Indicates a Motor Neuron/Oligodendrocyte Connection. Cell 2002, 109, 75–86, doi:10.1016/s0092-8674(02)00678-5. 21 Nakatani H ; Martin E ; Hassani H ; Clavairoly A ; Maire C L ; Viadieu A ; Kerninon C ; Delmasure A ; Frah M ; Weber M ; 21. Nakatani, H.; Martin, E.; Hassani, H.; Clavairoly, A.; Maire, C.L.; Viadieu, A.; Kerninon, C.; Delmasure, A.; Frah, M.; Weber, M.; et al. Ascl1/Mash1 Promotes Brain Oligodendrogenesis during Myelination and Remyelination. J. Neurosci. 2013, 33, 9752–9768, doi:10.1523/JNEUROSCI.0805-13.2013. 22. Gibson, E.M.; Purger, D.; Mount, C.W.; Goldstein, A.K.; Lin, G.L.; Wood, L.S.; Inema, I.; Miller, S.E.; Bieri, G.; Zuchero, J.B.; et al. Neuronal Activity Promotes Oligodendrogenesis and Adaptive Myelination in the Mammalian Brain. Science 2014, 344, 1252304, doi:10.1126/science.1252304. 23. Bacmeister, C.M.; Barr, H.J.; McClain, C.R.; Thornton, M.A.; Nettles, D.; Welle, C.G.; Hughes, E.G. Motor Learning Promotes Remyelination via New and Surviving Oligodendrocytes. Nat. Neurosci. 2020, 23, 819–831, doi:10.1038/s41593-020-0637-3. y g g y 24. Bengtsson, S.L.; Nagy, Z.; Skare, S.; Forsman, L.; Forssberg, H.; Ullén, F. Extensive Piano Practicing Has Regionally Specific Effects on White Matter Development. Nat. Neurosci. 2005, 8, 1148–1150, doi:10.1038/nn1516. 25. McKenzie, I.A.; Ohayon, D.; Li, H.; de Faria, J.P.; Emery, B.; Tohyama, K.; Richardson, W.D. Motor Active Central Myelination. Science 2014, 346, 318–322, doi:10.1126/science.1254960. 26. Duncan, I.D.; Radcliff, A. Remyelination Therapy for Demyelinating Disease. Nat. Rev. Neurol. 2020, doi:10.1038/s41582-020- 0341-7. 27. Houston, I.; Peter, C.J.; Mitchell, A.; Straubhaar, J.; Rogaev, E.; Akbarian, S. Epigenetics in the Human Brain. Neuropsychophar- macology 2013, 38, 183–197, doi:10.1038/npp.2012.78. References 2012, 26, 2749–2762, doi:10.1101/gad.203356.112. g 40. Fulton, M.D.; Zhang, J.; He, M.; Ho, M.-C.; Zheng, Y.G. The Intricate Effects of Alpha-Amino and Lysine Modifications on Arginine Methylation on the N-Terminal Tail of Histone H4. Biochemistry 2017, 56, 3539–3548, doi:10.1021/acs.biochem.7b00450. g 40. Fulton, M.D.; Zhang, J.; He, M.; Ho, M.-C.; Zheng, Y.G. The Intricate Effects of Alpha-Amino and Lysine Modifications on Arginine Methylation on the N-Terminal Tail of Histone H4. Biochemistry 2017, 56, 3539–3548, doi:10.1021/acs.biochem.7b00450. 41. Gu, B.; Lee, M.G. Histone H3 Lysine 4 Methyltransferases and Demethylases in Self-Renewal Anddifferentiation of Stem Cells. g y y 41. Gu, B.; Lee, M.G. Histone H3 Lysine 4 Methyltransferases and Demethylases in Self-Renewal Anddifferentiation of Stem Cells. Cell Biosci. 2013, 3, 39, doi:10.1186/2045-3701-3-39. 42. Guccione, E.; Bassi, C.; Casadio, F.; Martinato, F.; Cesaroni, M.; Schuchlautz, H.; Lüscher, B.; Amati, B. Methylation of Histone H3R2 by PRMT6 and H3K4 by an MLL Complex Are Mutually Exclusive. Nature 2007, 449, 933–937, doi:10.1038/nature06166. 43. Hyllus, D.; Stein, C.; Schnabel, K.; Schiltz, E.; Imhof, A.; Dou, Y.; Hsieh, J.; Bauer, U.-M. PRMT6-Media Histone H3 Antagonizes H3 K4 Trimethylation. Genes Dev. 2007, 21, 3369–3380, doi:10.1101/gad.44700 44. Margueron, R.; Reinberg, D. The Polycomb Complex PRC2 and Its Mark in Life. Nature 2011, 469, 343–349, doi:10.1038/na- ture09784. 45. Bernstein, B.E.; Kamal, M.; Lindblad-Toh, K.; Bekiranov, S.; Bailey, D.K.; Huebert, D.J.; McMahon, S.; Karlsson, E.K.; Kulbokas, E.J.; Gingeras, T.R.; et al. Genomic Maps and Comparative Analysis of Histone Modifications in Human and Mouse. Cell 2005, 120, 169–181, doi:10.1016/j.cell.2005.01.001. j 46. Zhang, P.; Torres, K.; Liu, X.; Liu, C.; Pollock, R.E. An Overview of Chromatin-Regulating Proteins in Cells. Curr. Protein Pept. Sci. 2016, 17, 401–410, doi:10.2174/1389203717666160122120310. 47. Creyghton, M.P.; Cheng, A.W.; Welstead, G.G.; Kooistra, T.; Carey, B.W.; Steine, E.J.; Hanna, J.; Lodato, M.A.; Frampton, G.M.; Sharp, P.A.; et al. Histone H3K27ac Separates Active from Poised Enhancers and Predicts Developmental State. Proc. Natl. Acad. Sci. USA 2010, 107, 21931–21936, doi:10.1073/pnas.1016071107. , , , /p 48. Jenuwein, T.; Allis, C.D. Translating the Histone Code. Science 2001, 293, 1074–1080, doi:10.1126/science.1063127. p n, T.; Allis, C.D. Translating the Histone Code. Science 2001, 293, 1074–1080, doi:10.1126/science.1063127. 48. Jenuwein, T.; Allis, C.D. Translating the Histone Code. Science 2001, 293, 1074–1080, doi:10.1126/sc 49. Lorenzo, A.D.; Bedford, M.T. Histone Arginine Methylation. FEBS Lett. 2011, 585, 2024–2031, doi:1 49. Lorenzo, A.D.; Bedford, M.T. Histone Arginine Methylation. FEBS Lett. 2011, 585, 2024–2031, doi:10.1016/j.febslet.2010.11.010. 50. References gy pp 28. Li, J.; Yang, X.; Qi, Z.; Sang, Y.; Liu, Y.; Xu, B.; Liu, W.; Xu, Z.; Deng, Y. The Role of MRNA M6A Methylation in the Nervous System. Cell Biosci. 2019, 9, doi:10.1186/s13578-019-0330-y. 29. Eden, S.; Cedar, H. Role of DNA Methylation in the Regulation of Transcription. Curr. Opin. Genet. Dev. 1994, 4, 255–259, doi:10.1016/S0959-437X(05)80052-8. 30. Ono, T.; Uehara, Y.; Kurishita, A.; Tawa, R.; Sakurai, H. Biological Significance of DNA Methylation in the Ageing Process. Age Ageing 1993, 22, S34–43, doi:10.1093/ageing/22.suppl_1.s34. Life 2021, 11, 62 22 of 30 31. Tawa, R.; Ono, T.; Kurishita, A.; Okada, S.; Hirose, S. Changes of DNA Methylation Level during Pre- and Postnatal Periods in Mice. Differentiation 1990, 45, 44–48, doi:10.1111/j.1432-0436.1990.tb00455.x. ff , , , j 32. Schübeler, D. Function and Information Content of DNA Methylation. Nature 2015, 517, 321–326, doi:10.1038/nature14192. ff j 32. Schübeler, D. Function and Information Content of DNA Methylation. Nature 2015, 517, 321–326, d 33. Smith, Z.D.; Meissner, A. DNA Methylation: Roles in Mammalian Development. Nat. Rev. G doi:10.1038/nrg3354. g 34. Tahiliani, M.; Koh, K.P.; Shen, Y.; Pastor, W.A.; Bandukwala, H.; Brudno, Y.; Agarwal, S.; Iyer, L.M.; Liu, D.R.; Aravind, L.; et al. Conversion of 5-Methylcytosine to 5-Hydroxymethylcytosine in Mammalian DNA by MLL Partner TET1. Science 2009, 324, 930–935, doi:10.1126/science.1170116. 35. Zhao, X.; Dai, J.; Ma, Y.; Mi, Y.; Cui, D.; Ju, G.; Macklin, W.B.; Jin, W. Dynamics of Ten-Eleven Transl Proteins and 5-Hydroxymethylcytosine in Oligodendrocyte Differentiation. Glia 2014, 62, 914–926, , J.; Ma, Y.; Mi, Y.; Cui, D.; Ju, G.; Macklin, W.B.; Jin, W. Dynamics of Ten-Eleven Translocation Hydroxylase 5-Hydroxymethylcytosine in Oligodendrocyte Differentiation. Glia 2014, 62, 914–926, doi:10.1002/glia.2264 36. Feng, J.; Shao, N.; Szulwach, K.E.; Vialou, V.; Huynh, J.; Zhong, C.; Le, T.; Ferguson, D.; Cahill, M.E.; Li, Y.; et al. Role of Tet1 and 5-Hydroxymethylcytosine in Cocaine Action. Nat. Neurosci. 2015, 18, 536–544, doi:10.1038/nn.3976. y y y y 37. Szulwach, K.E.; Li, X.; Li, Y.; Song, C.-X.; Wu, H.; Dai, Q.; Irier, H.; Upadhyay, A.K.; Gearing, M.; Levey, A.I.; et al. 5-HmC- Mediated Epigenetic Dynamics during Postnatal Neurodevelopment and Aging. Nat. Neurosci. 2011, 14, 1607–1616, doi:10.1038/nn.2959. 38. Kouzarides, T. Chromatin Modifications and Their Function. Cell 2007, 128, 693–705, doi:10.1016/j.cell.2007.02.005. 39. Dhar, S.S.; Lee, S.-H.; Kan, P.-Y.; Voigt, P.; Ma, L.; Shi, X.; Reinberg, D.; Lee, M.G. Trans-Tail Regulation of MLL4-Catalyzed H3K4 Methylation by H4R3 Symmetric Dimethylation Is Mediated by a Tandem PHD of MLL4. Genes Dev. References Mikkelsen, T.S.; Ku, M.; Jaffe, D.B.; Issac, B.; Lieberman, E.; Giannoukos, G.; Alvarez, P.; Brockman, W.; Kim, T.-K.; Koche, R.P.; et al. Genome-Wide Maps of Chromatin State in Pluripotent and Lineage-Committed Cells. Nature 2007, 448, 553–560, doi:10.1038/nature06008. 51. Rada-Iglesias, A.; Bajpai, R.; Swigut, T.; Brugmann, S.A.; Flynn, R.A.; Wysocka, J. A Unique Chrom Early Developmental Enhancers in Humans. Nature 2011, 470, 279–283, doi:10.1038/nature09692. 52. Shilatifard, A. Chromatin Modifications by Methylation and Ubiquitination: Implications in the Regulation of Gene Expression. Annu. Rev. Biochem. 2006, 75, 243–269, doi:10.1146/annurev.biochem.75.103004.142422. 53. Strahl, B.D.; Allis, C.D. The Language of Covalent Histone Modifications. Nature 2000, 403, 41–45, doi:10.1038/47412. 54 Martin C ; Zhang Y The Diverse Functions of Histone Lysine Methylation Nat Rev Mol Cell Biol 2005 6 838 849 53. Strahl, B.D.; Allis, C.D. The Language of Covalent Histone Modifications. Nature 2000, 403, 41–45, doi 53. Strahl, B.D.; Allis, C.D. The Language of Covalent Histone Modifications. Nature 2000, 403, 41–45, doi:10.1038/47412. 54. Martin, C.; Zhang, Y. The Diverse Functions of Histone Lysine Methylation. Nat. Rev. Mol. Cell Biol. 2005, 6, 838–849, doi:10 1038/nrm1761 55. Falcão, A.M.; Meijer, M.; Scaglione, A.; Rinwa, P.; Agirre, E.; Liang, J.; Larsen, S.C.; Heskol, A.; Frawley, R.; Klingener, M.; et al. PAD2-Mediated Citrullination Contributes to Efficient Oligodendrocyte Differentiation and Myelination. Cell Rep. 2019, 27, 1090–1102.e10, doi:10.1016/j.celrep.2019.03.108. 1090–1102.e10, doi:10.1016/j.celrep.2019.03.108. j p 56. Woodage, T.; Basrai, M.A.; Baxevanis, A.D.; Hieter, P.; Collins, F.S. Characterization of the CHD Fam Acad. Sci. USA 1997, 94, 11472–11477, doi:10.1073/pnas.94.21.11472. j p , M.A.; Baxevanis, A.D.; Hieter, P.; Collins, F.S. Characterization of the CHD Family of Proteins. Proc. Natl. 94, 11472–11477, doi:10.1073/pnas.94.21.11472. p 57. Bischof, M.; Weider, M.; Küspert, M.; Nave, K.-A.; Wegner, M. Brg1-Dependent Chromatin Remodelling Is Not Essentially Required during Oligodendroglial Differentiation. J. Neurosci. Off. J. Soc. Neurosci. 2015, 35, 21–35, doi:10.1523/JNEURO- SCI.1468-14.2015. 58. Hargreaves, D.C.; Crabtree, G.R. ATP-Dependent Chromatin Remodeling: Genetics, Genomics and Mechanisms. Cell Res. 2011, 21, 396–420, doi:10.1038/cr.2011.32. Life 2021, 11, 62 23 of 30 59. He, D.; Marie, C.; Zhao, C.; Kim, B.; Wang, J.; Deng, Y.; Clavairoly, A.; Frah, M.; Wang, H.; He, X.; et al. Chd7 Cooperates with Sox10 and Regulates the Onset of CNS Myelination and Remyelination. Nat. Neurosci. 2016, 19, 678–689, doi:10.1038/nn.4258. 60. Ho, L.; Crabtree, G.R. Chromatin Remodelling during Development. Nature 2010, 463, 474–484, doi:10.1038/nature08911. g y y 60. Ho, L.; Crabtree, G.R. Chromatin Remodelling during Development. Nature 2010, 463, 474–484, do g g p 61. References , C.R.; Zhong, N.; Amato, N.J.; Liu, Y.; Liu, S.; Cai, Q.; Ji, D.; Jin, S.-G.; Niedernhofer, L.J.; et al. Tet-Media Hydroxymethylcytosine in RNA. J. Am. Chem. Soc. 2014, 136, 11582–11585, doi:10.1021/ja505305z. 74. Fu, L.; Guerrero, C.R.; Zhong, N.; Amato, N.J.; Liu, Y.; Liu, S.; Cai, Q.; Ji, D.; Jin, S.-G.; Niedernhofer Formation of 5-Hydroxymethylcytosine in RNA. J. Am. Chem. Soc. 2014, 136, 11582–11585, doi:10.1021 75. Fu, Y.; Dominissini, D.; Rechavi, G.; He, C. Gene Expression Regulation Mediated through Reversible M6A RNA Methylation. Nat. Rev. Genet. 2014, 15, 293–306, doi:10.1038/nrg3724. g 76. Xiao, W.; Adhikari, S.; Dahal, U.; Chen, Y.-S.; Hao, Y.-J.; Sun, B.-F.; Sun, H.-Y.; Li, A.; Ping, X.-L.; Lai, W.-Y.; et al. Nuclear m 6 A Reader YTHDC1 Regulates MRNA Splicing. Mol. Cell 2016, 61, 507–519, doi:10.1016/j.molcel.2016.01.012. g 76. Xiao, W.; Adhikari, S.; Dahal, U.; Chen, Y.-S.; Hao, Y.-J.; Sun, B.-F.; Sun, H.-Y.; Li, A.; Ping, X.-L.; Lai, W.-Y.; et al. Nuclear m 6 A R d YTHDC1 R l MRNA S li i M l C ll 1 61 507 519 d i 10 1016/j l l 2016 01 012 76. Xiao, W.; Adhikari, S.; Dahal, U.; Chen, Y.-S.; Hao, Y.-J.; Sun, B.-F.; Sun, H.-Y.; Li, A.; Ping, X.-L.; Lai, A Reader YTHDC1 Regulates MRNA Splicing. Mol. Cell 2016, 61, 507–519, doi:10.1016/j.molcel.2016.01 77. Yue, Y.; Liu, J.; He, C. RNA N6-Methyladenosine Methylation in Post-Transcriptional Gene Expression Regulation. Genes Dev. 2015, 29, 1343–1355, doi:10.1101/gad.262766.115. 78. Zhao, X.; Yang, Y.; Sun, B.-F.; Shi, Y.; Yang, X.; Xiao, W.; Hao, Y.-J.; Ping, X.-L.; Chen, Y.-S.; Wang, W.-J.; et al. FTO-Dependent Demethylation of N6-Methyladenosine Regulates MRNA Splicing and Is Required for Adipogenesis. Cell Res. 2014, 24, 1403– 1419, doi:10.1038/cr.2014.151. 79. Liu, J.; Jia, G. Methylation Modifications in Eukaryotic Messenger RNA. J. Genet. Genom. 2014, 41, 21–33, doi:10.1016/j.jgg.2013.10.002. j jgg 80. Wei, J.; Liu, F.; Lu, Z.; Fei, Q.; Ai, Y.; He, P.C.; Shi, H.; Cui, X.; Su, R.; Klungland, A.; et al. Differential M6A, M6Am, and M1A Demethylation Mediated by FTO in the Cell Nucleus and Cytoplasm. Mol. Cell 2018, 71, 973–985.e5, doi:10.1016/j.mol- cel.2018.08.011. 81. Gage, F.H. Mammalian Neural Stem Cells. Science 2000, 287, 1433–1438, doi:10.1126/science.287.5457.1433. 82. Murao, N.; Noguchi, H.; Nakashima, K. Epigenetic Regulation of Neural Stem Cell Property from Em genetics 2016, 5, 1–10, doi:10.1016/j.nepig.2016.01.001. g j p g 83. Nakagawa, T.; Wada, Y.; Katada, S.; Kishi, Y. References Marie, C.; Clavairoly, A.; Frah, M.; Hmidan, H.; Yan, J.; Zhao, C.; Van Steenwinckel, J.; Daveau, R.; Zalc, B.; Hassan, B.; et al. Oligodendrocyte Precursor Survival and Differentiation Requires Chromatin Remodeling by Chd7 and Chd8. Proc. Natl. Acad. Sci. USA 2018, 201802620, doi:10.1073/pnas.1802620115. p 62. Yu, Y.; Chen, Y.; Kim, B.; Wang, H.; Zhao, C.; He, X.; Liu, L.; Liu, W.; Wu, L.M.; Mao, M.; et al. Olig2 Targets Chromatin Re- modelers to Enhancers to Initiate Oligodendrocyte Differentiation. Cell 2013, 152, 248–261, doi:10.1016/j.cell.2012.12.006. ns, B.R. The Biology of Chromatin Remodeling Complexes. Annu. Rev. Biochem. 2009, 78, 273–304, .biochem.77.062706.153223. 63. Clapier, C.R.; Cairns, B.R. The Biology of Chromatin Remodeling Complexes. Annu. Rev. Bioc doi:10.1146/annurev.biochem.77.062706.153223. 64. Catez, F.; Yang, H.; Tracey, K.J.; Reeves, R.; Misteli, T.; Bustin, M. Network of Dynamic Interactions between Histone H1 and High-Mobility-Group Proteins in Chromatin. Mol. Cell. Biol. 2004, 24, 4321–4328, doi:10.1128/MCB.24.10.4321-4328.2004. 65. Fusco, A.; Fedele, M. Roles of HMGA Proteins in Cancer. Nat. Rev. Cancer 2007, 7, 899–910, doi:10.1038/nrc2271. h Ad S A G ld l d Ch h S Ad E at, T.; Adam, S.A.; Goldman, R.D. Nuclear Lamins and Chromatin: When Structure Meets Function. Adv. 49, 157–166, doi:10.1016/j.advenzreg.2008.12.003. j g 67. Naetar, N.; Ferraioli, S.; Foisner, R. Lamins in the Nuclear Interior—Life Outside the Lamina. J. Cell Sci. 2017, 130, 2087–2096, doi:10.1242/jcs.203430. j 68. Yattah, C.; Hernandez, M.; Huang, D.; Park, H.; Liao, W.; Casaccia, P. Dynamic Lamin B1-Gene Association During Oligoden- drocyte Progenitor Differentiation. Neurochem. Res. 2020, 45, 606–619, doi:10.1007/s11064-019-02941-y. y g y 69. Derrien, T.; Guigó, R.; Johnson, R. The Long Non-Coding RNAs: A New (P)Layer in the “Dark Matter”. Front. Genet. 2012, 2, doi:10.3389/fgene.2011.00107. g 70. Mercer, T.R.; Dinger, M.E.; Sunkin, S.M.; Mehler, M.F.; Mattick, J.S. Specific Expression of Long Noncoding RNAs in the Mouse Brain. Proc. Natl. Acad. Sci. USA 2008, 105, 716–721, doi:10.1073/pnas.0706729105. 71. Qureshi, I.A.; Mehler, M.F. Emerging Roles of Non-Coding RNAs in Brain Evolution, Development, Plasticity and Disease. Nat. Rev. Neurosci. 2012, 13, 528–541, doi:10.1038/nrn3234. vine, D.V.; Martienssen, R.A. Noncoding RNAs and Gene Silencing. Cell 2007, 128, 763–776, 7.02.016. 72. Zaratiegui, M.; Irvine, D.V.; Martienssen, R.A. Noncoding RNAs and Gene Silencing. doi:10.1016/j.cell.2007.02.016. 73. Lim, L.P.; Lau, N.C.; Garrett-Engele, P.; Grimson, A.; Schelter, J.M.; Castle, J.; Bartel, D.P.; Linsley, P.S.; Johnson, J.M. Microarray Analysis Shows That Some MicroRNAs Downregulate Large Numbers of Target MRNAs. Nature 2005, 433, 769–773, doi:10.1038/nature03315. References Immunoprecipitation and Mass Spectrometry Define TET1 Interactome during Oligodendrocyte Differentiation. Cell Biosci. 2020, 10, 110, doi:10.1186/s13578-020-00473-5. 94. Sher, F.; Boddeke, E.; Olah, M.; Copray, S. Dynamic Changes in Ezh2 Gene Occupancy Underlie Its Involvement in Neural Stem Cell Self-Renewal and Differentiation towards Oligodendrocytes. PLoS ONE 2012, 7, doi:10.1371/journal.pone.0040399. 95. Wang, J.; Yang, L.; Dong, C.; Wang, J.; Xu, L.; Qiu, Y.; Weng, Q.; Zhao, C.; Xin, M.; Lu, Q.R. EED-Mediated Histone Methylation Is Critical for CNS Myelination and Remyelination by Inhibiting WNT, BMP, and Senescence Pathways. Sci. Adv. 2020, 6, eaaz6477, doi:10.1126/sciadv.aaz6477. 96. Sher, F.; Rößler, R.; Brouwer, N.; Balasubramaniyan, V.; Boddeke, E.; Copray, S. Differentiation of Neural Stem Cells into Oli- godendrocytes: Involvement of the Polycomb Group Protein Ezh2. Stem Cells 2008, 26, 2875–2883, doi:10.1634/stemcells.2008- 0121. 97. Liu, J.; Magri, L.; Zhang, F.; Marsh, N.O.; Albrecht, S.; Huynh, J.L.; Kaur, J.; Kuhlmann, T.; Zhang, W.; Slesinger, P.A.; et al. Chromatin Landscape Defined by Repressive Histone Methylation during Oligodendrocyte Differentiation. J. Neurosci. 2015, 35, 352–365, doi:10.1523/JNEUROSCI.2606-14.2015. 98. Hashimoto, M.; Murata, K.; Ishida, J.; Kanou, A.; Kasuya, Y.; Fukamizu, A. Severe Hypomyelination and Developmental Defects Are Caused in Mice Lacking Protein Arginine Methyltransferase 1 (PRMT1) in the Central Nervous System. J. Biol. Chem. 2016, 291, 2237–2245, doi:10.1074/jbc.M115.684514. j 99. Wang, J.; Weaver, I.C.G.; Gauthier-Fisher, A.; Wang, H.; He, L.; Yeomans, J.; Wondisford, F.; Kaplan, D.R.; Miller, F.D. CBP Histone Acetyltransferase Activity Regulates Embryonic Neural Differentiation in the Normal and Rubinstein-Taybi Syndrome Brain. Dev. Cell 2010, 18, 114–125, doi:10.1016/j.devcel.2009.10.023. j 100. Liu, A.; Han, Y.R.; Li, J.; Sun, D.; Ouyang, M.; Plummer, M.R.; Casaccia-Bonnefil, P. The Glial or Neuronal Fate Choice of Oli- godendrocyte Progenitors Is Modulated by Their Ability to Acquire an Epigenetic Memory. J. Neurosci. Off. J. Soc. Neurosci. 2007, 27, 7339–7343, doi:10.1523/JNEUROSCI.1226-07.2007. 101. Montgomery, R.L.; Hsieh, J.; Barbosa, A.C.; Richardson, J.A.; Olson, E.N. Histone Deacetylases 1 and 2 Control the Progression of Neural Precursors to Neurons during Brain Development. Proc. Natl. Acad. Sci. USA 2009, 106, 7876–7881, doi:10.1073/pnas.0902750106. p 102. Castelo-Branco, G.; Lilja, T.; Wallenborg, K.; Falcão, A.M.; Marques, S.C.; Gracias, A.; Solum, D.; Paap, R.; Walfridsson, J.; Teixeira, A.I.; et al. Neural Stem Cell Differentiation Is Dictated by Distinct Actions of Nuclear Receptor Corepressors and Histone Deacetylases. Stem Cell Rep. 2014, 3, 502–515, doi:10.1016/j.stemcr.2014.07.008. y p j 103. References Epigenetic Regulation for Acquiring Glial Identity by Neural Stem Cells during Cortical Development. Glia 2020, 68, 1554–1567, doi:10.1002/glia.23818. p g 84. Sanosaka, T.; Imamura, T.; Hamazaki, N.; Chai, M.; Igarashi, K.; Ideta-Otsuka, M.; Miura, F.; Ito, T.; Fujii, N.; Ikeo, K.; et al. DNA Methylome Analysis Identifies Transcription Factor-Based Epigenomic Signatures of Multilineage Competence in Neural Stem/Progenitor Cells. Cell Rep. 2017, 20, 2992–3003, doi:10.1016/j.celrep.2017.08.086. g p j p 85. Takizawa, T.; Nakashima, K.; Namihira, M.; Ochiai, W.; Uemura, A.; Yanagisawa, M.; Fujita, N.; Nakao, M.; Taga, T. DNA Methylation Is a Critical Cell-Intrinsic Determinant of Astrocyte Differentiation in the Fetal Brain. Dev. Cell 2001, 1, 749–758, doi:10.1016/s1534-5807(01)00101-0. 86. Fan, G.; Martinowich, K.; Chin, M.H.; He, F.; Fouse, S.D.; Hutnick, L.; Hattori, D.; Ge, W.; Shen, Y.; Wu, H.; et al. DNA Methyl- ation Controls the Timing of Astrogliogenesis through Regulation of JAK-STAT Signaling. Dev. Camb. Engl. 2005, 132, 3345– 3356, doi:10.1242/dev.01912. Life 2021, 11, 62 24 of 30 87. Hatada, I.; Namihira, M.; Morita, S.; Kimura, M.; Horii, T.; Nakashima, K. Astrocyte-Specific Genes Are Generally Demethylated in Neural Precursor Cells Prior to Astrocytic Differentiation. PLoS ONE 2008, 3, e3189, doi:10.1371/journal.pone.0003189. y j p 88. Namihira, M.; Kohyama, J.; Semi, K.; Sanosaka, T.; Deneen, B.; Taga, T.; Nakashima, K. Committed Neuronal Precursors Confer Astrocytic Potential on Residual Neural Precursor Cells. Dev. Cell 2009, 16, 245–255, doi:10.1016/j.devcel.2008.12.014. y j 89. Teter, B.; Rozovsky, I.; Krohn, K.; Anderson, C.; Osterburg, H.; Finch, C. Methylation of the glial fibrillary acidic protein gene shows novel biphasic changes during brain development. Glia 1996, 17, 195–205, doi:10.1002/(SICI)1098- 1136(199607)17:3<195::AID-GLIA2>3.0.CO;2-0. 90. Wheldon, L.M.; Abakir, A.; Ferjentsik, Z.; Dudnakova, T.; Strohbuecker, S.; Christie, D.; Dai, N.; Guan, S.; Foster, J.M.; Corrêa, I.R.; et al. Transient Accumulation of 5-Carboxylcytosine Indicates Involvement of Active Demethylation in Lineage Specifica- tion of Neural Stem Cells. Cell Rep. 2014, 7, 1353–1361, doi:10.1016/j.celrep.2014.05.003. p j p 91. Moyon, S.; Huynh, J.; Yoo, S.; Dutta, D.; Zhang, F.; Ma, D.; Yoo, S.; Lawrence, R.; Wegner, M.; John, G.; et al. Functional Char- acterization of DNA Methylation in the Oligodendrocyte Lineage. Cell Rep. 2016, doi:10.1016/j.celrep.2016.03.060. 92. Zhang, M.; Wang, J.; Zhang, K.; Lu, G.; Xu, L.; Ren, K.; Wang, W.; Liu, Y.; Xing, J.; Gao, X.; et al. TET1-Mediated Oligodendrocyte Homeostasis Regulates Myelination and Synaptic Functions. bioRxiv 2019, 821496, doi:10.1101/821496. 93. Zhang, M.; Zhang, K.; Wang, J.; Liu, Y.; Liu, G.; Jin, W.; Wu, S.; Zhao, X. 87. Hatada, I.; Namihira, M.; Morita, S.; Kimura, M.; Horii, T.; Nakashima, K. Astrocyte-Specific Genes Are Generally Demethylated in Neural Precursor Cells Prior to Astrocytic Differentiation. PLoS ONE 2008, 3, e3189, doi:10.1371/journal.pone.0003189. 88 N ihi M K h J S i K S k T D B T T N k hi K C itt d N l P C f References Ciliary Neurotrophic Factor Enhances the Rate of Oligodendrocyte Generation. Mol Cell Neurosci. 1996, 8, 146–156, doi:10.1006/mcne.1996.0053. 123. Canoll, P.D.; Musacchio, J.M.; Hardy, R.; Reynolds, R.; Marchionni, M.A.; Salzer, J.L. GGF/Neuregulin Is a Neuronal Signal That Promotes the Proliferation and Survival and Inhibits the Differentiation of Oligodendrocyte Progenitors. Neuron 1996, 17, 229– 243, doi:10.1016/s0896-6273(00)80155-5. 124. Diemel, L.T.; Jackson, S.J.; Cuzner, M.L. Role for TGF-Beta1, FGF-2 and PDGF-AA in a Myelination of CNS Aggregate Cultures Enriched with Macrophages. J. Neurosci. Res. 2003, 74, 858–867, doi:10.1002/jnr.10837. 125. Noble, M.; Murray, K.; Stroobant, P.; Waterfield, M.D.; Riddle, P. Platelet-Derived Growth Factor Promotes Division and Mo- tility and Inhibits Premature Differentiation of the Oligodendrocyte/Type-2 Astrocyte Progenitor Cell. Nature 1988, 333, 560– 562, doi:10.1038/333560a0. 126. Ohya, W.; Funakoshi, H.; Kurosawa, T.; Nakamura, T. Hepatocyte Growth Factor (HGF) Promotes Oligodendrocyte Progenitor Cell Proliferation and Inhibits Its Differentiation during Postnatal Development in the Rat. Brain Res. 2007, 1147, 51–65, doi:10.1016/j.brainres.2007.02.045. j 127. Barres, B.A.; Raff, M.C. Proliferation of Oligodendrocyte Precursor Cells Depends on Electrical Activity in Axons. Nature 1993, 361, 258–260, doi:10.1038/361258a0. 128. Demerens, C.; Stankoff, B.; Logak, M.; Anglade, P.; Allinquant, B.; Couraud, F.; Zalc, B.; Lubetzki, C. Induction of Myelination in the Central Nervous System by Electrical Activity. Proc. Natl. Acad. Sci. USA 1996, 93, 9887–9892, doi:10.1073/pnas.93.18.9887. 129. Gallo, V.; Zhou, J.M.; McBain, C.J.; Wright, P.; Knutson, P.L.; Armstrong, R.C. Oligodendrocyte Progenitor Cell Proliferation and Lineage Progression Are Regulated by Glutamate Receptor Mediated K+ Channel Block J Neurosci 1996 16 2659 2670 128. Demerens, C.; Stankoff, B.; Logak, M.; Anglade, P.; Allinquant, B.; Couraud, F.; Zalc, B.; Lubetzki, C. Induction of Myelination in the Central Nervous System by Electrical Activity. Proc. Natl. Acad. Sci. USA 1996, 93, 9887–9892, doi:10.1073/pnas.93.18.9887. 128. Demerens, C.; Stankoff, B.; Logak, M.; Anglade, P.; Allinquant, B.; Couraud, F.; Zalc, B.; Lubetzki, C. Induction of Myelination in the Central Nervous System by Electrical Activity. Proc. Natl. Acad. Sci. USA 1996, 93, 9887–9892, doi:10.1073/pnas.93.18.9887. 129. Gallo, V.; Zhou, J.M.; McBain, C.J.; Wright, P.; Knutson, P.L.; Armstrong, R.C. Oligodendrocyte Progenitor Cell Proliferation and Lineage Progression Are Regulated by Glutamate Receptor-Mediated K+ Channel Block. J. Neurosci. 1996, 16, 2659–2670, doi:10.1523/JNEUROSCI.16-08-02659.1996. 129. Gallo, V.; Zhou, J.M.; McBain, C.J.; Wright, P.; Knutson, P.L.; Armstrong, R.C. Oligodendrocyte Progenitor Cell Proliferation and Lineage Progression Are Regulated by Glutamate Receptor-Mediated K+ Channel Block. J. Neurosci. 1996, 16, 2659–2670, doi:10.1523/JNEUROSCI.16-08-02659.1996. 130. Li, Q.; Brus-Ramer, M.; Martin, J.H.; McDonald, J.W. References The SWI/SNF Chromatin Remodeling Protein Brg1 Is Requir esis and Mediates Transactivation of Ngn and NeuroD. Development 2005, 132, 105–115, doi:10.124 113. Gregg, R.G.; Willer, G.B.; Fadool, J.M.; Dowling, J.E.; Link, B.A. Positional Cloning of the Young Mutation Identifies an Essential Role for the Brahma Chromatin Remodeling Complex in Mediating Retinal Cell Differentiation. Proc. Natl. Acad. Sci. USA 2003, 100, 6535–6540, doi:10.1073/pnas.0631813100. p 114. Neo, W.H.; Yap, K.; Lee, S.H.; Looi, L.S.; Khandelia, P.; Neo, S.X.; Makeyev, E.V.; Su, I. MicroRNA MiR-124 Controls the Choice between Neuronal and Astrocyte Differentiation by Fine-Tuning Ezh2 Expression. J. Biol. Chem. 2014, 289, 20788–20801, doi:10.1074/jbc.M113.525493. j 115. Tsuyama, J.; Bunt, J.; Richards, L.J.; Iwanari, H.; Mochizuki, Y.; Hamakubo, T.; Shimazaki, T.; Okano, H. MicroRNA-153 Regu- lates the Acquisition of Gliogenic Competence by Neural Stem Cells. Stem Cell Rep. 2015, 5, 365–377, doi:10.1016/j.stemcr.2015.06.006. j 116. Naka-Kaneda, H.; Nakamura, S.; Igarashi, M.; Aoi, H.; Kanki, H.; Tsuyama, J.; Tsutsumi, S.; Aburatani, H.; Shimazaki, T.; Okano, H. The MiR-17/106–P38 Axis Is a Key Regulator of the Neurogenic-to-Gliogenic Transition in Developing Neural Stem/Progen- itor Cells. Proc. Natl. Acad. Sci. USA 2014, 111, 1604–1609, doi:10.1073/pnas.1315567111. 117. Dong, X.; Chen, K.; Cuevas-Diaz Duran, R.; You, Y.; Sloan, S.A.; Zhang, Y.; Zong, S.; Cao, Q.; Barres, B.A.; Wu, J.Q. Compre- hensive Identification of Long Non-Coding RNAs in Purified Cell Types from the Brain Reveals Functional LncRNA in OPC Fate Determination. PLoS Genet. 2015, 11, e1005669, doi:10.1371/journal.pgen.1005669. j pg 118. Katsel, P.; Roussos, P.; Fam, P.; Khan, S.; Tan, W.; Hirose, T.; Nakagawa, S.; Pletnikov, M.V.; Haroutunian, V. The Expression of Long Noncoding RNA NEAT1 Is Reduced in Schizophrenia and Modulates Oligodendrocytes Transcription. NPJ Schizophr. 2019, 5, 1–9, doi:10.1038/s41537-019-0071-2. , , , 119. Li, Y.; Guo, B.; Yang, R.; Xiao, Z.; Gao, X.; Yu, J.; Li, S.; Luo, Y. A Novel Long Noncoding RNA Lnc158 Promotes the Differenti- ation of Mouse Neural Precursor Cells into Oligodendrocytes by Targeting Nuclear Factor-IB. Neuroreport 2018, 29, 1121–1128, doi:10.1097/WNR.0000000000001083. 120. Mercer, T.R.; Qureshi, I.A.; Gokhan, S.; Dinger, M.E.; Li, G.; Mattick, J.S.; Mehler, M.F. Long Noncoding RNAs in Neuronal- Glial Fate Specification and Oligodendrocyte Lineage Maturation. BMC Neurosci. 2010, 11, 14, doi:10.1186/1471-2202-11-14. 121. Hughes, E.G.; Kang, S.H.; Fukaya, M.; Bergles, D.E. Oligodendrocyte Progenitors Balance Growth with Self-Repulsion to Achieve Homeostasis in the Adult Brain. Nat. Neurosci. 2013, 16, 668–676, doi:10.1038/nn.3390. 122. Barres, B.A.; Burne, J.F.; Holtmann, B.; Thoenen, H.; Sendtner, M.; Raff, M.C. References Ye, F.; Chen, Y.; Hoang, T.; Montgomery, R.L.; Zhao, X.H.; Bu, H.; Hu, T.; Taketo, M.M.; van Es, J.H.; Clevers, H.; et al. HDAC1 and HDAC2 Regulate Oligodendrocyte Differentiation by Disrupting the Beta-Catenin-TCF Interaction. Nat. Neurosci. 2009, 12, 829–838, doi:10.1038/nn.2333. , 104. Zhang, L.; He, X.; Liu, L.; Jiang, M.; Zhao, C.; Wang, H.; He, D.; Zheng, T.; Zhou, X.; Hassan, A.; et al. Hdac3 Interaction with P300 Histone Acetyltransferase Regulates the Oligodendrocyte and Astrocyte Lineage Fate Switch. Dev. Cell 2016, 36, 316–330, doi:10.1016/j.devcel.2016.01.002. j 105. Wu, M.; Hernandez, M.; Shen, S.; Sabo, J.K.; Kelkar, D.; Wang, J.; O’Leary, R.; Phillips, G.R.; Cate, H.S.; Casaccia, P. Differential Modulation of the Oligodendrocyte Transcriptome by Sonic Hedgehog and Bone Morphogenetic Protein 4 via Opposing Effects on Histone Acetylation. J. Neurosci. Off. J. Soc. Neurosci. 2012, 32, 6651–6664, doi:10.1523/JNEUROSCI.4876-11.2012. 106. Rafalski, V.A.; Ho, P.P.; Brett, J.O.; Ucar, D.; Dugas, J.C.; Pollina, E.A.; Chow, L.M.L.; Ibrahim, A.; Baker, S.J.; Barres, B.A.; et al. Expansion of Oligodendrocyte Progenitor Cells Following SIRT1 Inactivation in the Adult Brain. Nat. Cell Biol. 2013, 15, 614– 624, doi:10.1038/ncb2735. , 107. Kishi, Y.; Fujii, Y.; Hirabayashi, Y.; Gotoh, Y. HMGA Regulates the Global Chromatin State and Neurogenic Potential in Neo- cortical Precursor Cells. Nat. Neurosci. 2012, 15, 1127–1133, doi:10.1038/nn.3165. 108. Abraham, A.B.; Bronstein, R.; Reddy, A.S.; Maletic-Savatic, M.; Aguirre, A.; Tsirka, S.E. Aberrant Neural Stem Cell Proliferation and Increased Adult Neurogenesis in Mice Lacking Chromatin Protein HMGB2. PLoS ONE 2013, 8, doi:10.1371/jour- nal.pone.0084838. Life 2021, 11, 62 25 of 30 109. Bronstein, R.; Kyle, J.; Abraham, A.B.; Tsirka, S.E. Neurogenic to Gliogenic Fate Transition Perturbed by Loss of HMGB2. Front. Mol. Neurosci. 2017, 10, doi:10.3389/fnmol.2017.00153. 110. Rouhiainen, A.; Zhao, X.; Vanttola, P.; Qian, K.; Kulesskiy, E.; Kuja-Panula, J.; Gransalke, K.; Grönholm, M.; Unni, E.; Meistrich, M.; et al. HMGB4 Is Expressed by Neuronal Cells and Affects the Expression of Genes Involved in Neural Differentiation. Sci. Rep. 2016, 6, doi:10.1038/srep32960. p p 111. Matsumoto, S.; Banine, F.; Struve, J.; Xing, R.; Adams, C.; Liu, Y.; Metzger, D.; Chambon, P.; Rao, M.S.; Sherman, L.S. Brg1 Is Required for Murine Neural Stem Cell Maintenance and Gliogenesis. Dev. Biol. 2006, 289, 372–383, doi:10.1016/j.yd- bio.2005.10.044. G.A.; Kroll, K.L. The SWI/SNF Chromatin Remodeling Protein Brg1 Is Required for Vertebrate Neurogen- ransactivation of Ngn and NeuroD. Development 2005, 132, 105–115, doi:10.1242/dev.01548. 112. Seo, S.; Richardson, G.A.; Kroll, K.L. 130. Li, Q.; Brus-Ramer, M.; Martin, J.H.; McDonald, J.W. Electrical Stimulation of the Medullary Pyramid Promotes Proliferation and Differentiation of Oligodendrocyte Progenitor Cells in the Corticospinal Tract of the Adult Rat. Neurosci. Lett. 2010, 479, 128–133, doi:10.1016/j.neulet.2010.05.043. References The Essential Cofactor TRRAP Recruits the Histone Ace Myc. Mol. Cell. Biol. 2000, 20, 556–562, doi:10.1128/mcb.20.2.556-562.2000. 142. Doi, T.; Ogata, T.; Yamauchi, J.; Sawada, Y.; Tanaka, S.; Nagao, M. Chd7 Collaborates with Sox2 to Regulate Activation of Oligodendrocyte Precursor Cells after Spinal Cord Injury. J. Neurosci. 2017, 37, 10290–10309, doi:10.1523/JNEUROSCI.1109- 17.2017. 143. Moyon, S.; Frawley, R.; Marshall-Phelps, K.L.; Kegel, L.; Bostrand, S.M.; Sadowski, B.; Huang, D.; Jiang, Y.-H.; Lyons, D.; Mobius, W.; et al. TET1-Mediated DNA Hydroxy-Methylation Regulates Adult Remyelination. bioRxiv 2019, 819995, doi:10.1101/819995. 144. Hill, R.A.; Li, A.M.; Grutzendler, J. Lifelong Cortical Myelin Plasticity and Age-Related Degeneration in the Live Mammalian Brain. Nat. Neurosci. 2018, 21, 683–695, doi:10.1038/s41593-018-0120-6. 143. Moyon, S.; Frawley, R.; Marshall-Phelps, K.L.; Kegel, L.; Bostrand, S.M.; Sadowski, B.; Huang, D.; Jia W.; et al. TET1-Mediated DNA Hydroxy-Methylation Regulates Adult Remyelination. bioRxiv 2019 W.; et al. TET1-Mediated DNA Hydroxy-Methylation Regulates Adult Remyelination. bioRxiv 2019, 819995, doi:10.1101/819995. 144. Hill, R.A.; Li, A.M.; Grutzendler, J. Lifelong Cortical Myelin Plasticity and Age-Related Degeneration in the Live Mammalian Brain. Nat. Neurosci. 2018, 21, 683–695, doi:10.1038/s41593-018-0120-6. y y y g y 144. Hill, R.A.; Li, A.M.; Grutzendler, J. Lifelong Cortical Myelin Plasticity and Age-Related Degeneration in the Live Mammalian Brain. Nat. Neurosci. 2018, 21, 683–695, doi:10.1038/s41593-018-0120-6. , , , 145. Hughes, E.G.; Orthmann-Murphy, J.L.; Langseth, A.J.; Bergles, D.E. Myelin Remodeling through Experience-Dependent Oli- godendrogenesis in the Adult Somatosensory Cortex. Nat. Neurosci. 2018, 21, 696–706, doi:10.1038/s41593-018-0121-5. 146. Salami, M.; Itami, C.; Tsumoto, T.; Kimura, F. Change of Conduction Velocity by Regional Myelination Yields Constant Latency Irrespective of Distance between Thalamus and Cortex. Proc. Natl. Acad. Sci. USA 2003, 100, 6174–6179, doi:10.1073/pnas.0937380100. p 147. Seidl, A.H.; Rubel, E.W. Systematic and Differential Myelination of Axon Collaterals in the Mammalian Auditory Brainstem. Glia 2016, 64, 487–494, doi:10.1002/glia.22941. 148. Tomassy, G.S.; Berger, D.R.; Chen, H.-H.; Kasthuri, N.; Hayworth, K.J.; Vercelli, A.; Seung, H.S.; Lichtman, J.W.; Arlotta, P. Distinct Profiles of Myelin Distribution Along Single Axons of Pyramidal Neurons in the Neocortex. Science 2014, 344, 319–324, doi:10.1126/science.1249766. 149. Zonouzi, M.; Berger, D.; Jokhi, V.; Kedaigle, A.; Lichtman, J.; Arlotta, P. Individual Oligodendrocytes Show Bias for Inhibitory Axons in the Neocortex. Cell Rep. 2019, 27, 2799–2808.e3, doi:10.1016/j.celrep.2019.05.018. p j p 150. Grubinska, B.; Laszkiewicz, I.; Royland, J.; Wiggins, R.C.; Konat, G.W. Differentiation-Specific Demethylation of Myelin Asso- ciated Glycoprotein Gene in Cultured Oligodendrocytes. J. Neurosci. Res. 1994, 39, 233–242, doi:10.1002/jnr.490390302. 151. References Electrical Stimulation of the Medullary Pyramid Promotes Proliferation and Differentiation of Oligodendrocyte Progenitor Cells in the Corticospinal Tract of the Adult Rat. Neurosci. Lett. 2010, 479, 128–133, doi:10.1016/j.neulet.2010.05.043. Life 2021, 11, 62 26 of 30 26 of 30 131. Egawa, N.; Shindo, A.; Hikawa, R.; Kinoshita, H.; Liang, A.C.; Itoh, K.; Lok, J.; Maki, T.; Takahashi, R.; Lo, E.H.; et al. Differential Roles of Epigenetic Regulators in the Survival and Differentiation of Oligodendrocyte Precursor Cells. Glia 2019, 67, 718–728, doi:10.1002/glia.23567. g 132. Moyon, S.; Ma, D.; Huynh, J.L.; Coutts, D.J.C.; Zhao, C.; Casaccia, P.; Franklin, R.J.M. Efficient Re Methylation. eNeuro 2017, 4, ENEURO.0336-16.2017, doi:10.1523/ENEURO.0336-16.2017. y 133. Scaglione, A.; Patzig, J.; Liang, J.; Frawley, R.; Bok, J.; Mela, A.; Yattah, C.; Zhang, J.; Teo, S.X.; Zhou, T.; et al. PRMT5-Mediated Regulation of Developmental Myelination. Nat. Commun. 2018, 9, 2840, doi:10.1038/s41467-018-04863-9. 134. Elsesser, O.; Fröb, F.; Küspert, M.; Tamm, E.R.; Fujii, T.; Fukunaga, R.; Wegner, M. Chromatin Remodeler Ep400 Ensures Oli- godendrocyte Survival and Is Required for Myelination in the Vertebrate Central Nervous System. Nucleic Acids Res. 2019, 47, 6208–6224, doi:10.1093/nar/gkz376. g 135. Zhao, C.; Dong, C.; Frah, M.; Deng, Y.; Marie, C.; Zhang, F.; Xu, L.; Ma, Z.; Dong, X.; Lin, Y.; et al. Dual Requirement of CHD8 for Chromatin Landscape Establishment and Histone Methyltransferase Recruitment to Promote CNS Myelination and Repair. Dev. Cell 2018, 45, 753–768.e8, doi:10.1016/j.devcel.2018.05.022. j 136. Magri, L.; Gacias, M.; Wu, M.; Swiss, V.A.; Janssen, W.G.; Casaccia, P. C-Myc-Dependent Transcriptional Regulation of Cell Cycle and Nucleosomal Histones during Oligodendrocyte Differentiation. Neuroscience 2014, 276, 72–86, doi:10.1016/j.neurosci- ence.2014.01.051. 137. Marin-Husstege, M.; Muggironi, M.; Liu, A.; Casaccia-Bonnefil, P. Histone Deacetylase Activity Is Necessary for Oligodendro- cyte Lineage Progression. J. Neurosci. 2002, 22, 10333–10345, doi:10.1523/JNEUROSCI.22-23-10333.2002. 138. Dang, C.V. C-Myc Target Genes Involved in Cell Growth, Apoptosis, and Metabolism. Mol. Cell. Biol. 1999, 19, 1–11, doi:10.1128/mcb.19.1.1. 139. Guccione, E.; Martinato, F.; Finocchiaro, G.; Luzi, L.; Tizzoni, L.; Dall’ Olio, V.; Zardo, G.; Nervi, C.; Bernard, L.; Amati, B. Myc- Binding-Site Recognition in the Human Genome Is Determined by Chromatin Context. Nat. Cell Biol. 2006, 8, 764–770, doi:10.1038/ncb1434. 140. Martinato, F.; Cesaroni, M.; Amati, B.; Guccione, E. Analysis of Myc-Induced Histone Modifications on Target Chromatin. PLoS ONE 2008, 3, doi:10.1371/journal.pone.0003650. od, M.A.; Cole, M.D. The Essential Cofactor TRRAP Recruits the Histone Acetyltransferase HGCN5 to C- 2000, 20, 556–562, doi:10.1128/mcb.20.2.556-562.2000. 141. McMahon, S.B.; Wood, M.A.; Cole, M.D. References Kasuga, Y.; Fudge, A.D.; Zhang, Y.; Li, H. Characterization of a Long Noncoding RNA Pcdh17it as a Novel Marker for Immature Premyelinating Oligodendrocytes. Glia 2019, 67, 2166–2177, doi:10.1002/glia.23684. 164. Lecca, D.; Marangon, D.; Coppolino, G.T.; Méndez, A.M.; Finardi, A.; Costa, G.D.; Martinelli, V.; Furlan, R.; Abbracchio, M.P. MiR-125a-3p Timely Inhibits Oligodendroglial Maturation and Is Pathologically up-Regulated in Human Multiple Sclerosis. Sci. Rep. 2016, 6, doi:10.1038/srep34503. 165. Lin, S.-T.; Huang, Y.; Zhang, L.; Heng, M.Y.; Ptáček, L.J.; Fu, Y.-H. MicroRNA-23a Promotes Myelination in the Central Nervous System. Proc. Natl. Acad. Sci. USA 2013, 110, 17468–17473, doi:10.1073/pnas.1317182110. 166. Shin, D.; Shin, J.-Y.; McManus, M.T.; Ptáček, L.J.; Fu, Y.-H. Dicer Ablation in Oligodendrocytes Pro in Mice. Ann. Neurol. 2009, 66, 843–857, doi:10.1002/ana.21927. McManus, M.T.; Ptáček, L.J.; Fu, Y.-H. Dicer Ablation in Oligodendrocytes Provokes Neuronal Impairment l. 2009, 66, 843–857, doi:10.1002/ana.21927. 167. Shin, D.; Howng, S.Y.B.; Ptáček, L.J.; Fu, Y.-H. MiR-32 and Its Target SLC45A3 Regulate the Lipid Metabolism of Oligodendro- cytes and Myelin. Neuroscience 2012, 213, 29–37, doi:10.1016/j.neuroscience.2012.03.054. 168. Tripathi, A.; Volsko, C.; Garcia, J.P.; Agirre, E.; Allan, K.C.; Tesar, P.J.; Trapp, B.D.; Castelo-Branco, G.; Sim, F.J.; Dutta, R. Oli- godendrocyte Intrinsic MiR-27a Controls Myelination and Remyelination. Cell Rep. 2019, 29, 904–919.e9, doi:10.1016/j.celrep.2019.09.020. j p 169. Wang, H.; Moyano, A.L.; Ma, Z.; Deng, Y.; Lin, Y.; Zhao, C.; Zhang, L.; Jiang, M.; He, X.; Ma, Z.; et al. MiR-219 Cooperates with MiR-338 in Myelination and Promotes Myelin Repair in the CNS. Dev. Cell 2017, 40, 566–582.e5, doi:10.1016/j.devcel.2017.03.001. j p 169. Wang, H.; Moyano, A.L.; Ma, Z.; Deng, Y.; Lin, Y.; Zhao, C.; Zhang, L.; Jiang, M.; He, X.; Ma, Z.; et al. MiR-219 Cooperates with MiR-338 in Myelination and Promotes Myelin Repair in the CNS. Dev. Cell 2017, 40, 566–582.e5, doi:10.1016/j.devcel.2017.03.001. 170. Zhao, X.; He, X.; Han, X.; Yu, Y.; Ye, F.; Chen, Y.; Hoang, T.; Xu, X.; Mi, Q.S.; Xin, M.; et al. MicroRNA-Mediated Control of Oligodendrocyte Differentiation. Neuron 2010, 65, 612–626, doi:10.1016/j.neuron.2010.02.018. y y p , , , j 170. Zhao, X.; He, X.; Han, X.; Yu, Y.; Ye, F.; Chen, Y.; Hoang, T.; Xu, X.; Mi, Q.S.; Xin, M.; et al. MicroRNA-Mediated Control of Oligodendrocyte Differentiation. Neuron 2010, 65, 612–626, doi:10.1016/j.neuron.2010.02.018. 171. Conway, E.; Healy, E.; Bracken, A.P. PRC2 Mediated H3K27 Methylations in Cellular Identity and Cancer. Curr. Opin. Cell Biol. 2015, 37, 42–48, doi:10.1016/j.ceb.2015.10.003. 172. References Conway, G.D.; O’Bara, M.A.; Vedia, B.H.; Pol, S.U.; Sim, F.J. Histone Deacetylase Activity Is Required for Human Oligodendro- cyte Progenitor Differentiation. Glia 2012, 60, 1944–1953, doi:10.1002/glia.22410. y g g 152. Huang, J.; Vogel, G.; Yu, Z.; Almazan, G.; Richard, S. Type II Arginine Methyltransferase PRMT5 Regulates Gene Expression of Inhibitors of Differentiation/DNA Binding Id2 and Id4 during Glial Cell Differentiation. J. Biol. Chem. 2011, 286, 44424–44432, doi:10.1074/jbc.M111.277046. j 153. Liu, H.; Hu, Q.; D’ercole, A.J.; Ye, P. Histone Deacetylase 11 Regulates Oligodendrocyte-Specific Gene Expression and Cell Development in OL-1 Oligodendroglia Cells. Glia 2009, 57, 1–12, doi:10.1002/glia.20729. 154. Shen, S.; Li, J.; Casaccia-Bonnefil, P. Histone Modifications Affect Timing of Oligodendrocyte Progenitor Differentiation in the Developing Rat Brain. J. Cell Biol. 2005, 169, 577–589, doi:10.1083/jcb.200412101. 155. Shen, S.; Sandoval, J.; Swiss, V.; Li, J.; Dupree, J.; Franklin, R.; Casaccia-Bonnefil, P. Age-Dependent Epigenetic Control of Dif- ferentiation Inhibitors Is Critical for Remyelination Efficiency. Nat. Neurosci. 2008, 11, 1024–1034, doi:10.1038/nn.2172. Life 2021, 11, 62 27 of 30 156. Calabretta, S.; Vogel, G.; Yu, Z.; Choquet, K.; Darbelli, L.; Nicholson, T.B.; Kleinman, C.L.; Richard, S. Loss of PRMT5 Promotes PDGFRα Degradation during Oligodendrocyte Differentiation and Myelination. Dev. Cell 2018, 46, 426–440.e5, doi:10.1016/j.devcel.2018.06.025. j 157. Dai, J.; Bercury, K.K.; Jin, W.; Macklin, W.B. Olig1 Acetylation and Nuclear Export Mediate Oligoden Neurosci. 2015, 35, 15875–15893, doi:10.1523/JNEUROSCI.0882-15.2015. 158. Li, W.; Zhang, B.; Tang, J.; Cao, Q.; Wu, Y.; Wu, C.; Guo, J.; Ling, E.-A.; Liang, F. Sirtuin 2, a Mammalian Homolog of Yeast Silent Information Regulator-2 Longevity Regulator, Is an Oligodendroglial Protein That Decelerates Cell Differentiation through Deacetylating α-Tubulin. J. Neurosci. 2007, 27, 2606–2616, doi:10.1523/JNEUROSCI.4181-06.2007. y 159. Noack, M.; Leyk, J.; Richter-Landsberg, C. HDAC6 Inhibition Results in Tau Acetylation and Modula and Degradation in Oligodendrocytes. Glia 2014, 62, 535–547, doi:10.1002/glia.22624. 160. Lin, S.-T.; Fu, Y.-H. MiR-23 Regulation of Lamin B1 Is Crucial for Oligodendrocyte Development and Myelination. Dis. Model. Mech. 2009, 2, 178–188, doi:10.1242/dmm.001065. 161. Dugas, J.C.; Cuellar, T.L.; Scholze, A.; Ason, B.; Ibrahim, A.; Emery, B.; Zamanian, J.L.; Foo, L.C.; McManus, M.T.; Barres, B.A. Dicer1 and MiR-219 Are Required for Normal Oligodendrocyte Differentiation and Myelination. Neuron 2010, 65, 597–611, doi:10.1016/j.neuron.2010.01.027. j 162. He, D.; Wang, J.; Lu, Y.; Deng, Y.; Zhao, C.; Xu, L.; Chen, Y.; Hu, Y.-C.; Zhou, W.; Lu, Q.R. LncRNA Functional Networks in Oligodendrocytes Reveal Stage-Specific Myelination Control by an LncOL1/Suz12 Complex in the CNS. Neuron 2017, 93, 362– 378, doi:10.1016/j.neuron.2016.11.044. 163. References Wu, R.; Li, A.; Sun, B.; Sun, J.-G.; Zhang, J.; Zhang, T.; Chen, Y.; Xiao, Y.; Gao, Y.; Zhang, Q.; et al. A Novel m6A Reader Prrc2a Controls Oligodendroglial Specification and Myelination. Cell Res. 2019, 29, 23–41. g g p y 173. Xu, H.; Dzhashiashvili, Y.; Shah, A.; Kunjamma, R.B.; Weng, Y.; Elbaz, B.; Fei, Q.; Jones, J.S.; Li, Y.I.; Zhuang, X.; et al. M6A MRNA Methylation Is Essential for Oligodendrocyte Maturation and CNS Myelination. Neuron 2019, doi:10.1016/j.neu- ron.2019.12.013. 174. Barrera, K.; Chu, P.; Abramowitz, J.; Steger, R.; Ramos, R.L.; Brumberg, J.C. Organization of Myelin in the Mouse Somatosensory Barrel Cortex and the Effects of Sensory Deprivation. Dev. Neurobiol. 2013, 73, 297–314, doi:10.1002/dneu.22060. 175. Liu, J.; Dietz, K.; DeLoyht, J.M.; Pedre, X.; Kelkar, D.; Kaur, J.; Vialou, V.; Lobo, M.K.; Dietz, D.M.; Nestler, E.J.; et al. Impaired Adult Myelination in the Prefrontal Cortex of Socially Isolated Mice. Nat. Neurosci. 2012, 15, 1621–1623, doi:10.1038/nn.3263. 176. Makinodan, M.; Rosen, K.M.; Ito, S.; Corfas, G. A Critical Period for Social Experience–Dependent Olig and Myelination. Science 2012, 337, 1357–1360, doi:10.1126/science.1220845. y 177. Mangin, J.-M.; Li, P.; Scafidi, J.; Gallo, V. Experience-Dependent Regulation of NG2 Progenitors in the Developing Barrel Cortex. Nat. Neurosci. 2012, 15, 1192–1194, doi:10.1038/nn.3190. 178. Yang, S.M.; Michel, K.; Jokhi, V.; Nedivi, E.; Arlotta, P. Neuron Class–Specific Responses Govern Adaptive Myelin Remodeling in the Neocortex. Science 2020, 370, doi:10.1126/science.abd2109. 179. Nicholson, M.; Wood, R.J.; Fletcher, J.L.; Gonsalvez, D.G.; Hannan, A.J.; Murray, S.S.; Xiao, J. Remodeling of Pre-Existing Mye- linated Axons and Oligodendrocyte Differentiation Is Stimulated by Environmental Enrichment in the Young Adult Brain. bioRxiv 2020, doi:10.1101/2020.01.21.914572. 180. Lee, S.; Leach, M.K.; Redmond, S.A.; Chong, S.Y.; Mellon, S.H.; Tuck, S.J.; Feng, Z.Q.; Corey, J.M.; Chan, J.R. A Culture System to Study Oligodendrocyte Myelination Processes Using Engineered Nanofibers. Nat. Methods 2012, 9, 917–922, doi:10.1038/nmeth.2105. Life 2021, 11, 62 28 of 30 181. Mei, F.; Fancy, S.P.J.; Shen, Y.-A.A.; Niu, J.; Zhao, C.; Presley, B.; Miao, E.; Lee, S.; Mayoral, S.R.; Redmond, S.A.; et al. Micropillar Arrays as a High-Throughput Screening Platform for Therapeutics in Multiple Sclerosis. Nat. Med. 2014, 20, 954–960, doi:10.1038/nm.3618. / 182. Rosenberg, S.S.; Kelland, E.E.; Tokar, E.; De La Torre, A.R.; Chan, J.R. The Geometric and Spatial Constraints of the Microenvi- ronment Induce Oligodendrocyte Differentiation. Proc. Natl. Acad. Sci. USA 2008, 105, 14662–14667, doi:10.1073/pnas.0805640105. 183. Hernandez, M.; Patzig, J.; Mayoral, S.R.; Costa, K.D.; Chan, J.R.; Casaccia, P. References Mechanostimulation Promotes Nuclear and Epi- genetic Changes in Oligodendrocytes. J. Neurosci. 2016, 36, 806–813, doi:10.1523/JNEUROSCI.2873-15.2016. 184. Segel, M.; Neumann, B.; Hill, M.F.E.; Weber, I.P.; Viscomi, C.; Zhao, C.; Young, A.; Agley, C.C.; Thompson, A.J.; Gonzalez, G.A.; et al. Niche Stiffness Underlies the Ageing of Central Nervous System Progenitor Cells. Nature 2019, doi:10.1038/s41586-019- 1484-9. 185. Hartline, D.K.; Colman, D.R. Rapid Conduction and the Evolution of Giant Axons and Myelinated Fibers. Curr. Biol. 2007, 17, R29–R35, doi:10.1016/j.cub.2006.11.042. j 186. Fünfschilling, U.; Supplie, L.M.; Mahad, D.; Boretius, S.; Saab, A.S.; Edgar, J.; Brinkmann, B.G.; Kassmann, C.M.; Tzvetanova, I.D.; Möbius, W.; et al. Glycolytic Oligodendrocytes Maintain Myelin and Long-Term Axonal Integrity. Nature 2012, 485, 517– 521, doi:10.1038/nature11007. 187. Lee, Y.; Morrison, B.M.; Li, Y.; Lengacher, S.; Farah, M.H.; Hoffman, P.N.; Liu, Y.; Tsingalia, A.; Jin, L.; Zhang, P.W.; et al. Oligodendroglia Metabolically Support Axons and Contribute to Neurodegeneration. Nature 2012, 487, 443–448, doi:10.1038/na- ture11314. 188. Etxeberria, A.; Hokanson, K.C.; Dao, D.Q.; Mayoral, S.R.; Mei, F.; Redmond, S.A.; Ullian, E.M.; Chan, J.R. Dynamic Modulation of Myelination in Response to Visual Stimuli Alters Optic Nerve Conduction Velocity. J. Neurosci. 2016, 36, 6937–6948, doi:10.1523/JNEUROSCI.0908-16.2016. 189. Mitew, S.; Gobius, I.; Fenlon, L.R.; McDougall, S.J.; Hawkes, D.; Xing, Y.L.; Bujalka, H.; Gundlach, A.L.; Richards, L.J.; Kilpatrick, T.J.; et al. Pharmacogenetic Stimulation of Neuronal Activity Increases Myelination in an Axon-Specific Manner. Nat. Commun. 2018, 9, 306, doi:10.1038/s41467-017-02719-2. 190. Chacon-De-La-Rocha, I.; Fryatt, G.L.; Rivera, A.D.; Restani, L.; Caleo, M.; Raineteau, O.; Gomez-Nicola, D.; Butt, A.M. Synaptic Silencing Affects the Density and Complexity of Oligodendrocyte Precursor Cells in the Adult Mouse Hippocampus. bioRxiv 2020, doi:10.1101/2020.09.23.309682. , / 191. Mensch, S.; Baraban, M.; Almeida, R.; Czopka, T.; Ausborn, J.; El Manira, A.; Lyons, D.A. Synaptic Vesicle Release Regulates the Number of Myelin Sheaths Made by Individual Oligodendrocytes in Vivo. Nat. Neurosci. 2015, 18, 628–630, doi:10.1038/nn.3991. 192. Ortiz, F.C.; Habermacher, C.; Graciarena, M.; Houry, P.-Y.; Nishiyama, A.; Nait Oumesmar, B.; Angulo, M.C. Neuronal Activity in Vivo Enhances Functional Myelin Repair. JCI Insight 2019, 5, doi:10.1172/jci.insight.123434. 193. Steadman, P.E.; Xia, F.; Ahmed, M.; Mocle, A.J.; Penning, A.R.A.; Geraghty, A.C.; Steenland, H.W.; Monje, M.; Josselyn, S.A.; Frankland, P.W. Disruption of Oligodendrogenesis Impairs Memory Consolidation in Adult Mice. Neuron 2020, 105, 150–164.e6, doi:10.1016/j.neuron.2019.10.013. j 194. Stevens, B.; Porta, S.; Haak, L.L.; Gallo, V.; Fields, R.D. Adenosine: A Neuron-Glial Transmitter Promoting Myelination in the CNS in Response to Action Potentials. Neuron 2002, 36, 855–868, doi:10.1016/s0896-6273(02)01067-x. 195. References Chen, M.; Xu, Y.; Huang, R.; Huang, Y.; Ge, S.; Hu, B. N-Cadherin Is Involved in Neuronal Activity-Dependent Regulation of Myelinating Capacity of Zebrafish Individual Oligodendrocytes In Vivo. Mol. Neurobiol. 2016, 1–14, doi:10.1007/s12035-016- 0233-4. 208. Schnädelbach, O.; Blaschuk, O.W.; Symonds, M.; Gour, B.J.; Doherty, P.; Fawcett, J.W. N-Cadherin Influences Migration of Oligodendrocytes on Astrocyte Monolayers. Mol. Cell Neurosci. 2000, 15, 288–302, doi:10.1006/mcne.1999.0819. 209. Venkatesh, H.S.; Johung, T.B.; Caretti, V.; Noll, A.; Tang, Y.; Nagaraja, S.; Gibson, E.M.; Mount, C.W.; Polepalli, J.; Mitra, S.S.; et al. Neuronal Activity Promotes Glioma Growth through Neuroligin-3 Secretion. Cell 2015, 161, 803–816, doi:10.1016/j.cell.2015.04.012. j 210. Krasnow, A.M.; Ford, M.C.; Valdivia, L.E.; Wilson, S.W.; Attwell, D. Regulation of Developing Myelin Sheath Elongation by Oligodendrocyte Calcium Transients in Vivo. Nat. Neurosci. 2018, 21, 24, doi:10.1038/s41593-017-0031-y. 211. Butt, A.M.; Fern, R.F.; Matute, C. Neurotransmitter Signaling in White Matter. Glia 2014, 62, 1762–1779, doi:10.1002/glia.22674. 212. Fields, R.D. Imaging Single Photons and Intrinsic Optical Signals for Studies of Vesicular and Non-Vesicular ATP Release from Axons. Front. Neuroanat. 2011, 5, doi:10.3389/fnana.2011.00032. , ; , ; , g g , , , g 212. Fields, R.D. Imaging Single Photons and Intrinsic Optical Signals for Studies of Vesicular and Non-Vesicular ATP Release from Axons. Front. Neuroanat. 2011, 5, doi:10.3389/fnana.2011.00032. , , 213. Fields, R.D. Nonsynaptic and Nonvesicular ATP Release from Neurons and Relevance to Neuron–Glia Signaling. Semin. Cell Dev. Biol. 2011, 22, 214–219, doi:10.1016/j.semcdb.2011.02.009. 214. Fields, R.D.; Ni, Y. Nonsynaptic Communication through ATP Release from Volume-Activated Anion Channels in Axons. Sci. Signal. 2010, 3, ra73, doi:10.1126/scisignal.2001128. g g 215. Hrvatin, S.; Hochbaum, D.R.; Nagy, M.A.; Cicconet, M.; Robertson, K.; Cheadle, L.; Zilionis, R.; Ratner, A.; Borges-Monroy, R.; Klein, A.M.; et al. Single-Cell Analysis of Experience-Dependent Transcriptomic States in Mouse Visual Cortex. Nat. Neurosci. 2018, 21, 120–129, doi:10.1038/s41593-017-0029-5. 216. Hu, P.; Fabyanic, E.; Kwon, D.; Tang, S.; Zhou, Z.; Wu, H. Dissecting Cell-Type Composition and Activity-Dependent Tran- scriptional State in Mammalian Brains by Massively Parallel Single-Nucleus RNA-Seq. Mol. Cell 2017, 68, 1006–1015.e7, doi:10.1016/j.molcel.2017.11.017. j 217. Coppieters, N.; Dieriks, B.V.; Lill, C.; Faull, R.L.M.; Curtis, M.A.; Dragunow, M. Global Changes in DNA Methylation and Hydroxymethylation in Alzheimer’s Disease Human Brain. Neurobiol. Aging 2014, 35, 1334–1344, doi:10.1016/j.neurobiolag- ing.2013.11.031. g 218. Girdhar, K.; Hoffman, G.E.; Jiang, Y.; Brown, L.; Kundakovic, M.; Hauberg, M.E.; Francoeur, N.J.; Wang, Y.; Shah, H.; Kavanagh, D.H.; et al. References Wake, H.; Lee, P.R.; Fields, R.D. Control of Local Protein Synthesis and Initial Events in Myelination by Action Potentials. Science 2011, 333, 1647–1651, doi:10.1126/science.1206998. 196. Wake, H.; Ortiz, F.C.; Woo, D.H.; Lee, P.R.; Angulo, M.C.; Fields, R.D. Nonsynaptic Junctions on Myelinating Glia Promote Preferential Myelination of Electrically Active Axons. Nat. Commun. 2015, 6, 7844, doi:10.1038/ncomms8844. 197. Chen, T.-J.; Kula, B.; Nagy, B.; Barzan, R.; Gall, A.; Ehrlich, I.; Kukley, M. In Vivo Regulation of Oligodendrocyte Precursor Cell Proliferation and Differentiation by the AMPA-Receptor Subunit GluA2. Cell Rep. 2018, 25, 852–861.e7, doi:10.1016/j.celrep.2018.09.066. j p 198. Swire, M.; Kotelevtsev, Y.; Webb, D.J.; Lyons, D.A.; Ffrench-Constant, C. Endothelin Signalling Mediates Experience-Dependent Myelination in the CNS. eLife 2019, 8, e49493, doi:10.7554/eLife.49493. y f 199. Zhao, Y.-Y.; Shi, X.-Y.; Qiu, X.; Lu, W.; Yang, S.; Li, C.; Chen, L.; Zhang, L.; Cheng, G.-H.; Tang, Y Increases the Myelinated Nerve Fibers of Aged Rat Corpus Callosum. Anat. Rec. 2012, 295, 999–1005, d Increases the Myelinated Nerve Fibers of Aged Rat Corpus Callosum. Anat. Rec. 2012, 295, 999–1005, doi:10.1002/ar.22446. 200. Sampaio-Baptista, C.; Khrapitchev, A.A.; Foxley, S.; Schlagheck, T.; Scholz, J.; Jbabdi, S.; DeLuca, G.C.; Miller, K.L.; Taylor, A.; Thomas, N.; et al. Motor Skill Learning Induces Changes in White Matter Microstructure and Myelination. J. Neurosci. 2013, 33, 19499–19503, doi:10.1523/JNEUROSCI.3048-13.2013. 201. Xiao, L.; Ohayon, D.; McKenzie, I.A.; Sinclair-Wilson, A.; Wright, J.L.; Fudge, A.D.; Emery, B.; Li, H.; Richardson, W.D. Rapid Production of New Oligodendrocytes Is Required in the Earliest Stages of Motor-Skill Learning. Nat. Neurosci. 2016, doi:10.1038/nn.4351. 202. Stedehouder, J.; Brizee, D.; Shpak, G.; Kushner, S.A. Activity-Dependent Myelination of Parvalbumin Interneurons Mediated by Axonal Morphological Plasticity. J. Neurosci. 2018, 38, 3631–3642, doi:10.1523/JNEUROSCI.0074-18.2018. 203. Hines, J.H.; Ravanelli, A.M.; Schwindt, R.; Scott, E.K.; Appel, B. Neuronal Activity Biases Axon Selection for Myelination in Vivo. Nat. Neurosci. 2015, 18, 683–689, doi:10.1038/nn.3992. Life 2021, 11, 62 29 of 30 29 of 30 204. Bechler, M.E.; Byrne, L.; Ffrench-Constant, C. CNS Myelin Sheath Lengths Are an Intrinsic Property of Oligodendrocytes. Curr. Biol. CB 2015, doi:10.1016/j.cub.2015.07.056. j 205. Lee, S.; Chong, S.Y.; Tuck, S.J.; Corey, J.M.; Chan, J.R. A Rapid and Reproducible Assay for Modeling Myelination by Oligoden- drocytes Using Engineered Nanofibers. Nat. Protoc. 2013, 8, 771–782. y g g 206. Piscopo, D.M.; Weible, A.P.; Rothbart, M.K.; Posner, M.I.; Niell, C.M. Changes in White Matter in Mice Resulting from Low- Frequency Brain Stimulation. Proc. Natl. Acad. Sci. USA 2018, 115, E6339–E6346, doi:10.1073/pnas.1802160115. 207. References Cell-Specific Histone Modification Maps in the Human Frontal Lobe Link Schizophrenia Risk to the Neuronal Epige- nome. Nat. Neurosci. 2018, 21, 1126–1136, doi:10.1038/s41593-018-0187-0. 219. Huynh, J.L.; Garg, P.; Thin, T.H.; Yoo, S.; Dutta, R.; Trapp, B.D.; Haroutunian, V.; Zhu, J.; Donovan, M.J.; Sharp, A.J.; et al. Epigenome-Wide Differences in Pathology-Free Regions of Multiple Sclerosis-Affected Brains. Nat. Neurosci. 2014, 17, 121–130, doi:10.1038/nn.3588. 220. Jaffe, A.E.; Gao, Y.; Deep-Soboslay, A.; Tao, R.; Hyde, T.M.; Weinberger, D.R.; Kleinman, J.E. Mapping DNA Methylation across Development, Genotype and Schizophrenia in the Human Frontal Cortex. Nat. Neurosci. 2015, advance online publication, doi:10.1038/nn.4181. 221. Semick, S.A.; Bharadwaj, R.A.; Collado-Torres, L.; Tao, R.; Shin, J.H.; Deep-Soboslay, A.; Weiss, J.R.; Weinberger, D.R.; Hyde, T.M.; Kleinman, J.E.; et al. Integrated DNA Methylation and Gene Expression Profiling across Multiple Brain Regions Implicate Novel Genes in Alzheimer’s Disease. Acta Neuropathol. 2019, doi:10.1007/s00401-019-01966-5. p 222. Arenkiel, B.R.; Peca, J.; Davison, I.G.; Feliciano, C.; Deisseroth, K.; Augustine, G.J.; Ehlers, M.D.; Feng, G. In Vivo Light-Induced Activation of Neural Circuitry in Transgenic Mice Expressing Channelrhodopsin-2. Neuron 2007, 54, 205–218, doi:10.1016/j.neu- ron.2007.03.005. D.K.; Lambe, E.K. Chronic Social Isolation Reduces 5-HT Neuronal Activity via Upregulated SK3 Calcium- m Channels. eLife 2016, 5, e21416, doi:10.7554/eLife.21416. 223. Sargin, D.; Oliver, D.K.; Lambe, E.K. Chronic Social Isolation Reduces 5-HT Neuronal Activity via Activated Potassium Channels. eLife 2016, 5, e21416, doi:10.7554/eLife.21416. 224. Bonnefil, V.; Dietz, K.; Amatruda, M.; Wentling, M.; Aubry, A.V.; Dupree, J.L.; Temple, G.; Park, H.-J.; Burghardt, N.S.; Casaccia, P.; et al. Region-Specific Myelin Differences Define Behavioral Consequences of Chronic Social Defeat Stress in Mice. eLife 2019, 8, e40855, doi:10.7554/eLife.40855. 225. Jäkel, S.; Agirre, E.; Falcão, A.M.; van Bruggen, D.; Lee, K.W.; Knuesel, I.; Malhotra, D.; ffrench-Constant, C.; Williams, A.; Castelo-Branco, G. Altered Human Oligodendrocyte Heterogeneity in Multiple Sclerosis. Nature 2019, 566, 543–547, doi:10.1038/s41586-019-0903-2. 226. Marisca, R.; Hoche, T.; Agirre, E.; Hoodless, L.J.; Barkey, W.; Auer, F.; Castelo-Branco, G.; Czopka, T. Functionally Distinct Subgroups of Oligodendrocyte Precursor Cells Integrate Neural Activity and Execute Myelin Formation. Nat. Neurosci. 2020, 23, 363–374, doi:10.1038/s41593-019-0581-2. 30 of 30 Life 2021, 11, 62 30 of 30 227. Marques, S.; Zeisel, A.; Codeluppi, S.; van Bruggen, D.; Falcão, A.M.; Xiao, L.; Li, H.; Häring, M.; Hochgerner, H.; Romanov, R.A.; et al. Oligodendrocyte Heterogeneity in the Mouse Juvenile and Adult Central Nervous System. Science 2016, 352, 1326– 1329, doi:10.1126/science.aaf6463. 228. Maag, J.L.V.; Kaczorowski, D.C.; Panja, D.; Peters, T.J.; Bramham, C.R.; Wibrand, K.; Dinger, M.E. References Widespread Promoter Meth- ylation of Synaptic Plasticity Genes in Long-Term Potentiation in the Adult Brain in Vivo. BMC Genom. 2017, 18, 250, doi:10.1186/s12864-017-3621-x. 229. Sui, L.; Wang, Y.; Ju, L.-H.; Chen, M. Epigenetic Regulation of Reelin and Brain-Derived Neurotrophic Factor Genes in Long- Term Potentiation in Rat Medial Prefrontal Cortex. Neurobiol. Learn. Mem. 2012, 97, 425–440, doi:10.1016/j.nlm.2012.03.007. 230. Chawla, S.; Vanhoutte, P.; Arnold, F.J.L.; Huang, C.L.-H.; Bading, H. Neuronal Activity-Dependent N tling of HDAC4 and HDAC5. J. Neurochem. 2003, 85, 151–159, doi:10.1046/j.1471-4159.2003.01648.x. 231. He, X.-B.; Yi, S.-H.; Rhee, Y.-H.; Kim, H.; Han, Y.-M.; Lee, S.-H.; Lee, H.; Park, C.-H.; Lee, Y.-S.; Richardson, E.; et al. Prolonged Membrane Depolarization Enhances Midbrain Dopamine Neuron Differentiation via Epigenetic Histone Modifications. Stem Cells 2011, 29, 1861–1873, doi:10.1002/stem.739. 232. Martinowich, K.; Hattori, D.; Wu, H.; Fouse, S.; He, F.; Hu, Y.; Fan, G.; Sun, Y.E. DNA Methylation-Re eling in Activity-Dependent Bdnf Gene Regulation. Science 2003, 302, 890–893, doi:10.1126/science.109 g y p g 233. Sharma, R.P.; Tun, N.; Grayson, D.R. Depolarization Induces Downregulation of DNMT1 and DNMT3 in Primary Cortical Cultures. Epigenetics 2008, 3, 74–80, doi:10.4161/epi.3.2.6103. p g p 234. Berger, T.C.; Vigeland, M.D.; Hjorthaug, H.S.; Etholm, L.; Nome, C.G.; Taubøll, E.; Heuser, K.; Selmer, K.K. Neuronal and Glial DNA Methylation and Gene Expression Changes in Early Epileptogenesis. PLoS ONE 2019, 14, doi:10.1371/jour- nal.pone.0226575. p Guo, J.U.; Ma, D.K.; Mo, H.; Ball, M.P.; Jang, M.-H.; Bonaguidi, M.A.; Balazer, J.A.; Eaves, H.L.; Xie, B.; Fo Activity Modifies DNA Methylation Landscape in the Adult Brain. Nat. Neurosci. 2011, 14, 1345–1351, doi 236. Berry, K.; Wang, J.; Lu, Q.R. Epigenetic Regulation of Oligodendrocyte Myelination in Developmental Disorders and Neuro- degenerative Diseases. F1000Research 2020, 9, 105, doi:10.12688/f1000research.20904.1. g 237. Samudyata; Castelo-Branco, G.; Liu, J. Epigenetic Regulation of Oligodendrocyte Differentiation: From Development to Demy- elinating Disorders. Glia 2020, doi:10.1002/glia.23820. 238. Tiane, A.; Schepers, M.; Rombaut, B.; Hupperts, R.; Prickaerts, J.; Hellings, N.; van den Hove, D.; Vanmierlo, T. From OPC to Oligodendrocyte: An Epigenetic Journey. Cells 2019, 8, 1236, doi:10.3390/cells8101236.
https://openalex.org/W4310095828
https://econjournals.com/index.php/ijeep/article/download/13678/7023
English
null
Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach
International journal of energy economics and policy
2,022
cc-by
8,959
Accepted: 28 October 2022 Received: 15 August 2022 ABSTRACT Mining plays a very substantial role in the economy of South Africa. This sector remains to be amongst the greatest consumers of energy in the country, especially electricity. It is against these background that the main objective of this study is to determine the effects of energy pricing on the mining sector performance in South Africa from 1990 to 2019. Johansen and ADRL procedures were used to determine the effects of energy pricing on the mining sector performance in South Africa. The results of the Error correction model (ECM) under Johansen cointegration technique are negative as expected but they are not statistically significant. The study error term is -0.122975, which meant that cointegration relationship is established. The results show that capital stock and labour play an important role in balancing mining productivity, while energy prices, Gross Domestic Product and import prices, on the other hand, play a lesser role in balancing mining productivity. The results of the Error correction model (ECM) under ARDL designate the short-run coefficient for D (LNCT) and D (GDP2) are statistically significant at 1% level and the coefficient of error correction term ecm (-1) valued at -1.037410 is negative and highly significant signifying that in the short-run changes in Capital stock and Gross Domestic Product are associated with mining production. The long-run relationship is illustrated by a negative sign on the coefficient of the ECT. Policy implication of a long-run optimistic relationship between electricity pricing and Mining Sector productivity is that mines should invest on producing their own energy, in that way they will be responsible for reducing their costs and that could results in mines increasing their productive capacity in the long-term. Keywords: Energy Pricing, Mining Sector Performance, Autoregressive Distributed Lag, South Africa JEL Classifications: D04, C32, Q47, Q42, Q01 Department of Economics, North-West University, South Africa. *Email: hlalefangk@gmail.com Accepted: 28 October 2022 DOI: https://doi.org/10.32479/ijeep.13678 DOI: https://doi.org/10.32479/ijeep.13678 This Journal is licensed under a Creative Commons Attribution 4.0 International License Keywords: Energy Pricing, Mining Sector Performance, Autoregressive Distributed Lag, South Africa JEL Classifications: D04, C32, Q47, Q42, Q01 Tumelo Mmutle, Olebogeng David Daw, Hlalefang Khobai* Department of Economics, North-West University, South Africa. *Email: hlalefangk@gmail.com 2. LITERATURE REVIEW Kirikkaleli et al. (2021) Toda Yamamoto used causality and wavelet coherence tests to analyze the causal relationship between nuclear energy use and growth in the UK from 1998 to 2017. The study concluded that changes in growth will lead to changes in nuclear energy consumption in the UK over the long term. In the short term, between 2002 and 2006, there was a positive correlation between nuclear energy use and growth. 2.1. Empirical Studies of Developed Economies Rybak and Rybak (2016) researched Poland’s coal mining on possible strategies for hard coal mining because of Production Function analysis. They used productivity, the substitution of production factors and marginal productivity as their selected indicators. They compared the 2006 analysis done by Przybyla and Rybak (2008) with their 2010 results, and the comparison revealed that within 4 years, aggregate production (AE) decreased by 50%, marginal productivity (AE and PR) declined three times and the absence of economies of scale still occurred. 2.2. Empirical Studies of Developing Economies Agheli-Kohneshahri (2006) projected Iran’s production function. He used panel data of the mining industry in several provinces of Iran throughout 1996-2002. To estimate the studies, he used logarithmic Cobb-Douglas, Transcendental and Translog production functions. This was built on the estimates piloted by the Pooled Least Square (PLS) as well as Generalized Least Square (GLS). He discovered that the mines in Iran are more labour intensive and that the return to scale has marginally been bigger than 1. This meant that cumulative returns to scale in Iran’s mining exist. Lundberg (2009) developed the market feature for Swedish industrial energy usage as well as shifts in demand patterns over time by splitting the study into two phases (1960-1992 and 1993-2002). In his results, he found that production was a more substantial factor in the first sample, while the price in the second sample became more trivial. The more effective use of electricity in the second phase was a conceivable reason for this variation. Ghaderi et al. (2006) investigated the role of the energy demand market in Iran. While Egorova and Volchkova (2004) discovered that electricity prices were a factor in energy consumption, while other variables, such as industrial production, proved to be more important, carried out similar sectoral study of Russian industries. Kamin՛ski (2008) conducted a study on the impact of liberalisation of the electricity market on the hard coal mining sector in Poland. 1. INTRODUCTION by Oshikoya and Hussain (2001) indicate that the issue of unstable and high cost disruption of power supply leads to inefficiency in output and affects long-term growth and efficiency, as is the case in most African countries. South Africa is classified as a developing country and one of the most industrialized countries in Africa. The importance of energy towards the production phases cannot be ignored, as it plays a very essential part in the supply chain because it serves as an input in the production stages of businesses also as a final good for end-users. South Africa’s energy production and distribution competencies are extremely complex. “Minerals commodities continued to be the foundation of the South African economy throughout the 1990s and into the 21st century” (Ziramba, 2009). The world today is facing many energy challenges such as power outages experienced internationally and in South Africa. The energy demand is constantly increasing. This sentiment is echoed The energy sector is a critical vehicle in ensuring that the growth trajectory targeted is achievable and it is critical to the betterment of the lives of poor South Africans. Electricity infrastructure consists of generation, transmission and distribution. In terms of generation, Eskom (state-owned utility) is the major electricity Generation Sector in South Africa (Kohler, 2013). As stated by Creamer, Naidoo and Tyrer (2006), Eskom produces roughly 90% of the power consumed in South Africa from 26 power stations (the Eskom generation mix). Municipalities and redistributors, as 283 Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach Dagoumas et al. (2020) examined the relationship between energy prices and growth in Europe from 1990 to 2018 using the Engle Granger method to estimate annual data and using the VECM. The study found a causality between the crude oil price and the industrial electricity price to the electricity price for private households. The results also indicate that an increase in electricity prices would not negatively affect European growth rates. well as a private generator, produce the remaining 10%. Eskom is ranked the eleventh largest power utility globally, in terms of generating ability. It is also rated number nine in terms of sales, and boast the world’s largest cooling power station. It also owns and operates numerous power stations (coal, gas hydro and pumped storage and nuclear) (DoE, 2018). 2. LITERATURE REVIEW As a reference model, MON (integrated and centralized monopoly power sector structure) and LIB (fully liberalized electricity market) were used. Kamin՛ski (2008) discovered that the LIB power plant, which is independent of political control, no longer wanted to import coal from domestic sources because it was costly and that saved them money relative to the MON power plants. Results have shown that the demand for hard coal for electricity generation in the MON scenario is much higher, particularly after 2008; the discrepancies between scenarios are between 1, 5 and 3 Mt per year until 2020. The overall demand gap for the entire studied span is 31 Mt. Hosking and Keseke (2012) carried out an analysis on the cost of the discharge of electricity to mines in Zimbabwe using a Straightforward Assessment Method. One clear finding of the report is that the weak state of power supply in Zimbabwe has cost the mining industry tremendously. Lin and Zhu (2021) conducted a study on the energy efficiency of the mining sector in China. The two-way fixed effects model and the threshold panel model were used to calculate the marginal effect of industrial agglomeration on the total factor energy efficiency (TFEE) of the mining sector. Furthermore, the threshold panel model is used to calculate the influence of industrial agglomeration in regions with different levels of economic development on the energy efficiency of mining. The results show that increasing industrial agglomeration improves the energy efficiency of the mining sector. Katta et al. (2020) developed a study of the disaggregated energy use and greenhouse gas (GHG) emission footprint for Canada’s iron, gold, and potash mining sectors. The study found that bottom-up energy demand trees were developed for iron, gold, and potash mining. The energy intensities for each end-user were calculated and used in a bottom-up energy-environmental model to determine the associated GHG emissions of the end-use process using Sankey diagrams. The results revealed that the total energy and GHG emission intensities for iron, gold, and potash mining were 0.7, 149.8, 1.8 GJ/Mg and 33, 4922, 158 kg CO2eq/Mg, respectively. In iron mining, firing had the highest GHG emission share of 66%, in gold mining, ore transport had the highest GHG emission share of 22%, and 34% of potash mining GHG emissions came from product drying and steam generation. 2.3. Empirical Studies focused on South Africaf Little to no research have been done on the effects of electricity prices on the production of the mining industry in South Africa and only a handful have been done on power use and industrial development or economic development. Studies in South Africa include, but are not limited to, Szcygielski, Ensil and du Toit International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022 Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach (2018), Inglesi-Lotz and Pouris (2012), Kohler (2013) and Inglesi- Lotz and Blignaut (2011). for electricity. A  differential tariff system will increase the cost of energy inefficiency and promote the re-optimization of manufacturing processes so that more material inputs and less energy inputs will be required in energy-intensive industries. Szcygielski, Ensil and du Toit (2018) investigated a study on investment in gold mining stocks and its linkage to the gold price, where they looked at its benefits to investors in a developing economy. They applied a regression analysis tool to investigate the relationship between gold mining proceeds, the gold mining price and the exchange of rand-dollar using a multifactor model inspired by the arbitrage pricing theory. They discovered that there is a resilient, yet changing relationship between gold price, gold mining proceeds and the rand-dollar exchange. Part of their recommendation was that more research into individual gold betas within the ten mining companies must be conducted to achieve more results. Gonese et al. (2019) examined the impact of electricity prices on production at a sectoral level in South Africa from 1994 to 2015 using the generalized least squares method. The results show that electricity prices have a negative impact on production. Takentsi et al. (2022) investigated the causal relationship between energy prices and economic performance in South Africa using the Auto-Regressive Distributed Lag (ARDL) Bounds test technique for the period 1994 to 2019. The study established a long-term relationship between the variables. The results showed that electricity prices have a significant negative impact on economic growth in the short and long term, while crude oil prices have a significant positive correlation with economic growth in the long and short term. The investigation by Inglesi-Lotz and Pouris (2012) highlights the crucial need to increase energy production in the South African industry in the light of the high percentage of overall electricity usage in the sector. 3. METHODOLOGY To investigate the correlation between energy pricing and mining sector performance in South Africa, the study adopted Johanson cointegration technique and Autoregressive Distributed-Lag (ADRL) methods for analysis purpose. Selected variables used in this study are energy (electricity) pricing, labour (mining employment), capital stock, Gross Domestic Product (mining sector contribution), import prices, mining sector output (performance). The research used both formal and informal methods for stationarity analysis. Informal methods provide graphical presentation, while formal methods include all statistical analyses. Lastly, the Granger-Causality technique has been used to assess the causality of the variables. Inglesi-Lots and Blignaut (2011) examined the South African economic sector’s electricity consumption in response to fluctuations in electricity prices and economic output for the period 1993-2006. The study adopted a panel data analysis, and results found that the industrial sector was the only sector over that study period with statistically significant price elasticity. In addition, economic production contributed favorably to the commercial and industrial sectors (with strong and significant coefficients). This was in disparity with the other three industries, mining, agriculture and transport, whose electricity use was not influenced either by output or by their expense. 2.3. Empirical Studies focused on South Africaf According to Fawkes (2005), energy production developments are opening up to South African firms to raise revenues, improve environmental enforcement, to some degree alleviate competition from competitive suppliers and help resolve capital expenditure constraints. Industrial producers in the country should, as such, support the enhancement of electricity production. 3.3. Unit Root εt = Error term There are many ways of checking for unit root. These include the Durbin-Watson (DW) test, the Dickey-Fuller (DF) test (1979), the Augmented-Dickey Fuller (ADF) test (1979), the Phillip Perron (PP) test (1988) and the Kwiatkowski et al. (1992), among others. Nkoro and Uko (2016) define non-stationary time series as a stochastic practice with unit-roots or structural disturbances. Conversely, unit-roots is a primary source of non-stationarity. The presence of unit roots implies that a time series under reflection is not stationary, although the absence of it allows a time series to be stationary. The expected relationship between Capital stock and Mining performance is positive. This notion to use capital stock to explain mining performance is based on the assumption that as more capital stock increased these may translate into more output of performance (Cobb and Douglas, 1928). Labour was also considered to be in this model, therefore its expected relationship with mining performance is positive. The theory posed that an increase in labour will increase output in mining performance. There is an expected negative relationship between electricity pricing and mining performance. This relationship is viewed in a sense that electricity as an input in mining activity. Therefore, as input prices goes up this might translate in a sabotage in mining output. This is consistent with Adebola (2011); Bildirici et al. (2013); Masuduzzaman (2013); Ciarreta and Zarranga (2010) and Kohler (2013), who substantiate that an increases in electricity price have an unfavorable effect on productivity. This depicts that the root unit is one of the sources of non- stationarity. Trend Stationary (deterministic) Process (TSP) or Difference Stationary Process (DSP) can be a non-stationary stochastic system. A time series is said to be a steady-state pattern if the trend is completely linear and not variable where, if not constant, it is considered an integrated or stochastic difference trend. In the case of a deterministic trend, the divergence from the original value (non-stationary mean bodies) is fairly arbitrary and perishes rapidly. They may not influence or contribute to the long-term progress of the time series. The gross domestic product of a country is one of the main indicators used to measure performance of a country, therefore its expected relationship with mining performance is positive. According to Fedderke and Pirouz (2002), mining sector in particular the gold sector, remains the most essential foreign exchange earner for South African economy. 3.1. Model Specification To estimate a long-run relationship between energy pricing and mining sector performance in South Africa. The study adopts and modify a model used by Kohler (2013) on differential electricity pricing and energy efficiency in South Africa. Gonese et al. (2019) focused on electricity price impacts on aggregate GDP rather than on mining productivity and Kohler (2013) chose electricity consumption and electricity pricing as variables. Kohler (2013) has published a report on the differential prices of electricity and energy production in South Africa, with a specific focus on energy-intensive manufacturing and mines. The author claims that by applying a differential pricing strategy, the authorities will target energy-intensive industries by charging certain higher tariffs (as implemented in China after 2004) in order to improve output efficiency and lower overall demand for electricity. A  differential tariff system will increase the cost of energy inefficiency and promote the re-optimization of manufacturing processes so that more material inputs and less energy inputs will be required in energy-intensive industries. The literature on the relationship between electricity and output has largely focused on the direction of the causality between electricity consumption on industrial sectors rather than the electricity prices on Mining productivity. The model used in this study augments a Cobb-Douglas production function using the following variables: (Electricity price, mining employment (represented by labour), Capital Stock representing the machinery used by mines in the productions process, mining as a contribution to GDP at constant prices, import prices represents the prices of imported inputs in mining production and mining sector performance as the dependent variables). Kohler (2013) has published a report on the differential prices of electricity and energy production in South Africa, with a specific focus on energy-intensive manufacturing and mines. The author claims that by applying a differential pricing strategy, the authorities will target energy-intensive industries by charging certain higher tariffs (as implemented in China after 2004) in order to improve output efficiency and lower overall demand The implementation of electricity pricing relates with the research by Kohler (2013) and Gonese, Hompashe and Sithole (2019) Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach which also looked at the causality between electricity prices and sectoral outputs in South Africa. 3.3. Unit Root Various studies Atinay and Karagol, 2005; Blignaut et al., 2015; Goosh 2002; and Polemis and Dagoumas, 2013 submit that energy costs have an insignificant effect on the gross domestic product of South Africa. 3.1. Model Specification Capital stock and import prices are used in the model in such a way as to prevent relevant variables which, according to Gujarati and Porter (2009), may contribute to a bias in the outcome. According to Gonese, Hompashe and Sibanda (2019), there exists a relationship between electricity (energy) prices and production. The empirical model can be defined as follows: if the rand depreciate against foreign currency it will be expensive to buy machinery. A positive and significant impact of imports is consistent with Rahardja and Varela (2015) who mentioned that sectors import for various reasons such as quality, variety and value. The accessibility of imported inputs has contributed to improved product quality in the Indonesian Manufacturing sector. This entails that production can be improved through imports of cheaper quality machinery and advance technology. M f CT LAB EP GDP IMP p        ( , , , , ) /  (1) where MP = Mining sector performance CT = Capital stock LAB = Labour EP = Electricity prices GDP = Gross Domestic Prices IMP = Imported Prices where MP = Mining sector performance M CT LAB EP GDP IMP p t               0 1 2 3 4 5  (2) 3.5. Granger-causality h l f h l i The last step of the analysis is to observe the causality relationship amongst the variables. In economics, causality is well-defined as the ability of one variable to forecast the other. The study is focused particularly on two time series variables, which concentrated on analyzing the bilateral causality. The ECM specification is employed to test the causality as it guarantees the consistency of the variables. The null hypothesis to be confirmed here is that there is no existence between two variables of Granger causality. The Granger Causality Test is as follows: The trace test in Table 3.1 specifies that there are three cointegrating vectors, whereas Table 3.2 specifies that there is one cointegrating vector. Banerjee et al. (1993) specified that in a situation of there is different value of two tests, the results attained from Maximal Eigenvalue for stochastic matrix will be chosen. X X Y i i n x i t i n x i t x t             1 1 1 1 , , ,  (3) Y Y Y i i n y i t i n y i t y t             1 1 1 1 , , ,  (4) Akanbi (2014) mentioned that if we are faced in a conflicting cointegration set-up, the maximum eigenvalue test is accepted when estimating the error correction model as it has a sharper marginal hypothesis that pins down the number of cointegrating vectors. This study decided to choose one cointegrating equation since it makes economic sense. This provides an allowance to evaluate a long-term relationship and Error Correction Model (ECM). The long term equilibrium vector is estimated at Z=LNMP + 1.49 LNCT – 0.278 LNEP – 0.135LNIMP – 0.152 LNGDP2 + 0.447 LNLAB. The coefficient of LNCT, LNEP, LNIM, LNGDP2 and LNLAB are significant at 0.17, 0.03, 0.02, 0.02 and 0.06 respectively. (3) (4) where Xt is the first variable log and time t and Yt is the log of the second variable at time t. µx, t and µy, t are the white noise error terms at time t. αx, I is the past value parameter of value X, which explains to us how much past value of X describes the current value of X and βx,i is the past parameter of value Y, which explain to us how much past value of Y describes the current value of X. 3.4. Co-integration Test Granger (1981) and Engle and Granger (1987) initially developed the idea of co-integration. It is an econometric procedure that explores the correction of non-stationary variables. If two or more series are non-stationary, but the linear structure of them is stationary, then the series is co-integrated. According to Nkoro and Uko (2016), the cointegration test shows how time series, which can be individually non-stationary and drift far away from balance, can be coupled in such a way that the function of the balance powers certifies that they do not drift too far apart. Cointegration testing is a key step in the establishment From the import prices the relationship is expected to be either positive or negative, as the rand appreciate against foreign currencies it will be cheaper to buy input such as machinery and International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022 Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach 4.2. Co-integration Before testing for cointegration, lag selection test must be chosen. Before testing for cointegration, lag selection test must be chosen. The results demonstrated in Table 2 evidently indicate that VAR (1) is the most applicable. The LR test, SIC and HQ all selected one lag while FPE and AIC selected two lags. With VAR order having been recognised, Johansen’s (1988) test for cointegration can be applied. For this study 1 lag is chosen because 2 lags gave inconclusive results. 4.1. Unit Root Tests The Unit root test for stationarity presented in Table 1 shows that variable are I (1), which suggests that the series are non-stationary and they only become stationary after being differenced once except for LNEP which became stationary in level under ADF. Both LnCT and LnEP are not stationary at first difference under ADF and PP. They only became stationary at levels under KPSS. Variables are I (1) under KPSS, which suggests that the study rejects the null hypothesis. The null hypothesis of unit root is rejected when the test statistic is more negative than the critical value. Mostly, if two variables which are I (1) are regression and correlation, the combination will be I (1) as well. There are several methods to co-integration testing, including Autoregressive Distributed-Lag (ADRL), Engle-Granger, Engle Yoo and Johansen, among others. When a cointegrating vector exists, the method of cointegration of Johansen and Juselius (1990) cannot be implemented. It is therefore imperative to discover Pesaran and Shin (1995) and Pesaran et al. (1996) proposed an ADRL cointegration approach for a long-term relationship, regardless of whether the variables are I(0), I(1) or a combination of both. The research will therefore use the ADRL to provide practical and effective predictions, as well as the Johansen (1998) approach to the cointegration test. 4. FINDINGS OF THE STUDY if the model shows meaningful long-term relationships in an empirical way. Brooks (2008) stated that cointegration contracts with relationships between a set of variables where each has a root unit. 3.5. Granger-causality h l f h l i Source: own calculations Trace test indicates three (3) cointegrating eng (s) at the 0.05 level,* denotes rejection of hypothesis at the 0.05 level, **Mackinnon‑Haug‑Michelis (1999) P-values Table 3.2: Unrestricted cointegrated rank test (maximum eigenvalue) Hypothesized No. of CE (s) Eigen value Max‑Eigen Critical Value Probability** Statistic None* 0.779571 42.34109 40.07757 0.0273 At most 1 0.647251 29.17599 33.87687 0.1644 At most 2 0.576870 24.08215 27.58434 0.1319 At most 3 0.403664 14.47462 21.13162 0.3274 At most 4 0.316085 10.63782 14.26460 0.1733 At most 5 0.069743 2.024245 3.841466 0.1548 Source: own calculations Trace test indicates one (1) cointegrating eng (s) at the 0.05 level, *denotes rejection of hypothesis at the 0.05 level, **Mackinnon‑Haug‑Michelis (1999) P-values Table 3.2: Unrestricted cointegrated rank test (maximum eigenvalue) Source: own calculations Trace test indicates one (1) cointegrating eng (s) at the 0.05 level, *denotes rejection of hypothesis at the 0.05 level, **Mackinnon‑Haug‑Michelis (1999) P-values There was a consistent relationship between LMNP, LNCT, LNEP, LNIMP, LNGDP2 and LNLAB between 1990 and 2019. It can be inferred that there is a long-term relationship between variables, but that they could drift apart in the short run. The analysis will use residues from the long-term relationship to examine the short-term correction. The coefficients with negative figures suggest a stable long- term relationship between the dependent variable and its explanatory variables. Cointegration equation one reflects a favorable but not significant relationship between mining efficiency, capital stock and labour, whereas energy prices, Gross Domestic Product and import prices show a negative relationship. The results show that capital stock and labour play an important role in balancing mining productivity, while energy prices, Gross Domestic Product and import prices, on the other hand, play a lesser role in balancing mining productivity. Since the null hypothesis of no long-run cointegration relationship amongst variables has been rejected, the study estimates the ARDL model using univariate ARDL cointegration test selection with the maximum lag m=3. The ARDL model specifications nominated based on Akaike Information Criterion (AIC) and Schwarz Bayesian Criterion are the same. The ARDL estimates for these models are presented in Table 6 below. Table 7 presents the estimated coefficients of the long run relationship are significant for LNCT, LNGDP2 and LNIMP and are not significant for LNEP and LNLAB. The estimated coefficients are positive for LNCT, LNGDP2 and LNIMP and negative for LNEP and LNLAB. 3.5. Granger-causality h l f h l i Related explanations are used to αy,i and βy,i. Table 1: Unit root tests Variables Model ADF PP KPSS Levels Difference Levels Difference Levels Difference LnMp Intercept −2.018 −5.815* −2.107 −5.815* 0.273abc 0.064 Trend and Intercept −2.380 −5.733* −2.405 −5.731* 0.090abc 0.052 LnCT Intercept −2.035 −1.072 0.861 −1.276 0.686a 0.278 Trend and Intercept −2.574 −0.082 −2.221 0.225 0.167c 0.157 Ln Ep Intercept −1.050 −1.560 0.311 −1.712 0.545a 0.288 Trend and Intercept −2.736* −2.365 −1.283 −1.904 0.171c 0.098 Ln Gdp Intercept −3.455** −5.162** −8.473** −5.162** 0.351ab 0.281 Trend and Intercept −2.963* −5.589** −5.821** −5.614** 0.165c 0.127 Ln Imp Intercept −1.034 −4.313* −1.034 −4.313** 0.661 0.118 Trend and Intercept −1.810 −4.304* −1.680 −4.308*** 0.094ab 0.099 LnLab Intercept −6.150*** −4.306* −2.403 −4.300*** 0.300c 0.246 Trend and Intercept −3.293** −4.392* −2.003 −4.400*** 0.135abc 0.107 *, **, and *** designates the MacKinnon critical values for ADF and PP at 1%, 5% and 10% levels are−3.679, −2.967 and−2.622 respectively. a,band cdenotes rejection of null hypothesis of stationarity in KPSS at 1%, 5% and 10% levels at 0.739 ,0.463 and 0.347 respectively. Ln represents logarithms of variables Table 1: Unit root tests Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach Table 2: Selection order criteria Lag LogL LR FPE AIC SIC HQ 0 127.1513 N/A 7.03e‑12 −8.653663 −8.368191 −8.566391 1 338.7610 317.4146* 2.68e‑17 −21.19722 −19.19891* −20.58631* 2 381.5092 45.80161 2.46e‑17* −21.67923* −17.96809 −20.54469 Sample: 1990‑2019 LR: Sequential modified LR test statistic (each at 5% level), FPE: Final prediction error, AIC: Akaike information criterion, SIC: Schwarz information criterion, HQ: Hannan‑Quinn information criterion Source: own calculations *Indicates lag orders selected by criterion Table 2: Selection order criteria Sample: 1990‑2019 LR: Sequential modified LR test statistic (each at 5% level), FPE: Final prediction error, AIC: Akaike information criterion, SIC: Schwarz information criterion, HQ: Hannan‑Quinn information criterion Source: own calculations *Indicates lag orders selected by criterion Table 3.1: Unrestricted cointegration rank test (trace) Hypothesized No. 3.5. Granger-causality h l f h l i of CE (s) Eigen value Trace 0.05 Probability** Statistic Critical Value None * 0.779571 42.34109 40.07757 0.0273 At most 1 0.647251 29.17599 33.87687 0.1644 At most 2 0.576870 24.08215 27.58434 0.1319 At most 3 0.403664 14.47462 21.13162 0.3274 At most 4 0.316085 10.63782 14.26460 0.1733 At most 5 0.069743 2.024245 3.841466 0.1548 Source: own calculations Trace test indicates three (3) cointegrating eng (s) at the 0.05 level,* denotes rejection of hypothesis at the 0.05 level, **Mackinnon‑Haug‑Michelis (1999) P-values Table 3.1: Unrestricted cointegration rank test (trace) was piloted at one lag length order with one cointegrating vector at trend 3, Intercept NO tend CE in VAR. Table 4 below gives a summary of Error Correction Model. The Error Correction Model (ECM) of the Johansen test is negative as expected but it is not statistically significant; however, the study holds on the information that the long-run information variables are statistically significant and consistent to economic theory. The first stage of ARDL modelling identifies the relationship between mining productivity (LNMP) which is a dependent variable and electricity pricing (LNEP) and other explanatory variables. The research examines the existence of long-run cointegration relationship by computing the F statistic. Specified by few observations available for estimation, the maximum lag order for the numerous variables in the model is set at (m=3) and the period of estimation is from 1990 to 2019 presented in Table 3. The F statistic for testing the joint null hypothesis that there is no long run relationship between the variables as defined above is explained in as follows using the bound test. The computed F statistic is F=52.23110 as illustrated in Table 5. The appropriate critical value for this test as computed by Persan, Shin and Smith (2001) at the 99% level is specified by 4.68 upper bound I (1). Since the F statistic surpasses the upper bound of the critical value bound, the test rejects the null hypothesis of no long-run relationship between variables. This test propose that there is a long-run relationship between LNMP, LNCT, LNEP, LNGDP2, LNIMP and LNLAB. Henriksson, Sӧderholm and Wἁrell (2014) specified that long-run electricity demand in the mining industry is delicate to changes in electricity prices. According to Kummel (1982), the substantial relationship between energy (electricity) prices and sectoral production is constant with the Capital-Labour- Energy and Creativity (KLEC). 3.5. Granger-causality h l f h l i This specifies that Capital Stock, Gross Domestic Product and Import Prices have an optimistic statistically significant influence on Mining Productivity/Output rate at 5% level, whereas a change in Electricity price and labour have an insignificant influence on Mining productivity/Output at 5% level. Various studies (Blignaut et al., 2015; Gosh, 2002; Inglesi-Lotz, 2014; Polemis and Dagoumas, 2013) support the The significant error correction term between zero and negative two specifies a constant long term equilibrium. In this study the error term is −0.122975, which entails that cointegration relationship is established. The speed of adjustment is 12.3%. This is the speed at which mining productivity returns to equilibrium after a shock in independent variables like electricity pricing. It shows that 12.3% of the gap between mining sector productivity and its equilibrium value is eliminated in the short-run. VECM International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022 International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022 Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach Table 4: Vector error correlation model (johansen technique) Long‑run correlation Cointegrating equation LNMP(‑1) LNCT(‑1) LNEP(‑1) LNGDP(‑1) LNIMP(‑1) LNLAB(‑1) Coint. Eq. 1 C 1.000000 21.00824 −1.492399 (0.17966) (−8.30670) 0.278221 (0.03403) (8.17491) 0.015273 (0.01899) (0.80412) 0.135544 (0.02606) (5.20143) 0.447622 (0.06357) (−7.04127) Short‑run correlation Error correction D (LNMP) D (LNCT) D (LNEP) D (LNGDP) D (LNIMP) D (LNLAB) Coint. Eq. 1 −0.122975 (0.28759) (−0.42760) 0.151000 (0.04138) (3.64953) −0.83199 (0.36583) (−2.27431) 2.021939 (1.55618) (1.29930) 0.601098 (1.46779) (0.40953) 0.362331 (0.56639) (0.63972) Source: Own calculations Table 4: Vector error correlation model (johansen technique) short-run dynamics and the second section covers the estimates of the Error Correction Term (ECT) that measures the promptness of adjustment, where short-run dynamics join the long-run equilibrium path in the model. Table 5: ARDL Co‑integration test Test Statistic Value K F Statistic 52.23110 5 Critical values Significant I (0) Bound I (1) Bound 10% 2.26 3.35 5% 2.62 3.79 2.5% 2.96 4.18 1% 3.41 4.68 Source: Own calculations Table 5: ARDL Co‑integration test Short-run coefficients estimates display the dynamic changes of all variables. The Short-run coefficient D (LNCT) and D (GDP2) are statistically significant at 1%. The coefficient error correction term ECM (-1) appraised at −1.037410 is greatly significant, indicating that mining sector productivity and Capital Stock and Gross Domestic Product are cointegrated. 3.5. Granger-causality h l f h l i * * * 1 2589 0 2092 1 0464 2 0 1324 0 2851 27 8581 * * . . LNIMP LNLAB t t CointEg LNMP LNCT LNEP LNGDP t t t t      ( . . . . * * * 1 2589 0 2092 1 0464 2 0 1324 0 2851 27 8581 * * . . LNIMP LNLAB t t 3.5. Granger-causality h l f h l i The estimated coefficient of the error correct term is over adjusting as specified with the value of −1.037 and it is statistically significant. According to Jamil and Ahmed (2010), the error correction model works in a manner that error in the previous analyses the correction toward long-run equilibrium. Adebola (2011) indicated that the long-run relationship is illustrated by a negative sign on the coefficient of the ECT. notion that electricity prices have an insignificant effect on the Gross Domestic Product in South Africa. Tips (2014) stated that electricity price rises have had a wavering impact on the mining value chains in South Africa. The estimation point for the two ARDL models are equivalent and the estimated standard errors attained for the model designated by the SBC and AIC are the same. To evaluate if the adopted model in this study is reasonably appropriate for the data. Diagnostic tests were done and they include Serial Correlation, Ramsey’s RESET test (Linearity tests), Normality test and Heteroscedasticity. It is vital to conduct diagnostic tests when analysing because they reveal whether there is a problem in the evaluation of a model or not. If the problem is identified, it means that the model is not effective and this can also suggest that we have biased results. For this study, results were presented in a way that we can say they have economic significance and reasonable. As specified in Table 9 residuals are not serially correlated as indicated by LM-stat probability with 45.51%. Output from the Ramsey’s RESET test shows that null hypothesis cannot be rejected since the P-value is 0.1341 which is more than 0.1. The Jarque-Bera test statistic is 8.955 and the probability is 0.011. Therefore, we reject the null since residuals are not normally distributed. It has also been detected that the model is free from existence of heteroscedasticity threats as shown by probability of 82.07%. The long-run model matching the ARDL for the natural log of Mining Productivity/Output can be written as follows: CointEg LNMP LNCT LNEP LNGDP t t t t      ( . . . . * * * 1 2589 0 2092 1 0464 2 0 1324 0 2851 27 8581 * * . . LNIMP LNLAB t t CointEg LNMP LNCT LNEP LNGDP t t t t      ( . . . . 4.3. Error Correction Model (ECM) Estimates for ARDLfi of Dependent Variable 0.040372 Residual sum of squares 5.40E‑05 Equation Log‑Likelihood 138.8415 Akaike information criterion −8.506776 Schwarz Bayesian Criterion −7.354921 DW‑statistic 2.9668 Hannan‑Quinn Criterion −8.164269 Source: Own calculations Table 8: Error correction representation for ARDL model Regressor Coefficient Standard error t‑statistic (prob.) ARDL dependent variable in LNMP D (LNCT) 1.912626 0.676331 2.827944 (0.066) D (LNEP) −0.049188 0.057035 −0.862413 (0.452) D (GDP2) 1.081601 0.101894 10.614932 (0.002) D (LNIMP) 0.011602 0.021688 −1.622205 (0.203) D (LNLAB) −0.083702 0.079274 −1.055848 (0.368) ECM(‑1) −1.037410 0.174003 −5.962037 (0.009) Source: Own calculations Table 6: Autoregressive distributed lag estimates (ARDL) model Regressor Coefficient Standard Error T‑statistic (Prob.) Dependent variable is LNMP LNMP(‑1) −0.179329 0.147610 −1.214887 (0.311) LNMP(‑2) −0.171872 0.93861 −1.831125 (0.164) LNMP(‑3) 0.085458 0.077658 1.100438 (0.351) LNCT 1.912626 0.676331 2.827944 (0.066) LNCT(‑1) −0.970914 0.749160 −1.296003 (0.286) LNCT(‑2) −2.330395 2.781330 −0.837871 (0.436) LNCT(‑3) 2.694684 2.645147 1.018728 (0.383) LNEP −0.049188 0.057035 −0.862413 (0.452) LNEP(‑1) −0.047249 0.054323 −0.869780 (0.448) LNEP(‑2) −0.153554 0.106963 −1.435577 (0.246) LNEP(‑3) 0.032979 0.056113 0.587717 (0.598) LNGDP2 1.081601 0.101894 10.61493 (0.002) LNGDP2(‑1) 0.006321 0.021994 0.287403 (0.792) LNGDP2(‑2) 0.027878 0.017185 1.622205 (0.203) LNGDP2(‑3) −0.030228 0.017678 −1.709932 (0.186) LNIMP 0.011602 0.023687 0.489819 (0.658) LNIMP(‑1) −0.019267 0.029582 −0.651297 (0.561) LNIMP(‑2) −0.032650 0.021688 −1.505419 (0.229) LNIMP(‑3) −0.097087 0.035089 −2.766852 (0.070) LNLAB −0.083702 0.079274 −1.055848 (0.368) LNLAB(‑1) 0.023919 0.052673 0.454107 (0.680) LNLAB(‑2) 0.169096 0.081976 2.062739 (0.131) LNLAB(‑3) 0.186423 0.119623 1.558424 (0.217) C −28.90024 8.740591 −3.306440 (0.045) R‑Squared 0.998726 Adjust. R‑Squared 0.988957 S.E. of regression 0.004242 F‑Statistic 102.2391 (0.001) Mean of dependent variable 4.6186 S.D. of Dependent Variable 0.040372 Residual sum of squares 5.40E‑05 Equation Log‑Likelihood 138.8415 Akaike information criterion −8.506776 Schwarz Bayesian Criterion −7.354921 DW‑statistic 2.9668 Hannan‑Quinn Criterion −8.164269 Source: Own calculations Table 6: Autoregressive distributed lag estimates (ARDL) model Regressor Coefficient Table 6: Autoregressive distributed lag estimates (ARDL) modelfi Table 9: Short‑run diagnostics Test Null Hypothesis t‑statistic Probability Langrange multiplier (LM) No Serial Correlation 21.408 0.455 Linearity (Ramsey’s) Cannot reject null hypothesis 5.993 0.341 Jarque‑Bera (JB) There is normal distribution 8.955 0.011 Breusch‑Pagan (CH‑sq) No conditional heteroscedasticity 0.562 0.820 (A) Lagrange multiplier test of residual serial correlation. (B) Ramsey’s RESET test using the square of the fitted values. (C) Based on the skewness and kurtosis of residuals. (D) Based on the regression of squared residuals on squared fitted values. 4.3. Error Correction Model (ECM) Estimates for ARDLfi Source: Own calculation Table 7: Estimated long‑run coefficients using the ARDL approach for model Table 7: Estimated long‑run coefficients using the ARDL approach for model Regressor Coefficient Standard Error T‑ratio (Prob.) Dependent variable is LNMP LNCT 1.258907 0.413867 3.041813 (0.056) LNEP 0.209187 0.078609 −2.661104 (0.076) LNGDP2 1.046425 0.227782 4.593987 (0.019) LNIMP −0.132446 0.034354 −3.855328 (0.031) LNLAB 0.285072 0.118039 2.415071 (0.094) C −27.858076 7.094437 −3.926750 (0.029) Source: Own calculations (A) Lagrange multiplier test of residual serial correlation. (B) Ramsey’s RESET test using the square of the fitted values. (C) Based on the skewness and kurtosis of residuals. (D) Based on the regression of squared residuals on squared fitted values. Source: Own calculation Table 8: Error correction representation for ARDL model Regressor Coefficient Standard error t‑statistic (prob.) ARDL dependent variable in LNMP D (LNCT) 1.912626 0.676331 2.827944 (0.066) D (LNEP) −0.049188 0.057035 −0.862413 (0.452) D (GDP2) 1.081601 0.101894 10.614932 (0.002) D (LNIMP) 0.011602 0.021688 −1.622205 (0.203) D (LNLAB) −0.083702 0.079274 −1.055848 (0.368) ECM(‑1) −1.037410 0.174003 −5.962037 (0.009) Source: Own calculations Table 8: Error correction representation for ARDL model Regressor Coefficient Standard error t‑statistic (prob.) ARDL dependent variable in LNMP Table 8: Error correction representation for ARDL modelfi are presented in this chapter and the remaining other results are in the appendix (G). The presence of a long-run relationship amongst variables indicate that causality may exists. The study used pairwise (bilateral) approach which indicates the causal relationship between electricity pricing and mining performance (output) with f statistic of 2.36021 and probability of 0.1169, due to the insignificance of the probability, conclusion can be that electricity pricing does not granger cause mining performance. are presented in this chapter and the remaining other results are in the appendix (G). The presence of a long-run relationship amongst variables indicate that causality may exists. The study used pairwise (bilateral) approach which indicates the causal relationship between electricity pricing and mining performance (output) with f statistic of 2.36021 and probability of 0.1169, due to the insignificance of the probability, conclusion can be that electricity pricing does not granger cause mining performance. is a good measure of mining performance since residuals move somehow in the bent at 5% at the level of significance. Furthermore, the results indicate that mining performance with F statistics of 2.46586 and the probability of 0.1071 is significant at 10% probability does not granger cause electricity pricing. 4.3. Error Correction Model (ECM) Estimates for ARDLfi Furthermore, the results indicate that mining performance with F i Table 7: Estimated long‑run coefficients using the ARDL approach for model Regressor Coefficient Standard Error T‑ratio (Prob.) Dependent variable is LNMP LNCT 1.258907 0.413867 3.041813 (0.056) LNEP 0.209187 0.078609 −2.661104 (0.076) LNGDP2 1.046425 0.227782 4.593987 (0.019) LNIMP −0.132446 0.034354 −3.855328 (0.031) LNLAB 0.285072 0.118039 2.415071 (0.094) C −27.858076 7.094437 −3.926750 (0.029) Source: Own calculations Table 9: Short‑run diagnostics Test Null Hypothesis t‑statistic Probability Langrange multiplier (LM) No Serial Correlation 21.408 0.455 Linearity (Ramsey’s) Cannot reject null hypothesis 5.993 0.341 Jarque‑Bera (JB) There is normal distribution 8.955 0.011 Breusch‑Pagan (CH‑sq) No conditional heteroscedasticity 0.562 0.820 (A) Lagrange multiplier test of residual serial correlation. (B) Ramsey’s RESET test using the square of the fitted values. (C) Based on the skewness and kurtosis of residuals. (D) Based on the regression of squared residuals on squared fitted values. Source: Own calculation Table 6: Autoregressive distributed lag estimates (ARDL) model Regressor Coefficient Standard Error T‑statistic (Prob.) Dependent variable is LNMP LNMP(‑1) −0.179329 0.147610 −1.214887 (0.311) LNMP(‑2) −0.171872 0.93861 −1.831125 (0.164) LNMP(‑3) 0.085458 0.077658 1.100438 (0.351) LNCT 1.912626 0.676331 2.827944 (0.066) LNCT(‑1) −0.970914 0.749160 −1.296003 (0.286) LNCT(‑2) −2.330395 2.781330 −0.837871 (0.436) LNCT(‑3) 2.694684 2.645147 1.018728 (0.383) LNEP −0.049188 0.057035 −0.862413 (0.452) LNEP(‑1) −0.047249 0.054323 −0.869780 (0.448) LNEP(‑2) −0.153554 0.106963 −1.435577 (0.246) LNEP(‑3) 0.032979 0.056113 0.587717 (0.598) LNGDP2 1.081601 0.101894 10.61493 (0.002) LNGDP2(‑1) 0.006321 0.021994 0.287403 (0.792) LNGDP2(‑2) 0.027878 0.017185 1.622205 (0.203) LNGDP2(‑3) −0.030228 0.017678 −1.709932 (0.186) LNIMP 0.011602 0.023687 0.489819 (0.658) LNIMP(‑1) −0.019267 0.029582 −0.651297 (0.561) LNIMP(‑2) −0.032650 0.021688 −1.505419 (0.229) LNIMP(‑3) −0.097087 0.035089 −2.766852 (0.070) LNLAB −0.083702 0.079274 −1.055848 (0.368) LNLAB(‑1) 0.023919 0.052673 0.454107 (0.680) LNLAB(‑2) 0.169096 0.081976 2.062739 (0.131) LNLAB(‑3) 0.186423 0.119623 1.558424 (0.217) C −28.90024 8.740591 −3.306440 (0.045) R‑Squared 0.998726 Adjust. R‑Squared 0.988957 S.E. of regression 0.004242 F‑Statistic 102.2391 (0.001) Mean of dependent variable 4.6186 S.D. 4.3. Error Correction Model (ECM) Estimates for ARDLfi According to Arize (2016), Error Correction Coefficient must be between −1<P<0 and should be significantly negative. With the series within the unit interval, it records the speed of adjustment such that when the dependent variable outstrips the long-run relationship with the IV’s, they bend downwards at the rate within the range. Hence, it offers an indirect test of cointegration. Bhattacharya (2016) mentioned that if there is a negative Error Correction Coefficient and is significant than a long run relationship between dependent and independent variables, the long-run coefficients resulting from the cointegrating equation will indicate the long-run impact and the coefficient of the first differenced variables will display the short-run impact. Figure 1 indicates the residual’s pattern or performance with reverence to its stability. If the curved line which symbolises the residuals were to fall outside the two extreme lines signifying the critical regions, the residuals would have been viewed unstable. The stability of the model is supported from the results of the stability testing using CUSUM test. Meanwhile the residual plot did not fall outside the 5% significant borders, which means the estimates are considered to be stable over the period. Figure 2 shows the CUSUMSQ. The figure illustrates that electricity pricing Table 8 explains the results of the estimated ECM model which corresponds with the long-run estimates of Akaike Information Criterion. The estimated Error Correction Model (ECM) has two sections. The first section covers the estimated coefficients of 289 Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach ,f gy g g pp is a good measure of mining performance since residuals move somehow in the bent at 5% at the level of significance. are presented in this chapter and the remaining other results are in the appendix (G). The presence of a long-run relationship amongst variables indicate that causality may exists. The study used pairwise (bilateral) approach which indicates the causal relationship between electricity pricing and mining performance (output) with f statistic of 2.36021 and probability of 0.1169, due to the insignificance of the probability, conclusion can be that electricity pricing does not granger cause mining performance. 4.4. Granger Causality 4.4. Granger Causality 4.3. Error Correction Model (ECM) Estimates for ARDLfi This implies that there is no bidirectional relationship between mining performance and electricity prices for the observed period. Berk 4.4. Granger Causality Table 10 presents the granger causality results. Therefore, it should be noted that only results of the main variables of interest Table 10 presents the granger causality results. Therefore, it should be noted that only results of the main variables of interest International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022 Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach Table 10: Granger causality test Null hypothesis Obs. F‑statistics Prob. LNEP does not Granger cause LNMP 28 2.36021 0.1169 LNMP does not Granger cause LNEP 2.46586 0.1071*** *,**, and *** show significant at 1%, 5% and 10% respectively. Source: Own calculation Figure 2 : CUSUM of squares Figure 1: CUSUM Figure 1: CUSUM do not reflect accurate economic expenses of delivering power, which led to gross misallocation of resources and poor decision- making. After the 2008 power crisis, it has been proven that there is a direct relationship between electricity prices and mining sector productivity. The reality is that the South African mining sector has been experiencing serious challenges. Electricity prices increases resulted in a decrease in productivity, especially in the gold industry, power outages and a decline in GDP contribution. The sector is looking at mechanisation to address these problems. The study discovered that there was a cointegrating relationship between mining sector productivity, electricity pricing, capital stock, gross domestic product, import prices and labour. The results revealed that approximately −1.037 of the disequilibrium is corrected for mining sector output. Figure 2 : CUSUM of squares Figure 1: CUSUM Figure 2 : CUSUM of squares South Africa’s Mining Sector has been subjected to a deterioration in its output (especially gold mining) due to extreme increase in electricity tariffs since the 2008 economic crisis. To reduce electricity price hikes, South African Government should invest on renewable energy technologies. This is in line with National Development Plan (NDP) and South African climatic change (NPC, 2011) with the objective to support the scale-up of low carbon local technology market. The observed results on this study found that there is a long-run optimistic relationship between electricity pricing and Mining Sector productivity. 4.4. Granger Causality The study also agrees with Mineral Council South Africa that mines should invest on producing their own energy, in that way they will be responsible for reducing their costs and that could results in mines increasing their productive capacity in the long-term. Policy makers should also adopt Kohler (2013) proposal that authorities should target electricity intensive industries by employing a differential pricing policy. This will not only improve the mining sector output but it will also create further employment opportunities in the country. Table 10: Granger causality test Null hypothesis Obs. F‑statistics Prob. LNEP does not Granger cause LNMP 28 2.36021 0.1169 LNMP does not Granger cause LNEP 2.46586 0.1071*** *,**, and *** show significant at 1%, 5% and 10% respectively. Source: Own calculation Table 10: Granger causality test *,**, and *** show significant at 1%, 5% and 10% respectively. Source: Own calculation REFERENCES and Ϋetkiner (2013) used facts on the U.S energy prices and economic relationship for the period of 1951-2010 which revealed that the growth rate of energy prices may negatively influence growth rates of both GDP and energy demand. Narayan and Smyth (2005a) observed the relationship between electricity consumption, employment and real income for Australia using multivariate Granger causality for the period 1966-1999. They found a unidirectional relationship between GDP to electricity consumption and from GDP to employment. Wolde-Rufael (2009) found unidirectional causality between energy consumption and economic growth, while Ziramba (2009) discovered a bio-directional causality between the energy consumption and industrial output. Bah, M.M., Azam, M. (2017), Investigating the relationship between electricity consumption and economic growth: Evidence from South Africa. Renewable and Sustainable Energy Reviews, 80, 531-537. Bildirici, M.E., Bakirtas, T. (2014), The relationship among oil, natural gas and coal consumption and economic growth in BRICTS (Brazil, Russian, India, China, Turkey and South Africa) countries. Energy, 65, 134-144. Blignaut, J., Inglesi-Lotz, R., Weideman, J.C.P. (2015), Sectoral electricity elasticities in South Africa: Before and after the supply crisis of 2008. South African Journal of Science, 111(9/10), 7. Cronje, F. (2017), What does Great Mining Policy Look like? Address to Mining Indaba 2017, 5 October 2017. Available from: https://www. irr.org.za/media/articles-authored-by-the-institute/what-does-great- mining-policy-look-like Dagoumas, A.S., Polemis, M.L., Soursou, S.E. (2020), Revisiting the impact of energy prices on economic growth: Lessons learned from the European Union. Economic Analysis and Policy, 66(C), 85-95. International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022 5. CONCLUSION Dlamini, J., Balcilar, M., Gupta, R., Inglesi-Lotz, R. (2016), Relationship between energy consumption and economic growth in South Africa: Evidence from the bootstrap rolling-window approach. Energy Sources Part B Economics Planning and Policy, 11(7), 617-625. Mining remains a critical Industry with an enormous influence on the economy. South Africa instantly needs to advance the regulatory environment in the mining industry. High consumption costs led to retrenchments because of high tariffs hikes in electricity pricing. The experience of the South African electricity supply sector over the past 30 years has proven that electricity prices Engle, R.F., Granger, C.W.J. (1987), Cointegration and error correction representation, estimation and testing. Econometrica, 55(2), 251-276. Eskom Historical Average Price Increase. Eskom Tariff Increases vs. Engle, R.F., Granger, C.W.J. (1987), Cointegration and error correction representation, estimation and testing. Econometrica, 55(2), 251-276.f Eskom Historical Average Price Increase. Eskom Tariff Increases vs. 291 f Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach Mmutle, et al.: Effects of Energy Pricing on the Mining Sector Performance in South Africa: An Econometric Approach coherence approach. Journal of Public Affairs, 21(1), 21-30.ffi coherence approach. Journal of Public Affairs, 21(1), 21-30.ffi Inflation Since 1988 (with Projections to 2022). Available from: https://www.poweroptimal.com Inflation Since 1988 (with Projections to 2022). Available from: https://www.poweroptimal.com Kohler, M. (2013), Differential electricity pricing and energy efficiency in South Africa. Energy, 64, 524-532. Hosking, S., Keseke, N. (2012), Cost of electricity load shedding to mines in Zimbabwe: Direct assessment approach. International Journal of Physical and Social Sciences, 2(6), 207-237. Kwiatkowski, D., Phillips, P.C.B., Schmidt, P., Shin, Y. (1992), Testing the null hypothesis of stationarity against the alternative of a unit- root: How sure are we that economic time series have a unit root? Journal of Econometric, 54(1-3), 159-178.fi Inglesi-Lotz, R., Blignaut, J.N. (2011), South Africa’s electricity consumption: A sectoral decomposition analysis. Applied Energy, 88(12), 4779-4784.fi Lin, B., Zhu, R. (2021), Energy efficiency of the mining sector in China, what are the main influence factors? Resources, Conservation and Recycling, 167, 105321. Inglesi-Lotz, R., Blignaut, J.N. (2014), Improving the electricity efficiency in South Africa through a benchmark-and-trade system. Renewable and Sustainable Energy Reviews, 30, 833-840.fi Nkoro, E., Uko A.K. (2016), Autoregressive distributed lag (ARDL) cointegration technique: Application and interpretation. Journal of Statistical and Econometric Methods, 5(4), 3. Inglesi-Lotz, R., Pouris, A., (2012), Energy efficiency in South Africa: A decomposition exercise. Energy, 42(1), 113-120. 5. CONCLUSION Statistical and Econometric Methods, 5(4), 3. Johansen, S. (1988), Statistical analysis of cointegrating vectors. Journal of Economic Dynamic and Control, 12(2-3), 231-254. Pesaran, M.H., Shin, Y., Smith, R.J. (1996), Testing for the Existence of a Long Run Relationship. DAE Working Paper No. 9622. Department of Applied Economics. England: University of Cambridge. Johansen, S., Juselius, K. (1990), Maximum likelihood estimation and inference on cointegration with applications to the demand for money. Oxford Bulletin of Economics and Statistics, 52(2), 169-210. Rybak, A., Rybak, A. (2016). Possible strategies for hard coal mining in Poland as a result of production function analysis. Resources Policy, 50, 27-33. Katta, A.K., Davis, M., Kumar, A. (2020), Development of disaggregated energy use and greenhouse gas emission footprints in Canada’s iron, gold, and potash mining sectors. Resources Conservation and Recycling, 152, 104485. Takentsi, S., Sibanda, K., Hosu, Y. (2022), Energy prices and economic performance in South Africa: An ARDL bounds testing approach. Cogent Economics and Finance, 10(1), 2069905. Kirikkaleli, D., Adedoyin, F.F., Bekun, F.V. (2021), Nuclear energy consumption and economic growth in the UK: Evidence from wavelet Ziramba, E. (2009), Disaggregate energy consumption and industrial production in South Africa. Energy Policy, 37(6), 2214-2220. 292 International Journal of Energy Economics and Policy | Vol 12 • Issue 6 • 2022
https://openalex.org/W1969254103
https://journals.rbge.org.uk/ejb/article/download/638/529
English
null
An Illustrated Flora of North Cyprus. D.E. Viney.
Edinburgh journal of botany
2,010
cc-by
1,007
BOOK REVIEWS BOOK REVIEWS 93 example, from genera starting with 'A' there are 19 missed names, from 'B' 8, from ' C 24, etc. The data under each entry are, in the main, sound, and apart from the odd computer-generated hiccup it is generally accurate (for specific errors see Burtt's review in Kew Bull. 49(3): 592, 1994). There is a substantial amount of missing data (e.g. for Bolax, a common South American genus, the type is recorded as 'not indicated'), and I find that treating orthographic variants as full entries with the qualification 'Type: not indicated' is rather misleading. The authors have identified a real need for a work bringing together the huge amount of taxonomic information on the Umbelliferae in a concise form. This book is a brave attempt to fulfil that need, and, although very useful to Umbel researchers and others as it stands, I would hope that it forms the basis for a much more rigorous second edition. example, from genera starting with 'A' there are 19 missed names, from 'B' 8, from ' C 24, etc. The data under each entry are, in the main, sound, and apart from the odd computer-generated hiccup it is generally accurate (for specific errors see Burtt's review in Kew Bull. 49(3): 592, 1994). There is a substantial amount of missing data (e.g. for Bolax, a common South American genus, the type is recorded as 'not indicated'), and I find that treating orthographic variants as full entries with the qualification 'Type: not indicated' is rather misleading. The authors have identified a real need for a work bringing together the huge amount of taxonomic information on the Umbelliferae in a concise form. This book is a brave attempt to fulfil that need, and, although very useful to Umbel researchers and others as it stands, I would hope that it forms the basis for a much more rigorous second edition. In the Introduction to this work, the authors discuss in detail the GNOM database, and the philosophy behind storing this kind of taxonomic information in a flexible electronic medium, allowing constant updating. There is no doubt that storing taxo- nomic (and floristic) information in this form is an excellent way of handling such complex datasets. Furthermore it opens up exciting possibilities for data exchange and information-sharing on an international scale. R E F E R E N C E BURTT, B. L. (1991). Umbelliferae of Southern Africa: an introduction and annotated check-list. Edinb. J. Bot. 48: 133-282. BURTT, B. L. (1991). Umbelliferae of Southern Africa: an introduction and annotated check-list. Edinb. J. Bot. 48: 133-282. M. F. Watson An Illustrated Flora of North Cyprus. D.E. Viney. Koenigstein, Germany: Koeltz Scientific Books. 1994. xxix + 697pp, 339 black and white illustrations. ISBN 3 87429 364 5. Price DM 58.00 (paperback). Cyprus has a wild flora of some 2000 species of higher plants, including more than 350 naturalized adventives. This assemblage of plants comprises one of the most important island floras of the Mediterranean region, with 88 endemic species, and several subspecies and varieties. The great majority of the endemics are centred on the Troodos Mountains in the west, but a substantial endemic element occurs in northern Cyprus. Dr Deryck Viney has lived there for some years and has set up a North Cyprus herbarium, now curated by the Forestry Department of Turkish Cyprus. Geographically, Cyprus is an extension of the great Taurus range of Anatolia. This book therefore complements the two superb recent regional Floras - Desmond Meikle's Flora of Cyprus and the monumental Flora of Turkey and the East Aegean Region edited by the late Peter Davis. The north-eastern part of the island has been occupied by Turkey since the invasion of the island by Turkish forces in 1974. Two BOOK REVIEWS 94 particularly useful features of the present work are the inclusion of Turkish vernacu- lar plant names and a gazetteer of Greek vs. Turkish place names. Fortuitously, the regrettable political partition of Cyprus coincides to a great extent with natural geographical and ecological boundaries. North Cyprus is dominated by the Kyrenia Mountains, a limestone range extending 80km along the northern coast, which have a rich endemic flora. particularly useful features of the present work are the inclusion of Turkish vernacu- lar plant names and a gazetteer of Greek vs. Turkish place names. Fortuitously, the regrettable political partition of Cyprus coincides to a great extent with natural geographical and ecological boundaries. North Cyprus is dominated by the Kyrenia Mountains, a limestone range extending 80km along the northern coast, which have a rich endemic flora. Any illustrated book on Mediterranean plants is as welcome for its wider relevance as for local use. R E F E R E N C E This Flora has descriptions and keys, and the author's own accurate line drawings illustrate the text, opposite species descriptions wherever possible. It is good to see several rarely illustrated plants, not least members of confusing genera such as Crepis, Ranunculus and Silene, and important Cypriot endemics such as th white-flowered cabbage Brassica hilarionis of cliffs in the Kyrenia Mountains; B hilarionis is closely related to B. cretica of the southern Aegean and perhaps to the cultivated cauliflowers. I should like to have had even more information than is provided on the ecology, conservation status and uses of plants, especially where blank spaces lurk at the end of family accounts. The book's omission of ferns, grasses and sedges is apparently to be rectified in a supplementary volume being prepared by the author. This Flora is expensive for one so dowdy, and having just had a manuscript elegantly typeset into Times Roman in a West Cork living room, I am surprised that any publisher still lays out text in typescript when software for more sophisticated typography is widely available. That said, here is a solid addition to the regional floristic literature that ought to be much appreciated and used by Cypriot botanists, agronomists and foresters, as well as by the numerous visitors from Britain on spring flower tours. J. R. Akeroyd
https://openalex.org/W3042231548
https://europepmc.org/articles/pmc7362657?pdf=render
English
null
Risk literacy assessment of general practitioners and medical students using the Berlin Numeracy Test
BMC family practice
2,020
cc-by
5,683
Abstract Background: The responsibility for helping patients understand potential health benefits and risks, especially regarding screening tests, falls largely to general practitioners (GPs). The Berlin Numeracy Test (BNT) specifically measures risk literacy (i.e., the ability to understand different aspects of statistical numeracy associated with accurate interpretation of information about risks). This study explored the association between risk literacy levels and clinical experience in GPs vs. medical students. Additionally, the effect of GP risk literacy on evaluation of the predictive value of screening tests was examined. Methods: The participants were 84 GPs and 92 third-year medical students who completed the BNT (total score range 0–4 points). The GPs received an additional case scenario on mammography screening as a simple measure of performance in applying numeracy skills. Results: Despite having an average of 25.9 years of clinical experience, GPs scored no better than medical students on risk literacy (GPs: 2.33 points, 95% confidence interval [CI] 2.08–2.59; students: 2.34, 95% CI 2.07–2.61; P = .983). Of all GPs, 71.6% (n = 58) greatly overestimated the real predictive value. Results: Despite having an average of 25.9 years of clinical experience, GPs scored no better than medical students on risk literacy (GPs: 2.33 points, 95% confidence interval [CI] 2.08–2.59; students: 2.34, 95% CI 2.07–2.61; P = .983). Of all GPs, 71.6% (n = 58) greatly overestimated the real predictive value. Conclusions: In this study, we found no difference in risk literacy between current students and current GPs. GPs lack risk literacy and consequently do not fully understand numeric estimates of probability in routine screening procedures. Conclusions: In this study, we found no difference in risk literacy between current students and current GPs. GPs lack risk literacy and consequently do not fully understand numeric estimates of probability in routine screening procedures. Keywords: Risk literacy, Mass screening, Mammography, Medical education, Predictive value Keywords: Risk literacy, Mass screening, Mammography, Medical education, Predictive value research has shown that adequate risk communication can even increase the use of screening tests [8]. How- ever, some studies have demonstrated lownumeracy, the ability to perform basic quantitative calculations, among patients [9]. Because understanding risk requires a cer- tain level of numeracy, patients must often depend on their providers to interpret and communicate risk infor- mation. If providers also do not understand the quanti- tative aspects of risk, patients will be unable to comprehend or use the information well [10]. Friederichs et al. BMC Family Practice (2020) 21:143 https://doi.org/10.1186/s12875-020-01214-w Friederichs et al. BMC Family Practice (2020) 21:143 https://doi.org/10.1186/s12875-020-01214-w Open Access © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Abstract It is there- fore safe to assume that, as a rule, only physicians with adequate training will be able to assist their patients in this way. Risk literacy assessment of general practitioners and medical students using the Berlin Numeracy Test endrik Friederichs1* , Roman Birkenstein1, Jan C. Becker1, Bernhard Marschall1 and Anne Weisse Hendrik Friederichs1* , Roman Birkenstein1, Jan C. Becker1, Bernhard Marschall1 and Anne Weissenstein2 Background As previously described in detail [1], several studies sug- gest that patients want to be included in discussions about their health, diagnostics, and therapy [2].This patient-centered and shared decision-making approach between patients and doctors is the basis of modern medicine [3–7]. Individual risk calculations are an effect- ive method of risk communication, particularly in patient education for screening programs. Previous * Correspondence: friedeh@uni-muenster.de 1Study Hospital Münster, Institute for Education and Student Affairs, Medical Faculty of Münster, Malmedyweg 17-19, D-48149 Münster, Germany Full list of author information is available at the end of the article * Correspondence: friedeh@uni-muenster.de 1Study Hospital Münster, Institute for Education and Student Affairs, Medical Faculty of Münster, Malmedyweg 17-19, D-48149 Münster, Germany Full list of author information is available at the end of the article * Correspondence: friedeh@uni-muenster.de 1Study Hospital Münster, Institute for Education and Student Affairs, Medical Faculty of Münster, Malmedyweg 17-19, D-48149 Münster, Germany Full list of author information is available at the end of the article Page 2 of 8 Page 2 of 8 Friederichs et al. BMC Family Practice (2020) 21:143 Friederichs et al. BMC Family Practice (2020) 21:143 Sufficient competence in statistical numeracy is essen- tial to fully understand and communicate numeric esti- mates of probability [11]. The ability to understand statistics and the quantitative aspects of research varies tremendously among physicians and medical students [12, 13]. Additionally, because risk information is pri- marily quantitative, these individual differences arguably affect medical professionals’ comprehension of diagnos- tic tests and other risk-related communications [14]. The Berlin Numeracy Test [BNT] [15] is a multiple- choice instrument that specifically measures risk literacy; that is, the ability to understand different aspects of stat- istical numeracy associated with accurate interpretation of information about risks. The BNT was developed for use with educated and highly educated samples. In a previous study, we showed that the average risk literacy in medical students is slightly above average (Cohen’s d = .23) and does not increase during medical studies [1]. Little is known about whether risk literacy increases during postgraduate medical education.> numeracy assessed their risk of death with and without mammography more accurately than did patients with lower numeracy [20]. In Germany, the mammography participation rate is about 56%, so nearly half of the in- vited women decline the recommended appointment [21]. Therefore, an additional issue that needs to be ad- dressed in this context is the consequences of GPs’ risk literacy for their comprehension of and recommenda- tions for screening tests. Setting and subjects Our study was conducted in the winter semester of 2015/2016 with GPs of the Westfalen-Lippe State Cham- ber of Physicians, which is a registered corporation under public law and represents the interests of 42,000 physicians in this area (the fourth-largest state chamber in Germany). A postal questionnaire was our only finan- cially viable option, but the response rates of GPs to pos- tal surveys are often relatively low [22]. Expecting a large non-response rate, we contacted as many GPs as neces- sary to reach the a priori calculated sample size. The students were recruited at the medical school of the Westphalian Wilhelms University in Münster by ap- proaching them in the context of a curricular course in physical examination. This course prepares students for their clinical clerkships, so it takes part at the very be- ginning of their clinical experience. Research questions We performed a controlled study of GPs and medical students to investigate the following questions: 1. What is the risk literacy among GPs? 2. Is there a difference in risk literacy levels between clinically experienced GPs and medical students? y p 3. Does the level of risk literacy influence evaluation of the predictive value of screening tests? g p g Previous research has suggested suboptimal risk literacy in clinicians. Gigerenzer and colleagues [16] studied 160 gynecologists and found that one-third were ignorant of the benefits of mammography screening. Most of the in- vestigated gynecologists were unable to calculate the posi- tive predictive value, so their probability of choosing the best answer was “slightly less than chance.” In a different report published as a letter, Estrada et al. [17] assessed the numeracy skills of 45 health professionals (including phy- sicians, nurses, doctorate faculty, and medical students). Twenty percent of the participants gave incorrect answers to two or more of six absolutely basic numeracy questions. Bramwell and colleagues [18] reported that only about one-third of obstetricians were able to calculate the prob- ability of an unborn child actually having trisomy 21, given a positive test result. These findings suggest that a lack of risk literacy may account for medical students and novice physicians repeating unnecessary diagnostics [19], and also may lead to an overestimation of the value of diagnos- tic findings. Study design To our knowledge, very limited data exist on the risk literacy of general practitioners (GPs), despite the im- portance of their role in health services in general and in risk communication in particular. GPs are the first line of contact for questions about whether to use offered screenings; therefore, they bear the main responsibility for helping patients understand the potential benefits and risks of screenings. For example, in Germany, mam- mography screening is offered to all women aged 50 to 69 years. The invitation includes a recommended ap- pointment at one of the certified mammography centers. In a patient-centered study about understanding the risks and benefits of breast cancer screening, most par- ticipants overestimated the benefits. Patients with higher The participants were asked to complete the BNT volun- tarily and anonymously. The GPs received an additional case scenario on mammography screening to assess their performance in applying numeracy skills. One BNT item measures the same concept as the additional scenario (i.e., positive prediction in item 1), but in a non-clinical con- text. Students were not asked to complete the scenario be- cause it is already used twice in the medical curriculum to explain two aspects of mass screening (i.e., positive pre- dictive power and overdiagnosis). We did not ask medical students and GPs about their academic background in statistics and probability theory. However, completion of some statistical coursework is required before taking the preliminary medical exam Friederichs et al. BMC Family Practice (2020) 21:143 Page 3 of 8 Page 3 of 8 Page 3 of 8 (“Physikum”), which is a prerequisite for entering the clin- ical phase of medical school. Therefore, we can assume that all participants had received statistical training. All participants had to agree verbally to participate. Additionally, they provided informed consent prior to the study by reading the background information and choosing to provide data. Ethical approval was given by the Ethics Committee of the Chamber of Physicians at Westfalen-Lippe and the Medical School of Westphalian Wilhelms University in Münster (2015–308-f-S). in risk literacy levels between clinically experienced GPs and medical students (calculated a priori with G*Power 3.1) [24]. (“Physikum”), which is a prerequisite for entering the clin- ical phase of medical school. Therefore, we can assume that all participants had received statistical training. Acquired data were entered and analyzed using the statistical software IBM SPSS Statistics 21 (IBM Corp., Armonk, New York, USA). Outcome measures After the evaluation of all datasets, three major findings emerged. The first is that participants achieved an aver- age BNT score of 2.34 (SD 1.24) points, which is not meaningfully higher than the average score obtained by Cokely et al. [15] , who reported that a score of 2.2 points was achieved by “highly numerate individuals.” There was only a minimal difference between the scores of GPs (2.33 points [95% confidence intervals [CI] 2.08– 2.59]) and medical students (2.34 [95% CI 2.07–2.61]); accordingly, the Mann–Whitney U test results were not significant (P = .983). Figure 2 shows the distribution of scores by group. The proportion of participants achiev- ing the maximum score (4 points) was higher for the student group (27.2% vs. 19.0%, P = .20). Outcome measures were the total test scores students and GPs achieved on the BNT scenarios, and (for GPs) performance on the hypothetical case scenario. The GPs solved five case scenarios and the students solved four, as mentioned earlier. In addition, baseline characteristics (age, sex, clinical experience for GPs) were obtained from all participants. Data collection In the BNT, participants were required to solve four short case studies; two gave probabilities in relative number for- mat and two gave probabilities in absolute number format in the form of Bayesian inferences. To obtain the correct solution for these text problems, statistical information on the prevalence of a certain event had to be inserted into Bayes’ theorem. Participants received one point per cor- rect solution; thus, their total score could range from 0 to 4 points. The test format has been previously validated by Cokely and colleagues [15] against the original paper-and- pencil version of the BNT, which in turn has been bench- marked against several population datasets [15]. The aver- age score (standard deviation [SD]) in the original validation study with this test format is 1.6 (1.21) [15]. Even when used with highly numerate individuals, the multiple-choice test format provides good discriminability and remains well balanced [15]. Study design Descriptive means and standard deviations were calculated for participants’ age, and total test scores and frequencies were calculated for sex and for solving the case scenarios. Sample means and frequencies were compared with population means and frequencies using one-sample t-tests and chi-square tests, respectively. We used a Mann–Whitney U test to assess the difference in risk literacy between medical stu- dents and GPs as well as the difference between GPs who solved the case scenario successfully and those who did not. In addition, Kendall’s tau correlation coefficient was used to assess the association between BNT scores and successfully solved case scenarios. A chi-square test was used to compare the proportions of GPs and med- ical students achieving the maximum score. No adjust- ment for multiple testing was performed. All participants had to agree verbally to participate. Additionally, they provided informed consent prior to the study by reading the background information and choosing to provide data. Ethical approval was given by the Ethics Committee of the Chamber of Physicians at Westfalen-Lippe and the Medical School of Westphalian Wilhelms University in Münster (2015–308-f-S). Results Recruitment process and demographic characteristics The recruitment process is shown in Fig. 1. We obtained 84 complete data sets (return rate 10.7%) after contact- ing 785 GPs. The average age of the GPs was 54.7 (SD 8.0) years, and 24 (28.6%) were female. They had an average clinical experience of 25.9 (SD 8.6) years. Ninety-two complete questionnaires from 96 participat- ing medical students were entered into the analysis (re- turn rate 95.8%). The average age of the medical students was 23.2 (SD 4.0) years, and 55 (59.8%) were fe- male. The time needed to complete the test was less than 10 min. In the additional case scenario for the GPs, we presented mammography screening information about disease prevalence and conditional probabilities (sensitivity and 1- specificity) (see Additional file 1). The task was to provide a probability estimate for the positive predictive value. Only the correct answer (9.2%) with an error margin from 8 to 10 was scored as correct; all other responses and missing responses were considered incorrect. Each partici- pant who wrote down a number in this range received one point. As Table 1 reveals, the baseline characteristics of the study populations and the analyzed samples show no significant differences. Statistical methods We used the standard alpha level of .05 for significance and a power level of .80. Therefore, we needed a sample size of at least 82 participants to detect an effect size showing a minimally important difference (d = .40) [23] Friederichs et al. BMC Family Practice (2020) 21:143 Page 4 of 8 Fig. 1 Recruitment process for study samples. BNT, Berlin Numeracy Test Fig. 1 Recruitment process for study samples. BNT, Berlin Numeracy Test solve it (n = 62, 2.11 points,P < .01), a finding reflected in the positive correlation between BNT scores and solving the extra scenario (correlation coefficient Kendall’s tau .31, P < .01). Of the GPs, 58 (71.6%) en- tered values > 80%, a conclusion that greatly overesti- mates the real predictive value. Our second finding was that there were no significant differences between male and female participants (P ≥.05). Finally, GPs who successfully solved the extra scenario (n = 19, 23.5%) achieved significantly higher scores on the BNT (3.00 points) than did GPs who were unable to Table 1 Baseline characteristics of study populations and analyzed samples Study Population Analyzed Sample General Practitioners Total in the state chamber Participants P-value n 4848 84 Age in years, mean (SD) 55.1 (8.7) 54.7 (8.0) .65 Female sex, % (n) 33.3 (1613) 28.6 (24) .36 Clinical experience in years, mean (SD) 25.9 (8.6) – Medical Students Total in the cohort Participants P-value n 102 92 Age in years, mean (SD) 23.6 (4.5) 23.2 (4.0) .34 Female sex, % (n) 60.8 (62) 59.8 (55) .89 SD standard deviation Table 1 Baseline characteristics of study populations and analyzed samples Table 1 Baseline characteristics of study populations and analyzed samples Study Population Medical Students Friederichs et al. BMC Family Practice (2020) 21:143 Page 5 of 8 Fig. 2 Distribution of Berlin Numeracy Test scores for general practitioners and medical students Fig. 2 Distribution of Berlin Numeracy Test scores for general practitioners and medical students in the face of new data [29, 30] [see also Wegwarth & Gigerenzer [31] for a good summary]. In the previously mentioned study by Gigerenzer et al. [16], most gynecol- ogists grossly overestimated the probability of cancer after positive mammography screening results, and their probability estimates were highly variable. Not surpris- ingly, studies of medical students have shown similar re- sults. In 2002, Sheridan and Pignone provided 62 first- year medical students with three numeracy questions as- sociated with information about the baseline risk for de- veloping a hypothetical disease [32]. They reported that almost one-quarter of the students had trouble perform- ing basic numerical operations. Although showing a relatively high level of numeracy (77% answered all ques- tions correctly), only 61% of students correctly inter- preted quantitative data for treatment of a hypothetical disease. The authors concluded that medical students do not understand statistical concepts well, and our findings suggest that this conclusion also applies to fully trained physicians. Discussion Main findings and implications Limitations The main limitation of this study is the difference in questionnaire return rates between GPs and students. Whereas over 90% of the medical students returned questionnaires, far fewer GPs did so. Because the popu- lation of interest was a large, geographically dispersed one, a postal questionnaire was our only financially vi- able option. Although we tried to increase the response rate [43], only about 10% of the contacted GPs replied. Non-response reduces the effective sample size and may have introduced selection bias. In a shared decision-making approach, treatment deci- sions are made jointly by doctors and patients [34], with the patients’ wishes and needs regarding the therapy re- ceiving considerable weight. This is the core element of patient-centered medicine [34]. Adequate communica- tion of risks of disease and treatment modalities is an important factor in ensuring successful decision-making. Decision-making based on correct understanding of the underlying risks and benefits is central to the physician– patient relationship and is therefore an essential deter- minant of patient satisfaction and compliance [11, 35]. We were able to show that physicians clearly do not have the necessary numeracy to interpret cancer screen- ing statistics. As Moyer (2012) states, “Expecting them to communicate this information to patients is a stretch.”[36]. Selection bias severely limits the interpretation of em- pirical findings. The baseline characteristics provided no indication of systematic differences between the study populations and our analyzed samples. It is likely that if selection bias did occur in this study, it worked against the reported findings; that is, it is probable that GPs who answered the postal survey were much more confident about their numeracy skills than those who did not. Our results may therefore actually overestimate GPs’ competence in risk literacy and their understanding of routine screening procedures. Training in risk literacy may help to enhance students’ numeracy skills, but this sometimes does not seem to be the case [37]. GPs who successfully solved the case sce- nario attained significantly higher BNT scores (3.00 points) than those who did not, a finding reflected in the positive correlation between solving the extra scenario and BNT scores. According to Cohen, a correlation co- efficient of >.30 reflects a medium effect size [38]. Therefore, the BNT can help to identify individual weak- nesses in numeracy. The BNT is a rapid measure of indi- vidual deficiencies in risk literacy and could be used to enhance the effectiveness of training. Main findings and implications Our results show no difference in risk literacy between current students and current GPs. Although showing above-average test results, GPs lack understanding of the underlying success and failure probabilities of mam- mography screening and therefore cannot accurately re- late these to their patients. The first major finding was that GPs showed above- average test results on the BNT, but did not perform sig- nificantly better than medical students. GPs reached a mean risk literacy score of 2.33 points, which according to Cokely et al. [15] is not meaningfully better than the 2.2 points considered average for “highly numerate indi- viduals.” The likelihood that a positive result on a diag- nostic test is correct (the test’s positive predictive value) strongly depends on the prevalence of the disease. If prevalence is low, even the best diagnostic test will not be very effective. GPs require accumulated experience in accurate test interpretation and the corresponding risks. However, despite their work in settings with low preva- lence for many diseases, GPs show no better risk literacy than do average medical students. We found that GPs performed poorly in applying nu- meracy skills in the given case scenario. Nearly three- quarters of the GPs entered values of > 80%, a response quite different to the real predictive value. Risk literacy correlated significantly and positively with the ability to solve the case scenario. This is not only a problem with physicians. In the above-mentioned mammography study by Schwartz et al., even the most arithmetically able patients made accurate assessments of risk less than half the time [20]. In reality, physicians and patients are Although the ability to work with numbers is import- ant for physicians, only a handful of studies have ex- plored their numeracy skills. In these studies, physicians have repeatedly shown poor understanding of statistics and quantitative aspects of research [25–28]. They par- ticularly lack the ability to apply Bayesian inferences, which are used to estimate probabilities of a hypothesis Friederichs et al. BMC Family Practice (2020) 21:143 Friederichs et al. BMC Family Practice (2020) 21:143 Page 6 of 8 trained to generate natural frequencies by themselves from provided data. often confronted with additional irrelevant screening evi- dence, such as improved 5-year survival and increased early detection. Wegwarth et al. Limitations Several tools exist to increase understanding of risks [39, 40]. However, the introduction of such new training to an already crowded medical studies curriculum remains a challenge. Main findings and implications found that unfortunately “many physicians did not distinguish between irrelevant and relevant screening evidence”[33], a situation that al- most certainly hampers understanding of the benefits and risks of screening. Finally, it can be concluded that GPs lack risk literacy and therefore do not fully under- stand the numeric estimates of probability associated with routine screening procedures. Conclusions In this study, we found no difference in risk literacy be- tween current students and current GPs. GPs lack risk lit- eracy and consequently do not fully understand numeric estimates of probability in routine screening procedures. Supplementary information y Supplementary information accompanies this paper at https://doi.org/10. 1186/s12875-020-01214-w. Additional file 1. Case scenario of mammography screening Practice points  The responsibility to help patients understand potential health benefits and risks falls largely to GPs.  In this study, we found no difference in numeracy skills between current students and current GPs.  There is a pressing need to help physicians understand and correctly interpret risk information. The most promising way forward is the use of alterna- tive statistical formats, particularly natural frequencies (such as 100 of 1000 instead of 10%) for presenting the risks in screening tests. Therefore, developers of patient decision aids are recommended to preferentially use a natural frequency format for test and screening decisions [41]. The natural frequency method is very effective (Cohen’s d effect size .69; 95% CI 0.45–0.93) and should be used with patients and medical doctors [42]. Authors of publications about screenings should be encouraged by medical journal editors to ensure that the results are presented in terms of natural frequencies to avoid misin- terpretation. For all other cases, students should be  Training on risk literacy during medical school may help, but we also suggest presenting risk information as natural frequencies. Supplementary information Supplementary information Additional file 1. Case scenario of mammography screening Availability of data and materials h l d h h f d 16. Gigerenzer G, Gaissmaier W, Kurz-Milcke E, Schwartz LM, Woloshin S. Helping doctors and patients make sense of health statistics. Psychol Sci Public Interest. 2007;8:53–96. The original data that support the findings of this study are available from https://uni-muenster.sciebo.de/index.php/s/KqWs7nE2rGfnYgr. 17. Estrada C, Barnes V, Collins C, Byrd JC. Health literacy and numeracy. JAMA. 1999;282:527. References 28. Gramling R, Irvin JE, Nash J, Sciamanna C, Culpepper L. Numeracy and medicine: key family physician attitudes about communicating probability with patients. J Am Board Fam Pract. 2004;17:473. 1. Friederichs H, Schölling M, Marschall B. Assessment of risk literacy among German medical students: a cross-sectional study evaluating numeracy skills. Hum Ecol Risk Assess. 2014;20:1139–47. 29. Hoffrage U, Lindsey S, Hertwig R, Gigerenzer G. Medicine. Communicating statistical information. Science. 2000;290:2261–2. 2. Kiesler DJ, Auerbach SM. Optimal matches of patient preferences for information, decision-making and interpersonal behavior: evidence, models and interventions. Patient Educ Couns. 2006;61:319–41. 30. Jakob N, Sprenger M. Erhebung des Zahlenverständnisses sowie der Risk Literacy von Health Professionals in Österreich. In: Penz H, Martinz H, Klein R Krajic K, editors. Tagungsband Health Literacy als Basiskompetenz (in) der Gesundheitsgesellschaft. Fachhochschule Kärnten; 2014. p. 153–6. 3. Kashaf MS, McGill E. Does shared decision making in cancer treatment improve quality of life? A systematic literature review. Med Decis Mak. 2015; 35:1037–48. 31. Wegwarth O, Gigerenzer G. Statistical illiteracy in doctors. In: Gigerenzer G, Gray JAM, editors. Better doctors, better patients, better decisions: envisioning health care 2020. Cambridge: MIT Press; 2011. p. 137–51. 4. Serrano V, Rodriguez Gutierrez R, Hargraves I, Gionfriddo MR, Tamhane S, Montori VM. Shared decision-making in the care of individuals with diabetes. Diabet Med. 2016;33:742–51. 32. Sheridan SL, Pignone M. Numeracy and the medical student's ability to interpret data. Effective Clin Pract. 2002;5:35–40. 5. Sheridan SL, Harris RP, Woolf SH. Shared Decision-Making Workgroup of the U.S. Preventive Services Task Force. Shared decision making about screening and chemoprevention. A suggested approach from the U.S. Preventive Services Task Force. Am J Prev Med. 2004;26:56–66. 33. Wegwarth O, Schwartz LM, Woloshin S, Gaissmaier W, Gigerenzer G. Do physicians understand cancer screening statistics? A national survey of primary care physicians in the United States. Ann Intern Med. 2012;156:340–9. 6. Picano E. Informed consent and communication of risk from radiological and nuclear medicine examinations: how to escape from a communication inferno. BMJ. 2004;329:849–51. 34. Weston WW. Informed and shared decision-making: the crux of patient- centered care. CMAJ. 2001;165:438–9. 35. Epstein RM, Alper BS, Quill TE. Communicating evidence for participatory decision making. JAMA. 2004;291:2359–66. 7. Apter AJ, Paasche-Orlow MK, Remillard JT, Bennett IM, Ben-Joseph EP, Batista RM, et al. Numeracy and communication with patients: they are counting on us. J Gen Intern Med. 2008;23:2117–24. 36. Moyer VA. Received: 14 August 2019 Accepted: 7 July 2020 27. Berwick DM, Fineberg HV, Weinstein MC. When doctors meet numbers. Am J Med. 1981;71:991–8. Acknowledgments 9. Schwartz LM, Woloshin S, Welch HG. Can patients interpret health information? An assessment of the medical data interpretation test. Med Decis Mak. 2005;25:290–300. The authors wish to thank Saima Ghazal and Mirta Galesic for the provision of the Berlin Numeracy Test, and Gerd Gigerenzer for the suggestion of using a case scenario to evaluate GPs. We thank Diane Williams, PhD, from Edanz Group (https://en-author-services.edanzgroup.com/) for editing a draft of this manuscript 10. Ancker JS, Kaufman D. Rethinking health numeracy: a multidisciplinary literature review. J Am Med Inform Assoc. 2007;14:713–21. 11. Ghosh AK, Ghosh K. Translating evidence-based information into effective risk communication: current challenges and opportunities. J Lab Clin Med. 2005;145:171–80. Abbreviations GP( ) G l GP(s): General practitioner(s); BNT: Berlin Numeracy Test Page 7 of 8 Page 7 of 8 Page 7 of 8 Friederichs et al. BMC Family Practice (2020) 21:143 Friederichs et al. BMC Family Practice (2020) 21:143 Consent for publication Not applicable. 22. Cottrell E, Roddy E, Rathod T, Thomas E, Porcheret M, Foster NE. Maximising response from GPs to questionnaire surveys: do length or incentives make a difference? BMC Med Res Methodol. 2015;15:1. Competing interests The authors declare that they have no competing interests. 23. Hattie J. Visible learning. London: Routledge; 2008. 24. Faul F, Erdfelder E, Lang AG, Buchner A. G*power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39:175–91. Authors’ contributions 12. Bryant GD, Norman GR. Expressions of probability: words and numbers. N Engl J Med. 1980;302:411–2. 12. Bryant GD, Norman GR. Expressions of probability: words and numbers. N Engl J Med. 1980;302:411–2. HF conceived the study and participated in its design and coordination. RB participated in the data acquisition and data analysis. JCB and BM participated in the study design. AW participated in the design and coordination of the study. All authors helped to draft the manuscript. The authors read and approved the final manuscript. 13. Ghosh AK, Ghosh K, Erwin PJ. Do medical st understand probability? QJM. 2004;97:53–5. 14. Johnston BC, Alonso-Coello P, Friedrich JO, Mustafa RA, Tikkinen KAO, Neumann I, et al. Do clinicians understand the size of treatment effects? A randomized survey across 8 countries. CMAJ. 2016;188:25–32. Funding 15. Cokely ET, Galesic M, Schulz E, Ghazal S, Garcia-Retamero R. Measuring risk literacy: the Berlin numeracy test. Judgm Decis Mak. 2012;7:25–47. Not applicable. Ethics approval and consent to participate 18. Bramwell R, West H, Salmon P. Health professionals’ and service users’ interpretation of screening test results: experimental study. BMJ. 2006;333:284 Participants were asked to complete the Berlin Numeracy Test voluntarily and anonymously. To achieve maximum transparency, all participants had to verbally agree to participate. Additionally, they provided their informed consent prior to the study by reading the background information and choosing to provide data. An extra written consent was not obtained. The study and the use of only verbal consent was approved by the Ethics Committee of the Chamber of Physicians at Westfalen-Lippe and the Medical School of Westphalian Wilhelms University in Münster (2015–308-f-S). erpretation of screening test results: experimental study. BMJ. 2006;333:2 19. Noguchi Y, Matsui K, Imura H, Kiyota M, Fukui T. Quantitative evaluation of the diagnostic thinking process in medical students. J Gen Intern Med. 2002;17:839–44. 20. Schwartz LM, Woloshin S, Black WC, Welch HG. The role of numeracy in understanding the benefit of screening mammography. Ann Intern Med. 1997;127:966–72. 21. Kooperationsgemeinschaft Mammographie Deutsches Mammographie Screening. Jahresbericht Evaluation 2012. Berlin: Deutsches Mammographie Screening Programm; 2012. Author details 1 1Study Hospital Münster, Institute for Education and Student Affairs, Medical Faculty of Münster, Malmedyweg 17-19, D-48149 Münster, Germany. 2 1Study Hospital Münster, Institute for Education and Student Affairs, Medical Faculty of Münster, Malmedyweg 17-19, D-48149 Münster, Germany. 25. Wulff HR, Andersen B, Brandenhoff P, Guttler F. What do doctors know about statistics? Stat Med. 1987;6:3–10. y y g y 2Department of Internal Medicine, Marien-Hospital, Erftstadt, NRW, Germany. 26. Windish DM, Huot SJ, Green ML. Medicine residents' understanding of the biostatistics and results in the medical literature. JAMA. 2007;298:1010–22. Received: 14 August 2019 Accepted: 7 July 2020 Received: 14 August 2019 Accepted: 7 July 2020 References What we don't know can hurt our patients: physician innumeracy and overuse of screening tests. Ann Intern Med. 2012;156:392–3. 8. Edwards AGK, Naik G, Ahmed H, Elwyn GJ, Pickles T, Hood K, et al. Personalised risk communication for informed decision making about taking screening tests. Cochrane Database Syst Rev. 2013;2:CD001865. 37. Noguchi Y, Matsui K, Imura H, Kiyota M, Fukui T. A traditionally administered short course failed to improve medical students' diagnostic performance. A Page 8 of 8 Page 8 of 8 Friederichs et al. BMC Family Practice (2020) 21:143 Friederichs et al. BMC Family Practice (2020) 21:143 quantitative evaluation of diagnostic thinking. J Gen Intern Med. 2004;19: 427–32. quantitative evaluation of diagnostic thinking. J Gen Intern Med. 2004;19: 427–32. 38. Cohen J. Statistical power analysis for the behavioral sciences. New York: Academic press; 2013. 39. Friederichs H, Ligges S, Weissenstein A. Using tree diagrams without numerical values in addition to relative numbers improves students' numeracy skills: a randomized study in medical education. Med Decis Mak. 2014;34:253–7. 40. Garcia-Retamero R, Hoffrage U. Visual representation of statistical information improves diagnostic inferences in doctors and their patients. Soc Sci Med. 2013;83:27–33. 41. Trevena LJ, Zikmund-Fisher BJ, Edwards A, Gaissmaier W, Galesic M, Han PKJ, et al. Presenting quantitative information about decision outcomes: a risk communication primer for patient decision aid developers. BMC Med Informatics Decision Making. 2013;13(Suppl 2):S7. 42. Akl EA, Oxman AD, Herrin J, Vist GE, Terrenato I, Sperati F, et al. Using alternative statistical formats for presenting risks and risk reductions. Cochrane Database Syst Rev. 2011:CD006776. 43. Edwards PJ, Roberts I, Clarke MJ, DiGuiseppi C, Wentz R, Kwan I, et al. Methods to increase response to postal and electronic questionnaires. Cochrane Database Syst Rev. 2009:MR000008. 43. Edwards PJ, Roberts I, Clarke MJ, DiGuiseppi C, Wentz R, Kwan I, et al. Methods to increase response to postal and electronic questionnaires. Cochrane Database Syst Rev. 2009:MR000008. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W3020403849
https://europepmc.org/articles/pmc7215883?pdf=render
English
null
Correlation of RAS-Pathway Mutations and Spontaneous Myeloid Colony Growth with Progression and Transformation in Chronic Myelomonocytic Leukemia—A Retrospective Analysis in 337 Patients
International journal of molecular sciences
2,020
cc-by
8,551
Correlation of RAS-Pathway Mutations and Spontaneous Myeloid Colony Growth with Progression and Transformation in Chronic Myelomonocytic Leukemia—A Retrospective Analysis in 337 Patients Sperr 17, Paul Knöbl 17, Ulrich Jäger 17 , Gregor Hörmann 18 and Peter Valent 17,19 1 Medical School, Sigmund Freud University, 1020 Vienna, Austria 1 Medical School, Sigmund Freud University, 1020 Vienna, Austria g y partment of Internal Medicine V with Hematology, Oncology and Palliative Medicine, Hospital Hietzing, 30 Vi A t i f h hi t i @ il (T G ) hi t i f h @ il (E G ) g y 2 Department of Internal Medicine V with Hematology, Oncology and Palliative Medicine, Hospital Hietzing, 1130 Vienna, Austria; forschung.hietzing@gmail.com (T.G.); hietzing.forschung@gmail.com (E.G.) 3 Department of Laboratory Medicine, Medical University of Vienna, 1090 Vienna, Austria; eva.jaeger@akhwien.at (E.J.); michael.gurbisz@meduniwien.ac.at (M.G.); 3 Department of Laboratory Medicine, Medical University of Vienna, 1090 Vienna, Austria eva.jaeger@akhwien.at (E.J.); michael.gurbisz@meduniwien.ac.at (M.G.); ilse schwarzinger@meduniwien ac at (I S ) ilse.schwarzinger@meduniwien.ac.at (I.S.) 4 Blood Transfusion Service, Blood Transfusion Service for Upper Austria, Austrian Red Cross, 4020 Linz, Austria; agnes.barna@o.roteskreuz.at 4 Blood Transfusion Service, Blood Transfusion Serv 4020 Linz, Austria; agnes.barna@o.roteskreuz.at 4 Blood Transfusion Service, Blood Transfusion Service for Upper Austria, Austrian Red Cross, 4020 Linz Austria; agnes barna@o roteskreuz at nsfusion Service for Upper Austria, Austrian Red Cros 4 Blood Transfusion Service, Blood Transfusion Service for Upper Austria, Austrian Red Cross, 5 Department of Internal Medicine III, Hanusch Hospital, 1140 Vienna, Austria; thomas.noesslinger@wgkk.at (T.N.); michael.pfeilstoecker@wgkk.at (M.P.); tuechler@stht.at (H.T.); thamersliwa@gmail.com (T.S.); felix.keil@wgkk.at (F.K.) g g 6 Department of Laboratory Medicine, Hospital Hietzing, 1130 Vienna, Austria; christoph.geissler@wienkav.at 7 Department of Internal Medicine IV Hospital Wels Grieskirchen 4600 Wels Austria; 6 Department of Laboratory Medicine, Hospital Hietzing, 1130 Vienna, Austria; christoph.geissler@wienkav.at 7 Department of Internal Medicine IV, Hospital Wels-Grieskirchen, 4600 Wels, Austria; j h ibl@kli ik t (S H ) j f th l @kli ik l t (J T ) p y , p g, , ; p g 7 Department of Internal Medicine IV, Hospital Wels-Grieskirchen, 4600 Wels, Austria; sonja.heibl@klinikum-wegr.at (S.H.); josef.thaler@klinikum-wels.at (J.T.) 8 Department of Internal Medicine I with Hematology with Stem Cell Transplantation, Hemostaseology and Medical Oncology, Ordensklinikum Linz Barmherzige Schwestern - Elisabethinen, 4020 Linz, Austria; Sigrid.Machherndl-Spandl@elisabethinen.or.at (S.M.-S.); Otto.Zach@Ordensklinikum.at (O.Z.); ansgar.weltermann@elisabethinen.or.at (A.W.); peter@bettelheim.eu (P.B.) 9 Internal Medicine V with Hematology and Oncology, Medical University of Innsbruck, 6020 Innsbruck, Austria; reinhard.stauder@i-med.ac.at 10 Department of Internal Medicine, Division of Hematology, Medical University of Graz, 8036 Graz, Austria; armin.zebisch@medunigraz.at (A.Z.); heinz.sill@medunigraz.at (H.S.) 10 Department of Internal Medicine, Division of Hematology, Medical Universit armin.zebisch@medunigraz.at (A.Z.); heinz.sill@medunigraz.at (H.S.) 10 Department of Internal Medicine, Division of Hematology, Medical University of Graz, 8036 Graz, Austria; 11 Otto-Loewi-Research Center for Vascular Biology, Immunology and Inflammation, 11 Otto-Loewi-Research Center for Vascular Biology, Immunology and Inflammation, Division of Pharmacology Medical University of Graz 8036 Graz Austria 11 Otto-Loewi-Research Center for Vascular Biology, Immunology and Inflamm Division of Pharmacology, Medical University of Graz, 8036 Graz, Austria 12 Otto Loewi Research Center for Vascular Biology, Immunology and Inflam Division of Pharmacology, Medical University of Graz, 8036 Graz, Austria 12 12 Department of Internal Medicine, Hospital Kirchdorf, 4560 Kirchdorf, Austria; Bruno.Schneeweiss@gespag.at 13 Department of Internal Medicine/Oncology, St. Correlation of RAS-Pathway Mutations and Spontaneous Myeloid Colony Growth with Progression and Transformation in Chronic Myelomonocytic Leukemia—A Retrospective Analysis in 337 Patients Klaus Geissler 1,2,*, Eva Jäger 3, Agnes Barna 4, Michael Gurbisz 3, Temeida Graf 2, Elmir Graf 2, Thomas Nösslinger 5, Michael Pfeilstöcker 5, Heinz Tüchler 5, Thamer Sliwa 5, Felix Keil 5, Christoph Geissler 6, Sonja Heibl 7, Josef Thaler 7, Sigrid Machherndl-Spandl 8, Otto Zach 8 , Ansgar Weltermann 8, Peter Bettelheim 8, Reinhard Stauder 9 , Armin Zebisch 10,11 , Heinz Sill 10 , Ilse Schwarzinger 3, Bruno Schneeweiss 12, Leopold Öhler 13, Ernst Ulsperger 14, Rajko Kusec 15, Ulrich Germing 16, Wolfgang R. International Journal of Molecular Sciences International Journal of Molecular Sciences Received: 5 April 2020; Accepted: 22 April 2020; Published: 24 April 2020 Abstract: Although the RAS-pathway has been implicated as an important driver in the pathogenesis of chronic myelomonocytic leukemia (CMML) a comprehensive study including molecular and functional analyses in patients with progression and transformation has not been performed. A close correlation between RASopathy gene mutations and spontaneous in vitro myeloid colony (CFU-GM) growth in CMML has been described. Molecular and/or functional analyses were performed in three cohorts of 337 CMML patients: in patients without (A, n = 236) and with (B, n = 61) progression/transformation during follow-up, and in patients already transformed at the time of sampling (C, n = 40 + 26 who were before in B). The frequencies of RAS-pathway mutations (variant allele frequency ≥20%) in cohorts A, B, and C were 30%, 47%, and 71% (p < 0.0001), and of high colony growth (≥20/105 peripheral blood mononuclear cells) 31%, 44%, and 80% (p < 0.0001), respectively. Increases in allele burden of RAS-pathway mutations and in numbers of spontaneously formed CFU-GM before and after transformation could be shown in individual patients. Finally, the presence of mutations in RASopathy genes as well as the presence of high colony growth prior to transformation was significantly associated with an increased risk of acute myeloid leukemia (AML) development. Together, RAS-pathway mutations in CMML correlate with an augmented autonomous expansion of neoplastic precursor cells and indicate an increased risk of AML development which may be relevant for targeted treatment strategies. Keywords: CMML; AML; RAS-pathway mutations; CFU-GM Keywords: CMML; AML; RAS-pathway mutations; CFU-GM Correlation of RAS-Pathway Mutations and Spontaneous Myeloid Colony Growth with Progression and Transformation in Chronic Myelomonocytic Leukemia—A Retrospective Analysis in 337 Patients Josef Hospital, 1130 Vienna, Austria; leopold.oehler@sjk-wien.at p j 14 Department of Internal Medicine, Hospital Horn, 3580 Horn, Austria; ernst.ulsperger@aon.at ep e o e e c e, osp o , 3580 o , s ; e s spe ge @ o 15 School of Medicine, University of Zagreb, University Hospital Dubrava, 10000 Zagreb, Croatia; Rajko.Kusec@irb.hr 16 Department of Hematology, Oncology, and Clinical Immunology, Heinrich-Heine-University, 40225 Düsseldorf, Germany; germing@med.uni-duesseldorf.de 16 Department of Hematology, Oncology, and Clinical Immunology, H 40225 Düsseldorf, Germany; germing@med.uni-duesseldorf.de Int. J. Mol. Sci. 2020, 21, 3025; doi:10.3390/ijms21083025 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2020, 21, 3025 2 of 14 17 Department of Internal Medicine I, Division of Hematology and Hemostaseology, Medical University of Vienna, 1090 Vienna, Austria; wolfgang.r.sperr@meduniwien.ac.at (W.R.S.); paul.knoebl@meduniwien.ac.at (P.K.); ulrich.jaeger@meduniwien.ac.at (U.J.); peter.valent@meduniwien.ac.at (P.V.) p 18 Central Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Innsbruck, 6020 Innsbruck, Austria; gregor.hoermann@meduniwien.ac.at g g 19 Ludwig Boltzmann Institute for Hematology and Oncology (LBI HO), Medical University of Vienna, 1090 Vienna, Austria * Correspondence: klaus.geissler@wienkav.at; Tel.: +43-01-80110-3122; Fax: +43-01-80110-2671 Keywords: CMML; AML; RAS-pathway mutations; CFU-GM 1. Introduction p y yp pp In juvenile myelomonocytic leukemia (JMML), a RAS-pathway driven hematologic malignancy in children, growth factor-independent formation of granulocyte/macrophage colony-forming units (CFU-GM) in semisolid cultures is considered a hallmark of the disease. Therefore, autonomous CFU-GM formation has been included as a diagnostic criterion in previous WHO classifications [2,4]. If considering this test as a functional parameter of RAS-pathway hyperactivation indications for aberrant RAS-pathway signaling in CMML can actually be traced back for 30 years when we described this in vitro phenomenon in 1988 in a subset of our CMML patients [18]. Later, we have shown that spontaneous CFU-GM formation in CMML is a GM-CSF-related in vitro phenomenon [19]. It has also been described that CMML progenitors are hypersensitive against GM-CSF in a study of Padron et al. [20]. In preclinical mouse models, molecular alterations of RASopathy genes in murine hematopoietic cells are leading to a myelomonocytic leukemia like phenotype in vivo and to spontaneous myeloid colony formation due to GM-CSF hypersensitivity in vitro [21–25]. Recently we were able to demonstrate a close correlation between increased spontaneous colony formation in CMML patients and the presence of RAS-pathway mutations [26]. Together these findings strongly suggest that high spontaneous in vitro CFU-GM formation in CMML reflect RAS-pathway hyperactivation at a functional level. Although a correlation of mutations in RAS-pathway genes and spontaneous myeloid colony formation has been shown by us in CMML patients without transformation, a comprehensive analysis of the RAS-pathway in patients with CMML derived AML has not been performed. Molecular as well as functional data on the RAS-pathway aberrations, however, in this particular cohort would be of significant interest considering the dismal prognosis of patients and the availability of RAS pathway inhibitors. In the “Austrian Biodatabase for Chronic Myelomonocytic Leukemia” (ABCMML) we retrospectively and prospectively collect hematological, clinical, molecular, and biological information of patients with CMML from different centers in a real world setting [27]. Due to the retrospective character of our database it contains data that are obtained from patients being in different phases of CMML evolution at the time of inclusion and subsequent follow up. 1. Introduction Chronic myelomonocytic leukemia (CMML) is a hematopoietic malignancy of the elderly that is characterized by overlapping features of myelodysplastic syndromes (MDS) and myeloproliferative neoplasms (MPN) and an inherent risk of transformation to secondary acute myeloid leukemia (sAML) [1–7]. The median overall survival of CMML patients is approximately 30 months. Allogeneic hematopoietic stem cell transplantation is the only curative therapy approach, but this form of treatment is only rarely feasible because of age and/or comorbidities that are often found in patients with CMML. In patients ineligible for transplantation, intensive chemotherapy results in low response rates and short response duration. Hydroxyurea is commonly used as palliative drug to control myeloproliferation in advanced CMML and sAML [8]. The cytidine analogues azacitidine (AZA) and decitabine have demonstrated some efficacy in delaying disease progression in advanced CMML and have been approved for the treatment of this disease [9,10]. Since many patients eventually die from transformation and/or progression of disease a better understanding of disease evolution is crucial to improve treatment and prognosis. At the molecular level CMML is a heterogeneous disease similar to other hematologic neoplasms. However, many of the gene alterations detected in CMML affect similar or the same oncogenic machineries and signaling pathways [11]. Over the past few years, a large number of mutations in genes encoding epigenetic regulators (TET2, ASXL1, EZH2, UTX, IDH1, IDH2, and DNMT3A), 3 of 14 Int. J. Mol. Sci. 2020, 21, 3025 splicing factors (SF3B1, SRSF2, ZRSF2, and U2AF1), and signaling molecules (NRAS, KRAS, CBL, JAK2, and FLT3) have been identified in clonal cells in CMML [12]. In a few patients, progression of CMML was associated with an increment of RAS mutational burden, suggesting a role of RAS-pathway hyperactivation in progression and transformation to AML [13]. Although the impact of RAS-pathway mutations in the risk to develop secondary leukemia has been studied in larger cohorts of patients with CMML, patients who had already transformed to AML were usually not included and thus these studies could not provide an answer to the question to what extent the RAS-pathway can indeed contribute to transformation [14–16]. In one study, the molecular features and mutational patterns were analyzed during blast transformation of CMML and the RAS-pathway was apparently involved [17]. In none of these studies, however, functional tests of RAS-pathway hyperactivation were applied. 1. Introduction Therefore, we divided patients into three cohorts based on criteria recently proposed by an international consortium: [28] patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to sAML at the time of sampling (cohort C). Using data (molecular, n = 313; CFU-GM, n = 196) from 337 CMML patients we compared the frequencies of RASopathy gene mutations (variant allele frequency (VAF) ≥20%) and of high CFU-GM growth (≥20/105 peripheral blood mononuclear cells (PBMNC)) in patient cohorts A, B and C and were able to monitor disease evolution in individual patients in whom serial samples were available. 2.1. Impact of Disease Stage on Survival in Patients with CMML The total cohort comprised 209 (62%) men and 128 (38%) women with a median age of 73.0 years (range: 36–93 years) which is comparable to other cohorts reported in the literature [14–17]. 4 of 14 Int. J. Mol. Sci. 2020, 21, 3025 As mentioned above the patients were grouped into one of three categories: patients without evidence of AML development and/or progression related death at any time (cohort A, n = 236), patients who developed AML and/or progression related death during follow up (cohort B, n = 61) and patients after transformation to sAML (cohort C, n = 40). In order to demonstrate the clinical significance of this categorization we first calculated Kaplan Meier plots for each of the patient categories. As shown in supplemental Figure S1 there was a clear discrimination between the three categories with a median survival of 30 months in cohort A, 21 months in cohort B, and 5 months in cohort C (p < 0.0001). These finding suggest that the three categories that we have chosen in this study really represent different stages of CMML evolution. 2.2. Hematologic and Cytogenetic Characteristics in Patient-Subgroups The laboratory characteristics of these patient groups are shown in Supplementary Table S1, including white blood count (WBC), hemoglobin (Hb) level, platelet count, percentage of PB blasts, and percentage of monocytes. As compared to CMML patients without evidence of transformation and/or disease-related death at any time (cohort A) patients who had already transformed to sAML (cohort C) had significantly (<0.05) higher WBC counts and PB blast cell percentages, and significantly lower Hb levels and platelet counts. The WBC counts and PB blast cell percentages were also significantly higher in CMML patients who later progressed during follow up (group B) than in patients of group A. However, cohort B was not significantly different to cohort A with regard to Hb levels, and platelet counts. Since group B was relatively small we cannot exclude the possibility that significant differences would have been found in this category when the number of these patients would have been larger. The frequencies of high risk cytogenetic abnormalities such as + 8, −7/del(7q), and complex karyotype (≥3 abnormalities) in cytogenetic analyses were 16% (19/122) in cohort A, 41% (16/39) in cohort B, and 30% (8/27) in cohort C (p = 0.003). 2.3. RAS-Pathway Mutations Correlate with AML Evolution in Patients with CMML Molecular analysis by NGS was performed in 295 patients. Because of their functional significance RAS signaling pathway components including NRAS, KRAS, NF-1, PTPN11, and CBL were summarized as RASopathy genes within one category. As shown in Figure 1 there was a continuous increase in the frequencies of RAS-pathway mutations in patients who never developed AML (A) to patients with overt transformation, with 30% (59/198) in cohort A, 47% (27/57) in cohort B, and 71% (41/58) in cohort C, respectively (p < 0.0001; A vs. B, p = 0.134; B vs. C, p = 0.011). The proportions of mutations of individual genes can be seen in the stacked histogram. The numbers for the different cohorts and genes, respectively, were: cohort A: NRAS 9.1%, KRAS 6.1%, CBL 11.6%, NF1 2.5%, and PTPN11 0.5%; cohort B: NRAS 22.8%, KRAS 5.3%, CBL 12.3%, NF1 3.5%, and PTPN11 3.5%; cohort C: NRAS 25.9%, KRAS 19.0%, CBL 12.0%, NF1 7.0%, and PTPN11 7.0%. Supplementary Table S2 shows the frequencies of non-RASopathy gene mutations with a frequency of at least 10% in patient samples. There was a significant increase in the frequency of RUNX1 mutations towards progression/transformation of CMML into AML but no increase in the prevalence of mutations of SETBP1, TET2, EZH2, ASXL1, SRSF2, and TP53. As shown in Figure 2 the presence of mutations in RASopathy genes prior to transformation w associated with an increased risk of AML development (p = 0.007). A comprehensive mutation status of CMML patients with sAML (cohort C) is shown in Figure 3. Among RAS-pathway-related mutations NRAS mutations (n = 15) were the most common ones followed by KRAS (n = 11), CBL (n = 8), NF1 (n = 6), and PTPN11 (n = 4), respectively. Detailed information regarding the variants detected in these samples are given in Supplementary Table S3. In six patients, serial samples during progression were available to perform NGS analysis (Figure 4). In five/six patients, there was an increase by more than two-fold in the allele burden of RAS-pathway aberrations during disease progression/transformation. 5 of 14 Int. J. Mol. Sci. 2020, 21, 3025 Figure 1. Frequencies of RASopathy gene mutations in the 3 patient cohorts: patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to secondary acute myeloid leukemia (AML) at the time of sampling (cohort C). 2.3. RAS-Pathway Mutations Correlate with AML Evolution in Patients with CMML Individual genes are indicated by different colors. Cohort C includes 18 patients from cohort B who initially had no evidence of transformation but developed AML during observation. In patients with more than one mutation in RASopathy genes the mutation with the highest variant allele frequency (VAF) was used for this analysis. Figure 1. Frequencies of RASopathy gene mutations in the 3 patient cohorts: patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to secondary acute myeloid leukemia (AML) at the time of sampling (cohort C). Individual genes are indicated by different colors. Cohort C includes 18 patients from cohort B who initially had no evidence of transformation but developed AML during observation. In patients with more than one mutation in RASopathy genes the mutation with the highest variant allele frequency (VAF) was used for this analysis. Figure 2. Time to AML transformation in chronic myelomonocytic leukemia (CMML) patients stratified by the presence or absence of RASopathy gene mutations. Figure 2. Time to AML transformation in chronic myelomonocytic leukemia (CMML) patients stratified by the presence or absence of RASopathy gene mutations. Int. J. Mol. Sci. 2020, 21, 3025 6 of 14 Figure 3. Comprehensive mutation status of genes in patients with CMML derived AML. Each column corresponds to one patient. Colored squares indicate mutated, white squares wild-type genes. The colors of mutant genes indicate the most affected functional categories. Red, green, blue, and yellow represent the RAS-pathway, epigenetic regulators, spliceosome, and other components, respectively. Missing data are indicated by gray squares. Figure 3. Comprehensive mutation status of genes in patients with CMML derived AML. Each column corresponds to one patient. Colored squares indicate mutated, white squares wild-type genes. The colors of mutant genes indicate the most affected functional categories. Red, green, blue, and yellow represent the RAS-pathway, epigenetic regulators, spliceosome, and other components, respectively. Missing data are indicated by gray squares. Figure 3. Comprehensive mutation status of genes in patients with CMML derived AML. Each column corresponds to one patient. Colored squares indicate mutated, white squares wild-type genes. The colors of mutant genes indicate the most affected functional categories. Red, green, blue, and yellow represent the RAS-pathway, epigenetic regulators, spliceosome, and other components, respectively. Missing data are indicated by gray squares. Figure 3. 2.3. RAS-Pathway Mutations Correlate with AML Evolution in Patients with CMML The boxplot in Figure 5b shows a large variation in colony numbers between single patients in the different cohorts, however, median CFU-GM numbers per 105 MNC clearly increased with 4.5 in cohort A, 19 in cohort B, and 287 in cohort C (p < 0.0001). Figure 5. (a) Frequencies of high spontaneous myeloid colony formation (>20/105 peripheral blood mononuclear cells (PBMNC)) in patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to secondary AML at the time of sampling (cohort C). Cohort C includes 13 patients from cohort B who initially had no evidence of transformation but developed AML during observation. (b) Box plots showing the distribution of spontaneous colony numbers in the 3 patient cohorts including median values, minimum values, maximum values, as well as upper and lower quartiles, respectively. Cultures were plated in duplicates or triplicates, respectively, at 25–100 × 103 PBMNC/mL. Aggregates with more than 40 translucent, dispersed cells were counted as CFU-GM. CFU-GM data from patients are expressed as mean values from cultures. Figure 5. (a) Frequencies of high spontaneous myeloid colony formation (>20/105 peripheral blood mononuclear cells (PBMNC)) in patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to secondary AML at the time of sampling (cohort C). Cohort C includes 13 patients from cohort B who initially had no evidence of transformation but developed AML during observation. (b) Box plots showing the distribution of spontaneous colony numbers in the 3 patient cohorts including median values, minimum values, maximum values, as well as upper and lower quartiles, respectively. Cultures were plated in duplicates or triplicates, respectively, at 25–100 × 103 PBMNC/mL. Aggregates with more than 40 translucent, dispersed cells were counted as CFU-GM. CFU-GM data from patients are expressed as mean values from cultures. As shown in Figure 6 the presence of high spontaneous CFU-GM in patients prior to transformation was associated with an increased risk of AML development (p = 0.011). As shown in Figure 6 the presence of high spontaneous CFU-GM in patients prior to transformation was associated with an increased risk of AML development (p = 0.011). 2.3. RAS-Pathway Mutations Correlate with AML Evolution in Patients with CMML Comprehensive mutation status of genes in patients with CMML derived AML. Each column corresponds to one patient. Colored squares indicate mutated, white squares wild-type genes. The colors of mutant genes indicate the most affected functional categories. Red, green, blue, and yellow represent the RAS-pathway, epigenetic regulators, spliceosome, and other components, respectively. Missing data are indicated by gray squares. Figure 4. Serial mutation analysis in individual patients. The variant allele frequencies of 6 individual patients are shown at different time points during the course of their disease. Open squares indicate time points pre transformation and closed squares time points after transformation. Initiation of treatment is indicated by arrows: AZA—azacitidine, HU—hydoxyurea, LEN—lenalidomide, and MIDAC—mitoxantrone + cytarabine. Figure 4. Serial mutation analysis in individual patients. The variant allele frequencies of 6 individual patients are shown at different time points during the course of their disease. Open squares indicate time points pre transformation and closed squares time points after transformation. Initiation of treatment is indicated by arrows: AZA—azacitidine, HU—hydoxyurea, LEN—lenalidomide, and MIDAC—mitoxantrone + cytarabine. High Autonomous in Vitro Colony Formation Correlates with AML Evolution in Patients with CMML In vitro cultures were performed in 183 patients with CMML. Recently we demonstrated a close correlation between high spontaneous colony formation in CMML patients and the presence of RAS-pathway mutations [26]. This correlation was also seen in this study using 162 samples in which both NGS data and colony data were available. The median number of spontaneously 7 of 14 Int. J. Mol. Sci. 2020, 21, 3025 formed CFU-GM/105 MNC was 47 (range 0–1127) in RAS-positive patients as compared to 4 (0–812) in RAS-negative patients (p < 0.0001). Unstimulated in vitro myeloid colony formation in RAS-positive CMML patients is also much higher than the spontaneous formation of CFU-GM in normal individuals (median 4.8/105 PBMNC, range 3.5–8.5), which has been reported by us previously [29]. The frequency of spontaneous CFU-GM growth ≥0/105 in the 3 patient cohorts is shown in Figure 5a. There was a continuous increase in the frequency of high spontaneous myeloid colony formation (≥20/105 PBMNCs) from patients in group A (42/135, 31%) to patients in group B (16/36, 44%) and to patients in group C (20/25, 80%), respectively (p < 0.0001; A vs. B, p = 0.141; B vs. C, p = 0.005). 2.3. RAS-Pathway Mutations Correlate with AML Evolution in Patients with CMML Table 1 shows the numbers of spontaneously formed CFU-GM in eight patients in whom in vitro cultures could be performed before and after transformation to sAML. Seven of eight patients were characterized by NGS, and in all of them mutations of RASopathy genes were detected (NRAS 4, KRAS 1, CBL 1, PTPN11 1). As compared to pre transformation values, a marked increase of colony numbers following transformation to sAML can be seen in all eight patients (median [range] CFU-GM/105 MNC pre transformation 41.5 (1–622), post transformation 263 (48–4553); p = 0.012). 8 of 14 Int. J. Mol. Sci. 2020, 21, 3025 Figure 6. Time to AML transformation in CMML patients stratified by the presence or absence of spontaneous CFU-GM growth >20/105 mononuclear cells. i T AML f CMML fid b h Figure 6. Time to AML transformation in CMML patients stratified by the presence or absence of spontaneous CFU-GM growth >20/105 mononuclear cells. Table 1. White blood cell counts and numbers of spontaneously formed CFU-GM in 8 patients in whom in vitro cultures could be performed before and after transformation to secondary acute myeloid leukemia. Table 1. White blood cell counts and numbers of spontaneously formed CFU-GM in 8 patients in whom in vitro cultures could be performed before and after transformation to secondary acute myeloid leukemia. Table 1. White blood cell counts and numbers of spontaneously formed CFU-GM in 8 patients in whom in vitro cultures could be performed before and after transformation to secondary acute myeloid leukemia. Patient Genotype WBC G/L Pretransform. (%PB blasts) WBC G/L Posttransform. (%PB blasts) CFU–GM/ 105 MNC Pretransform. CFU-GM/ 105 MNC Posttransform. UPN1-026 CBL 12.3 (2%) 178 (60%) 35 302 PN1-033 NRAS 7.8 (4%) 160 (10%) 200 533 UPN1-038 NA 93.8 (1%) 50.7 (30%) 622 4553 UPN1-071 PTPN11 74.0 (8%) 161 (18%) 11 48 UPN1-128 NRAS 57.5 (3%) 155 (13%) 59 272 UPN1-142 NRAS 20.3 (0%) 80.7 (1%) 48 202 UPN1-171 NRAS 55.5 (0%) 50.0 (54%) 1 254 UPN1-468 NRAS 8.5 (0%) 24.5 (57%) 8 381 NA—not available. 3. Discussion These data are in agreement with data from others demonstrating that mutations of critical genes involved in signaling cascades, including the RAS-pathway, were also acquired as late events in MDS [34–37]. Moreover, recent preclinical models suggest that activating RAS mutations and somatic loss-of-function mutations in TET2 exert cooperating effects and accelerated disease progression [38,39]. During the progression of CMML to AML spontaneous CFU-GM formation was found to be continuously increasing. There is now sufficient evidence to consider spontaneous CFU-GM formation as a functional test of RAS-pathway hyperactivation in CMML [26]. Due to the fact that in CMML more than one functionally relevant molecular aberration can be detected in many patients we think that the spontaneous in vitro colony formation of CFU-GM may help predict that the RAS-pathway is involved in disease evolution which may in turn have clinical (prognostic and therapeutic) implications. Another functional assay for RAS-pathway hyperactivity may be an increased expression of phosphorylated STAT5 which is a downstream target of the RAS-pathway. Although we did not perform such experiments in this study, Padron et al. have demonstrated that neoplastic cells in CMML exhibit increased STAT5 activity [20]. We confirm previous studies which have reported that RAS-pathway mutations are common in CMML but are usually rare in other myeloid malignancies such as CML or Ph-negative MPN [30]. In one study, NRAS mutations were not found in any of 86 CML patients in blast crisis examined [40]. Only one patient, in whom the initial diagnosis of CML blast crisis had been revised to CMML, displayed an NRAS mutation within codon 13 [40]. In another study in which targeted cancer exome sequencing was performed in BCR-ABL-negative MPNs, NRAS mutations were found in only 4.7% of 168 patients with primary myelofibrosis (MF) and in none of the patients with polycythemia vera (PV) [41]. In addition to mutations of the RAS-pathway, a large number of mutations in other genes were found in our study which is in line with previous reports [12]. In contrast to RAS-pathway mutations, however, aberrations in the majority of other genes were not significantly different in our three patient groups. This argues against a major role of these aberrations in the transformation process although contributions of such mutations in individual patients cannot be excluded. Significant changes were only observed for RUNX1. 3. Discussion CMML is an incurable stem cell-derived MDS/MPN overlap neoplasm characterized by monocyte expansion and an increased risk to transform to AML [1–7]. Although the RAS pathway has been implicated in the pathogenesis of CMML, little is known about mechanistic and functional correlates and how progression impacts survival in these patients. We have analyzed data from CMML patients collected in our ABCMML-registry and found that RAS-pathway mutations and thus RAS hyperactivation correlates with autonomous expansion of neoplastic stem/progenitor cells and with disease progression. Moreover, we show that the allelic burden of RAS-pathway-related mutations increase during progression in individual patients with CMML. The frequency of RAS-pathway mutations in our patients were within the range of other published series in which marked variations can be found ranging from 23%–60% [30–32]. Reasons for these variations may be differences in the criteria applied for CMML diagnosis, differences in disease stages investigated, and differences in the number of target genes/regions covered by the NGS panels applied. In this regard, it is worth noting that the diagnostic criteria for CMML have significantly changed during the last three decades, in Int. J. Mol. Sci. 2020, 21, 3025 9 of 14 particular when comparing the former FAB classification with the most recent WHO classification of CMML [1–5]. In view of the lack of specific disease markers, the diagnosis of CMML is still based primarily on clinical parameters and morphologic features of clonal cells. The hybrid nature of CMML with features of both a MDS and a MPN creates also diagnostic difficulties. A most important diagnostic parameter is the presence of PB monocytosis which is a prerequisite diagnostic criterion of CMML [7,33]. Previous observations have already implicated the RAS-pathway in the pathogenesis of CMML [12]. Our data confirm this notion and provide evidence that RAS pathway mutations are critical involved in CMML evolution and progression. The frequency of RASopathy genes in our patients was clearly increasing during progression and was observed in more the 70% of all patients with CMML-derived sAML. In six of these patients, serial mutation analyses could be performed. The fact that VAFs of RASopathy genes in all of these cases were lower at first presentation than VAFs at the time of AML and in general lower than VAFs of epigenetic and splicing factors suggests that mutations in genes relevant to the RAS signaling pathway were later acquired and were driving events for disease progression. 3. Discussion A trend towards a higher risk to transform to AML was reported for CMML patients with RUNX1 mutated cells, especially patients with C-terminal mutations [42]. An association between ASXL1 mutations and acute transformation in CMML has been described but there was only a trend found in our study [43]. We are aware of the limitations of our study. For example, most of the information used in this study was derived from real world data that were not collected systematically or prospectively. In addition, data from patient records were obtained over many years and from many different centers. However, real world data have recently been recognized as an important way to get insights into the natural history of rare diseases [44]. CMML is a rare disease and adequate patient numbers for a Int. J. Mol. Sci. 2020, 21, 3025 10 of 14 systematic and prospective study are not easy to collect within a limited time frame. In a substantial number of patients we had serial samples from several time points giving us the possibility to determine the variant allele frequencies of gene mutations at different time points during the course of their disease. Since many patients had already died when we started our analyses, germline controls were not available in most patients. However, almost all mutations in RASopathy genes which were found in our study have already been reported as somatic aberrations, and the allelic burden of the mutant forms detected in our patients also argue against germline defects. Our clinical data confirm the dismal prognosis of CMML patients who transform to AML. Such patients had a median survival of only 5.0 months which is in accordance with other published series [45]. These data clearly indicate the medical need for improving treatment concepts for these patients. The findings of this study may contribute to achieve this goal by suggesting that hyperactivation of the RAS signaling pathway plays a major role in the transformation process from CMML into AML. The study of RAS-pathway inhibitors seems to be a logical step in our efforts. In fact, a recent study has investigated the effect of such an inhibitor in patients with RAS-pathway driven hematological malignancies including CMML and has shown some responses in these poor risk patients who are extremely difficult to treat especially when they progress to AML [46]. 4.1. Patients 2020, 21, 3025 11 of 14 11 of 14 4.3. Molecular Studies Genomic DNA was isolated from mononuclear cell (MNC) fractions of these blood samples according to standard procedures. The mutational status of CMML-related protein coding genes was determined by targeted amplicon sequencing using the MiSeq platform (Illumina, San Diego, CA, USA). Details regarding gene panel, library preparation and data processing have been reported previously [27]. To minimize the chance to capture mutations which may be associated with non-neoplastic conditions we chose a VAF cutoffof ≥20% which has been shown to provide a specificity of 86% [48]. 4.4. Colony Assay In one of our centers (Medical University of Vienna) the assessment of hematopoietic colony formation in vitro has been an integral part of the diagnostic work up in patients with suspected myeloid malignancies for many years [49]. CFU-GM growth was assessed in semisolid cultures without growth factors as previously described [19]. A detailed description of the technique is provided in the Supplementary Materials. 4.2. Cytogenetic Analysis Cytogenetic studies were performed using G-banding according to standard techniques on BM cells for 24–48 h in unstimulated culture. Chromosome aberrations were classified according to the International System for Human Cytogenetic Nomenclature (ISCN). CMML-specific cytogenetic risk classification was low for normal karyotype and isolated -Y, intermediate for other abnormalities and high for trisomy 8, complex karyotype (≥3 abnormalities), and abnormalities of chromosome 7 [47]. In general samples were taken before any disease-modifying treatment such as allogeneic stem cell transplantation, aggressive chemotherapy, or hypomethylating agents. 4.1. Patients We employed data from 337 CMML patients from the ABCMML which has been shown to be a representative and useful real-life data source for biomedical research [27]. In this database, we retrospectively collected epidemiologic, hematologic, biochemical, clinical, immunophenotypic, cytogenetic, molecular and biologic data of patients with CMML from different centers. Internal Review Board approval was obtained at each institution (ethics committee of the city of Vienna, EK 15-059-VK). Clinical and laboratory routine parameters were obtained from patient records. A detailed central manual retrospective chart review was carried out to ensure data quality before analysis of data from institutions. Data curation included the extraction of discrete data elements from patient records, a check for accuracy and consistency of data, and a verification that baseline data were reflective of CMML that was strictly defined according to WHO criteria. Progression and/or transformation was defined as disease related mortality, transformation to AML, or a combination of criteria as noted by Savona et al. [28]. Briefly the proposed criteria for disease progression include major criteria such as increase in blast count, evidence of cytogenetic evolution, new extramedullary disease and minor criteria such as transfusion dependence, significant loss of maximal response on cytopenias, reduction in Hb by ≥1.5 g/dL from best response or from baseline, increasing symptoms, and evidence of clonal evolution, respectively. A combination of 2 major criteria, 1 major and 2 minor criteria, or 3 minor criteria have to be met. By using these criteria our study population was categorized in patients without (A, n = 236) and with (B, n = 61) progression/transformation during follow-up, and in patients already transformed at the time of sampling (C, n = 40 + 26 who were before in B). Clinical and laboratory routine parameters were obtained from the patients’ records. Blood counts were taken at the time of sampling and survival analysis was calculated from the sampling date. The specimens investigated in order to analyze the frequencies of RASopathy gene mutations included individual samples obtained from 277 patients, and two serial samples, one at the time before (group B) and another after transformation (group C) from 18 patients, respectively. Likewise, the specimens investigated in order to analyze the frequencies of high colony growth included individual samples obtained from 170 patients, and two serial samples, one at the time before (group B) and another after transformation (group C) from 13 patients, respectively. Int. J. Mol. Sci. References 1. Bennett, J.M.; Catovsky, D.; Daniel, M.T.; Flandrin, G.; Galton, D.A.; Gralnick, H.R.; Sultan, C. Proposals for the Classification of the Myelodysplastic Syndromes. Br. J. Haematol. 1982, 51, 189–199. [CrossRef] [PubMed] 2. Vardiman, J.W.; Harris, N.L.; Brunning, R.D. The World Health Organization (WHO) Classification of the Myeloid Neoplasms. Blood 2002, 100, 2292–2302. [CrossRef] [PubMed] 3. Vardiman, J.W.; Thiele, J.; Arber, D.A.; Brunning, R.D.; Borowitz, M.J.; Porwit, A.; Harris, N.L.; Le Beau, M.M.; Hellström-Lindberg, E.; Tefferi, A.; et al. The 2008 Revision of the World Health Organization (WHO) Classification of Myeloid Neoplasms and Acute Leukemia: Rationale and Important Changes. Blood 2009, 114, 937–951. [CrossRef] 4. Arber, D.A.; Orazi, A.; Hasserjian, R.; Thiele, J.; Borowitz, M.J.; Le Beau, M.M.; Bloomfield, C.D.; Cazzola, M.; Vardiman, J.W. The 2016 Revision to the World Health Organization Classification of Myeloid Neoplasms and Acute Leukemia. Blood 2016, 127, 2391–2405. [CrossRef] 5. Padron, E.; Steensma, D.P. Cutting the Cord from Myelodysplastic Syndromes: Chronic Myelomonocytic Leukemia-Specific Biology and Management Strategies. Curr. Opin. Hematol. 2015, 22, 163–170. [CrossRef] 5. Padron, E.; Steensma, D.P. Cutting the Cord from Myelodysplastic Syndromes: Chronic Myelomonocytic Leukemia-Specific Biology and Management Strategies. Curr. Opin. Hematol. 2015, 22, 163–170. [CrossRef] 6. Itzykson, R.; Fenaux, P.; Bowen, D.; Cross, N.C.P.; Cortes, J.; De Witte, T.; Germing, U.; Onida, F.; Padron, E.; Platzbecker, U.; et al. Diagnosis and Treatment of Chronic Myelomonocytic Leukemias in 6. Itzykson, R.; Fenaux, P.; Bowen, D.; Cross, N.C.P.; Cortes, J.; De Witte, T.; Germing, U.; Onida, F.; Padron, E.; Platzbecker, U.; et al. Diagnosis and Treatment of Chronic Myelomonocytic Leukemias in Adults: Recommendations From the European Hematology Association and the European LeukemiaNet. HemaSphere 2018, 1. [CrossRef] 7. Valent, P.; Orazi, A.; Savona, M.R.; Patnaik, M.M.; Onida, F.; van de Loosdrecht, A.A.; Haase, D.; Haferlach, T.; Elena, C.; Pleyer, L.; et al. Proposed Diagnostic Criteria for Classical Chronic Myelomonocytic Leukemia (CMML), CMML Variants and Pre-CMML Conditions. Haematologica 2019, 104, 1935–1949. [CrossRef] 8. Wattel, E.; Guerci, A.; Hecquet, B.; Economopoulos, T.; Copplestone, A.; Mahé, B.; Couteaux, M.E.; Resegotti, L.; Voglova, V.; Foussard, C.; et al. A Randomized Trial of Hydroxyurea versus VP16 in Adult Chronic Myelomonocytic Leukemia. Groupe Français Des Myélodysplasies and European CMML Group. Blood 1996, 88, 2480–2487. [CrossRef] 9. Fenaux, P.; Mufti, G.J.; Hellstrom-Lindberg, E.; Santini, V.; Finelli, C.; Giagounidis, A.; Schoch, R.; Gattermann, N.; Sanz, G.; List, A.; et al. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. 4.5. Statistical Analysis The log-rank test was used to determine whether individual parameters were associated with overall survival (OS) and time to AML transformation. OS was defined as the time from sampling to death (uncensored) or last follow up (censored). Time to AML transformation was defined as the time of sampling to the time of transformation to secondary AML (uncensored) or death/last contact (censored). Dichotomous variables were compared between different groups with the use of the chi-square test. The Mann-Whitney-U-test was used to compare 2 and the Kruskal–Wallis test to compare more than 2 unmatched groups when continuous variables were not normally distributed. Results were considered significant at p < 0.05. Statistical analyses were performed with the SPSS version 19.0.0 (SPSS Inc, Chicago, Illinois, USA); the reported p-values were 2-sided. Supplementary Materials: The following are available online at http://www.mdpi.com/1422-0067/21/8/3025/s1, Figure S1: Kaplan-Meier survival curves of the 3 patient cohorts. Table S1: Laboratory values of the 3 patient cohorts: patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to secondary AML at the time of sampling (cohort C). Table S2: Frequencies of other than RASopathy gene mutations in the 3 patient cohorts: patients without evidence of progression (cohort A), patients who developed disease progression (transformation and/or disease-related death) during follow up (cohort B), and patients who had already transformed to secondary AML at the time of sampling (cohort C). Table S3: Detailed informations (region, ENST, ENSP, variant allele frequency) of molecular aberrations detected in RASopathy genes in samples of patients with CMML derived AML. Author Contributions: K.G. designed the project, collected, analyzed and interpreted the data and wrote the manuscript. P.V. collected data and wrote the manuscript. E.J., A.B., and M.G. performed the experiments. T.G., and E.G. performed the administration of data. T.N., M.P., H.T., T.S., F.K., C.G., S.H., J.T., S.M.-S., O.Z., A.W., P.B., R.S., A.Z., H.S., I.S., B.S., L.Ö., E.U., R.K., U.G., W.R.S., P.K., U.J., and G.H. collected data. All authors reviewed the draft manuscript and approved the final version for submission. All authors have read and agreed to the published version of the manuscript. Funding: This study was supported by the “Gesellschaft zur Erforschung der Biologie und Therapie Tumorkrankheiten - ABCMML-112015” and the “Austrian Science Fund (FWF) – grant F4704-B20”. 12 of 14 Int. J. Mol. Sci. 2020, 21, 3025 References Padron, E.; Painter, J.S.; Kunigal, S.; Mailloux, A.W.; McGraw, K.; McDaniel, J.M.; Kim, E.; Bebbington, C.; Baer, M.; Yarranton, G.; et al. GM-CSF–Dependent PSTAT5 Sensitivity Is a Feature with Therapeutic Potential in Chronic Myelomonocytic Leukemia. Blood 2013, 121, 5068–5077. [CrossRef] 21. Wang, J.; Liu, Y.; Li, Z.; Du, J.; Ryu, M.-J.; Taylor, P.R.; Fleming, M.D.; Young, K.H.; Pitot, H.; Zhang, J. Endogenous Oncogenic Nras Mutation Promotes Aberrant GM-CSF Signaling in Granulocytic/Monocytic Precursors in a Murine Model of Chronic Myelomonocytic Leukemia. Blood 2010, 116, 5991–6002. [CrossRef] [PubMed] 22. Van Meter, M.E.M.; Díaz-Flores, E.; Archard, J.A.; Passegué, E.; Irish, J.M.; Kotecha, N.; Nolan, G.P.; Shannon, K.; Braun, B.S. K-RasG12D Expression Induces Hyperproliferation and Aberrant Signaling in Primary Hematopoietic Stem/Progenitor Cells. Blood 2007, 109, 3945–3952. [CrossRef] [PubMed] 23. Li, Q.; Haigis, K.M.; McDaniel, A.; Harding-Theobald, E.; Kogan, S.C.; Akagi, K.; Wong, J.C.Y.; Braun, B.S.; Wolff, L.; Jacks, T.; et al. Hematopoiesis and Leukemogenesis in Mice Expressing Oncogenic NrasG12D from the Endogenous Locus. Blood 2011, 117, 2022–2032. [CrossRef] [PubMed] 24. Chan, R.J.; Leedy, M.B.; Munugalavadla, V.; Voorhorst, C.S.; Li, Y.; Yu, M.; Kapur, R. Human Somatic PTPN11 Mutations Induce Hematopoietic-Cell Hypersensitivity to Granulocyte-Macrophage Colony-Stimulating Factor. Blood 2005, 105, 3737–3742. [CrossRef] [PubMed] 25. Le, D.T.; Kong, N.; Zhu, Y.; Lauchle, J.O.; Aiyigari, A.; Braun, B.S.; Wang, E.; Kogan, S.C.; Le Beau, M.M.; Parada, L.; et al. Somatic Inactivation of Nf1 in Hematopoietic Cells Results in a Progressive Myeloproliferative Disorder. Blood 2004, 103, 4243–4250. [CrossRef] 26. Geissler, K.; Jäger, E.; Barna, A.; Alendar, T.; Ljubuncic, E.; Sliwa, T.; Valent, P. Chronic Myelomonocytic Leukemia Patients with RAS Pathway Mutations Show High in Vitro Myeloid Colony Formation in the Absence of Exogenous Growth Factors. Leukemia 2016, 30, 2280–2281. [CrossRef] 27. Geissler, K.; Jäger, E.; Barna, A.; Gurbisz, M.; Marschon, R.; Graf, T.; Graf, E.; Borjan, B.; Jilch, R.; Geissler, C.; et al. The Austrian Biodatabase for Chronic Myelomonocytic Leukemia (ABCMML): A Representative and Useful Real-Life Data Source for Further Biomedical Research. Wien Klin Wochenschr 2019, 131, 410–418. [CrossRef] 28. Savona, M.R.; Malcovati, L.; Komrokji, R.; Tiu, R.V.; Mughal, T.I.; Orazi, A.; Kiladjian, J.-J.; Padron, E.; Solary, E.; Tibes, R.; et al. An International Consortium Proposal of Uniform Response Criteria for Myelodysplastic/Myeloproliferative Neoplasms (MDS/MPN) in Adults. Blood 2015, 125, 1857–1865. [CrossRef] 29. Oehler, L.; Foedinger, M.; Koeller, M.; Kollars, M.; Reiter, E.; Bohle, B.; Skoupy, S.; Fritsch, G.; Lechner, K.; Geissler, K. References Efficacy of Azacitidine Compared with That of Conventional Care Regimens in the Treatment of Higher-Risk Myelodysplastic Syndromes: A Randomised, Open-Label, Phase III Study. Lancet Oncol. 2009, 10, 223–232. [CrossRef] 10. Aribi, A.; Borthakur, G.; Ravandi, F.; Shan, J.; Davisson, J.; Cortes, J.; Kantarjian, H. Activity of Decitabine, a Hypomethylating Agent, in Chronic Myelomonocytic Leukemia. Cancer 2007, 109, 713–717. [CrossRef] 1. Ciriello, G.; Cerami, E.; Sander, C.; Schultz, N. Mutual Exclusivity Analysis Identifies Oncogenic Netw Modules. Genome Res. 2012, 22, 398–406. [CrossRef] 12. Patnaik, M.M.; Tefferi, A. Cytogenetic and Molecular Abnormalities in Chronic Myelomonocytic Leukemia. Blood Cancer J. 2016, 6, e393. [CrossRef] [PubMed] 13. Ricci, C.; Fermo, E.; Corti, S.; Molteni, M.; Faricciotti, A.; Cortelezzi, A.; Lambertenghi Deliliers, G.; Beran, M.; Onida, F. RAS Mutations Contribute to Evolution of Chronic Myelomonocytic Leukemia to the Proliferative Variant. Clin. Cancer Res. 2010, 16, 2246–2256. [CrossRef] [PubMed] 14. Itzykson, R.; Kosmider, O.; Renneville, A.; Gelsi-Boyer, V.; Meggendorfer, M.; Morabito, M.; Berthon, C.; Adès, L.; Fenaux, P.; Beyne-Rauzy, O.; et al. Prognostic Score Including Gene Mutations in Chronic Myelomonocytic Leukemia. JCO 2013, 31, 2428–2436. [CrossRef] [PubMed] 15. Patnaik, M.M.; Itzykson, R.; Lasho, T.L.; Kosmider, O.; Finke, C.M.; Hanson, C.A.; Knudson, R.A.; Ketterling, R.P.; Tefferi, A.; Solary, E. ASXL1 and SETBP1 Mutations and Their Prognostic Contribution in Chronic Myelomonocytic Leukemia: A Two-Center Study of 466 Patients. Leukemia 2014, 28, 2206–2212. [CrossRef] [PubMed] 16. Elena, C.; Gallì, A.; Such, E.; Meggendorfer, M.; Germing, U.; Rizzo, E.; Cervera, J.; Molteni, E.; Fasan, A.; Schuler, E.; et al. Integrating Clinical Features and Genetic Lesions in the Risk Assessment of Patients with Chronic Myelomonocytic Leukemia. Blood 2016, 128, 1408–1417. [CrossRef] Int. J. Mol. Sci. 2020, 21, 3025 13 of 14 13 of 14 17. Patnaik, M.M.; Wassie, E.A.; Lasho, T.L.; Hanson, C.A.; Ketterling, R.; Tefferi, A. Blast Transformation in Chronic Myelomonocytic Leukemia: Risk Factors, Genetic Features, Survival, and Treatment Outcome: Blast Transformation in Chronic Myelomonocytic Leukemia. Am. J. Hematol. 2015, 90, 411–416. [CrossRef] 18. Geissler, K.; Hinterberger, W.; Bettelheim, P.; Haas, O.; Lechner, K. Colony Growth Characteristics in Myelomonocytic Leukemia. Leuk. Res. 1988, 12, 373–377. [CrossRef] 19. Geissler, K.; Ohler, L.; Födinger, M.; Virgolini, I.; Leimer, M.; Kabrna, E.; Kollars, M.; Skoupy, S.; Bohle, B.; Rogy, M.; et al. Interleukin 10 Inhibits Growth and Granulocyte/Macrophage Colony-Stimulating Factor Production in Chronic Myelomonocytic Leukemia Cells. J. Exp. Med. 1996, 184, 1377–1384. [CrossRef] y y p 20. References Interleukin-10 Inhibits Spontaneous Colony-Forming Unit-Granulocyte-Macrophage Growth from Human Peripheral Blood Mononuclear Cells by Suppression of Endogenous Granulocyte-Macrophage Colony-Stimulating Factor Release. Blood 1997, 89, 1147–1153. [CrossRef] 30. Hirsch-Ginsberg, C.; LeMaistre, A.C.; Kantarjian, H.; Talpaz, M.; Cork, A.; Freireich, E.J.; Trujillo, J.M.; Lee, M.S.; Stass, S.A. RAS Mutations Are Rare Events in Philadelphia Chromosome-Negative/Bcr Gene Rearrangement-Negative Chronic Myelogenous Leukemia, but Are Prevalent in Chronic Myelomonocytic Leukemia. Blood 1990, 76, 1214–1219. [CrossRef] 31. Gelsi-Boyer, V.; Trouplin, V.; Adélaïde, J.; Aceto, N.; Remy, V.; Pinson, S.; Houdayer, C.; Arnoulet, C.; Sainty, D.; Bentires-Alj, M.; et al. Genome Profiling of Chronic Myelomonocytic Leukemia: Frequent Alterations of RAS and RUNX1genes. BMC Cancer 2008, 8, 299. [CrossRef] [PubMed] 32. Tyner, J.W.; Erickson, H.; Deininger, M.W.N.; Willis, S.G.; Eide, C.A.; Levine, R.L.; Heinrich, M.C.; Gattermann, N.; Gilliland, D.G.; Druker, B.J.; et al. High-Throughput Sequencing Screen Reveals Novel, Transforming RAS Mutations in Myeloid Leukemia Patients. Blood 2009, 113, 1749–1755. [CrossRef] [PubMed] Int. J. Mol. Sci. 2020, 21, 3025 14 of 14 14 of 14 33. Orazi, A.; Chiu, R.; O’Malley, D.P.; Czader, M.; Allen, S.L.; An, C.; Vance, G.H. Chronic Myelomonocytic Leukemia: The Role of Bone Marrow Biopsy Immunohistology. Mod. Pathol. 2006, 19, 1536–1545. [CrossRef] [PubMed] 34. Mossner, M.; Jann, J.-C.; Wittig, J.; Nolte, F.; Fey, S.; Nowak, V.; Obländer, J.; Pressler, J.; Palme, I.; Xanthopoulos, C.; et al. Mutational Hierarchies in Myelodysplastic Syndromes Dynamically Adapt and Evolve upon Therapy Response and Failure. Blood 2016, 128, 1246–1259. [CrossRef] 35. Constantinidou, M.; Chalevelakis, G.; Economopoulos, T.; Koffa, M.; Liloglou, T.; Anastassiou, C.; Yalouris, A.; Spandidos, D.A.; Raptis, S. Codon 12 Ras Mutations in Patients with Myelodysplastic Syndrome: Incidence and Prognostic Value. Ann. Hematol. 1997, 74, 11–14. [CrossRef] 36. De Souza Fernandez, T.; Menezes de Souza, J.; Macedo Silva, M.L.; Tabak, D.; Abdelhay, E. Correlation of N-Ras Point Mutations with Specific Chromosomal Abnormalities in Primary Myelodysplastic Syndrome. Leuk. Res. 1998, 22, 125–134. [CrossRef] 37. Shih, L.-Y.; Huang, C.-F.; Wang, P.-N.; Wu, J.-H.; Lin, T.-L.; Dunn, P.; Kuo, M.-C. Acquisition of FLT3 or N-Ras Mutations Is Frequently Associated with Progression of Myelodysplastic Syndrome to Acute Myeloid Leukemia. Leukemia 2004, 18, 466–475. [CrossRef] 38. Kunimoto, H.; Meydan, C.; Nazir, A.; Whitfield, J.; Shank, K.; Rapaport, F.; Maher, R.; Pronier, E.; Meyer, S.C.; Garrett-Bakelman, F.E.; et al. Cooperative Epigenetic Remodeling by TET2 Loss and NRAS Mutation Drives Myeloid Transformation and MEK Inhibitor Sensitivity. Cancer Cell 2018, 33, 44–59.e8. [CrossRef] 39. References Jin, X.; Qin, T.; Zhao, M.; Bailey, N.; Liu, L.; Yang, K.; Ng, V.; Higashimoto, T.; Coolon, R.; Ney, G.; et al. Oncogenic N-Ras and Tet2 Haploinsufficiency Collaborate to Dysregulate Hematopoietic Stem and Progenitor Cells. Blood Adv. 2018, 2, 1259–1271. [CrossRef] 40. Watzinger, F.; Gaiger, A.; Karlic, H.; Becher, R.; Pillwein, K.; Lion, T. Absence of N-Ras Mutations in Myeloid and Lymphoid Blast Crisis of Chronic Myeloid Leukemia. Cancer Res. 1994, 54, 3934–3938. 41. Tenedini, E.; Bernardis, I.; Artusi, V.; Artuso, L.; Roncaglia, E.; Guglielmelli, P.; Pieri, L.; Bogani, C.; Biamonte, F.; Rottuno, G.; et al. Targeted Cancer Exome Sequencing Reveals Recurrent Mutations in Myeloproliferative Neoplasms. Leukemia 2014, 28, 1052–1059. [CrossRef] [PubMed] 42. Kuo, M.-C.; Liang, D.-C.; Huang, C.-F.; Shih, Y.-S.; Wu, J.-H.; Lin, T.-L.; Shih, L.-Y. RUNX1 Mutations Are Frequent in Chronic Myelomonocytic Leukemia and Mutations at the C-Terminal Region Might Predict Acute Myeloid Leukemia Transformation. Leukemia 2009, 23, 1426–1431. [CrossRef] [PubMed] 43. Gelsi-Boyer, V.; Trouplin, V.; Roquain, J.; Adélaïde, J.; Carbuccia, N.; Esterni, B.; Finetti, P.; Murati, A.; Arnoulet, C.; Zerazhi, H.; et al. ASXL1 Mutation Is Associated with Poor Prognosis and Acute Transformation in Chronic Myelomonocytic Leukaemia: Gene Mutations in CMML. Br. J. Haematol. 2010, 151, 365–375. [CrossRef] 44. Khozin, S.; Blumenthal, G.M.; Pazdur, R. Real-World Data for Clinical Evidence Generation in Oncology. JNCI J. Natl. Cancer Inst. 2017, 109. [CrossRef] 45. Patnaik, M.M.; Pierola, A.A.; Vallapureddy, R.; Yalniz, F.F.; Kadia, T.M.; Jabbour, E.J.; Lasho, T.; Hanson, C.A.; Ketterling, R.P.; Kantarjian, H.M.; et al. Blast Phase Chronic Myelomonocytic Leukemia: Mayo-MDACC Collaborative Study of 171 Cases. Leukemia 2018, 32, 2512–2518. [CrossRef] 46. Borthakur, G.; Popplewell, L.; Boyiadzis, M.; Foran, J.; Platzbecker, U.; Vey, N.; Walter, R.B.; Olin, R.; Raza, A.; Giagounidis, A.; et al. Activity of the Oral Mitogen-Activated Protein Kinase Kinase Inhibitor Trametinib in RAS-Mutant Relapsed or Refractory Myeloid Malignancies: Trametinib in RAS-Mutant Malignancies. Cancer 2016, 122, 1871–1879. [CrossRef] 47. Such, E.; Germing, U.; Malcovati, L.; Cervera, J.; Kuendgen, A.; Della Porta, M.G.; Nomdedeu, B.; Arenillas, L.; Luño, E.; Xicoy, B.; et al. Development and Validation of a Prognostic Scoring System for Patients with Chronic Myelomonocytic Leukemia. Blood 2013, 121, 3005–3015. [CrossRef] 48. Malcovati, L.; Gallì, A.; Travaglino, E.; Ambaglio, I.; Rizzo, E.; Molteni, E.; Elena, C.; Ferretti, V.V.; Catricalà, S.; Bono, E.; et al. Clinical Significance of Somatic Mutation in Unexplained Blood Cytopenia. Blood 2017, 129, 3371–3378. [CrossRef] 49. Öhler, L.; Geissler, K.; Hinterberger, W. References Diagnostic and Prognostic Value of Colony Formation of Hematopoietic Progenitor Cells in Myeloid Malignancies. Wien. Klin. Wochenschr. 2003, 115, 537–546. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
W1968647478.txt
https://zenodo.org/records/1553780/files/article.pdf
de
Neue Grundlagen der Holzmessung
Forstwissenschaftliches Centralblatt
1,922
public-domain
2,460
@o'~enabl: 9tette NnmbIagen bet @oI~mef[tmg. 233 ~ette ~ r u n b l a g e n bet s ~on ~. ~oI~eaabl, ~6erltborb b) ~ar~teIImt~ bet ~un{tionen bet ~n~?aIt~attoren burd) ~ourierf~e iRei~)en. Bur ~ar~'telhmg bet ~erteil[ungghm~tion Nmten an ~teIle bet ~,{ReiIle att~ ~ourieri~e ~ei~)ett benfit~t unb ~iegu toieber Ne | reitle gem~i[')lt toerbert. ~ i r biirfen annet)mea, ba~ l'i~ hie ~urg?meger bet ~imtiiglen in einem ~e~'tanbe tmfl)anbenen ~tfmme goilgjea ben t~renben e a-- a t', unb e a + a ~,, betoegen, roeg~db mir hie ~eriobe ~=6~e to{i~)l[en. ~erner erroetft eg ~i~ atg groet~m~ihig, toenn mir loieber hie !~ogafitI)men bet ~nIldiglattomt ~ur ~ar[ielItmg bur~j hie ~inueret~)e benfi~en. ~ a bie ~oelfi~ienten bet ~,~lei~e gan~ IMfimmten ~oei[i~ienten be~: | ertti~re~en, i~ ~6nrte~ bern gdorberten ~enauigl~eit~grctbe enti~0rec~en~ ~ei bet (~ntroic~fmtg bet ~ourierlc~en ~ei~e in f = In N -1- .~. -l- k~ sin ~ -t- A~ sin 2 ~ -t- . . . . . . . . 67) (~----- (~ -t- 3 ~,~) : 6 ~ ) mt~ be~ ~--~Rei~)e In f = l u N [ ' -1- lnh q- In r (u) -t- g ha ~,s~r 1 -t- . . . . 68) ~fir alIe larat~ti~) in ~3etvad0t ~ommenbett ~erte bet 9.Rsmente /*,, ~'a, ~,~ ... attm !8orang die ~oei[isienten A~ bet ~irttt~ieiI)e bere~uet nnb tabellarij~ 8urn ~ebraa~e berdtge~'telIt merben. ~egett bet regelm~igig t'teigenben ober ~allenbert 5genben3 bet ~t;~)en, ~anb ~orm~a~lhm~ett mirb ben ~inagrei~?en, womit bieje beiben Rttrt~en bei~rielben roerben, goe~tmti~ig aa~er einer Ron~'iantea ein tmn ~: linem: abl)~ingigeg ~lieb ange{iigt. ~ieie beiben Rm~ert er~d)einert bann, gIei~ gulammengefa~t, in fotgenber {~om: in (1 Z)= B q-C Jr- (Bo q- Co)r -q- (B~ -}- Ct) sin ~,-}- (B~-}- C~)sin 2 V~q-... 69) ,~nbem ~u bieien 61Id~jangen bet ~ti~len~ unb ~orm~a~jihmae hie 7l: ~leic~urtg bet ~tamragrttrtb~lti~e in g---~ In i -t- 2 a nt- 2 r abbiert ~virb, erljalten wit hie ~lei~ang bet ~tammint)alte Iron ben ein~elnen ~atert~dIen r mtb roenn bann ~ier~tt noa) hie ffOleidgnngbet ~erteihmg~un[tion abbim Ivi~b, bann bdommen roir ~iir hie ~e~'tanb~in0altghtrDe hie gClei~ung: 7~ l n ( f g l Z ) : l n N - [ - l n ~ - - [ - 2a-}-R-l-eo~+qtsiu~P-l-q~sin2~p... 70) 234 ,~o~enabl: roobei A + B + 0 = R ; Bo+Oo+2=0o; .Am'~-Bm"{-Om=0m flea ~e~t ill ~ i t bez @leic~ung (70) ift bet ~effanbgin~alt lbe~a~tiebea, melvin: bern ,~nter~alI gmifc~en r tmb r + dr ~agd)~rt. c) ~;ran~forrnation bet ~an~timten yon ~a~alt~fatEtoven~ ~dton be~ bin: ~edeitaafl bet alIgemeinen ~tamm~alflforme[ f~---Nh(~o~ ( u ) - - ~ h ,% ~,(u)...) roarben ,erfc~iebene Umformmtgen vorgenommen: .~nbem x ~ a + r ober x - - a ~---$ flekO.t rouzbe, ifi bie ~rbinatenac~fe am hie @rfi~e- a ~er-!c~oben roorben. ~erne~ ift ein anberer g[bfSi~'fenmaf3ftabu = h r ein-gef/iDrt unb bamit ei~e ~treffang be~ ~Irgumentgebiete~ l)erbeiflef/i~)rt morben. ~ i r f6nnert mm auf hie | ~cangfo~matiDn: f f: r/N; ~ = ~ - (67) tmb (68) aug nocD bie . . . . . . . . . . . . 71) anroenbea, roobur~ Wit baart l~ie ~Ie/~ang be~ relatiVen | f ~:,,:h (% {u) l h ' , @ % ( u ) + ...) . . . . 72) ober In ~ = In ~ = A + As sin ,~ + .A~sin 2 g/-I- . . . . . . 73) e~Dalten, roel[c~enun angibt, rodc~er uaenblidt ~leiner ~rad)teil ~ bet ~tammga~)lea aug bern ~eftanbe mit bet @tamm~a~[ 1 aal bag ~nter, ,alI ~roii~en r unb r + de f/ilIt. ~ i t bec ~mngformatimt x----- a + ~ attf hie ~6~en-- unb ~o~m~ab[farOen 1: fi (x) ~ego. z = f~ (x) ezfc~einen biele ~qa~uen in bet ~orm: mobei mit 1~ 5e~ro. Z~ bie g~6De beam. hie ~om3aO[ be~ ~tammeg mit bern ~urc~me[fer ea b. i. beg 9IrgumentmitteIffammeg be~eid3ttet fit. ~i~ t6nnen aaf hie ~i~[/m~-- unb ~ormsaI)l[~arve aa~ nod) hie ~ran~formationen 1=1~1~ unb Z~---gaZ~ anmenben, roobei mit 1~ unb g bie zelativen ~6~)en, begin, relativen ~o~m-~aI)fen, bqogen aug hie! ~tamm~)6~e beg .~[rgumentmittelftammeg, begger~ :datiVe , ~ e nab ~om~aD1 1 i[i. ~eue ~runblasea be~ @Qlbme[~m~8. 235 ~Iu~ br @Iei~ung bet ~tammg~ttnb~i~en 71: ]ng--- In ~- -4- 2a + 2 ~ Ing~ -4- 2~ i[t o~ne weiterr abbune~mem ba~ hie @Ieit~un0 bet relatiuen ~tamm~ ~runbg~i~en In (-~-) --~-2 ~ [autet. ~ie @M~ung be~ retati,en ~e~'tanbgin~alte ift bann: /fv~ In l-~vJ = In (vr~3-- q + qo ~ + qt sin ~p+ q~ sin 2 g,... 74), (R - - In (la Za) -= q) ~ie gtnaiogie ~toifd)enbet @leigounqbet relatioen ~eftanbgini)aite (74) unb her relatiuen ~tamm~ abet ~ormtttrue (48) liegt .or 9.Ittgen: ~entt mir 3ttm 9htmmtg iiberge~en, bann ~)abett roir mit bet @leig)ung d g = N v a e e + e~~ + ~ sintp + O~sin2 ~v+"" " d~ 75) bag ~i~erential[ beg ~et'tanb~inOalteg, n~imii~ ben anenbli@ fteinert SeiI ~om ~otggd?a[te beg ~Seftanbeg, bet gum ~ntert~alI 5roilg?ert ~ unb ~ + d geOSrt. ~urc~ ~ntegration ert)alten mir V=Nv~fee+e~ . 76)- ~2it bern anter bern ~ntegratgeid)en ftet)enben 91ugbrudl i]'t bet retatit~e ~ni/al[t beg ~e[tanbeg befd)rieben, b. i. bet ~oI~ge~alt t~eg ~e~ [tanbeg, be~'~en ~tamragaIfl 1 i~'i, begett 9.Irgumentmittel~'tamm hie qttabratiigle ~uerff~i~e, g2fi~e tmb ~om~alll ie 1 gu~ommt unb be[[en ~e[ati~e ,3nt)altgMto~enfttr~en mit bert angegebenett ~teiDertbelt~rieben [ink ~biejen ~ntegralwert jet~eittoir glei~ .z/ nnb ttennen -//bie ~e~tanbg, ~omga~JI, a&o jene gaOL melee mit bern ,~n~)ttlt beg 9IrgumentmitteI, t'iammeg ttnb bet ~tamm~a~fl N mul[tipli~iert ben ~o[bget)att beg ~e, ftanbeg gibt. V = Nv2-// d) ~ie ~eftant~gmittei[t~imme. 1. ~)er geometriJ~e ~2ittel~'tamm. ~er NrgttmentburfOjfOnitt a, b. i. bet ~l~onent a ronrbe atg arit~metiJ~eg ~ittel bet im ~e~'lanbe~orltommenben x beftimmt. ~g lit al~o: 2nx Oa~__e N ~ nix t =e T e ~ = 1 7 o ~1 ~ ' . o N ~ x~ .e N n~ X~... ... 236 @oljenab[: ba~ I~ei[~t e 9 if/ ba~ 6r ~ittel[ au~ ben i~mtlid?en im ~effanbe ~offommenben ~ur~me~'[ern. ~ i r nemtett be~aIb ben ~lrgumentmittet, ~tamm mit bern ~)urdlme[[er e9 bert ,geometriidjen ~J~ittet[tamm". !3ei eitter ~tamms@Iuerteilung n. b. ~. gr liegt bieier ~tamm in ~itte be~ ~rgamentgebiete~; oaf hie | a ~/ilIt hie g)b~ft~@I bet in eiae~ ~tafe t~ereinigten ~tamm~@len. ~)er geometrildje ~XitteI~'tamm liegt in biefem ~alI bei 50~ bet ~tamm3a~ert. 2. ~)er Rrei~fi~idjenmitte[~'tamm. ~er ~tamm mit be~ ariff)-metild) mittIerert ~rei6f~id)e, fur~ bet a~it~)meti~ff~e~J~ittetftamm, farm bei einer ~tammsa~)fuerteilang n. b. @. (~. r~ie folgt befiimmt roerben: ~ie RreiB~i~iCe eine~ ~t~mme~ ~om ~nteroalIe x i~'t: z~ 2a+2~ ~er ~eN~ac~enittmme ift ~i normat~ | ~ie ~r ,~erteihmfl: ~ e 2a~2~ N h - - k ~ , _ o.ratt roir buv ~ 2a he2~--h:~-~ nte ratio e alte.: --(113 1 G----~N ~ e ~ (a +~''~) . . . . . . . . . . 78) ~)er ~)ar~mege~ beg ~tammeg mit be~ arit~)metiid) mittleren ~rei#, ,frame ift alio: d~,~---e a + t ~ ~ere~net man ttad~ bie[er ~ome[ ~on ~eft/inben mit beliebiger ~tammga~l~erteihtng ben ~)ard)me[~e~ dg, Io ert)~ilt man fair attl ben ~iIIimete~ genan hie auf gerof~)alid)em ~ege erre~nete 9Xittel~'t~ffe ~ a n farm al[io iagea, b@ bie ~emad)10ilfigung ber t)@eren ~omente ~a, #~. . . . 8toedg gettauerer ~ei69reiBattg be~ ~3erteifungglttnftion f(x) fiir ben in bet ~ragig erforberlid)en @ra~ t~on @enattig[eit bet ,.~tamm, grtmbg~idlenberea~nang bebeutunggIo~ ift. Ni~ Nnnen jet~t aac~ tmterlud)ett, our roelc~er | bet | .,~@IIeite~ bet | mit be~ arit0metild) mittleml Rrei~fi~if0e Iiegt. ,Negebett ift bet ~urd)mege~ d -= e a + ~'~" ~teae ~mxblasen bet @ollme~ung. 237 (~elac~t ~irb hie rehti~e ~tamm~a~I, roeI~e bern ~tamm ~om ~nter~ ~alIe } ~---~.2 ~atvramt, mit anberen ~o~text: e~ ift bec ~ e ~ be~ ~ntegraI~ h f --h~j~ V~J ~u bere~aen. ~n ~npa~jun~ an bie ~a~e~n ~fir ba~ ~et)Ierinteorat fc~reiben roic: 0 u ~ hp~ 2 z,_- fe-u"du + fe--U'du laO 0 + 9 (o/--, + . . . . . . 79) ~ i r ieOelt, bah hie ~umme bet reIati~en ~tamm~aOlen, ober roenn mi~ ang in her /iblic~en ~eife attgbrfic[en, bah bag ~tammSaDIpro~ent, bei meld~em bet ~tamm mit her aritDmetilc~ mittIeren ~ceigg~ic~e ~a itt~en i~'t, ~on bet ~c6~e bet ~treaung ~ ab~6ngt. ~)er a~it~metif~e g~runbg~i~enmitteiftamm liqlt bet einem um [o D6~eren ~tamm~a~b pro~ent je gr6~er bag logacitDmif~ abgeftu~te ~rgumentgebiet ift, fiber me[~e~ [i~ hie ~tomm~a~len Verteiten, ~e gr~iger bag ~erD~ittnig ~roii~en ben fc~ro~ic~ftettunb [t6fffien ~t~irfdla[[en beg ~eftanbeg fit. ~enn D~xt ~ic~tenbe~'t~inbettbe[~auptet mirb, bah bet ff~ranbft~ic~en~ mittet~tamm bei 60% bet ~tctmm~a~tett liegt, fo er~alten mir aug bert ~aMn ~fi~ bag ~eDIe~integra[ / x q) (u) = 0,20 = 9 (0,1792) g~ ~ 0,1792. ]/Y ~ 0,2535; h ~---2,79 ~ i t biele~rbe~'fimmten~ri~ge t~on ~t~ ~ber h ~irtb ~ic~tenbe)'t~inbe~on gan~ beftimmtec ~e~af~hmfl befa~rieben. ~ie ~age beg Rcetgg~i~en~ mittelftammeg me~gielt ni~)t nttr mit bern ~trettattflgmage #~ ober bern l~ieraa~ berec~neten ~arameter h, jonbertt auc~ mit ben t)6~eren @[ementen fta, ~q roenn anbere ~.Serteilanggianttionen al[g iol~e n. b. ~. ~. ~or[iegert. ~)ag ~tamm~aDlpro~ettt mie her aritDmetiic~e ~itteQ'iamm [inb ieDr f~man~enbe ~effmale bet ~efianb~bei~reihmg. ~ageflert ifi mit ben beiben $2effmaIett a unb ~,~, bie beibe aug ben flleic~en ~rDebtmgett be[timmt merben ~/~nnen, mie bet S~rei~g~ic~enmitteI)'tamm,nicer nttr bie g~riJf~e bet beiben ~ittel[t/imme o tmb ea + t~ ~nbera auc~ ein me~entli~e~ ~erfmaI bet ~e~iattb.~ftruttur gemonnen. 238 bol~ettaM: 3. ~)er ,~nI~a[t~ unb ~ert~mitteIftamm. tim eine fftSOerung~ ~orme[ ~fir ben ~tamm mittIeren ~ol~geOalter [~erieiten ~u t[~nnen, ben~en roir un~ bie @[ei~nngen bet f~)iil]en~ wie bet ~orm~aOIStrve in 9teiOen uon ~ote~en $ entwktdt, bie mir mit $~ abbre~en. ,~ur ~amme bider beiben 9~eihen abbieren wit bie @Ieid~nng bet Rrei~fi~ic~en, bonn Iantet b/e @l[eic~ttng bet ~tamminbalte In v~ ~ In v~ + (Pl + Pl + 2) $ + (ql + %) ~ ~v~f/ir ~oir bet ~info~Deit megen leben: In v ~ lnv~ + p~ + q~ ~ei normaler ~tamm~ahIt~erteihmg ift bonn bie @l[d~ttn~ bet ~e~tonb~in[~oIt~l[nrUe: fv~dV=Nvaep~+q~ h Go --00 p~ _ _h . q e 4 (h~-q) . . . . . . . 80) V~-~)ie ~nr~en bet ~nhatt~aftoren l'inb bier bur~ ie 2 ~on)'tante bet ~nhoIt~ nnb ~orm[urt~e [oroie burd~ bo~ ~frennn~mo~ ~,, a[[o ~u~ foremen bnr~ 5 t~on !Be[tonb ~n ~ei'tonb roed~elnbe $~on~'iante, neDen ben 2bme[[nn6en be~ 6eometri[COen ~ittelD'tomme~, bOimmt. ~iir ?or(ffi~c 3roedr 5cheater bier, f~on dnett rodt~et)enben @tab yon @enani~feit. @ezabe barin abet, baw id~on 5nr annCiOernben ~e~ Id~rdbang bet ~c~ffinbc cine io gro~e ~oh[ yon ~arametern notmenbig i~'t, [ii~t ~i~ bet @rob t~on lln~id~erOeit ohne~mcn, toddle hie ~et~obe bet ~n~ol[t~berr nad~ bern aritt~metifdOen~ittefftomme in ~icOtdigt. 9"lit bet ~ormc[ (80)hahen mi~: einen [~e~timmten ~ert ~iir ben ,,reloti~en ~e[ionb~in~)oIt" ober hie ~eftonb~[orm~aOl: pS h 4 (h~- cO, -//:-- h ~ e bet a~ na~ bet ~ormel (76/ n~iherunggmei~e ober, menn gemini,)t, mit fir6[~erer @enaui~eit technetium, ~raphi!~ ober me~aniI~ bQimmt Werben tann. ~)cr ~oI~ehatt v be~ ~nhaIt~mitteI~'tammr ift: v~v~.//. ~ie ~el'ianb*~orm~aDr ~ [/i~t ~i~ and), mie hie ~tomm~orm3ah[ Xt att~ ,~t~imme anbe~er 9.Ir~nmentr nmlm:mon. ~est)gen au~ ben ~nha~t v~ V -~- Nv~ 9~ene @tmxbla~en t~et Ovlsme~juuO. 239 ~)e~ ~tammee mit her arit~metif~ mitfleren ~ei~{id~e g i~t bei~piet~ ~eile bet ~eftanb~inf~alt v : Va v~ vg ~en~en wit un$.an ~telle bet ~tammin~al:te, hie ~tammwerte w~ lmrc~ hie @leic~unO: In w~ --- In w~ -[- p~ -]- q~Z baroe~fellt[, jo ij't bei einer ~erteilun~ bet ~tamm~ct[~len nac~ bern ~ponentia[~ele~e bet ~eftanb~mert W: p~ W = Nwa ( ~ o4(h'---~q'), wobei mit w~ bec ~ert bee geometrifd~en ~ittd~tammee unb mit bern in ~[ammer ~efe~ten ~a~[tor bet ,rdafive ~e~tanberoert" (~e[tanb~mert~ ~orm~aO[) be~eid~net i~'t. ~Iac~ bieler ~ert ~[ann nacO ber ~orme[ (76) mit befiebi~er @enaai~ hit bere~net unb ou~ anbere ~r~umente umfle~ormt merben. e) Ys her ~.Irflumente. ~ie[~er i~'t fiber hie ~e~fj~Oe, roorauf ~ic~ hie ~erteilurtO bet ~tamm~aD[en be~ie~t, nic~t~ au~efa~t morben. ~fir bie ~ar~iellun~ bet ~ertei[uno~unOion m/ire ee ~Ieic~filfi~, ob hie ~tamm~al~[en naCO ~t~den ~eorbnet merben, hie bei 1,~ m abet bei 0,1 1 iiber hem ~oben ~eme~en mor~en ~inb, ee [inb abet nut jene ~e~t/inbe mitein~nber ~er~~Ieicl~bar, beren ~r~umente [i~ au~ ~tbme[~anoen bei relati~ ~leic~en ~tamm~i~Oen ~1 ~id) be~ie~en. ~a e~ unmi~oIic~ i~'t, alh ~t~imme bei relatiu ~leic~en ~tamm~i~en ~u mermen, ~o ~rirb hie ~ej'tanb~au~naOme nac~ ~urc~me~errt bei 1,~ m (be[~er bei 1,5) fiber bent ~oben volI~o~en anb bann nac~ bieler eine ~tammfi~te nad~ ~[roumenten bei 0,t bet ~tammf~ii~e ~eroe[telIt. 2)amit roirb hie ~ei'tanbeaulnaOme bet ~e~un~ fie~enber ~t/imme an~epa~t, hie ~oI~erunoen au~ ben ~r~ebni~en bet ~e[~un~ fieoenber ~t/imme ~iirmen bei bet ~e~un~ ~teOenber ~t/imme unb Oart~er~ej't/inbe ~erroertet ~erben. ~ie ~rane~ormation bet ~r~umente bei 1,3 m fiber bern ~vben in jene bei 0,1 I ~e~c~ie~t mit ~iffe bet ~tammhxr~en~Iei~un~ lnd ~---],~k "Jr-mx ~vomit hie burc~Ic~nittfic~e ~orm bee ~ar~e[an][aa~e~ bet ~t~imme ~ij~en 1,3 unb 0,1 I it~er bern ~oben Dinreic~enb ~enau be[cDrieben with. ~urc~ ~e[~un~ ~on 2 ~)urd~me[[ern 8. ~. bei L3 unb bei 3,~ m 18" 240 Ool~enabI: ~e~e (~mnb~a~en bet Ool~rae~f~g. fi6er bern ~Ooben an ~ro~e~t~mmen ~er~iebener ~t~r~e~'tnfen roerben hie ~ar~nitt~werte Don m m i t f~itfe bet (~Iei~mtg r~--- 2 m = Inds.s - - In da.s, ober m ~-- (ds.s -- din) : 2 d~,s ~ere~net. ~er ~ur~me~er d~ ~e~ 0,1 1 i~t bann: d~ -~- dl,s -]- m (~ 1 ~ 1,3) ~iir c~ g~eic~ 0,1 1 f~lIt bie ~e~bf~oe uon ~t/immen bi~ ~tt 33 In irtrterI~a~fi l~er beibert ~e~tel~ett 5e~ 13 ttnb 3,3 m, bet -~ttr~me~er d~ mirb interpoliert. ~ei ~fiimmen unter 13 m uttb fiber $3 m ~60e roirb bet $)urd~mej'[er bei 0,1 1 e~trapoliert, e~ 16nnen abet [fir ~ej'flhtbe mit ~einerert ober gr6geren bur~Id~nittlid~en ~tamm~)~)en att~ anbere ~e~b~f~ett ~ero/iOIt roerben. -~ i t bet ~ran~Iormation bet ~rgumente f~e~1,3 in Iol~e 5e~ ,~ I voirb nocO eine onbere roi~tige ~u~gabe bei bet ~e~'tanb~me~tmg tei~meiie gd6~'t, tt~imlfcO bie -~effimmtmg bet ~c~ait~orm. ~ i t bieler ~ran~; ~:ormatiott mirb n~im~ic~ hie ~c~ctft~orm -- tmb bie ~orm~orO~ ~ bern ~ingtt[~e bet tmg~ei~en ~c~a~tO6~)en unb ben me~idnben ~ormen be~ ~ttr~eIan~au~e~, a~io ~roei ~afloren, mdc~e an ben gro~en ~martfltngen bec unec~ten ~orm~arSett Oaupfl~ic~ic~ Ic~alb ~'iub, ettt~ogen. ~ie burc~ic~nitflic~ett ~orm~d)~ett besogen attf bie ~tammb~ibe ~1, ba~ [inb aflo bie ec~tert ~orm~abIen, roeic~en bei nicer 8tt t~einen ~ ffir bie uerfc~iebenett ~tamm~[a[fert be~fdfiett gelc~of~enen ~ge~'tanbe~ in her ~efld nttr feOr menifl ~oneinanber aft, ~vetut attd~ bie ~orm~ttr~en nic~t ttner~)e~i~ ~er~c~iebett finb. ~iir bie ~n~att~berec~nuug bet ~e[t6nbe geniigt hie ~rmitttang bet e~ten ~orm~ta~)len, hie fi~ k~5~'t bar~ ~O~i~ung inner,rib geringer ~e~lergrett~ett (~+ 2%) ermittdn Ia~en. ~iir hie ~ertSere~ntmg ftet)en~ bet ~t~imme, melee mie hie ~ertOere~ntmg {~egenber ~fiimme, abI~nitt ~roeile att~geIfi~)rt merbett mu~, i~'t abet hie ~rmitthmg bet ~ormhtr~e notroenbig. ~u~ hie ~)ierbei einsttlc~agenben ~8erla~)ren tanrt bier nicer ein~egangen roerbett. (~c~ta~ ~ot~t.) IV. ~leti~en. ~uub~ebun~ bes ~eic~forftwirtfc~aft~rc, ts ~um IDieberaufbau. ~ie jeber anbere ~r~oerbetmeifl, ift hie bmti~e ~or~n~irt~d~aft I~ereit nnb lieiirebt, an bern roirtic~aft[ic~en ~ieberanff~au bee ~atertanbee ntittuarbeitett. 3m r ]tt foer iI~r fo na~e ~er~anbten 2anb~irt|~aft ~ottn ee ~@ bei i~r aider am due f~tt~ midenbe ~ r t r a ~ e r ~ t m g ~Tanbe~n, roir miiffen ~ielmd~r in bet ~orft~irt~aft mit [anon ~r re,hen. ~)ie na~elTenbe ~tmbgelhm 8, corn ~eic~eferft~oirtf~afterat bttr6 |einert ~t~;ttbigen g[aei~u~ am 22. ~lR~ir~ 1922 l~e~c~Iof~en, ~ellt ben je~i~en ~tanb bet ~ei~uni~f~t~i~eit
https://openalex.org/W3144881688
https://www.researchsquare.com/article/rs-327318/latest.pdf
English
null
Experimental and Theoretical Characterization of Thio coumarin by NLO single crystal in DFT Calculations
Research Square (Research Square)
2,021
cc-by
7,929
Experimental and Theoretical Characterization  of Thio coumarin by NLO single crystal in DFT Calculations K.SAMBATH KUMAR  (  sa975kumar@gmail.com ) Research Article 1 1Post Graduate and Research Department of Physics, A.A. Govt.Arts College, Villupuram, Tamilnadu, India – 605602. 2Post Graduate and Research Department of Physics, Thiru.A.Govindasamy Govt Arts College, Tindivanam, Tamilnadu, India – 604002.Research scholar R& D centre, Bharathiar university,Coimbatore.Tamilnadu,India. 3P.G.Department of Physics, Cauvery College for Women, Tiruchirappalli,India. K.Sambathkumar1*, N.Rajkamal2,M.Venkatachalapthy 2& R.Gayathri 3 *Corresponding Author. Tel.:+91 9994188861. E-mail Address:sa975kumar@gmail.com 2 Abstract Abstract Experimental and theoretical electron densities in thiocoumarin have been analyzed based on the X- ray diffraction data at 90 K. These compound pack in the crystal lattice with weak C-H…O and C- H….π interactions, and variations in charge density properties and derived local energy densities have been investigated in the regions of intermolecular interactions. Theoretical charge density calculations on crystals using the B3LYP/6-311++G(d,p) method show remarkable agreement with the derived properties and energy densities from the experiment. The intermolecular interactions follow an exponential dependence of electron density and energy densities at the bond critical points. The Laplacian follows a “Morse-like” dependence on the length of the interaction line. Based on the set of criteria defined using the theory of “atoms in molecules”, it has become possible to distinguish between a hydrogen bond (C-H…O) and a van der Waals interaction (C-H….π). This has resulted in the identification of a “region of overlap” in terms of electron densities, energy densities, and mutual penetration of the hydrogen and acceptor atoms with respect to the interaction length. This approach suggests a possible tool to distinguish between the two types of interactions. High field indicates that this molecule exhibit considerable electrical conductivity in atomic charges.The ESP map is found to be positive throughout the backbone of the molecule. The negative charges have a tendency to drift from left to right. Thermodynamic parameters like heat capacities (Cºp,m), entropies (Sºm) and enthalpies changes (Hºm) are used for various electrical field. Theoretical FT-IR and FT-Raman spectra of the title molecule have been constructed which show good agreement with recorded spectra. Key words: TCN, XRD, HOMO – LUMO, MEP. 3 Introduction Coumarins constitutes large group of phenolic compounds found in medicinal plants that share a common chemical structure of 2H-thiochromen-2-one. They are produced as secondary plant metabolites, coumarin are mono- or deoxygenated on the heteroaromatic ring. Such as Umbelliferae, Apiaceae and Rutaceae coumarins accumulate in free form or as glycosides. In addition 2- hydroxycinnamic acid glycosides are enzymatically hydrolyzed and then lactonized into coumarins. Most of them have important biological effects including antioxidant, anti-inflammatory, anticoagulant, anti allergic, antiviral, anticancer and enzyme inhibitor activities although many are not suitable for therapeutic use because of their toxicity. Coumarin has been use in synthetic chemistry, medicinal chemistry, and photochemistry. Coumarin and its derivatives have contributed immensely to the area of solid-state photochemistry [1]. Some substituted coumarin derivatives find application in the dye and dye –sensitized solar cell industry and also in the area of LASER dyes based on the compound showing dependent variation in the static dipole moment. Coumarin has used to probe ultrafast solvation effects. Coumarin and its sulfur derivatives possess well-defined dipole moments and it exhibit second harmonic generation effects since they crystallize in noncentrosymmetric space groups. We have studied the geometry and molecular packing patterns of coumarin and their derivatives in order to evaluate the features of non covalent interactions. We have earlier reported the charge density distribution in 2H-chromene-thione (thiocoumarin) and have evaluated the topological properties using the “atoms in molecules” approach. X-ray diffraction data at 90 K on the compound have been subjected to multipole atom refinements followed by the “atoms in molecules” approach to derive bond critical points [2-4]. The theoretical charge densities including the periodicity for the crystal have been calculated using the HF/DFT approach based on the experimentally derived structural model. C-H·····O interactions in crystal engineering have been shown to influence packing model, and in a recent review weak hydrogen bonding is yet unclear. But the packing features of C- Coumarins constitutes large group of phenolic compounds found in medicinal plants that share a common chemical structure of 2H-thiochromen-2-one. They are produced as secondary plant metabolites, coumarin are mono- or deoxygenated on the heteroaromatic ring. Such as Umbelliferae, Apiaceae and Rutaceae coumarins accumulate in free form or as glycosides. In addition 2- hydroxycinnamic acid glycosides are enzymatically hydrolyzed and then lactonized into coumarins. Introduction Ab initio HF and density functional theory (DFT) calculations have been performed to support our wavenumber assignments. The HOMO and LUMO analysis have been used to elucidate information regarding charge transfer within the molecule. The thermodynamic parameters like heat capacities, entropies and enthalpies changes at constant pressure within the electrical field. But isoelectronic molecular electrostatic potential surfaces (MEPSs) and electron density surfaces were calculated using 6-311++G(d,p) basis set. Structures resulting from the plot of electron density surface mapped with electrostatic potential surface depict the shape, size, charge density distribution and the site of chemical reactivity of a molecule. Introduction Most of them have important biological effects including antioxidant, anti-inflammatory, anticoagulant, anti allergic, antiviral, anticancer and enzyme inhibitor activities although many are not suitable for therapeutic use because of their toxicity. Coumarin has been use in synthetic chemistry, medicinal chemistry, and photochemistry. Coumarin and its derivatives have contributed immensely to the area of solid-state photochemistry [1]. Some substituted coumarin derivatives find application in the dye and dye –sensitized solar cell industry and also in the area of LASER dyes based on the compound showing dependent variation in the static dipole moment. Coumarin has used to probe ultrafast solvation effects. Coumarin and its sulfur derivatives possess well-defined dipole moments and it exhibit second harmonic generation effects since they crystallize in noncentrosymmetric space groups. We have studied the geometry and molecular packing patterns of coumarin and their derivatives in order to evaluate the features of non covalent interactions. We have earlier reported the charge density distribution in 2H-chromene-thione (thiocoumarin) and have evaluated the topological properties using the “atoms in molecules” approach. X-ray diffraction data at 90 K on the compound have been subjected to multipole atom refinements followed by the “atoms in molecules” approach to derive bond critical points [2-4]. The theoretical charge densities including the periodicity for the crystal have been calculated using the HF/DFT approach based on the experimentally derived structural model. C-H·····O interactions in crystal engineering have been shown to influence packing model and in a recent review weak hydrogen bonding is yet unclear But the packing features of C 4 H·····π interactions, which bring out the features of such weak van der Waals interactions. Charge density measurement has characterized the nature of C-H·····π interactions in terms of electron densities, bond critical points, and Laplacians. In this article, we looked into the characterization of C- H·····O and C-H·····π interactions. Koch and Popelier’s criteria, for the first time we have classified C- H····· O interactions into two categories (hydrogen bonded and weak van der Waals interaction) and C-H·····π interactions as weak van der Waals interactions. The literature survey reveals that to the best of our knowledge, no ab initio HF and DFT method of the vibrational analysis of TCN have been reported so far. Therefore, the present investigation is undertaken to study the vibrational spectra of this molecule completely and to identify the various normal modes with greater wavenumber accuracy. Experimental Section All chemicals were purchased from Sigma–Aldrich or Fluka Chemical Company with a stated purity of greater than 99% and it was used as such without further purification. The FT-IR spectra were recorded in the region of 400–4000 cm−1 on Perkin-Elmer Spektrum 100 BX, FT-IR spectropho tometer. The FT-Raman spectrum is recorded in the wavenumber region 3500–100 cm−1 on a NEXUS 670 spectrophotometer equipped with Raman module accessory operating at 1.5W power with Nd:YAG laser of wavelength 1064 nm is used as an excitation source.The high-resolution single- 5 crystal X-ray diffraction data were collected on a Bruker AXS SMART APEX CCD diffractometer using Mo Kα radiation (50 kV, 40 mA). To achieve the final temperature the ramp rate set to 40 K/hr. And the temperature was maintained at 90.0(2) K by using the Oxford Cryo Systems. Data were collected with different scan times of (20-60 s for thiocoumarin) to cover the full-sphere reciprocal space different 2θ settings of the detector (-25°, -50°, and -75°) and φ settings (0°, 90°, 180°, and 270°) of the goniometer. The scanning angle ω was set to 0.3° for each 606 frames. The crystal-to- detector distance was kept at 6.03 cms. This strategy provides high resolution, large redundancy, and better completeness in data sets, which are the key factors for multipole refinement model. The structures were resolved by direct methods using SHELXS97 and refined in the spherical atom approximation (based on F2) by using SHELXL97 as included in a complete package WinGX. The molecular diagrams were generated using ORTEP. Computational methods HF/DFT calculations for TCN are performed using GAUSSIAN 09W [5] program package without any constraint on the geometry. The molecular structure of the title compound in the ground state is computed both ab-initio-HF and DFT (B3LYP) with 6–311++G (d, p) basis set. The minimum energy of geometrical structure obtained by using same basis set of HF and DFT methods. All the computations have been done by adding polarization function d and diffuse function on heavy atoms and polarization function p and diffuse function on hydrogen atoms, in addition to triple split valence basis set (6–311++G (d, p), for better treatment of polar bonds. Synthesis methods A solution of 2.0g (18mmol) of freshly distilled thiophenol 3.27g (19mmol) of triethyl orthopropiolate and 1.60g (15mmol) of pivalic acid in 10mL of p-cymene was heated at reflux for 26 h. The reaction mixture was cooled to room temperature and then column chromatographed (hexane/Et2O) on silica gel to give 2.79g (75%yield) of pale–yellow liquid (2-mercapto cinnamate). 2-mercapto cinnamate(1.00g,5.4mmol) dissolved in 10mL of 95%. Et OH was added 0.75g (13.4mmol) KOH. The reaction mixture was heated at reflux for 2h, then cooled to 25°C, and acidified with 5% aqueous phase was extracted with ether (3x20mL), and the combined organic fractions were washed with H2O (20mL) and brine (20mL), dried (Na2SO4), and evaporated. An 87% yield (0.85g) of crystalline was obtained. 6 The 2-mercaptocinnamic acid (50 mg, 0.27mmol) was stirred with 7mL of polyphosphoric acid at 100-105 °C for 35 min. The reqaction mixture was poured on ice, the aqueous phase was extracted with ether (4x25mL), and the combined, dried (Na2SO4) organic phase was evaporated. Preparative thin- layer chromatography (CHCl3/CH3OH) of the residue afforded 25mg (57% yields) of crystalline (Table 1). Suitable crystals with reasonable sizes were mounted in a Lindeman capillary and allowed to stabilize at final temperature for one hour. The unit-cell parameters were determined repeatedly until the estimated standard deviations in cell dimensions did not vary beyond acceptable limits. Preparation of thiocoumarin from mercaptocinnamic acid and polyphosphoric acid Preparation of thiocoumarin from mercaptocinnamic acid and polyphosphoric acid Preparation of thiocoumarin from mercaptocinnamic acid and polyphosphoric acid Preparation of thiocoumarin from mercaptocinnamic acid and polyphospho Prediction of Raman intensities The prediction of Raman intensities are carried out by following the procedure outlined below: The Raman activities (Si) calculated by the GAUSSIAN 09W program and adjusted during the scaling procedure with MOLVIB are converted by relative Raman Intensities (Ii) using the following relationship derived from the basic theory of Raman scattering [6-8]. ( ) ( ) 4 o i i i i i f S I = 1 exp hc / kT ν −ν ν − − ν     7 where νo is the exciting frequency (in cm-1 units), νi is the vibrational wave number of the ith normal mode, h, c and k are the universal constants and f is a suitably chosen common normalization factor for all peak intensities. Molecular geometry The molecular structure of the TCN belongs to CS point group symmetry. The optimized monomer and dimer molecular structure of title molecule is obtained from GAUSSIAN 09W and GAUSSVIEW programs are shown in Fig. 1. The compound contains S and O connected with hetero aromatic ring. The structure optimization zero point vibrational energy of the title compound in HF and B3LYP/6–311++G (d, p) are -5572.0698 and -5577.9193 Hartrees, respectively. Experimental details of the unit cell parameter and multipole refinement parameter including the residual densities over the asymmetric unit are listed in Table 1. Fig 2 gives the ORTEP diagrams of the title compound, showing the thermal ellipsoids at 50% probability level along with atoms.The final atomic coordinates, thermal parameters, bond lengths, and angles of the compound are provided as Supporting Information. The bond structures satisfy Hirshfeld’s rigid bond test [9]. The largest differences of mean-square displacement amplitudes ΔA,B for thiocoumarin it is 8x10-4 Å2 for the S1- C1 bond. The experimental topological parameters of the covalent bonds within the molecule for the compound along with the values obtained from periodic theoretical calculations are given in Table 2. It is note that the values obtained from experiment and theoretical calculations are in good agreement. So the maximum difference being 0.06, and 0.08 Å for the C6-C5 and C8-S1 bonds, respectively, in the location of the bond critical points. The electron density and Laplacian values at the bond critical points are in good agreement, demonstrating that both experimental and theoretical methodologies provide comparative measures of topological and charge density properties. The C=O bond length and the associated bond critical points differ in each case, which is a consequence of the nature of 8 intermolecular interaction originating from the O-atom. Thiocoumarin has also three C-H….O interactions at this O-atom. Fig 3 traces the bond paths (experimental) between O-atoms and H-atoms with the (3, -1) bond critical points along with representative bond path between C-H and the Cπ for each compound. Laplacian maps from experimental analysis showing the distribution in the region of the bond paths along C-H…O and C-H…..Cπ are given in Fig 4. The multipole population parameters (Plm± and PV) along with k and k´ from experimental and theoretical refinements are provided Information’s. This study is to characterize intermolecular interactions involving C-H….O and C- H….π in each structure and evaluate them based on Koch and Popelier’s criteria. Molecular geometry The C-H….π interactions considered for this study are limited to the aromatic carbon atom regions of the compound. The details of all the parameters characterizing and interactions are given in Table 3. The inclusion of C-H….π interactions (van der Waals in nature) in our evaluation of the eight criteria arises from the fact that it provides an immediate comparison with a weak hydrogen bond such as C- H….O. Indeed C-H…O interactions define the possible limit of the hydrogen bond as can be seen in the evaluation of the eight criteria as follows. Thus these two interactions allow for the definition of a “region of overlap” between the hydrogen bond and a van der Waals interaction. The evaluation of integrated properties over the basin of the H-atoms involves the interactions forms the basis of these criteria. It determins the charge, dipolar polarization, potential energy, and volume of H-atoms involved in the C-H·····O and C-H·····π interactions considering the crystal (experimental and theoretical) and the isolated molecule. The theoretical values have been calculated for isolated molecules using both HF and DFT methods. Since the theoretical calculation in the crystal is evaluated using only DFT, the integration properties appear most reliable with respect to the DFT calculation for isolated molecule. The Table 3 has the value for the C-H·····O hydrogen bonds and from the C-H·····π van der Waals interactions. For example, the difference in charge on the H-atom increases from a value of 0 0376 e for C H·····π to a value of 0 0825 e for C H·····O hydrogen bond in the structure of 9 thiocoumarin. The intermediate values are taken by those H-atoms, which display both C-H·····π and C-H····· O character. Comparable electron losses were found in case of formation of a dihydrogen bond and also a set of C-H····· O bonds containing van der Waals complexes, indicating that in general smaller electron loss reflects weak interaction. The differences in the potential energy (PE) in all cases depict the expected trend destabilizing the H-atom upon crystal. So calculations of nuclear PE values, and the kinetic energy output from the calculation on isolated molecules is subtracted from the corresponding estimate of the total energy. It shows that the increase in the PE for C-H·····O hydrogen bond is larger compared to that in the C-H·····π regime (for thiocoumarin, 0.0363 autos -0.0126 au). The energy changes are between C-H····· O hydrogen bond and C-H·····π interaction. Molecular geometry A measure of the extent and direction of dipolar polarization of the atomic electron density is given by the magnitude of the moment, and this value decreases by 0.0755 au for the C-H·····O bond and 0.0290 au for a C- H·····π interaction in thiocoumarin. The final criterion, decrease of H-atom’s volume, is quite significant to bring in the difference in bond strength between the two interactions. The volume shrinks by 7.78 au for C-H·····O as compared to 2.90 au for C-H·····π interaction for thiocoumarin, respectively. All these criteria show similar trends and allow the types of interactions on quantitative steps are in Table [3]. 10 Optimized monomer and dimer molecular structure of Thiocoumarin. 1: Optimized monomer and dimer molecular structure of Thiocoumarin. Fig. 1: Fig. 1: p 11 Fig. 2: ORTEP diagram with labels for atoms of Tthiocoumarin at 90 K with 50% ellipsoid probability (non H-atoms). Fig. 2: ORTEP diagram with labels for atoms of Tthiocoumarin at 90 K with 50% ellipsoid probability (non H-atoms). Fig. 2: Fig. 2: 12 Fig 3: Bond path character in thiocoumarin showing the critical point locations along the C- H····· O and C-H····· Cπ interactions shown in red. Fig 3: 13 Table1 Crystal data and structure refinement for Thiocoumarin. Crystal data and structure refinement for Thiocoumarin. Molecular geometry Empirical formula C9H6OS Formula weight (g.mol-1) 162.2 Crystal size/mm 0.40×0.21×0.18 Crystal system Monoclinic Space group Pc Unit cell dimensions a (Å) 3.8056(3) b (Å) 8.4552(7) c (Å) 11.3651(10) α (°) 90 β (°) 95.629(4) γ (°) 90 V (Å3) 363.93(5) Z 2 μ (mm−1) 0.369 Dc/g cm-1 1.48 (sinθ /λ )max/Å-1 1.08 Reflections collected 6373 R(F); Rw(F) 0.0158; 0.0145 S 1.41 Nobs/Npar 20.14 Largest difference in peak and hole (eÅ-3) -0.203; 0.324 14 Table 2 Intramolecular Bond Critical Points for the Compounds and Bond Ellipticity Є = (λ1/λ2 - 1)a Bond (A-B) atom ρb 2ρb Rij d1(A-Cp) d2(Cp-B) λ1 λ2 λ3 Є S(1)-C(1) S(1) 1.359(38) -2.076(45) 1.7736 0.9107 0.8628 -8.45 -6.72 13.09 0.26 C(1) 1.293 -4.894 1.7715 0.9633 0.8082 -7.34 -6.19 8.64 0.19 S(1)-C(8) S(1) 1.336(43) - 1.219(51) 1.7472 0.8771 0.8700 -8.24 -6.44 13.47 0.28 C(8) 1.401 -6.132 1.7456 0.8596 0.7860 -7.99 -6.82 8.68 0.17 O(1)-C(1) O(1) 2.979(16) -23.635(83) 1.2229 0.7993 0.4236 -25.93 -24.50 26.79 0.06 C(1) 2.769 24.862 1.2241 0.7859 0.4381 -23.54 -22.24 20.92 0.06 C(1)-C(2) C(1) 2.008(13) -15.771(27) 1.4504 0.7208 0.7296 -14.94 -12.04 11.21 0.24 C(2) 1.894 -14.336 1.4508 0.7666 0.6842 -13.46 -11.69 10.82 0.15 C(2)-C(3) C(2) 2.401(15) -22.244(31) 1.3578 0.6674 0.6904 -18.03 -14.09 9.88 0.28 C(3) 2.224 -20.458 1.3558 0.6899 0.6659 -16.82 -13.35 9.71 0.26 C(9)-C(3) C(9) 1.987(14) -14.111(30) 1.4429 0.6981 0.7447 -14.52 -11.47 11.88 0.27 C(3) 1.901 -15.085 1.4415 0.7188 0.7227 -13.86 -11.85 10.63 0.17 C(9)-C(4) C(9) 2.184(14) -17.383(31) 1.4110 0.7054 0.7056 -15.46 -13.42 11.50 0.15 C(4) 2.005 -16.725 1.4103 0.7199 0.6904 -14.95 -12.13 10.36 0.23 C(5)-C(4) C(5) 2.217(18) -18.958(36) 1.3853 0.6747 0.7106 -17.03 -13.23 11.30 0.29 C(4) 2.130 -18.644 1.3837 0.6816 0.7020 -16.01 -12.80 10.17 0.25 C(6)-C(5) C(6) 2.241(17) -18.632(39) 1.4015 0.6364 0.7651 -15.60 -13.54 10.51 0.15 C(5) 2.091 -17.860 1.4019 0.6936 0.7084 -15.40 -12.94 10.48 0.19 C(6)-C(7) C(6) 2.288(14) -19.371(33) 1.3894 0.6759 0.7136 -16.49 -13.51 10.64 0.22 C(7) 2.111 -17.845 1.3892 0.6812 0.7080 -15.49 -12.74 10.38 0.22 C(8)-C(7) C(8) 2.308(13) -20.965(31) 1.3998 0.7317 0.6681 -18.05 -13.71 10.79 0.32 C(7) 2.090 -17.497 1.3984 0.7282 0.6702 -15.31 -12.61 10.42 0.21 C(8)-C(9) C(8) 2.109(14) -16.831(30) 1.4059 0.6632 0.7427 -15.38 -12.73 11.28 0.21 C(9) 2.037 -16.982 1.4058 0.7175 0.6883 -14.96 -12.39 10.37 0.21 a The values from periodic calculation using B3LYP/6-311++G method are given in Red colour. Molecular geometry 15 Table 3: Atomic Net Charges (q), Atomic Potential Energy (PE), Atomic Dipolar Polarization (M), and Atomic Volume (V) of the H-Atoms in the Crystal and in the Isolated Molecule and Their Corresponding Differences (in au). Atomic Net Charges (q) Interactions atom q (crystal) q (isolated) Δq (crystal - isolated) expt.(E) theo.(T) HF DFT E-HF E-DFT T-HF T-DFT O(1)...H(3) H(3) 0.0991 0.0980 -0.0164 -0.0155 0.1155 0.0836 0.1144 0.0825 O(1)...H(4) H(4) 0.2872 0.1092 -0.0334 0.0058 0.3206 0.2814 0.1426 0.1034 O(1)...H(6) H(6) 0.3029 0.0809 -0.0250 0.0126 0.3279 0.29.3 0.1059 0.0683 C(4)...H(7) H(7) 0.2810 0.0561 -0.0193 0.0185 0.3003 0.2625 0.0754 0.0376 Atomic Potential Energy (PE) Interactions atom q (crystal) q (isolated) Δq (crystal - isolated) expt.(E) theo.(T) HF DFT E-HF E-DFT T-HF T-DFT O(1)...H(3) H(3) -1.2408 -1.2098 -1.2865 -1.2461 0.0457 0.0053 0.0767 0.0363 O(1)...H(4) H(4) -1.0306 -1.2061 -1.2960 -1.2498 0.2654 0.2192 0.0899 0.0437 O(1)...H(6) H(6) -0.9684 -1.2293 -1.2887 -1.2454 0.3203 0.2770 0.0594 0.1610 C(4)...H(7) H(7) -0.9787 -1.2526 -1.2837 -1.2400 0.3050 0.2613 0.0311 -0.0126 Atomic Dipolar Polarization (M) Interactions atom q (crystal) q (isolated) Δq (crystal - isolated) expt.(E) theo.(T) HF DFT E-HF E-DFT T-HF T-DFT O(1)...H(3) H(3) 0.1033 0.0953 0.0986 0.1253 0.0047 -0.0220 -0.0033 -0.0300 O(1)...H(4) H(4) 0.0478 0.0505 0.0978 0.1260 -0.0500 -0.0782 -0.0473 -0.0755 O(1)...H(6) H(6) 0.1021 0.0743 0.0969 0.1230 0.0052 -0.0209 -0.0226 -0.0487 C(4)...H(7) H(7) 0.1561 0.0943 0.0971 0.1233 0.0590 0.0328 -0.0028 -0.0290 Atomic Volume (V) Interactions atom q (crystal) q (isolated) Δq (crystal - isolated) expt.(E) theo.(T) HF DFT E-HF E-DFT T-HF T-DFT O(1)...H(3) H(3) 34.45 41.27 47.44 47.96 -12.99 -13.51 -6.17 -6.69 O(1)...H(4) H(4) 27.97 40.90 48.53 48.68 -20.56 -20.71 -7.63 -7.78 O(1)...H(6) H(6) 34.80 43.28 48.54 48.52 -13.74 -13.72 -5.26 -5.24 C(4)...H(7) H(7) 38.90 45.16 47.94 48.06 -9.04 -9.16 -2.78 -2.90 4.2. Vibrational assignments The TCN consists of 17 atoms, and belongs to Cs symmetry. Hence the numbers of normal modes of vibrations for TCN are 45. The bands that belong to the in-plane modes are represented as A’ while the out-of-plane modes as A” Thus the 45 normal modes of vibrations are distributed as Vib = 31A’+14A”. The internal and symmetric co-ordinates of TCN are shown in Tables S1-S2, 16 respectively. The harmonic-vibrational frequencies calculated for TCN at HF and DFT-B3LYP levels using the triple split valence basis set along with the diffuse and polarization functions, 6–311++G (d, p), observed FT-IR and FT-Raman frequencies for various modes of vibrations are presented in Table S3. Although basis set are marginally sensitive as observed in the HF and DFT values using 6– 311++G (d, p), reduction in the computed harmonic vibrational frequencies are noted. Without affecting the basic level of calculations, it is customary to scale down the calculated harmonic frequencies in order to get an agreement with the experimental values. The scaled calculated frequencies minimize the root-mean square difference between calculated and experimental frequencies for bands with definite identifications. The IR and Raman spectra of experimental and are given in the Figures.S1-S2., respectively. C–H vibrations The aromatic organic compounds structure shows the presence of asymmetric C–H stretching vibrations in the region 3100–3000cm-1 [10], which is the characteristics region for recognition on C– H stretching vibrations. In the title molecule, bands have been observed at 3090, 2172 and 2842,2821, 2801, 2157cm-1 assigned to C–H stretching vibrations in FT-IR and FT-Raman spectra, respectively. The C–H in-plane and out of- plane bending vibrations normally take place as a number of intensity sharp bends in the region of 1300–1000 and 1000–750cm-1 , respectively. The bands for C–H in plane bending vibrations are identified at 1499, 1472, 1456,1442,1421,1392 and 1455 cm-1 in FT-IR and FT- Raman spectra, respectively and the C–H out-of-plane bending vibrations are found at 721 and 958,788,604,582,570cm-1. Theoretical Positive field and negative field for C–H in-plane and C–H out- of-plane bending vibrations by B3LYP/ 6–311++G (d, p) method showed excellent agreement with recorded spectrum. C- C vibrations 17 Generally the C-C stretching vibrations in aromatic compounds are seen in the region of 1430– 1650cm-1. The presence of conjugate substituent such as C-C causes a heavy doublet formation around the region 1625–1575cm-1[11]. The six ring carbon atoms undergo coupled vibrations, which are known as skeletal vibrations give a maximum of four bands in the region 1660–1420cm-1. As predicted in the earlier references [12], the prominent peaks are due to strong C-C stretching 1999, 1971, 1852, 1720,1709,1672,1656 cm-1 and 1952,1914,1342 cm-1 in FT-IR and Raman, respectively are due to C–C vibrations for the title compound. The C−C in-plane and out-of-plane bending modes of TCN are in Table S3. C−S vibrations The C−S stretching bands are usually observed in the range 670-930 cm-1 with a moderate intensity. In general, the assignment of band due to the C−S stretching vibration in different compounds is difficult in the infrared since the band is of variable intensity and may be found over the wide region 1035-245 cm-1, whereas, in general, C−S stretching vibrations in IR spectra which are normally easy to identify [12]. The C-S group is less polar than carbonyl links and has a considerably weaker band. In consequence, the band is not intense, and it falls at lower frequencies, where it is much more susceptible to coupling effects. Identification is therefore difficult and uncertain. The absorption of C−S group with other substituents usually appears in between 1250-1020 cm-1. Consideration of these factors leads to assign the IR band at 1366, 1347 cm-1 in TCN to C-S stretching vibration. Besides the stretching vibrations, the sulfur atom gives raise to bending vibrations with the neighbouring atoms. The respective peaks for in-plane and out-of-plane bending vibrations C−S group are listed in Table S3. All the C−S vibrational bands of TCN are in line with the literature values and nearly in agreement with the computed values by B3LYP/6-311++G(d,p). C= O vibrations 18 The interaction of carbonyl group with other groups present in the system did not produce such a drastic and characteristic change in the frequency of C- O stretch as did by interaction of C–S stretch. The carbon–oxygen double bond is formed by pπ–pπ between carbon and oxygen. Because of the different electron negativities of carbon and oxygen atoms, the bonding electrons are not equally distributed between the two atoms. The lone pair of electrons on oxygen also determines the nature of the carbonyl group. The position of the C-O stretching vibration is very sensitive to various factors such as the physical state, electronic effects by substituent, ring strains. Normally carbonyl group vibrations occur in the region 1850–1600cm−1 [12]. In this study, the C-O stretching vibrations of TCN are observed at 2148 cm−1 in FTIR spectrum and these modes are confirmed by their PED values. The in-plane and out-of-plane bending vibrations of C-O group have also been identified and presented in Table S3 for the title molecule. 5. Prediction of first hyperpolarizability The first hyperpolarizability (β0) of this novel molecular system and the related properties (β0,α0) of TCN are calculated using the B3LYP/6-311++G(d,p) basis set, based on the finite field approach. From the applied electric field, the energy of a system is a function of the electric field. So the first hyperpolarizability is a third-rank tensor that can be described by a 3 × 3 × 3 matrix. The 27 components of the 3D matrix can be reduced to 10 components due to the Kleinman symmetry [12]. It can be given in the lower tetrahedral. The components of β are defined as the coefficients in the Taylor series expansion of the energy in the external electric field. When the external electric field is weak and homogenous, this expansion becomes: E = E0– μαFα – 1/2ααβFαFβ – 1/6βαβγ Fα FβFγ +....... E = E0– μαFα – 1/2ααβFαFβ – 1/6βαβγ Fα FβFγ +....... Where E0 is the energy of the unperturbed molecules, Fα the field at the origin and μα, ααβ and βαβγ are the components of dipole moment, polarizability and the first hyperpolarizabilities, 19 respectively. The total static dipole moment μ, the mean polarizability α0 and the mean first hyperpolarizability β0, using the x, y, z components they are defined as follows: μ = (μx2+μy2+μz2)1/2 α0 = (αxx+αyy+αzz)/3 α = 2-1/2 [(αxx– αyy)2 + (αyy – αzz)2 + (αzz – αxx)2 + 6αxx2 )1/2 β0 = (βx2+βy2+βz2)1/2 βvec = 3/5[(βx2+βy2+βz2)1/2] where βx = βxxx + βxyy +βxzz βy = βyyy + βyxx +βyzz βz = βzzz + βzxx +βzyy βx = βxxx + βxyy +βxzz βy = βyyy + βyxx +βyzz βz = βzzz + βzxx +βzyy The β0 components of GAUSSIAN 09W output are reported in atomic units and therefore the calculated values are converted into e.s.u. units (1 a.u. = 8.3693 × 10-33 e.s.u.). The calculated value of hyperpolarizability and polarizability of TCN are tabulated in Table S4. From this conclusion that the title molecules are an attractive object for future studies of NLO properties. 6. HOMO–LUMO energy gap and related molecular properties The properties of the frontier molecular orbitals(FMOs) are very useful for physicists and chemists. The electron densities of these FMOs were used for predicting the most reactive position in π‒electron systems and also explained several types of reactions in conjugated systems [13]. Moreover, eigen values of the lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) and their energy gap reflect the chemical reactivity of the molecule. Recently the energy gap between HOMO and LUMO has been used to prove the bioactivity from intramolecular charge transfer (ICT) [13].The HOMO‒LUMO energy gap for the studied compound is calculated by B3LYP/6‒311G(d,p). The HOMO and LUMO pictures are shown in Figure S4. It is found that the HOMO level is mainly localized on the benzene ring π-system and extended to the C1-C2, C5-S6, 20 O11-C1, C9-C8 and C10-C5 bonds. The LUMO is located mainly on the O group of the heterocycles. The EHOMO and ELUMO are calculated to be -0.24549a.u and -0.08935a.u espectively. The HOMO‒LUMO energy gap (ΔE) represents the lowest energy electronic transition. In the studied compound, the HOMO‒LUMO energy gap of the studied compound is 0.15614a.u. (Figures S4) . In this investigation, the more relevant electronic potential (IP), electron affinities (EA), chemical potential (μ) it is the negative of electro negativity (χ), hardness (η), softness (S) and the electrophilic index(ω) were calculated. The ionization potential is calculated as the energy difference between the energy of the molecule derived from electron-transfer (radical cation) and the respective neutral molecule; IP = Ecation - En. 6. HOMO–LUMO energy gap and related molecular properties The EA was computed as the energy difference between the neutral molecule and the anion molecule: EA = En+ Eanion [16].The HOMO and LUMO energy was also used to estimate the IP and EA in the framework of Koopmans’ theorem: IP = -εHOMO and EA= - εLUMO Within the framework of the density functional theory (DFT), one of the global quantities is chemical potential (μ), which is measures the escaping tendency of an electronic cloud, and equals the slope of the Energy versus N(number of electrons) curve at external potential ν(r) : µ = (∂E/∂N)V(r) µ = (∂E/∂N)V(r) Finite difference approximation to Chemical Potential gives, Finite difference approximation to Chemical Potential gives, Finite difference approximation to Chemical Potential gives, χ = -µ = -(∂E/∂N)V(r) χ = -µ = -(∂E/∂N)V(r) The theoretical definition of chemical hardness has been provided by the density functional theory as the second derivative of electronic energy with respect to the number of electrons N, for a constant external potential ν(r): η = ½(∂2E/∂N2)V(r) = ½( ∂µ/∂N)V(r) η = ½(∂2E/∂N2)V(r) = ½( ∂µ/∂N)V(r) Finite difference approximation to Chemical hardness gives, Finite difference approximation to Chemical hardness gives, Finite difference approximation to Chemical hardness gives, Finite difference approximation to Chemical hardness gives, 21 η = ( I-A )/2 η = ( I-A )/2 η = ( I-A )/2 For Insulator and semiconductor, hardness is half of the energy gap (εHOMO - εLUMO ), and the softness is given as [14] : S = 1/2η=(∂2E/∂N2)V(r)= (∂E/∂N)V(r) S = 1/2η=(∂2E/∂N2)V(r)= (∂E/∂N)V(r) Electrophilicity index is a measure of energy lowering due to maximal electron flow between donor and acceptor. Electrophilicity index (ω) is defined as, Electrophilicity index is a measure of energy lowering due to maximal electron flow between donor and acceptor. Electrophilicity index (ω) is defined as, ω= µ2/2 η ω= µ2/2 η B3LYP functional used in this study has a high efficient to calculate the electronic properties for the organic studied molecules, such as ionization potentials (IP), electron affinities (EA), electro negativity (χ), absolute hardness (η), absolute softness (S), electrophilic index (ω). The behavior of electro negativity, softness and electrophilic index for the studied molecules shows the magnitude large than these for the original ring, adding the radicals give the molecule more softness. The calculated value of electrophilicity index describes the biological activity of TCN. 6. HOMO–LUMO energy gap and related molecular properties All the calculated values of hardness, potential, softness and electrophilicity index are shown in Table S5. 7. Multipole Model and Deformation Densities The final refinement, of the Hirshfeld rigid bond test was applied to the covalent bonds involving non-hydrogen atoms. The values of maximum differences of mean-square displacement amplitudes are found to be 9(2) x10-4 Ǻ2 at C8-C9 respectively, which are below the standard limit of 10-3 Ǻ2. The residual electron densities calculated (with I >3 σ (I)) over the molecular planes are almost minimum and maximum densities are in the range of -0.223 to 0.130 e Ǻ-3. The residuals are very similar to the reported values. The static deformation electron density maps obtained from experimental and theoretical analyses of the crystal shown in Figure S5 The electron lone pairs of O atom, except the atoms affected by disorder, are clearly visible in the deformation electron density maps. The deformation electron density maps associated with the other part of the molecule were also shown to 22 have high accurate bonding features. The statistical analysis was performed on the static deformation electron density grids obtained from the experimental and theoretical multipole modeling. The correlation between the experimental and theoretical deformation electron densities Δ ρ is 90% and 87%. The diffraction data of the crystal are of slightly better quality in terms of resolution and R- factors, which is in accordance with the higher correlation [15]. 8. Electrostatic Potentials Analysis of deformation electrostatic potential (ESP) derived from the deformation electron densities on the molecular surfaces was performed to highlight the effect of crystalline environment and also to point out the differences and the similarities between the two polymorphic forms. The construction of a three-dimensional ESP map plotted over the molecular surfaces from experimental charge densities clearly brings out the differences of electrostatic nature of the three forms (Figures.S6-S8). The electropositive and electronegative surfaces are well separated in the forms. It displays a larger electronegative surface in ESP mapping. This is due to the conformational difference at the C-O side chain and additionally due to the involvement of the π electrons of three bonded atoms C1-C2-H12 in C-H… π type of contacts and the presence of π ….π contacts. The electronegative surface is mainly seen around the oxygen atoms, which are involved in C-H…O type of intermolecular contacts. However, the atom O11 shown to have less prominent electronegative surface compared to other S atom in the structures. This is because the atom O11 is not involved in any intermolecular contacts, whereas the corresponding atom in the S is seen to interact remotely with the neighbour molecule (Figure S9). The corresponding maps from the theoretical analysis revealed similar features. The ESP maps clearly emphasize the preferred binding sites to form the networks of interactions and also highlight the difference in nature of interactions [16]. 9. THERMODYNAMIC PROPERTIES 9. THERMODYNAMIC PROPERTIES 23 Based on the vibrational analysis at B3LYP/6-311++G(d,p) level and statistical thermodynamics, the standard thermodynamic functions; heat capacities (Cºp,m), entropies (Sºm) and enthalpies changes (Hºm) are obtained at constant pressure for zero (monomer and dimer) and applied electric field in Table S6. As seen from Table S6, the standard thermodynamic functions increase with increasing temperature, due to the intensities of molecular vibration increase; it is observed that the entropy increases but the enthalpy and heat capacity remain constant. According to Boyle’s law for gases, a molecule is compressed at constant temperature its volume decreases [17-18]. Due to decreasing volume, the number of possible sites that occupied by particles of the molecule may be restricted. Thus, the entropy tends to decrease with increasing pressure at constant temperature. This investigation will be helpful for the further studies of the title molecule. The correlation graph for zero (monomer and dimer) applied EFs are shown in Figure S10 and the corresponding fitting equations are as follows: For 0.00 VÅ−1 0 m S = 210.23396+ 0.71772T -0.00238× 10−5 T2 (R2 = 0.99884) 0 p,m C = 9.94698+ 0.48974T +3.33885× 10−5 T2 (R2 = 0.99886) 0 m H = -2.8146+1 .31434T -2.62041× 10−5T2 (R2 = 0.9994) For 0.00 VÅ−1 For 0.01 VÅ−1 or 0.01 VÅ 0 m S = 227.93312+ 0.41634T +8.49829× 10−5 T2 (R2 = 0.99906) 0 p,m C = 1.81101-6.67441T −0.06719× 10−4 T2 (R2 = 0.99988) 0 m H = 3.68006+ 1.71434T + 2.01163× 10−4 T2 (R2 = 0.99942) For 0.01 VÅ 0 m S = 227.93312+ 0.41634T +8.49829× 10−5 T2 (R2 = 0.99906) 0 p,m C = 1.81101-6.67441T −0.06719× 10−4 T2 (R2 = 0.99988) 0 m H = 3.68006+ 1.71434T + 2.01163× 10−4 T2 (R2 = 0.99942) For 0.02 VÅ−1 0 m S = 228.93312+ 0.23581T +7.55761× 10−5 T2 (R2 = 99928) 0 p m C = 1.61301+ 0.4950 T −2.1175 × 10−4 T2 (R2 = 0.9988) For 0.02 VÅ−1 p, 0 m H = −3.80606-1.1436T + 2.09860× 10−4 T2 (R2 = 0.99947) 24 For Dimer 0 m S = 311.40643+ 1.13227T – 3.22688× 10−4 T2 (R2 = 0.99981) 0 p,m C = 6.48149+ 5.36266T −0.0397× 10−4 T2 (R2 = 0.99866) 0 m H = −5.87053+ 2.49217T - 0.0397× 10−4 T2 (R2 = 0.99960) For Dimer 0 m S = 311.40643+ 1.13227T – 3.22688× 10−4 T2 (R2 = 0.99981) 0 p,m C = 6.48149+ 5.36266T −0.0397× 10−4 T2 (R2 = 0.99866) 0 m H = −5.87053+ 2.49217T - 0.0397× 10−4 T2 (R2 = 0.99960) For Dimer 10. ATOMIC CHARGES Atomic charges for molecule can be derived in multiple ways with a high level of convergence to the same values and only then, knowing the remaining uncertainty, are suited for molecular simulations. Mulliken charges arise from the mulliken population analysis [19] and provide a means of estimating partial atomic charges from calculations carried out by the methods of computational chemistry. Generally, Mulliken population analysis (MPA) and Natural Population analysis (NPA) methods are used to calculate the atomic charges of atoms in molecules. The difference of MPA charges distribution for zero and various applied EFs (±0.02VÅ−1) are listed in Table S7. The MPA charges of all C-atoms for zero field vary from 2.0632 a.u to –1.0486 a.u. When the applied field increases, the charge value decreases for C-atoms. The variations in MPA charge are small due to be systematic and almost uniform. Additionally, the magnitudes of the hydrogen atomic charges were noted to be only positive values, indicating the charge transfer from hydrogen to carbon atom. The results show that: The results show that: • All the hydrogen atoms in the molecules lost electrons. • The sulphur atoms in the molecules lost electrons. • Carbon atomic charges were found to be both positive values except C4. • Charge migration to heavy atoms can be related to molecular interactions. • Oxygen atoms accept electrons while the atomic charges were found to be negative. 25 Conclusion Experimental and theoretical charge density calculation are belonging to the thiocoumarin suggests that there exists a well-defined “region of overlap” between hydrogen bond and van der Waals interaction. But the experimental results are closer to that at a critical distance (donor)-H·····A (acceptor) interaction is switched from “hydrogen bond” to “van der Waals” type. Theoretically it is sufficient to describe both qualitative and quantitative measures to evaluate the nature of C-H·····O and C-H ····· π interactions. HOMO - LUMO energy gap explains the eventual charge transfer interaction taking place within the molecule.The summation of total electrostatic interaction energy values of the monomer suggests electrostatic interaction. The plotting of 3 dimensional deformation electrostatic potential maps over the molecular surfaces elucidates the nature of interactions.The result describes the variation in electrostatic and transport properties for zero and various external applied field. The variation in MPA charges are small due to the application of EFs: however, in most cases it is found to be systematic and almost uniform. The results clearly show the superiority of MPA scheme. This study may be useful to design new molecules with more electrical conductivity. A complete vibrational analysis of TCN have been investigated by FT-IR and FT-Raman spectroscopies and are based on ab initio and DFT calculation at the HF/6-311++G(d,p) and B3LYP/6-311++G(d,p) levels and their frequencies are compared. In particular, the results of DFT-B3LYP method indicates better fit to experimental values. The optimized geometrical parameters calculated at B3LYP method are slightly larger than calculated HF level. The predicted first hyperpolarizability reveals that the title molecule is an attractive object for future studies of non-linear optical properties. The results of the TED calculation are also useful for understanding the delocalisation of electrons involved in this compound. Experimental and theoretical charge density calculation are belonging to the thiocoumarin suggests that there exists a well-defined “region of overlap” between hydrogen bond and van der Waals interaction. But the experimental results are closer to that at a critical distance (donor)-H·····A (acceptor) interaction is switched from “hydrogen bond” to “van der Waals” type. Theoretically it is sufficient to describe both qualitative and quantitative measures to evaluate the nature of C-H·····O and C-H ····· π interactions. HOMO - LUMO energy gap explains the eventual charge transfer interaction taking place within the molecule.The summation of total electrostatic interaction energy values of the monomer suggests electrostatic interaction. 11. Potential energy surface scan The potential surface scans of theoretical approximations were performed for title molecule is shown in Figure S11. The bond angle C1-S6-O11 for TCN is also the relevant coordinate for conformational flexibility within the molecule. During the calculations all the geometrical parameters were simultaneously relaxed during the calculations while the C1-S6- O11 angle was varied in steps of 10o, 20o, 30o, ….., 70o. The barrier of this rotation mildly increases shows that the sulfonyl group cannot be hindered by halogen atom present in the position (due to intramolecular bonding). The large energy barrier indicates that the rotation of C=O group is very much restricted. 26 Conclusion The plotting of 3 dimensional deformation electrostatic potential maps over the molecular surfaces elucidates the nature of interactions.The result describes the variation in electrostatic and transport properties for zero and various external applied field. The variation in MPA charges are small due to the application of EFs: however, in most cases it is found to be systematic and almost uniform. The results clearly show the superiority of MPA scheme. This study may be useful to design new molecules with more electrical conductivity. A complete vibrational analysis of TCN have been investigated by FT-IR and FT-Raman spectroscopies and are based on ab initio and DFT calculation at the HF/6-311++G(d,p) and B3LYP/6-311++G(d,p) levels and their frequencies are compared. In particular, the results of DFT-B3LYP method indicates better fit to experimental values. The optimized geometrical parameters calculated at B3LYP method are slightly larger than calculated HF level. The predicted first hyperpolarizability reveals that the title molecule is an attractive object for future studies of non-linear optical properties. The results of the TED calculation are also useful for understanding the delocalisation of electrons involved in this compound 27 Reference Reference [1] Parthapratim Munshi, Christian Jelsch, Venkatesha R. Hathwar, and Tayur N. Guru Row. Experimental and Theoretical Charge Density Analysis of Polymorphic Structures: The Case of Coumarin 314 Dye Crystal Growth & Design. DOI: 10.1021/cg900649m.2010. [2] Nishio, M . CH/π hydrogen bonds in crystals Cryst. Eng. Commun. 2004, 6, 130-158. [3] Ellena, J.; Goeat, A. E.; Howard, J. A. K.; Punte, G. Role of the Hydrogen Bonds in Nitroanilines Aggregation:Charge Density Study of 2-Methyl-5-nitroanilineJ. Phys. Chem.2001, 105, 8696-8708 [4] Axel Diefenbach and F. Matthias Bickelhaupt. Activation of H−H, C−H, C−C, and C−Cl Bonds by Pd(0). Insight from the Activation Strain Model J. Phys. Chem. A 2004, 108(40), 8460–8466. [5] M. J. Frisch. G. W. Trucks, H.B Schlegal et al., GAUSSIAN 09,Revision A.02, Gaussian, Inc.,Wallingford CT,2009. [6] P.L.Polavarapu, Ab initio vibrational Raman and Raman optical activity spectra J.Phys.Chem.4 (1990) 8106–8112. [7] G. Keresztury, S. Holly, J. Varga, G. Besenyei, A.V. Wang, J.R. Durig,Spectrochim. Acta A 49 (1993) 2007. [8] G. Keresztury, in: J.M. Chalmers, P.R. Griffiths (Eds.), Handbook of Vibrational Spectroscopy, vol. 1, John Wiley & Sons Ltd., 2002. [9] Parthapratim Munshi and Tayur N. Guru Row. Exploring the Lower Limit in Hydrogen Bonds: Analysis of Weak C−H···O and C−H···π Interactions in Substituted Coumarins from Charge Density Analysis J. Phys. Chem. A 2005, 109, 659-672. [10] D.Cecily Mary Glory, R.Madivanane and K.Sambathkumar Electronic Structure Investigations of 3 and 5- Diamino-1,2,4-Triazole by UV–Visible , NMR Spectral Studies and Homo-Lumo Analysis by AB Initio and DFT Calculations Elixir Comp. Chem. 89 (2015) 36730-36741. [11] Kuppusamy Sambathkumar Vibrational spectra, NBO, HOMO–LUMO and conformational stability studies of 4-hydroxythiobenzamideSpectrochim. Acta A 147 (2015) 51-66. [12] K. Sambathkumar, Density Functional Theory Studies of Vibrational Spectra, Homo- Lumo, NBO and NLO Analysis of Some Cyclic and Heterocyclic Compounds (Ph.D. Thesis), Bharathidasan University, Tiruchirappalli, August 2014. [13] S.Manimaran, K.Sambathkumar,,N.Rajkamal M. Venkatachalapathy& K.Chandrasekaran Indian J. Pure Appl. Phys., 57 (2019) 236-2649. 28 [14] D. Cecily Mary Glory, K. Sambathkumar, R. Madivanane , G. Velmurugan,R. Gayathri , S. Nithiyanantham, M. Venkatachalapathy, N. Rajkamal J of Molecular Structure 1163 (2018) 480-495. [15] U. Salzner, J.B. Lagowski, P.G. Pickup and R.A. Poirier, Density-functional theory of atoms and molecules, J. Comput. Chem., 18(1997). [16] Parthapratim Munshi, Christian Jelsch, Venkatesha R. Hathwar and Tayur N. Guru Row Revealing the Polarizability of Organic Fluorine in the Trifluoromethyl Group: Implications in Supramolecular Chemistry Cryst. [19] K Sambathkumar,Molecular structure and vibrational spectra of 2, 4, 6 -trimethylbenzene sulphonyl chloride (FTIR & Raman) by quantum chemical calculations Indian J. Pure Appl. Phys., 58 (2020) 589-598. Reference Growth Des.,2014, 14 (11), 5366–5369 [17] K Sambathkumar, S Nithiyanantham, Synthesis, characterization and theoretical properties of coumarin NLO single crystal by DFT method J Mater Sci: Mater Electron 2017,28, 6529– 6543. [18] S. Manimaran,• K. SambathKumar,R. Gayathri,K. Raja,N. Rajkamal,M.Venkatachalapathy,G. Ravichandran,C.Lourdu EdisonRaj,Medicinal Plant Using Ground State Stabilization of Natural Antioxidant Curcumin by Keto-Enol Tautomerisation Natural Products and Bioprospecting (2018) 8:369–390. [19] [19] K Sambathkumar,Molecular structure and vibrational spectra of 2, 4, 6 -trimethylbenzene sulphonyl chloride (FTIR & Raman) by quantum chemical calculations Indian J. Pure Appl. Phys., 58 (2020) 589-598. Figures Figure 1 Optimized monomer and dimer molecular structure of Thiocoumarin. Figure 1 Optimized monomer and dimer molecular structure of Thiocoumarin. Optimized monomer and dimer molecular structure of Thiocoumarin. Figure 2 Figure 2 ORTEP diagram with labels for atoms of Tthiocoumarin at 90 K with 50% ellipsoid probability (non H- atoms). ORTEP diagram with labels for atoms of Tthiocoumarin at 90 K with 50% ellipsoid probability (non H- atoms). Figure 3 Bond path character in thiocoumarin showing the critical point locations along the C-H····· O and C-H····· Cπ interactions shown in red. Figure 3 Bond path character in thiocoumarin showing the critical point locations along the C-H····· O and C-H····· Cπ interactions shown in red. Bond path character in thiocoumarin showing the critical point locations along the C-H····· O and C-H····· Cπ interactions shown in red.
https://openalex.org/W1985871285
https://europepmc.org/articles/pmc3109408?pdf=render
English
null
The Involvement of Neuroinflammation and Kynurenine Pathway in Parkinson's Disease
Parkinson's disease
2,011
cc-by
9,179
SAGE-Hindawi Access to Research Parkinson’s Disease Volume 2011, Article ID 716859, 11 pages doi:10.4061/2011/716859 SAGE-Hindawi Access to Research Parkinson’s Disease Volume 2011, Article ID 716859, 11 pages doi:10.4061/2011/716859 SAGE-Hindawi Access to Research Parkinson’s Disease Volume 2011, Article ID 716859, 11 pages doi:10.4061/2011/716859 Anna Zinger,1 Carlos Barcia,2 Maria Trinidad Herrero,2 and Gilles J. Guillem f , y f , , p 3St Vincent’s Centre for Applied Medical Research, Darlinghurst, NSW 2010, Australia Correspondence should be addressed to Gilles J. Guillemin, g.guillemin@unsw.edu.au Received 1 December 2010; Accepted 31 January 2011 Received 1 December 2010; Accepted 31 January 2011 Academic Editor: Heinz Reichmann Copyright © 2011 Anna Zinger et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Parkinson’s disease (PD) is a common neurodegenerative disorder characterised by loss of dopaminergic neurons and localized neuroinflammation occurring in the midbrain several years before the actual onset of symptoms. Activated microglia themselves release a large number of inflammatory mediators thus perpetuating neuroinflammation and neurotoxicity. The Kynurenine pathway (KP), the main catabolic pathway for tryptophan, is one of the major regulators of the immune response and may also be implicated in the inflammatory response in parkinsonism. The KP generates several neuroactive compounds and therefore has either a neurotoxic or neuroprotective effect. Several of these molecules produced by microglia can activate the N-methyl- D-aspartate (NMDA) receptor-signalling pathway, leading to an excitotoxic response. Previous studies have shown that NMDA antagonists can ease symptoms and exert a neuroprotective effect in PD both in vivo and in vitro. There are to date several lines of evidence linking some of the KP intermediates and the neuropathogenesis of PD. Moreover, it is likely that pharmacological modulation of the KP will represent a new therapeutic strategy for PD. 1. Introduction Microglia respond to a wide range of immunologic stimuli or CNS injuries and either initiate protective and/or neuroinflammatory processes [11]. The SN contains the highest concentration of microglia compared with other brain areas [12]. 2.2. Human Studies. A large epidemiological study on approximately 150,000 men and women has shown that the use of nonsteroidal anti-inflammatory drugs (NSAIDs) can prevent or delay the onset of PD [28]. Chen et al. have also observed a similar effect in chronic users of ibuprofen, a NSAID acting on cyclooxygenase (COX) [29]. A correlation has also been found between high plasma concentrations of interleukin-6, a proinflammatory cytokine, and an increased risk of developing PD [30]. Moreover, in vivo imaging studies on patients with idiopathic PD have shown an increase in neuroinflammatory areas in basal ganglia, striatum, and frontal and temporal cortical regions compared with age- matched healthy controls [31]. All of these studies suggest that microglial activation occurs at an early stage of the disease either before (or in parallel with) the important loss of dopaminergic neurons. In postmortem PD tissues, acti- vated microglial cells have been detected around impaired dopaminergic neurons in the SN, thus demonstrating the presence of neuroinflammation [32]. As previously dis- cussed, MPTP causes Parkinsonism in both humans and primates. This leads to the chronic presence of activated microglia around dopaminergic neurons in the SN for up to 10 years after exposure [33, 34], even without L-DOPA treatment [35]. Substantial evidence of microglial activation associated with dopaminergic neuronal damage suggests that degenerating neurons initiate microgliosis, which then leads to further neuronal loss. Microglial activation represents an initiator and/or a secondary responder in this disease process. Therefore, suppressing neuroinflammation by pre- venting microglial activation could potentially slow down or stop this continuous and deleterious cycle which damages neurons. Resting microglia have a characteristic ramified mor- phology; the small cell body remains stationary whilst the long branches are constantly moving and are sensitive to any minor physiological changes [13, 14]. At the site of inflammation, activated microglia change their morphology becoming amoeboid and may act similarly to macrophages: they possibly perform phagocytosis, express increased levels of major histocompatibility complex (MHC) antigens, and secrete various cytotoxins, which may ultimately activate additional microglia to remove harmful stimuli and even initiate healing processes [15, 16]. 1. Introduction pathway, followed by stratial dopaminergic denervation which causes pathological changes in neurotransmission of basal ganglia motor circuit and results in characteristic Parkinsonian symptoms [5]. Another pathological hallmark of the disease is the presence of protein inclusions called Lewy bodies (LBs), which are abnormal intracellular α- synuclein (SYN) aggregates in the cytoplasm and axons of the remaining neurons [6]. Neurons containing LBs undergo neurodegenerative processes and subsequently die. Parkinson’s disease (PD) is the most common movement disorder and is the second most common chronic progressive neurodegenerative disorder after Alzheimer’s disease. PD is a sporadic and age-dependent disease in 90% of cases and affects more than 1% of the world population over the age of 65 [1]. PD is characterised by motor symptoms including bradykinesia, tremor, rigidity, postural instability as well as nonmotor symptoms such as dementia, sleep disturbance, neurobehavioral, and sensory abnormalities [2]. To date, there is no available cure for PD. However, L-Dopa and dopaminergic agonists are useful in treating PD symptoms. This type of therapy mainly aims to replace dopamine in the striatum but does not slow neurodegener- ative processes. Moreover, long-term use is associated with serious side effects such as dyskinesia and motor fluctuations [7] resulting in a diminished effect of treatment [8]. PD is neuropathologically characterized by the loss of midbrain-pigmented neurons in the substantia nigra pars compacta (SNpc). Under normal conditions, these neurons produce dopamine at the striatum and other basal ganglia nuclei [3]. It has been estimated that at the onset of PD symp- toms, up to 70% of dopaminergic neurons have been lost. Postmortem examinations have also shown that more than 90% of these neurons have been depleted [4]. Dopaminergic loss leads to an irreversible degeneration of the nigrostriatal Although the aetiology of PD is relatively unknown, it has been suggested that there is an association with mito- chondrial dysfunction in nigral neurons and neurotoxicity from excess glutamate and reactive oxygen species (ROS) 2 2 Parkinson’s Disease toxic to dopaminergic neurons and is often used to induce an in vivo PD-like disease in animal models [25]. Moreover, inhibition of microglial activation results in a strong decrease in neurotoxicity in both MPTP mouse and LPS rat models [26, 27]. production [9, 10]. Microglia are the prime immune cells of the central nervous system (CNS) and are important pro- ducers of neuroactive molecules involved in oxidative stress, excitotoxicity and neuroinflammation. 1. Introduction The total number of MHC class II microglia has been shown to be significantly increased not only in SN and putamen but also in the hippocampus, transentorhinal cortex, cingulate cortex, and temporal cortex in PD brains [17]. This implies that microglia are activated and are likely to be associated with the neuropathological phenomenon, which ultimately damages neurons [17, 18]. The microglial reaction is a very tightly regulated process which is essential for a precise immune response; excessive microglial activation leads to a continuous release of inflam- matory mediators such as cytokines, chemokines, reactive free radicals, and proteases [19]. This process is referred to as “reactive microgliosis” and involves the proliferation, recruitment, and activation of microglia which is then followed by neuronal damage [20], all of which are secondary to actual neuronal injury. Thus, initial, acute damage from microgliosis may provoke a continuous cycle of events, which then develops into chronic, progressive neurodegeneration which is a common characteristic of Parkinson’s disease [21]. However, the initial stimulus driving excessive inflamma- tion is still unknown. There are several compounds released by damaged neurons, which are able to induce microgliosis and ROS production. These include (i) matrix metallo- proteinase 3 (released by damaged dopaminergic neurons), which induces superoxide production by microglia leading to neuronal death [36]. (ii) Neuromelanin, a neuronal pigment released in PD by dying neurons which is capable of activat- ing microglia [37]. (iii) SYN, a component of LB neurons, typically found in PD that is toxic to neurons but only in the presence of microglia. (iv) Aggregated SYN-activated microglia are toxic to dopaminergic neurons isolated from embryonic mouse brain. Importantly, its toxicity is depen- dant on the presence of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase following ROS formation [38]. Another study has shown that neuroinflammation is accom- panied by dopaminergic loss and aggregation of oxidized SYN in the cytoplasm of SN neurons when human SYN is present in the mouse brain [39]. Taken together, these studies suggest that there is a link between protein aggregation and the production of ROS by activated microglia. Parkinson’s Disease GPi and SNpr are the output nuclei of the basal ganglia, which projects to the thalamus and from there to the cortex. The indirect pathway connects the striatum to output nuclei through external segment of the globus pallidus (GPe) and then subthalamic nucleus (STN). In Parkinson’s disease (PD), the dopaminergic input from SNpc is progressively lost, causing a reduction in the direct pathway signal. Indirect pathway increases its activity through STN in the output nuclei and has inhibitory influence on the thalamus. It leads to a reduction of thalamic glutamateric input on the motor cortex and subsequent reduction in movement, as rigidity and bradykinesia are observed in PD patients. sign in PD [62]. It can also be induced by microglia activated through the NMDA receptor [61]. NMDA recep- tors have been linked with disturbed energy metabolism and glutamate transmission leading to neuronal death, and have therefore been investigated as important therapeutic targets in pharmacological PD research [63]. Accordingly, reducing glutamatergic transmission may lead to an “anti- PD activity”. Indeed, injections of the NMDA antagonist, MK-801, reverses parkinsonian symptoms in MPTP-treated monkeys [64]. Several studies using rodent PD models have shown that glutamate antagonists have both symptomatic and neuroprotective effects in PD [59]. Recently, PD patients treated with memantine, another NMDA receptor antagonist have shown moderate but significant improvements in terms of cognitive symptoms [65]. The use of amantadine as an adjuvant to levodopa has demonstrated beneficial effects on motor response complications [66]. Additional evidence has been reviewed and has demonstrated the potential of NMDA receptor blockade in reversing parkinsonian symptoms [59]. p p g [ ] On the other hand, the excitatory neurotransmitter glutamate plays a critical role in glutamatergic transmission in basal ganglia functions [55]. The action of glutamate on neurons is mediated by ionotropic and metabotropic gluta- mate receptors. Ionotropic N-methyl-D-aspartate (NMDA) receptors are known to mediate excitotoxicity caused by high levels of glutamate and can be found on dopaminergic neurons [56]. Activation of NMDA receptors located on DA neurons leads to neurotoxicity both in vitro and in vivo [57, 58]. The functional organisation of basal ganglia also contributes to the genesis of symptoms observed in move- ment disorder. The striatum (the input nucleus of the basal ganglia circuit) is the main recipient of dopaminergic fibres from the SN. Parkinson’s Disease The reduction in dopaminergic innervations of the striatum and changes in the activity of basal ganglia induces complex changes in the structure and function of basal ganglia NMDA receptor [59]. Glutamatergic excitation is increased and glutamatergic neurons become uninhibited under PD conditions, especially due to the excessive firing from the subthalamic nucleus to the SN [60] (Figure 1). It has been shown that the neurotoxicity of activated microglia is primarily mediated by glutamate released through NMDA receptor signalling [61]. Neuritic beading (a focal bead- like swelling in dendrites and axons) is a neuropathological Parkinson’s Disease Parkinson’s Disease 3 SNpc Cortex Glu Glu SNpr/ GPi GPe STN Thalamus Glu Glu Basal ganglia PD PD PD PD? PD PD Striatum DA neurons Over production of ROS by microglia has been directly linked to neuronal toxicity and death via the nitric oxide (NO) mechanism [40, 41]. NO induces oxidative stress, a major cause of neuronal injury, which is strongly linked to the pathogenesis of PD and physiological aging [42, 43]. For example, NO can react with dopamine to generate quinone products, which are known to have a damaging effect on brain mitochondria [44]. Basal level of lipid peroxidation is increased in the SN of PD patients, suggesting a higher sensitivity of this area to free radicals and ROS [45]. Aging also contributes to microglial “priming”: activated microglia in healthy aged brains release excessive quantities of proin- flammatory cytokines compared to younger individuals [46, 47]. Furthermore, there is an increased probability of developing a neurodegenerative disorder after 60 years of age due to age-related increases in oxidative, metabolic, or inflammatory activation [48]. Basal ganglia Basal ganglia y Inflammatory cytokines (IL-1β, TNF-α, IL-6, and IFN- γ) are also released by activated microglia and amplify the inflammatory response. Excessive production of these cytokines has been reported in the SN of PD patients [49, 50] as well as in cerebrospinal fluid (CSF) and blood compart- ments [51, 52]. Cytokines can stimulate inactivated microglia and also directly bind to receptors on the cellular surface of dopaminergic neurons thereby promoting apoptotic cell death and subsequent phagocytosis of DA neurons [53]. Neurons in the midbrain, unlike those in the hippocampus or cortex, exhibit a greater sensitivity to proinflammatory cytokines. Moreover, this sensitivity has been directly related to a high degree of oxidative processes [19]. In contrast, acti- vated microglia also produce anti-inflammatory cytokines such as TGF-β1, IL-10, and IL-1. These cytokines play a role in the inhibition of the inflammatory response. Importantly, the balance between pro- and anti-inflammatory cytokine production is impaired during neuroinflammation [54]. Figure 1: Basal ganglia motor circuit in Parkinson’s disease: dopaminergic neurons (DA) create a direct pathway between Substantia Nigra pars compacta (SNpc) and striatum—the input nuclei of the basal ganglia. Another direct pathway connects the striatum to the internal segment of globus pallidus (GPi) and the substantia nigra pars reticulata (SNpr). 2. The Role of Neuroinflammation in the Pathogenesis of PD A large number of studies involving cells, animal models, and human patients indicate the involvement of neuroinflamma- tion in the neuropathogenesis of PD. 2.1. In Vitro/In Vivo. To demonstrate the delayed and pro- gressive nature of neuroinflammation observed in PD, lipo- polysaccharide (LPS) was administered to rodents as a single dose or a chronic infusion [22]. While LPS has no direct effect on neurons, it is capable of initiating a chronic inflam- mation and a delayed, secondary progressive degeneration of dopaminergic neurons in the SN [22, 23]. An in vitro study has also shown that 1-methyl-4-phenyl-1,2,3,6-tetrahydro- pyridine (MPTP) can initiate direct neuronal injury in neuron-glia cultures which is then followed by the induction of reactive microgliosis [24]. Furthermore, in a microglia free neuronal-astrocytic coculture, MPTP induced only acute, non-progressive neurotoxicity [21]. MPTP is selectively 3. The Kynurenine Pathway The kynurenine pathway (KP) represents the main catabolic pathway of the essential amino acid tryptophan (TRP), Parkinson’s Disease Tryptophan Kynurenine 3-hydroxykynurenine Nicotinamide Quinolinic acid 3-hydroxyanthranilic acid Kynurenic acid 2-amini-3-carboxymuconate- semialdehyde Picolinic acid 2-amino-3-carboxymuconate- semialdehyde decarboxylase (ACMSD) Anthranilic acid 5-hydroxy-anthranilic acid Indoleamine 2,3- dioxygenase I, II (IDO) Kynurenine 3- hydroxilase (KMO) Kynurenine aminotransferase I, II, III (KAT) Kynureninase Kynureninase 3-hydroxylanthraniate 3,4- dioxygenase (3-HAO) Nonenzymatic Quinolinate phosphoribosyl tran ferase (QPRT) s Figure 2: Simplified diagram of Kynurenine pathway: during neuroinflammation, 95% of the dietary tryptophan is metabolized along the KP within the brain. The remaining 5% serves as a precursor to the synthesis of the neurotransmitter serotonin. IDO catalyses the initial and rate-limiting step in the degradation of tryptophan through KP that ultimately leads to the production of nicotinamide. Tryptophan Tryptophan Indoleamine 2,3- dioxygenase I, II (IDO) Kynurenine aminotransferase I, II, III (KAT) Kynurenine Kynureninase Kynurenine 3- hydroxilase (KM Anthranilic acid 5-hydroxy-anthranilic acid Nicotinamide Quinolinic acid 2-amini-3-carboxymuconate- semialdehyde Picolinic acid 2-amino-3-carboxymuconate- semialdehyde decarboxylase (ACMSD) 3-hydroxylanthraniate 3,4- dioxygenase (3-HAO) Nonenzymatic Quinolinate phosphoribosyl tran ferase (QPRT) s 3-hydroxylanthraniate 3,4- dioxygenase (3-HAO) 2-amino-3-carboxymuconate- semialdehyde decarboxylase (ACMSD) semialdehyde Nonenzymatic Quinolinic acid Quinolinate phosphoribosyl tran ferase (QPRT) s Figure 2: Simplified diagram of Kynurenine pathway: during neuroinflammation, 95% of the dietary tryptophan is metabolized along the KP within the brain. The remaining 5% serves as a precursor to the synthesis of the neurotransmitter serotonin. IDO catalyses the initial and rate-limiting step in the degradation of tryptophan through KP that ultimately leads to the production of nicotinamide. producing excitation and which ultimately causes selective neuronal lesions in the rat brain [80, 81]. Acute QUIN production can lead to human neuronal death and chronic production causes dysfunction by at least six separate mecha- nisms [82, 83]. In pathophysiological concentrations, QUIN activates the NMDA receptor [84]. QUIN also increases glutamate release in neurons and inhibits glutamate uptake and catabolism in astrocytes. QUIN can potentiate its own toxicity and that of other excitotoxins, for example, NMDA and glutamate thus producing progressive mitochondrial dysfunction [85]. Finally, QUIN can increase free radical generation by inducing nitric oxide synthase production (NOS) in astrocytes and neurons which in turn leads to oxidative stress [86, 87]. Within the brain, QUIN is produced by activated microglia and infiltrating macrophages [70]. Neurons and astrocytes do not produce QUIN [88, 89]. 3. The Kynurenine Pathway Recent findings have demonstrated that QUIN excitotoxicity in human astrocytes and neurons is mediated through activation of an NMDA-like receptor [87]. In addition, QUIN-induced damage can be increased in the presence of 3-HK, 6-hydroxidopamine, a specific dopaminergic neuron toxin, or ROS [90–92]. Human glial cells, such as astrocytes and microglia produce most components of the KP [93]. The KP components are also present in macrophages that are capable of penetrating the blood-brain barrier (BBB) in the presence of brain damage or infection [94]. Thus, up-regulation of QUIN production alone or with additional neurotoxic factors during inflammation could easily lead to over activation of the NMDA receptor. This is followed by oxidative stress, which occurs in early PD development. which ultimately leads to the production of the cen- tral metabolic cofactor, nicotinamide adenine dinucleotide (NAD+) (Figure 2). The KP is also one of the major regulatory mechanisms of the immune response [67]. Two nonmutually exclusive theories have been proposed: (1) that TRP degradation suppresses T-cell proliferation by dramatically depleting the supply of this critical amino acid and (2) that various downstream KP metabolites suppress certain immune cells [67]. Induction of the KP by the rate- limiting enzyme, indoleamine 2,3 dioxygenase (IDO1) in dendritic cells completely inhibits clonal expansion of T cells [68]. Moreover, TRP depletion and IDO1/KP activa- tion have been implicated in the facilitation of immune tolerance associated with pregnancy and tumour persistence [69]. The cellular expression of the KP in the brain is only partially understood. It is complete in cells of monocytic lineage, including macrophages and microglia [70], but only partially present in human astrocytes [71], neurons [72], and endothelial cells [73]. The various KP metabolites can have either neurotoxic or neuroprotective effects and occasionally both depending on their concentration. The neurotoxicity of several KP metabolites has been investigated in relation to oxidative stress generation and neuronal death in vitro and in vivo in animal models of neurodegener- ative disorders [74–77]. 3-hydroxykynurenine (3-HK), 3- hydroxyanthranilic acid (3HAA) and 5-hydroxyanthranilic acid (5HAA) are known to induce cell death in cultures of rat neurons [78]. 3-HK is toxic to stratial neuronal cultures, mainly due to its ability to generate ROS and initiate apoptosis [79]. Quinolinic acid (QUIN) however, is likely to be the most important in terms of biological activity. 3. The Kynurenine Pathway QUIN can selectively activate NMDA receptors In contrast to the neurotoxic activity of QUIN, kynurenic acid (KYNA) is a neuroprotective metabolite, antagonising Parkinson’s Disease 5 Iba-1 Merge IFN-γ Figure 3: Activated microglial cells express IFN-γ in Parkinsonism: confocal images of the immunofluorescence of IFN-γ (red) com- bined with microglia cells marker—Iba-1 (green) in the SNpc of a parkinsonian monkey. Scale bar: 35 mm. all ionotropic glutamate receptors (including NMDA) and thus blocks some of the neurotoxic effects of QUIN and other excitotoxins. KYNA is produced from kynurenine by the kynurenine aminotransferase enzymes (KAT) I, KAT II, and KAT III, in astrocytes [71]. Endogenous generation of KYNA in rat brain has been shown to be more effective than KYNA applied exogenously, suggesting the importance of localised KYNA production and physical proximity to NMDA recep- tors [95]. An increase in endogenous KYNA levels can prevent SN dopaminergic loss caused by focal infusion of QUIN or NMDA [96]. Nanomolar concentrations of KYNA significantly reduce glutamate output from striatal neurons in rat brain, similar to the kynurenine hydroxylase (KMO) inhibitors [97]. Both, KYNA and QUIN are produced in the SN or the adjoining striatum region [98, 99]. Based on previous studies, it can be hypothesised that under normal conditions local concentrations of KYNA and QUIN are low and physiologically regulate NMDA receptor function. However, in disease states, where QUIN production is high, it is thought that there is insufficient KYNA concentration to block QUIN production [100]. Figure 3: Activated microglial cells express IFN-γ in Parkinsonism: confocal images of the immunofluorescence of IFN-γ (red) com- bined with microglia cells marker—Iba-1 (green) in the SNpc of a parkinsonian monkey. Scale bar: 35 mm. in the frontal cortex, putamen, and SNpc of PD patients [107]. KYNA, but not the highly selective NMDA antagonist 7-chlorokynurenic acid exhibits partial protection against MPP+ toxin on dopaminergic terminals of rat striatum [110]. However, increased KAT II activity, which is an enzyme responsible for 75% of the KYNA synthesis in the brain, has been found in peripheral red blood cells of PD patients. It is not however found in plasma [111]. The increased KAT II activity correlates with higher blood KYNA concentrations; this elevation may be caused by 3-HK released from the CNS. 3. The Kynurenine Pathway As KYNA has limited abilities to cross the BBB, it has been suggested that peripheral KYNA is likely to be transported to the brain by large neutral amino acid carriers and there it has neuroprotective effects [112]. Another recent study has shown that KYNA is involved in leukocyte recruitment and the investigators hypothesised that KYNA might therefore have an anti-inflammatory action [113]. Based on preclinical and clinical data, KYNA or its analogues are thought to have neuroprotective effects in PD trough binding as antagonists to the NMDA receptor. This in turn causes slow neuronal excitotoxic damage [114]. Picolinic acid is another endogenous neuroprotective compound [101] and is also the main metal chelator in the brain [102]. Previously, we have shown that it is produced in micromolar concentrations by human primary neurons [72]. PD is associated with neuropathological features such as protein aggregation and oxidative stress associated with the involvement of metal ions [103]. Therefore, use of chelating agents has also been suggested as a form of therapy for PD. The KP, under normal physiological conditions is well balanced and produces all KP intermediates leading ulti- mately to NAD+ production. However, under pathologic conditions, IDO1 is activated and astrocytes produce kynure- nine (KYN) and KYNA, [104], neurons produce PIC [88] and activated microglia/infiltrating macrophages produce QUIN [89]. It is important to note that PIC and KYNA can partly antagonise the neurotoxic effects of QUIN [105]. However, astroglial secretion of large quantities of KYN can lead to further synthesis of QUIN by microglia, suggesting that the cerebral synthesis of QUIN largely overtakes the neuroprotective effects of PIC and KYNA [106]. Unpublished data from our group shows an increase in the production of IFN-γ by microglia in the SN of MPTP- treated macaques’ brain (Figure 3). This is of particular significance, as IFN-γ is also a potent inducer of the KP [115]. In the same study, we have also shown that QUIN is produced and accumulated by activated microglia. These microglia colocalise with dopaminergic neurons in the SN of MPTP-treated macaques. Several other studies have shown extensive evidence of activated microglial cells and NMDAR+ dopaminergic neurons in the SNpc. This suggests that the NMDA receptor is likely to be activated by endogenous QUIN released by microglia and followed closely by glutamate [116, 117] (Figure 4). 4. Evidence for the Involvement of the KP in PD Impaired KP metabolism and altered KYNA levels have been previously reported in the brain of PD patients. This occurs when the KYNA/TRP ratio in serum and cerebrospinal fluid (CSF) is significantly increased together with 3-HK levels, a neurotoxic compound that contributes to oxidative damage in the putamen and SNpc [107, 108]. These findings suggest that endogenous KYNA concentrations are decreased and unable to effectively block NMDA receptor and prevent neurotoxicity induced by 3-HK. KAT I expression, the KP enzyme which leads to KYNA formation, is decreased in the SNpc of MPTP-treated mice [109]. KAT-I immunoreac- tivity in dopaminergic neurons and surrounding microglia has been linked to increased vulnerability of SN neurons to toxicity. Lowered KYNA concentrations have also been found 5. Recent KP Inhibitors for the Treatment of PD Several drugs that block the KP are currently under thera- peutic investigation both in our laboratory and by other investigators. For example, 4-chlorokynurenine crosses the BBB and blocks QUIN toxicity at the glycine site on NMDA receptors [118]. Kynurenic acid analogues are currently due to enter clinical trials for the treatment of epilepsy, 6 Parkinson’s Disease 6 Activated microglia cells TRP KYN Intermediates QUIN INF-γ Neurotoxicity Dopaminergic neurons TRP KYN KAT 2,3 KYNA QUIN Neuroprotection TRP Glutamate GluT Astrocytes NMDAR IDO IDO Figure 4: Model for Kynurenine pathway interactions between astrocytes, neurons, and microglia during brain inflammation. Abbreviations: TRP: tryptophan; IDO: Indoleamine 2,3-dioxygenase; KYN: kynurenine: QUIN: quinolinic acid; NMDAR: NMDA receptor; KAT: Kynurenine aminotransferase; GluT: glutamate transporter. Activated microglia cells TRP KYN Intermediates QUIN INF-γ Neurotoxicity Dopaminergic neurons TRP KYN KAT 2,3 KYNA QUIN Neuroprotection TRP Glutamate GluT Astrocytes NMDAR IDO IDO Astrocytes Activated microglia cells Figure 4: Model for Kynurenine pathway interactions between astrocytes, neurons, and microglia during brain inflammation. Abbreviations: TRP: tryptophan; IDO: Indoleamine 2,3-dioxygenase; KYN: kynurenine: QUIN: quinolinic acid; NMDAR: NMDA receptor; KAT: Kynurenine aminotransferase; GluT: glutamate transporter. Resting microglia Rotenone LPS Activated microglia DA neuron Cytokines free radicals Injured DA neuron Injury signal Stimuli: α-synuclein Neuromelanin MMP3 NMDAR QUIN ROS Neuronal death PD progression Reactive microgliosis auto-amplifying loop IDO PIC 3HK Oxidative stress ACMSD INFγ Figure 5: The possible role of Kynurenine pathway involvement in dopaminergic neurodegenerative process through microglia activation: Parkinson’s disease is associated with chronic activation of microglia, which also can be induced by LPS or Rotenone treatments. Classic microglia activation release neurotoxic substances including reactive oxygen species (ROS) and proinflammatory cytokines as INF-γ, potent activator of Kynurenine pathway (KP). KP in activated microglia leads to upregulation of 3HK and QUIN. 3HK is toxic primarily as a result of conversion to ROS. The combined effects of ROS and NMDA receptor-mediated excitotoxicity by QUIN contribute to the dysfunction of neurons and their death. However, picolinic acid (PIC) produced through KP activation in neurons, has the ability to protect neurons against QUIN-induced neurotoxicity, being NMDA agonist. Microglia can become overactivated, by proinflammatory mediators and stimuli from dying neurons and cause perpetuating cycle of further microglia activation microgliosis. Excessive microgliosis will cause neurotoxicity to neighbouring neurons and resulting in neuronal death, contributing to progression of Parkinson’s disease. References stroke, and possibly PD as potential neuroprotective agents [119]. Two KP analogues are at present under investi- gation in a phase III clinical trial. These are Terifluno- mide (Sanofi-Aventis) and Laquinimod (Teva Neuroscience) [120]. Recently, one KP analogue reached the Japanese mar- ket as a potent immunomodulatory drug for the treatment of arthritis, asthma, and dermatitis [120]: Tranilast/Rizaben (Kissei Pharmaceutics) is an anthranilic acid derivative and it has been proposed as a treatment for autoimmune diseases such as Multiple Sclerosis [121]. Finally, 8-OH Quinoli- none metal attenuating compounds—Clioquinol and PBT2 (Prana) rapidly decrease soluble brain amyloid-beta and improve cognitive performance [122]. Interestingly, these 2 compounds share close structural similarity and similar biochemical properties with KYNA and QUIN. [1] C. M. Tanner, “Epidemiology of Parkinson’s disease,” Neuro- logic Clinics, vol. 10, no. 2, pp. 317–329, 1992. [2] J. Jankovic, “Parkinson’s disease: clinical features and diag- nosis,” Journal of Neurology, Neurosurgery and Psychiatry, vol. 79, no. 4, pp. 368–376, 2008. [3] D. J. Gelb, E. Oliver, and S. Gilman, “Diagnostic criteria for Parkinson disease,” Archives of Neurology, vol. 56, no. 1, pp. 33–39, 1999. [4] E. Bezard, S. Dovero, C. Prunier et al., “Relationship between the appearance of symptoms and the level of nigrostriatal degeneration in a progressive 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine-lesioned macaque model of Parkinson’s disease,” Journal of Neuroscience, vol. 21, no. 17, pp. 6853– 6861, 2001. [5] T. Wichmann and M. R. DeLong, “Functional neuroanatomy of the basal ganglia in Parkinson’s disease,” Advances in neurology, vol. 91, pp. 9–18, 2003. Conjugates of KYNA analogues with D-glucose or D- galactose increase its ability to cross the BBB and prevent excitotoxicity and seizures in an animal model [123]. Kynurenine 3-hydroxylase inhibitors significantly reduce the severity of dystonia in hamsters and may therefore be a potential candidate for managing dyskinesia associated with striatal dysfunction [124]. There is also an increasing interest in the use of pharmacological modulation of the KP in treating numerous disorders like AIDS-dementia and many other neurodegenerative diseases, diabetes, depression, infections, tumour development, glaucoma, and cataract formation [116]. [6] M. G. Spillantini, R. A. Crowther, R. Jakes, M. Hasegawa, and M. Goedert, “α-synuclein in filamentous inclusions of Lewy bodies from Parkinson’s disease and dementia with Lewy bodies,” Proceedings of the National Academy of Sciences of the United States of America, vol. 95, no. 11, pp. 6469–6473, 1998. [7] J. L. Montastruc, O. Rascol, and J. M. References Senard, “Current status of dopamine agonists in Parkinson’s disease management,” Drugs, vol. 46, no. 3, pp. 384–393, 1993. [8] A. H. V. Schapira, “Molecular and clinical pathways to neu- roprotection of dopaminergic drugs in Parkinson disease,” Neurology, vol. 72, no. 7, pp. S44–S50, 2009. [9] W. Dauer and S. Przedborski, “Parkinson’s disease: mecha- nisms and models,” Neuron, vol. 39, no. 6, pp. 889–909, 2003. 5. Recent KP Inhibitors for the Treatment of PD Resting microglia Rotenone LPS Activated microglia DA neuron Cytokines free radicals Injured DA neuron Injury signal Stimuli: α-synuclein Neuromelanin MMP3 NMDAR QUIN ROS Neuronal death PD progression Reactive microgliosis auto-amplifying loop IDO PIC 3HK Oxidative stress ACMSD INFγ Resting microglia Reactive microgliosis auto-amplifying loop microglia Reactive microgliosis auto-amplifying loop DA neuron Activated microglia Rotenone LPS Injury signal NMDAR Stimuli: α-synuclein Neuromelanin MMP3 Injured DA neuron Figure 5: The possible role of Kynurenine pathway involvement in dopaminergic neurodegenerative process through microglia activation: Parkinson’s disease is associated with chronic activation of microglia, which also can be induced by LPS or Rotenone treatments. Classic microglia activation release neurotoxic substances including reactive oxygen species (ROS) and proinflammatory cytokines as INF-γ, potent activator of Kynurenine pathway (KP). KP in activated microglia leads to upregulation of 3HK and QUIN. 3HK is toxic primarily as a result of conversion to ROS. The combined effects of ROS and NMDA receptor-mediated excitotoxicity by QUIN contribute to the dysfunction of neurons and their death. However, picolinic acid (PIC) produced through KP activation in neurons, has the ability to protect neurons against QUIN-induced neurotoxicity, being NMDA agonist. Microglia can become overactivated, by proinflammatory mediators and stimuli from dying neurons and cause perpetuating cycle of further microglia activation microgliosis. Excessive microgliosis will cause neurotoxicity to neighbouring neurons and resulting in neuronal death, contributing to progression of Parkinson’s disease. 7 Parkinson’s Disease References 6. Conclusions [10] S. Przedborski, V. Jackson-Lewis, M. Vila et al., “Free radical and nitric oxide toxicity in Parkinson’s disease,” Advances in neurology, vol. 91, pp. 83–94, 2003. PD seems to be associated with an imbalance between the two main branches of the KP within the brain. KYNA synthesis by astrocytes is decreased and concomitantly, QUIN production by microglia is increased (Figure 5). There are many therapeutic opportunities for intervention and modification of an impaired KP that may prevent the pro- gression of neurodegenerative disorders such as PD. Using specific KP enzyme inhibitors, it may be possible to reinstate a physiologically normal KP, which is neuroprotective. This neuroprotective state might also be synergistically improved by concomitantly blocking the NMDA receptor using its antagonists, such as memantine or MK801. Additionally, neuroprotection may be achieved by designing KYNA analogues that are able to penetrate the BBB and deliver neuroprotective compounds to brain pools thus reducing hyperactivation of glutamatergic receptors. [11] G. W. Kreutzberg, “Microglia: a sensor for pathological events in the CNS,” Trends in Neurosciences, vol. 19, no. 8, pp. 312–318, 1996. [12] P. L. McGeer, S. Itagaki, B. E. Boyes, and E. G. McGeer, “Reac- tive microglia are positive for HLA-DR in the substantia nigra of Parkinson’s and Alzheimer’s disease brains,” Neurology, vol. 38, no. 8, pp. 1285–1291, 1988. [13] A. Nimmerjahn, F. Kirchhoff, and F. Helmchen, “Neuro- science: resting microglial cells are highly dynamic surveil- lants of brain parenchyma in vivo,” Science, vol. 308, no. 5726, pp. 1314–1318, 2005. [14] B. Liu and J. S. Hong, “Role of microglia in inflammation- mediated neurodegenerative diseases: mechanisms and strategies for therapeutic intervention,” Journal of Pharma- cology and Experimental Therapeutics, vol. 304, no. 1, pp. 1–7, 2003. [15] J. Gehrmann, Y. Matsumoto, and G. W. Kreutzberg, “Microglia: intrinsic immuneffector cell of the brain,” Brain Research Reviews, vol. 20, no. 3, pp. 269–287, 1995. Acknowledgments The authors thank Nady Braidy, Seray Adams, and Louise Pemberton for their critical reading and comments on this paper. They also thank Parkinson NSW for supporting their research on PD. This work was also supported by grants from: the Spanish Ministry of Science and Innovation (FIS PI10 02827, SAF2010-2127, and RYC-2010-06729), Fundaci´on S´eneca (15329/PI/10), CIBERNED (Centro de Investigaci´on Biom´edica en Red sobre Enfermedades Neu- rodegenerativas). [16] G. M. Hayes, M. N. Woodroofe, and M. L. Cuzner, “Microglia express MHC class II in normal and demyelinating human white matter,” Annals of the New York Academy of Sciences, vol. 540, pp. 501–503, 1988. [17] K. Imamura, N. Hishikawa, M. Sawada, T. Nagatsu, M. Yos- hida, and Y. Hashizume, “Distribution of major histocom- patibility complex class II-positive microglia and cytokine profile of Parkinson’s disease brains,” Acta Neuropathologica, vol. 106, no. 6, pp. 518–526, 2003. 8 Parkinson’s Disease [34] J. W. Langston, L. S. Forno, J. Tetrud, A. G. Reeves, J. A. Kaplan, and D. Karluk, “Evidence of active nerve cell de- generation in the substantia nigra of humans years after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine exposure,” Annals of Neurology, vol. 46, no. 4, pp. 598–605, 1999. [18] W. F. Hickey and H. Kimura, “Perivascular microglial cells of the CNS are bone marrow-derived and present antigen in vivo,” Science, vol. 239, no. 4837, pp. 290–292, 1988. [19] M. L. Block, L. Zecca, and J. S. Hong, “Microglia- mediated neurotoxicity: uncovering the molecular mecha- nisms,” Nature Reviews Neuroscience, vol. 8, no. 1, pp. 57–69, 2007. [35] C. Barcia, A. S´anchez Bahillo, E. Fern´andez-Villalba et al., “Evidence of active microglia in substantia nigra pars compacta of parkinsonian monkeys 1 year after MPTP exposure,” GLIA, vol. 46, no. 4, pp. 402–409, 2004. [20] W. J. Streit, S. A. Walter, and N. A. Pennell, “Reactive micro- gliosis,” Progress in Neurobiology, vol. 57, no. 6, pp. 563–581, 1999. [36] Y. S. Kim, D. H. Choi, M. L. Block et al., “A pivotal role of matrix metalloproteinase-3 activity in dopaminergic neuronal degeneration via microglial activation,” The FASEB Journal, vol. 21, no. 1, pp. 179–187, 2007. [21] H. M. Gao and J. S. Hong, “Why neurodegenerative diseases are progressive: uncontrolled inflammation drives disease progression,” Trends in Immunology, vol. 29, no. 8, pp. 357– 365, 2008. [37] L. Zecca, F. A. Zucca, H. Wilms, and D. Sulzer, “Neuromelanin of the substantia nigra: a neuronal black hole with protective and toxic characteristics,” Trends in Neurosciences, vol. Acknowledgments 26, no. 11, pp. 578–580, 2003. [22] A. Casta˜no, A. J. Herrera, J. Cano, and A. Machado, “Lipopolysaccharide intranigral injection induces inflam- matory reaction and damage in nigrostriatal dopaminergic system,” Journal of Neurochemistry, vol. 70, no. 4, pp. 1584– 1592, 1998. [38] W. Zhang, T. Wang, Z. Pei et al., “Aggregated α-synuclein activates microglia: a process leading to disease progression in Parkinson’s disease,” The FASEB Journal, vol. 19, no. 6, pp. 533–542, 2005. [23] H. M. Gao, J. Jiang, B. Wilson, W. Zhang, J. S. Hong, and B. Liu, “Microglial activation-mediated delayed and progressive degeneration of rat nigral dopaminergic neurons: relevance to Parkinson’s disease,” Journal of Neurochemistry, vol. 81, no. 6, pp. 1285–1297, 2002. [39] H. M. Gao, P. T. Kotzbauer, K. Uryu, S. Leight, J. Q. Trojanowski, and V. M. Y. Lee, “Neuroinflammation and oxidation/nitration of α-synuclein linked to dopaminergic neurodegeneration,” Journal of Neuroscience, vol. 28, no. 30, pp. 7687–7698, 2008. [24] H. M. Gao, B. Liu, W. Zhang, and J. S. Hong, “Critical role of microglial NADPH oxidase-derived free radicals in the in vitro MPTP model of Parkinson’s disease,” The FASEB Journal, vol. 17, pp. 1954–1956, 2003. [40] Y. S. Kim and T. H. Joh, “Microglia, major player in the brain inflammation: their roles in the pathogenesis of Parkinson’s disease,” Experimental and Molecular Medicine, vol. 38, no. 4, pp. 333–347, 2006. [25] M. Gerlach, P. Riederer, H. Przuntek, and M. B. H. Youdim, “MPTP mechanisms of neurotoxicity and their implications for Parkinson’s disease,” European Journal of Pharmacology, vol. 208, no. 4, pp. 273–286, 1991. [41] C. C. Chao, S. Hu, T. W. Molitor, E. G. Shaskan, and P. K. Peterson, “Activated microglia mediate neuronal cell injury via a nitric oxide mechanism,” Journal of Immunology, vol. 149, no. 8, pp. 2736–2741, 1992. [26] D. C. Wu, V. Jackson-Lewis, M. Vila et al., “Blockade of microglial activation is neuroprotective in the 1-methyl-4- phenyl-1,2,3,6-tetrahydropyridine mouse model of Parkin- son disease,” Journal of Neuroscience, vol. 22, no. 5, pp. 1763– 1771, 2002. [42] P. Jenner, “Oxidative stress in Parkinson’s disease,” Annals of Neurology, vol. 53, no. 3, pp. S26–S38, 2003. [43] E. Koutsilieri, C. Scheller, E. Gr¨unblatt, K. Nara, J. Li, and P. Riederer, “Free radicals in Parkinson’s disease,” Journal of Neurology, vol. 249, supplement 2, pp. II1–II5, 2002. [27] B. Liu, L. Du, and J. S. Acknowledgments Hong, “Naloxone protects rat dopaminergic neurons against inflammatory damage through inhibition of microglia activation and superoxide generation,” Journal of Pharmacology and Experimental Ther- apeutics, vol. 293, no. 2, pp. 607–617, 2000. [44] S. Jana, A. K. Maiti, M. B. Bagh et al., “Dopamine but not 3,4-dihydroxy phenylacetic acid (DOPAC) inhibits brain respiratory chain activity by autoxidation and mitochondria catalyzed oxidation to quinone products: implications in Parkinson’s disease,” Brain Research, vol. 1139, no. 1, pp. 195–200, 2007. [28] H. Chen, S. M. Zhang, M. A. Hern´an et al., “Nonsteroidal anti-inflammatory drugs and the risk of Parkinson disease,” Archives of Neurology, vol. 60, no. 8, pp. 1059–1064, 2003. [45] D. T. Dexter, C. J. Carter, F. R. Wells et al., “Basal lipid peroxi- dation in substantia nigra is increased in Parkinson’s disease,” Journal of Neurochemistry, vol. 52, no. 2, pp. 381–389, 1989. [29] H. Chen, E. Jacobs, M. A. Schwarzschild et al., “Nonsteroidal antiinflammatory drug use and the risk for Parkinson’s disease,” Annals of Neurology, vol. 58, no. 6, pp. 963–967, 2005. [46] R. N. Dilger and R. W. Johnson, “Aging, microglial cell priming, and the discordant central inflammatory response to signals from the peripheral immune system,” Journal of Leukocyte Biology, vol. 84, no. 4, pp. 932–939, 2008. [30] H. Chen, E. J. O’Reilly, M. A. Schwarzschild, and A. Ascherio, “Peripheral inflammatory biomarkers and risk of Parkinson’s disease,” American Journal of Epidemiology, vol. 167, no. 1, pp. 90–95, 2008. [47] Y. Huang, C. J. Henry, R. Dantzer, R. W. Johnson, and J. P. Godbout, “Exaggerated sickness behavior and brain proinflammatory cytokine expression in aged mice in response to intracerebroventricular lipopolysaccharide,” Neurobiology of Aging, vol. 29, no. 11, pp. 1744–1753, 2008. [31] A. Gerhard, N. Pavese, G. Hotton et al., “In vivo imaging of microglial activation with [11C](R)-PK11195 PET in idiopathic Parkinson’s disease,” Neurobiology of Disease, vol. 21, no. 2, pp. 404–412, 2006. [32] P. L. McGeer and E. G. McGeer, “Glial reactions in Parkin- son’s disease,” Movement Disorders, vol. 23, no. 4, pp. 474– 483, 2008. [48] R. von Bernhardi, J. E. Tichauer, and J. Eugen´ın, “Aging- dependent changes of microglial cells and their relevance for neurodegenerative disorders,” Journal of Neurochemistry, vol. 112, no. 5, pp. 1099–1114, 2010. [33] P. L. McGeer, C. Schwab, A. Parent, and D. Doudet, “Presence of reactive microglia in monkey substantia nigra years after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine administra- tion,” Annals of Neurology, vol. 54, no. 5, pp. 599–604, 2003. Acknowledgments [49] J. E. Merrill and E. N. Benveniste, “Cytokines in inflammatory brain lesions: helpful and harmful,” Trends in Neurosciences, vol. 19, no. 8, pp. 331–338, 1996. Parkinson’s Disease 9 [50] T. Nagatsu, M. Mogi, H. Ichinose, and A. Togari, “Cytokines in Parkinson’s disease,” Journal of Neural Transmission. Supplementa, no. 58, pp. 143–151, 2000. [67] J. R. Moffett and M. A. Namboodiri, “Tryptophan and the immune response,” Immunology and Cell Biology, vol. 81, no. 4, pp. 247–265, 2003. [51] M. Mogi, M. Harada, P. Riederer, H. Narabayashi, K. Fujita, and T. Nagatsu, “Tumor necrosis factor-α (TNF-α) increases both in the brain and in the cerebrospinal fluid from parkinsonian patients,” Neuroscience Letters, vol. 165, no. 1-2, pp. 208–210, 1994. [68] A. L. Mellor, B. Baban, P. Chandler et al., “Cutting edge: induced indoleamine 2,3 dioxygenase expression in dendritic cell subsets suppresses T cell clonal expansion,” Journal of Immunology, vol. 171, no. 4, pp. 1652–1655, 2003. [69] D. H. Munn and A. L. Mellor, “IDO and tolerance to tumors,” Trends in Molecular Medicine, vol. 10, no. 1, pp. 15–18, 2004. [52] G. Stypuła, J. Kunert-Radek, H. Stepien, K. Zyli´nska, and M. Pawlikowski, “Evaluation of interleukins, ACTH, cortisol and prolactin concentrations in the blood of patients with Parkinson’s disease,” NeuroImmunoModulation, vol. 3, no. 2-3, pp. 131–134, 1996. [70] G. J. Guillemin, D. G. Smith, G. A. Smythe, P. J. Armati, and B. J. Brew, “Expression of the kynurenine pathway enzymes in human microglia and macrophages,” Advances in Experi- mental Medicine and Biology, vol. 527, pp. 105–112, 2003. [53] E. C. Hirsch, “Glial cells and Parkinson’s disease,” Journal of Neurology, vol. 247, supplement 2, no. 2, pp. 58–62, 2000. [71] G. J. Guillemin, S. J. Kerr, G. A. Smythe et al., “Kynurenine pathway metabolism in human astrocytes: a paradox for neuronal protection,” Journal of Neurochemistry, vol. 78, no. 4, pp. 842–853, 2001. [54] T. Nagatsu and M. Sawada, “Inflammatory process in Parkin- son’s disease: role for cytokines,” Current Pharmaceutical Design, vol. 11, no. 8, pp. 999–1016, 2005. [72] G. J. Guillemin, K. M. Cullen, C. K. Lim et al., “Characteriza- tion of the kynurenine pathway in human neurons,” Journal of Neuroscience, vol. 27, no. 47, pp. 12884–12892, 2007. [55] P. Ravenscroft and J. Brotchie, “NMDA receptors in the basal ganglia,” Journal of Anatomy, vol. 196, no. 4, pp. 577–585, 2000. [73] R. Owe-Young, N. L. Webster, M. Acknowledgments Mukhtar et al., “Kynu- renine pathway metabolism in human blood-brain-barrier cells: implications for immune tolerance and neurotoxicity,” Journal of Neurochemistry, vol. 105, no. 4, pp. 1346–1357, 2008. [56] E. A. Waxman and D. R. Lynch, “N-methyl-D-aspartate receptor subtype mediated bidirectional control of p38 mitogen-activated protein kinase,” Journal of Biological Chemistry, vol. 280, no. 32, pp. 29322–29333, 2005. [57] S. Kikuchi and S. U. Kim, “Glutamate neurotoxicity in mesencephalic dopaminergic neurons in culture,” Journal of Neuroscience Research, vol. 36, no. 5, pp. 558–569, 1993. [74] H. Q. Wu, P. Guidetti, J. H. Goodman et al., “Kynurenergic manipulations influence excitatory synaptic function and excitotoxic vulnerability in the rat hippocampus in vivo,” Neuroscience, vol. 97, no. 2, pp. 243–251, 2000. [58] B. P. Connop, R. J. Boegman, K. Jhamandas, and R. J. Benin- ger, “Excitotoxic action of NMDA agonists on nigrostriatal dopaminergic neurons: modulation by inhibition of nitric oxide synthesis,” Brain Research, vol. 676, no. 1, pp. 124–132, 1995. [75] A. Chiarugi, E. Meli, and F. Moroni, “Similarities and differences in the neuronal death processes activated by 3OH-kynurenine and quinolinic acid,” Journal of Neuro- chemistry, vol. 77, no. 5, pp. 1310–1318, 2001. [59] P. J. Hallett and D. G. Standaert, “Rationale for and use of NMDA receptor antagonists in Parkinson’s disease,” Pharma- cology & Therapeutics, vol. 102, no. 2, pp. 155–174, 2004. [76] R. Schwarcz and R. Pellicciari, “Manipulation of brain kynurenines: glial targets, neuronal effects, and clinical opportunities,” Journal of Pharmacology and Experimental Therapeutics, vol. 303, no. 1, pp. 1–10, 2002. [60] W. J. Schmidt, M. Bubser, and W. Hauber, “Behavioural pharmacology of glutamate in the basal ganglia,” Journal of Neural Transmission. Supplementa, no. 38, pp. 65–89, 1992. [77] A. J. Smith, T. W. Stone, and R. A. Smith, “Neurotoxicity of tryptophan metabolites,” Biochemical Society Transactions, vol. 35, no. 5, pp. 1287–1289, 2007. [61] H. Takeuchi, T. Mizuno, G. Zhang et al., “Neuritic beading induced by activated microglia is an early feature of neuronal dysfunction toward neuronal death by inhibition of mitochondrial respiration and axonal transport,” Journal of Biological Chemistry, vol. 280, no. 11, pp. 10444–10454, 2005. [78] A. J. Smith, R. A. Smith, and T. W. Stone, “5-hydro- xyanthranilic acid, a tryptophan metabolite, generates oxi- dative stress and neuronal death via p38 activation in cul- tured cerebellar granule neurones,” Neurotoxicity Re- search, vol. 15, no. 4, pp. 303–310, 2009. [62] P. M. Mattila, J. O. Rinne, H. Helenius, and M. Acknowledgments Rigor, “Neurotoxicity of qui- nolinic acid and its derivatives in hypoxic rat hippocampal slices,” Brain Research, vol. 568, no. 1-2, pp. 199–204, 1991. [85] Y. M. Bordelon, M. F. Chesselet, D. Nelson, F. Welsh, and M. Ereci´nska, “Energetic dysfunction in quinolinic acid-lesio- ned rat striatum,” Journal of Neurochemistry, vol. 69, no. 4, pp. 1629–1639, 1997. [101] K. Jhamandas, R. J. Boegman, R. J. Beninger, and M. Bialik, “Quinolinate-induced cortical cholinergic damage: modulation by tryptophan metabolites,” Brain Research, vol. 529, no. 1-2, pp. 185–191, 1990. [86] H. Baran, B. Kepplinger, M. Herrera-Marschitz, K. Stolze, G. Lubec, and H. Nohl, “Increased kynurenic acid in the brain after neonatal asphyxia,” Life Sciences, vol. 69, no. 11, pp. 1249–1256, 2001. [102] U. Testa, F. Louache, M. Titeux, P. Thomopoulos, and H. Rochant, “The iron-chelating agent picolinic acid enhances transferrin receptors expression in human erythroleukaemic cell lines,” British Journal of Haematology, vol. 60, no. 3, pp. 491–502, 1985. [87] N. Braidy, R. Grant, S. Adams, B. J. Brew, and G. J. Guillemin, “Mechanism for quinolinic acid cytotoxicity in human astrocytes and neurons,” Neurotoxicity Research, vol. 16, no. 1, pp. 77–86, 2009. [103] F. Molina-Holgado, R. C. Hider, A. Gaeta, R. Williams, and P. Francis, “Metals ions and neurodegeneration,” BioMetals, vol. 20, no. 3-4, pp. 639–654, 2007. [88] G. J. Guillemin, K. M. Cullen, C. K. Lim et al., “Characteriza- tion of the kynurenine pathway in human neurons,” Journal of Neuroscience, vol. 27, no. 47, pp. 12884–12892, 2007. [104] G. J. Guillemin, S. J. Kerr, L. A. Pemberton et al., “IFN- β induces kynurenine pathway metabolism in human macrophages: potential implications for multiple sclerosis treatment,” Journal of Interferon and Cytokine Research, vol. 21, no. 12, pp. 1097–1101, 2001. [89] G. J. Guillemin, G. Smythe, O. Takikawa, and B. J. Brew, “Expression of indoleamine 2,3-dioxygenase and production of quinolinic acid by human microglia, astrocytes, and neurons,” GLIA, vol. 49, no. 1, pp. 15–23, 2005. [105] R. J. Beninger, A. M. Colton, J. L. Ingles, K. Jhamandas, and R. J. Boegman, “Picolinic acid blocks the neurotoxic but not the neuroexcitant properties of quinolinic acid in the rat brain: evidence from turning behaviour and tyrosine hydroxylase immunohistochemistry,” Neuroscience, vol. 61, no. 3, pp. 603–612, 1994. [90] P. Guidetti and R. Schwarcz, “3-hydroxykynurenine potentiates quinolinate but not NMDA toxicity in the rat striatum,” European Journal of Neuroscience, vol. 11, no. 11, pp. 3857–3863, 1999. [91] W. M. H. Acknowledgments Behan and T. W. Stone, “Enhanced neuronal damage by co-administration of quinolinic acid and free radicals, and protection by adenosine A2A receptor anta- gonists,” British Journal of Pharmacology, vol. 135, no. 6, pp. 1435–1442, 2002. [106] R. Owe-Young, N. L. Webster, M. Mukhtar et al., “Kynu- renine pathway metabolism in human blood-brain-barrier cells: implications for immune tolerance and neurotoxicity,” Journal of Neurochemistry, vol. 105, no. 4, pp. 1346–1357, 2008. [92] I. Ghorayeb, Z. Puschban, P. O. Fernagut et al., “Simulta- neous intrastriatal 6-hydroxydopamine and quinolinic acid injection: a model of early-stage striatonigral degeneration,” Experimental Neurology, vol. 167, no. 1, pp. 133–147, 2001. [107] T. Ogawa, W. R. Matson, M. F. Beal et al., “Kynurenine path- way abnormalities in Parkinson’s disease,” Neurology, vol. 42, no. 9, pp. 1702–1706, 1992. [108] M. Flint Beal, W. R. Matson, E. Storey et al., “Kynurenic acid concentrations are reduced in Huntington’s disease cerebral cortex,” Journal of the Neurological Sciences, vol. 108, no. 1, pp. 80–87, 1992. [93] M. G. Espey, O. N. Chernyshev, J. F. Reinhard, M. A. A. Namboodiri, and C. A. Colton, “Activated human microglia produce the excitotoxin quinolinic acid,” NeuroReport, vol. 8, no. 2, pp. 431–434, 1997. [109] E. Knyih´ar-Csillik, B. Csillik, M. P´ak´aski et al., “Decreased expression of kynurenine aminotransferase-I (KAT-I) in the substantia nigra of mice after 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine (MPTP) treatment,” Neuroscience, vol. 126, no. 4, pp. 899–914, 2004. [94] M. P. Heyes, K. Saito, and S. P. Markey, “Human macro- phages convert L-tryptophan into the neurotoxin quinolinic acid,” Biochemical Journal, vol. 283, no. 3, pp. 633–635, 1992. [95] H. E. Scharfman, P. S. Hodgkins, S. C. Lee, and R. Schwarcz, “Quantitative differences in the effects of de novo produced and exogenous kynurenic acid in rat brain slices,” Neuroscience Letters, vol. 274, no. 2, pp. 111–114, 1999. [110] M. Merino, M. L. Vizuete, J. Cano, and A. Machado, “The non-NMDA glutamate receptor antagonists 6-cyano-7- nitroquinoxaline-2,3-dione and 2,3-dihydroxy-6-nitro-7- sulfamoylbenzo(f )quinoxaline, but not NMDA antagonists, block the intrastriatal neurotoxic effect of MPP+ ,” Journal of Neurochemistry, vol. 73, no. 2, pp. 750–757, 1999. [96] A. F. Miranda, R. J. Boegman, R. J. Beninger, and K. Jhamandas, “Protection against quinolinic acid-mediated excitotoxicity in nigrostriatal dopaminergic neurons by endogenous kynurenic acid,” Neuroscience, vol. 78, no. 4, pp. 967–975, 1997. [111] Z. Hartai, P. Klivenyi, T. Janaky, B. Penke, L. Dux, and L. Acknowledgments R¨oytt¨a, “Neuritic degeneration in the hippocampus and amygdala in Parkinson’s disease in relation to Alzheimer pathology,” Acta Neuropathologica, vol. 98, no. 2, pp. 157–164, 1999. [79] S. Okuda, N. Nishiyama, H. Saito, and H. Katsuki, “3-hydro- xykynurenine, an endogenous oxidative stress generator, causes neuronal cell death with apoptotic features and region selectivity,” Journal of Neurochemistry, vol. 70, no. 1, pp. 299–307, 1998. [63] C. Ikonomidou and L. Turski, “Neurodegenerative disorders: clues from glutamate and energy metabolism,” Critical Reviews in Neurobiology, vol. 10, no. 2, pp. 239–263, 1996. [80] T. W. Stone and M. N. Perkins, “Quinolinic acid: a potent endogenous excitant at amino acid receptors in CNS,” Euro- pean Journal of Pharmacology, vol. 72, no. 4, pp. 411–412, 1981. [64] J. T. Greenamyre and C. F. O’Brien, “N-methyl-D-aspartate antagonists in the treatment of Parkinson’s disease,” Archives of Neurology, vol. 48, no. 9, pp. 977–981, 1991. [65] D. Aarsland, C. Ballard, Z. Walker et al., “Memantine in patients with Parkinson’s disease dementia or dementia with Lewy bodies: a double-blind, placebo-controlled, multicentre trial,” The Lancet Neurology, vol. 8, no. 7, pp. 613–618, 2009. [81] R. Schwarcz, W. O. Whetsell, and R. M. Mangano, “Quino- linic acid: an endogenous metabolite that produces axon- sparing lesions in rat brain,” Science, vol. 219, no. 4582, pp. 316–318, 1983. [66] L. Verhagen Metman, P. Del Dotto, P. van den Munckhof, J. Fang, M. M. Mouradian, and T. N. Chase, “Amantadine as treatment for dyskinesias and motor fluctuations in Parkin- son’s disease,” Neurology, vol. 50, no. 5, pp. 1323–1326, 1998. [82] G. J. Guillemin and B. J. Brew, “Implications of the kynu- renine pathway and quinolinic acid in Alzheimer’s disease,” Redox Report, vol. 7, no. 4, pp. 199–206, 2002. 10 Parkinson’s Disease [99] R. Schwarcz, F. Du, W. Schmidt et al., “Kynurenic acid: a potential pathogen in brain disorders,” Annals of the New York Academy of Sciences, vol. 648, pp. 140–153, 1992. [83] G. J. Guillemin, L. Wang, and B. J. Brew, “Quinolinic acid selectively induces apoptosis of human astocytes: poten- tial role in AIDS dementia complex,” Journal of Neuroinflam- mation, vol. 2, article 16, 2005. [100] A. C. Foster, A. Vezzani, E. D. French, and R. Schwarcz, “Kynurenic acid blocks neurotoxicity and seizures induced in rats by the related brain metabolite quinolinic acid,” Neuroscience Letters, vol. 48, no. 3, pp. 273–278, 1984. [84] A. Schurr, C. A. West, and B. M. Acknowledgments Vecsei, “Kynurenine metabolism in plasma and in red blood cells in Parkinson’s disease,” Journal of the Neurological Sciences, vol. 239, no. 1, pp. 31–35, 2005. [97] R. Carpenedo, A. Pittaluga, A. Cozzi et al., “Presynaptic kynurenate-sensitive receptors inhibit glutamate release,” European Journal of Neuroscience, vol. 13, no. 11, pp. 2141–2147, 2001. [112] S. Fukui, R. Schwarcz, S. I. Rapoport, Y. Takada, and Q. R. Smith, “Blood-brain barrier transport of kynurenines: implications for brain synthesis and metabolism,” Journal of Neurochemistry, vol. 56, no. 6, pp. 2007–2017, 1991. [98] R. C. Roberts, K. E. McCarthy, F. Du, E. Okuno, and R. Schwarcz, “Immunocytochemical localization of the quino- linic acid synthesizing enzyme, 3-hydroxyanthranilic acid oxygenase, in the rat substantia nigra,” Brain Research, vol. 650, no. 2, pp. 229–238, 1994. [113] M. C. Barth, N. Ahluwalia, T. J. T. Anderson et al., “Kynurenic acid triggers firm arrest of leukocytes to vascular 11 Parkinson’s Disease 11 endothelium under flow conditions,” Journal of Biological Chemistry, vol. 284, no. 29, pp. 19189–19195, 2009. [114] H. N´emeth, J. Toldi, and L. V´ecsei, “Kynurenines, Parkinson’s disease and other neurodegenerative disorders: preclinical and clinical studies,” Journal of Neural Transmission. Supplementa, no. 70, pp. 285–304, 2006. [115] S. Fujigaki, K. Saito, K. Sekikawa et al., “Lipopolysaccharide induction of indoleamine 2,3-dioxygenase is mediated dom- inantly by an IFN-γ-independent mechanism,” European Journal of Immunology, vol. 31, no. 8, pp. 2313–2318, 2001. [116] T. W. Stone and L. G. Darlington, “Endogenous kynurenines as targets for drug discovery and development,” Nature Reviews Drug Discovery, vol. 1, no. 8, pp. 609–620, 2002. [117] L. McNally, Z. Bhagwagar, and J. Hannestad, “Inflammation, glutamate, and glia in depression: a literature review,” CNS Spectrums, vol. 13, no. 6, pp. 501–510, 2008. [118] H. Q. Wu, S. C. Lee, and R. Schwarcz, “Systemic admini- stration of 4-chlorokynurenine prevents quinolinate neuro- toxicity in the rat hippocampus,” European Journal of Phar- macology, vol. 390, no. 3, pp. 267–274, 2000. [119] T. W. Stone, “Development and therapeutic potential of kynurenic acid and kynurenine derivatives for neuro- protection,” Trends in Pharmacological Sciences, vol. 21, no. 4, pp. 149–154, 2000. [120] M. Platten, P. P. Ho, and L. Steinman, “Anti-inflammatory strategies for the treatment of multiple sclerosis— tryptophan catabolites may hold the key,” Drug Discovery Today: Therapeutic Strategies, vol. 3, no. 3, pp. 401–408, 2006. [121] M. Platten, P. P. Ho, S. Acknowledgments Youssef et al., “Treatment of autoim- mune neuroinflammation with a synthetic tryptophan metabolite,” Science, vol. 310, no. 5749, pp. 850–855, 2005. [122] C. W. Ritchie, A. I. Bush, and C. L. Masters, “Metal-protein attenuating compounds and Alzheimer’s disease,” Expert Opinion on Investigational Drugs, vol. 13, no. 12, pp. 1585–1592, 2004. [123] J. R. Moffett, T. Els, M. G. Espey, S. A. Walter, W. J. Streit, and M. A. A. Namboodiri, “Quinolinate immunoreactivity in experimental rat brain tumors is present in macrophages but not in astrocytes,” Experimental Neurology, vol. 144, no. 2, pp. 287–301, 1997. [124] M. Hamann, S. E. Sander, and A. Richter, “Effects of the kynurenine 3-hydroxylase inhibitor Ro 61-8048 after intrastriatal injections on the severity of dystonia in the dt mutant,” European Journal of Pharmacology, vol. 586, no. 1–3, pp. 156–159, 2008.
https://openalex.org/W4246503143
https://www.qeios.com/read/ASA3FW/pdf
English
null
Interferon Regulatory Factor 9
Definitions
2,020
cc-by
95
Qeios · Definition, February 2, 2020 Open Peer Review on Qeios Interferon Regulatory Factor 9 National Cancer Institute National Cancer Institute Qeios ID: ASA3FW · https://doi.org/10.32388/ASA3FW Source National Cancer Institute. Interferon Regulatory Factor 9. NCI Thesaurus. Code C104656. Interferon regulatory factor 9 (393 aa, ~44 kDa) is encoded by the human IRF9 gene. Interferon regulatory factor 9 (393 aa, ~44 kDa) is encoded by the human IRF9 gene. This protein plays a role in activating transcription via type I interferon signaling. This protein plays a role in activating transcription via type I interferon signaling. 1/1
https://openalex.org/W4239770121
https://scholarshare.temple.edu/bitstream/20.500.12613/6963/1/Jang-JournalArticle-2021-06.pdf
English
null
The Role of Local Communities and Well-Being in UNESCO World Heritage Site Conservation: An Analysis of the Operational Guidelines, 1994–2019
Sustainability
2,021
cc-by
10,330
sustainability sustainability sustainability   Citation: Jang, H.; Mennis, J. The Role of Local Communities and Well-Being in UNESCO World Heritage Site Conservation: An Analysis of the Operational Guidelines, 1994–2019. Sustainability 2021, 13, 7144. https://doi.org/ 10.3390/su13137144 Keywords: UNESCO; operational guidelines; local communities; well-being; world heritage sites The Role of Local Communities and Well-Being in UNESCO World Heritage Site Conservation: An Analysis of the Operational Guidelines, 1994–2019 Hanbyeol Jang * and Jeremy Mennis Hanbyeol Jang * and Jeremy Mennis Department of Geography and Urban Studies, Temple University, Philadelphia, PA 19122, USA; jmennis@temple.edu * Correspondence: hanbyeol.geo@temple.edu Abstract: UNESCO’s world heritage program aims to protect sites of cultural and natural heritage worldwide. Issues of local communities and well-being have been given increasing attention by heritage conservation scholars, but a systemic review of UNESCO guidelines has not been performed. Here, we examine the evolution of the ‘Operational Guidelines for the Implementation of the World Heritage Convention,’ documents representing the heritage conservation policies of UNESCO over the period 1994–2019. Using keyword analysis and document analysis, the findings show evidence of an increasing emphasis on local communities, growing primarily since 2005. However, the theme of well-being only first emerged in the operational guidelines in 2019. Political, economic, and environmental challenges idiosyncratic to specific places often complicate the role of local communities and well-being in heritage conservation priorities. Future research should investigate the potential implementation and implications of these changes for the guidelines at specific UNESCO world heritage sites.   Citation: Jang, H.; Mennis, J. The Role of Local Communities and Well-Being in UNESCO World Heritage Site Conservation: An Analysis of the Operational Guidelines, 1994–2019. Sustainability 2021, 13, 7144. https://doi.org/ 10.3390/su13137144 Academic Editor: Asterios Bakolas Received: 26 May 2021 Accepted: 19 June 2021 Published: 25 June 2021 1. Introduction Since its establishment in 1945, the United Nations Educational Scientific and Cultural Organization (UNESCO) and its governing bodies, the General Assembly of States Parties to the World Heritage Convention and the World Heritage Committee, have worked to protect the world’s diverse world heritage sites. As of 2021, the total number of enlisted heritage sites is 1121, including cultural, natural, and mixed categories, registered in 167 countries [1]. The protection and management of registered sites depends on the mutual collaboration and reconciliation of multiple stakeholders. Among those stakeholders, the active participation of the local communities residing nearby or within world heritage sites is key for properly conserving the enlisted heritage places [2–4]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. y p p y g g p The participation of local residents has received growing attention from heritage re- searchers, primarily focusing on the designation and implementation of heritage sites [5–7]. Negative effects on local communities in the process of heritage management have been documented in the literature [8,9]. Recently, some scholars studying heritage sites maintain that it is necessary to include local communities into the sites’ operation and management, which is both beneficial to conserving heritage sites and to the sustainability of local com- munities [10–12]. The scholars argue that natural and cultural conservation efforts through the world heritage program should be considered in collaboration with local residents because world heritage sites also serve as important resources and livelihoods in local residents’ daily lives [8]. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). While there is growing attention to incorporating local people and communities into the established procedure for implementing UNESCO heritage sites, well-being, which is a key element of a heritage site’s impact on local communities, has garnered little https://www.mdpi.com/journal/sustainability Sustainability 2021, 13, 7144. https://doi.org/10.3390/su13137144 Sustainability 2021, 13, 7144 2 of 14 attention from heritage studies and other pertinent scholarship [13]. Broadly identified, the notion of well-being incorporates physical, mental, economic, and social aspects of health [14] (p.240) [13,15], and can be understood to include facets of thinking, feeling, and imagining that affect health [16]. 1. Introduction A handful of studies have emphasized the role of well-being in heritage sites, especially in the context of community-based conservation as a way of examining the status and condition of concerned local people [13,15,17]. However, this prior research is primarily centered around internal networks of local communities and does not explicitly address the external relations of those local communities with government agencies and with relevant international organizations, such as UNESCO, by which local communities’ traditional ways of life have been impacted. We argue here that heritage conservation should be considered along with local communities’ well-being, of which a critical component is the relationship between local communities and other stakeholders (including UNESCO and its governing bodies) in heritage site management. The present research seeks to address this lacuna in the literature by examining how the positioning of local communities is reflected in UNESCO heritage site management and conservation guidelines. By utilizing keyword analysis and document analysis of official UNESCO publications, we investigate how attention to, acknowledgement of, and issues of local communities and their well-being have been identified and changed over the past two decades. Examining these changes will contribute to an understanding of evolving UNESCO heritage conservation objectives and will provide guidance for scholars and policy-makers regarding heritage conservation, indigenous populations, and their well-being. 2. Literature Review Much research related to UNESCO and heritage conservation has focused on interro- gating registered natural and cultural sites in specific regions and countries [17–20], or on the ways in which the conservation efforts of UNESCO and World Heritage Committee have contributed to economic development, mainly through natural or cultural tourism development [21–24]. Research concerning local communities’ involvement associated with UNESCO world heritage has only started to emerge in the past decade [25–27]. We consider three categories of the ways in which researchers have addressed the role of local communities in heritage conservation: (1) the influence of heritage sites on local communi- ties; (2) conflicts and cooperation between local communities and other agents concerning heritage sites; and (3) the significant roles of local communities in heritage conservation. Regarding the first category, researchers examine how heritage site registration and opera- tion have positively or negatively influenced local people’s lives [9,27]. For instance, in the registration processes of marine parks in Western Australia, local governments catalyzed coastal fishery prohibitions by extending reserve areas to protect the coastal environment, which damaged the traditional livelihoods of local communities dependent on a small-scale fishery [9]. The second category addresses conflicts and cooperation between local communities and other stakeholders such as visitors, different levels of governments, and international organizations [28–31]. For example, Zhang and colleagues [31] articulate the multiple layers of conflicts regarding heritage tourism in China. With the increase of the Chinese tourism industry through heritage site conservation, an encounter of local cultural, social, and economic values with the external values of the conservation program catalyzed conflicts regarding new governmental regulations and guidelines, which intermixed with the conflicts of interests among multiple stakeholders (e.g., governments, local raft workers, tourism agency, and tourists) in Wuyi Mountain, China [31]. Another study touches on how tourists visiting a certain heritage site will result in the degradation of existing local peoples’ economic, social, and environmental conditions because, in many developing countries, heritage tourism is often funded by international organizations or supported by foreign capital with a top-down development model [30]. Thus, local communities’ concerns become relegated and compromised for the sake of tourism development [30]. Sustainability 2021, 13, 7144 3 of 14 This research also suggests that this tension can be resolved when different actors are identified and ‘take responsibility for governance and development of the designated area’ [30] (p.44). 2. Literature Review The third category incorporates the role of local communities in the sustainability of heritage site conservation [11,32–34]. Deacon and Smeets [34], for example, investigate in what manner local people are involved with the operations of UNESCO heritage sites, including nomination, registration, and management. This research indicates that local communities are often neglected in terms of decision-making processes [34]. By recognizing that the notion of ‘community’ was largely neglected in the criteria of early heritage sites’ nomination and management processes, Rössler [33] also takes into account that community involvement should be prioritized in order to sustain a desirable heritage conservation. Highlighted as one of the key Sustainable Development Goals (SDGs) by the United Nations (Goal 3: Good health and well-being for people), well-being has recently received increasing attention in heritage conservation, especially concerning the well-being of those who live in developing countries [35,36]. Although many heritage sites are located in such countries, there have been few attempts to engage with issues of well-being among heritage studies researchers [15]. A handful of research on well-being and heritage conservation can be classified into research focusing on visitors (i.e., tourists) and on local commu- nities. Regarding the former, researchers have found that visitors benefit from heritage conservation sites, including from physical exercises and mental restoration [37–39]. On the other hand, the latter has shown negative consequences of heritage site conservation on local communities’ well-being. For example, Su and colleagues [40,41] pay attention to the resettlement of local communities as a result of the protection of heritage areas. The authors argue that resettlement plans often lack sufficient and concrete strategies, including the provision of health care facilities at the new settlement, which can negatively affect the long-term well-being of the local residents of the Yinhuwan village, in Mount Sanqingshan National Park, China [40,41]. Other researchers noted that local residents’ subjective well-being is associated with their socio-economic status, indicating that people with high income through heritage tourism are more likely to show positive subjective well-being regarding tourism development [42]. Research on Australian heritage sites and Aboriginal Australians’ well-being high- lights the mental and cognitive connections between Aborigines and the cultural values of spiritual heritage sites [15,43–46]. 2. Literature Review In analogous research on the domestic heritage site in the East Anglia region of England, Power and colleagues [13] explored the social well-being of local participants toward local community-based heritage conservation, such as personal enrichment, social learning, and satisfaction from conservation-related activities. These findings suggest that heritage sites and their management have the potential to provide positive effects for the well-being of local communities, by enhancing the level of their physical and spiritual satisfaction. 3. Data and Methods To investigate the extent to which the issue of local communities and their well-being is recognized by UNESCO, we examine the ‘Operational Guidelines for the Implementation of the World Heritage Convention’ (referred to hereafter as the ‘operational guidelines’), that have been published by UNESCO on an occasional basis since 1977. The operational guidelines specify the overall procedure regarding the designation, management, monitor- ing, and support of the UNESCO world heritage program [47]. In the present research, the operational guidelines are used for gathering data on the recognition of local communities in heritage site conservation as well as local community well-being in the operation of the world heritage program. For the analysis, the operational guidelines for 1994–2019 (a total of 14 documents) were downloaded from the UNESCO website and examined. Offering fundamental protocols for the overall heritage maintenance, the operational guidelines reflect UNESCO’s policy changes through updates to related concepts, knowledge, and experiences [47]. The transformation of the operational guidelines reflects the results of the Sustainability 2021, 13, 7144 4 of 14 negotiations taking place during the sessions of the World Heritage Committee. The opera- tional guidelines have been analyzed previously in multiple academic papers, presenting pivotal stances of UNESCO on heritage site management [27,33], but not with a focus on local communities or well-being, as in the present research. Using these operational guidelines, we conduct a keyword analysis and a document analysis to demonstrate how the interest of UNESCO in the themes of local community and well-being has changed over time. While a quick overview of the changes of the operational guidelines is avail- able in a separate report of the sessions of the World Heritage Committee, it does not provide a detailed explanation of what has been updated in which contexts. Thus, a sys- temic examination of the operational guidelines will complement this gap by identifying more details. Stemming from the field of linguistics, keyword analysis has been identified as a mixed quantitative and qualitative analytical technique [48–50]. A quantitative step involves collecting the number of words indicating a particular theme [51]. In the qualitative step, a researcher identifies each world within the context of the main themes [50]. 3. Data and Methods After this two-step verification is completed, the counted words are used for investigating the annual transitions in the themes of the operational guidelines from 1994 to 2019; the more frequently specific words are used, the greater the assumed importance of these words in the document [52]. The process of selecting keywords is based in part on the ‘World Heritage POLICY COMPENDIUM’ webpage, a portion of which represents the policies regarding community engagement: 6.1 Participation of local communities and other stakeholders, 6.2 Human rights and right-based approach, 6.3 Gender equality, 6.4 Indigenous people, 6.5 Youth, and 6.6 Fostering of peace and security. Among the policies, ‘local communities,’ ‘human rights,’ and ‘indigenous people’ served as keywords that indicate a focus on local com- munities. Regarding the selection of ‘human rights,’ we referred to literature invoking the interactions of human rights and well-being (health) where well-being is considered a necessary component to facilitate human rights [53–55], through which we also include ‘well-being’ as one of keywords for the analysis. Six other keywords are reviewed and identified based on the pertinent literature, including ‘local community’ and ‘local commu- nities’ [27,34], ‘traditional societies’ and ‘indigenous societies’ [56], and ‘local people’ and ‘local population’ [27,34]. These eight keywords are employed to undertake a word count for each by browsing the number of words in the PDF formats of the papers published in 1997–2019 (a total of 12 documents), and manually browsing the papers published in 1977–1996 (a total of 11 documents), since tracking electronic versions word by word were not available for these years. y As a complement to the keyword analysis, we also interpret the text of the operational guidelines in light of the selected keywords to place them in context [57], to provide a more detailed and nuanced understanding of UNESCO’s attention to local communities and their well-being. Furthermore, we examine a series of other related official documents: decision papers adopted in the ordinary sessions of the World Heritage Committee from 1977 to 2019 (a total of 43 documents), focusing on the context of the keywords, since the operational guidelines mirror the decisions adopted in these annual sessions [33]. 3. Data and Methods In our examination of the decision papers, we interrogate where the selected keywords of the operational guidelines occur in the decision papers and in which contexts, such as the background topic within which the keyword appears, the reason why the keyword is mentioned, and the progress noted in advancing the UNESCO’s objectives. This comparative review of these various sources assists in identifying pertinent narratives that are bound up with the changes in the operational guidelines. 4. Results Participation of local people in the nomination process is essential to make them feel a shared responsibility with the State Party in the maintenance of the site [60]. (Article 14, 1996) (part of Article 14, 1994). 14. Participation of local people in the nomination process is essential to make  them feel a shared responsibility with the State Party in the maintenance of the  site [60] (Article 14 1996) 14. Participation of local people in the nomination process is essential to make them feel a shared responsibility with the State Party in the maintenance of the site [60]. (Article 14, 1996) (part of Article 14, 1994). 14. Participation of local people in the nomination process is essential to make  them feel a shared responsibility with the State Party in the maintenance of the  site [60] (Article 14, 1996). 0 0 0 0 0 0 3 3 3 3 3 3 3 3 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 local community 0 1 1 1 1 8 7 8 8 8 7 7 7 12 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 local communities 0 0 0 0 0 0 0 0 0 0 5 5 5 18 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 indigenous peoples 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 traditional socieities Figure 1. Cont. 4. Results The operational guidelines from 1977 to 1992 (a total of eight documents) do not include any of the keywords. Figure 1 displays the annual frequency of the keywords in the operational guidelines from 1994 to 2019. The emergence of the keywords begins Sustainability 2021, 13, 7144 5 of 14 with the appearance of the term ‘local people’ in the 1994 document, which serves as the first acknowledgment of the role of local people in the maintenance of heritage sites (The authors highlighted and indented the keywords inside the quotations below.). However, the context of the 1994 term, shown below, suggests that local concerns are considered secondary to the committee’s decision-making processes. In the 1996 operational guidelines, the corresponding part became revised with a separate article 14 that eliminates the clause specifying the secondary status of local communities’ participation in the nomination process, reflecting the decision of the 19th session of the World Heritage Committee (1995) that endorses the inclusion of ‘cultural properties in the world heritage list’ [58] (p. 74). operational guidelines from 1994 to 2019. The emergence of the keywords begins with the  appearance of the term ‘local people’ in the 1994 document, which serves as the first ac‐ knowledgment of the role of local people in the maintenance of heritage sites (The authors  highlighted and indented the keywords inside the quotations below.). However, the con‐ text of the 1994 term, shown below, suggests that local concerns are considered secondary  to the committee’s decision‐making processes. In the 1996 operational guidelines, the cor‐ responding part became revised with a separate article 14 that eliminates the clause spec‐ ifying the secondary status of local communities’ participation in the nomination process,  reflecting the decision of the 19th session of the World Heritage Committee (1995) that  d h l f l l h ld h l ( ) 14. Participation of local people in the nomination process is essential to make them feel a shared responsibility with the State Party in the maintenance of the site [60]. (Article 14, 1996) (part of Article 14, 1994). 14. Participation of local people in the nomination process is essential to make  them feel a shared responsibility with the State Party in the maintenance of the  site [60] (Article 14 1996) 14. 4. Results 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 traditional socieities Figure 1. Cont. Sustainability 2021, 13, 7144 Sustainability 2021, 13, x FOR 6 of 14 6 of 14 Figure 1. The frequency of the keywords in the operational guidelines (1994–2019); Data Source: [59–72]. 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 local people 0 0 0 0 0 1 1 1 1 1 1 1 0 0 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 local population 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 well-being 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 2 4 6 8 10 12 14 16 18 1994 1995 1996 1997 1998 1999 2000 2011 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 human rights Figure 1. The frequency of the keywords in the operational guidelines (1994–2019); Data Source: [59–72]. The frequency of the keywords in the operational guidelines (1994–2019); Data Source: [59–72]. Figure 1. The frequency of the keywords in the operational guidelines (1994–2019); Data Source: [59–72]. The term ‘local communities’ appears for the first time in the 1996 operational guide‐ lines, emphasizing that the approval of local communities, as well as the collaboration  with those communities, are important in the nomination of heritage places. 4. Results The term ‘local communities’ appears for the first time in the 1996 operational guide- lines, emphasizing that the approval of local communities, as well as the collaboration with those communities, are important in the nomination of heritage places. …… It is important that due attention be paid to the full range of values repre‐ sented in the landscape, both cultural and natural. The nominations should be  prepared in collaboration with and the full approval of local communities [60]  (Article 41, 1996). . . . . . . It is important that due attention be paid to the full range of values represented in the landscape, both cultural and natural. The nominations should be prepared in collaboration with and the full approval of local communities [60]. (Article 41, 1996) The frequency of keywords concerning the theme of local communities did not in‐ crease further until the 2005 publication. The frequency of the term ‘local communities’  surges to eight occurrences in 2005 (see Figure 1). In addition, the terms ‘traditional soci‐ eties’ and ‘local population’ also make an appearance that same year (see Figure 1). The frequency of keywords concerning the theme of local communities did not increase further until the 2005 publication. The frequency of the term ‘local communities’ surges to eight occurrences in 2005 (see Figure 1). In addition, the terms ‘traditional societies’ and ‘local population’ also make an appearance that same year (see Figure 1). Partners in the protection and conservation of World Heritage can be those in‐ dividuals and other stakeholders, especially local communities, governmental,  nongovernmental and private organizations and owners who have an interest  and  involvement  in  the  conservation  and  management  of  a  World  Heritage  property [64] (Article 40, 2005). Partners in the protection and conservation of World Heritage can be those individuals and other stakeholders, especially local communities, governmental, nongovernmental and private organizations and owners who have an interest and involvement in the conservation and management of a World Heritage property [64]. (Article 40, 2005) ….. Human activities, including those of traditional societies and local commu‐ nities, often occur in natural areas. These activities may be consistent with the  outstanding universal value of the area where they are ecologically sustainable  [64] (Article 90, 2005). Wh l h d f h k d f l d . . . . . . Human activities, including those of traditional societies and local communities, often occur in natural areas. 4. Results These activities may be consistent with the outstanding universal value of the area where they are ecologically sustainable [64]. (Article 90, 2005) While the increased appearance of the keywords in 2005 is meaningful in demon‐ strating the increasing attention given to local communities, their context within the op‐ erational guidelines suggests that the emphasis is on the role and responsibility of local  communities in conserving UNESCO heritage sites sustainably, rather than the impact of  heritage conservation on local communities or the challenges to local communities’ liveli‐ While the increased appearance of the keywords in 2005 is meaningful in demonstrat- ing the increasing attention given to local communities, their context within the operational guidelines suggests that the emphasis is on the role and responsibility of local communi- ties in conserving UNESCO heritage sites sustainably, rather than the impact of heritage Sustainability 2021, 13, 7144 7 of 14 conservation on local communities or the challenges to local communities’ livelihoods or well-being presented by heritage conservation. conservation on local communities or the challenges to local communities’ livelihoods or well-being presented by heritage conservation. g p y g This argument is supported by the decision papers between 1996 and 2004. These decision papers indicate recurrent revision processes of the operational guidelines that are finally adopted in the 2005 decision paper. For example, the 1999 decision paper shows a growing reference to local communities and their participation in heritage management, as cultural heritage lists have increased more than ever before. The Director-General of UNESCO, in the address of the 1999 session of the World Heritage Committee, also empha- sized the role of local communities as ‘the true custodians of the World Heritage sites’ [73] (p.74). Several parts of the 2000 decision paper also provide broader contexts in which political negotiations for revising the operational guidelines played out, partly recognizing global institutions’ special attention to indigenous people and their communities by accept- ing the recommendations of the World Heritage Indigenous Peoples Council of Experts (WHIPCOE) [74] (p.96), and from the Principle 22 of the Rio Declaration on Environment and Development and Chapter 26 of Agenda 21 [74] (p.98). The keyword count remains relatively constant between 2005 and 2015, at which time the term ‘indigenous peoples’ first emerges. Compared with the passages from 2005 in Article 40 and 90, above, the passages from 2015 reveal the development of a more sophisticated appreciation of local communities. 4. Results Importantly, this year also marks the first use of the terms ‘well-being’ and ‘human rights,’ which suggests a greater concern over the impact of the conservation of Sustainability 2021, 13, 7144 8 of 14 natural or cultural heritage sites on local communities. For example, in the Article 214bis (2019) [72], below, there is a greater recognition of the economic benefits and opportunities offered to local communities associated with protected sites. natural or cultural heritage sites on local communities. For example, in the Article 214bis (2019) [72], below, there is a greater recognition of the economic benefits and opportunities offered to local communities associated with protected sites. . . . . . . The programmes may be based on innovation and local entrepreneurship, and aimed in particular at medium/small/micro scale levels, to promote sus- tainable and inclusive economic benefits for local communities and indigenous peoples and to identify and promote opportunities for public and private invest- ment in sustainable development projects, including those that promote use of local materials and resources and foster local cultural and creative industries and safeguarding intangible heritage associated with World Heritage properties [72]. (Article 214bis, 2019) . . . . . . The programmes may be based on innovation and local entrepreneurship, and aimed in particular at medium/small/micro scale levels, to promote sus- tainable and inclusive economic benefits for local communities and indigenous peoples and to identify and promote opportunities for public and private invest- ment in sustainable development projects, including those that promote use of local materials and resources and foster local cultural and creative industries and safeguarding intangible heritage associated with World Heritage properties [72]. (Article 214bis, 2019) Moreover, Article 215 (2019) [72] indicates that local communities’ knowledge and traditional ways of resolving conflicts are also regarded as legitimate and valuable, con- tributing to ‘the conservation and management of World Heritage properties.’ . . . . . . States Parties are encouraged to support scientific studies and research methodologies, including traditional and indigenous knowledge held by local communities and indigenous peoples, with all necessary consent. Such studies and research are aimed at demonstrating the contribution that the conservation and management of World Heritage properties, their buffer zones and wider setting make to sustainable development, such as in conflict prevention and resolution, including, where relevant, by drawing on traditional ways of dispute resolution that may exist within communities [72]. 4. Results While the concern about local communities continues to be centered around conserving heritage sites, as it was in the 2005 document, Article 123 suggests that local (indigenous) communities have gained more concrete and indispensable roles in the operation of heritage sites, through the use of ‘prior and informed consent, public consultations, and hearings,’ as compared to prior years’ documents. Partners in the protection and conservation of World Heritage can be those individuals and other stakeholders, especially local communities, indigenous peoples, governmental, non-governmental and private organizations and owners who have an interest and involvement in the conservation and management of a World Heritage property [69]. (Article 40, 2015) Participation in the nomination process of local communities, indigenous peo- ples, governmental, non-governmental and private organizations and other stake- holders is essential to enable them to have a shared responsibility with the State Party in the maintenance of the property. States Parties are encouraged to prepare nominations with the widest possible participation of stakeholders and to demon- strate, as appropriate, that the free, prior and informed consent of indigenous peoples has been obtained, through, inter alia making the nominations publicly available in appropriate languages and public consultations and hearings [69]. (Article 123, 2015) The decision papers during this period also present a growing recognition of indige- nous peoples and their communities. According to the 2012 decision paper, as more natural sites with indigenous residents become enlisted, the references to indigenous people ac- cordingly become more noticeable [75]. With respect to the revision of the operational guidelines, there were several meetings dedicated to discussing how to deal with indige- nous peoples [76] (p.7). The 2014 decision paper also shows that the World Heritage Committee will continue to focus on local communities and indigenous peoples, as they are part of ‘the eight key dimensions of sustainable development’ [77] (p.7). This transition of the World Heritage Committee to local communities and indigenous peoples contributes to adopting a number revisions of the operational guidelines in the 2015 decision paper entailing the emphasis on those keywords [78] (p.243). g p y p The keyword count increases abruptly again in the 2019 operational guidelines, where the frequency of the terms ‘local communities’ and ‘indigenous people’ increases to 12 and 18, respectively. 5. Discussion and Conclusions The findings from the operational guidelines over two decades illustrate that UNESCO and the World Heritage Committee have increasingly recognized that local people and their communities are to be included in the overall initiatives and implementations of UNESCO heritage sites. The rise in the number of keywords referring to local communities, shown in Figure 1, indicates these pertinent organizations’ efforts to recognize local communities as a vital component of sustainable heritage site management. These findings suggest that UNESCO has made some attempts to respond to prior critiques regarding the lack of attention to local communities in heritage conservation. Our findings regarding well- being, however, suggest that UNESCO has only just begun to engage with issues of local communities’ health and well-being. g The growing emphasis on local communities by UNESCO can be attributed to specific decisions of UNESCO’s governing body, the World Heritage Committee, arising from broader interests related to human rights at the UN. Deacon and Smeets [34], for instance, claim that the UNESCO 2003 convention on intangible cultural heritage was committed to emphasizing the roles of local communities for collaborative and sustainable heritage conservation. Other scholars also argue that the ‘United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP)’ in the general assembly of 2007 enabled the growing concern for local communities at heritage conservation sites [16,27]. These efforts reflect the interest of heritage conservation researchers on local communities, such as Bortolotto [81], who argues that protecting ‘intangible cultural heritage’ is inseparable from the concerns of local communities, and Logan [82], who argues that the rights of local communities’ cultural practices ought to be recognized as a form of human rights. Those shifts in the emphasis on local communities are consistent with the records of the meeting sessions of the World Heritage Committees, with increasing appearances of the term ‘human right’ in recent years [78,79,83–87]. Several scholars have noted that UNESCO has engaged with the idea of local communities in heritage preservation for a number of years, albeit without putting the prioritization of local communities into practice [88–90]. p g p p Given that the concept of well-being has only appeared in the operational guidelines in 2019, the recognition of well-being as an important aspect of heritage conservation is only emerging and deserves continued monitoring and attention, particularly regarding the impact of heritage conservation on the well-being of local communities. 4. Results (Article 215, 2019) Although the emergence of both terms, ‘well-being’ and ‘human rights’ incorporate multiple stakeholders, and are not limited to local communities. This conspicuous shift toward an emphasis on well-being and human rights suggests that the impacts of con- servation efforts on local communities receive more attention from UNESCO than they have before. States Parties are encouraged to mainstream into their programmes and activities related to the World Heritage Convention the principles of the relevant policies adopted by the World Heritage Committee, the General Assembly of States Par- ties to the Convention and the UNESCO Governing Bodies, such as the Policy Document for the Integration of a Sustainable Development Perspective into the Processes of the World Heritage Convention and the UNESCO policy on engaging with indigenous peoples, as well as other related policies and docu- ments, including the 2030 Agenda for Sustainable Development and international human rights standards [72]. (Article 14bis, 2019) . . . . . . Properties may support a variety of ongoing and proposed uses that are ecologically and culturally sustainable and which may enhance the quality of life and well-being of communities concerned . . . . . . [72]. (Article 119, 2019) In recognizing the diversity mentioned above, common elements of an effective management system could include: b) a respect for diversity, equity, gender equality and human rights and the use of inclusive and participatory planning and stakeholder consultation processes; [72]. (Article 111b, 2019) The keyword well-being appears for the first time in the 2011 decision paper, but its context is not related to local communities, but to the general context of ‘the economic and social well-being of all the inhabitants’ [14] (p. 240, p. 242). Although well-being has been invoked when deciding to inscribe certain sites on the World Heritage Lists ([79] (p. 208); [80] (p. 169)), its articulation is limited in the recent decision papers. In terms of human rights, the decision papers alternatively use descriptions such as ‘their rights to maintain sustainable traditional use of resources’ [78] (p. 64), and ‘user access rights, cultural rights’ [79] (p. 149) to express the specific dimensions of human rights. The decision papers also show that the recent increased attention to human rights is a Sustainability 2021, 13, 7144 9 of 14 9 of 14 reflection of global concerns raised by the Office of the United Nations High Commissioner for Human Rights [78] (p. 4. Results 158) and the African Commission on Human and Peoples’ Right (ACHPR) [78] (p. 67). In summary, the UNESCO’s operational guidelines display an increasing transition to the theme of local communities, with 2005, 2015, and 2019 standing out as important years when key phrases related to local communities’ emerged or increased in number. Importantly, however, much of the attention on local communities has been focused on the role of such communities in heritage conservation practice and management, as opposed to the potential impacts of conservation on local communities. This emphasis remained until 2019, when issues of well-being and human rights began to receive attention in the operation guidelines. 5. Discussion and Conclusions This transition is also identified in the sessions of the World Heritage Committees (The references of the illustrated years are as follows: [91–96] (The documents of [93] and [95] were searched using the term ‘wellbeing’ instead of ‘well-being.’). Such an effort follows previous efforts to place the issue of well-being and human rights in international organizations with respect to sites of cultural heritage [16] and sustainable development [97]. These efforts could be extended to considerations of different mental and physical health aspects of well- being [98,99] as well as to the challenges and implications of implementing considerations of well-being in natural and cultural heritage site management [8,100]. There are a number of challenges to involving local communities as stakeholders in heritage management. For instance, a large heritage site may include several community Sustainability 2021, 13, 7144 10 of 14 10 of 14 groups, and protecting the rights of one group may neglect the rights of other groups [9,16]. Deacon and Smeets [34] argue that in the heritage designation process, local communities are unlikely to participate in the ultimate decisions of heritage management, even though the authenticity and cultural importance of a heritage site often rests with that local community. Moreover, it should be taken into account that, while we identified a growing recognition of local communities both in the operational guidelines and decisions papers, incidental observation suggests that associated practices at the heritage sites themselves do not seem to adhere to the guidelines. In this regard, local people and communities neglected in heritage conservation may exemplify a continued mismatch between the practical reality and the administrative ideal [101] regarding the role of local communities and well-being in heritage conservation. These challenges raise questions about the capability of UNESCO to address the interests of local communities, even though the issue receives greater attention in the operational guidelines, as the operational guidelines serve primarily as recommendations and do not carry any regulatory consequences. Thus, enhancing the well-being of local communities in heritage conservation may require more than simple changes to the operational guidelines. g p g We acknowledge several limitations to this research. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. References 1. UNESCO. UNESCO World Heritage Centre—World Heritage List. Available online: https://whc.unesco.org/en/list/ (accessed on 15 June 2021). 1. UNESCO. UNESCO World Heritage Centre—World Heritage List. Available online: https://whc.unesco.org/en/list/ (accessed on 15 June 2021). 2. Stenseke, M. Local Participation in Cultural Landscape Maintenance: Lessons from Sweden. Land Use Policy 2009, 26, 214–223. [C R f] on 15 June 2021). 2. Stenseke, M. Local Participation in Cultural Landscape Maintenance: Lessons from Sweden. Land Use Policy 2009, 26, 214–223. [CrossRef] J ) 2. Stenseke, M. Local Participation in Cultural Landscape Maintenance: Lessons from Sweden. Land [CrossRef] 3. Stoll-Kleemann, S.; de la Vega-Leinert, A.C.; Schultz, L. The Role of Community Participation in the Effectiveness of UNESCO Biosphere Reserve Management: Evidence and Reflections from Two Parallel Global Surveys. Environ. Conserv. 2010, 227–238. [CrossRef] 4. Amir, S.; Osman, M.M.; Bachok, S.; Ibrahim, M. Sustaining Local Community Economy through Tou World Heritage City. Procedia Environ. Sci. 2015, 28, 443–452. [CrossRef] 5. Labadi, S. Representations of the Nation and Cultural Diversity in Discourses on World Heritage. J. Soc. Archaeol. 2007, 7, 147–170. [CrossRef] 6. Galla, A. World Heritage: Benefits Beyond Borders; Cambridge University Press: Cambridge, UK, 2012. Galla, A. World Heritage: Benefits Beyond Borders; Cam 7. Meskell, L.; Liuzza, C.; Bertacchini, E.; Saccone, D. Multilateralism and UNESCO World Heritage: Decisi and Political Processes. Int. J. Herit. Stud. 2015, 21, 423–440. [CrossRef] 7. Meskell, L.; Liuzza, C.; Bertacchini, E.; Saccone, D. Multilateralism and UNESCO World Heritage: Decision-Making, States Parties and Political Processes. Int. J. Herit. Stud. 2015, 21, 423–440. [CrossRef] 8 Oviedo G ; Puschkarsky T World Heritage and Rights Based Approaches to Nature Conservation Int J Herit Stud 2012 18 Oviedo, G.; Puschkarsky, T. World Heritage and Rights-Based Approaches to Nature Conservation. Int. J 285–296. [CrossRef] [ ] 9. Jones, R.; Shaw, B. Thinking Locally, Acting Globally? Stakeholder Conflicts Over UNESCO World Heritage Inscription in Western Australia. J. Herit. Tour. 2012, 7, 83–96. [CrossRef] 10. Nicholas, L.N.; Thapa, B.; Ko, Y.J. RESIDENTS’PERSPECTIVES OF a World Heritage Site: The Pitons Management Area, St. Lucia. Ann. Tourism Res. 2009, 36, 390–412. [CrossRef] 11. Disko, S. World Heritage Sites and Indigenous Communities: The Importance of Adopting a Human Rights-Based Approach. In Community Development through World Heritage; Albert, M.-T., Richon, M., Viñals, M.J., Witcomb, A., Eds.; United Nations Educational, Scientific and Cultural Organization: Paris, France, 2012; pp. 16–26. 12. Su, M.M.; Wall, G. 5. Discussion and Conclusions 11 of 14 Sustainability 2021, 13, 7144 Conflicts of Interest: The authors declare no conflict of interest. References Community Participation in Tourism at a World Heritage Site: Mutianyu Great Wall, Beijing, China. Int. J. Tour. Res. 2014, 16, 146–156. [CrossRef] 13. Power, A.; Smyth, K. Heritage, Health and Place: The Legacies of Local Community-Based Heritage Conservation on Social Wellbeing. Health Place 2016, 39, 160–167. [CrossRef] g 14. WHC-11/35.COM/20. Decisions Adopted by the World Heritage Committee at its 35th Session; United Nations Educational, Scientific and Cultural Organization: Paris, France, 2011. New and Emerging Challenges to Heritage and Well-being: A Critical Review. Heritage 2019, 2, 1300–1315 15. Taçon, P.S.; Baker, S. New and Emerging Challenges to Heritage and Well-being: A Critical Review [CrossRef] 16. Hodder, I. Cultural Heritage Rights: From Ownership and Descent to Justice and Well-Being. Anthropol. Q. 2010, 83, 861–882. [CrossRef] [ ] 17. Coningham, R.; Lewer, N. Paradise Lost: The Bombing of the Temple of the Tooth—a UNESCO World Heritage Site in Sri Lanka. Antiquity 1999, 73, 857–866. [CrossRef] q y [ ] 18. Subade, R.F. Mechanisms to Capture Economic Values of Marine Biodiversity: The Case of Tubbataha Reefs UNESCO World Heritage Site, Philippines. Mar. Policy 2007, 31, 135–142. [CrossRef] R.F. Mechanisms to Capture Economic Values of Marine Biodiversity: The Case of Tubbataha Reefs UNES Site, Philippines. Mar. Policy 2007, 31, 135–142. [CrossRef] Subade, R.F. Mechanisms to Capture Economic Values of Marine Biodiversity: The Case of Tubbataha R h l l [ f] 18. Subade, R.F. Mechanisms to Capture Economic Values of Marine Biodiversity: The Case of Tubba Heritage Site Philippines Mar Policy 2007 31 135–142 [CrossRef] 18. Subade, R.F. Mechanisms to Capture Economic Values of Marine Biodiversity: The Case of Tubbatah Heritage Site, Philippines. Mar. Policy 2007, 31, 135–142. [CrossRef] 19. Shamsuddin, S.; Sulaiman, A.B.; Amat, R.C. Urban Landscape Factors that Influenced the Character of George Town, Penang UNESCO World Heritage Site. Procedia-Soc. Behav. Sci. 2012, 50, 238–253. [CrossRef] 20. Gullino, P.; Larcher, F. Integrity in UNESCO World Heritage Sites. A Comparative Study for Rural Landscapes. J. Cult. Herit. 2013, 14, 389–395. [CrossRef] M.A. The Heritage-Scape: UNESCO, World Heritage, and Tourism; Lexington Books: Lanham, MD, USA, 2008. 21. Di Giovine, M.A. The Heritage-Scape: UNESCO, World Heritage, and Tourism; Lexington Books: Lanham 22. Piotrowski, J.M.; Arezki, R.; Cherif, R. Tourism Specialization and Economic Development: Evidence from the UNESCO World Heritage List. IMF Working Paper. 2009, 09/176. Available online: https://ssrn.com/abstract=1457599 (accessed on 6 April 2021). 23. VanBlarcom, B.L.; Kayahan, C. 5. Discussion and Conclusions First, our study is limited to the analysis of the UNESCO’s operational guidelines documents, and thus, while this analysis does reveal important transitions to the recognition of the concepts of local com- munities and well-being in UNESCO’s objectives in heritage conservation, we do not have information on the causes or consequences of these changes to the guidelines, despite the interrogation of the decision papers, whose decisions are reflected on the changes to the operational guidelines. Additionally, the relationship between other official documents (e.g., charters, recommendations, and declarations, etc.) of UNESCO, and its governing bodies, and their explicit impact on the operational guidelines is also unknown. An analy- sis of this relationship might contribute to gaining a more contextualized understanding of noticeable changes of the operational guidelines in 2005, 2015, and 2019. Second, our choice of selected keywords may have limited our recognition of all instances of concepts relating to local communities and well-being. Notably, none of the selected keywords appeared in the operational guidelines from 1977 to 1993 (a total of nine documents). Other potential keywords implicating local communities or well-being more implicitly may reveal a recognition of these concepts in earlier UNESCO documents. g p To our knowledge this is the first study to systemically examine UNESCO heritage conservation guidelines regarding the emphasis on local communities and well-being. The findings show a growing emphasis of UNESCO on local communities beginning primarily in 2005, and the emergence of issues of well-being and human rights, just in 2019. We are convinced that these issues are increasingly recognized as key components of heritage conservation, though we recognize that political, economic, and environmental challenges idiosyncratic to specific places often complicate the place of local communities and well- being in heritage conservation priorities. Future research should investigate the potential implementation and implications of these changes for the guidelines at specific UNESCO world heritage sites. Author Contributions: Conceptualization, H.J. and J.M.; methodology, H.J. and J.M.; formal analysis, H.J.; writing—original draft preparation, H.J.; writing—review and editing, H.J. and J.M.; All authors have read and agreed to the published version of the manuscript. Funding: The publication of this article is funded in part by pending support from the Temple University Libraries Open Access Publishing Fund. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Publicly available datasets were analyzed in this study. These data can be found here: [https://whc.unesco.org/en/guidelines/ and https://whc.unesco.org/en/ sessions/]. References Assessing the Economic Impact of a UNESCO World Heritage Designation. J. Herit. Tour. 2011, 6, 143–164. [CrossRef] C.; Rizzo, I. The Effects of UNESCO World Heritage List Inscription on Tourism Destinations Performance in n. Model. 2016, 53, 494–508. [CrossRef] 24. Cuccia, T.; Guccio, C.; Rizzo, I. The Effects of UNESCO World Heritage List Inscription on Tourism Des Italian Regions. Econ. Model. 2016, 53, 494–508. [CrossRef] 25. Bertacchini, E.; Saccone, D.; Santagata, W. Embracing Diversity, Correcting Inequalities: Towards a New Global Governance for the UNESCO World Heritage. Int. J. Cult. Policy 2011, 17, 278–288. [CrossRef] g y 26. Brumann, C. Community as Myth and Reality in the UNESCO World Heritage Convention. In Between Imagined Communities and Communities of Practice; Tauschek, M., Ed.; Göttingen University Press: Göttingen, Germany, 2015; pp. 273–289. Available online: http://books.openedition.org/gup/235 (accessed on 3 March 2021). 27. Labadi, S. UNESCO, world heritage, and sustainable development: International discourses and local impacts. In Collision or Collaboration, Collision or Collaboration: Archaeology Encounters Economic Development ed.; Gould, P., Pyburn, A., Eds.; Springer: Cham, Switzerland, 2017; pp. 45–60. [CrossRef] 12 of 14 12 of 14 Sustainability 2021, 13, 7144 28. Maikhuri, R.K.; Nautiyal, S.; Rao, K.S.; Saxena, K.G. Conservation Policy–people Conflicts: A Case Study from Nanda Devi Biosphere Reserve (a World Heritage Site), India. For. Policy Econ. 2001, 2, 355–365. [CrossRef] p ( g ) y 29. Ingram, C.B. Parks, People and Planning: Local Perceptions of Park Management on the Ningaloo Coast, North West Cape, Western Australia. Master’s Thesis, Curtin University of Technology, Bentley, Australia, 2008. 30. Viñals, M.J.; Morant, M. Heritage, Tourism and Local Community Interactions within the Framework of Site Management. In Community Development through World Heritage; Albert, M.-T., Richon, M., Viñals, M.J., Witcomb, A., Eds.; United Nations Educational, Scientific and Cultural Organization: Paris, France, 2012; pp. 40–48. g pp 31. Zhang, C.; Fyall, A.; Zheng, Y. Heritage and Tourism Conflict within World Heritage Sites in China: A Longitudinal Study. Curr. Issues Tour. 2015, 18, 110–136. [CrossRef] 32. Landorf, C. Managing for Sustainable Tourism: A Review of Six Cultural World Heritage Sites. J. Sustain. Tour. 2009, 17, 53–70. [CrossRef] 33. Rössler, M. Linking Nature and Culture: World Heritage Cultural Landscapes. In Cultural Landscapes: The Challenges of Conservation, Proceedings of the World Heritage 2002 Shared Legacy, Common Responsibility Associated Workshops, Ferrara, Italy, 11–12 November 2002; pp. 10–15. y 34. Deacon, H.; Smeets, R. References [CrossRef] ng to the Ancestors: Why Rock Art is Important for Indigenous Australians and their Well-Being. Rock Art t Res. Assoc. (AURA) 2019, 36, 5–14. 46. Tacon, P.S. Connecting to the Ancestors: Why Rock Art is Important for Indigenous Australians and t Res. J. Aust. Rock Art Res. Assoc. (AURA) 2019, 36, 5–14. 47. Wiener, A.S. (Ed.) The Evolution of the Condition of Authenticity within the Operational Guidelines; Academia: San Francisco, CA, USA, 2016; pp. 1841–1851. pp 48. Seale, C.; Ziebland, S.; Charteris-Black, J. Gender, Cancer Experience and Internet use: A Comparative Keyword Analysis of Interviews and Online Cancer Support Groups. Soc. Sci. Med. 2006, 62, 2577–2590. [CrossRef] pp p 49. Wezel, A.; Soldat, V. A Quantitative and Qualitative Historical Analysis of the Scientific Discipline of Agroecology. Int. J. Agric. Sustain. 2009, 7, 3–18. [CrossRef] 50. Seale, C.; Charteris-Black, J. Keyword Analysis: A New Tool for Qualitative Research. In The SAGE Handbook of Qualitative Methods in Health Researc; SAGE Publications Ltd.: London, UK, 2010; pp. 536–665. [CrossRef] pp 51. Ferilli, S.; Biba, M.; Basile, T.M.A.; Esposito, F. Combining Qualitative and Quantitative Keyword E 51. Ferilli, S.; Biba, M.; Basile, T.M.A.; Esposito, F. Combining Qualitative and Quantitative Keyword Extraction Methods with Document Layout Analysis. In Proceedings of the Ircdl, Austin, TX, USA, 15–19 June 2009; pp. 22–33. , ; , ; , ; p , g Q Q y Document Layout Analysis. In Proceedings of the Ircdl, Austin, TX, USA, 15–19 June 2009; pp. 22–33. 52. Leech, N.L.; Onwuegbuzie, A.J. Qualitative Data Analysis: A Compendium of Techniques and a School Psychology Research and Beyond. Sch. Psychol. Q. 2008, 23, 587. [CrossRef] 53. Susser, M. Health as a Human Right: An Epidemiologist’s Perspective on the Public Health. Am. J. Public Health 1993, 83, 418–426. [CrossRef] 54. Evans, T. A Human Right to Health? Third World Q. 2002, 23, 197–215. [CrossRef] 54. Evans, T. A Human Right to Health? Third World Q. 2002, 23, 197–215. [CrossRef] 54. Evans, T. A Human Right to Health? Third World Q. 2002, 23, 197–215. [CrossRef] 55. Thiele, B. The Human Right to Adequate Housing: A Tool for Promoting and Protecting Individual and Community Health. Am. 55. Thiele, B. The Human Right to Adequate Housing: A Tool for Promoting and Protecting Individual and J. Public Health 2002, 92, 712–715. [CrossRef] [PubMed] 56. Seale, R.G. A Perspective from Canada on Heritage and Tourism. Ann. Tourism Res. 1996, 23, 484–488. References Authenticity, Value and Community Involvement in Heritage Management under the World Heritage and Intangible Heritage Conventions. Herit. Soc. 2013, 6, 129–143. [CrossRef] 35. Hickel, J. The Contradiction of the Sustainable Development Goals: Growth Versus Ecology on a Finite Planet. Sustainable Dev. 2019, 27, 873–884. [CrossRef] 36. De Neve, J.; Sachs, J.D. The SDGs and Human Well-being: A Global Analysis of Synergies, Trade-Offs, and Regional Differences. Sci. Rep. 2020, 10, 1–12. [CrossRef] Pretty, J. The Health Benefits of Walking in Greenspaces of High Natural and Heritage Value. J. Integr. Environ . [CrossRef] 37. Barton, J.; Hine, R.; Pretty, J. The Health Benefits of Walking in Greenspaces of High Natural and Her Sci. 2009, 6, 261–278. [CrossRef] 38. Armaitiene, A.; Bertuzyte, R.; Vaskaitis, E. Conceptual Framework for Rethinking of Nature Heritage Management and Health Tourism in National Parks. Procedia-Soc. Behav. Sci. 2014, 148, 330–337. [CrossRef] 39. Grossi, E.; Blessi, G.T.; Sacco, P.L. Magic Moments: Determinants of Stress Relief and Subjective Wellbeing from Visiting a Cultural Heritage Site. Cult. Med. Psychiatry 2019, 43, 4–24. [CrossRef] [PubMed] g y y 40. Su, M.M.; Wall, G.; Xu, K. Heritage Tourism and Livelihood Sustainability of a Resettled Rural Community: Mount Sanqingshan World Heritage Site, China. J. Sustain. Tour. 2016, 24, 735–757. [CrossRef] 41. Su, M.M.; Wall, G.; Xu, K. Tourism-Induced Livelihood Changes at Mount Sanqingshan World Heritage Site, China. Environ. Manag. 2016, 57, 1024–1040. [CrossRef] 42. Chi, C.G.; Cai, R.; Li, Y. Factors Influencing Residents’ Subjective Well-being at World Heritage Sites. Tour. Manag. 2017, 63, 209–222. [CrossRef] 43. Grieves, V. Aboriginal Spirituality: Aboriginal Philosophy, the Basis of Aboriginal Social and Emotional Wellbei Centre for Aboriginal Health Darwin: Casuarina, Australia, 2009. 44. Dockery, A.M. Culture and Wellbeing: The Case of Indigenous Australians. Soc. Indicators Res. 2010, 99, 315–332. [CrossRef] 44. Dockery, A.M. Culture and Wellbeing: The Case of Indigenous Australians. Soc. Indicators Res. 2010, 99, 315–332. [CrossRef] 45. Prout, S. Indigenous Wellbeing Frameworks in Australia and the Quest for Quantification. Soc. Indicators Res. 2012, 109, 317–336. [CrossRef] 44. Dockery, A.M. Culture and Wellbeing: The Case of Indigenous Australians. Soc. Indicators Res. 2010, 99, 315–332. [CrossRef] 45. Prout, S. Indigenous Wellbeing Frameworks in Australia and the Quest for Quantification. Soc. Indicators Res. 2012, 109, 317–336. y g g 45. Prout, S. Indigenous Wellbeing Frameworks in Australia and the Quest for Quantification. Soc. Indicators Res. 2012, 109, 317–336. References Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2019. 73. UNESCO. WHC-99/CONF.209/22. In Proceedings of the Report of the Rapporteur of the 23rd Session of the Committee, Marrakesh, Morocco, 29 November–4 December 1999. 74. UNESCO. WHC-2000/CONF.204/21. In Proceedings of the Report of the Twenty-Fourth Session of the W Cairns, Australia, 27 November–2 December 2000. 75. UNESCO. WHC-12/36.COM/19. In Proceedings of the Decisions Adopted by the World Heritage Committee at Its 36th Session, Saint Petersburg, Russian Federation, 24 June–6 July 2012. g y 76. UNESCO. WHC-13/37.COM/20. In Proceedings of the Decisions Adopted by the World Heritage Committee at Its 37th Session, Phnom Penh, Cambodia, 16–27 June 2013. 77. UNESCO. WHC-14/38.COM/16. In Proceedings of the Decisions Adopted by the World Heritage Committee at Its 38th Session, Doha, Qatar, 15–25 June 2014. 78. UNESCO. WHC-15/39.COM/19. In Proceedings of the Decisions Adopted by the World Heritage Committee at Its 39th Session, Bonn, Germany, 28 June–8 July 2015. y y 79. UNESCO. WHC-16/40.COM/19. In Proceedings of the Report of the Decisions Adopted during the 4 Heritage Committee, Istanbul, Turkey, 10–17 July 2016. /40.COM/19. In Proceedings of the Report of the Decisions Adopted during the 40th Session of the World e, Istanbul, Turkey, 10–17 July 2016. g y y 80. UNESCO. WHC-18/42.COM/18. In Proceedings of the Decisions Adopted during the 42nd Session of the World Heritage Committee, Manama, Bahrain, 24 June–4 July 2018. y 81. Bortolotto, C. From Objects to Processes: UNESCO’S ‘Intangible Cultural Heritage’. J. Mus. Ethnogr. 2007, 21–33. Available online: https://www.jstor.org/stable/40793837 (accessed on 15 November 2020). 82. Logan, W.S. Closing Pandora’s box: Human rights conundrums in cultural heritage protection. In Cultural Heritage and Human Rights, Cultural heritage and human rights ed.; Silverman, H., Ruggles, D.F., Eds.; Springer: New York, NY, USA, 2007; pp. 33–52. [CrossRef] [ ] 83. CONF 002 XI. 1991. Preparation of the Celebration of the Twentieth Anniversary of the Adoption of the World Heritage Convention. In Proceedings of the 15th Session of the World Heritage Committee, Carthage, Tunisia, 9–13 December 1991. [ ] 83. CONF 002 XI. 1991. Preparation of the Celebration of the Twentieth Anniversary of the Adoption of the World Heritage Convention. In Proceedings of the 15th Session of the World Heritage Committee, Carthage, Tunisia, 9–13 December 1991. 84. 38 COM 5D. World Heritage and Sustainable Development. References [CrossRef] 56. Seale, R.G. A Perspective from Canada on Heritage and Tourism. Ann. Tourism Res. 1996, 23, 484–488. [CrossRef] 57. Bowen, G.A. Document Analysis as a Qualitative Research Method. Qual. Res. J. 2009. [CrossRef] 57. Bowen, G.A. Document Analysis as a Qualitative Research Method. Qual. Res. J. 2009. [CrossRef] 13 of 14 13 of 14 Sustainability 2021, 13, 7144 58. UNESCO. WHC-95/CONF.203/16. In Proceedings of the Report of the Rapporteur of the 19th Session of the Committee, Berlin, Germany, 4–9 December 1995; 1995. y 59. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 1994. g 60. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 1996. g 61. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 1997. 62. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 1999. g 63. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Heritage Centre: Paris, France, 2002. g 64. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2005. g 65. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2008. 66. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2011. g 67. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2012. g 68. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Heritage Centre: Paris, France, 2013. ritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World ris, France, 2013. g 69. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2015. g 70. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2016. g 71. UNESCO World Heritage Centre. Operational Guidelines for the Implementation of the World Heritage Convention; UNESCO World Heritage Centre: Paris, France, 2017. 72. UNESCO World Heritage Centre. References In Proceedings of the 38th Session of the World Heritage Committee, Doha Qatar 15 25 June 2014 g g g 84. 38 COM 5D. World Heritage and Sustainable Development. In Proceedings of the 38th Session of the World Heritage Committee, Doha, Qatar, 15–25 June 2014. 85. 43 COM 7B.30. Sangha Trinational (Cameroon, Central African Republic, Congo) (N 1380rev). In Proceedings of the 43rd Session of the World Heritage Committee, Baku, Republic of Azerbaijan„ 30 June–10 July 2019. 86. 39 COM 8B.5. Kaeng Krachan Forest Complex Thailand. In Proceedings of the 39th Session of the World Heritage Committee, Bonn, Germany, 28 June–8 July 2015. 14 of 14 14 of 14 Sustainability 2021, 13, 7144 87. 40 COM 8B.11. Examination of Nominations of Natural Properties to the World Heritage List. In Proceedings of the 40th Session of the World Heritage Committee, Istanbul, Turkey, 10–17 July 2016. Paris, France, 24–26 October 2016. g y J y 88. Meskell, L. The Rush to Inscribe: Reflections on the 35th Session of the World Heritage Committee, UNESCO Paris, 2011. J. Field Archaeol. 2012, 37, 145–151. [CrossRef] 89. Disko, S. Indigenous Peoples’ Rights and the World Heritage Convention. In A Companion to Heritage Studies; Logan, W., Craith, M.N., Kockel, U., Eds.; Wiley Online Library: Hoboken, NJ, USA, 2015; pp. 355–372. [CrossRef] Disko, S. Indigenous Peoples’ Rights and the World Heritage Convention. In A Companion to Heritage Stud M N Kockel U Eds ; Wiley Online Library: Hoboken NJ USA 2015; pp 355 372 [CrossRef] Disko, S. Indigenous Peoples Rights and the World Heritage Convention. In A Companion to Heritage Stud M.N., Kockel, U., Eds.; Wiley Online Library: Hoboken, NJ, USA, 2015; pp. 355–372. [CrossRef] , g p g g p g ; g Kockel, U., Eds.; Wiley Online Library: Hoboken, NJ, USA, 2015; pp. 355–372. [CrossRef] 90. Cameron, C. UNESCO and Cultural Heritage: Unexpected Consequences. In A Companion to Heritage Studies; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2016; pp. 323–336. [CrossRef] 91. 35 COM 8B.43. Cultural Properties—Coffee Cultural Landscape of Colombia (Colombia). In Proceedings of the 35th Session of the World Heritage Committee, Paris, France, 19–29 June 2011. 92. 35 COM 8B.46. Natural Properties—Examination of Minor Boundary Modifications—Selous Game Res Tanzania). In Proceedings of the 35th Session of the World Heritage Committee, Paris, France, 19–29 Ju 93. 39 COM 8B.38. Aqueduct of Padre Tembleque Hydraulic System, Mexico. References In Proceedings of the 39th Session of the World Heritage Committee, Bonn, Germany, 28 June–8 July 2015. 94. 40 COM 8B.17. Examination of Nominations of Mixed Properties to the World Heritage List. In Proceedings of the 40th Session of the World Heritage Committee, Istanbul, Turkey, 10–17 July 2016. Paris, France, 24–26 October 2016. 95. 41 COM 8B.17. Historic City of Ahmadabad (India). In Proceedings of the 41st Session of the World Heritage Committee, Krakow, Poland, 2–12 July 2017. 96. 42 COM 7B.91. Simien National Park (Ethiopia) (N9). In Proceedings of the 42nd Session of the World Heritage Committee, Manama, Bahrain, 24 June–4 July 2018. 97. Nocca, F. The Role of Cultural Heritage in Sustainable Development: Multidimensional Indicators a Sustainability 2017, 9, 1882. [CrossRef] of Cultural Heritage in Sustainable Development: Multidimensional Indicators as Decision-Making Tool 9, 1882. [CrossRef] y 98. Su, M.M.; Wall, G. Global–local Relationships and Governance Issues at the Great Wall World Heritage Site, China. J. Sustain. Tour. 2012, 20, 1067–1086. [CrossRef] 99. Osipova, E.; Wilson, L.; Blaney, R.; Shi, Y.; Fancourt, M.; Strubel, M.; Salvaterra, T.; Brown, C.; Verschuuren, B. The Benefits of Natural World Heritage: Identifying and Assessing Ecosystem Services and Benefits Provided by the World’s most Iconic Natural Places; IUCN: Gland, Switzerland, 2014. 100. Ripp, M.; Rodwell, D. Governance in UNESCO World Heritage Sites: Reframing the role of management plans as a tool to improve community engagement. In Aspects of Management Planning for Cultural World Heritage Sites; Makuvaza, S., Ed.; Springer: Berlin/Heidelberg, Germany, 2018; pp. 241–253. [CrossRef] 101. Bertacchini, E.; Liuzza, C.; Meskell, L. Shifting the Balance of Power in the UNESCO World Heritage Committee: An Empirical Assessment. Int. J. Cult. Policy 2017, 23, 331–351. [CrossRef]
https://openalex.org/W3120220474
https://europepmc.org/articles/pmc7826896?pdf=render
English
null
Using a B-Phycoerythrin Extract as a Natural Colorant: Application in Milk-Based Products
Molecules/Molecules online/Molecules annual
2,021
cc-by
11,340
  Keywords: B-phycoerythrin; colorimetry; milks; natural food colorants; color stability Citation: García, A.B.; Longo, E.; Murillo, M.C.; Bermejo, R. Using a B-Phycoerythrin Extract as a Natural Colorant: Application in Milk-Based Products. Molecules 2021, 26, 297. https://doi.org/10.3390/ molecules26020297 Article Using a B-Phycoerythrin Extract as a Natural Colorant: Application in Milk-Based Products Ana Belén García 1, Eleonora Longo 2, Mª Carmen Murillo 1 and Ruperto Bermejo 1,* 1 Department of Physical and Analytical Chemistry, High Engineering Polytechnic School of Linares, University of Jaen, 23700 Linares, Spain; abgr0013@red.ujaen.es (A.B.G.); mcmc0031@red.ujaen.es (M.C.M.) 2 Department of Pharmaceutical and Pharmacological Sciences, University of Padova, 35131 Padova, Italy; eleonora.longo@studenti.unipd.it * Correspondence: rbermejo@ujaen es 1 Department of Physical and Analytical Chemistry, High Engineering Polytechnic School of Linares, University of Jaen, 23700 Linares, Spain; abgr0013@red.ujaen.es (A.B.G.); mcmc0031@red.ujaen.es (M.C.M.) 2 Department of Pharmaceutical and Pharmacological Sciences, University of Padova, 35131 Padova, Italy; eleonora.longo@studenti.unipd.it University of Jaen, 23700 Linares, Spain; abgr0013@red.ujaen.es (A.B.G.); mcmc0031@red.ujaen.es (M.C.M.) 2 Department of Pharmaceutical and Pharmacological Sciences, University of Padova, 35131 Padova, Italy; eleonora.longo@studenti.unipd.it g p * Correspondence: rbermejo@ujaen.es Abstract: Nowadays, there is a growing interest in finding new coloring molecules of natural origin that can increase and diversify the offer of natural food dyes already present in the market. In the present work, a B-phycoerythrin extract from the microalgae Porphyridium cruentum was tested as a food colorant in milk-based products. Using spectroscopy and colorimetry, the extract was character- ized and gave evidence of good properties and good stability in the pH range between 4.0 and 9.0. Coloring studies were conducted to demonstrate that samples carrying the pink extract could be used for simulating the pink color of marketed milk-based products. The staining factors, representing the amount of pink protein to be added to reproduce the color of strawberry commercial products, ranged between 1.6 mg/L and 49.5 mg/L, being sufficiently low in all samples. Additionally, color stability during a short period of cold storage was studied: it demonstrated that the three tested types of dairy products remained stable throughout the 11-day analysis period with no significant changes. These results prove the potential of the B-phycoerythrin extract as a natural colorant and alternative ingredient to synthetic coloring molecules. molecules molecules molecules 1. Introduction Natural colorants are derived from plants, fruits, or animals and are then extracted and purified. Natural colorants might be sensitive to light exposure, might change color when exposed to different pH, and degrades at high temperatures, and they are generally available in a limited color range. All of these factors limit the use of natural colorants at the industry level. Generally, consumers are more skeptical towards the use of synthetic colorants and prefer natural food dyes for their potential benefits [9]. Therefore, there is a need for new natural colorants for use in food to extend the coloring range of natural food dyes. If possible, food dyes should be extracted from plants or fruits in order to be suitable for all diets. Several investigations have been made in recent years regarding the use of natural col- orants in food and drinks, which include studying aspects of preparation, characterization, application, stability, kinetics, and color evaluation [10–14]. In this context, milk-based products (yogurts, liquid yogurts, and milkshakes) are consumed throughout the world due to their organoleptic and nutritional value. The incorporation of natural colorants to these products can give a better alternative to the use of synthetic molecules by appealing to the growing health-related awareness among consumers [15–20]. Other researchers have reported works involving the use of natural colorants in bakery products, soft drinks, and candies [21–25]. Phycobiliproteins have been incorporated in desserts, puddings, sour- dough, soft drinks, and dairy products, although their use needs to be further supported by toxicity testing. Moreover, additional research focusing on bioavailability and on the interaction of phycobiliproteins with food, beverages, matrixes, and stabilizers among others is required [26–29]. q In light of the above, we have focused our attention on using a phycoerythrin extract (B-PE extract) as a natural pink colorant, searching for new molecules which can allow for expansion of the natural colorants offered. Another pivotal reason for trying to exploit this dye is its antioxidant effect described in several publications, which adds value to the milk-based products obtained [7,30–32]. Phycoerythrin is attractive as an antioxidant and anti-inflammatory drug, having properties to protect against physiological changes under oxidative stress and great effects against aging. Additionally, this compound has fewer side effects than chemical drugs, and algae are eco-friendly sources of this compound and are receiving increasing attention. 1. Introduction Rhodophyta (or red algae) are microalgae particularly exploited for potential applica- tion of their components. Porphyridium cruentum (P. cruentum) is a red algae from which B-phycoerythrin (B-PE) is obtained. B-PE is a phycobiliprotein in which a phycoerythro- bilin chromophore is linked to the protein by a thioether linkage [1]. Phycobiliproteins are water-soluble and colored proteins present in microalgae that play an important role in photosynthesis [2]. Phycobiliproteins are organized in phycobilisomes on the outer surface of thylakoid membranes, and their role is to absorb light in regions where chlorophyll does not, transferring energy to the photochemical reaction centers. In the reaction centers, the electronic energy is then transformed into chemical energy. B-PE is a αβ heterodimer with chains of α-helix structure. Recently, B-PE has gained interest for its application as a fluorescent probe in research and clinical diagnosis and as natural pink colorant in food, fabrics, cosmetics, and drugs [3–5]. Only a few studies have used phycoerythrin directly in food application [6,7]. Academic Editor: Peter Karuso Received: 8 December 2020 Accepted: 5 January 2021 Published: 8 January 2021 Academic Editor: Peter Karuso Received: 8 December 2020 Accepted: 5 January 2021 Published: 8 January 2021 Publisher’s Note: MDPI stays neu- tral with regard to jurisdictional clai- ms in published maps and institutio- nal affiliations. pp Coloring additives are dyes that confer color to food, drinks, or drugs to which they are added. Color is a driving feature in the choice of food and can deeply condition consumers on what to buy. According to the European Food Safety Authority (EFSA), food colorants are one category of food additives that is used in the food industry to render processed foods more attractive, to restore the color of food after processing, and to give color to otherwise colorless food. Food colorants can be divided in two groups depending on their origin: synthetic and natural colorants [8]. Natural colorants have gained more interest among consumers who demand natural products. In line with an Copyright: © 2021 by the authors. Li- censee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and con- ditions of the Creative Commons At- tribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/molecules Molecules 2021, 26, 297. https://doi.org/10.3390/molecules26020297 Molecules 2021, 26, 297 2 of 13 increasing trend of healthy diets and lifestyle, consumers have shifted their preference from synthetic additives towards natural ones. 1. Introduction Hence, the outcome of this study would enhance the organoleptic and nutritional properties of milk-based products using a B-PE extract from red algae. g In the present work, a B-Phycoerythrin extract (B-PE extract) from the red algae P. cruentum was studied as a natural colorant in different milk-based products. This study represents a first approach that aims to characterize the B-PE extract by spectroscopy in order to understand the coloring properties of the protein solution. Additionally, a colorimetric study was developed for the color evaluation of samples containing the natural protein extract which reproduce the color of commercially available products. 2. Results and Discussion 2.1. Characterization of B-Phycoerythrin Extract 2.1. Characterization of B-Phycoerythrin Extract The B-PE extract from P. cruentum is composed of B-PE, R-phycocyanin (R-PC), and allophycocyanin (APC) [33]. Preliminary analysis for the estimation of phycobiliprotein content in the extract revealed that this colorant extract solution contained 0.8 mg B-PE/mL as the major component (93% w/w). Other proteins are present in a significantly minor proportion, such as APC (4% w/w) and R-PC (3% w/w). The B-phycoerythrin purity ratio (A545/A280) was 2, which represents a more than sufficient purity grade for its use as a colorant in the food sector. The absorbance spectrum of the coloring extract shows two main peaks at 545 and 565 nm (Figure 1a), which correspond to B-PE being the main component present in the extract. The absorbance ratio (A545/A280) was used as a purity indicator. When its value is lower than 4, it indicates the presence of other proteins and that it is not a pure B- Molecules 2021, 26, 297 3 of 13 5 and PE solution. In this case, R-PC and APC are present in a minor proportion, not having significant repercussions on the final pink color of the B-PE extract. lower than 4, it indicates the presence of other proteins and that it is not a pure B-PE so- lution. In this case, R-PC and APC are present in a minor proportion, not having signifi- cant repercussions on the final pink color of the B-PE extract. PE solution. In this case, R-PC and APC are present in a minor proportion, not having significant repercussions on the final pink color of the B-PE extract. lower than 4, it indicates the presence of other proteins and that it is not a pure B-PE so- lution. In this case, R-PC and APC are present in a minor proportion, not having signifi- cant repercussions on the final pink color of the B-PE extract. Figure 1. (a) Absorption and fluorescence spectra of the B-phycoerythrin (B-PE) extract from P. cruentum at pH = 7.0, and (b) variation of absorbance (■) (at 545 nm) and relative fluorescence in- tensity (▲) (at 575 nm) of the B-PE extract solution at various pH values in 20 mM phosphate buffer: for absorbance and fluorescence measurements, the protein concentration was 19·10−3 mg B-PE/mL. 2.1. Characterization of B-Phycoerythrin Extract 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0 100 200 300 400 500 600 700 800 250 350 450 550 650 FI A Wavelength (nm) FI A 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0 100 200 300 400 500 600 700 800 3 4 5 6 7 8 9 10 11 12 13 FI A pH FI A (a) (b) Figure 1. (a) Absorption and fluorescence spectra of the B-phycoerythrin (B-PE) extract from P. cruen- tum at pH = 7.0, and (b) variation of absorbance (■) (at 545 nm) and relative fluorescence intensity (▲) (at 575 nm) of the B-PE extract solution at various pH values in 20 mM phosphate buffer: for absorbance and fluorescence measurements, the protein concentration was 19·10−3 mg B-PE/mL. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0 100 200 300 400 500 600 700 800 250 350 450 550 650 FI A Wavelength (nm) FI A (a) (a) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0 100 200 300 400 500 600 700 800 3 4 5 6 7 8 9 10 11 12 13 FI A pH FI A (b) (b) Figure 1. (a) Absorption and fluorescence spectra of the B-phycoerythrin (B-PE) extract from P. cruentum at pH = 7.0, and (b) variation of absorbance (■) (at 545 nm) and relative fluorescence in- tensity (▲) (at 575 nm) of the B-PE extract solution at various pH values in 20 mM phosphate buffer: for absorbance and fluorescence measurements, the protein concentration was 19·10−3 mg B-PE/mL. Figure 1. (a) Absorption and fluorescence spectra of the B-phycoerythrin (B-PE) extract from P. cruen- tum at pH = 7.0, and (b) variation of absorbance (■) (at 545 nm) and relative fluorescence intensity (▲) (at 575 nm) of the B-PE extract solution at various pH values in 20 mM phosphate buffer: for absorbance and fluorescence measurements, the protein concentration was 19·10−3 mg B-PE/mL. The protein extract obtained in this study is mostly constituted by B-PE thanks to the removal of contaminant proteins by precipitation with ammonium sulphate. A small shoulder is observed at 620 nm, which corresponds to the R-PC maximum. However, no peak corresponding to APC is observed at 650 nm. Therefore, B-phycoerythrin is the main constituent of the extract, and it confers the characteristic pink color to the extract. Regard- ing fluorescence, the corresponding spectrum (Figure 1b) shows a maximal fluorescence intensity emission at 575 nm. 2.1. Characterization of B-Phycoerythrin Extract Both absorption and fluorescence spectra for the B-PE extract demonstrate that B-phycoerythrin is the main protein found and therefore the predomi- nant one in the extract. The other two phycobiliproteins (APC and R-PC) appear in a small i The protein extract obtained in this study is mostly constituted by B-PE thanks to the removal of contaminant proteins by precipitation with ammonium sulphate. A small shoulder is observed at 620 nm, which corresponds to the R-PC maximum. However, no peak corresponding to APC is observed at 650 nm. Therefore, B-phycoerythrin is the main constituent of the extract, and it confers the characteristic pink color to the extract. Regarding fluorescence, the corresponding spectrum (Figure 1b) shows a maximal fluorescence intensity emission at 575 nm. Both absorption and fluorescence spectra for the B-PE extract demonstrate that B-phycoerythrin is the main protein found and therefore the predominant one in the extract. The other two phycobiliproteins (APC and R-PC) appear in a small proportion. proportion. The pH is a relevant factor that influences the stability of phycobiliproteins. UV-Vis spectroscopy and fluorescence spectroscopy were used to determine the stability of the pink B-PE extract and its optimal pH range of application (Figure 1b). The milk-based products utilized in this work presented pH values of 6.71, 4.21, and 4.16 for milk, liquid p p The pH is a relevant factor that influences the stability of phycobiliproteins. UV-Vis spectroscopy and fluorescence spectroscopy were used to determine the stability of the pink B-PE extract and its optimal pH range of application (Figure 1b). The milk-based products utilized in this work presented pH values of 6.71, 4.21, and 4.16 for milk, liquid yogurt, and yogurt, respectively. Accordingly, it is necessary to know the behavior of the colorant extract inside alimentary samples presenting these pH values. Molecules 2021, 26, 297 4 of 13 When the coloring extract is mixed with phosphate buffer at pH values below 3, proteins aggregates form immediately. This is explained by the precipitation of proteins due to the very acidic pH of the solution. A similar behavior was obtained when the B-PE extract was tested in alkaline solutions at pH values above 11: proteins are unstable, and the spectrum of the extract does not retain any shape, being completely disrupted (data not shown). Therefore, the B-PE coloring extract is spectroscopically measured in the range from pH 3.80 to pH 11.0. 2.1. Characterization of B-Phycoerythrin Extract Every extract sample presents two characteristic picks at 545 and 565 nm, and a third pick is present at 280 nm (proteins total peak). The absorbance spectrum preserves its characteristic profile when the protein extract is in the pH range 4.0–8.0. B-PE was excited at 465 nm to record the fluorescence spectra of the protein extract at different pH (Figure 1b). At neutral pH, the fluorescence maximum is registered at 575 nm. The spectrum maintains its profile and registers high values of fluorescence intensity (FI) between pH 6.0 and pH 9.0. The intensity of the peak decreases at pH lower than 6 and higher than 9. These results are in accordance with absorbance determinations showing instability in highly acidic or basic media. y g y Additionally, the B-PE extract was analyzed to obtain color parameters according to the Color Space defined by the International Commission on Illumination (CIELAB-color system: L*, a* and b* values) as a function of pH. The results are summarized in Table 1. Therefore, variations in pH values did not cause important variations in the L*, a*, or b* indexes, showing the stability of the lightness, redness, and yellowness of B-PE extract. This observation is crucial because the solution maintains an attractive pink color in the pH range assayed. Table 1. Color parameters (CIELAB-color system) for the B-PE extract at different pH values (media ± SD). L*, a* and b* are the three CIELAB values to defining the color. PH L*(D65) a*(D65) b*(D65) 3 97.85 ± 0.02 4.28 ± 0.02 −2.82 ± 0.02 4 97.85 ± 0.02 4.37 ± 0.03 −2.75 ± 0.02 5 97.80 ± 0.02 4.54 ± 0.05 −2.83 ± 0.03 6 97.80 ± 0.03 4.49 ± 0.05 −2.77 ± 0.03 7 97.81 ± 0.02 4.58 ± 0.06 −2.77 ± 0.04 8 97.88 ± 0.15 4.39 ± 0.30 −2.55 ± 0.15 9 97.94 ± 0.04 4.33 ± 0.09 −2.48 ± 0.05 It is important to emphasize that the B-PE extract keeps color parameters unchanged in the pH range assayed and that, when the protein extract is added, coloring parame- ters are maintained highly similar in the three different groups of milk-based products selected. Therefore, the stability of the redness for the B-PE extract is related to a higher pH stability for B-phycoerythrin biliprotein. 2.1. Characterization of B-Phycoerythrin Extract Previous works have studied the pH-induced conformational and functional dynamics of B-phycoerythrin isolated from different red microalgae by means of absorption, fluorescence, and other common techniques [33,34]. B- phycoerythrin has a strong functional stability in the pH range 3.5–10, with no remarkable changes in absorbance or fluorescence, suggesting the ability of stable energy transfer or the protein even at low pH values. p p In light of the above and taking into account the characterization carried out using absorbance, fluorescence, and colorimetry, applications of this pink B-PE extract may be useful in acidic and neutral food samples, such as in milk-based products. 2.2. Milk-Based Products Pigmentation with B-PE Extract Product Commercial Brand Extract Volume Added (µL) ∆E*ab Reached Staining Factor (mg/L) Strawberry Milkshakes Extract Concentration 1.12 mg B-PE/mL A 205 4.93 16.12 B 260 4.77 20.36 C 650 4.73 49.55 D 410 5.00 31.78 E 390 4.87 30.27 F 60 4.43 4.77 Strawberry Liquid Yogurts Extract Concentration 1.12 mg B-PE/mL G 90 4.85 7.15 H 50 4.89 3.98 I 65 5.45 5.17 J 50 5.53 3.98 K 1120 4.87 82.90 Strawberry yogurts Extract concentration 0.8 mg B-PE/mL L 325 4.82 17.97 LL 180 4.59 10.05 M 285 4.95 15.80 N 170 4.95 9.50 In global terms, liquid yogurts required less B-PE extract volume than milkshakes and semisolid yogurts to reproduce the colors of commercial products. The products that required the lowest amount of protein were “H” and “J” liquid yogurts, which obtained staining factors of 3.98 mg/L. On the contrary, the product that required the highest amount of protein was “K” liquid yogurt, which achieved a staining factor of 82.9 mg/L. These results show the potential of the B-PE extract as a natural colorant and its potential economic viability due to the low values of the obtained staining factors. The progressive addition of the B-PE extract to standard commercial products (white products) promotes the gradual decrease of the ∆E*ab value as the color of the standard progressively approaches the color of the commercial product. The final sample is the one in which the lowest ∆E*ab value is reached (values close to ∆E*ab = 4). It is important to note that, when a ∆E*ab value lower than 5 is reached, the color of the sample is nearly the same as that of the reference when viewed by the naked human eye [15]. Figure 2 shows a vivid presentation of the color samples obtained using the B-PE extract compared to the reference product found in the market. Three representative samples were selected as examples of the milk-based products assayed. As previously commented, the color of each sample was very close to the color of its reference commercial product. tors for different milk-based products: these data are obtained according to the coloring curves obtained gures 2–4) Table 2. Staining factors for different milk-based products: these data are obtained according to the coloring curves obtained for each product (Figures 2–4). 2.2. Milk-Based Products Pigmentation with B-PE Extract According to the ingredients described on the labels of pink milk-based products found in the market, the most commonly used synthetic dyes are Ponceau 4R (FD&C Red n◦1), Bordeaux (FD&C Red nº9), and Allura Red (E-129). Regarding natural molecules, the only one that is usually found in dairy products is carminic acid (E-120) [3,9,16]. Molecules 2021, 26, 297 5 of 13 Therefore, the color given by the pink B-PE extract to standard products is compared to the color of reference products found in grocery stores which contain other types of colorants, synthetic or natural. It is important to highlight that the current trend in the food industry for human consumption is favoring the use of natural colorants instead of those produced by chemical synthesis. In coloring studies, the standard product was colored with the pink B-PE extract to achieve a color most similar to the one of the reference product. From coloring assays, a coloring curve is obtained: it plots the difference in color (∆E*ab) between the sample and the reference against the volume of B-PE extracts added to the standard product (µL). In order to match the color of marketed products in the three types of milk-based products, it was necessary to add a specific volume of B-PE extract to 14 ml of the milk-based product (Table 2). Because the amount of added protein was kept at <50 mg/L in all experiments, its effect on texture and viscosity was negligible, and these results are in agreement with previous studies [15]. its effect on texture and viscosity was negligible, and these results are in agreement with previous studies [15]. Table 2. Staining factors for different milk-based products: these data are obtained according to the coloring curves obtained for each product (Figures 2–4). 2.2. Milk-Based Products Pigmentation with B-PE Extract Photographs showing a vivid presentation of the market products and milk-based products obtained using the B-PE extract as a natural colorant: in this figure, three representative samples are shown: sample A, representing milkshakes (MS-A); sample G, representing liquid yogurts (LY-G); and sample LL, representing yogurts (Y-LL). In each pair of photos, on the left is the reference product found in the market and on the right is the sample obtained using the B-PE extract. of commercial products. Staining factors for each sample were calculated as the amount of B-PE protein needed for each liter (of product) to obtain the same color as of the reference commercial product (Table 2). Generally, sample staining factors are proving that good colors can be obtained with a fairly low quantity of the B-PE colo extract. 2.2.1. Milkshakes Six different commercial milkshakes were studied to test the potential coloring pacity of the B-PE extract (Table 2). Thus, as most of the final ΔE*ab values reached w below or very close to 5 (Figure 3), coloring curves obtained from milkshakes suggest t the B-PE coloring extract has good coloring properties. Milkshake “C” registered the highest staining factor (Table 2), and its coloring cu suggests that a large volume of the B-PE extract (650 μL) was necessary to obtain the fi sample (Figure 3). On the other hand, milkshake “F” required the lowest volume of B coloring solution (60 μL) to achieve the desired color. Differently, other milksha showed intermediate volumes of the coloring extract to reach colors very close to the o of commercial products. Staining factors for each sample were calculated as the to amount of B-PE protein needed for each liter (of product) to obtain the same color as t of the reference commercial product (Table 2). Generally, sample staining factors are lo proving that good colors can be obtained with a fairly low quantity of the B-PE color extract. 40 0 5 10 15 20 25 30 35 40 0 100 200 300 400 500 600 700 ∆E*ab Extract volume (μL) A B C D E F EVIEW 7 With the exception of product “K”, the rest of the samples required very small ex volumes for achieving the desired color. Thus, staining factors were between 3.98 and 2.2.1. Milkshakes Six different commercial milkshakes were studied to test the potential coloring ca- pacity of the B-PE extract (Table 2). 2.2. Milk-Based Products Pigmentation with B-PE Extract Product Commercial Brand Extract Volume Added (µL) ∆E*ab Reached Staining Factor (mg/L) Strawberry Milkshakes Extract Concentration 1.12 mg B-PE/mL A 205 4.93 16.12 B 260 4.77 20.36 C 650 4.73 49.55 D 410 5.00 31.78 E 390 4.87 30.27 F 60 4.43 4.77 Strawberry Liquid Yogurts Extract Concentration 1.12 mg B-PE/mL G 90 4.85 7.15 H 50 4.89 3.98 I 65 5.45 5.17 J 50 5.53 3.98 K 1120 4.87 82.90 Strawberry yogurts Extract concentration 0.8 mg B-PE/mL L 325 4.82 17.97 LL 180 4.59 10.05 M 285 4.95 15.80 N 170 4.95 9.50 Table 2. Staining factors for different milk-based products: these data are obtained according to the coloring curves obtained for each product (Figures 2–4). In global terms, liquid yogurts required less B-PE extract volume than milkshakes and semisolid yogurts to reproduce the colors of commercial products. The products that required the lowest amount of protein were “H” and “J” liquid yogurts, which obtained staining factors of 3.98 mg/L. On the contrary, the product that required the highest amount of protein was “K” liquid yogurt, which achieved a staining factor of 82.9 mg/L. These results show the potential of the B-PE extract as a natural colorant and its potential economic viability due to the low values of the obtained staining factors. The progressive addition of the B-PE extract to standard commercial products (white products) promotes the gradual decrease of the ∆E*ab value as the color of the standard progressively approaches the color of the commercial product. The final sample is the one in which the lowest ∆E*ab value is reached (values close to ∆E*ab = 4). It is important to note that, when a ∆E*ab value lower than 5 is reached, the color of the sample is nearly the same as that of the reference when viewed by the naked human eye [15]. Figure 2 shows a vivid presentation of the color samples obtained using the B-PE extract compared to the reference product found in the market. Three representative samples were selected as examples of the milk-based products assayed. As previously commented, the color of each sample was very close to the color of its reference commercial product. 6 of 13 Molecules 2021, 26, 297 Figure 2. 2.2. Milk-Based Products Pigmentation with B-PE Extract Thus, as most of the final ΔE*ab values reached were below or very close to 5 (Figure 3), coloring curves obtained from milkshakes suggest that the B-PE coloring extract has good coloring properties. Milkshake “C” registered the highest staining factor (Table 2), and its coloring curve suggests that a large volume of the B-PE extract (650 μL) was necessary to obtain the final sample (Figure 3). On the other hand, milkshake “F” required the lowest volume of B-PE coloring solution (60 μL) to achieve the desired color. Differently, other milkshakes showed intermediate volumes of the coloring extract to reach colors very close to the ones of commercial products. Staining factors for each sample were calculated as the total amount of B-PE protein needed for each liter (of product) to obtain the same color as that of the reference commercial product (Table 2). Generally, sample staining factors are low, proving that good colors can be obtained with a fairly low quantity of the B-PE coloring extract. 35 40 Figure 3. Coloring curves of milkshakes resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). 0 5 10 15 20 25 30 35 40 0 100 200 300 400 500 600 700 ∆E*ab Extract volume (μL) A B C D E F Figure 3. Coloring curves of milkshakes resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). EVIEW 7 of 13 With the exception of product “K”, the rest of the samples required very small extract volumes for achieving the desired color. Thus, staining factors were between 3.98 and 7.15 mg/L from products “J” and “G”, respectively. As an exception, the staining factor of “K” liquid yogurt was quite high (82.9), being 21 times higher than the staining factor of “H” liquid yogurt. 35 Figure 3. Coloring curves of milkshakes resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). Figure 3. Coloring curves of milkshakes resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). mg/L from products J and G , respectively. As an exception, the staining factor of K liquid yogurt was quite high (82.9), being 21 times higher than the staining factor of “H” liquid yogurt. Figure 3. 2.2. Milk-Based Products Pigmentation with B-PE Extract Photographs showing a vivid presentation of the market products and milk-based products obtained using the B-PE extract as a natural colorant: in this figure, three representative samples are shown: sample A, representing milkshakes (MS-A); sample G, representing liquid yogurts (LY-G); and sample LL, representing yogurts (Y-LL). In each pair of photos, on the left is the reference product found in the market and on the right is the sample obtained using the B-PE extract. Figure 2. Photographs showing a vivid presentation of the market products and milk-based products obtained using the B-PE extract as a natural colorant: in this figure, three representative samples are shown: sample A, representing milkshakes (MS-A); sample G, representing liquid yogurts (LY-G); and sample LL, representing yogurts (Y-LL). In each pair of photos, on the left is the reference product found in the market and on the right is the sample obtained using the B-PE extract. pacity of the B-PE extract (Table 2). Thus, as most of the final ΔE*ab values reached w below or very close to 5 (Figure 3), coloring curves obtained from milkshakes suggest t the B-PE coloring extract has good coloring properties. Milkshake “C” registered the highest staining factor (Table 2), and its coloring cu suggests that a large volume of the B-PE extract (650 μL) was necessary to obtain the fi sample (Figure 3). On the other hand, milkshake “F” required the lowest volume of B coloring solution (60 μL) to achieve the desired color. Differently, other milksha showed intermediate volumes of the coloring extract to reach colors very close to the o of commercial products. Staining factors for each sample were calculated as the to amount of B-PE protein needed for each liter (of product) to obtain the same color as t of the reference commercial product (Table 2). Generally, sample staining factors are l proving that good colors can be obtained with a fairly low quantity of the B-PE color extract. Figure 2. Photographs showing a vivid presentation of the market products and milk-based products obtained using the B-PE extract as a natural colorant: in this figure, three representative samples are shown: sample A, representing milkshakes (MS-A); sample G, representing liquid yogurts (LY-G); and sample LL, representing yogurts (Y-LL). In each pair of photos, on the left is the reference product found in the market and on the right is the sample obtained using the B-PE extract. Figure 2. 2.2. Milk-Based Products Pigmentation with B-PE Extract Coloring curves of milkshakes resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). 2.2.2. Liquid Yogurts In the same way, five different commercial liquid yogurts were studied for testing the potential coloring capacity of the B-PE extract. In this case, coloring curves obtained from milkshakes reflect a wide variety of results (Figure 4). 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 ∆E*ab Extract volume (μL) A B C D E F 2.2.2. Liquid Yogurts In the same way, five different commercial liquid yogurts were studied for testing the potential coloring capacity of the B-PE extract. In this case, coloring curves obtained from milkshakes reflect a wide variety of results (Figure 4). Figure 4. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). 2 2 3 Semisolid Yogurts 0 5 10 15 20 25 30 35 40 45 0 200 400 600 800 1000 1200 ∆E*ab Extract volume (μL) G H I J K Figure 4. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). Figure 3. Coloring curves of milkshakes resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). 2.2.2. Liquid Yogurts In the same way, five different commercial liquid yogurts were studied for testing the potential coloring capacity of the B-PE extract. In this case, coloring curves obtained f ilk h k fl t id i t f lt (Fi 4) 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 ∆E*ab Extract volume (μL) A B C D E F 2.2.2. Liquid Yogurts In the same way, five different commercial liquid yogurts were studied for testing the potential coloring capacity of the B-PE extract. In this case, coloring curves obtained from milkshakes reflect a wide variety of results (Figure 4). Figure 4. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). 0 5 10 15 20 25 30 35 40 45 0 200 400 600 800 1000 1200 ∆E*ab Extract volume (μL) G H I J K Figure 4. 2.2.2. Liquid Yogurts 0 0 200 In the same way, five different commercial liquid yogurts were studied for testing the potential coloring capacity of the B-PE extract. In this case, coloring curves obtained from milkshakes reflect a wide variety of results (Figure 4). Figure 4. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). Extract volume (μL) With the exception of product “K”, the rest of the samples required very small extract volumes for achieving the desired color. Thus, staining factors were between 3.98 and 7.15 mg/L from products “J” and “G”, respectively. As an exception, the staining factor of “K” liquid yogurt was quite high (82.9), being 21 times higher than the staining factor of “H” liquid yogurt. 2.2.3. Semisolid Yogurts Finally, four different pink semisolid yogurts were found in the market and assayed to test the potential coloring capacity of the B-PE extract (Figure 5). Coloring curves of all assayed semisolid yogurts showed excellent ΔE*ab values. All 2.2.1. Milkshakes 45 Six different commercial milkshakes were studied to test the potential coloring capacity of the B-PE extract (Table 2). Thus, as most of the final ∆E*ab values reached were below or very close to 5 (Figure 3), coloring curves obtained from milkshakes suggest that the B-PE coloring extract has good coloring properties. 30 35 40 Milkshake “C” registered the highest staining factor (Table 2), and its coloring curve suggests that a large volume of the B-PE extract (650 µL) was necessary to obtain the final sample (Figure 3). On the other hand, milkshake “F” required the lowest volume of B-PE coloring solution (60 µL) to achieve the desired color. Differently, other milkshakes showed intermediate volumes of the coloring extract to reach colors very close to the ones of commercial products. Staining factors for each sample were calculated as the total amount of B-PE protein needed for each liter (of product) to obtain the same color as that of the reference commercial product (Table 2). Generally, sample staining factors are low, proving that good colors can be obtained with a fairly low quantity of the B-PE coloring extract. 5 10 15 20 25 30 ∆E*ab G H I J K 2.2. Milk-Based Products Pigmentation with B-PE Extract Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P cruentum ([B PE] = 1 12 mg B PE/mL) In the same way, five different commercial liquid yogurts were studied for testing the potential coloring capacity of the B-PE extract. In this case, coloring curves obtained from milkshakes reflect a wide variety of results (Figure 4). Figure 4. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). Figure 4. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 1.12 mg B-PE/mL). 7 of 13 r of K of “H” Molecules 2021, 26, 297 2.2.1. Milkshakes 45 2.2.3. Semisolid Yogurts final ΔE*ab values were was highly similar to th Finally, four different pink semisolid yogurts were found in the market and assayed to test the potential coloring capacity of the B-PE extract (Figure 5). was highly similar to the color of its reference commercial product. Overall, staining fac- tors were low, with sample “N” requiring the lowest amount of extract and sample “L” requiring the highest (Table 2). to test the potential coloring capacity of the B-PE extract (Figure 5). p q g p requiring the highest (Table 2). Figure 5. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 0.8 mg B-PE/mL). 0 5 10 15 20 25 30 35 40 0 50 100 150 200 250 300 350 ΔE*ab Extract volume (μL) L LL N M Figure 5. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 0.8 mg B-PE/mL). Figure 5. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 0.8 mg B-PE/mL). Figure 5. Coloring curves of liquid yogurts resulted from coloring assays using the B-PE extract from P. cruentum ([B-PE] extract = 0.8 mg B-PE/mL). Coloring curves of all assayed semisolid yogurts showed excellent ∆E*ab values. All final ∆E*ab values were equal to or lower than 5, suggesting that the color of each sample was highly similar to the color of its reference commercial product. Overall, staining Molecules 2021, 26, 297 8 of 13 ted for t to re- factors were low, with sample “N” requiring the lowest amount of extract and sample “L” requiring the highest (Table 2). Figure 6c shows the variations in a index (redness) for the prepared samples. It is possible to appreciate very slight variations, with values remaining practically constant from 0 to 11 days. In this case, the milkshake sample showed the highest values of a* while factors were low, with sample “N” requiring the lowest amount of extract and sample “L” requiring the highest (Table 2). Figure 6c shows the variations in a index (redness) for the prepared samples. It is possible to appreciate very slight variations, with values remaining practically constant from 0 to 11 days. In this case, the milkshake sample showed the highest values of a* while 2.3. Stability Studies of Colored Milk-Based Samples the yogurt sample registered the smallest. The liquid yogurt sample had the highest L* The stability of food products is a crucial aspect to take into account, and because of that, color stability determinations have been developed for milk-based products added with the B-PE extract. Figure 6 shows variations of the color coordinates L*, a*, and b* as well as the color difference ∆E*ab registered for milkshakes, liquid yogurts, and yogurts added with B-PE extract during the storage period of 11 days. The samples selected for each type of milk-based product were obtained by the addition of the B-PE extract to reproduce the color of “D” milkshake, “G” liquid yogurt, and “N” yogurt. q y g p g of a* (Figures 6b–d), compared with the milkshake and the yogurt. Moreover, it was ob- served that values of L* were stable from 0 to 9 days for the three types of milk-based products and slightly increased up to 11 days for liquid yogurts. However, the yogurt sample registered L* values slightly increasing up to 6 days and then remaining constant for 11 days. Despite the above changes, the movements of the values of L* were not ob- served to have a drastic behavior and ΔL* between 11 days and 0 days did not exceed 1.5 units for all the 3 types of products assayed. Figure 6. Changes in color indexes (L*, a*, b*, and ∆E*ab are the CIELAB color system values) of milk shake (■), liquid yogurt (♦), and yogurt (▲) pigmented with B-PE extract during a short period cold storage: the values plotted are the means of triplicate ± standard deviations.Changes in ∆E*ab (a); changes in L* (b); changes in a* (c) and changes in b* (d). Figure 6. Changes in color indexes (L*, a*, b*, and ∆E*ab are the CIELAB color system values) of milk shake (■), liquid yogurt (♦), and yogurt (▲) pigmented with B-PE extract during a short period cold storage: the values plotted are the means of triplicate ± standard deviations.Changes in ∆E*ab (a); changes in L* (b); changes in a* (c) and changes in b* (d). Figure 6. 2.3. Stability Studies of Colored Milk-Based Samples the yogurt sample registered the smallest. The liquid yogurt sample had the highest L* Changes in color indexes (L*, a*, b*, and ∆E*ab are the CIELAB color system values) of milk shake (■), liquid yogurt (♦), and yogurt (▲) pigmented with B-PE extract during a short period cold storage: the values plotted are the means of triplicate ± standard deviations.Changes in ∆E*ab (a); changes in L* (b); changes in a* (c) and changes in b* (d). Figure 6. Changes in color indexes (L*, a*, b*, and ∆E*ab are the CIELAB color system values) of milk shake (■), liquid yogurt (♦), and yogurt (▲) pigmented with B-PE extract during a short period cold storage: the values plotted are the means of triplicate ± standard deviations.Changes in ∆E*ab (a); changes in L* (b); changes in a* (c) and changes in b* (d). The ΔE*ab values (Figure 6a) were very low for liquid yogurt and milkshake, regis- tering ΔE*ab = 0.25 and ΔE*ab = 0.6 (1 days), respectively. Despite their variations and upward trends, ΔE*ab did not exceed a value of 2.0 for the entire study period. A different Figure 6c shows the variations in a* index (redness) for the prepared samples. It is possible to appreciate very slight variations, with values remaining practically constant from 0 to 11 days. In this case, the milkshake sample showed the highest values of a* while the yogurt sample registered the smallest. The liquid yogurt sample had the highest L* and b* values and intermediate values of a* (Figure 6b–d), compared with the milkshake and the yogurt. Moreover, it was observed that values of L* were stable from 0 to 9 days for the three types of milk-based products and slightly increased up to 11 days for liquid yogurts. However, the yogurt sample registered L* values slightly increasing up to 6 days and then remaining constant for 11 days. Despite the above changes, the movements of the values of L* were not observed to have a drastic Molecules 2021, 26, 297 9 of 13 behavior and ∆L* between 11 days and 0 days did not exceed 1.5 units for all the 3 types of products assayed. behavior and ∆L* between 11 days and 0 days did not exceed 1.5 units for all the 3 types of products assayed. The ∆E*ab values (Figure 6a) were very low for liquid yogurt and milkshake, regis- tering ∆E*ab = 0.25 and ∆E*ab = 0.6 (1 days), respectively. 2.3. Stability Studies of Colored Milk-Based Samples the yogurt sample registered the smallest. The liquid yogurt sample had the highest L* Despite their variations and upward trends, ∆E*ab did not exceed a value of 2.0 for the entire study period. A different trend was observed for yogurt, which started with a value of ∆E*ab = 0.25 at day 1 and concluded with ∆E*ab = 3.0 at 11 days. However, regardless of its high fluctuation, the difference between 11 days and 1 day was 2.75 for yogurt, whereas it was 1.4 and 1.25 for milkshake and liquid yogurt, respectively. It is well known that ∆E*ab values lower than 3.0 cannot be easily detected by the naked human eye and are taken as the same color by consumers [35]. According to these data, it is deduced that the sample color is quite stable and that this fact could be due to the protection afforded by the lipid–protein matrix of milk. It is well known that the majority of natural pigments have antioxidant activity. Concerning stability studies of colored milk-based samples, the data obtained showed that the sample color is quite stable. This effect could be explained by the antioxidant activity of the B-PE extract as reported by others authors [30–32]. The antioxidant effect of phycoerythrin was also shown in a previous study using this colorant in ice-cream [7]. y y y g The high stability of the phycoerythrin pigment within the protein–lipid matrix of milk-based products found in this study is in contrast with the statement on storage studies as a generality reported for others authors on storage studies who pointed out that the colors from natural sources tend to lose dyeing strength or to disappear with time [36]. According to them, a greater weight of plant material would be needed for direct use or for extraction of the natural dye compared to the weight of the synthetic dye in order to obtain the same depth of color. Incorporation of natural colorants to food systems faces different challenges such as their relatively low stability in processing and storage conditions and the presence of undesirable odor or flavor characteristics [37]. All this indicates that phycoerythrin maintained its stability throughout the refriger- ated storage time for the three types of milk-based products without significant changes in any of the three color coordinates. Values of ∆E*ab ≥5.0 were not obtained at any moment during storage. This is the threshold value used by other researchers to indicate the onset of instability [21]. 2.3. Stability Studies of Colored Milk-Based Samples the yogurt sample registered the smallest. The liquid yogurt sample had the highest L* In addition to its antioxidant capacity, another aspect to consider is the toxicological connotation involving phycoerythrin application as a natural colorant. This molecule is gaining more importance in present days as it plays a crucial role in preventing or delaying diseases that involve toxicity in certain organs such as the liver [38]. By and large, artificial antioxidants such as butylhydroxynisole and butylhydroxytoluene possess potential health risk and toxicity, which makes it crucial to substitute them with new harmless natural antioxidants like phycoerythrin. 3.3. Alimentary Samples Alimentary samples are constituted by milk-based products (milkshakes, liquid yo- gurts, and semisolid yogurts) available in grocery stores. The selected commercial products have a pink color; while standard products do not carry any coloring additive and therefore have a white color. Regarding milkshakes, semi-skimmed white milk of the same brand was used as a standard. On the other hand, for liquid yogurts and semisolid yogurts, a natural white liquid and semisolid yogurt of the same brand were used as a standard. Different commercial products were utilized to simulate colors using the B-phycoerythrin extract: six milkshakes, five liquid yogurts, and four yogurts. Storage of the samples was done in a domestic refrigerator (Fagor, 3Ffk-6625X, Mondragón, Spain) at T = 5 ± 2 ◦C. Only semisolid yogurts were manipulated before colorimetric analysis due to their consistency. They were weighed and then placed on the magnetic stirrer (HB502, Bibby, Stone, UK) for 10 min to obtain a liquid matrix. The pink protein extract was added to the matrix under constant stirring to allow its homogenous mixing. 3.2. Characterization of the B-Phycoerythrin Extract 3.2. Characterization of the B-Phycoerythrin Extract Absorbance spectra were recorded on the UV-visible spectrophotometer (Lambda 20, Perkin Elmer, software UV Data Manager, Waltham, Massachussetts, U.S.) between 250 and 750 nm. Fluorescence spectra were recorded on the spectrofluorimeter (FP-6500, Jasco, soft- ware Spectra Manger, Madrid, Spain) between 500 and 700 nm. Recording the absorption and fluorescence spectra of aqueous solutions in a large pH range served to monitor the influence of pH on the stability of the spectral parameters of the B-phycoerythrin extract. The protein extract was dissolved in 20 mM phosphate buffer at different pHs (ranging between 1.0 and 13.0) to achieve an appropriate concentration. Absorption spectra of the B-PE extract were determined after gradual falls or increases in pH (pH 7–1 and pH 7–12), and the stability was monitored by measuring the absorbance at 545 nm. Fluorescence spectra were determined with the same solutions, and a fluorescence intensity at 576 nm was registered. All spectra were recorded at room temperature. g p p The amounts of different phycobiliproteins (B-PE, R-PC, and APC) in the B-PE extract were calculated from measurements of the absorbance at 280, 545, 565, 620, and 650 nm using the following equations [33]: R-PC (mg/mL) = (A620 nm −0.7·A650 nm)/7.38 (1) APC (mg/mL) = (A650 nm −0.19·A620 nm)/5.65 (2) B-PE (mg/mL) = (A565 nm−2.8·[R-PC] −1.34·[APC])/12.7 (3) (3) 3.1. Production of Phycoerythrin Extract The B-phycoerythrin extract was obtained using the corresponding methodology [34,39] from cells of the microalgae P. cruentum, generously provided by the Chemical Engineering Department of Almería University (Spain). The phycobiliprotein solution was extracted from the lyophilized microalgae using phosphate buffer 0.1 M pH 5.5 for reconstitution. The intracellular material was obtained by disrupting the cell wall by osmotic shock. The sample was centrifuged (Medtronic BL-S, P-Selecta, Barcelona, Spain) to recover the phy- cobiliproteins in the supernatant. The supernatant was precipitated with 60% saturation ammonium sulphate ((NH4)2SO4). The precipitate was centrifuged, and the pellet was recovered and resuspended in a small volume of the suitable buffer. The protein solution was dialyzed to obtain the final B-phycoerythrin extract. The concentrated protein solution was placed in a clean container, and 1% w/w of sodium azide (NaN3) was added as a preservative. The sample was stored at 4 ◦C. 10 of 13 10 of 13 Molecules 2021, 26, 297 3.4. Color Analysis The color of the samples was analyzed by a colorimeter (CM-5, Konica Minolta, data software CM-S100W SpectraMagic TM NX, Version 2.2, Osaka, Japan) operated in the transmittance mode. The color was registered according to the CIELAB color space, and the color space values were obtained: L* represents the light-dark pair (lightness), a* represents the red-green pair (redness) and b* represents the yellow–blue pair (yellowness). The recommended illuminant/observer pair D65/10◦was used following the method of measurement for translucent semisolid samples [15]. First, the color of the reference product (commercial product) was registered by the colorimeter, followed by registration of the color of the standard sample (sample to be added with B-PE extract). Successively, few microliters of B-PE extract were added to the standard sample, and the color change was registered. After each B-PE extract addition, the color difference between the alimentary commercial product and the colored sample was defined by ∆E*ab. For determination of the CIELAB color difference ∆E*ab, the following equation was used: ∆E*ab = [(∆L*)2 + (∆a*)2 + (∆b*)2]1/2 = [(L*i −L*0)2 + (a*i −a*0)2 + (b*i −b*0)2]1/2 (4) (4) Molecules 2021, 26, 297 11 of 13 11 of 13 where L*0, a*0, and b*0 were the values of the alimentary commercial product and L*i, a*I, and b*i were the measured values of each sample added with the B-PE extract. where L*0, a*0, and b*0 were the values of the alimentary commercial product and L*i, a*I, and b*i were the measured values of each sample added with the B-PE extract. 4. Conclusions The present study demonstrates that the B-PE extract can be used as a coloring additive for milk-based products. The color of commercial reference products containing synthetic and natural colorants was compared to the color of samples dyed with the pink B-PE extract. Colors and color differences were measured according to the CIELAB color space, and the colorimetry-based approach was suitable for the purpose of the study. Coloring assays conducted using a colorimeter prove that the colors of the samples are similar to the colors of the reference samples taken from the market. Staining factors determine that low amounts of B-PE are necessary to obtain acceptable samples colors. pH influences the spectroscopic properties of the protein extract and the stability of B-PE. The optimal pH range for stability of the B-PE extract isolated from P. cruentum is 4.0–9.0. The study proves that the B-PE coloring extract could be an interesting natural colorant for its attractive coloring properties, and it could expand the list of natural coloring additives now available on the market. Author Contributions: A.B.G. and E.L. performed the experiments and analyzed the data. E.L. and M.C.M. developed samples preparation and instrumental methods and helped in some experiments. R.B. conceived and supervised the work and wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the University of Jaén (research project PID2019-107816GB- I00). Data Availability Statement: Not applicable. Acknowledgments: The authors would like to thank the Department of Chemical Engineering of the University of Almería for providing the P. cruentum microalgae biomass. Ana Belén García would like to thank the University of Jaén for her pre-doctoral fellowship. Eleonora Longo would like to thank the University of Jaén and Padua University (Italy) for her pre-doctoral fellowship. Conflicts of Interest: The authors declare no conflict of interest. Sample Availability: Samples of the compounds are available from the authors. References 1. Camara-Artigas, A.; Bacarizo, J.; Andújar-Sánchez, M.; Ortiz-Salmerón, E.; Valle, C.M.; Cuadri, C.; Martin-Garcia, J.M.; Martínez- Rodríguez, S.; Sobczuk, T.M.; Ibañez, M.J.; et al. pH-dependent structural conformations of B-phycoerythrin from Porphyridium cruentum. FEBS J. 2012, 279, 3680–3691. [CrossRef] [PubMed] 1. Camara-Artigas, A.; Bacarizo, J.; Andújar-Sánchez, M.; Ortiz-Salmerón, E.; Valle, C.M.; Cuadri, C.; Martin-Garcia, J.M.; Martínez- Rodríguez, S.; Sobczuk, T.M.; Ibañez, M.J.; et al. pH-dependent structural conformations of B-phycoerythrin from Porphyridium cruentum. FEBS J. 2012, 279, 3680–3691. [CrossRef] [PubMed] 1. Camara-Artigas, A.; Bacarizo, J.; Andújar-Sánchez, M.; Ortiz-Salmerón, E.; Valle, C.M.; Cuadri, C.; Martin-Garcia, J.M.; Martínez- Rodríguez, S.; Sobczuk, T.M.; Ibañez, M.J.; et al. pH-dependent structural conformations of B-phycoerythrin from Porphyridium cruentum. FEBS J. 2012, 279, 3680–3691. [CrossRef] [PubMed] 2. Fernández-Rojas, B.; Hernández-Juárez, J.; Pedraza-Chaverri, J. Nutraceutical properties of phycocyanin. J. Funct. Foods 2014, 11, 375–392. [CrossRef] 2. Fernández-Rojas, B.; Hernández-Juárez, J.; Pedraza-Chaverri, J. Nutraceutical properties of phycocyanin. J. Funct. Foods 2014, 11, 375–392. [CrossRef] 3. Munier, M.; Jubeau, S.; Wijaya, A.; Morançais, M.; Dumay, J.; Marchal, L.; Jaouen, P.; Fleurence, J. Physicochemical factors affecting the stability of two pigments: R-phycoerythrin of Grateloupia turuturu and B-phycoerythrin of Porphyridium cruentum. Food Chem. 2014, 150, 400–407. [CrossRef] [PubMed] [ ] [ ] 4. Martínez, J.M.; Delso, C.; Álvarez, I.; Raso, J. Pulsed electric field permeabilization and extraction of phycoerythrin from Porphyridium cruentum. Algal Res. 2019, 37, 51–56. [CrossRef] 4. Martínez, J.M.; Delso, C.; Álvarez, I.; Raso, J. Pulsed electric field permeabilization and extraction of phycoerythrin from Porphyridium cruentum. Algal Res. 2019, 37, 51–56. [CrossRef] p y 5. Wang, X.; Guo, Y.; Li, Z.; Ying, W.; Chen, D.; Deng, Z.; Peng, X. Dual emission from nanoconfined R-phycoerythrin fluorescent proteins for white light emission diodes. RSC Adv. 2019, 9, 9777–9782. [CrossRef] p y 5. Wang, X.; Guo, Y.; Li, Z.; Ying, W.; Chen, D.; Deng, Z.; Peng, X. Dual emission from nanoconfined R-phycoerythrin fluorescent proteins for white light emission diodes. RSC Adv. 2019, 9, 9777–9782. [CrossRef] 6. Sudhakar, M.P.; Jagatheesan, A.; Perumal, K.; Arunkumar, K. Methods of phycobiliprotein extraction from Gracilaria crassa and its applications in food colourants. Algal Res. 2015, 8, 115–120. [CrossRef] 6. Sudhakar, M.P.; Jagatheesan, A.; Perumal, K.; Arunkumar, K. Methods of phycobiliprotein extraction from Gracilaria crassa and its applications in food colourants. Algal Res. 2015, 8, 115–120. [CrossRef] pp g 7. Ganesan, A.R.; Shanmugam, M. Isolation of phycoerythrin from Kappaphycus alvarezii: A poten cream. Environ. Biol. Fishes 2020, 32, 4221–4233. [CrossRef] 8. References Carocho, M.; Barreiro, M.F.; Morales, P.; Ferreira, I.C.F.R. Adding Molecules to Food, Pros and Cons: A Review on Synthetic and Natural Food Additives. Compr. Rev. Food Sci. Food Saf. 2014, 13, 377–399. [CrossRef] 8. Carocho, M.; Barreiro, M.F.; Morales, P.; Ferreira, I.C.F.R. Adding Molecules to Food, Pros and Cons: A Review on Synthetic and Natural Food Additives. Compr. Rev. Food Sci. Food Saf. 2014, 13, 377–399. [CrossRef] p f , , [ ] 9. Carocho, M.; Morales, P.; Ferreira, I.C. Natural food additives: Quo vadis? Trends Food Sci. Technol. 2015, 45, 284–295. [CrossRef] 10. Pereira, T.; Barroso, S.; Mendes, S.; Gil, M.M. Stability, kinetics, and application study of phycobiliprotein pigments extracted from red algae Gracilaria gracilis. J. Food Sci. 2020, 85, 3400–3405. [CrossRef] p f [ ] 9. Carocho, M.; Morales, P.; Ferreira, I.C. Natural food additives: Quo vadis? Trends Food Sci. Technol. 2015, 45, 284–295. [CrossRef] 9. Carocho, M.; Morales, P.; Ferreira, I.C. Natural food additives: Quo vadis? Trends Food Sci. Technol. 2015, 45, 284–295. [CrossRef] 10. Pereira, T.; Barroso, S.; Mendes, S.; Gil, M.M. Stability, kinetics, and application study of phycobiliprotein pigments extracted from red algae Gracilaria gracilis. J. Food Sci. 2020, 85, 3400–3405. [CrossRef] 12 of 13 12 of 13 Molecules 2021, 26, 297 11. Montibeller, M.J.; Monteiro, P.D.L.; Rios, A.D.O.; Rios, A.D.O.; Manfroi, V. Stability assessment of anthocyanins obtained from skin grape applied in kefir and carbonated water as a natural colorant. J. Food Process. Preserv. 2018, 42, e13698. [CrossRef] g p pp J , , [ ] 12. Ngamwonglumlert, L.; Devahastin, S.; Chiewchan, N.; Raghavan, V.G.S. Color and molecular structure alterations of brazilein g p pp J 12. Ngamwonglumlert, L.; Devahastin, S.; Chiewchan, N.; Raghavan, V.G.S. Color and molecular structure alterations of brazilein extracted from Caesalpinia sappan L under different pH and heating conditions Sci Rep 2020 10 1 10 [CrossRef] [PubMed] g p pp J 12. Ngamwonglumlert, L.; Devahastin, S.; Chiewchan, N.; Raghavan, V.G.S. Color and molecular structure alterations of brazilein extracted from Caesalpinia sappan L. under different pH and heating conditions. Sci. Rep. 2020, 10, 1–10. [CrossRef] [PubMed] 12. Ngamwonglumlert, L.; Devahastin, S.; Chiewchan, N.; Raghavan, V.G.S. Color and molecular structu extracted from Caesalpinia sappan L. under different pH and heating conditions. Sci. Rep. 2020, 10, 1– g g , ; , ; , ; g , extracted from Caesalpinia sappan L. under different pH and heating conditions. Sci. Rep. 2020, 10, 1–10. References The effect of pH treatment and refrigerated storage on natural colourant preparations (betacyanins) from red pitahaya and their potential application in yoghurt. LWT 2017, 80, 437–445. [CrossRef] 21. Obón, J.; Castellar, M.; Alacid, M.; Fernández-López, J. Production of a red–purple food colorant from Opuntia stricta fruits by spray drying and its application in food model systems. J. Food Eng. 2009, 90, 471–479. [CrossRef] 22. Da Silva, L.P.; Pereira, E.; Prieto, M.A.; Simal-Gandara, J.; Pires, T.C.; Alves, M.J.; Calhelha, R.C.; Barros, L.; Ferreira, I.C. Rubus ulmifolius Schott as a Novel Source of Food Colorant: Extraction Optimization of Coloring Pigments and Incorporation in a Bakery Product. Molecules 2019, 24, 2181. [CrossRef] [PubMed] y 23. Dewi, E.N.; A Kurniasih, R.; Purnamayati, L. The Application of Microencapsulated Phycocyanin as a Blue Natural Colorant to the Quality of Jelly Candy. IOP Conf. Ser. Earth Environ. Sci. 2018, 116, 012047. [CrossRef] 24. Otálora, M.C.; Barbosa, H.D.J.; Perilla, J.E.; Osorio, C.; Nazareno, M.A. Encapsulated betalains (Opuntia ficus-indica) as natural colorants. Case study: Gummy candies. LWT 2019, 103, 222–227. [CrossRef] 25. Leong, H.Y.; Su, C.-A.; Lee, B.-S.; Lan, J.C.-W.; Law, C.L.; Chang, J.-S.; Show, P.L. Development of Aurantiochytrium limacinum SR21 cultivation using salt-rich waste feedstock for docosahexaenoic acid production and application of natural colourant in food product. Bioresour. Technol. 2019, 271, 30–36. [CrossRef] 26. Chentir, I.; Kchaou, H.; Hamdi, M.; Jridi, M.; Li, S.; Doumandji, A.; Nasri, M. Biofunctional gelatin-based films incor- porated with food grade phycocyanin extracted from the Saharian cyanobacterium Arthrospira sp. Food Hydrocoll. 2019, 89, 715–725. [CrossRef] [ ] 27. Galetovi´c, A.; Seura, F.; Gallardo, V.; Graves, R.; Cortés, J.; Valdivia, C.; Núñez, J.; Tapia, C.; Neira, I.; Sanzana, S.; et al. Use of Phycobiliproteins from Atacama Cyanobacteria as Food Colorants in a Dairy Beverage Prototype. Foods 2020, 9, 244. [CrossRef] 27. Galetovi´c, A.; Seura, F.; Gallardo, V.; Graves, R.; Cortés, J.; Valdivia, C.; Núñez, J.; Tapia, C.; Neira, I.; Sanzana, S.; et al. Use of Phycobiliproteins from Atacama Cyanobacteria as Food Colorants in a Dairy Beverage Prototype. Foods 2020, 9, 244. [CrossRef] 28. Niccolai, A.; Venturi, M.; Galli, V.; Pini, N.; Rodolfi, L.; Biondi, N.; D’Ottavio, M.; Batista, A.P.; Raymundo, A.; Granchi, L.; et al. Development of new microalgae-based sourdough “crostini”: Functional effects of Arthrospira platensis (spirulina) addition. Sci. Rep. 2019, 9, 1–12. [CrossRef] y p y y g yp 28. References extracted from Caesalpinia sappan L. under different pH and heating conditions. Sci. Rep. 2020, 10, 1–10. [CrossRef] [PubMed] 13. Lobo, F.A.T.; Silvana, R.; Domingues, J.; Rodrigues, S.; Costa, V.; Falcão, D.; Araujo, K.G.D.L. Inclusion complexes of yellow bell pepper pigments with β -cyclodextrin: Preparation, characterisation and application as food natural colorant. J. Sci. Food Agric. 2018, 98, 2665–2671. [CrossRef] [PubMed] 14. Chen, X.; Ding, Y.; Forrest, B.; Oh, J.; Boussert, S.M.; Hamann, M.T. Lemon yellow #15 a new highly stable, water soluble food colorant from the peel of Citrus limon. Food Chem. 2019, 270, 251–256. [CrossRef] [PubMed] 15. Mezquita, P.C.; Huerta, B.E.B.; Ramírez, J.C.P.; Hinojosa, C.P.O. Milks pigmentation with astaxanthin and determination of colour stability during short period cold storage. J. Food Sci. Technol. 2015, 52, 1634–1641. [CrossRef] [PubMed] uita, P.C.; Huerta, B.E.B.; Ramírez, J.C.P.; Hinojosa, C.P.O. Milks pigmentation with astaxanthin and determin 15. Mezquita, P.C.; Huerta, B.E.B.; Ramírez, J.C.P.; Hinojosa, C.P.O. Milks pigmentation with astaxanthin and determination of colour stability during short period cold storage J Food Sci Technol 2015 52 1634 1641 [CrossRef] [PubMed] Mezquita, P.C.; Huerta, B.E.B.; Ramírez, J.C.P.; Hinojosa, C.P.O. Milks pigmentation with astaxanthin and d stability during short period cold storage. J. Food Sci. Technol. 2015, 52, 1634–1641. [CrossRef] [PubMed] ty during short period cold storage. J. Food Sci. Technol. 2015, 52, 1634–1641. [CrossRef] [PubMed] 16. De O Moreira, I.; Passos, T.S.; Chiapinni, C.; Silveira, G.K.; Souza, J.C.M.; Coca-Vellarde, L.G.; Deliza, R.; Araújo, K.G.D.L. Colour evaluation of a phycobiliprotein-rich extract obtained from Nostoc PCC9205 in acidic solutions and yogurt. J. Sci. Food Agric. 2011, 92, 598–605. [CrossRef] 17. Pires, T.C.S.P.; Dias, M.I.; Carocho, M.; Barreira, J.C.; Santos-Buelga, C.; Barros, L.; Ferreira, I.C. Extracts from Vaccinium myrtillus L. fruits as a source of natural colorants: Chemical characterization and incorporation in yogurts. Food Funct. 2020, 11, 3227–3234. [CrossRef] [PubMed] [ ] [ ] 18. De Almeida, H.H.S.; Barros, L.; Barreira, J.C.; Calhelha, R.C.; Heleno, S.A.; Sayer, C.; Miranda, C.G.; Leimann, F.V.; Barreiro, M.F.; Ferreira, I.C. Bioactive evaluation and application of different formulations of the natural colorant curcumin (E100) in a hydrophilic matrix (yogurt). Food Chem. 2018, 261, 224–232. [CrossRef] 19. Mourtzinos, I.; Prodromidis, P.; Grigorakis, S.; Makris, D.P.; Biliaderis, C.G.; Mourtzinos, I. Natural foo onion wastes: Application in a yoghurt product. Electrophoresis 2018, 39, 1975–1983. [CrossRef] pp y g p p 20. Gengatharan, A.; Dykes, G.A.; Choo, W.-S. 36. Kammerer, D.R.; Schillmöller, S.; Maier, O.; Schieber, A.; Carle, R. Colour stability of canned strawberries using black carrot and elderberry juice concentrates as natural colourants. Eur. Food Res. Technol. 2006, 224, 667–679. [CrossRef] y j 37. Wallace, T.; Giusti, M. Determination of Color, Pigment, and Phenolic Stability in Yogurt Systems Colored with Nonacylated Antho- cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 2008, 73, C241–C248. [CrossRef] 38. Afreen, S.; Fatma, T. Extraction, purification and characterization of phycoerythrin from Michrochaete and its biological activities. Biocatal. Agric. Biotechnol. 2018, 13, 84–89. [CrossRef] g 39. Bermejo-Román, R.; Ruiz, E.; Ramos, A.; Acién, F.G. Pilot-Scale Recovery of Phycoerythrin from Porphyridium cruentum using Expanded Bed Adsorption Chromatography. Sep. Sci. Technol. 2013, 48, 1913–1922. [CrossRef] cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 2008, 73, C241–C248. [CrossRef] 38. Afreen, S.; Fatma, T. Extraction, purification and characterization of phycoerythrin from Michrochaete and its biological activities. Biocatal. Agric. Biotechnol. 2018, 13, 84–89. [CrossRef] 37. Wallace, T.; Giusti, M. Determination of Color, Pigment, and Phenolic Stability in Yogurt Systems Colored cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 2008, 7 cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 200 38. Afreen, S.; Fatma, T. Extraction, purification and characterization of phycoerythrin from Michrochae Biocatal. Agric. Biotechnol. 2018, 13, 84–89. [CrossRef] , ; , , g , y g y cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 2008, 7 38. Afreen, S.; Fatma, T. Extraction, purification and characterization of phycoerythrin from Michrochaete a Biocatal. Agric. Biotechnol. 2018, 13, 84–89. [CrossRef] y j , , [ ] 37. Wallace, T.; Giusti, M. Determination of Color, Pigment, and Phenolic Stability in Yogurt Systems Colored with Nonacylated Antho- cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 2008, 73, C241–C248. [CrossRef] 38. Afreen, S.; Fatma, T. Extraction, purification and characterization of phycoerythrin from Michrochaete a Biocatal. Agric. Biotechnol. 2018, 13, 84–89. [CrossRef] 36. Kammerer, D.R.; Schillmöller, S.; Maier, O.; Schieber, A.; Carle, R. Colour stability of canned strawberries using black carrot and elderberry juice concentrates as natural colourants. Eur. Food Res. Technol. 2006, 224, 667–679. [CrossRef] 37. Wallace, T.; Giusti, M. Determination of Color, Pigment, and Phenolic Stability in Yogurt Systems Colored with Nonacylated Antho- cyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J. Food Sci. 2008, 73, C241–C248. [CrossRef] 38. Afreen, S.; Fatma, T. Extraction, purification and characterization of phycoerythrin from Michrochaete and its biological activities. Biocatal. Agric. Biotechnol. 2018, 13, 84–89. [CrossRef] 39. Bermejo-Román, R.; Ruiz, E.; Ramos, A.; Acién, F.G. Pilot-Scale Recovery of Phycoerythrin from Porphyridium cruentum using Expanded Bed Adsorption Chromatography. Sep. Sci. Technol. 2013, 48, 1913–1922. [CrossRef] References Niccolai, A.; Venturi, M.; Galli, V.; Pini, N.; Rodolfi, L.; Biondi, N.; D’Ottavio, M.; Batista, A.P.; Raymundo, A.; Granchi, L.; et al. Development of new microalgae-based sourdough “crostini”: Functional effects of Arthrospira platensis (spirulina) addition. Sci. Rep. 2019, 9, 1–12. [CrossRef] p 29. Hsieh-Lo, M.; Castillo, G.; Ochoa-Becerra, M.A.; Mojica, L. Phycocyanin and phycoerythrin: Strategies to improve production yield and chemical stability. Algal Res. 2019, 42, 101600. [CrossRef] 30. Patel, S.N.; Sonani, R.R.; Jakharia, K.; Bhastana, B.; Patel, H.M.; Chaubey, M.G.; Singh, N.K.; Madamwar, D. Antioxidant activity and associated structural attributes of Halomicronema phycoerythrin. Int. J. Biol. Macromol. 2018, 111, 359–369. [CrossRef] p y y 31. Jung, S.-M.; Park, J.S.; Shim, H.J.; Kwon, Y.S.; Kim, H.G.; Shin, H.S. Antioxidative effect of phycoerythrin derived from Grateloupia filicina on rat primary astrocytes. Biotechnol. Bioprocess Eng. 2016, 21, 676–682. [CrossRef] 32. Sonani, R.R.; Singh, N.K.; Kumar, J.; Thakar, D.; Madamwar, D. Concurrent purification and antioxidant activity of phyco- biliproteins from Lyngbya sp. A09DM: An antioxidant and anti-aging potential of phycoerythrin in Caenorhabditis elegans. Process. Biochem. 2014, 49, 1757–1766. [CrossRef] 33. Bermejo, R.; Fernández, E.; Alvarez-Pez, J.M.; Talavera, E.M. Labeling of cytosine residues with biliproteins for use as fluorescent DNA probes. J. Lumin. 2002, 99, 113–124. [CrossRef] p 34. Bermejo, R.; Ruiz, E.; Acien, F.; Acién-Fernández, F.G. Recovery of B-phycoerythrin using expanded bed adsorption chromatogra- phy: Scale-up of the process. Enzym. Microb. Technol. 2007, 40, 927–933. [CrossRef] 34. Bermejo, R.; Ruiz, E.; Acien, F.; Acién-Fernández, F.G. Recovery of B-phycoerythrin using expanded bed adsorption chromatogra- phy: Scale-up of the process. Enzym. Microb. Technol. 2007, 40, 927–933. [CrossRef] p y p p y 35. Hong, S.-I.; Han, J.H.; Krochta, J.M. Optical and surface properties of whey protein isolate coatings on plastic films as influenced by substrate, protein concentration, and plasticizer type. J. Appl. Polym. Sci. 2004, 92, 335–343. [CrossRef] p y p p 35. Hong, S.-I.; Han, J.H.; Krochta, J.M. Optical and surface properties of whey protein isolate coatings on plastic films as influenced by substrate, protein concentration, and plasticizer type. J. Appl. Polym. Sci. 2004, 92, 335–343. [CrossRef] 13 of 13 13 of 13 Molecules 2021, 26, 297 g 39. Bermejo-Román, R.; Ruiz, E.; Ramos, A.; Acién, F.G. Pilot-Scale Recovery of Phycoerythrin from Porphyridium cruentum using Expanded Bed Adsorption Chromatography. Sep. Sci. Technol. 2013, 48, 1913–1922. [CrossRef]
https://openalex.org/W4319721567
https://journals.sfu.ca/ornneo/index.php/ornneo/article/download/929/627
English
null
NESTING OF THE GREATER YELLOW-HEADED VULTURE (CATHARTES MELAMBROTUS) IN EASTERN PERU
Ornitología neotropical
2,022
cc-by
2,475
Key words: Amazon basin · Breeding · Cathartidae · New World Vultures · Phenology · Rocky cliff Key words: Amazon basin · Breeding · Cathartidae · New World Vultures · Phenology · Rocky cliff The Greater Yellow-headed Vulture (Cathartes melambrotus) is a little studied Cathartidae endemic to tropical South America (Ferguson-Lees & Christie 2001). It is largely restricted to the Amazon basin, ranging from southern Colombia and northeastern Venezuela, east to the Guianas and south to southeastern Peru, northern Bolivia, and central and eastern Brazil, where it main- ly occupies primary forests (Jones 2020). Due to its wide distribution, the IUCN considers it a Least Concern species, even though its population appears to be decreasing (BirdLife International 2021). ii Although the Greater Yellow-headed Vulture is common in unbroken primary forests throughout its range (Thiollay 1989), there is little information on its reproductive biology (Monsalvo et al. 2020, Jones 2020). Perhaps due to its preference for un- disturbed lowland humid forest that is difficult to access, its nests and nesting sites have been poorly documented (Jones 2020). Available information is fragmentary, superficial, and controversial. Breeding activity of Greater Yellow-headed Vultures has been reported in Amazonas state, Brazil (Cintra & Naka 2012, Monsalvo et al. 2020), but the information provided is lim- ited to the mention that the species breeds in a natural reserve, and remarks the presence of fledglings and dates of observa- tions. Monsalvo et al. (2020) speculated that a Lesser Yellow-headed Vulture (Cathartes burrovianus) egg collected from a nest on the ground “between prickly thickness” in Paraguay could instead, according to its size, belong to a Greater Yellow-headed Vulture. Ferguson-Less & Christie (2001) described a nest as in a large tree cavity, but did not provide details on location; thus, Jones (2020) questioned this report. Here I describe, for the first time, a nesting record of the Greater Yellow-headed Vulture from eastern Peru by providing unquestionable evidence. While scanning a rocky cliff in search of raptor nests (Figure 1) on 31 October 2019 at 9:00 h, close to the town of San Luis de Shuaro (10°53'S, 75°17'O; 1,230 m a.s.l., Junín, Peru, I repeatedly observed two Greater Yellow-headed Vulture adults flying in front of the cliff. At around 15:00 h , one of the vultures approached the cliff and perched on a nearby tree for 5 min before moving to a rock on the cliff wall and then jumping to the entrance of a cavity, from where a well-developed nestling appeared (Figure 2). NESTING OF THE GREATER YELLOW-HEADED VULTURE (CATHARTES MELAMBROTUS) IN EASTERN PERU NESTING OF THE GREATER YELLOW-HEADED VULTURE (CATHARTES MELAMBROTUS) IN EASTERN PERU Submitted 7 April 2021 ∙ First decision 29 April 2022 ∙ Acceptance 11 May 2022 ∙ Online publication 7 August 2022 Communicated by Carlos Bosque © Neotropical Ornithological Society Tomás Rivas-Fuenzalida1,2 1 Avenida Independencia, 259, apartamento 1, La Merced, Chanchamayo, Junín, Perú (2019-2021) 2 Fundación Ñankulafkén, Reserva Natural El Natri, ruta p-60, km 42, Contulmo, Chile. E-mail: Tomás Rivas-Fuenzalida · trivasfuenzalida@gmail.com Abstract · The Greater Yellow-headed Vulture (Cathartes melambrotus) is one of the least studied cathartids (Cathartiformes, Cathartidae). Its reproductive biology is little known, being the only species in the family whose nest has not been described. Here, I report the finding of a nest located in a cavity 55 m above the ground on a high cliff, at 1,230 m a.s.l. in the rainforest of central Peru. The nest was active during the late dry season (October 31, 2019) and hosted a well-developed chick (50-60 days old) being fed by parents. The nesting site's habitat was thinned forests for shade coffee, banana, and citrus crops and remnants of mature secondary forests. Resumen · Nidificación del gallinazo de cabeza amarilla mayor (Cathartes melambrotus) en el este de Perú ii i El gallinazo de cabeza amarilla mayor (Cathartes melambrotus), es uno de los catártidos (Cathartiformes, Cathartidae) menos estudiados. Su biología reproductiva es poco conocida, siendo la única especie en la familia que no cuenta con una descripción del nido. Aquí describo el hallazgo de un nido ubicado en una cavidad a 55 m de altura en un acantilado a 1.230 m s.n.m. en la selva central de Perú. El nido estaba activo al final de la estación seca (31 octubre 2019) y contenía un polluelo bien desarrollado (50–60 días) que estaba siendo alimentado por sus padres. El hábitat alrededor del sitio nido presentó bosques raleados para sombrear cultivos de café, plantaciones de plátano y cítricos, y remanentes de bosque secundario maduro. Key words: Amazon basin · Breeding · Cathartidae · New World Vultures · Phenology · Rocky cliff The adult fed the nestling while fluttering awkwardly, and 2 min later it flew away (Figure 3). At 17:30 h, an adult vulture visited the nest, apparently intending to feed the nestling, but the birds were out of sight. At 18:00 h, I watched an adult vulture walking on the ground on a secondary road near the nest site, presumably foraging (Figure 5). Proper identifica- tion of the Greater Yellow-headed Vultures was possible due to the pale undersides to the secondaries, and by its apparent larger size compared to Turkey Vultures (C. aura), the only other vulture of this genus in the region. The similar Lesser Yellow- 100 RNITOLOGÍA NEOTROPICAL (2022) 33: 101—104 gure 1. Greater Yellow-headed Vulture (Cathartes melambrotus) nest-site in eastern Peru. Photo: Tomás Rivas-Fuenzalida. ORNITOLOGÍA NEOTROPICAL (2022) 33: 101—104 Figure 1. Greater Yellow-headed Vulture (Cathartes melambrotus) nest-site in eastern Peru. Photo: Tomás Rivas-Fuenzalida. igure 2. Greater Yellow-headed Vulture (Cathartes melambrotus) nestling in eastern Peru. Photo: Tomás Rivas-Fuenzalida. Figure 2. Greater Yellow-headed Vulture (Cathartes melambrotus) nestling in eastern Peru. Photo: Tomás Rivas-Fuenzalida. should be similar. According to Nelson et al. (2009), Turkey Vulture nestlings have an appearance similar to that of the Greater Yellow-headed Vulture chick shown in Figure 2, are 50 to 60 days old, and are close to fledging. Furthermore, incubation in the Turkey Vulture lasts 38–40 days (Kirk & Mossman 2020). Thus, assuming a 40-day incubation period should be similar. According to Nelson et al. (2009), Turkey Vulture nestlings have an appearance similar to that of the Greater Yellow-headed Vulture chick shown in Figure 2, are 50 to 60 days old, and are close to fledging. Furthermore, incubation in the Turkey Vulture lasts 38–40 days (Kirk & Mossman 2020). Thus, assuming a 40-day incubation period headed Vultures (C. burrovianus) does not occur in this area. Based on the information available for other Cathartes species, the time of reproduction and the age of the ob- served nestling could be estimated. Since Turkey and Greater Yellow-headed Vultures have similar body masses (c. 1,500 g; Jones 2020, Kirk & Mossman 2020), their nestling periods 101 NESTING OF CATHARTES MELAMBROTUS gure 3. Adult Greater Yellow-headed Vulture (Cathartes melambrotus) leaving the nest in eastern Peru. Photo: Tomás Rivas-Fuenzalida. Figure 3. Adult Greater Yellow-headed Vulture (Cathartes melambrotus) leaving the nest in eastern Peru. Photo: Tomás Rivas-Fuenzalida. Figure 4. Key words: Amazon basin · Breeding · Cathartidae · New World Vultures · Phenology · Rocky cliff Adult Greater Yellow-headed Vulture (Cathartes melambrotus), foraging in a secondary road near the nesting-site in eastern Peru. Photo: Tomá Rivas-Fuenzalida. Figure 4. Adult Greater Yellow-headed Vulture (Cathartes melambrotus), foraging in a secondary road near the nesting-site in eastern Peru. Photo: Tomás Rivas-Fuenzalida. w-headed Vulture (Cathartes melambrotus), foraging in a secondary road near the nesting-site in eastern Peru. Photo: Tomás for the Greater Yellow-headed Vulture and that the nestling was 50–60 days old, hatching must have occurred in the first week of September, and egg-laying between the third and fourth week of July. This breeding timing is consistent with the scant data from other regions. Tostain et al. (1992) watched copulation of Greater Yellow-headed Vultures in August in French Guiana. They suggested that breeding oc- curred during the dry season, with fledglings being active at 102 ORNITOLOGÍA NEOTROPICAL (2022) 33: 101—104 Figure 5. Greater Yellow-headed Vulture (Cathartes melambrotus) nest cavity, photographed by a drone in eastern Peru. Photo: Tomás Rivas-Fuenzalida. Figure 5. Greater Yellow-headed Vulture (Cathartes melambrotus) nest cavity, photographed by a drone in eastern Peru. Photo: Tomás Rivas-Fuenzalida. Figure 5. Greater Yellow-headed Vulture (Cathartes melambrotus) nest cavity, photographed by a drone in easte the beginning of the rainy season. Additionally, the estimat- ed timing of breeding of the Greater Yellow-headed Vulture coincides with that of many diurnal and nocturnal birds of prey within my study area in central-eastern Peru (author’s pers. observ.) and in other regions of the Neotropics (Whitacre & Burnham 2012). This is likely related to a better chance of survival of chicks in exposed nests by avoiding pe- riods of heavy rains (Gerhardt & Gerhardt 2012) or by great- er availability of food at the beginning of the rainy season (Whitacre & Burnham 2012). if the beginning of the rainy season. Additionally, the estimat- ed timing of breeding of the Greater Yellow-headed Vulture coincides with that of many diurnal and nocturnal birds of prey within my study area in central-eastern Peru (author’s pers. observ.) and in other regions of the Neotropics (Whitacre & Burnham 2012). This is likely related to a better chance of survival of chicks in exposed nests by avoiding pe- riods of heavy rains (Gerhardt & Gerhardt 2012) or by great- er availability of food at the beginning of the rainy season (Whitacre & Burnham 2012). if Vulture in Peru (1,100 m a.s.l; Walker et al. 2006, Schulen- berg et al. ACKNOWLEDGEMENTS I would like to thank SERFOR for the research permission (373-2018) and José Luis Gagliardi for his support. I’d also wish to thank Ricardo Figueroa Rojas and Francisco Rivas for their valuable comments and suggestions on the original version of this manuscript, and to the Editor-in-Chief Carlos Bosque and an anonymous reviewer for their comments, which greatly improved this work. The nesting site was a rocky cliff approximately 130 m wide and 70 m high, facing south, thus protecting the nest against solar radiation, strong winds, and rain. The nesting cavity was located approximately 55 m above a river course, 7 m from the cliff’s top and 5 m from the lateral verge of the cliff, and vegetation was dense above and around the cliff face. The cavity entrance (Figure 5) was about 80 cm high and 60 cm wide, and inaccessible to humans and terrestrial predators. Cliff and nest dimensions were estimated from altitude and distance recordings made by a drone and by comparing the entrance hole with the size of the parent vul- tures when perching at the nest entrance. It is of interest that two pairs of Black Vultures (Coragyps atratus) and one of Turkey Vultures (Cathartes aura) were also nesting in crev- ices and ledges on the same cliff as the Yellow-headed Vul- ture nested. Nesting in rocky substrates is also known for other Cathartidae, such as the Turkey Vulture (Kirk and Mossman 2020), the Black Vulture (Buckley 2020), the King Vulture (Sarcoramphus papa; Holste et al. 2020), the Califor- nia Condor (Gymnogyps californianus; Finkelstein et al. 2020), and the Andean Condor (Vultur gryphus, Houston et al. 2020). i Key words: Amazon basin · Breeding · Cathartidae · New World Vultures · Phenology · Rocky cliff 2010), revealing that it is not an exclusive breeder of lowland forests. It also suggests that Greater Yellow- headed Vultures may tolerate disturbed habitats, as ob- served in other areas of its distribution (Olmos et al. 2006, Thiollay 2007). REFERENCES BirdLife International (2021) Species factsheet: Cathartes mel- ambrotus. Available at http://www.birdlife.org [Accessed 3 June 2021] Buckley, NJ (2020) Black Vulture (Coragyps atratus), version 1.0. In: Poole, AF & FB Gill (eds). Birds of the World. Cornell Lab of Orni- thology, Ithaca, New York, USA. ii Cintra, R & LN Naka (2012) Spatial variation in bird community com- position in relation to topographic gradient and forest heteroge- neity in a Central Amazonian rainforest. International Journal of Ecology 1–25. i Ferguson-Lees J & D Christie (2001) Raptors of the world. Houghton Mifflin, Boston, USA t Finkelstein, M, Z Kuspa, NF Snyder, NJ Schmitt (2020) California Condor (Gymnogyps californianus), version 1.0. In: PG Rodewald (ed). Birds of the World. Cornell Lab of Ornithology, Ithaca, NY, USA. My finding shows that breeding occurred at the upper limit of the altitudinal range of the Greater Yellow-headed Gerhardt, RP & DM Gerhardt (2012) Mexican Owl. Pp. 313–319 in 103 NESTING OF CATHARTES MELAMBROTUS Whitacre DF (ed). Neotropical Birds of Prey, biology and ecology of a forest raptor community. Cornell University Press. Ithaca, New York, USA. Olmos, F, JF Pacheco & LF Silveira (2006) Notas sobre aves de rapina (Cathartidae, Acciptridae e Falconidae) brasileiras. Revista Brasi- leira de Ornitologia 14: 401–404. Schulenberg, TS, DF Stotz, DF Lane, JP O’Neill & TA Parker III (2010) Birds of Peru. Princeton University Press, New Jersey, USA. Holste, M, JM Ruth & JC Eitniear (2020) King Vulture (Sarcoramphus papa), version 1.0. In: Schulenberg, TS (ed.) Birds of the World. Cornell Lab of Ornithology, Ithaca, New York, USA. i Thiollay, JM (1989) Censusing of diurnal raptors in a primary rain forest: comparative methods and species detectability. Journal of Raptor Research 23: 72–84. i Houston, D, GM Kirwan, DA Christie & CJ Sharpe (2020) Andean Condor (Vultur gryphus), version 1.0. In: J del Hoyo, A Elliott, J Sargatal, DA Christie, & E de Juana (eds.) Birds of the World. Cornell Lab of Ornithology, Ithaca, New York, USA. Thiollay, JM (2007) Raptor communities in French Guiana: distribu- tion, habitat selection, and conservation. Journal of Raptor Research 41: 90–105. Jones, MF (2020) Greater Yellow-headed Vulture (Cathartes mel- ambrotus), version 1.0. In: Schulenberg, TS (ed.) Birds of the World. Cornell Lab of Ornithology, Ithaca, NY, USA. Tostain, O, JL Dujardin, C Érard & JM Thiollay (1992) Oiseaux de Guyane: Biologie, Écologie, Protection, Répartition. Société d’Etudes Ornithologiques & Laboratoire d’Ecologie Générale, Muséum National d’Histoire Naturelle, Brunoy, France. REFERENCES Kirk, DA & MJ Mossman (2020) Turkey Vulture (Cathartes aura), version 1.0. In: Poole, AF & FB Gill (eds.) Birds of the World. Cornell Lab of Ornithology, Ithaca, New York, USA. Walker, B, D F. Stotz, T Pequeño & JW Fitzpatrick (2006) Birds of the Manu Biosphere Reserve. Pp. 23–49 in Patterson BD, Stotz DF, Solari S (eds). Mammals and Birds of the Manu Biosphere Re- serve, Peru. Field Museum of Natural History, Chicago, USA. i Monsalvo, JAB, MAA Silva, NM Heming & MA Marini (2020) Geo- graphical variation and current knowledge on breeding traits of vultures in the neotropics. https://doi.org/10.1007/s43388-020- 00003-4 Whitacre, DF & WA Burnham (2012) Ecology and Conservation of Tikal´s Raptor Fauna. Pp. 328–360 in Whitacre DF (ed). Neotrop- ical Birds of Prey, biology and ecology of a forest raptor commu- nity. Cornell University Press, Ithaca, New York, USA. Nelson, RW, D Moore, F Kunnas & R Morse (2009) Turkey Vultures: A Photographic Guide for Aging Nestlings. Alberta Species at Risk Report No. 124. Alberta Sustainable Resource Develop- ment, Fish and Wildlife Division, Edmonton, Canada. 104
https://openalex.org/W3033024644
https://europepmc.org/articles/pmc7356636?pdf=render
English
null
Yeast Sphingolipid-Enriched Domains and Membrane Compartments in the Absence of Mannosyldiinositolphosphorylceramide
Biomolecules
2,020
cc-by
15,078
Yeast Sphingolipid-Enriched Domains and Membrane Compartments in the Absence of Mannosyldiinositolphosphorylceramide Andreia Bento-Oliveira 1,† , Filipa C. Santos 1,† , Joaquim Trigo Marquês 1,† , Pedro M. R. Paulo 2, Thomas Korte 3, Andreas Herrmann 3 , H. Susana Marinho 1 and Rodrigo F. M. de Almeida 1,* Andreia Bento-Oliveira 1,† , Filipa C. Santos 1,† , Joaquim Trigo Marquês 1,† , Pedro M. R. Paulo 2, Thomas Korte 3, Andreas Herrmann 3 , H. Susana Marinho 1 and Rodrigo F. M. de Almeida 1,* 1 Centro de Química Estrutural, Faculdade de Ciências, Universidade de Lisboa, 1749-016 Lisbon, Portugal; abdoliveira@fc.ul.pt (A.B.-O.); fpsantos@fc.ul.pt (F.C.S.); jmtmarques@fc.ul.pt (J.T.M.); hsmarinho@fc.ul.pt (H.S.M.) 1 Centro de Química Estrutural, Faculdade de Ciências, Universidade de Lisboa, 1749-016 Lisbon, Portugal; abdoliveira@fc.ul.pt (A.B.-O.); fpsantos@fc.ul.pt (F.C.S.); jmtmarques@fc.ul.pt (J.T.M.); hsmarinho@fc.ul.pt (H.S.M.) 2 Centro de Química Estrutural, Instituto Superior Técnico, 1049-001 Lisbon, Portugal; pedro.m.paulo@tecnico.ulisboa.pt 3 Department of Biology, Molecular Biophysics, IRI Life Sciences, Humboldt-Universität zu Berlin, 10115 Berlin, Germany; thomas.korte@rz.hu-berlin.de (T.K.); andreas.herrmann@rz.hu-berlin.de (A.H.) * Correspondence: rfalmeida@fc ul pt; Tel : +351 217 500 925 3 Department of Biology, Molecular Biophysics, IRI Life Sciences, Humboldt-Universität zu Berlin, 10115 Berlin, Germany; thomas.korte@rz.hu-berlin.de (T.K.); andreas.herrmann@rz.hu-berlin.de (A.H.) * Correspondence: rfalmeida@fc.ul.pt; Tel.: +351-217-500-925 10115 Berlin, Germany; thomas.korte@rz.hu-berlin.de (T.K.); andreas.herrmann@rz.hu-berlin.de (A.H.) * Correspondence: rfalmeida@fc.ul.pt; Tel.: +351-217-500-925 y * Correspondence: rfalmeida@fc.ul.pt; Tel.: +351-217-500-925 * Correspondence: rfalmeida@fc.ul.pt; Tel.: +351-217-500-925 † These authors contributed equally to this work. biomolecules biomolecules biomolecules 1. Introduction The plasma membrane of fungi contains a multitude of actively maintained membrane compartments, which have distinct protein markers, are stable in space and time, and are responsible for specific physiological functions [1–3]. Despite the growing interest in these compartments, the mechanism of their formation, their lipid composition, and their interrelations are still largely unknown [4]. In yeast and other fungi, as well as in plants, many plasma membrane proteins display slow lateral diffusion or are immobile, when compared to vertebrate biological membranes or artificial fluid lipid bilayers [2]. In some cases, this low mobility is dependent on plasma membrane sterols and sphingolipids [1,3,5–8]. In this regard, sphingolipid-enriched microdomains seem to be of particular relevance [9–11]. These domains found in the plasma membrane of Saccharomyces cerevisiae differ from the prototypical mammalian lipid raft both in composition, given they are ergosterol-depleted, and biophysically, as they are highly rigid gel domains [11]. Hence, the study of these domains is particularly important, as they may explain fundamental differences between fungal and mammalian membranes [9], providing potential targets that can be explored concerning antifungal drug development [12]. p g p g p g g g p In the plasma membrane of yeast S. cerevisiae two major non-overlapping compartments can be found: The membrane compartment containing Can1p (MCC), an arginine permease, and the membrane compartment containing Pma1p (MCP), the plasma membrane H+-ATPase. MCC is currently designated MCC/eisosome, as its characteristic furrow-like invaginating shape is stabilized by a large cytosolic protein scaffold, the eisosome, at the bottom of the MCC furrow [2,13,14]. A typical S. cerevisiae cell in exponential phase contains around 1.5–2.5 such invaginations per µm2 of cell surface [8,13] covering roughly 25% of the membrane area. Fluorescence labeling with the 3′-β-sterol-binding filipin suggested that the MCC/eisosome is ergosterol-enriched [15]. The deletion of proteins that are part of the MCC/eisosome, or are associated with it, compromises the formation of intact MCC/eisosome domains and influences ergosterol distribution in the plasma membrane [2]. Some of the well-known functions of the MCC/eisosome include the sensing of local membrane disturbances and deformations, e.g., due to mechanical or osmotic stress [16], involvement in membrane adaptation to other types of stress [17–20], regulation of lipid homeostasis, in particular sensing sphingolipid levels [2], and protection against endocytosis/starvation [3]. Notably, membrane potential has been suggested to modulate the organization of MCC domains. Received: 23 March 2020; Accepted: 2 June 2020; Published: 6 June 2020 Abstract: The relevance of mannosyldiinositolphosphorylceramide [M(IP)2C] synthesis, the terminal complex sphingolipid class in the yeast Saccharomyces cerevisiae, for the lateral organization of the plasma membrane, and in particular for sphingolipid-enriched gel domains, was investigated by fluorescence spectroscopy and microscopy. We also addressed how changing the complex sphingolipid profile in the plasma membrane could influence the membrane compartments (MC) containing either the arginine/ H+ symporter Can1p (MCC) or the proton ATPase Pma1p (MCP). To achieve these goals, wild-type (wt) and ipt1∆cells, which are unable to synthesize M(IP)2C accumulating mannosylinositolphosphorylceramide (MIPC), were compared. Living cells, isolated plasma membrane and giant unilamellar vesicles reconstituted from plasma membrane lipids were labelled with various fluorescent membrane probes that report the presence and organization of distinct lipid domains, global order, and dielectric properties. Can1p and Pma1p were tagged with GFP and mRFP, respectively, in both yeast strains, to evaluate their lateral organization using confocal fluorescence intensity and fluorescence lifetime imaging. The results show that IPT1 deletion strongly affects the rigidity of gel domains but not their relative abundance, whereas no significant alterations could be perceived in ergosterol-enriched domains. Moreover, in these cells lacking M(IP)2C, a clear alteration in Pma1p membrane distribution, but no significant changes in Can1p distribution, were observed. Thus, this work reinforces the notion that sphingolipid-enriched domains distinct from ergosterol-enriched regions are present in the S. cerevisiae plasma membrane and suggests that M(IP)2C is important for a proper hydrophobic chain packing of sphingolipids in the gel domains of wt cells. Furthermore, our results strongly support the involvement of sphingolipid domains in the formation and stability of the MCP, possibly being enriched in this compartment. Keywords: Saccharomyces cerevisiae; membrane compartments; sphingolipids; Pma1p; Can1p; fluorescence lifetime imaging microscopy (FLIM); fungal plasma membrane; inositolphosphorylceramides; fluorescence spectroscopy; giant unilamellar vesicles (GUVs) Biomolecules 2020, 10, 871; doi:10.3390/biom10060871 www.mdpi.com/journal/biomolecules 2 of 24 Biomolecules 2020, 10, 871 1. Introduction Readily after membrane depolarization, proteins including Can1p are no longer confined to the MCC, acquiring a more homogeneous distribution in the plasma membrane [15,21]. Concomitantly, sphingolipid-enriched microdomains dissipate [22]. Thus, despite the stability of MCC and the low mobility of Can1p inside the MCC, its distribution in the plasma membrane is dynamic and regulated. The MCP was first identified in S. cerevisiae by confocal fluorescence microscopy as a network-like compartment containing Pma1p [23], the most abundant plasma membrane protein in fungi [24]. This protein is responsible for the creation and maintenance of membrane potential of fungal cells, as well as regulation of the intracellular pH and ion concentration [25]. More recently, it has been suggested to be of relevance for signaling [3]. The electrochemical gradient established by Pma1p drives nutrient uptake, namely by symporters localized at the MCC, such as Can1p, establishing an important functional relation between MCP and MCC. On the other hand, in S. cerevisiae [26,27] as well as in Cryptococcus neoformans [28,29], alterations of sphingolipid metabolic balance affect the oligomerization, traffic, activity, and plasma membrane stability of Pma1p [30–33]. These results point to an intimate relation between the MCP and the sphingolipids. Pma1p has relatively long hydrophobic transmembrane segments and is one of the yeast plasma membrane constituents with the longest residence time [34]. Thus, this compartment seems to be a good candidate to host sphingolipids. With very long hydrocarbon chains, they exhibit tendency to form highly ordered gel domains, and relatively low levels of ergosterol [15] are putatively found in the MCP. However, whether sphingolipids are major lipid components of those domains is 3 of 24 Biomolecules 2020, 10, 871 also unproved, as is the sphingolipid enrichment in the MCC. Moreover, the labelling of sphingolipids is not straightforward, and only indirect evidence is available to support that hypothesis. Based on the structure of their headgroup, complex sphingolipids are classified in three types in yeast, inositolphosphorylceramide (IPC), MIPC (mannosylinositolphosphorylceramide), and M(IP)2C (mannosyldiinositolphosphorylceramide). M(IP)2C is the most abundant complex sphingolipid in S. cerevisiae, regardless of whether the whole cell or specifically the plasma membrane is considered [35,36]. Still, MIPC is also present in the plasma membrane of wild-type (wt) S. cerevisiae cells. The formation of terminal complex sphingolipid M(IP)2C from MIPC is catalyzed by an inositol phosphotransferase encoded by the IPT1 gene [35,36]. 1. Introduction In ipt1∆cells, despite the major changes in sphingolipid composition, that are characterized by the lack of M(IP)2C and its replacement by MIPC, the sterol composition is similar to that of wt cells [35]. Some studies suggest that this gene deletion may enhance resistance to both nystatin and miconazole [35], and confers markedly higher resistance to zymocin [37], syringomycin E [38], and the plant defensin DmAMP1 produced by Dahlia merckii seeds [39]. The authors of the latter study proposed that the antifungal peptide targets membrane patches containing M(IP)2C, although the specific contribution of each complex sphingolipid for the organization of the plasma membrane is still unclear. In this work, in an attempt to clarify some of the issues described above, we have carried out a thorough biophysical characterization of the plasma membrane in S. cerevisiae ipt1∆versus wt cells, addressing both lipid domains and membrane compartments. We found that the plasma membrane of ipt1∆cells has more compact sphingolipid-enriched domains, whereas dielectric properties that are more specifically modulated by ergosterol remained essentially unchanged. Moreover, we present evidence that in ipt1∆cells, where M(IP)2C is absent, the plasma membrane distribution of Pma1p is much more affected than that of Can1p. 2.3. Plasma Membrane Isolation The plasma membrane isolation was performed as described in [43] with minor adjustments. Briefly, cells were collected by centrifugation (5 min, 5000× g) and washed twice with 0.4 M sucrose in buffer A (25 mM imidazole HCl, pH 7.0) at 3500× g for 10 min. Harvested cells were resuspended in buffer A containing sucrose at 0.4 M and a mixture of protease inhibitors (Halt™protease inhibitor cocktail (100×) from Thermo Scientific (Waltham, MA, USA)), and lysed gently with glass beads. To obtain a crude membrane extract, lysed cells were centrifuged at low speed (530× g) for 20 min, the supernatant was recovered and centrifuged at 22,000× g for 30 min. For the total lipid extracts preparation, the pellet containing the crude membrane extract was stored, while for the plasma membrane isolation it was resuspended in 2 mL of buffer A containing the protease inhibitor mixture. Then, it was applied on the top of a 10 mL discontinuous sucrose gradient (consisting of three layers of 2.25, 1.65, and 1.1 M sucrose in buffer A). The gradient centrifugation was carried out in a CP80NX Hitachi ultracentrifuge (Berkshire, United Kingdom) for 18 h at 80,000× g. Pure plasma membrane fractions were obtained from the interface between the 2.25 and 1.65 M sucrose layers and then they were resuspended in either buffer A (for lipid extraction) or B (for fluorescence analysis; 100 mM NaH2PO4.H2O, 100 mM NaCl, 1 mM EDTA, pH 7.4) and pelleted by centrifugation at 30,000× g for 40 min. 2.1. Materials and Strains Table 1 lists the S. cerevisiae strains used in this work. The wt and ipt1∆strains were transformed using the lithium acetate procedure [40], with the integrative plasmid Ylp211-CAN1-GFP (URA3) or Ylp128-PMA1-mRFP (LEU2) kindly provided by Dr. Guido Grossmann and Professor Widmar Tanner (University of Regensburg, Germany). Table 1. Saccharomyces cerevisiae strains used in this work, genotype description, and growth media. Strain Genotype Source Growth Media wild-type (wt; BY4741) MATa; his3∆1; leu2∆0; met15∆0; ura3∆0 EUROSCARF (Frankfurt, Germany) Synthetic Complete medium (SC) ipt1∆ BY4741; YMR272c::kanMX4 EUROSCARF (Frankfurt, Germany) SC wt-Can1p-GFP BY4741; Ylp211CAN::GFP This study SC ura- wt-Pma1p-mRFP BY4741; Ylp128PMA1:mRFP This study SC leu- ipt1∆-Can1p-GFP BY4741; YMR272c::kanMX4; Ylp211CAN::GFP This study SC ura- ipt1∆-Pma1p-mRFP BY4741; YMR272c::kanMX4; Ylp128PMA1::mRFP This study SC leu- Table 1. Saccharomyces cerevisiae strains used in this work, genotype description, and growth media. 4 of 24 Biomolecules 2020, 10, 871 Yeast extract, bactopeptone, yeast nitrogen base, and bacto-agar were purchased from Difco (Hampton, NH, USA). Amino acids, nucleotides and glucose used in the synthetic complete (SC) medium, sucrose, and organic solvents (of analytical grade for lipid extraction and spectroscopic grade for lipids and probes stock solutions) were obtained from Merck (Darmstadt, Germany). Trans-parinaric acid (t-PnA) was purchased from Santa Cruz Biotech. (Santa Cruz, CA, USA), 1,6-diphenylhexatriene (DPH) from Invitrogen (Carlsbad, CA, USA), 4-[2-[6-(dioctylamino)-2-naphthalenyl]ethenyl]-1-(3-sulfopropyl)-pyridinium (di-8-ANEPPS) was obtained from Biotium (Fremont, CA, USA), 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (Rhod-DOPE) from Avanti Polar Lipids (Alabaster, AL, USA), and Ludox (TM-50 colloidal silica, 50 wt.% suspension in water) acquired from Sigma-Aldrich (St. Louis, MO, USA). The fluorescent probes were quantified spectrophotometrically using the respective molar absorption coefficients: t-PnA, ε299.4 nm ethanol = 89,000 M−1cm−1; DPH, ε350 nm methanol = 88,000 M−1cm−1; di-8-ANEPPS, ε498 nm methanol = 37,000 M−1cm−1, Rhod-DOPE, ε559 nm chloro form = 95,000 M−1cm−1 [41]. 2.2. Media and Growth Conditions S. cerevisiae cells were inoculated at an A600nm of 0.15 and cultured in SC medium (2% (w/v) glucose, 0.68% (w/v) yeast nitrogen base and amino acid composition given in [42]), except when specifically stated and for microscopy experiments using di-8-ANEPPS and Rhod-DOPE as membrane probes. In these cases, a similar growth procedure was performed using a culture medium containing 1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) glucose (YPD). Yeast cells were grown at 30 ◦C and 160 rpm overnight and reinoculated into the culture medium at an A600nm of ~0.15. Upon reaching an A600nm of 0.6, after a 5–6 h incubation, i.e., in the mid-exponential phase, cells were harvested. 2.5. Giant Unilamellar Vesicles (GUVs) Preparation GUVs were prepared by the electroformation method of Angelova et al. [47]. For GUV preparation, a 1 mM solution of the lipids extracted from yeast plasma membrane in chloroform: methanol (2:1, v/v) was used. The adequate volume of probe (di-8-ANEPPS and/ or Rhod-DOPE) stock solution was mixed to obtain a final lipid/probe ratio of 1:500 (mol:mol) and a chloroform:methanol ratio of (2:1, v/v). Then, 25 µL of this mixture, previously homogenized, was spread with a gas-tight syringe on the titanium plates of the GUV chamber. To remove any traces of organic solvent, the plates were placed on a vacuum desiccator for 30 min. The titanium chambers were assembled and filled with 700 µL of a 200 mM sucrose solution with sodium azide 15 mM, previously heated to 60 ◦C in a heat block, sealed, connected to the signal generator and placed in the heat block maintaining the system at 60 ◦C. A 10 Hz sinusoidal wave with a peak-to-peak amplitude of 0.4 V was initially applied and the voltage was increased to 2.0 V and kept overnight. The sinusoidal signal was then changed to 4 Hz frequency and 2.6 V voltage for 20 min. The vesicle suspension was kept at room temperature (24 ◦C) in the dark until use. For GUV visualization, 100 µL of each GUV suspension and 150 µL of glucose solution 200 mM were added to a well of an eight-well plastic plate from Ibidi® with glass like coverslip bottom. 2.6. Fluorescence Spectroscopy Measurements and Data Analysis 2.4. Lipid Extraction The extraction of the total lipids of the cell and of the plasma membrane was performed using a slightly modified Bligh–Dyer method [44]. The isolated plasma membrane (IPM) and the crude membrane extracts were separately incubated with 1 mL of buffer A and 19 mL of chloroform:methanol (1:2, v/v) at 20 ◦C with shaking at 140 rpm for 1 h. Chloroform was further added in order to have a chloroform:methanol ratio of 1:1 (v/v). The same volume of a 0.04% (w/v) CaCl2 solution was added to the mixture, which was then vigorously stirred for 1 min (while relieving the pressure). Once two distinct 5 of 24 Biomolecules 2020, 10, 871 phases were visible, both the organic phase and the interface were collected separately. The whole process was then repeated for the interface, from which the organic phase was collected, added to the previous one, dried under a stream of nitrogen and then stored at −20 ◦C. In plasma membrane suspensions and in lipid extracts, total phospholipid concentration was determined by phosphorus analysis [45]. Ergosterol molar proportion was assumed to be ca. 2:1 (phospholipid:ergosterol) [19,46]. Before use, total lipid extracts were redissolved in chloroform/methanol (2:1, v/v) and the adequate volume was taken to obtain suspensions of multilamellar vesicles (MLVs) in buffer B, at a final lipid concentration of ca. 0.3 mM (see [11] for further details). 2.6. Fluorescence Spectroscopy Measurements and Data Analysis Fluorescence measurements were performed with a Fluorolog Model 3.22 spectrofluorimeter from Horiba Jobin Yvon (Villeneuve D’ascq, France) at 24.0 ± 1.0 ◦C in a sample compartment with temperature-control and magnetic stirring. Harvested cells (wt and ipt1∆) were washed twice with sterile water and then suspended in buffer B. Before starting the measurements, the probes DPH or t-PnA were added to the cell suspension to a final concentration of 2 µM and incubated at 24 ◦C (room temperature) for 20 and 5 min, respectively [11]. For fluorescence measurements, the IPM suspensions were incubated with t-PnA, DPH or di-8-ANEPPS for more than 5 min, 20 min, and 1 h respectively, at room temperature, to a final ratio of probe/lipid of 1/100, for DPH and t-PnA, and 1/200 for di-8-ANEPPS. The MLV suspensions made from total lipid extracts were labelled with DPH at a probe:phospholipid ratio of 1/200. following a procedure previously described [48]. For fluorescence in the steady-state, the excitation and emission wavelengths were set to, respectively, 320 and 404 nm for t-PnA and 358 and 430 nm for DPH. The value of steady-state fluorescence anisotropy, ⟨r⟩, was determined through Equation (1): ⟨r⟩= (IVV −G × IVH) (IVV + 2G × IVH) (1) (1) in which the subscripts indicate the orientation of the excitation and emission polarizers, where V and H correspond to vertical and horizontal orientations and G is a correction for the instrument different sensitivity to each light polarization plane. An adequate blank was subtracted from each intensity reading. in which the subscripts indicate the orientation of the excitation and emission polarizers, where V and H correspond to vertical and horizontal orientations and G is a correction for the instrument different sensitivity to each light polarization plane. An adequate blank was subtracted from each intensity reading. Time-resolved fluorescence measurements were carried out by the single photon timing technique, using a nanoLED N-320 from Horiba Jobin Yvon (Villeneuve D’ascq, France) for the excitation of t-PnA, 6 of 24 Biomolecules 2020, 10, 871 and 404 nm as the emission wavelength. The scattering agent used to obtain the instrument response function (IRF) was Ludox®. The experimental fluorescence intensity decays were analyzed with the software TRFA® version 1.4 (Minsk, Belarus). To completely separate the emission of the probe from the autofluorescence of the cells, a method of global analysis was used [49]. 2.6. Fluorescence Spectroscopy Measurements and Data Analysis The fluorescence intensity decays of the probe were described by a multiexponential model according to Equation (2): I(t) = Xn i=1 αi exp(−t τi ). (2) (2) After subtracting the contribution of the cells autofluorescence to the pre-exponential factor in each component i with lifetime τi, and after the thus obtained pre-exponential factors were normalized to obtain their relative amplitude αi. Equation (3) was used to calculate the mean fluorescence lifetime (intensity weighted): 2 ⟨τ⟩= P αiτ2 i P αiτi , (3) (3) and Equation (4) was used to compute the amplitude-weighted mean fluorescence lifetime: and Equation (4) was used to compute the amplitude-weighted mean fluorescence lifetime τ = Xn i=1 αiτi. (4) (4) The criteria for good quality of the fitting were the reduced global χ2 close to 1 and the random distributions of weighted residuals and of residuals autocorrelation (see supplemental Figure S1). The criteria for good quality of the fitting were the reduced global χ2 close to 1 and the random distributions of weighted residuals and of residuals autocorrelation (see supplemental Figure S1). Changes in membrane dipole potential were evaluated in IPM suspensions using the ratio of the steady-state fluorescence intensity values obtained from excitation of di-8-ANEPPS at 420 nm and at 520 nm, both with emission at 635 nm [50,51]. 2.7. Fluorescence Intensity and Lifetime Imaging by Confocal Microscopy 2.7.2. Yeast Living Cells Tagged with GFP and mRFP 2.7.2. Yeast Living Cells Tagged with GFP and mRFP FLIM measurements as well as confocal images of harvested S. cerevisiae wt and ipt1∆cells with GFP-tagged Can1p and mRFP-tagged Pma1p were performed on a time-resolved confocal fluorescence microscope, model MicroTime 200, from PicoQuant GmbH (Berlin, Germany), with details of the equipment setup described in [52]. The point-spread function of the setup was measured using fluorescent beads of size 100 nm, which afforded a lateral resolution of about 400 nm. Intensity images were obtained concomitantly with FLIM measurements. GFP and mRFP were excited at 480 nm using a pulsed-laser diode and detected by a SPAD using a the longpass filter above 510 nm. The images were obtained with a scanning resolution of 0.156 µm/pixel. All images were acquired with an excitation power of ca. 0.54 kW/cm2 and further processed by creating a mask in which the cell membrane regions were visually selected using the region-of-interest (ROI) tools of SymPhoTime program, version 5.3.2.2. The fluorescence decays corresponding to the regions identified as cell membranes were obtained from the histograms of photon arrival times concerning only the pixels contained in these regions. Here, the same criteria as in Section 2.7.1. for the analysis of the FLIM data of proteins were employed. The results are averaged over ca. 400 cells from 3 independent experiments. Confocal images were used to assess the distribution of Can1p-GFP and Pma1p-mRFP in the plasma membrane of yeast living cells. For numerical data extraction a line profile of the fluorescence intensity of the plasma membrane was obtained using Image Pro-Plus v6.0 and the fluorescence heterogeneity, i.e., the fraction of pixels which have intensity higher or lower than the average plus or minus 10%, respectively (or plus or minus the standard deviation—both calculations yield the same relative results when comparing proteins or strains). It reflects to which extent the fluorescence intensity is concentrated in specific pixels and is independent on the total fluorescence intensity (see supplemental Figure S2). The heterogeneity was determined for the plasma membrane of each cell. For each transformed yeast strain, 4 independent experiments were performed, and more than 190 cells were analyzed. 2.8. Statistical Analysis The results are presented as mean ± standard deviation (S.D.), unless stated otherwise, and the sample dimension and number of biologically independent replicates are given with the results. The statistical significance was determined using Student’s t-test. Mean values were considered significantly different for p values below or equal to 0.05. 2.7.1. GUV and Living Cells Labeled with Di-8-ANEPPS or Rhod-DOPE For fluorescence lifetime imaging microscopy (FLIM) with di-8-ANEPPS, after cell harvesting, the cellular suspension was labeled at room temperature (24 ◦C) with the probe at 1 µM, which was added from a concentrated methanol stock solution. After a 5 min incubation with the probe, the cells were centrifuged and suspended in buffer B. Intensity images as well as FLIM measurements were taken in the confocal mode using an Olympus Fluoview 1000 (Olympus, Tokyo, Japan) upgraded with a LSM time-resolved kit (PicoQuant GmbH, Berlin, Germany) and an oil immersion objective, ×60 (1.35 N.A.). For intensity images, Di-8-ANEPPS was excited by a 488 nm line of a multiline argon laser and its emission detected in the range of 565–665 nm. Rhod-DOPE intensity images were acquired with excitation at 559 nm using a line of a multiline argon laser and detection in the range of 590–690 nm. Images with a frame size of 512 × 512 pixels were acquired. For FLIM measurements, di-8-ANEPPS was excited at 483 nm and Rhod-DOPE at 560 nm using pulsed-laser diodes and detected by a single photon avalanche photodiode (SPAD) using a 620 ± 30 nm bandpass filter. FLIM images were acquired for 60 s (45 frames with an average photon count rate of 2–4 × 104 count/s). Analysis of FLIM data was performed using the software from PicoQuant, SymPhoTime 64, (version 2.5), considering the convolution of the fluorescence decays with the IRF. The time distribution of photon counts in each pixel retrieved from the regions of interest were combined into a single decay curve which was analyzed by using a 2 exponentials model (Equation (2) with i = 2), judging the fitting quality from the distribution of the residuals and reduced χ2 value. The intensity-weighted mean fluorescence lifetime τ and the amplitude weighted mean fluorescence lifetime τ were calculated using Equations (3) and (4). The results are averaged over more than 50 GUV from at least 3 independent experiments and more than 400 cells from 4 independent experiments. The fluorescence lifetime of the probe was independent of the incubation time. 7 of 24 7 of 24 Biomolecules 2020, 10, 871 3.1. Sphingolipid-Enriched Domains are more Compact in ipt1∆Cells of at least four biological replicates. ** p ≤0.01. Regarding the relative abundance of the sphingolipid-enriched domains, the amplitude of the long lifetime component was somewhat lower in ipt1∆cells, but the differences were not statistically significant. Taken together the results of packing and abundance of the sphingolipid-enriched domains contribute to the identical values obtained for the steady state fluorescence anisotropy of t-PnA in living cells or IPM of S. cerevisiae wt and ipt1∆cells (Supplementary Figure S4). 3.1. Sphingolipid-Enriched Domains are more Compact in ipt1∆Cells To understand how sphingolipid-enriched gel domains may change with an alteration of the sphingolipid profile, the plasma membrane of S. cerevisiae living cells was labelled with t-PnA. The fluorescence properties of this probe, especially the time-resolved, are sensitive to changes in the amount and composition of highly ordered gel domains, into which it partitions very favorably in membrane model systems, undergoing a concomitant increase of its fluorescence quantum yield [48,53,54]. Additionally, the long lifetime component in t-PnA fluorescence intensity decay can be used as a fingerprint for gel domains when longer than 30 ns [11]. The fluorescence intensity decay of t-PnA incorporated in the plasma membrane of S. cerevisiae cells is properly described by four exponentials, where the long component lifetime (τlong) gives information about the relative degree of hydrophobic packing of gel domains and its amplitude (normalized pre-exponential) about their relative abundance [55]. For wt cells, the long lifetime component was 40.9 ± 0.4 ns, a value similar to that obtained in our previous work (~41 ns) [11], while for ipt1∆cells it was 45.1 ± 2.4 ns (Figure 1–τlong). Since this parameter is higher than 30 ns for both strains, it reveals the presence of gel domains in the plasma membrane [11]. Moreover, both the long-lifetime component and the intensity weighted mean fluorescence lifetime (Figure 1—⟨τ⟩) 8 of 24 Biomolecules 2020, 10, 871 were even longer in ipt1∆cells than wt cells. Similar results were obtained for cells grown in YPD medium (Supplementary Material Figure S3). Figure 1. The gel lipid domains of the plasma membrane of intact S. cerevisiae cells are more compact in ipt1∆than in wt cells. The fluorescence intensity decay of the probe t-PnA in S. cerevisiae cellular suspensions was obtained at 24 ◦C, as described in the “experimental procedures”. The long-component lifetime (τlong) and normalized amplitude (αlong; see inset) and the mean fluorescence lifetime (⟨τ⟩) are shown. The values are the mean ± S.D. of at least four biological replicates. ** p ≤0.01. Figure 1. The gel lipid domains of the plasma membrane of intact S. cerevisiae cells are more compact in ipt1∆than in wt cells. The fluorescence intensity decay of the probe t-PnA in S. cerevisiae cellular suspensions was obtained at 24 ◦C, as described in the “experimental procedures”. The long-component lifetime (τlong) and normalized amplitude (αlong; see inset) and the mean fluorescence lifetime (⟨τ⟩) are shown. The values are the mean ± S.D. 3.2. Membrane Fluidity Is Altered in ipt1∆Cells To further understand how the plasma membrane responds to alterations in the sphingolipid profile, the membrane of S. cerevisiae was labelled with DPH. Since DPH partitions equally throughout different lipid phases, it is sensitive to the membrane global properties, rather than to a particular type of domain [56,57], with the notable exception of poorly hydrated ceramide-rich phases [54]. DPH steady-state fluorescence anisotropy, ⟨r⟩, is a well-established parameter to report alterations in the global fluidity of the plasma membrane [11,58,59]. The fluorescence anisotropy of DPH changes linearly with composition (and temperature) along the gel/ liquid disordered (ld) and ld/ liquid ordered (lo) phase coexistence region, i.e., responds linearly to the fraction of each phase, in several binary and ternary lipid systems [12,54,58], including in the 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)/ergosterol binary system [60]. Moreover, in the POPC/phytoceramide (the yeast sphingolipid backbone) system, the fluorescence anisotropy and intensity of DPH behave similarly to typical gel/ ld systems in the region of coexistence of ld and a phytoceramide-rich gel phase where the hydrophobic packing is comparable to the one found for the gel domains in yeast [48]. All these studies support the use of DPH anisotropy as a reporter of global membrane fluidity in yeast membranes. As can be seen in Figure 2, for both the IPM suspensions and total lipid extracts reconstituted into multilamellar liposomes the steady-state anisotropy, ⟨r⟩, of DPH was significantly lower for ipt1∆cells than for wt cells, but the values of ⟨r⟩for IPM and total lipid extracts were similar for each strain. 9 of 24 Biomolecules 2020, 10, 871 Figure 2. The global order of the plasma membrane of S. cerevisiae is higher in wt than in ipt1∆ cells. The diphenylhexatriene (DPH) steady-state fluorescence anisotropy at 24 ◦C was obtained as described under “experimental procedures” for wt and ipt1∆cells isolated plasma membrane (IPM) and multilamellar vesicles (MLVs) made from total lipid extracts obtained from cells in mid-exponential phase. The values are the mean ± S.D. of at least four biological replicates. * p ≤0.05. Figure 2. The global order of the plasma membrane of S. cerevisiae is higher in wt than in ipt1∆ cells. The diphenylhexatriene (DPH) steady-state fluorescence anisotropy at 24 ◦C was obtained as described under “experimental procedures” for wt and ipt1∆cells isolated plasma membrane (IPM) and multilamellar vesicles (MLVs) made from total lipid extracts obtained from cells in mid-exponential phase. The values are the mean ± S.D. 3.2. Membrane Fluidity Is Altered in ipt1∆Cells of at least four biological replicates. * p ≤0.05. In conclusion, the results suggest that the lack of M(IP)2C in ipt1∆cells and/or the accumulation of MIPC [35,36] may lead to an alteration of the global fluidity of cellular membranes. In conclusion, the results suggest that the lack of M(IP)2C in ipt1∆cells and/or the accumulation of MIPC [35,36] may lead to an alteration of the global fluidity of cellular membranes. 3.3. Plasma Membrane Fluid Domains Present Similar Properties in wt and ipt1∆cells Electrochromic dyes, such as di-8-ANEPPS, are suitable to report sterol-related properties [50,61]. In fact, the partition of another probe from ANEPPS family, di-4-ANEPPS, for the lo phase formed in the presence of ergosterol is higher (~2.1×) than for the corresponding ld phase in the same POPC/ergosterol phase coexistence tie-line. The probe also displays a higher fluorescence quantum yield in the lo phase [62]. On the other hand, as already mentioned, t-PnA partitions preferentially towards gel domains [11,53,54]. Thus, complementary information can be retrieved by labelling the membranes with di-8-ANEPPS [59]. Unlike t-PnA, di-8-ANEPPS can be used in fluorescence microscopy studies due to its excitation and emission in the visible range and photostability. The imaging techniques enable to directly conclude from the labelling pattern if the probe is homogeneously distributed in the membrane or if brighter regions can be readily distinguished. We also used FLIM, which allows to selectively analyze the fluorescence lifetime for each region of interest (ROI) of the plasma membrane of living cells. In addition to intact cells, GUV reconstituted from plasma membrane lipid extracts pertaining to either wt or ipt1∆cells were also used. These reconstituted systems were double-labelled both with di-8-ANEPPS and with Rhod-DOPE. Rhod-DOPE was only used to label GUVs since when added externally to an aqueous solution it will not incorporate efficiently into the lipid bilayer [49]. Rhod-DOPE, due to the unsaturated acyl chains of DOPE moiety, partitions preferentially towards ld phase versus both gel [63] and ergosterol-induced lo [62] domains. In a three-phase coexistence situation, it will be mostly excluded from the gel, but still with some partition into the lo regions of the membrane [63]. Thus, since di-8-ANEPPS labels preferentially the lo phase and Rhod-DOPE the ld phase, their distribution in the membrane allows to evaluate the phase behavior of the GUV prepared from the plasma membrane lipid extracts. Both GUV and living cells were analyzed by confocal microscopy and FLIM (Figure 3). 10 of 24 Biomolecules 2020, 10, 871 Figure 3. S. cerevisiae plasma membrane lipids have complex phase behavior. Representative confocal microscopy images of giant unilamellar vesicles (GUVs) reconstituted from plasma membrane lipid extracts obtained from wt cells (A,B) or ipt1∆cells (C,D) double-labelled with both di-8-ANEPPS and Rhod-DOPE. Images from di-8-ANEPPS channel (A,C) and Rhod-DOPE channel (B,D) are shown. White arrows indicate regions that were poorly labelled by both probes. 3.3. Plasma Membrane Fluid Domains Present Similar Properties in wt and ipt1∆cells The blue arrow points to a region that was poorly labelled by Rhod-DOPE but not at all by di-8-ANEPPS. Figure 3. S. cerevisiae plasma membrane lipids have complex phase behavior. Representative confocal microscopy images of giant unilamellar vesicles (GUVs) reconstituted from plasma membrane lipid extracts obtained from wt cells (A,B) or ipt1∆cells (C,D) double-labelled with both di-8-ANEPPS and Rhod-DOPE. Images from di-8-ANEPPS channel (A,C) and Rhod-DOPE channel (B,D) are shown. White arrows indicate regions that were poorly labelled by both probes. The blue arrow points to a region that was poorly labelled by Rhod-DOPE but not at all by di-8-ANEPPS. Confocal images of double-labelled GUVs prepared with plasma membrane lipid extracts from wt or ipt1∆cells revealed that the fluorescence intensity of Rhod-DOPE and di-8-ANEPPS was not uniform throughout the GUV membrane. There were regions where the Rhod-DOPE signal was brighter, which indicated the presence of ld domains. The same was observed for di-8-ANEPPS, which suggested the presence of regions where lo predominates. Probe distribution throughout the membrane was clearly heterogeneous and despite the existence of regions where either the Rhod-DOPE or the di-8-ANEPPS signal was brighter it was not possible to accurately determine the relative proportion of ld and lo since these probes label most of the membrane, though with different intensity. Eventually, the size of the domains may be below the resolution of confocal fluorescence microscopy. Nonetheless, there were regions that were not labelled at all. These dark regions reveal lipid domains from which both probes were excluded. Both these probes are efficiently excluded from gel domains in coexistence with ld and lo domains but have different distribution between ld and lo phases [49,64,65]. Therefore, those dark regions that are simultaneously unlabeled by both probes likely correspond to rigid gel domains, in agreement with the results obtained from the fluorescence spectroscopy of t-PnA (Figure 1). FLIM experiments were carried out both on GUV and living yeast cells to study in more detail the biophysical properties of the plasma membrane regions probed by di-8-ANEPPS in wt versus ipt1∆cells. In FLIM, the microscopy image generated is based on the fluorescence lifetime rather than fluorescence intensity. Electrochromic dyes such as ANEPPS display a complex relaxation kinetics, which translates into a multiexponential fluorescence decay. 3.3. Plasma Membrane Fluid Domains Present Similar Properties in wt and ipt1∆cells It is highly dependent on the density of dipoles and dielectric constant of the membrane interfacial region, polarity properties that can be strongly affected by the presence of sterols [61,66]. In the case of cells, ROI corresponding to the plasma membrane were selected (Figure 4A–D). The parameters describing the fluorescence intensity decays, namely the normalized pre-exponentials and the fluorescence lifetimes of the two 11 of 24 Biomolecules 2020, 10, 871 exponential components, as well as the amplitude-weighted and intensity-weighted mean fluorescence lifetimes are presented in Table 2. Figure 4. Di-8-ANEPPS parameters are identical for wt and ipt1∆cells in the different systems studied. Representative fluorescence lifetime imaging microscopy (FLIM) images of living S. cerevisiae wt (A,B) and ipt1∆cells (C,D) labelled with di-8-ANEPPS. Panels (A,C) show representative intensity images, whereas in panels (B,D) the region-of-interest (ROI) selecting the plasma membrane is represented with the lifetime color scale obtained by FLIM. Lifetime components in GUVs obtained from plasma membrane lipids of wt and ipt1∆cells and in living cells, (E) lifetime of the short component, (F) lifetime of the long component and (G) amplitude of the long component. The amplitude of the short component is 1-αlong. The values are the mean ± S.D. of at least three biological replicates with a total of at least 50 GUVs and 400 cells analyzed per strain. ***, p ≤0.001. In panel (H) Rex (420 nm/520 nm), which is linearly dependent on the membrane dipole potential, is shown for IPM suspensions labeled with di-8-ANEPPS (n ≥3). Figure 4. Di-8-ANEPPS parameters are identical for wt and ipt1∆cells in the different systems studied. Representative fluorescence lifetime imaging microscopy (FLIM) images of living S. cerevisiae wt (A,B) Figure 4. Di-8-ANEPPS parameters are identical for wt and ipt1∆cells in the different systems studied. g p p y Representative fluorescence lifetime imaging microscopy (FLIM) images of living S. cerevisiae wt (A,B) and ipt1∆cells (C,D) labelled with di-8-ANEPPS. Panels (A,C) show representative intensity images, whereas in panels (B,D) the region-of-interest (ROI) selecting the plasma membrane is represented with the lifetime color scale obtained by FLIM. Lifetime components in GUVs obtained from plasma membrane lipids of wt and ipt1∆cells and in living cells, (E) lifetime of the short component, (F) lifetime of the long component and (G) amplitude of the long component. The amplitude of the short component is 1-αlong. The values are the mean ± S.D. 3.3. Plasma Membrane Fluid Domains Present Similar Properties in wt and ipt1∆cells of at least three biological replicates with a total of at least 50 GUVs and 400 cells analyzed per strain. ***, p ≤0.001. In panel (H) Rex (420 nm/520 nm), which is linearly dependent on the membrane dipole potential, is shown for IPM suspensions labeled with di-8-ANEPPS (n ≥3). 12 of 24 Biomolecules 2020, 10, 871 Table 2. Fluorescence intensity decay parameters of di-8-ANEPPS incorporated in S. cerevisiae wt and ipt1∆cells at 24 ◦C. The fluorescence intensity decays measured in cells were described by the sum of two exponentials, with amplitudes α1 and α2 and lifetimes τ1 and τ2. τav is the amplitude-weighted mean fluorescence lifetime and ⟨τ⟩is the intensity-weighted mean fluorescence lifetime (Equations (3) and (4)). The values are the mean ± S.D. of more than 50 GUVs from three biological replicates and ca. 400 cells from four biological replicates. Membrane System Strain τ1 (ns) α1 τ2 (ns) α2 ¯τ (ns) ⟨τ⟩(ns) GUVs wt 1.48 ± 0.33 0.63 ± 0.12 3.07 ± 0.53 0.37 ± 0.12 2.05 ± 0.32 2.35 ± 0.35 ipt1∆ 1.37 ± 0.29 0.66 ± 0.09 3.11 ± 0.54 0.34 ± 0.09 1.95 ± 0.33 2.30 ± 0.37 Living cells wt 1.43 ± 0.40 0.38 ± 0.07 3.00 ± 0.31 0.62 ± 0.07 2.43 ± 0.12 2.70 ± 0.06 ipt1∆ 1.37 ± 0.20 0.36 ± 0.06 3.03 ± 0.07 0.64 ± 0.06 2.43 ± 0.15 2.69 ± 0.07 The lifetimes of the components describing the fluorescence decays obtained in GUV and in living cells are identical (Figure 4E,F) though their amplitudes are not (Figure 4G). Nonetheless, when comparing the two types of cells, wt and ipt1∆, the decay parameters obtained were all very similar with no statistical differences found either in the plasma membrane of living cells or in reconstituted GUV, with a percent variation of less than 2%, if at all. These results imply that the properties of the membrane reported by di-8-ANEPPS were not significantly affected by altering the composition of complex sphingolipids. To further confirm that the properties of ergosterol-enriched domains are not affected by changes in the sphingolipid profile, the ratiometric value Rex (420 nm/520 nm), which has a linear relationship with the membrane dipole potential [50], was determined from the excitation spectra of di-8-ANEPPS in the IPM of S. cerevisiae (Figure 4H). This parameter was not significantly different between wt and ipt1∆cells, corroborating the previous conclusion. 3.3. Plasma Membrane Fluid Domains Present Similar Properties in wt and ipt1∆cells The ⟨τ⟩of Rhod-DOPE was also measured in GUVs reconstituted from lipids of wt cells yielding a value of 2.36 ± 0.02 ns, whereas in those prepared from lipids of ipt1∆cells a value of 2.45 ± 0.04 ns was obtained. These values lay within those measured in phosphatidylcholine/ergosterol lipid bilayers, where a longer fluorescence lifetime points to a more fluid environment [62]. Thus, the fact that the fluorescence lifetime of Rhod-DOPE in GUVs made from plasma membrane lipids is slightly longer for ipt1∆cells than for wt cells, suggests a more fluid membrane in the deletion mutant, in agreement with the results obtained with the probe DPH. 3.4. The Lateral Organization of Pma1p But Not Can1p Is Dependent on the Sphingolipid Profile Considering, e.g., that the refractive index of a lipid membrane varies with the length and packing of the lipid hydrophobic chains [71] and also with polar headgroup chemistry [72], the change obtained in the mean fluorescence lifetime is particularly relevant since it would represent a variation equivalent to ~0.03 in terms of refractive index, or an increase of 20% (v/v) of glycerol in a water/glycerol mixture for a GFP in solution [70], which, in turn, is also comparable e.g., to a change of about 6 carbons in the length of a fatty acid chain monolayer [71]. Figure 5. The lack in M(IP)2C leads to a change in the microenvironment surrounding Pma1p but not Can1p. Representative FLIM images of ROI comprising S. cerevisiae wt cells and ipt1∆cells plasma membrane. Intensity-weighted mean fluorescence lifetime for S. cerevisiae cells. (A) wt-Can1p-GFP and ipt1∆-Can1p-GFP cells and (B) wt-Pma1p-mRFP and ipt1∆Pma1p-mRFP cells. The values are the mean ± S.D. of three independent biological replicates with a total of ca. 400 cells analyzed for each strain. **** p ≤0.0001. Figure 5. The lack in M(IP)2C leads to a change in the microenvironment surrounding Pma1p but not Can1p. Representative FLIM images of ROI comprising S. cerevisiae wt cells and ipt1∆cells plasma membrane. Intensity-weighted mean fluorescence lifetime for S. cerevisiae cells. (A) wt-Can1p-GFP and ipt1∆-Can1p-GFP cells and (B) wt-Pma1p-mRFP and ipt1∆Pma1p-mRFP cells. The values are the mean ± S.D. of three independent biological replicates with a total of ca. 400 cells analyzed for each strain. **** p ≤0.0001. The heterogeneity of Can1p and Pma1p distribution in the plasma membrane of S. cerevisiae was also assessed (see Section 2.7.2) to further evaluate the possible influence of the complex sphingolipids in the yeast membrane compartments MCC and MCP. The distribution heterogeneity was determined by confocal fluorescence microscopy [15,23], as described in the experimental section. These fusion proteins have been used previously to visualize and detect alterations in both membrane compartments in S. cerevisiae [15,23], or similar proteins, e.g., Pma1p-GFP in other fungi such as Neurospora Crassa [73] or Pma1p tagged with hemagglutinin in C. neoformans [28]. In Figure 6, representative confocal images for the two strains labeled with either Pma1p-mRFP or Can1p-GFP are shown, as well as values obtained for the heterogeneity of their distribution in the plasma membrane for all the replicates. 3.4. The Lateral Organization of Pma1p But Not Can1p Is Dependent on the Sphingolipid Profile In order to evaluate the impact of changing sphingolipid composition of S. cerevisiae plasma membrane in the organization of MCC and MCP, the time-resolved fluorescence properties of Can1p tagged with GFP (Can1p-GFP) and Pma1p tagged with mRFP (Pma1p-mRFP) were studied by FLIM both in living wt and ipt1∆cells. As already mentioned, Can1p is found in MCC domains that co-localize with ergosterol staining by filipin, whereas Pma1p is the hallmark of MCP which seems to be intimately related with sphingolipids, but with no direct evidence for sphingolipid enrichment [3]. The fluorescence decay parameters of the proteins mentioned above were collected for both wt and ipt1∆cells (Figure 5 and Table S1). The decays are bi-exponential as expected when exciting GFP using a wavelength outside the absorption band of the protonated form and detecting emission above 500 nm [67–69]. Moreover, the lifetime and amplitude values obtained for each component of the fluorescence decay are in line with those reported in the literature [67–69]. The mean fluorescence lifetime obtained for GFP suggests that the protein is close to a medium with a high refractive index [70] as anticipated considering the membrane location of Can1p. For Can1p-GFP no differences were found between wt cells and ipt1∆cells. On the contrary, for Pma1p-mRFP significant differences in the fluorescence decay parameters were found between the two strains studied (Supplementary Table S1). A statistically significant decrease was also observed in the intensity-weighted mean fluorescence 13 of 24 13 of 24 Biomolecules 2020, 10, 871 lifetime, from 2.09 ± 0.04 ns in wt cells to 1.96 ± 0.07 ns in ipt1∆cells (Figure 5B). This corresponds to, approximately, a 6% variation. This alteration reflects a different environment surrounding the protein in the two strains, which could be related to changes in the membrane refractive index, or to alterations in the pH or local ion concentration in the vicinity of the protein. 3.4. The Lateral Organization of Pma1p But Not Can1p Is Dependent on the Sphingolipid Profile On the left, representative confocal fluorescence images of living S. cerevisiae cells with or without deletion of IPT1 and expressing either Can1p-GFP or Pma1p-mRFP, as indicated, are shown. 3.4. The Lateral Organization of Pma1p But Not Can1p Is Dependent on the Sphingolipid Profile The fact that the heterogeneity values for Can1p and Pma1p distribution in the plasma membrane in wt cells are very similar, 0.68 ± 0.10 and 0.66 ± 0.12 respectively, can be due to the reported complementarity of the two membrane compartments. The heterogeneity of Can1p distribution is only slightly—if at all—affected by 3% in ipt1∆cells relatively to wt cells (Figure 6A). In contrast, the heterogeneity of Pma1p distribution in the plasma membrane is 15% higher in ipt1∆cells than in wt cells (Figure 6B), a prominent change between 14 of 24 Biomolecules 2020, 10, 871 strains especially when compared to the one found for Can1p. Interestingly, the quantification of fluorescence intensity at the plasma membrane (see supplementary Figure S5), reveals that Can1p-GFP has slightly higher average intensity in ipt1∆plasma membrane than in the wt, whilst for Pma1p-mRFP, the intensity is significantly weaker in ipt1∆cells. This difference in fluorescence intensity between strains indicates a change in the amount of protein at the plasma membrane, since the fluorescence quantum yield of Pma1p-RFP is only marginally lower in ipt1∆cells, according to the measured amplitude-weighted mean fluorescence lifetimes (Supplementary Table S1) and pinpoints a potential defect in Pma1p accumulation at the plasma membrane in ipt1∆cells. The mechanism underlying the lower Pma1p-mRFP levels at the plasma membrane of ipt1∆cells can be related to alterations in the protein traffic or turnover or protein expression levels. Figure 6. MCP distribution in the plasma membrane is different in wt cells and ipt1∆cells. The heterogeneity of Can1p-GFP and Pma1p-mRFP distribution along the plasma membrane was determined for wt cells (A) and ipt1∆cells (B) as described in the experimental procedures. The values are the median ± S.D. of at least four biological replicates with a total of at least 190 cells analyzed per replicate. **, p ≤0.01; ****, p ≤0.0001. On the left, representative confocal fluorescence images of living S. cerevisiae cells with or without deletion of IPT1 and expressing either Can1p-GFP or Pma1p-mRFP, as indicated, are shown. Figure 6. MCP distribution in the plasma membrane is different in wt cells and ipt1∆cells. The heterogeneity of Can1p-GFP and Pma1p-mRFP distribution along the plasma membrane was determined for wt cells (A) and ipt1∆cells (B) as described in the experimental procedures. The values are the median ± S.D. of at least four biological replicates with a total of at least 190 cells analyzed per replicate. **, p ≤0.01; ****, p ≤0.0001. 4. Discussion Sphingolipid-enriched gel domains in the plasma membrane of yeast [11] have been proposed to act as diffusion barriers [9] possibly contributing to the stability of the plasma membrane compartments [48] and to the separation of mother and daughter cell components during S. cerevisiae cell division that occurs at the budding neck [74]. Another possible function might relate to cell viability in the absence of a cell wall since in the filamentous fungus N. crassa, a much tighter packing of sphingolipid-enriched domains occurs in the cell wall-less slime strain when compared to the wt strain under similar growth conditions [59] and membrane compartmentalization of lipids and proteins is interconnected to the cell wall and responds to cell wall stress [4]. Although it has been shown that sphingolipids are crucial for the formation and properties of such gel domains [10,11], it is still unclear if these domains contain a major sphingolipid class or if they are composed by a mixture of different sphingolipids. M(IP)2C is Biomolecules 2020, 10, 871 15 of 24 the most abundant complex sphingolipid in S. cerevisiae, particularly at the plasma membrane [35,75], but experimental support allowing the unequivocal identification of the components of gel domains is missing. Considering the numerous relations between sphingolipids and/or C26 very long chain fatty acids with Pma1p [27,76], it is reasonable to predict that sphingolipid-enriched regions (gel domains) and MCP may co-localize at the yeast plasma membrane, but no direct evidence to confirm this hypothesis is available. Moreover, the impact of the sphingolipid headgroup on Pma1p distribution and microenvironment in the plasma membrane of S. cerevisiae has not been elucidated. Recent observations support the hypothesis that MCC domains are ergosterol-enriched [7]. However, it is still unknown if complex sphingolipids, particularly M(IP)2C, play a direct role in the organization of the MCC [3,15]. To address these questions, wt cells and ipt1∆cells were compared regarding plasma membrane biophysical properties and distribution and organization of proteins referring to distinct membrane compartments. The deletion of the IPT1 gene, which leads to the lack of M(IP)2C and its replacement by MIPC [35,36], allows to evaluate how changing the polar headgroup of complex sphingolipids influences the different plasma membrane lipid domains, namely, the sphingolipid-enriched gel domains, and also how it impacts on protein distribution and microenvironment. Thus, we aimed at establishing a more direct correlation between sphingolipid-enriched domains and MCP or MCC. 4. Discussion We showed in this work that the formation of gel domains was not hampered by the absence of M(IP)2C as the relative abundance of these domains in the plasma membrane was similar for both S. cerevisiae wt cells and ipt1∆cells. In fact, in ipt1∆cells, plasma membrane gel domains are even more rigid than those found in wt cells, as reported by t-PnA time-resolved fluorescence. This can be rationalized considering that M(IP)2C is a constituent of gel domains and in ipt1∆cells is replaced by MIPC in those domains. The lack of a second inositol phosphate group makes the polar head of MIPC smaller and less charged than that of M(IP)2C. This may allow a tighter packing of these complex sphingolipids due to less electrostatic repulsion and/or steric hindrance at the headgroup level, which may result in a tighter packing and a more rigid environment at the hydrophobic membrane core. A similar proposal has been made to explain the tighter packing of gel domains found in the plasma membrane of scs7∆cells, which lack the 2-OH group in the fatty acyl chain of all sphingolipids, when compared with wt cells [11]. On the other hand, the polar headgroups of M(IP)2C and MIPC may confer different average transversal positions to the sphingolipid amide group and sphingolipid backbone hydroxyl groups, which leads to distinct H-bonding network at the membrane surface [77] affecting the compactness of gel domains. The biophysical differences reported in this work between wt and ipt1∆cells suggest that M(IP)2C is an important component of gel domains in wt cells. If sphingolipid-enriched domains were mainly composed by MIPC only, and independent of M(IP)2C, then in ipt1∆cells, which have a higher concentration of MIPC in the plasma membrane [46,78], we would observe an increase in the abundance of sphingolipid-enriched domains, but no changes in their packing. However, what we found is the opposite, i.e., an increased packing and either no changes or a very small decrease in the abundance of sphingolipid-enriched domains. Thus, in ipt1∆cells the sphingolipid-enriched domains are probably enriched in MIPC, as compared to wt cells. This can be due either to a direct effect, i.e., the absence of M(IP)2C and its replacement by MIPC or an indirect effect that results from changing the balance of interactions between sphingolipids, including IPC, or between sphingolipids and other membrane lipids. 4. Discussion The loss of the IPT1 gene leads to an increase in calcium tolerance [36,79], and differential drug resistance/sensitivity phenotypes [80], or an impaired uptake of miconazole [81,82]. Such results indicate that M(IP)2C differentially affects membranes permeability or the activity of membrane proteins [80], which suggests that different direct/indirect mechanisms may be at play. It has also been suggested that in ipt1∆S. cerevisiae cells, lowered levels of M(IP)2C or altered levels of other sphingolipids may regulate the activity of some ABC transporters such as Pdr5p and Yor1p, which have been associated with controlling phospholipid content or distribution at the plasma membrane [83,84]. 16 of 24 16 of 24 Biomolecules 2020, 10, 871 Thus, alterations of membrane components other than sphingolipids could be responsible, e.g., for the decreased global fluidity of the plasma membrane in ipt1∆cells. Sphingolipid-enriched gel domains have been proposed to serve as diffusion barriers [74], which may account, e.g., for the slow diffusion of plasma membrane proteins in yeast. Thus, the alteration in their rigidity may influence the diffusion properties of several membrane components in the plasma membrane [85], conformational freedom of proteins residing in these domains [86] or the partition of proteins in/out of these domains, namely glycosylphosphatidylinositol-anchored proteins [11]. Thus, alterations of membrane components other than sphingolipids could be responsible, e.g., for the decreased global fluidity of the plasma membrane in ipt1∆cells. Sphingolipid-enriched gel domains have been proposed to serve as diffusion barriers [74], which may account, e.g., for the slow diffusion of plasma membrane proteins in yeast. Thus, the alteration in their rigidity may influence the diffusion properties of several membrane components in the plasma membrane [85], conformational freedom of proteins residing in these domains [86] or the partition of proteins in/out of these domains, namely glycosylphosphatidylinositol-anchored proteins [11]. While changing the sphingolipid headgroup has a measurable impact on the compactness of gel domains, it has no obvious effect on the lipid-dependent dipolar properties of sterol-enriched domains. With di-8-ANEPPS, our results for both fluorescence lifetime and membrane dipole potential measurements, were nearly identical for wt and ipt1∆cells, regardless of the model system studied—GUVs, living cells or IPM. This suggests that those biophysical properties, which are strongly dependent on sterol and sterol-lipid interactions [51,61], are less dependent on the major sphingolipid headgroup. 4. Discussion In erg6∆cells, which have an altered sterol profile, but no noticeable changes in sphingolipid profile when compared to wt cells [87,88], the rigidity of the gel domains is not changed [11], but the fluorescence lifetime of a sterol sensitive dye is significantly different from wt cells [51]. This behavior is opposite to that found for ipt1∆cells where the rigidity of the gel domains is increased but the fluorescence properties of di-8-ANEPPS remain unchanged when compared to wt cells. Hence, sphingolipid-enriched and ergosterol-enriched regions seem to be to a certain extent independent in the plasma membrane of yeast. Those results thus support the recent observation that sphingolipids and ergosterol do not predominate in the same membrane leaflet [89]. Nonetheless, the occurrence of sphingolipid-enriched domains in the outer leaflet does not rule out the presence of ergosterol molecules in the inner leaflet of MCP. Regarding the protein markers of membrane compartments, it was found that Pma1p plasma membrane distribution heterogeneity and fluorescence lifetime are changed in ipt1∆cells when compared to wt cells, whereas the same parameters remain unaltered for Can1p. Since Pma1p- and Can1p-specific membrane compartments have been described as spatially separated [1,3], and considering that changes were observed in the biophysical properties of sphingolipid-enriched domains but not in ergosterol-enriched domains, the results obtained in this work for Pma1p and Can1p support an independent compartmentalization of sphingolipid-enriched and sterol-enriched regions, and that these regions could be lipid domain counterparts of the MCP and the MCC, respectively. Currently, it is not certain which lipids compose either MCP or MCC apart from the lower ergosterol content of the former and the higher ergosterol content of the latter [15,23,86]. However, by using a lipidomic approach with styrene-maleic-acid polymers (SMAPLs) to determine which lipids surround Can1p and Pma1p, and other MCC or MCP specific proteins (periprotein lipidome), Klooster et al. [86] demonstrated that in fact the sphingolipid/ergosterol ratio is much higher in the MCP than in the MCC, while they do not differ much in periprotein phospholipid composition. The distribution pattern of Pma1p as isolated foci in the plasma membrane has been associated to the honeycomb-like hexagonal lattices observed on the plasma membrane of S. cerevisiae [1]. Moreover, it has been suggested that the lipids contained in these structures should be very compact and display a gel arrangement [1]. 4. Discussion It is possible that each M(IP)2C molecule can recruit a higher number of glycerophospholipid molecules, not only due to its double negative charge that could interact with the positive choline moiety of phosphatidylcholine, but especially because of the large size of its polar headgroup. MIPC, which is less charged and contains a smaller headgroup compared to M(IP)2C, would recruit a lower number of glycerophospholipid molecules than M(IP)2C. Thus, in ipt1∆cells, where M(IP)2C is missing and MIPC accumulates instead, gel domains would be more rigid due to the lower content in glycerophospholipids. On another hand, these low melting temperature lipids would become more available for the formation of disordered domains, contributing to the higher global fluidity of the membrane in ipt1∆cells. Since gel domains are more rigid in ipt1∆cells, this would suggest that liquid domains are more disordered in this strain. It was previously confirmed with DPPC and POPC membranes with and without cholesterol, labelled with Rhod-DOPE at different probe/lipid ratios [62,90], that the higher the ⟨τ⟩values for Rhod-DOPE the more disordered are the membrane regions into which the probe is incorporating. Therefore, the fact that ⟨τ⟩of Rhod-DOPE in GUVs reconstituted from lipids of wt cells is lower than those prepared from lipids of ipt1∆cells supports the hypothesis that liquid regions in the plasma membrane of ipt1∆cells are more disordered when compared to wt cells. Additionally, considering the recent results on the Pma1p periprotein lipidome [86], where this region of the membrane is sphingolipid-enriched, but there seems to be some ergosterol present, one might hypothesize that in ipt1∆cells the sphingolipid-enriched domains /Pma1p periprotein lipidome become even more depleted in ergosterol, which could also lead to an increased rigidity of the domains, because ergosterol may actually fluidify these domains [11]. However, it is unlikely that the changes are due to ergosterol, because there were no changes in the fluorescence parameters of di-8-ANEPPS in ipt1∆cell membrane and, in the erg6∆cells strain lacking ergosterol mentioned above, the rigidity of the gel domains was not different from the wt. This work points to a role of the polar headgroup of complex sphingolipids in the organization of Pma1p in the plasma membrane of S. cerevisiae. Also, in C. neoformans the complex sphingolipid headgroup has been related with Pma1p oligomerization and transport [28]. By regulating the rigidity/order of gel domains, sphingolipid polar headgroups may thus contribute to control Pma1p stability, distribution, and oligomerization in the plasma membrane. 4. Discussion The results here presented, and considering the body of knowledge on the relation between sphingolipids and sterols in MCP organization, e.g., [1,26–33,86], are consistent with the presence of gel-forming sphingolipids in the composition of the MCP, providing the most simple explanation for the concomitant changes here observed in gel domains packing and Pma1p-mRFP, but not Can1p-GFP, distribution and fluorescence lifetime. The different fluorescence lifetime of Pma1p-mRFP in wt vs. ipt1∆, cells reflects a different environment for the protein. This could be due to a different refractive index in the vicinity of the protein, which could be attributed to its relocation into different domains or changes in the thickness and packing of those domains, or it could also be due to misfunction of Pma1p leading to alterations of pH and ion concentration in the vicinity of the protein. This is a subject that certainly deserves further investigation. Our data strongly suggest that the MCC 17 of 24 Biomolecules 2020, 10, 871 does not contain gel domains, since an alteration of their compactness did not affect the properties measured for Can1p. However, it should be noted that Can1p MCC partitioning is less affected than other MCC transporters by sphingolipid stress [3]. The labelling of GUV reconstituted from plasma membrane lipid extracts of wt cells or ipt1∆cells with both di-8-ANEPPS and Rhod-DOPE revealed a complex phase behavior where three distinct types of domains can be distinguished and related to a three-phase coexistence (ld/lo/gel). DPH steady-state fluorescence anisotropy was lower in the IPM of ipt1∆cells, pointing to a lower global order of the membrane, which seems to contradict the higher order of the gel domains. Therefore, it seems that the remainder of the membrane is more fluid. In particular, there could be a small increase in the relative abundance and/or fluidity of ld-like domains. Upon IPT1 deletion no alterations in lo-like domains were apparent, since fluorescence decays of di-8-ANEPPS yielded similar components lifetimes and amplitudes in ipt1∆and wt cells, and the gel domains became more rigid (and maybe slightly less abundant). Phytoceramide, the backbone of complex sphingolipids in yeast, can recruit up to 3 glycerophospholipid molecules per molecule of phytoceramide to form rigid gel phases [48]. 4. Discussion However, a specific interaction of M(IP)2C, having an extra inositol phosphate group as compared to MIPC, with the protein cannot be ruled out. If the lipid annulus surrounding the protein is different, e.g., in terms of number of lipid rings, the fraction of annular or non-annular complex sphingolipid may change, resulting in a different overall packing within the sphingolipid-enriched domains. The present work shows how the headgroup structure of complex sphingolipids can be an important determinant in the organization of plasma membrane lipids and proteins in S. cerevisiae. Recently, we showed that the polyene antifungal nystatin forms active pores in membranes lacking any sterol, but containing gel domains [12], challenging the classical model which describes the molecular 18 of 24 18 of 24 Biomolecules 2020, 10, 871 mode of action of nystatin through an ergosterol-dependent mechanism. Moreover, the action of other antifungal drugs has been shown to involve an interplay with sphingolipids that is conditioned by their headgroup structure. S. cerevisiae cells defective in M(IP)2C are resistant to zymocin [37]. Moreover, zymocin requires an active Pma1p proton pump to exert its toxicity. These observations can be related to our results supporting a close relation between MCP and sphingolipid-enriched domains. The sphingolipid headgroup also influences the antifungal activity of the lipopeptide syringomycin E [38]. Fungi lacking M(IP)2C, and accumulating MIPC, become resistant to syringomycin E. Syringomycin E acts on the membrane forming channels causing K+ efflux and Ca2+ influx. The complete absence of M(IP)2C hampers syringomycin E toxicity. Since in ipt1∆cells, gel domains become more tightly packed, this might hinder the insertion of syringomycin E acyl chain into the membrane. An independent study supporting the importance of sphingolipid-dependent biophysical properties shows that scs7∆cells, where sphingolipid-enriched domains are also more rigid than in wt cells [11], are also more resistant to the action of syringomycin E. Since the lack of M(IP)2C leads to a tightening of gel domains, as well as to an altered distribution of Pma1p, it is possible that membrane proteins, alongside lipid composition, are also important for the susceptibility to syringomycin E. This is further supported by the observation that Pma1p proper functioning is crucial for the virulent activity of pathogenic fungi [28,91,92]. 4. Discussion The apparent requirement of intact MCP/M(IP)2C domains for the action of syringomycin E is supported by the fact that cumulative conditions of MIPC accumulation and absence of M(IP)2C lead to resistance; yet higher levels of MIPC, when M(IP)2C is still produced, do not change the sensitivity phenotype [93]. A better understanding of sphingolipids role in yeast plasma membrane organization, particularly, in domain formation, may thus be an important contribute to clarify antifungal molecular targets and mode of action. Gournas and collaborators showed that sphingolipids are required for the slower diffusion of Can1p in the MCC [8] and, presented different hypothesis for the still unknown mechanism [3]. Our results seem to rule out the hypothesis that M(IP)2C is directly required for the conformation-dependent anchoring of the protein transporters in the MCC, since its absence does not alter any measured property of Can1p. Also, in this work we observed diverse membrane reorganizations upon IPT1 deletion, including changes in global plasma membrane fluidity, and yet no significant alterations were detected for Can1p, which seems to discard the hypothesis that the partitioning of Can1p into the MCC could be indirectly affected by a general alteration of plasma membrane organization. Our results did not confirm neither reject the possibility that sphingolipids depletion could lead to a concomitant decrease of ergosterol levels in eisosomes, since ergosterol levels are similar in both ipt1∆cells and wt cells and we show that some of the sterol-dependent properties are not significantly changed in the plasma membrane of ipt1∆cells when compared to wt cells. However, another scenario is still possible, where sphingolipid-enriched domains act as diffusional barriers and are neighboring the MCC, thus preventing movement of Can1p. The change in sphingolipid levels may induce an alteration of the relative arrangement of yeast compartments, indirectly affecting those that are not sphingolipid-enriched. 5. Conclusions There are still many unaddressed questions regarding lipid–protein interactions in the plasma membrane of fungi. Disclosing the mechanism of membrane compartments formation, their lipid composition and interrelations will hopefully lead to significant advances in membrane biology, but their investigation is also driven by the importance of some of these compartments in the pathogenicity of clinically relevant fungal species. The identification of gel domains under physiological conditions was an important finding but the specific molecular constitution of those domains is unknown. Four different fluorescent probes, t-PnA, DPH, di-8-ANEPPS, and Rhod-DOPE, were used to label yeast plasma membrane, in living cells and IPM and in reconstituted systems, namely GUV made from plasma membrane lipid extracts. The set of probes employed yields complementary information that enables a comprehensive picture of 19 of 24 Biomolecules 2020, 10, 871 the biophysical changes taking place at the plasma membrane upon the loss of M(IP)2C synthesis ability. With this work, it became clear that M(IP)2C is important for the regulation of gel domains in the plasma membrane of yeast, since in cells unable to synthesize this complex sphingolipid these domains are more rigid. In the future, co-localization studies with Pma1p-mRFP, and Can1p-GFP-containing wt cells [23], ipt1∆cells and other strains with mutations in sphingolipid (and sterol) metabolism will provide invaluable insights to further understand the intricate relation between membrane lipid composition, biophysical properties, and membrane compartments organization and function in yeast. p p y p p p g y Our results also help to clarify the interrelation between sphingolipids, M(IP)2C in this case, and Pma1p or Can1p. Diverse results in the literature point to a dependence of Pma1p proper sorting with intact sphingolipid synthesis. Here, we demonstrate a direct correlation between sphingolipid polar headgroup and protein distribution. Pma1p distribution is more heterogeneous in ipt1∆cells, whereas Can1p distribution, if affected at all, becomes slightly more homogeneous. This suggests that M(IP)2C is directly involved in the organization of MCP, but possibly not of MCC. The concomitant alteration in the rigidity of gel domains and Pma1p distribution supports a close communication between gel domains and MCP. On the other hand, MCC and ergosterol-enriched domains did not undergo any noticeable changes upon a major alteration of complex sphingolipid profile, which corroborates that MCC and ergosterol co-localize, further supporting that M(IP)2C is not an important constituent of ergosterol-enriched domains. 5. Conclusions Supplementary Materials: The following are available online at http://www.mdpi.com/2218-273X/10/6/871/s1, Figure S1: Fluorescence intensity decay of t-PnA in wt and ipt1∆cell suspensions at 24 ◦C. Figure S2: Relation between the mean intensity of fluorescence along the plasma membrane of each protein, Can1p or Pma1p, for each cell and respective heterogeneity. Figure S3: The lifetime of the long component of t-PnA is essentially identical in growth media SC and YPD. Figure S4: The deletion of IPT1 does not lead to significant differences between intact cells and isolated plasma membrane (IPM) in the sphingolipid-enriched domains. Figure S5: Can1p-GFP and Pma1p-mRFP mean fluorescence intensity in the plasma membrane of wt and ipt1∆cells. Table S1: Parameters describing the fluorescence intensity decays of Can1p-GFP and Pma1p-mRFP in the plasma membrane of S. cerevisiae wt and ipt1∆cells. Author Contributions: Conceptualization, R.F.M.d.A.; methodology, H.S.M., T.K., P.M.R.P. and R.F.M.d.A.; formal analysis, A.B.-O., F.C.S., J.T.M., H.S.M.; T.K., and P.M.R.P.; investigation, A.B.-O., F.C.S. and J.T.M.; resources, H.S.M., A.H., P.M.R.P. and R.F.M.d.A.; data curation, H.S.M., T.K. and P.M.R.P.; writing—original draft preparation, A.B.-O., F.C.S., J.T.M. and R.F.M.d.A.; writing—review and editing, H.S.M. and A.H.; visualization, A.B.-O., F.C.S. and J.T.M.; supervision, R.F.M.d.A.; funding acquisition, R.F.M.d.A. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Fundação para a Ciência e a Tecnologia, F.C.T., I.P., through research grants PTDC/BBB-BQB/6071/2014, UID/Multi/00612/2019 and UIDB/00100/2020, and doctoral scholarships to F.C.S. (SFRH/BD/108031/2015) and A.B.-O. (SFRH/BD/145600/2019). Acknowledgments: The authors thank Guido Grossmann and Widmar Tanner (University of Regensburg, Germany) that generously provided plasmid Ylp211-CAN1-GFP (URA3) and Ylp128-PMA1-mRFP (LEU2). We wish to thank Luísa Cyrne (Universidade de Lisboa, Portugal) for helping with the yeast cells transformation. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. References 1. Malinsky, J.; Opekarova, M. New Insight into the Roles of Membrane Microdomains in Physiological Activities of Fungal Cells. Int. Rev. Cell Mol. Biol. 2016, 325, 119–180. [CrossRef] [PubMed] 1. Malinsky, J.; Opekarova, M. New Insight into the Roles of Membrane Microdomains in Physiological Activities of Fungal Cells. Int. Rev. Cell Mol. Biol. 2016, 325, 119–180. [CrossRef] [PubMed] 2. Zahumensky, J.; Malinsky, J. Role of MCC/Eisosome in Fungal Lipid Homeostasis. Biomolecules 2019, 9, 305. [CrossRef] [PubMed] 2. Zahumensky, J.; Malinsky, J. Role of MCC/Eisosome in Fungal Lipid Homeostasis. Biomolecules 2019, 9, 305. [CrossRef] [PubMed] 3. Athanasopoulos, A.; André, B.; Sophianopoulou, V.; Gournas, C. Fungal Plasma Membrane Domains. FEMS Microbiol. Rev. 2019. [CrossRef] [PubMed] 3. Athanasopoulos, A.; André, B.; Sophianopoulou, V.; Gournas, C. Fungal Plasma Membrane Domains. FEMS Microbiol. Rev. 2019. [CrossRef] [PubMed] 4. De Almeida, R.F.M. A route to understanding yeast cellular envelope—Plasma membrane lipids interplaying in cell wall integrity. FEBS J. 2018. [CrossRef] 4. De Almeida, R.F.M. A route to understanding yeast cellular envelope—Plasma membrane lipids interplaying in cell wall integrity. FEBS J. 2018. [CrossRef] 5. Ganguly, S.; Singh, P.; Manoharlal, R.; Prasad, R.; Chattopadhyay, A. Differential dynamics of membrane proteins in yeast. Biochem. Biophys. Res. Commun. 2009, 387, 661–665. [CrossRef] 5. Ganguly, S.; Singh, P.; Manoharlal, R.; Prasad, R.; Chattopadhyay, A. Differential dynamics of membrane proteins in yeast. Biochem. Biophys. Res. Commun. 2009, 387, 661–665. [CrossRef] 20 of 24 20 of 24 Biomolecules 2020, 10, 871 6. Valdez-Taubas, J.; Pelham, H.R. Slow diffusion of proteins in the yeast plasma membrane allows polarity to be maintained by endocytic cycling. Curr. Biol. CB 2003, 13, 1636–1640. [CrossRef] 7. Bianchi, F.; Syga, Ł.; Moiset, G.; Spakman, D.; Schavemaker, P.E.; Punter, C.M.; Seinen, A.-B.; van Oijen, A.M.; Robinson, A.; Poolman, B. Steric exclusion and protein conformation determine the localization of plasma membrane transporters. Nat.Commun. 2018, 9, 501. [CrossRef] 8. Gournas, C.; Gkionis, S.; Carquin, M.; Twyffels, L.; Tyteca, D.; André, B. Conformation-dependent partitioning of yeast nutrient transporters into starvation-protective membrane domains. Proc. Natl. Acad. Sci. USA 2018, 115, E3145–E3154. [CrossRef] 9. De Almeida, R.F.; Joly, E. Crystallization around solid-like nanosized docks can explain the specificity, diversity, and stability of membrane microdomains. Front. Plant Sci. 2014, 5, 72. [CrossRef] 10. Vecer, J.; Vesela, P.; Malinsky, J.; Herman, P. Sphingolipid levels crucially modulate lateral microdomain organization of plasma membrane in living yeast. FEBS Lett. 2014, 588, 443–449. [CrossRef] 11. References Aresta-Branco, F.; Cordeiro, A.M.; Marinho, H.S.; Cyrne, L.; Antunes, F.; de Almeida, R.F. Gel domains in the plasma membrane of Saccharomyces cerevisiae: Highly ordered, ergosterol-free, and sphingolipid-enriched lipid rafts. J. Biol. Chem. 2011, 286, 5043–5054. [CrossRef] [PubMed] 12. Dos Santos, A.G.; Marquês, J.T.; Carreira, A.C.; Castro, I.R.; Viana, A.S.; Mingeot-Leclercq, M.P.; de Almeida, R.F.M.; Silva, L.C. The molecular mechanism of Nystatin action is dependent on the membrane biophysical properties and lipid composition. Phys. Chem. Chem. Phys. 2017, 19, 30078–30088. [CrossRef] [PubMed] 13. Stradalova, V.; Stahlschmidt, W.; Grossmann, G.; Blazikova, M.; Rachel, R.; Tanner, W.; Malinsky, J. Furrow-like invaginations of the yeast plasma membrane correspond to membrane compartment of Can1. J. Cell Sci. 2009, 122, 2887–2894. [CrossRef] [PubMed] 14. Walther, T.C.; Brickner, J.H.; Aguilar, P.S.; Bernales, S.; Pantoja, C.; Walter, P. Eisosomes mark static sites of endocytosis. Nature 2006, 439, 998–1003. [CrossRef] 15. Grossmann, G.; Opekarova, M.; Malinsky, J.; Weig-Meckl, I.; Tanner, W. Membrane potential governs lateral segregation of plasma membrane proteins and lipids in yeast. EMBO J. 2007, 26, 1–8. [CrossRef] 16. Berchtold, D.; Piccolis, M.; Chiaruttini, N.; Riezman, I.; Riezman, H.; Roux, A.; Walther, T.C.; Loewith, R. Plasma membrane stress induces relocalization of Slm proteins and activation of TORC2 to promote sphingolipid synthesis. Nat. Cell Biol. 2012, 14, 542–547. [CrossRef] p g p y 17. Douglas, L.M.; Konopka, J.B. Plasma membrane architecture protects Candida albicans from killing by copper. PLoS Genet. 2019, 15, e1007911. [CrossRef] 18. Douglas, L.M.; Wang, H.X.; Keppler-Ross, S.; Dean, N.; Konopka, J.B. Sur7 promotes plasma membrane organization and is needed for resistance to stressful conditions and to the invasive growth and virulence of Candida albicans. mBio 2012, 3. [CrossRef] 19. Pedroso, N.; Matias, A.C.; Cyrne, L.; Antunes, F.; Borges, C.; Malho, R.; de Almeida, R.F.M.; Herrero, E.; Marinho, H.S. Modulation of plasma membrane lipid profile and microdomains by H2O2 in Saccharomyces cerevisiae. Free Radic. Biol. Med. 2009, 46, 289–298. [CrossRef] 20. Dupont, S.; Beney, L.; Ritt, J.F.; Lherminier, J.; Gervais, P. Lateral reorganization of plasma membrane is involved in the yeast resistance to severe dehydration. Biochim. Biophys. Acta 2010, 1798, 975–985. [CrossRef] 21. Grossmann, G.; Malinsky, J.; Stahlschmidt, W.; Loibl, M.; Weig-Meckl, I.; Frommer, W.B.; Opekarova, M.; 20. Dupont, S.; Beney, L.; Ritt, J.F.; Lherminier, J.; Gervais, P. Lateral reorganization of plasma membrane is involved in the yeast resistance to severe dehydration. Biochim. Biophys. Acta 2010, 1798, 975–985. [CrossRef] 21. pp 5. Kane, P.M. Proton Transport and pH Control in Fungi. Adv. Exp. Med. Biol. 2016, 892, 33–68. [CrossRef References Grossmann, G.; Malinsky, J.; Stahlschmidt, W.; Loibl, M.; Weig-Meckl, I.; Frommer, W.B.; Opekarova, M.; Tanner, W. Plasma membrane microdomains regulate turnover of transport proteins in yeast. J. Cell Biol. 2008, 183, 1075–1088. [CrossRef] [PubMed] 22. Herman, P.; Vecer, J.; Opekarova, M.; Vesela, P.; Jancikova, I.; Zahumensky, J.; Malinsky, J. Depolarization affects the lateral microdomain structure of yeast plasma membrane. FEBS J. 2015, 282, 419–434. [CrossRef] [PubMed] 23. Malinska, K.; Malinsky, J.; Opekarova, M.; Tanner, W. Visualization of protein compartmentation within the plasma membrane of living yeast cells. Mol. Biol. Cell 2003, 14, 4427–4436. [CrossRef] [PubMed] 24. Serrano, R. Transport across yeast vacuolar and plasma membrane. In The Molecular and Cellular Biology of the Yeast Saccharomyces: Genome Dynamics, Protein Synthesis, and Energetics; Broach, J.R., Jones, E.W., Pringle, J.R., Eds.; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, USA, 1991; Volume 21, pp. 523–585. pp 25. Kane, P.M. Proton Transport and pH Control in Fungi. Adv. Exp. Med. Biol. 2016, 892, 33–68. [Cross 21 of 24 21 of 24 Biomolecules 2020, 10, 871 26. Bagnat, M.; Chang, A.; Simons, K. Plasma membrane proton ATPase Pma1p requires raft association for surface delivery in yeast. Mol. Biol. Cell 2001, 12, 4129–4138. [CrossRef] 27. Gaigg, B.; Toulmay, A.; Schneiter, R. Very long-chain fatty acid-containing lipids rather than sphingolipids per se are required for raft association and stable surface transport of newly synthesized plasma membrane ATPase in yeast. J. Biol. Chem. 2006, 281, 34135–34145. [CrossRef] 28. Farnoud, A.M.; Mor, V.; Singh, A.; Del Poeta, M. Inositol phosphosphingolipid phospholipase C1 regulates plasma membrane ATPase (Pma1) stability in Cryptococcus neoformans. FEBS Lett. 2014, 588, 3932–3938. [CrossRef] 29. Munshi, M.A.; Gardin, J.M.; Singh, A.; Luberto, C.; Rieger, R.; Bouklas, T.; Fries, B.C.; Del Poeta, M. The Role of Ceramide Synthases in the Pathogenicity of Cryptococcus neoformans. Cell Rep. 2018, 22, 1392–1400. [CrossRef] 30. Wang, Q.; Chang, A. Sphingoid base synthesis is required for oligomerization and cell surface stability of the yeast plasma membrane ATPase, Pma1. Proc. Natl. Acad. Sci. USA 2002, 99, 12853–12858. [CrossRef] 31. Lauwers, E.; Andre, B. Association of yeast transporters with detergent-resistant membranes correlates with their cell-surface location. Traffic 2006, 7, 1045–1059. [CrossRef] 32. Lauwers, E.; Grossmann, G.; Andre, B. Evidence for coupled biogenesis of yeast Gap1 permease and sphingolipids: Essential role in transport activity and normal control by ubiquitination. Mol. Biol. Cell 2007, 18, 3068–3080. [CrossRef] [PubMed] 33. Lee, M.C.; Hamamoto, S.; Schekman, R. References Ceramide biosynthesis is required for the formation of the oligomeric H+-ATPase Pma1p in the yeast endoplasmic reticulum. J. Biol. Chem. 2002, 277, 22395–22401. [CrossRef] [PubMed] 34. Ferreira, T.; Mason, A.B.; Slayman, C.W. The yeast Pma1 proton pump: A model for understanding the biogenesis of plasma membrane proteins. J. Biol. Chem. 2001, 276, 29613–29616. [CrossRef] [PubMed] 35. Leber, A.; Fischer, P.; Schneiter, R.; Kohlwein, S.D.; Daum, G. The yeast mic2 mutant is defective in the formation of mannosyl-diinositolphosphorylceramide. FEBS Lett. 1997, 411, 211–214. [CrossRef] 36. Dickson, R.C.; Nagiec, E.E.; Wells, G.B.; Nagiec, M.M.; Lester, R.L. Synthesis of mannose-(inositol-P)(2)-ceramide, the major sphingolipid in Saccharomyces cerevisiae, requires the IPT1 (YDR072c) gene. J. Biol. Chem. 1997, 272, 29620–29625. [CrossRef] [PubMed] 37. Zink, S.; Mehlgarten, C.; Kitamoto, H.K.; Nagase, J.; Jablonowski, D.; Dickson, R.C.; Stark, M.J.R.; Schaffrath, R. Mannosyl-diinositolphospho-ceramide, the major yeast plasma membrane sphingolipid, governs toxicity of Kluyveromyces lactis zymocin. Eukaryot. Cell 2005, 4, 879–889. [CrossRef] 38. Stock, S.D.; Hama, H.; Radding, J.A.; Young, D.A.; Takemoto, J.Y. Syringomycin E inhibition of Saccharomyces cerevisiae: Requirement for biosynthesis of sphingolipids with very-long-chain fatty acids and mannose- and phosphoinositol-containing head groups. Antimicrob. Agents Chemother. 2000, 44, 1174–1180. [CrossRef] 39. Thevissen, K.; Cammue, B.P.; Lemaire, K.; Winderickx, J.; Dickson, R.C.; Lester, R.L.; Ferket, K.K.; Van Even, F.; Parret, A.H.; Broekaert, W.F. A gene encoding a sphingolipid biosynthesis enzyme determines the sensitivity of Saccharomyces cerevisiae to an antifungal plant defensin from dahlia (Dahlia merckii). Proc. Natl. Acad. Sci. USA 2000, 97, 9531–9536. [CrossRef] 40. Gietz, D.; St Jean, A.; Woods, R.A.; Schiestl, R.H. Improved method for high efficiency transformation of intact yeast cells. Nucleic Acids Res. 1992, 20, 1425. [CrossRef] 41. Johnson, I.; Spence, M.T.Z. (Eds.) The Molecular Probes Handbook—A Guide to Fluorescent Probes and Technologies; Life Technologies Corporation: Carlsbad, CA, USA, 2010. 42. Branco, M.R.; Marinho, H.S.; Cyrne, L.; Antunes, F. Decrease of H2O2 Plasma Membrane Permeability during Adaptation to H2O2 in Saccharomyces cerevisiae. J. Biol. Chem. 2004, 279, 6501–6506. [CrossRef] 42. Branco, M.R.; Marinho, H.S.; Cyrne, L.; Antunes, F. Decrease of H2O2 Plasma Membrane Permeability during Adaptation to H2O2 in Saccharomyces cerevisiae. J. Biol. Chem. 2004, 279, 6501–6506. [CrossRef] 43. Panaretou, B.; Piper, P. Isolation of yeast plasma membranes. Methods Mol. Biol. 2006, 313, 27–32. [PubMed] during Adaptation to H2O2 in Saccharomyces cerevisiae. J. Biol. Chem. 2004, 279, 6501–6506. [CrossRef] 43. Panaretou, B.; Piper, P. Isolation of yeast plasma membranes. Methods Mol. Biol. 2006, 313, 27–32. 46. Daum, G.; Tuller, G.; Nemec, T.; Hrastnik, C.; Balliano, G.; Cattel, L.; Milla, P.; Rocco, F.; Conzelmann, A.; Vionnet, C.; et al. Systematic analysis of yeast strains with possible defects in lipid metabolism. Yeast 1999, 15, 601–614. [CrossRef] References [PubMed] 43. Panaretou, B.; Piper, P. Isolation of yeast plasma membranes. Methods Mol. Biol. 2006, 313, 27–32. [PubMed] 44. Bligh, E.G.; Dyer, W.J. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 1959, 37, 911–917. [CrossRef] [PubMed] 45. McClare, C.W. An accurate and convenient organic phosphorus assay. Anal. Biochem. 1971, 39, 527–530. [CrossRef] 46. Daum, G.; Tuller, G.; Nemec, T.; Hrastnik, C.; Balliano, G.; Cattel, L.; Milla, P.; Rocco, F.; Conzelmann, A.; Vionnet, C.; et al. Systematic analysis of yeast strains with possible defects in lipid metabolism. Yeast 1999, 15, 601–614. [CrossRef] 22 of 24 Biomolecules 2020, 10, 871 47. Angelova, M.I.; Soléau, S.; Méléard, P.; Faucon, F.; Bothorel, P. Preparation of giant vesicles by external AC electric fields. Kinetics and applications. In Trends in Colloid and Interface Science VI; Steinkopff: Darmstadt, Germany, 1992; pp. 127–131. 48. Marques, J.T.; Cordeiro, A.M.; Viana, A.S.; Herrmann, A.; Marinho, H.S.; de Almeida, R.F. Formation and Properties of Membrane-Ordered Domains by Phytoceramide: Role of Sphingoid Base Hydroxylation. Langmuir ACS J. Surf. Colloids 2015, 31, 9410–9421. [CrossRef] 49. Bastos, A.E.; Scolari, S.; Stockl, M.; Almeida, R.F. Applications of fluorescence lifetime spectroscopy and imaging to lipid domains in vivo. In Methods Enzymol; Conn, P.M., Ed.; Academic Press: Burlington, MA, USA, 2012; Volume 504, pp. 57–81. [CrossRef] 50. Haldar, S.; Kanaparthi, R.K.; Samanta, A.; Chattopadhyay, A. Differential Effect of Cholesterol and Its Biosynthetic Precursors on Membrane Dipole Potential. Biophys. J. 2012, 102, 1561–1569. [CrossRef] 51. Khmelinskaia, A.; Marques, J.M.; Bastos, A.E.; Antunes, C.A.; Bento-Oliveira, A.; Scolari, S.; Lobo, G.M.; Herrmann, A.; Marinho, H.S.; de Almeida, R.F. Liquid-ordered phase formation by mammalian and yeast sterols: A common feature with organizational differences. Front. Cell Dev. Biol. 2020. Accepted. [CrossRef] 52. Francisco, A.P.; Botequim, D.; Prazeres, D.M.F.; Serra, V.V.; Costa, S.M.B.; Laia, C.A.T.; Paulo, P.M.R. Extreme Enhancement of Single-Molecule Fluorescence from Porphyrins Induced by Gold Nanodimer Antennas. J. Phys. Chem. Lett. 2019, 10, 1542–1549. [CrossRef] y 53. Sklar, L.A.; Hudson, B.S.; Simoni, R.D. Conjugated polyene fatty acids as fluorescent probes: Synthetic phospholipid membrane studies. Biochemistry 1977, 16, 819–828. [CrossRef] 54. De Almeida, R.F.; Loura, L.M.; Prieto, M. Membrane lipid domains and rafts: Current applications of fluorescence lifetime spectroscopy and imaging. Chem. Phys. Lipids 2009, 157, 61–77. [CrossRef] 55. Mateo, C.R.; Brochon, J.C.; Lillo, M.P.; Acuna, A.U. Lipid Clustering in Bilayers Detected by the Fluorescence Kinetics and Anisotropy of Trans-Parinaric Acid. Biophys. J. 1993, 65, 2237–2247. References [CrossRef] 56. Davenport, L. [24] Fluorescence probes for studying membrane heterogeneity. In Methods in Enzymology; Academic Press: Cambridge, MA, USA, 1997; Volume 278, pp. 487–512. 57. Florine-Casteel, K.; Feigenson, G.W. On the use of partition coefficients to characterize the distribution of fluorescent membrane probes between coexisting gel and fluid lipid phases: An analysis of the partition behavior of 1,6-diphenyl-1,3,5-hexatriene. Biochim. Biophys. Acta (BBA) Biomembr. 1988, 941, 102–106. [CrossRef] 58. Lentz, B.R. Membrane “Fluidity” from Fluorescent Anisotropy Measurements. In Spectroscopic Membrane Probes; Loew, M.L., Ed.; CRC Press: Boca Raton, FL, USA, 1988; pp. 13–41. 59. Santos, F.C.; Fernandes, A.S.; Antunes, C.A.; Moreira, F.P.; Videira, A.; Marinho, H.S.; de Almeida, R.F. Reorganization of plasma membrane lipid domains during conidial germination. Biochim. Biophys. Acta 2017, 1862, 156–166. [CrossRef] [PubMed] 60. Silva, L.; Coutinho, A.; Fedorov, A.; Prieto, M. Competitive binding of cholesterol and ergosterol to the polyene antibiotic nystatin. A fluorescence study. Biophys. J. 2006, 90, 3625–3631. [CrossRef] [PubMed] 61. Amaro, M.; Reina, F.; Hof, M.; Eggeling, C.; Sezgin, E. Laurdan and Di-4-ANEPPDHQ probe different properties of the membrane. J. Phys. D Appl. Phys. 2017, 50. [CrossRef] [PubMed] 62. Bastos, A.E.P.; Marinho, H.S.; Cordeiro, A.M.; de Soure, A.M.; de Almeida, R.F.M. Biophysical properties of ergosterol-enriched lipid rafts in yeast and tools for their study: Characterization of ergosterol/phosphatidylcholine membranes with three fluorescent membrane probes. Chem. Phys. Lipids 2012, 165, 577–588. [CrossRef] 63. De Almeida, R.F.; Borst, J.; Fedorov, A.; Prieto, M.; Visser, A.J. Complexity of lipid domains and rafts in giant unilamellar vesicles revealed by combining imaging and microscopic and macroscopic time-resolved fluorescence. Biophys. J. 2007, 93, 539–553. [CrossRef] 64. González-Ramírez, E.J.; Goñi, F.M.; Alonso, A. Mixing brain cerebrosides with brain ceramides, cholesterol and phospholipids. Sci. Rep. 2019, 9, 13326. [CrossRef] 65. Pinto, S.N.; Fernandes, F.; Fedorov, A.; Futerman, A.H.; Silva, L.C.; Prieto, M. A combined fluorescence spectroscopy, confocal and 2-photon microscopy approach to re-evaluate the properties of sphingolipid domains. Biochim. Biophys. Acta 2013, 1828, 2099–2110. [CrossRef] 66. Fluhler, E.; Burnham, V.G.; Loew, L.M. Spectra, membrane binding, and potentiometric responses of new charge shift probes. Biochemistry 1985, 24, 5749–5755. [CrossRef] 23 of 24 23 of 24 Biomolecules 2020, 10, 871 67. Cotlet, M.; Hofkens, J.; Maus, M.; Gensch, T.; Van der Auweraer, M.; Michiels, J.; Dirix, G.; Van Guyse, M.; Vanderleyden, J.; Visser, A.J.W.G.; et al. Excited-State Dynamics in the Enhanced Green Fluorescent Protein Mutant Probed by Picosecond Time-Resolved Single Photon Counting Spectroscopy. J. References Phys. Chem. B 2001, 105, 4999–5006. [CrossRef] 68. Heikal, A.A.; Hess, S.T.; Webb, W.W. Multiphoton molecular spectroscopy and excited-state dynamics of enhanced green fluorescent protein (EGFP): Acid–base specificity. Chem. Phys. 2001, 274, 37–55. [CrossRef] 68. Heikal, A.A.; Hess, S.T.; Webb, W.W. Multiphoton molecular spectroscopy and excited-state dynamics of enhanced green fluorescent protein (EGFP): Acid–base specificity. Chem. Phys. 2001, 274, 37–55. [CrossRef] 69. Suhling, K.; Siegel, J.; Phillips, D.; French, P.M.W.; Lévêque-Fort, S.; Webb, S.E.D.; Davis, D.M. Imaging enhanced green fluorescent protein (EGFP): Acid–base specificity. Chem. Phys. 2001, 274, 37–55. [CrossRef] 69. Suhling, K.; Siegel, J.; Phillips, D.; French, P.M.W.; Lévêque-Fort, S.; Webb, S.E.D.; Davis, D.M. Imaging the Environment of Green Fluorescent Protein. Biophys. J. 2002, 83, 3589–3595. [CrossRef] 69. Suhling, K.; Siegel, J.; Phillips, D.; French, P.M.W.; Lévêque-Fort, S.; Webb, S.E.D.; Davis, D.M. Imaging the Environment of Green Fluorescent Protein. Biophys. J. 2002, 83, 3589–3595. [CrossRef] 70. Suhling, K.; Davis, D.; Petrasek, Z.; Siegel, J.; Phillips, D. Influence of the refractive index on EGFP fluorescence lifetimes in mixtures of water and glycerol. Proc. SPIE 2001, 4259, 92–101. [CrossRef] 71. Petrov, J.G.; Pfohl, T.; Möhwald, H. Ellipsometric Chain Length Dependence of Fatty Acid Langmuir Monolayers. A Heads-and-Tails Model. J. Phys. Chem. B 1999, 103, 3417–3424. [CrossRef] 72. Kienle, D.F.; de Souza, J.V.; Watkins, E.B.; Kuhl, T.L. Thickness and refractive index of DPPC and DPPE monolayers by multiple-beam interferometry. Anal. Bioanal. Chem. 2014, 406, 4725–4733. [CrossRef] 73. Fajardo-Somera, R.A.; Bowman, B.; Riquelme, M. The plasma membrane proton pump PMA-1 is incorporated into distal parts of the hyphae independently of the Spitzenkorper in Neurospora crassa. Eukaryot. Cell 2013, 12, 1097–1105. [CrossRef] 74. Clay, L.; Caudron, F.; Denoth-Lippuner, A.; Boettcher, B.; Frei, S.B.; Snapp, E.L.; Barral, Y. A sphingolipid-dependent diffusion barrier confines ER stress to the yeast mother cell. Elife 2014, 3. [CrossRef] 75. Ejsing, C.S.; Sampaio, J.L.; Surendranath, V.; Duchoslav, E.; Ekroos, K.; Klemm, R.W.; Simons, K.; Shevchenko, A. Global analysis of the yeast lipidome by quantitative shotgun mass spectrometry. Proc. Natl. Acad. Sci. USA 2009, 106, 2136–2141. [CrossRef] 76. Gaigg, B.; Timischl, B.; Corbino, L.; Schneiter, R. Synthesis of sphingolipids with very long chain fatty acids but not ergosterol is required for routing of newly synthesized plasma membrane ATPase to the cell surface of yeast. J. Biol. Chem. 2005, 280, 22515–22522. [CrossRef] 77. Marques, J.T.; Marinho, H.S.; de Almeida, R.F.M. Sphingolipid hydroxylation in mammals, yeast and plants—An integrated view. Prog. Lipid Res. References 2018, 71, 18–42. [CrossRef] [PubMed] 78. Tani, M.; Toume, M. Alteration of complex sphingolipid composition and its physiological significance in yeast Saccharomyces cerevisiae lacking vacuolar ATPase. Microbiology 2015, 161, 2369–2383. [CrossRef] [PubMed] 79. Dickson, R.C. Thematic Review Series: Sphingolipids. New insights into sphingolipid metabolism and function in budding yeast. J. Lipid Res. 2008, 49, 909–921. [CrossRef] [PubMed] 80. Hallstrom, T.C.; Lambert, L.; Schorling, S.; Balzi, E.; Goffeau, A.; Moye-Rowley, W.S. Coordinate control of sphingolipid biosynthesis and multidrug resistance in Saccharomyces cerevisiae. J. Biol. Chem. 2001, 276, 23674–23680. [CrossRef] 81. Tanaka, S.; Tani, M. Mannosylinositol phosphorylceramides and ergosterol coodinately maintain cell wall integrity in the yeast Saccharomyces cerevisiae. FEBS J. 2018, 285, 2405–2427. [CrossRef] 82. Morimoto, Y.; Tani, M. Synthesis of mannosylinositol phosphorylceramides is involved in maintenance of cell integrity of yeast Saccharomyces cerevisiae. Mol. Microbiol. 2015, 95, 706–722. [CrossRef] 83. Kean, L.S.; Grant, A.M.; Angeletti, C.; Mahé, Y.; Kuchler, K.; Fuller, R.S.; Nichols, J.W. Plasma membrane translocation of fluorescent-labeled phosphatidylethanolamine is controlled by transcription regulators, PDR1 and PDR3. J. Cell Biol. 1997, 138, 255–270. [CrossRef] 84. Decottignies, A.; Grant, A.M.; Nichols, J.W.; de Wet, H.; McIntosh, D.B.; Goffeau, A. ATPase and Multidrug Transport Activities of the Overexpressed Yeast ABC Protein Yor1p. J. Biol. Chem. 1998, 273, 12612–12622. [CrossRef] 85. Ratto, T.V.; Longo, M.L. Obstructed diffusion in phase-separated supported lipid bilayers: A combined atomic force microscopy and fluorescence recovery after photobleaching approach. Biophys. J. 2002, 83, 3380–3392. [CrossRef] 86. Van‘t Klooster, J.S.; Cheng, T.Y.; Sikkema, H.R.; Jeucken, A.; Moody, B.; Poolman, B. Periprotein lipidomes of Saccharomyces cerevisiae provide a flexible environment for conformational changes of membrane proteins. Elife 2020, 9. [CrossRef] 24 of 24 24 of 24 Biomolecules 2020, 10, 871 87. Guan, X.L.; Souza, C.M.; Pichler, H.; Dewhurst, G.; Schaad, O.; Kajiwara, K.; Wakabayashi, H.; Ivanova, T.; Castillon, G.A.; Piccolis, M.; et al. Functional Interactions between Sphingolipids and Sterols in Biological Membranes Regulating Cell Physiology. Mol. Biol. Cell 2009, 20, 2083–2095. [CrossRef] [PubMed] 88. Swain, E.; Baudry, K.; Stukey, J.; McDonough, V.; Germann, M.; Nickels, J.T. Sterol-dependent regulation of sphingolipid metabolism in Saccharomyces cerevisiae. J. Biol. Chem. 2002, 277, 26177–26184. [CrossRef] [PubMed] 89. Solanko, L.M.; Sullivan, D.P.; Sere, Y.Y.; Szomek, M.; Lunding, A.; Solanko, K.A.; Pizovic, A.; Stanchev, L.D.; Pomorski, T.G.; Menon, A.K.; et al. Ergosterol is mainly located in the cytoplasmic leaflet of the yeast plasma membrane. Traffic 2018, 19, 198–214. [CrossRef] [PubMed] 90. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References Castro, B.M.; de Almeida, R.F.M.; Fedorov, A.; Prieto, M. The photophysics of a Rhodamine head labeled phospholipid in the identification and characterization of membrane lipid phases. Chem. Phys. Lipids 2012, 165, 311–319. [CrossRef] [PubMed] 91. Stewart, E.; Gow, N.A.; Bowen, D.V. Cytoplasmic alkalinization during germ tube formation in Candida albicans. J. Gen. Microbiol. 1988, 134, 1079–1087. [CrossRef] [PubMed] 92. Mahanty, S.K.; Gupta, P.; Banerjee, U.; Fotedar, R.; Prasad, R. Defective plasma membrane H(+)-ATPase or orthovanadate resistant mutants from Candida albicans, a pathogenic yeast. Biochem. Int. 1990, 22, 11–20. 93. Stock, S.D.; Hama, H.; DeWald, D.B.; Takemoto, J.Y. SEC14-dependent secretion in Saccharomyces cerevisiae. Nondependence on sphingolipid synthesis-coupled diacylglycerol production. J. Biol. Chem. 1999, 274, 12979–12983. [CrossRef] 92. Mahanty, S.K.; Gupta, P.; Banerjee, U.; Fotedar, R.; Prasad, R. Defective plasma membrane H(+)-ATPase or orthovanadate resistant mutants from Candida albicans, a pathogenic yeast. Biochem. Int. 1990, 22, 11–20. 93. Stock, S.D.; Hama, H.; DeWald, D.B.; Takemoto, J.Y. SEC14-dependent secretion in Saccharomyces cerevisiae. Nondependence on sphingolipid synthesis-coupled diacylglycerol production. J. Biol. Chem. 1999, 274, 12979–12983. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2103080724
https://dspace.lib.cranfield.ac.uk/bitstream/1826/9814/3/Principle_of_duality_in_prognostics-2014.pdf
English
null
Principle of Duality on Prognostics
International journal of advanced computer science and applications/International journal of advanced computer science & applications
2,014
cc-by
8,780
Principle of Duality on Prognostics Mohammad Samie School of Applied Sciences Cranfield University Bedford, UK Alireza Alghassi School of Applied Sciences Cranfield University Bedford, UK Suresh Perinpanayagam School of Applied Sciences Cranfield University Bedford, UK Suresh Perinpanayagam School of Applied Sciences Cranfield University Bedford, UK Amir M. S. Motlagh Faculty of Science and Technology University of Westminster London, UK Epaminondas Kapetanios Faculty of Science and Technology University of Westminster London, UK Epaminondas Kapetanios Faculty of Science and Technology University of Westminster London, UK Abstract—The accurate estimation of the remaining useful life (RUL) of various components and devices used in complex systems, e.g., airplanes remain to be addressed by scientists and engineers. Currently, there area wide range of innovative proposals put forward that intend on solving this problem. Integrated System Health Management (ISHM) has thus far seen some growth in this sector, as a result of the extensive progress shown in demonstrating feasible and viable techniques. The problems related to these techniques were that they often consumed time and were too expensive and resourceful to develop. In this paper we present a radically novel approach for building prognostic models that compensates and improves on the current prognostic models inconsistencies and problems. Broadly speaking, the new approach proposes a state of the art technique that utilizes the physics of a system rather than the physics of a component to develop its prognostic model. A positive aspect of this approach is that the prognostic model can be generalized such that a new system could be developed on the basis and principles of the prognostic model of another system. This paper will mainly explore single switch dc-to-dc converters which will be used as an experiment to exemplify the potential success that can be discovered from the development of a novel prognostic model that can efficiently estimate the remaining useful life of one system based on the prognostics of its dual system. system’s aircraft, regardless of whether the adverse event had been erupted by one of the subsystems. In order to diligently address this problem, it is important to develop technologies capable of integrating large heterogeneous distributed system [2] and asynchronous data streams from multiple subsystems; hence making it easier to detect a potential adverse event. (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 DOI: 10.14569/IJACSA.2014.050609 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 DOI: 10.14569/IJACSA.2014.050609 www.ijacsa.thesai.org Copyright Statement: This is an open access article licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, even commercially as long as the original work is properly cited. - See more at: http://thesai.org/Publications/ViewPaper?Volume=5&Issue=1&Code=IJACSA&SerialNo=13#sthash.c2KMuq1m.dpuf Keywords—Prognostic Model; Integrated System Health Management (ISHM); Degradation; Duality; Cuk Converter Keywords—Prognostic Model; Integrated System Health Management (ISHM); Degradation; Duality; Cuk Converter Principle of Duality on Prognostics The following technologies will later be used for diagnosing what caused the event, forecasting what consequences the event will have on the RUL of the system (i.e.,whether the entire system will be put at risk), and lastly to take appropriate precautions to mitigate the event [1]. Moving on, in order to accurately estimate the remaining useful life of devices solely depends on developing prognostic models. This will require additional care and attention to be invested towards preparing the degradation profiles and establishing the physics of failure for every component. Therefore, it’s necessary to gather and obtain the degradation profiles of every subsystem, including their individual components. This further result’s as a new degradation profile being formulated for whenever a component is upgraded. This degradation profile is calculated from either analysing the accumulated damage or the data driven. A drawback of calculating the degradation profile is that any changes made in the design of the system willconsume time and incur additional costs, since the prognostics model will need to be re-upgraded. It’s therefore safe to say that the proposals discussed above are very expensive and consume a lot of time to process while also being unreliable, noisy and inaccurate [3]. I. INTRODUCTION Integrated System Health Management (ISHM) [1] is a future advancement incondition based asset management that attempts to create automated prognostic and diagnostic systems to maintain and improve the integrity and readiness expected from legacy Health and Usage Monitoring Systems. ISHM is functioned (over a certain period of time) to detect, predict, diagnose and mitigate adverse events caused by degradation, fatigue and faults in components. For instance, the following problems may occur during an important function related to a We intend to overcome these problems by developing a System – Level Reasoning (SLR) to at least provide the system with significant capabilities that can potentially reduce costs by adding a System Integrated Prognostic Reasoner (SIPR) to the system prognostics [1][4]. For Instance, a Vehicle Integrated Prognostic Reasoner (VIPR) is a project funded by NASA for developing the next generation VLRS. A typical functional module within the SLR is a System Reference Model. The www.ijacsa.thesai.org Copyright Statement: This is an open access article licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, even commercially as long as the original work is properly cited. - See more at: http://thesai.org/Publications/ViewPaper?Volume=5&Issue=1&Code=IJACSA&SerialNo=13#sthash.c2KMuq1m.dpuf 43 | P a g e 43 | P a g e (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 System Reference Model divides the system into partitions. It also provides the necessary relationships between subsystems, which is required for the inference process. This partitioning allows the inference engine to reuse and link the same prognostic models to multiple subsystems and further minimize certification and qualification costs [1][4]. propagation models.Failure mechanisms are often used in conjunction with system lifecycle management to create prognostics and health management (PHM) disciplines. PHM is also sometimes known as system health management (SHM) or within the field of transportation applications; it is either known as vehicle health management (VHM) or engine health management (EHM). Building prognostic models constitutes of three main technical approaches which fall within the categories of data-driven approaches, model-based approaches, and hybrid approaches [1][4][5]. Although various methods and techniques including: neural network, fuzzy, statistics, semantic computing, graph theory, etc., have been thus far utilized for the development and enhancement of ISHM; however, ISHM continues to suffer from problems related to inefficient models, uncertainties and inadequate reasoning. A. Data-Driven Prognostics Data-driven prognostics [6] are mainly based on pattern recognition and machine learning approaches that help identify and detect trends and changes in the individual phases of a system’s state. A way to predict trends in nonlinear systems is by using classical data-driven methods,such as stochastic models, an autoregressive model, the bilinear model, the projection pursuit, etc. Soft computing techniques that involve using various types of neural networks (NNs) and neural fuzzy (NF) systems have also been commonly adopted to deal with data-driven forecasting of a system state [7][8]. This prognostic approach applies to applications that havecomplicated system architecture, i.e., systems that incur high amount of cost when developing an accurate prognostic model. So by adopting this approach to deal with complex systems will lead the prognostics of a system to be much faster and cheaper to set up as compared to other approaches. Contrarily, data driven approaches may have a wider confidence intervals than other approaches which mean it will require a substantial amount of data for training purposes [9]. There are various strategies used to develop data-driven prognostics which involve the analysis of either (1) modelling cumulative damage and then extrapolating out to a damage threshold, or (2) directly learning from the data based on the remaining useful life. The prognostics of the system shall be further explained in section 2. The principles of duality in electrical systems, along with brainstorming the duality concept of system’s prognostics, are covered in section 3. Section 4 covers the prognostics of dc-to-dc converters with details of Cuk and its dual circuit. The proposed algorithm to develop prognostics for dc-to-dc converters using duality concept is presented in section 5. Simulation results are discussed in section 6. Section 7 discusses future work. Lastly, section 8 concludes the major points discussed in the paper. As it is a lengthy and rather costly process to fail each and every system one by one, we thus seek to obtain the run-to- failure data which refers to the main fundamental setback, especially for new systems. In order to retrieve adequate data- driven prognostics, the accelerated aging data should be extracted cautiously from a number of similar/related products by using appropriate measuring tools. A. Data-Driven Prognostics This means that both the quality and quantity aspects of the data driven prognostics will add to expenses; especially since the data sources may have been derived from a wide range of factors including: temperature, pressure, oil debris, currents, voltages, power, vibration and acoustic signal, spectrometric data, as well as calibration and calorimetric data. As a result, it is important to fully understand what parameters and signals will be necessary to measure, and which features will need be extracted from the noisy and high-dimensional data [6][7][9]. II. PROGNOSTICS In condition-based maintenance, prognostics can be defined as a controlled engineering discipline that focuses on the prediction and estimation of the future course of a system or component that tries to establish when the system/component starts to slowly develop irregularities and faults to the point where it eventually malfunctions. A system or component that malfunctions means that it can no longer operate accordingly. The predicted lifecycle of a system or component is referred to as the Remaining Useful Life (RUL). RUL is used in decision making for contingency mitigation and maintenance. There are various scientific techniques used that help construct the prognostics of a system or component including: failure mode analyses, early detection of aging signs, and damage I. INTRODUCTION Additionally, prognostic models also remain to be very costly and time consuming to prepare. The reason these problems exist is mainly because of the systems prognostics heavily relying on the physics of failure models and degradation profiles which are often considered to be inaccurate, inconsistent or even very noisy. We therefore believe that the ISHM system will greatly benefit if the prognostic of a component and system were to be perceived as a feature of a system rather than being perceived as the physics of components. The advantage of this approach is that it will enable SLR to develop prognostics for a new subsystem based on a collection of features (encompassing various models/patterns) already known from the previous prognostics of subsystems; hence saving a lot of time and resources. In order to successfully fulfil this task, SLR may need to employ various techniques associated with Soft Computing (SC),such as fuzzy and neural network within its Inference Engine and System Reference Model units, so that the subsystems properties can be linked to one another. In regards to this project, we expect that a duality connection will be found between the prognostics of dual systems, assuming that the prognostics of dual systems are seen as its parameters and features rather than physics of components. B. Model-Based Prognostics Attempts made towards integrating a physical model of a system which is (either accomplished via micro or macro levels) into the estimated remaining useful life (RUL) is referred to as model-based prognostics [5]. The micro level (also known as material level) is often referred to as damage propagation model which is a physical model that is integrated 44 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 in a series of dynamic equations. The following dynamic equations define the very relationships between damage and degradation of a system or component. They further define how the system or component operates under environmental and operational conditions. Despite it being almost impossible to measure many critical damage properties, an alternative solution would be to use sensed system parameters. However, it is possible that the level of uncertainty and inaccuracy to be increased. Even though uncertainty and inaccuracy is added as a result of sensed system parameters, uncertainty management would be considered with the realistic assumptions and simplifications, which may potentially overcome the limitations caused by the sensed system parameter [4][5][10]. TABLE I. DUALITY PRINCIPLE IN ELECTRICAL SYSTEMS System Dual of System Voltage of nodes or across device Current of branch or mesh Current of branch or mesh Voltage of nodes or across device Resistor (R) Conductivity (1/R) Capacitor (C) Inductance (C) Inductance (L) Capacitor (L) Voltage Source (Vs) Current Source (Vs) Current Source (Is) Voltage Source (Is) Kirchhoff’s Current Law Kirchhoff’s Voltage Law Kirchhoff’s Voltage Law Kirchhoff’s Current Law Mesh/Loop Node In regards to duality concepts, a duality relationship between two electrical circuits is expected to be found, if the values of the parameters and topologies of both circuits are linked to one another based on details in Table 1. Looking at it from a mathematical perspective, dual circuits are known to have the same mathematical model, apart from their parameter differing. Even though the function of systems are different, their prognostics still can be assigned to each other on the basis of dual relationships found between the systems, along with having the same mathematical model with dual parameters shown in Table 1. This provides us with the required facilities to develop the prognostics of a system based on the prognostics of its dual system. C. Hybrid Approaches In reality, it is almost impossible to either have a purely data-driven or purely model-based approach. However, both these models do share parts of one another’s mechanisms. The intention of hybrid approaches is to show the strength of both ‘data-driven’ approaches and ‘model-based’ approaches into one prognostic strategy. Two well-known categories of Hybrid approaches are, 1) Pre-estimate fusion and 2) Post-estimate fusion. The first technique applied, hardly has any ‘ground truth’ data or ‘run-to-failure’ data available. The second technique is fitted for situations where uncertainty management is required. This means that the second technique helps to narrow down uncertainty intervals of data-driven or model- based approaches while also improving accuracy [11][12]. B. Model-Based Prognostics In contrast to physical expressions used in micro-levels, macro-level models alternatively use mathematical models at a system level that help define the relationship among system input, system state, and system measurable variables. This mathematical model is often a simplified representation of the system. Simplification may lead to making prototyping faster; but the trade-off to this is that although the coverage of the model is increased, the accuracy of a particular degradation model is consequently decreased. In addition, within a complex application, such as a gas turbine engine, there would be a lack of knowledge in attempting to develop the proper mathematics for all subsystems or components. Again, this leads to uncertainty and inaccuracy, similar to micro-level models; which means simplifications must be considered by performing uncertainty management procedures [1][4][10]. Graph theory [13] well established that the behaviour and the functionality of a system can be recognized by knowing the topology of a system without having to know the components and devicesused in the system (considering we already know the nodes voltages and currents of the branches in the circuits). It can thus be expected that graph theory provides us with the capability to construct the prognostic of a system based on its topology rather than concentrating on the integrated devices and components within the system. It is also to be expected, systems that share similar or dual topologies and mathematical models will also share similar prognostics regardless of the components integrated within the system. This makes it possible to investigate how prognostic models can be constructed from knowing the topology of system rather than having to know thephysics of failure of a system. This therefore makes the process of modelling the prognostics of a system much more feasible and realistic, as it saves a substantial amount of resources and time, since you wouldn’t have to go through the process of individually testing every system in order to identify its prognostics. IV. PROGNOSTICS OF DC-TO-DC CONVERTER In reality, this represents the duality principles shown in Table 1 which proves that the resistor is a dual of a conductive; or in regards to this example, it can be known since the short circuit is a dual of an open circuit. A basic building block for power convertor type systems is dc-to-dc converter. There are many dc-to-dc voltage and current converters that have various topologies. These topologies can be defined algebraically [17]-[20], graphically, [21][22] or in a matrix form [23]-[25]. It is of significant interest in unifying converter topological characteristics, relationships, and analysis [26]. In regards to health management, the aim would be to develop a basic structure, model or concept that shows where all the converters, including their prognostics may have been derived and mapped. This unified model can lead to many advantages in developing conditional based monitoring, and System- Level Reasoning (SLR). Fig. 1. a, b) Resistive circuit with duality relationship, c,d) Graphs for circuits 1-a and 1-b. The underlying concepts related to basic converters can be unified on the basis of what has been already presented in [27] with regard to duality principles and in relation with current and voltage-source converters. The authors in [27] used circuit transformations to unify the basic converters, ultimately showing that other converters are derivable transformation topologies of the basic converter. Fig. 1. a, b) Resistive circuit with duality relationship, c,d) Graphs for circuits 1-a and 1-b. The same rules can be applied to more complicated circuits where various components, including capacitors and inductances are also used. The most critical point that needs to be taken into account is the fact that degradation and failure mechanisms of dual components are not truly related to one another. For instance, the degradation mechanism of capacitor is not in any shape or form related to degradation mechanisms of inductance. This section shows how duality concept can be used to develop prognostic models for Cuk converter and its dual circuit. The following simulations were all conducted with ORCAD and MATLAB. Schematic of Cuk converter and its dual circuit are shown in Fig. 2-a and 2-b. Fig. 2. a) Cuk Converter, b) Dual circuit for Cuk converter in 2-a. A way to deal with this problem is to rely on the well- known physics principles, such as Ohm’s and Kirchhoff’s laws. IV. PROGNOSTICS OF DC-TO-DC CONVERTER In reference to these two laws, it can confidently be said that any electric component can be formulated by using voltage across the component and current through the component. Alternatively, in regards to graph theory’s basic principles of circuit and system design, it has been well known that the behaviour of a system is fully formulated if the voltage of all nodes and current through all the branches in the circuit are also known. This means that behaviour and the function of circuit can be fully formulated no matter what components are used in the circuit, as long as all the voltages and currents are known Fig. 1-c and 1-d, respectively show the graph of the equivalent circuits in Fig. 1-a and 1-b. Fig. 2. a) Cuk Converter, b) Dual circuit for Cuk converter in 2-a. We use certain values for Cuk converter devices, as well as all the equations depicted in reference [28] for all the simulations in this paper. Cuk is a step-down/step-up converter that shares a similar switching topology with buck-boost converter. It thereby presents the voltage ratio of a buck-boost converter [28]: Perceiving it from a circuit level, the details required for the development of a prognostics model for a circuit does not necessarily need to be known. Essentially, sensors are used to measure voltages, currents, temperature, etc. By basing it on the meaning of the sensed values, allows the experiences of a degraded circuit or system of any form, to be interpreted as a circuit not functioning properly. Although this principle can be applied for greater purposes, i.e., to design a device independent prognostic model, this paper will for now mainly concentrate on presenting a realization of duality principles for the development of prognostics for dual circuits. ⁄ ( ) ⁄    where vo is output voltage, vg is the input voltage, Ds is the duty cycle of the switch ton/(ton+toff); and ton and toff are durations for when the switch is on and off. Equation 3 is calculated from the principle of conservative energy and the fact that the inductor currents relate to the input and output currents. This equation shows that the output voltage can be controlled by maintaining the duty cycle of the switch. In addition, duality concept has already been recommended for diagnosing faults. III. PROGNOSTICS OF DUAL SYSTEMS Duality is one of the fundamental properties which can be consistently seen in physical systems, such as, electrical, mechanical systems, etc. [13][14]. It has an interesting history in mathematics, engineering and science. Duality relations have been identified between geometric objects, algebraic structures, topological constructs and various other scientific constructs. In regards to electrical systems, duality relations have appeared in the core principles for any theorem within an electrical circuit analysis, for situations where there is a dual theorem that replaces one of the quantities with dual quantities. Fig. 1 shows an example of dual circuits. Using Kirchhoff’s laws, it can be said that both circuits have the same mathematical model as shown in equation 1 for circuit in Fig. 1-a; and equation 2 for circuit in Fig. 1-b: ( ⁄ ⁄ ⁄ )   ( )     If for instance a degradation mechanisms is added, R2 in circuit of Fig. 1-b is aged towards a short circuit (R2→ 0), this is turned as (1/R2→∞) in circuit of Fig. 1-a. Examples of such dual quantities in electrical systems are current and voltage, impedance and admittance, meshes and nodes found in electrical systems (shown in Table 1) [15]. 45 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 IV. PROGNOSTICS OF DC-TO-DC CONVERTER Run to failure data for four different IGBTs. Fig. 3. Real model for a) Capacitor, b) Inductance. Run to failure data for four different IGBTs. Fig. 3. Real model for a) Capacitor, b) Inductance. Fig. 4. Real model for a) Capacitor, b) Inductance. Inside converters, the components that are mainly damaged refer to the IGBTs and capacitors. Alghassi et al has discussed different failures mechanisms related to IGBT and they have also presented prognosis model for the dominant failure at a component level in [29][30]. IGBT experiences a number of failure mechanisms including: bond wire fatigue, bond wire lift up, corrosion of the wires, static and dynamic latch up, loose gate control voltage, etc. The resulting affects mentioned are too complex, but we assume that these failure mechanisms can cause IGBT to behave as either an open circuit on a collector- emitter or a device encountering malfunction on its gate- emitter control. For instance, IGBTs thermal junction is increased due to solder crack which turns to wire bond lift off that increases the resistor relating to the collector-emitter. On the other hand, hot carrier injection is increased due to electrical stress. This causes short circuit on the IGBTs gate- emitter junction. The result of this failure, leads to IGBT’s gate controllability being missed (loose gate control voltage) causing IGBT to malfunction. The result of this effect is an increase in current through collector-emitter which means that the resistor of collector emitter is decreased. Therefore, it can be established that wire bond lift off and loose gate control voltage are failure mechanisms that presents some kind of duality relationship. While one of them increases the resistor, the other one decreases the resistor. Generally, we assume that IGBT’s failure and malfunction mechanisms are parameters that have duality relationships. Fig. 3. Real model for a) Capacitor, b) Inductance. Fig. 3. Real model for a) Capacitor, b) Inductance. Fig. 4. Real model for a) Capacitor, b) Inductance. IV. PROGNOSTICS OF DC-TO-DC CONVERTER Reference [16] proposes a fault diagnoser based on the duality principle and the optimal control theory for linear systems. However, this paper will present duality applications in system prognostics. 46 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 Basedon the type of switching scheme, output voltage can be either higher or lower than the input voltage. The state equations for Cuk converter are: By the time that the IGBTs are damaged, Cs and Ls are fully charged, as well as the other energy storage components lose energy, so Vo would be 0. It is, however, impossible to have a real short circuit in IGBT, thus we assume that it may have happened when the current through the collector-emitter exceeds over its limit just before the IGBT is burned out.    [ ]    Based on the level of accuracy, there are number of models that can be applied to a real capacitor and an inductance. To simplify a simulation, we assume that the capacitor and the inductance can both be modelled like Fig. 4 for the purposes of this paper. These models will present duality relationship between capacitance and inductance while also presenting the energy lost by the resistors. R1 typically has had very large values, while R2 has a small value; but due to degradation, these resistors are changed towards either open or short circuits. [ ]  [ ]  The Cuk converter has two inputs, a control input (Vc) and an input from the power supply (vs) and one output (vo). Therefore, matrix [A B C D] relates to ‘state space matrices’ for the open-loop model from the vs to the vo. Similarly, [A Bc C D] is the state space matrices from the control input d to the output vo. Values for A, B, Bc, C, and D are given in [17]. The same equation can be extracted for dual circuit of Cuk converter in Fig. 2-b; however, parameters are used in a dual form as shown in table 1. Switches in Fig. 2 are IGBT with a control voltage Vc. Yin and Zin are input admittance and input impedance of Cuk circuit and its dual circuit. Run to failure data for four different IGBTs. Fig. 3. Real model for a) Capacitor, b) Inductance. V. ALGORITHM TO DEVELOP PROGNOSTICS Fig. 5 illustrates the proposed algorithm devised to develop this prognostics model. The same process that possesses different sets of run to failure degradation and malfunction profiles is repeated for both Cuk and its dual circuit. At first the components of the circuits are set to be in a good condition. Then as soon as the time step for the circuit is increased, the values of the components are changed by using a series of values provided within the degradation profile for the new time step. Signals, such as v1, v2, vo, i1, i2, io, are measured at each time step phase. Fig. 3 shows IGBT run to failure data relating to four different IGBTs. This data is very noisy and needs to be filtered, but there are still a number of states that can be seen in the data. These states refer to cracks or wires that were lifted up due to degradation mechanisms. The resulting effects are changes in the IGBT’s functionality; and changes in the channel resistor of that IGBT. We assume that degradation is processed in a form of duality for Cuk and its dual circuit, so that if IGBT of Cuk experiences degradation towards its open circuit, IGBT of dual circuit of Cuk is degraded towards short circuit. The following signals are used for calculating the properties of the system, such as transfer functions, input and output impedances and admittances. Subsequently, the system degradation is turned according to changes it has encountered during the transfer functions (Zc(d,t), Yc(d,t), Zdc(d,t), Ydc(d,t)). 47 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 The resulted time dependent transfer function which is known as prognostic model is excited with step function (u(t)) during the RUL estimation. Step function u(t) provides a fixed input of  for the converter over the period of t>0. We later assign fuzzy values to the output of transfer function excited with u(t). The fuzzy values represents whether there are a small, medium, normal, transient and big changes experienced at the output of the converter. So where d is an index of a selected degradation profile, c is Cuk and dc is the dual circuit of the Cuk converter. V. ALGORITHM TO DEVELOP PROGNOSTICS Whenever d is altered, time step (t) is reset to zero which resets the process of the circuit to a healthy condition for the new degradation scheme. The reason for measuring the mentioned signals and parameters is that it would make it possible to understand how energy is transferred between capacitances and inductances; and how that transferred energy is lost when the system is also degraded. The term ‘normal’ in the fuzzy set, means that changes in the signals can be ignored and transient means that the circuit is in a transient mode and should be settled in a steady state during a specific time constant. RUL is estimated using the MAX fuzzy function which is applied on the triggered fuzzy values. MAX fuzzy function selects the maximum fuzzy value among the fired membership functions. Fig. 5. Algorithm used to develop prognostic model. d/m: degradation/mulfunction number; t: time step or cycle. During this process, RUL is estimated in a fuzzy form, and therefore needs to be de-fuzzified. During the de-fuzzification step, RUL is also scaled up, so that the integration of the estimated RUL (in fuzzy form) reflects the maximum life cycle of the circuit, Fig. 7. Confidence levels are implemented using fuzzy adjectives and adverbs found in fuzzy base knowledge and fuzzy rules. Fig. 6. Tee and Pi Models Fig. 7. Simplified Algorithm for RUL Estimation. VI. SILUMATION RESULTS AND DISCUSSION Fig. 6. Tee and Pi Models Fig. 7. Simplified Algorithm for RUL Estimation. Fig. 5. Algorithm used to develop prognostic model. d/m: degradation/mulfunction number; t: time step or cycle. To successfully estimate the remaining useful life of both converters, the same process is to be repeated for numbers of different degradation and malfunction profiles. The implication of such a process is to obtain standard Tee and Pi models for Cuk converter and its dual circuit, as shown in Fig. 6. This means that there would be a number of time dependent Tee and Pi models, one for each degradation and malfunction profile. There are many different techniques, such as neural network, fuzzy, statistics, etc., that can be utilised in order to generate a universal prognostic model for the converters (Cuk and its dual) out of all time dependent Tee/Pi models needed to be trained. Here, we just use a mean value to simplify and speed up the process. Fig. 7. Simplified Algorithm for RUL Estimation. Fig. 7. V. ALGORITHM TO DEVELOP PROGNOSTICS Simplified Algorithm for RUL Estimation. VI. SILUMATION RESULTS AND DISCUSSION Simulation results, using ORCAD 16.6 that reflect the circuits shown in Fig. 3 are presented in Fig. 8 and 9. Looking at these figures, it is apparent that Vo1 has the same trend as Io2; and the same for Iin1 and Vin2. 48 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 These in turn reflect the similarities encountered within the ransfer functions, such as Z, Y, Av and Ai shown in Fig. 10. As shown in Fig. 11, we used the IGBT model for our imulations. To add degradation to our simulations, we hanged the IGBT’s parameters, such as Ron in such a way that a trend of failure in Fig. 2 will be obtained from the IGBT model in Fig. 11. To speed up the simulation, we intend to have all 4500 ycles shown in Fig. 2 in just 25 ms. The result from this mapping is that the degradation will be accelerated in such a way that the first degradation will be experienced around 8 ms after exciting the circuit with step function, u(t); however, the hreshold needed to estimate whether the IGBT has aged enough to incur damage in an earlier time is around 10 ms. The ame life time and threshold can be expected from energy ransfer and power of Co and Lo, respectively shown in Fig. 12 and 13 for Cuk and its dual circuit. The figures illustrate that as a result of degradation, energy s not sufficiently transferred in the circuit. So this informs us hat the health state of a circuit can just as well be understood rom the state equations of a circuit, whereby energy signals rom storing elements, capacitors and inductors are used as tate variables. Fig. 8. Changes in Iin1 and Vin2 due to degradation in IGBTs. Fig. 9. Changes in Vo1 and Io2 due to degradation in IGBTs. Fig. 10. Changes in transfer functions due to degradation. International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 Fig. 8. Changes in Iin1 and Vin2 due to degradation in IGBTs. Fig. 9. Changes in Vo1 and Io2 due to degradation in IGBTs. (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. V. ALGORITHM TO DEVELOP PROGNOSTICS Fig. 11. IGBT Model for Simulation. Fig. 11. IGBT Model for Simulation. further care to be conducted, such as filtering in order to reduce uncertainties for accurate RUL estimation. For instance, instead of making direct decisions based on monitored signals, the monitored signals can be shifted in the FIFO (First-In First- Out stack) one by one, and the mean value of available data in the FIFO could be hence used for the RUL estimation. FIFO has a fixed storage length, so that shifting a new sample to the FIFO will release the sample that had been already shifted into the stack at the earliest time. Mean value of FIFO captures the trend of signals and eliminates noise, unwanted information and uncertainties. The following FIFO will increase system reasoning within RUL estimation. The length of FIFO has had a great impact on eliminating noise, but it normally shouldn’t take that long to lose trend of system degradation. Additionally, implementing a mean value on the stored data in the FIFO may add DC value (i.e., mean value) to the estimated RUL. As DC value is constant (i.e., meaningless information), it will make it easier to eliminate the DC value from RUL. Fig. 11. IGBT Model for Simulation. Fig. 11. IGBT Model for Simulation. Fig. 12. Energy in Co of Cuk converter in Fig. 3. In order to simplify the process, we use Av1 and Ai2 for the RUL estimation using fuzzy logic techniques to estimate the remaining life time of circuit, as shown from the algorithm in Fig. 7. All the input and output membership functions are set in Gaussian form with input fuzzy values as {small, medium, normal, transient and big} and output fuzzy values as {health_state_1, health_state_2, health_state_3, health_state_4 and health_state_5}. Fuzzy values at input refer to the changes in the trend of Av1 and Ai2. Fuzzy values at output refer to the life state of circuit, such as young for health_state_1 and, aged for health_state_5. We also use number of adverbs and adjective to address 10% and 90% confidence levels in life estimation. Fig. 14 shows the final RUL in fuzzy form. This figure shows how the circuit behaves in different life/health states during its life cycle. LF1 to LF5 show the life domain of each health state. LF1 comes from having huge transient period at the beginning of Av1 and Ai2. V. ALGORITHM TO DEVELOP PROGNOSTICS 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 These in turn reflect the similarities encountered within the transfer functions, such as Z, Y, Av and Ai shown in Fig. 10. As shown in Fig. 11, we used the IGBT model for our simulations. To add degradation to our simulations, we changed the IGBT’s parameters, such as Ron in such a way that a trend of failure in Fig. 2 will be obtained from the IGBT model in Fig. 11. Fig. 8. Changes in Iin1 and Vin2 due to degradation in IGBTs. To speed up the simulation, we intend to have all 4500 cycles shown in Fig. 2 in just 25 ms. The result from this mapping is that the degradation will be accelerated in such a way that the first degradation will be experienced around 8 ms after exciting the circuit with step function, u(t); however, the threshold needed to estimate whether the IGBT has aged enough to incur damage in an earlier time is around 10 ms. The same life time and threshold can be expected from energy transfer and power of Co and Lo, respectively shown in Fig. 12 and 13 for Cuk and its dual circuit. Fig. 8. Changes in Iin1 and Vin2 due to degradation in IGBTs. Fig. 8. Changes in Iin1 and Vin2 due to degradation in IGBTs. Fig. 9. Changes in Vo1 and Io2 due to degradation in IGBTs. The figures illustrate that as a result of degradation, energy is not sufficiently transferred in the circuit. So this informs us that the health state of a circuit can just as well be understood from the state equations of a circuit, whereby energy signals from storing elements, capacitors and inductors are used as state variables. Fig. 9. Changes in Vo1 and Io2 due to degradation in IGBTs. Fig. 9. Changes in Vo1 and Io2 due to degradation in IGBTs. g g g Fig. 10. Changes in transfer functions due to degradation. Fig. 10. Changes in transfer functions due to degradation. 49 | P a g e 49 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 Fig. 11. IGBT Model for Simulation. Fig. 12. Energy in Co of Cuk converter in Fig. 3. Fig. 13. Energy in Lo for dual circuit of Cuk in Fig. 3. VIII. CONCLUSION In conclusion, this paper shows that the prognostics of systems that share similar properties in the form of duality can be applied to one another. A prognostic model is developed in the form of a time dependant transfer function where based on the degradation mechanisms related to a system’s components, the values are subsequently altered over a certain period of time. So by having the prognostics assigned to a system’s property will thereby reflect the duality connection found within the degradation and malfunction profiles of a system. So if we were to consider that the components of a system are aged, this will mean that their dual components in the dual circuit will befaced with malfunction. Fig. 15. Resulting RUL after testing prognostic model with data test. We realized that if a degradation profile is used for Cuk, such that it’s converted to a malfunction profile for its dual circuit so that the IGBTs in both circuits always remain in dual forms; then a duality relationship would be found between the transferred functions of these two circuits. For instance, Zc(t) is equal to Ydc(t). This is because as the degradation profile changes the IGBT of Cuk towards an open circuit; its malfunction profile also changes the IGBT of dual circuit towards a short circuit. If the malfunction profile for dual circuit of Cuk is not extracted from the degradation profile of a Cuk circuit, then Zc(t) is not identical to Ydc(t). However, we come to a conclusion that if the whole process is repeated for number of different degradation and malfunction profiles and that the mean value of Zc(t) and Ydc(t) are used for comparison; leads to meaningful similarity patterns to be found between Zc(t) and Ydc(t). Zmc(t) can be used for the mean value of Zc(d,t) and Ymdc(t) can be used for the mean value of Ydc(d,t), in situations where m refers to the mean value. Zmc(t) and Ymdc(t) can be both used as prognostic models for Cuk and its dual circuit. However, these two transfer function are not exactly identical, but they would be more similar to one another if the process that is required to be executed to obtain the functions is repeated for various numbers of degradation and malfunction profiles for both circuits. V. ALGORITHM TO DEVELOP PROGNOSTICS In reality, we are not faced with such a big transient period, but it is included in our simulation just because of the acceleration in the degradation process. LF2 represents the how long the system works without the inclusion of degradation, and the rest refer to durations to which the circuit does experience degradation. The reason LF4 appears twice is that the circuit is not experiencing big changes in its transfer functions, but there are meaningful transient periods found that split the LF4 into two separated Gaussian fired fuzzy values. Fig. 12. Energy in Co of Cuk converter in Fig. 3. Fig. 13. Energy in Lo for dual circuit of Cuk in Fig. 3. Fig. 13. Energy in Lo for dual circuit of Cuk in Fig. 3. Fig. 13. Energy in Lo for dual circuit of Cuk in Fig. 3. The timings presented here are not realistic as we accelerated the aging mechanisms in the simulation. Nevertheless, it does clearly prove that the concept of the circuits with duality relationships or even with similarities in components and topologies may also have similar prognostics models. Fig. 14. Resulting RUL after testing prognostic model with data test. Every signal, energy and transfer function monitored in the figures have similar information with regard to the effects of degradation of a circuit. However, some of these parameters are rapidly changing due to the switching scheme of IGBT and energy transfer in Capacitors and Inductors, but in most cases the same trend can be found in all these signals. This experiment only refers to the degradation profile concerning the IGBT which refers to the component that mostly experiences degradation during real time; while simultaneously all other components are assumed to behave as non-aged devices (in all simulation). As shown in Fig. 10, Av1 and Ai2 seem to be the best for RUL calculation. Other transfer functions and signals are viewed as noisy data, thus requiring Fig. 14. Resulting RUL after testing prognostic model with data test. 50 | P a g e 50 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 The results reflecting the prognostic model is tested with an additional degradation profile. V. ALGORITHM TO DEVELOP PROGNOSTICS This will be handled as a test data which will assist us in estimating the remaining useful life time for the converter. Fig. 15 shows the de-fuzzified RUL that represents the remaining useful life with 10% and 90% confidence levels. Ideally, it is expected that the life of a circuit is decreased as a negative ramp in Fig. 15; however, our simulation shows that the RUL is slightly wavy. levels. In summary, by using the prognostic model of a system for other systems where similarities in their properties (like duality) are found, would give us a more accurate and reliable representation of the state and condition of the system. This is assuming that the prognostics are developed from adequate number of degradation profiles, and that they also have the right minimum and maximum confidence levels. VII. FUTURE WORK Fig. 15. Resulting RUL after testing prognostic model with data test. In this paper, we have looked at the IGBT in a converter as a critical component, thus meaning that the life expectancy of the converter is dependent on the remaining useful life of the IGBT. However, as for another component, such as a capacitor, it is also susceptible to thermal and mechanical stress. Thereby we must investigate whether it is classed as a dominant component failure in a converter or not. So in order to improve the novelty of duality in prognosis, requires one to have a cluster of components. This may overall have a remarkable impact on developing ISHM for critical applications. REFERENCES [1] I. K. Jennions, “Integrated Vehicle Health Management Perspectives on an emerging field”, SAE International, Warrendale, pennsylvania, USA 2011, pp. 100-110. [17] R. W. Erickson, “Synthesis of switched-mode converters”,IEEE Power Electron, PESC '83; Annual Power Electronics Specialists Conference, 14th, Albuquerque, NM, June 6-9, 1983, pp. 9–22. [2] A. El-Sayed and M. El-Helw, “Distributed Component-Based Framework for Unmanned”, Proceeding of the IEEE International Conference on Information and Automation Shenyang, China, June 2012, pp. 45-50. [18] D. Maksimovic, “Synthesis of PWMand quasi-resonant DC-to-DC power converters,” Ph.D. dissertation, California Inst. Technol., Pasadena, CA, USA, 1989. [19] Y. S. Lee, “A systematic and unified approach to modeling switches in switch-mode power supplies,” IEEE Trans. Ind. Electron., vol. IT-32, no. 4, Nov. 1985, pp. 445–450. [3] W. Wenbin and M. Carr, “A Stochastic Filtering Based Data Driven Approach for Residual Life prediction and Condition Based Maintenance Decision Making Support” Prognostics & Systems Health Management, IEEE Conference, Macao, China, Jan. 2010, pp. 1-10. [20] A. K. S. Bhat and F. D. Tan, “A unified approach to characterization of PWM and quasi-PWM switching converters: Topological constraints, classification, and synthesis,” IEEE Trans. Power Electron., vol. 6, no. 4, Oct. 1991, pp. 719–726. [4] I. K. Jennions, “Integrated Vehichle Health Management The Technology”, SAE International, Warrendale, pennsylvania, USA, 2013, pp. 139-154. [21] M. Ogata and T. Nishi, “Topological criteria for switched mode dc-dc converters,” in Proc. ISCAS, May 2003, vol. 3, pp. 184–187. [5] M. Daigle and K. Goebel, “Model-Based Prognostics under Limited Sensing”, IEEE Aerospace Conference, Big Sky Resort, USA, March 2012, pp. 1-12. [22] D. H. Wolaver, “Basic constraints from graph theory for DC-to-DC conversion networks,” IEEE Trans. Circuit Theory, vol. CT-19, no. 6, Nov. 1972, pp. 640–650. [6] C. Chen and M. Pecht, “Prognostics of Lithium-Ion Batteries Using Model Based and Data-Driven Methods”, 2012 Prognostics & System Health Management Conference, IEEE PHM Conference, Beijing, China, May 2012, pp. 1-6. [23] Y. Berkovich, A. Shenkman, A. Loinovici, and B. Axelrod, “Algebraic representation of DC-DC converters and symbolic method of their analysis,” in Proc. IEEE 24th Convent. Elect. Electron. Eng., Eilat, Israel, 2006, pp. 47–51. [7] H. Chao, D. Byeng, K. Youn, and K. Taejin, "Semi-Supervised Learning with Co-Training for Data-Driven Prognostics", Conference on Prognostics and Health Management, IEEE PHM Conference, 2012, pp. 1-10. [24] M. Ogata and T. Nishi, “Topological conditions for passive switches in switching converters,” in Proc. 18th IEEE ECCTD, 2007, pp. 898–901. [25] T. ACKNOWLEDGMENT [14] D. Y. Gao, "Duality Principles in Nonconvex systems: Theory, Methods and Applications", Kluwer Academic Publishers, USA, ISBN 9781441948250, 2000. The authors would like to sincerely thank Professor C Mark Johnson and Dr Paul Evans from the Power Electronics, Machines and Control Group, University of Nottingham for the contribution of failure data of the IGBTs and the power cycling test rig configuration. [15] G. E. Sharpe and, G. P. H. Styan, "Circuit Duality and the General Network Inverse", IEEE Transactions on Circuit Theory, Vol 12, IEEE 1965, pp. 22-27. [16] J. Li, "Optimal Fault Diagnosis Based on Duality Principle for Linear Systems", Control and Decision IEEE Conference, July 2008, pp. 573- 577. (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 Circuits", Mcgraw-Hill, 1st Ed. Edition, March 1987. Circuits", Mcgraw-Hill, 1st Ed. Edition, March 1987. VIII. CONCLUSION By implementing more intelligent algorithms that use stochastic, neural network, fuzzy and other techniques instead of a simple mean value function will increase the accuracy of this prognostic model. Implementing such intelligent algorithms also reflects the future aim and direction of our research. Additionally, we should be aware that prognostics have always been a way to estimate the life time of devices and systems within different confidence levels. Confidence levels provide assurance, so that we can comfortably rely on the performance of an aged system. The point is the accuracy of prognostic models has always been under doubt and remains to be under margins of confidence The accuracy of the developed prognostic model is highly dependent on the number of degradation profilesavailable; and the methodology used to train the time dependant transfer function. The minimum and maximum confidence levels are used to guarantee and express the accuracy of this model. However, this approach is presented just for Cuk converter and its dual circuit, but it seems that the same technique can be used for systems that have slightly similar mechanisms, properties topologies and degradation. Thereby, further research needs to be conducted for systems that are not in dual forms, especially for the purposes of exploring how the prognostic model of a system could be mapped to the prognostic model of another system. The advantage and usage of such a technique is emphasized in the implementation stage of the inference engine for System- Level Reasoning (SLR) and System Integrated Prognostic Reasoner (SIPR). It additionally provides us with the required facility to transfer degradation knowledge and experiences between systems. This means that the development of prognostics for huge systems, such as heterogeneous distributed systems used in applications like aircraft will be much faster, while decreasing the cost assigned to accelerated aging tests and preparing degradation profiles. We ultimately intend on pushing forward with our research, in order to apply this technique to the development of the prognostic inference engine and reasoned for aircraft. 51 | P a g e www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 6, 2014 REFERENCES Nishi, T. Oghishima, and M. Ogata, “Topological conditions on switched mode DC-DC converters,” in Proc. ITC-CSCC, Jul. 2002, pp. 1129–1132. [8] T. Sreenuch, A. Alghassi, S. Perinpanayagam, and Y. Xie; "Probabilistic Monte-Carlo Method for Modelling and Prediction of Electronics Component Life", (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 5, No. 1, 2014, pp. 96-104. [26] R. Severns, “Switch mode converter topologies make them work for you,” Intersil, Inc., Milpitas, CA, USA, Appl. Bull. A035, 1980. [9] S. Sarkar, X. Jin, and A. Ray, "Data-Driven Fault Detection in Aircraft Engines With Noisy Sensor Measurements", Journal of Engineering for Gas Turbines and Power, Vol. 133, ASME, August 2011, pp. 1-10. [27] B. W. Williams; "Generation and Analysis of Canonical Switching Cell DC-to-DC Converters". IEEE Transactions on Industrial Electronics, VOL. 61, NO. 1, January 2014, pp. 329-346. [10] L. Jianhui, M. Madhavi, K. Pattipati, Q. Liu, M. Kawamooto, and S. Chigusa, "Model-based Prognostic Techniques Applied to Suspention System", IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE 2008, Vol. 38, Issue 5, pp. 1156-1168. [28] F. J. Rytkonen and R. Tymerski, "Modern Control Regulator Design for DC-DC Converters", Electrical and Computer Engineering Department Portland State University. [online]. Available from http://web.cecs.pdx. edu/~tymerski/ece451/Cuk_Control.pdf,2014.05. 14. [11] P. Shetty, D. Mylaraswamy, and T. Ekambaram, "A hybrid prognostic model formulation system identification and health estimation of auxiliary power units”, Aerospace Conference, IEEE, Big Sky, MT, USA, March 2006, pp. 1-10. [29] A. Alghassi, S. Perinpanayagam, I. K. Jennions; "Prognostic capability evaluation of power electronic modules in transportation electrification and vehicle systems", IEEE, 15th IEEE Conference on Power Electronics and Applications (EPE), 2-6 Sept. 2013, 9 pages. [12] A. K. Garga, K. T. McClintic, R. L. Campbell, Y. Chih-Chung, M. S. Lebold, T. A. Hay, and et al., "Hybrid reasoning for prognostic learning in CBM systems", Digital Object Identifier: IEEE Proceedings on Aerospace Conference, Big Sky, Montana, USA, vol.6, March 2001, pp 2957-2969. [30] A. Alghassi, S. Perinpanayagam, I. K. Jennions; "A simple state-based prognostic model for predicting remaining useful life of IGBT power module", IEEE, 15th European Conference on Power Electronics and Applications (EPE), 2013, 7 pages. [13] L. O. Chua , C. A. Desoer, and E. S. Kuh, "Linear and Nonlinear 52 | P a g e 52 | P a g e www.ijacsa.thesai.org
W3126675401.txt
http://journals.uran.ua/tarp/article/download/4850/4501
en
Використання комп’ютерно-інтегрованих технологій проектування у ливарному виробництві
Technology audit and production reserves
2,012
cc-by
1,362
ISSN 2226-3780 Математическое моделирование — прикладные аспекты УДК 621.74 Е. Н. Сапегина Использование компьютерноинтегрированных технологий проектирования в литейном производстве В статье приведены результаты математического моделирования усадки и формирования усадочных дефектов в литых корпусных деталях при использовании различных материалов и конструктивных элементов литейной оснастки. Ключевые слова: математическая модель, литая деталь, компьютерно-интегрированная технология. 1. Введение Исследования, о которых идет речь в докладе, относятся к области математического моделирова­ ния в процессах проектирования. Этап проекти­ рования должен быть неразрывно связан с этапом технологии в контексте конструкторско-техноло­ гической подготовки производства. Например, при отработке технологических процессов изготовления сложных отливок из цветных сплавов под давле­ нием [1—8] или при газодинамическом воздействии в процессе кристаллизации отливки, необходимо на стадии проектирования учитывать много су­ щественных факторов. 2. Постановка проблемы В данном докладе рассматривается решение проблем по обеспечению качества литых деталей путем оптимального изменения параметров мате­ риала и конструктивных элементов оснастки на основе математического моделирования процессов заполнения литейной формы. 3. Основная часть 3.1. Анализ литературных источников по теме исследования. В работах [9—12] было показано, что эффективными инструментами для решения поставленной задачи является программный пакет трехмерного проектирования Solid Works и про­ грамма LVM Flow. Конечно-разностная модель ти­ повой по габаритам и конструкции литой детали может быть построена по таким исходным дан­ ным: размер ячейки; количество ячеек; материал отливки; температура металла; способ заливки. Методология построения математических моде­ лей, описывающих влияние перечисленных выше параметров проектирования (входных перемен­ ных) на качество заполнения формы расплавом, как описано в работах [13—15], включает в себя: нормирование входных переменных; построение планов полного или дробного факторного экспери­ ментов и проведение соответствующих компьютер­ ных экспериментов (в случае, если есть возмож­ ность самостоятельно менять входные переменные); или искусственной ортогонализации при наличии результатов пассивного эксперимента (взятых, на­ пример, из литературных источников). Реализация компьютерного эксперимента по заливке формы расплавом и процесса заполнения формы возможна также, как показано в работах [16—19], с помощью симплекс-метода и гребневого анализа по Хёрлю, причем в первом случае получение оптимального решения непосредственно связано с проведением эксперимента, а во втором — требует дополнитель­ ного анализа полученной поверхности отклика. 3.2. Результаты исследований. На основе рассмо­ трения и моделирования влияния описанных вход­ ных переменных на процесс образования дефек­ тов усадочного характера была выявлена важная роль состояния поверхности раздела «отливка — форма» с точки зрения качества заполнения формы. Полученные результаты математического и компью­ терного моделирования могут быть использованы на этапе проектирования литейной технологии, так как позволяют избежать лишних затрат произ­ водства на проведение натурных экспериментов — вместо этого можно использовать полученную ма­ тематическую модель. Література 1. Доценко Ю. В. Влияние комплексной технологии на свойства отливок из сплава АК7ч с повышенным со­ держанием железа [Текст] / Ю. В. Доценко, В. Ю. Се­ ливерстов // Восточно-Европейский журнал передовых технологий. — 2011. — № 6/5(54). — С. 45—48. 2. Доценко Ю. В. Особенности затвердевания отливок из алюминиевых сплавов при нарастающем давлении и модифицировании [Текст] / Ю. В. Доценко, В. Ю. Се­ ливерстов // Восточно-Европейский журнал передовых технологий. — 2012. — № 1/5(55). — С. 18—22. Technology audit and production reserves — № 5/2(7), 2012, © E. Sapegina 61 Математическое моделирование — прикладные аспекты 3. Доценко Ю. В. Затвердевание отливок из сплава АК5М при комплексном воздействии на расплав [Текст] / Ю. В. Доценко, В. Ю. Селиверстов // Вестник Нацио­ нального технического университета «ХПИ». Сборник научных трудов. Тематический выпуск «Новые решения в современных технологиях». — Харьков : НТУ «ХПИ», 2012. — № 1 — С. 3—8. 4. Доценко Ю. В. Особенности оценки эффективности получения отливок способом литья под. высоким дав­ лением [Текст] / Ю. В. Доценко, В. Ю. Селиверстов, В. В. Мацийчук, С. В. Малых // Вісник Національного технічного університету «ХПІ». Збірник наукових праць. Тематичний випуск: Нові рішення в сучасних техноло­ гіях. — Харків : НТУ «ХПІ». — 2012. — № 9. — С. 21—29. 5. Селиверстов В. Ю. Разработка рациональной конструк­ ции блока ЛВМ при газодинамическом воздействии в процессе кристаллизации отливки [Текст] / В. Ю. Се­ ливерстов, П. Д. Кущ, Ю. В. Доценко // Восточно-Ев­ ропейский журнал передовых технологий. — 2012. — № 2/13(56). — С. 9—12. 6. Доценко Ю. В. Анализ эффективности комплексных тех­ нологических решений по повышению качества литей­ ных алюминиевых сплавов с повышенным содержанием железа [Текст] / Ю. В. Доценко, В. Ю. Селиверстов, К. В. Шейдаев // XIII International scientific conference. New technologies and achievements in metallurgy and materials engineering. A collective monograph edited by Henryk Dyja, Anna Kawalek. Chapter 1. Series: Monographs No 24. Czestochova 2012. — P. 211—216. 7. Seliverstov V. About modifying action of gas-dynamic in­ fluence on the A356 aluminium alloy [Text] / V. Seliverstov, Y. Dotsenko // XIII International scientific conference. New technologies and achievements in metallurgy and materials engineering. A collective monograph edited by Henryk Dyja, Anna Kawalek. Chapter 1. Series: Mono­ graphs No 24. Czestochova 2012. — P. 265—271. 8. Селиверстов В. Ю. Технологические особенности реа­ лизации газодинамического воздействия на металл, за­ твердевающий в литейной форме [Текст] / В. Ю. Се­ ливерстов, Ю. В. Доценко, К. А. Думенко // Вестник Национального технического университета «ХПИ». Сборник научных трудов. Тематический выпуск «Но­ вые решения в современных технологиях». — Харьков : НТУ «ХПИ», 2012. — № 26 — С. 5—14. 9. Акимов О. В. Экспериментальные исследования и ком­ пьютерное моделирование материалов для блок-картера ДВС [Текст] / О. В. Акимов, А. П. Марченко // СхідноЄвропейський журнал передових технологій. — 2008. — № 5/1(35). 10. Акимов О. В. Компьютерное моделирование процес­ сов при производстве литых деталей двигателя [Текст] / О. В. Акимов, В. И. Алёхин, А. П. Марченко // Ли­ тейное производство. — 2010. — № 9. — С. 31—33. 11. Акимов О. В. Применение методик конструкторско-тех­ нологического проектирования деталей ДВС в модели­ ровании литейных процессов изготовления автомобиль­ ных поршней [Текст] / О. В. Акимов, В. И. Алёхин, А. П. Марченко // Цветные металлы. — 2010. — № 8. 12. Акимов О. В. Анализ влияния некоторых факторов на качество индукционной тепловой сборки подшипни­ ковых узлов [Текст] / О. В. Акимов, М. К. Кравцов, В. Т. Акимов // Східно-Європейський журнал передових технологій. — 2010. — № 5/5(47). 13. Дёмин Д. А. Обработка экспериментальных данных и построение математической модели технологического процесса методом наименьших квадратов (МНК) [Текст] / Д. А. Дёмин // Восточно-Европейский журнал пере­ довых технологий. — 2006. — № 3/1. — С. 47—50. 14. Серая О. В. Оценивание параметров уравнения регрес­ сии в условиях малой выборки [Текст] / О. В. Серая, Д. А. Дёмин // Восточно-Европейский журнал передовых технологий. — 2009. — № 6/4(42). — 2009. — С. 14—19. 62 ISSN 2226-3780 15. Серая О. В. Оценка представительности усеченных ортогональных подпланов плана полного факторно­ го эксперимента [Текст] / О. В. Серая, Д. А. Дёмин // Системні дослідження та інформаційні технології. — 2010. — № 3. — С. 84—88. 16. Дёмин Д. А. Оптимизация технологического процесса в цехе предприятия [Текст] / Д. А. Дёмин // ВосточноЕвропейский журнал передовых технологий. — 2005. — № 6. — С. 48—59. 17. Дёмин Д. А. Оптимизация технологических режи­ мов [Текст] / Д. А. Дёмин // Восточно-Европейский журнал передовых технологий. — 2006. — № 2/1(20). — С. 32—35. 18. Коваленко Б. П. Оптимизация состава холоднотвер­ деющих смесей (ХТС) с пропиленкарбонатом [Текст] / Б. П. Коваленко, Д. А. Дёмин, А. Б. Божко // Вос­ точно-Европейский журнал передовых технологий. — 2006. — № 6. — С. 59—61. 19. Дёмин Д. А. Принятие решений в процессе управле­ ния электроплавкой с учетом факторов нестабильно­ сти технологического процесса [Текст] / Д. А. Дёмин // Вісник національного технічного університету «ХПІ». — Харків : НТУ «ХПІ». — 2010. — № 17. — С. 67—72. Використання комп’ютерно-інтегрованих технологій проектування у ливарному виробництві О. М. Сапегіна У статті наведені результати математичного моделювання усадки і формування усадочних дефектів в литих корпусних деталях при використанні різних матеріалів і конструктивних елементів ливарного оснащення. Ключові слова: математична модель, лита деталь, комп’ю­ терно-інтегрована технологія. Олена Миколаївна Сапегіна, магістр кафедри ливарного виробництва Національного технічного університету «Хар­ ківський політехнічний інститут», тел.: (057) 707-68-54, e-mail: litvo11@kpi.kharkov.ua. Use of computer-integrated technologies of designing in foundry E. Sapegina The results of mathematical modeling of shrinkage and formation of shrink defects in the moulded body parts, using a variety of materials and structural components of foundry equipment, are given in the article. Keywords: mathematical model, moulded piece, computer-integ­ rated technology Elena Sapegina, master of Department of Technology and Equip­ ment, National Technical University «Kharkiv Polytechnic Insti­ tute», tel.: (057) 707-68-54, e-mail: litvo11@kpi.kharkov.ua. Технологический аудит и резервы производства — № 5/2(7), 2012
https://openalex.org/W1856099249
https://environmentalevidencejournal.biomedcentral.com/track/pdf/10.1186/s13750-015-0044-5
English
null
How effective are on-farm mitigation measures for delivering an improved water environment? A systematic map
Environmental evidence
2,015
cc-by
13,473
Abstract Background:  Agricultural activities are estimated to contribute 70% of nitrates, 28% of phosphates and 76% of sediments measured in UK rivers. Catchments dominated by agriculture also have elevated levels of pesticides and bacterial pathogens. European member states have a policy commitment to tackle this pollution through the water framework directive. Here we report on the results of a systematic map to investigate and describe the nature and coverage of research pertaining to the effectiveness of 6 on-farm mitigation measures, slurry storage, cover/catch crops, woodland creation; controlled trafficking, subsoiling and vegetated buffer strips for delivering an improved water environment in terms of a reduction in nitrogen (N), phosphorus (P), sediment, pesticides and faecal indicator organisms (FIOs) or pathogens from faecal material. Methods:  Research evidence for the effectiveness of the 6 on-farm mitigation measures for delivering an improved water environment (as detailed above) was collated using English language search terms for temperate farming systems in Europe, Canada, New Zealand and northern states of the United States of America. Searches for literature were made from online publication databases, search engines, specialist websites and bibliographies of topic specific reviews. Recognised experts, authors and practitioners were also contacted to identify unpublished literature. Articles were screened for relevance at title, abstract and full text using predefined inclusion criteria set out in an a priori published protocol. All relevant articles were mapped in a searchable database using pre-defined coding and critically appraised for relevance and reliability. Articles reporting the same study were removed. All full text studies without confounding factors were identified and coded for in a separate searchable database. Results:  A total of 718 articles were included in the database. Buffer strips were the most commonly studied inter- vention followed by cover crops and slurry storage. Little evidence was found for woodland creation and sub-soiling. No studies were found for controlled trafficking on grassland. Nitrogen was most frequently measured, followed by P, sediment, pesticides and FIOs or pathogens from faecal material. Conclusions:  The majority of the evidence collated in this map investigated the effectiveness of buffer strips and cover crops for improving water quality. This evidence was predominantly focussed on reducing N pollution. An evi- dence gap exists for the impact of cover/catch crops in reducing leaching of pesticides, FIOs and pathogens, and for organic forms of N and P. Randall et al. Environ Evid (2015) 4:18 DOI 10.1186/s13750-015-0044-5 Open Access Nicola P Randall*†, Louise M Donnison†, Paul J Lewis and Katy L James Nicola P Randall*†, Louise M Donnison†, Paul J Lewis and Katy L James © 2015 Randall et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. *Correspondence: nrandall@harper‑adams.ac.uk †Nicola P Randall and Louise M Donnison contributed equally to this work Centre for Evidence Based Agriculture, Harper Adams University, Newport, Shropshire TF10 8NB, UK Abstract There was limited research investigating the effectiveness of buffer strips for reducing leach- ing of organic forms of N or P, or for pesticides that are currently authorised for use/commonly used in UK agriculture. Further, long term studies across different seasons with controls, pre and post water quality measurements and mul- tiple sampling points from both field and rivers would improve the evidence base. Evidence gaps exist for woodland creation, subsoiling and controlled trafficking on grassland. *Correspondence: nrandall@harper‑adams.ac.uk †Nicola P Randall and Louise M Donnison contributed equally to this work Centre for Evidence Based Agriculture, Harper Adams University, Newport, Shropshire TF10 8NB, UK © 2015 Randall et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Page 2 of 15 Randall et al. Environ Evid (2015) 4:18 Keywords:  Buffer strip, Cover crop, Catch crop, Slurry, Woodland creation, Subsoiling, Controlled trafficking, Water quality, Nitrogen, Phosphorus, Pesticide, Sediment, Faecal indicator organism, Pathogen Keywords:  Buffer strip, Cover crop, Catch crop, Slurry, Woodland creation, Subsoiling, Controlled trafficking, Water quality, Nitrogen, Phosphorus, Pesticide, Sediment, Faecal indicator organism, Pathogen 4. Controlled trafficking (confinement of farm machin- ery to certain areas of a field) of grasslands to reduce soil erosion and compaction, and water runoff [16]. 4. Controlled trafficking (confinement of farm machin- ery to certain areas of a field) of grasslands to reduce soil erosion and compaction, and water runoff [16]. 4. Controlled trafficking (confinement of farm machin- ery to certain areas of a field) of grasslands to reduce soil erosion and compaction, and water runoff [16]. Objective of the systematic maph The primary question of this systematic map was:f h How effective are cover or catch crops, woodland creation, controlled trafficking on grassland, subsoiling, buffer strips and slurry storage as on-farm mitigation measures for delivering an improved water environment?h A decline in water quality has increased water cleaning costs, reduced reservoir capacities and can have nega- tive impacts on wildlife and flood defences [6]. Climate change scenarios suggest that the UK will experience wetter winters, and warmer, drier summers. Increased extreme weather events may increase the likelihood of heavy rains washing soil and pollutants into river sys- tems, and drier summers will concentrate levels of pollut- ants in rivers [7]. This question has the following components:i h Population Controlled waters as defined in section 104 of the Water Resorces Act and include territorial, coastal, inland and ground waters [20]. Interventions Slurry storage, cover/catch crops, wood- land creation, controlled trafficking, subsoiling, buffer strips. Comparator Absence of intervention or variation of intervention. European member states have a policy commitment to tackle water pollution through the water framework directive (WFD) and its integral components namely, the Nitrates, Ground Water and Bathing Water Direc- tives [7]. During the last 10 years the UK Department for Environment Food and Rural Affairs (Defra) and the Environment Agency (England and Wales) have funded catchment projects to improve water quality at a cost of around 70 million pounds [8]. This included studies to assess the effectiveness of mitigation meas- ures [9].h Outcome Impact on water quality in terms of change in any form of N or P, sediment, pesticides and FIOs or pathogens from faecal material. Background European agriculture has intensified over the last 50  years, leading to increased usage of fertilisers and agrochemicals [1]. Soil compaction and reductions in organic matter content resulting from intensive practice have increased the risk of soil erosion and water run-off. Nutrient applications in excess of plant needs, coupled with increased run-off from agricultural land, has con- tributed to a decline in water quality [2]. In the UK, for example, agricultural activities are estimated to be the source of 28% of phosphates, 70% of nitrates and 76% of sediments in rivers [3, 4], and catchments dominated by agricultural land use have elevated levels of bacterial pathogen counts [5]. f 5. Subsoiling (breaking up of compacted soil layers) to reduce compaction, and water runoff [17]. f 5. Subsoiling (breaking up of compacted soil layers) to reduce compaction, and water runoff [17]. f 6. Vegetated buffer strips to trap sediments, reduce pollutants and immobilise soluble nutrients through plant uptake or microbial degradation [18, 19]. f 6. Vegetated buffer strips to trap sediments, reduce pollutants and immobilise soluble nutrients through plant uptake or microbial degradation [18, 19]. Specialist sources The search terms for the database and web searches are listed in Table 1. Websites of relevant specialist organisations were identi- fied by the review team and stakeholders, and were also searched for relevant material. Websites were searched manually, by navigating through the site ‘Publications’ sections, if available, and also by using any provided auto- mated search with a number of key search terms. The first 50 hits from organisational websites were examined for appropriate data. Wildcards (*) were used where accepted by data- bases/search engines to pick up multiple word end- ings. Keywords were also made more restrictive by the addition of qualifiers. Keyword and qualifier combina- tions varied for each intervention. Where not already used as a qualifier, each search string was appended with ‘AND water’ if more than 900 search results were retrieved. A full list of organisational websites searched is given in Table 2. Topic specific bibliographies of meta-analyses and reviews were searched for relevant articles missed by the previous searches [15, 19, 24–26], as well as refer- ence lists e.g. the list of buffer strip studies maintained by Corell [27] (http://www.unl.edu/nac/riparianbibliogra- phy.htm). The exact search strings used differed between data- bases. Details of the terms used in each of the search facilities employed are provided in Additional file 1. Databasesh Recognised experts, practitioners and authors were contacted for further recommendations and the provi- sion of relevant unpublished material or missing data. The following online databases were searched to identify relevant literature for the primary question: ISI Web of Knowledge involving the following products: ISI Web of Science; ISI Proceedings, Science Direct, Wiley Online Library, Ingenta Connect, Index to Theses Online, CAB Abstracts, Agricola, Copac and Directory of Open Access Journals. Methodsh This question was commissioned by the UK Depart- ment of Environment, Food and Rural Affairs (Defra) to describe the nature and coverage of research pertaining to the effectiveness of 6 on-farm mitigation measures to deliver an improved water environment. Study design was discussed with the stakeholder group comprising; Defra, the UK Natural Environment Research Council (NERC), the Environment Agency (UK) and the Forestry Commission (UK). The methods used in the development of the systematic map followed the Collaboration for Environmental Evidence Systematic Review Guidelines [21] and from an existing systematic map report [22]. A scoping search was performed to validate the methodol- ogy, and is detailed in a review protocol [23], which was used to inform the final methodology. Only the system- atic map element from the published a priori protocol is presented here. This study reports on the results of a systematic map to investigate the effectiveness of 6 on-farm mitigation measures for delivering an improved water environ- ment in terms of a reduction in levels of any form of N or P, sediments, pesticides and faecal indicator organisms (FIOs) or pathogens from faecal material: 1. Slurry storage to reduce pathogens in slurry, and pol- lution incidents from spills and leaks [10]. 2. Fast-growing cover/catch crops, planted over win- ter to minimise soil erosion and reduce runoff and ensure that nutrients stay in the root zone [11–13]. 3. Woodland creation to improve soil structure and water infiltration and reduce runoff [14, 15]. Page 3 of 15 Page 3 of 15 Randall et al. Environ Evid (2015) 4:18 Search engines h A comprehensive search of multiple information sources attempted to capture an un-biased sample of literature to encompass both published and grey literature. Searches were conducted in 2012. Further internet searches were performed using the search engines: http://www.Scirus and http://scholar. google.com. The first 50 hits were examined for appropri- ate data. Table 2  List of specialist organisations websites searched Table 2  List of specialist organisations websites searched Defra Online Databases Environment Agency NERC Open Research Archive Forestry Commission/Forestry Research Centre for Ecology and Hydrology Natural England Countryside Council for Wales Scottish Natural Heritage Scottish Environment Agency Northern Ireland Environment Agency European Environment Agency European Commission Joint Research Centre Finnish Environment Agency Ministry of Agriculture and Forestry (Finland) Swedish Environment Agency Danish Environment Agency Ministry of Food, Agriculture and Fisheries (Denmark) Government Norway Portal Flemish Environment Agency Ministry of Food, Agriculture and Agri-food Canada Environment Canada US Department of Agriculture US Environment Protection Agency Agency of the Environment and Energy (France) Federal Environment Agency (Germany) Federal Ministry of Food, Agriculture and Consumer Protection (Germany) Netherlands Environmental Assessment Agency Department for the Environment, Transport, Energy and Communication (Switzerland) Environmental Protection Authority (New Zealand) Ministry of Agriculture and Fisheries (New Zealand) Food and Agriculture Organisation of the United Nations Ecologic Institute and EU Cost (European Cooperation in Science and Technology) p g Defra Online Databases Environment Agency NERC Open Research Archive Forestry Commission/Forestry Research Centre for Ecology and Hydrology Natural England Countryside Council for Wales Scottish Natural Heritage Scottish Environment Agency Northern Ireland Environment Agency European Environment Agency European Commission Joint Research Centre Finnish Environment Agency Ministry of Agriculture and Forestry (Finland) Swedish Environment Agency Danish Environment Agency Ministry of Food, Agriculture and Fisheries (Denmark) Government Norway Portal Flemish Environment Agency Ministry of Food, Agriculture and Agri-food Canada Environment Canada US Department of Agriculture US Environment Protection Agency Agency of the Environment and Energy (France) Federal Environment Agency (Germany) Federal Ministry of Food, Agriculture and Consumer Protection (Germany) Netherlands Environmental Assessment Agency Department for the Environment, Transport, Energy and Communication (Switzerland) Environmental Protection Authority (New Zealand) Ministry of Agriculture and Fisheries (New Zealand) Food and Agriculture Organisation of the United Nations Ecologic Institute and EU Cost (European Cooperation in Science and Technology) Inclusion criteria were applied by one reviewer, except where there was uncertainty, when two reviewers exam- ined the text and a consensus agreement was made. To assess and limit the effects of between-reviewer differ- ences in determining relevance, two reviewers (LD and NR) applied the inclusion criteria to a random sample of 50 articles at the title or abstract level. A kappa statistic was calculated using the online calculator http://www. graphpad.com/quickcalcs/kappa1.cfm. Duplicates and irrelevant articles were removed from Google search results using the procedure outlined for the main search results. Inclusion criteria All retrieved articles were assessed for relevance using the following inclusion criteria, which were developed in collaboration with funders, stakeholders and subject experts. • • Population: Articles that investigated the effective- ness of one of the on-farm mitigation measures to improve water quality of controlled waters [20] irre- spective of scale. Stakeholders agreed that the review should focus on temperate countries with similar farming systems to the United Kingdom (UK). Those countries were: UK, Ireland, France, Belgium, Swit- zerland, Germany, Holland, Luxembourg, Liechten- stein, Denmark, Sweden, Norway, Finland, Austria, Slovakia, Poland, Hungary, Czech Republic, Romania, Lithuania, Latvia, Estonia, Belarus, Ukraine, Canada and New Zealand and northern states of the United States of America (USA) as defined as all states that were entirely above the bottom of Oklahoma (so excluding states such as Georgia, Mississippi, Texas and California).f • • Population: Articles that investigated the effective- ness of one of the on-farm mitigation measures to improve water quality of controlled waters [20] irre- spective of scale. Stakeholders agreed that the review should focus on temperate countries with similar farming systems to the United Kingdom (UK). Those countries were: UK, Ireland, France, Belgium, Swit- zerland, Germany, Holland, Luxembourg, Liechten- stein, Denmark, Sweden, Norway, Finland, Austria, Slovakia, Poland, Hungary, Czech Republic, Romania, Lithuania, Latvia, Estonia, Belarus, Ukraine, Canada and New Zealand and northern states of the United States of America (USA) as defined as all states that were entirely above the bottom of Oklahoma (so excluding states such as Georgia, Mississippi, Texas and California).f single library. Google Scholar and organisational web search results were imported into spread sheets. Table 2  List of specialist organisations websites searched Search results from organisational web sites were checked by title for relevance. Those that passed the inclusion criteria were then examined at abstract/full text by following the web links. The remaining Google scholar and web site search results were combined with the main search results before the final stage of screening at full text, and any duplicates removed. Endnote databaseh The results of each search were imported into separate EndNote X2TM library files and then combined into a Table 1  Keywords and qualifiers used in the literature search Exact keyword and qualifier combinations varied in order to optimise searching efficiency and were informed by a scoping search. Mitigation Keyword AND Qualifier Slurry storage Slurr* stor* Animal waste lagoon* Animal waste stor* Slurr* lagoon* Slurr* tank* Dairy lagoon* Water qualit* Water pollut* Control of pollut* Nitrat* OR Nitrogen Phosph* Nutrient loss* Bacter* Fecal OR faecal Pesticid* Sediment* River* OR Stream* OR Catchment* Leak* OR Seap* OR Spill* Ground* water* Run off OR runoff Directive* OR Europe* Infiltrat* Leach* Water AND (Erosion OR Erod*) Eutrophication Water Woodland Afforest* (Wooded OR woodland*) AND (agricult* OR arable OR grass*) (Shelterbelt* OR windbreak* OR hedge*) Spray drift and tree* Buffer Riparian AND (buffer* OR zone* OR filter* Or strip* Filter strip* Vegetat* AND(buffer* OR barrier*) Controlled trafficking Wheel* AND compact* AND grass* Traffic* AND compact* AND grass* Soil compact* AND grass* Controlled traffic* AND grass* Subsoiling loosening compacted soil “Subsoiling” Loosen* Compact* Deep ripping Cover crop/catch crop “Cover crop” OR “Cover crops” OR “Covercrop” OR “Covercrops” “Catch crop” OR “Catch crops” OR “Catchcrop” OR “Catchcrops” Table 1  Keywords and qualifiers used in the literature search Page 4 of 15 Randall et al. Environ Evid (2015) 4:18 Table 2  List of specialist organisations websites searched Table 2  List of specialist organisations websites searched Table 2  List of specialist organisations websites searched Table 2  List of specialist organisations websites searched Defra Online Databases Environment Agency NERC Open Research Archive Forestry Commission/Forestry Research Centre for Ecology and Hydrology Natural England Countryside Council for Wales Scottish Natural Heritage Scottish Environment Agency Northern Ireland Environment Agency European Environment Agency European Commission Joint Research Centre Finnish Environment Agency Ministry of Agriculture and Forestry (Finland) Swedish Environment Agency Danish Environment Agency Ministry of Food, Agriculture and Fisheries (Denmark) Government Norway Portal Flemish Environment Agency Ministry of Food, Agriculture and Agri-food Canada Environment Canada US Department of Agriculture US Environment Protection Agency Agency of the Environment and Energy (France) Federal Environment Agency (Germany) Federal Ministry of Food, Agriculture and Consumer Protection (Germany) Netherlands Environmental Assessment Agency Department for the Environment, Transport, Energy and Communication (Switzerland) Environmental Protection Authority (New Zealand) Ministry of Agriculture and Fisheries (New Zealand) Food and Agriculture Organisation of the United Nations Ecologic Institute and EU Cost (European Cooperation in Science and Technology) insufficient information to exclude an article, at any stage, it was carried forward to the next stage. A record was made of the number of articles included and excluded at each stage of the screening processes. Study inclusion Study screening An iterative process was used to filter out non-relevant articles in a two stage process, firstly to screen out com- pletely irrelevant subjects (e.g. subjects such as mining and medicine) and secondly to filter out more closely related but irrelevant articles (e.g. air pollution and plant pathology). All articles were manually examined by at least title before being excluded. • • Intervention: Articles measuring the effectiveness of the following on-farm interventions in improving water quality were included: • Buffer strips: Studies measuring the impact on water quality of buffer strips composed of trees/ grass/shrubs, including shelterbelts and hedges. Articles retained were screened for relevance apply- ing inclusion criteria in three stages; title only, abstract, and then full text (where available). Where there was Page 5 of 15 Randall et al. Environ Evid (2015) 4:18 Studies of wetlands (unless wetland adjacent to buffer strip) or floodplains were excluded. Studies of wetlands (unless wetland adjacent to buffer strip) or floodplains were excluded. in water. Therefore in this review studies report- ing FIOs have been grouped together with studies investigating pathogens from faecal material. Studies were included that estimated water quality from soil samples, or that measured slurry leakage or changes in counts of FIOs or pathogens over time in slurry. Studies that measured inferred impacts (e.g. soil infiltration rates, crop yields, plant biomass, denitri- fication rates, mineralisation of soil N and pesticide drift) were excluded. fl •  Slurry storage: Studies measuring seepage of slurry from slurry storage, studies measuring changes in counts of FIOs or pathogens over time with slurry storage (excludes changes in N or P or air pollu- tion studies), and studies measuring the impact on water quality of the timing and amount of slurry applications. Studies of solid manure storage were excluded. •• Types of study: Only studies that reported primary research investigating the effect of an intervention on water quality were considered for inclusion in the review, which therefore excluded review articles and modelling studies. •• Types of study: Only studies that reported primary research investigating the effect of an intervention on water quality were considered for inclusion in the review, which therefore excluded review articles and modelling studies. •  Cover/catch crops: Studies of cover/catch crops or crops grown for winter cover and effects on water quality. Study inclusion Study screening Winter wheat or volunteer weeds were categorised as cover/catch crops if they provided ground cover in the same manner as a traditional cover/catch crop. •  Woodland creation: Studies measuring changes in water quality after afforestation of former agricul- tural land. Studies growing trees for biomass and testing their potential in cleaning waste water were also included, as were studies measuring the impact of crops intercropped with trees on water quality. Studies that compared water quality between dif- ferent land uses (forest, urban, arable, grassland) or measured changes in soil nutrient cycling after afforestation were excluded. Woodland buffer strip studies were also excluded as they were included in the intervention ‘buffer strips’ •  Woodland creation: Studies measuring changes in water quality after afforestation of former agricul- tural land. Studies growing trees for biomass and testing their potential in cleaning waste water were also included, as were studies measuring the impact of crops intercropped with trees on water quality. Studies that compared water quality between dif- ferent land uses (forest, urban, arable, grassland) or measured changes in soil nutrient cycling after afforestation were excluded. Woodland buffer strip studies were also excluded as they were included in the intervention ‘buffer strips’ Database and repository searches were conducted in the English language. No date restrictions were applied. Systematic map Coding Key wording was used to describe, categorise and code articles in the systematic map database. Keywords were generated from the primary question, the scoping study, topics reported in the included research, existing sys- tematic maps and expert knowledge. Articles were either coded on full text, abstract or title depending upon the availability of text. The definitions of the categories and codes used in the systematic map are detailed in Addi- tional file  2. For some categories multiple codes were applied, (for example, articles that reported results from more than one country or had multiple water quality measurements). Coding was moderated between review- ers (25% of articles were checked by both reviewers and any ambiguities discussed). f •  Subsoiling: Subsoiling studies that measured water quality. Studies that measured water quality after the break up/loosening of compacted soil layers. • Controlled trafficking on grassland: Studies that measured the effect on water quality of controlled traffic on grasslands. • • Types of comparator: Absence of intervention or var- iation of intervention. Studies with a no mitigation treatment (e.g. cropped or bare ground plots) were categorised as controlled studies, whereas studies using measurements over time and space were con- sidered to have within treatment comparative meas- ures and are referred to in this text as ‘with compara- tor’. Trends available in the evidence for all mitigations A schematic showing the numbers of articles that were included and excluded at each stage of the systematic mapping process is shown in Fig. 1. Preliminary trends were investigated using all the 718 articles. These trends remained consistent after articles were removed to meet the additional criteria. Studies often investigated multiple mitigations or outcomes or were performed in more than one country. This results in discrepancies in the total number of articles reported. Table 3  Scoring system used to  assess hierarchy of  evi- dence Adapted from: Pullin and Knight [30]. Category Score Hierarchy of evidence Randomized 1 Yes—randomized (includes partial) 0 Not randomized Control 3 Controlled BACI 2 Control 1 Comparator 0 None Study length 1 Study length greater than or equal to a year 0 Study length less than a year Replicates 2 Replicate temporal (includes time series) and spatial 1 Replicate temporal or spatial 0 No replicates Study type 3 Manipulative study 2 Correlative study 1 Monitoring study 0 Sampling study Table 3  Scoring system used to  assess hierarchy of  evi- dence Intervention typef Buffer strips (including woodland buffers) were the most commonly reported intervention (n  =  364), followed by cover/catch crop (n  =  245). Fewer articles reported slurry storage (n = 93), woodland creation (n = 24) and subsoiling (n = 10). No articles were found for controlled trafficking on grasslands. Hierarchy of evidence An overall indication of the relevance and reliability of evidence available for each intervention was calculated based on scoring of standard categories for included articles for each intervention. Every article coded in the systematic map at full text (including articles with confounding factors) was given a value according to a hierarchy of evidence adapted from systematic review guidelines used in public health [28] and conservation [29], and using a system adapted from a method out- lined by Pullin and Knight [30]. Articles were given val- ues for their design, based on categories applied in the systematic map database (Table  3). The values for each category were combined for each article, and these were then collated (mean and standard deviation) for each intervention. Of the 718 articles, 467 met our additional criteria that articles should be at full text and not report on a study already included in another article.(Where a study was reported more than once, the article with the most com- prehensive information was carried forward to the next stage). These articles were used to calculate the scientific rigour of the evidence for each intervention. The 467 articles were further assessed to remove any with con- founding factors resulting in 410 articles. The number of articles removed with confounding factors was similar for each intervention. Systematic map database Data in this table were used to calculate the hierarchy of evidence by filtering for articles coded at full text and with articles reporting the same study removed. 2. Water Quality Map Full Text: All studies with- out confounding factors coded at full text. Articles reporting same study were also removed. 2. Water Quality Map Full Text: All studies with- out confounding factors coded at full text. Articles reporting same study were also removed. Systematic map database A searchable database of coded articles for the primary question was created in Microsoft Access to describe the water quality research for the topic specific mitigation measures. Two database tables were created (Additional file 3): 1. Water Quality Map Title Abstract Full Text: All arti- cles coded, at title, abstract or full text. All evidence was coded with country of study, mitigation and water quality measurement, if the information was missing, ‘not clear’ was recorded. In addition full text articles were coded for study design, and those articles without confounding factors (i.e. where the effects of interventions or contribution of causal fac- tors can be separated and there are no other factors influencing the outcome of the study) were coded for • • Types of outcome: Water quality (irrespective of experimental scale) was measured by changes in the level of any form of N, P, sediment, pesticide, FIO (e.g. total coliforms, faecal coliforms, Escherichia coli, faecal streptococci, enterococci or clostridia spe- cies) or pathogen from faecal material (e.g. protozoan pathogens such as Cryptosporidium species and bac- terial pathogens such as Salmonella and Campylo- bacter species and Yersinia enterocolitica). FIOs are used to provide an indication of pathogen presence Randall et al. Environ Evid (2015) 4:18 Page 6 of 15 A total of 718 articles were judged to have met the inclusion criteria based on either title, abstract or full text review and were included in the systematic map (Database 1 of Additional file  3). A Kappa score of 0.588 indicated moderate agreement [31] for inclusion between reviewers and this was considered accept- able for this type of systematic map based on CEE guidelines [21]. The majority of articles were journal papers (n = 494), followed by conference papers/post- ers (n  =  118), reports (n  =  44) and theses (n  =  27). The remaining articles (n  =  35) were either books or the article type was unclear. Four meta-analyses were also found in the search process [17, 22, 23, 32]. The earliest article retrieved, dated from 1950 after that date no articles were published until 1971. A substan- tial increase in articles was seen from the early 1990s onwards. outcome. Data in this table were used to calculate the hierarchy of evidence by filtering for articles coded at full text and with articles reporting the same study removed. outcome. Geographical locationh Likewise most articles that reported measurements of P were from buffer strips (n  =  136), with less research found for slurry storage (n = 42) and cover/catch crops (n  =  24). Pesticides were most often Geographical locationh The majority of articles originated from the northern states of the USA (n  =  256) and these predominantly investigated buffer strips. The remainder of the articles originated from Europe, and most were from the UK (n = 80) where cover/catch crops were reported margin- ally more frequently than buffer strips. Randall et al. Environ Evid (2015) 4:18 Page 7 of 15 Records identified through database searching (excluding Google scholar and web searches) (n=146,941) Identification Screening Eligibility Identification Scoring Records identified with Google scholar no duplicates (n=5,430) Duplicates removed (n=72,655) Records excluded (n=54,101) Records excluded (n=12,634) Records excluded (n=6,100) Records excluded: Full text (n=467); Abstract (n=202) Records removed (n=223) Records excluded (n=28) Records excluded (n=57) Records added after screening: Google scholar (n=90); organisational websites (n=18) Records excluded (n=9,490) Records after duplicates removed (n=74,086) Records after keyword exclusion of irrelevant topics (n=19,985) Records after keyword exclusion of irrelevant biological topics (n=7,351) Records after screening on title/abstract and manual duplicate removal (n=1,359) Records after screening on full text/abstract: 718 articles mapped; 495 full text; 147 abstract 76 title Records identified through systematic mapping (n=718) Full text articles after removal of text read at title, abstract (n=495) Records after removal of duplicate article reporting same study Hierarchy of evidence (n=467) Records after removal of articles with confounding factors (n=410) Bibliographic records added (n=28) Fig. 1  Schematic of stages for the systematic map. Records removed (n=223) Records excluded (n=28) Records excluded (n=57) Fig. 1  Schematic of stages for the systematic map. N were recorded in buffer strip (n  =  209), cover/catch crop (n = 203) and slurry storage (n = 58) articles. Meas- urements of sediment were mainly reported in buffer strip articles (n = 128), although a few articles discussed cover/catch crops measuring sediments derived from soil erosion (n  =  28). Likewise most articles that reported measurements of P were from buffer strips (n  =  136), with less research found for slurry storage (n = 42) and cover/catch crops (n  =  24). Pesticides were most often N were recorded in buffer strip (n  =  209), cover/catch crop (n = 203) and slurry storage (n = 58) articles. Meas- urements of sediment were mainly reported in buffer strip articles (n = 128), although a few articles discussed cover/catch crops measuring sediments derived from soil erosion (n  =  28). Buffer strip study design Over half of the studies were manipulative (n  =  148) with the remainder mainly correlative (n = 74). Approxi- mately, half of the manipulative studies had temporal replication (n = 73) and most of the rest (n = 55) had times series data. The majority of the correlative stud- ies were conducted for at least a year or longer (n = 52), whereas, over a third of manipulative studies were con- ducted for less than a year (n = 83). Very few buffer strip studies were conducted for longer than 10 years (n = 5). One hundred and four studies had a control, most fre- quently bare ground (n = 38) or of cropped/alternative vegetation (n  =  27). Only 1 study used a Before-After Control-Impact experiment. Studies without a con- trol usually had a comparator (n = 121), often compar- ing water quality at different points along the width of a buffer, starting from the inflow of water and ending at the outflow (n = 82). Some controlled studies reported results in relation to inflow measurements rather than controls. Three studies measured changes in water qual- ity over time. Over half of the studies were manipulative (n  =  148) with the remainder mainly correlative (n = 74). Approxi- mately, half of the manipulative studies had temporal replication (n = 73) and most of the rest (n = 55) had times series data. The majority of the correlative stud- ies were conducted for at least a year or longer (n = 52), whereas, over a third of manipulative studies were con- ducted for less than a year (n = 83). Very few buffer strip studies were conducted for longer than 10 years (n = 5). Studies most often measured P in surface waters (n  =  50). Little research was found for P measured in subsurface (n = 5) or groundwaters (n = 13). This possi- bly reflects the properties of phosphorus, which has a low mobility in soil. Thirteen studies measured P in multiple flow paths. A total of 97 studies assessed the effective- ness of buffer strips for reducing sediment in water. The remaining studies measured sediment as ‘sediment loss’ and ‘turbidity’. One hundred and four studies had a control, most fre- quently bare ground (n = 38) or of cropped/alternative vegetation (n  =  27). Only 1 study used a Before-After Control-Impact experiment. Outcomes measured One hundred and thirty-nine studies assessed the effec- tiveness of buffer strips for reducing N. Nitrate (n = 120) and ammonium (n = 46) were the most commonly meas- ured forms of N. Nitrate was most often recorded in groundwater whilst ammonium was most often recorded in surface water (Fig. 2). Very few studies investigating organic forms (n  =  10). Measurements of groundwa- ter/subsurface flow were more common for woodland buffer studies (n  =  39/52), than grass buffer studies (n = 14/102). Nitrate was commonly measured in both woodland and grass buffer strips but Total N and ammo- nium were more often measured in grass buffers (Fig. 3). Ninety-four studies assessed the effectiveness of buffer strips for reducing P. Total P and orthophosphate were the most commonly recorded measurements. Study outcome Overall, 36 pesticides and 7 types of FIO and 4 pathogens were included in the map. Eight different forms of N, and 10 forms of P were recorded. Four forms of sediment were recorded (see Additional file 3). i The dominant water quality outcome measured was N (n  =  473), followed by P (n  =  178) and sediment (n = 165). Less research was found for pesticides (n = 71) and FIOs or pathogens (n = 61). Most measurements of Page 8 of 15 Randall et al. Environ Evid (2015) 4:18 or larger scale studies at catchment, regional, country or international level (n = 38). Only twenty-three stud- ies sampled river water, which may reflect the difficulty in eliminating the impact of confounding factors in river/ catchment studies. Seventy-four studies had data for all 4 seasons, and more studies were conducted in sum- mer (n = 62) than autumn (n = 21) or winter (n = 13). Recording of soil type was often varied or absent (n = 72) but most commonly fell within the loam range (n = 129). Buffer strips were composed of either grass (n = 154), trees (n  =  55) or a mixture of trees, grasses or shrubs (n = 69). More studies recorded buffer strips composed of deciduous trees (n = 80) than conifer species (n = 9). Vegetation manipulation was a common experimental factor e.g. species, density, cutting (n = 98). Other fac- tors included type or amount of fertilizer applied to plots, buffer width and soil type. recorded in buffer strip articles (n  =  63). Evidence for FIOs or pathogens was limited and came from arti- cles reporting slurry storage (n  =  34) and buffer strips (n = 32). or larger scale studies at catchment, regional, country or international level (n = 38). Only twenty-three stud- ies sampled river water, which may reflect the difficulty in eliminating the impact of confounding factors in river/ catchment studies. Seventy-four studies had data for all 4 seasons, and more studies were conducted in sum- mer (n = 62) than autumn (n = 21) or winter (n = 13). Recording of soil type was often varied or absent (n = 72) but most commonly fell within the loam range (n = 129). Buffer strips (including tree buffers) Description of studiesif Two hundred and twenty-five buffer strip studies were included in the database after studies with confound- ing factors and articles reporting the same study were removed. Almost two-thirds of the studies were con- ducted in the USA (n = 139), the rest were mostly from Europe (n = 62). Study quality h Cover/catch crop studies had the highest scientific rig- our values [mean value of 6.8, standard deviation (SD) 3.1] compared to buffer strip (mean 5.9, SD 2.4) and slurry storage studies (mean 4.1, SD 3.1). These values reflected that the majority of cover/catch crop and buffer strip studies were manipulative, often controlled and fully replicated, and were of longer duration compared to slurry storage studies. Furthermore, fewer buffer strip and slurry storage studies were randomised compared to cover/catch crop studies. Buffer strips were composed of either grass (n = 154), trees (n  =  55) or a mixture of trees, grasses or shrubs (n = 69). More studies recorded buffer strips composed of deciduous trees (n = 80) than conifer species (n = 9). Vegetation manipulation was a common experimental factor e.g. species, density, cutting (n = 98). Other fac- tors included type or amount of fertilizer applied to plots, buffer width and soil type. Study quality was not assessed for woodland creation and subsoiling studies due to the low number of studies found. Buffer strip study design Studies without a con- trol usually had a comparator (n = 121), often compar- ing water quality at different points along the width of a buffer, starting from the inflow of water and ending at the outflow (n = 82). Some controlled studies reported results in relation to inflow measurements rather than controls. Three studies measured changes in water qual- ity over time. Nineteen studies assessed the effectiveness of buffer strips for reducing FIOs or pathogens and the majority of these investigated surface flow (n = 17). Total faecal coliforms and E. coli were the commonest forms of FIO measured (Fig. 4).hff Thirty-eight studies assessed the effectiveness of buffer strips for reducing pesticides in water, 15 measured pes- ticides in surface flow, 9 in subsurface flow and 8 in both flow paths. Most of the studies were conducted at single sites/ farms (n = 129) with fewer multi-site studies (n = 41), Randall et al. Environ Evid (2015) 4:18 Page 9 of 15 Thirty-five different pesticides were investigated, with the herbicide atrazine (no longer approved for use in the European Union) the most commonly studied (n = 26). Cover/catch crops D i i f di Cover/catch crop study design Studies were mainly manipulative (n = 125) with a few correlative studies (n  =  6) and one monitoring study. Most studies were conducted for at least two winter seasons (n = 102) with 8 lasting for more than 10 years. Grass (n = 61) and cereals (n = 51) were the most com- monly studied cover/catch crop and were either grown 28 23 24 57 1 12 14 3 2 0 10 20 30 40 50 60 Nitrate Total N Ammonium Number of studies Surface water Groundwater Subsurface water Fig. 2  Number of buffer strip studies measuring different forms of nitrogen (N) in surface, subsurface and ground waters. Studies with multiple flow paths or studies where flow paths were not clear were excluded. When flow path was not stated and measurements were taken below ground a default coding of subsurface was used, therefore the distinction between groundwater and subsurface may not be valid. 42 5 13 39 23 21 0 5 10 15 20 25 30 35 40 45 Nitrate Total N Ammonium Number of studies Tree buffers Grass buffers Fig. 3  Number of tree and grass buffer strip studies for each form of nitrogen (N) measured. Buffer strip study design Studies with multiple buffer types were excluded. Grass-shrub buffers were excluded. Tree buffer strips included tree; tree-grass; tree, grass, shrub; grass tree buffers. Studies where tree buffers were compared to grass buffers were excluded. Fig. 4  Number of buffer strip studies for each form of faecal indicator organism or pathogen from faecal material measured. Fig. 4  Number of buffer strip studies for each form of faecal indicator organism or pathogen from faecal material measured. Cover/catch crop study design Thirty-five different pesticides were investigated, with the herbicide atrazine (no longer approved for use in the European Union) the most commonly studied (n = 26). Studies were mainly manipulative (n = 125) with a few correlative studies (n  =  6) and one monitoring study. Most studies were conducted for at least two winter seasons (n = 102) with 8 lasting for more than 10 years. Grass (n = 61) and cereals (n = 51) were the most com- monly studied cover/catch crop and were either grown alone, intercropped with a winter crop, or drilled into the stubble left from the previous crop. Fallow, bare ground or cropped plots were commonly used controls. A few studies did not have a control, either measuring changes in water quality over time or between different cover/ catch crop types. Volunteer weeds and winter wheat were Buffer strip study design Studies with multiple buffer types were excluded. Grass-shrub buffers were excluded. Tree buffer strips included tree; tree-grass; tree, grass, shrub; grass tree buffers. Studies where tree buffers were compared to grass buffers were excluded. 11 3 3 8 1 0 2 4 6 8 10 12 Number of studies Fig. 4  Number of buffer strip studies for each form of faecal indicator organism or pathogen from faecal material measured. 28 23 24 57 1 12 14 3 2 0 10 20 30 40 50 60 Nitrate Total N Ammonium Number of studies Surface water Groundwater Subsurface water Fig. 2  Number of buffer strip studies measuring different forms of nitrogen (N) in surface, subsurface and ground waters. flow paths or studies where flow paths were not clear were excluded. When flow path was not stated and measurements w a default coding of subsurface was used, therefore the distinction between groundwater and subsurface may not be valid 42 5 13 39 23 21 0 5 10 15 20 25 30 35 40 45 Nitrate Total N Ammonium Number of studies Tree buffers Grass buffers Fig. 3  Number of tree and grass buffer strip studies for each form of nitrogen (N) measured. Studies with multiple buffer types were excluded. Grass-shrub buffers were excluded. Tree buffer strips included tree; tree-grass; tree, grass, shrub; grass tree buffers. Studies where tree buffers were compared to grass buffers were excluded. 11 3 3 8 1 0 2 4 6 8 10 12 Number of studies Fig. 4  Number of buffer strip studies for each form of faecal indicator organism or pathogen from faecal material measured. 42 5 13 39 23 21 0 5 10 15 20 25 30 35 40 45 Nitrate Total N Ammonium Number of studies Tree buffers Grass buffers Fig. 3  Number of tree and grass buffer strip studies for each form of nitrogen (N) measured. Studies with multiple buffer types were excluded. Grass-shrub buffers were excluded. Tree buffer strips included tree; tree-grass; tree, grass, shrub; grass tree buffers. Studies where tree buffers were compared to grass buffers were excluded. 11 3 3 8 1 0 2 4 6 8 10 12 Number of studies Fig. 4  Number of buffer strip studies for each form of faecal indicator organism or pathogen from faecal material measured. Fig. 3  Number of tree and grass buffer strip studies for each form of nitrogen (N) measured. Study design Th d Commonly used experimental factors were crop type (n = 62), date and amount of fertilizer application (n = 45), the date and technique for removing the cover/ catch crop (n = 6), type of tillage (n = 27) and soil type (n = 18). Study design The 42 studies could be divided into 3 categories: The 42 studies could be divided into 3 categories: • • Studies that measured leakage from under or nearby to slurry storage (n = 23). These types of studies were mainly correlative using measurements over time and distance as comparators. Slurry was normally sourced from swine or dairy farms. Most of the slurry stores studied were earth lined and below ground. Only 4 of the articles were less than 10 year old and only 6 of the studies were conducted in Europe (the other studies were from the USA and Canada). Slurry storage legislation and drinking water standards may therefore not be comparable across studies. • • Studies that measured leakage from under or nearby to slurry storage (n = 23). These types of studies were mainly correlative using measurements over time and distance as comparators. Slurry was normally sourced from swine or dairy farms. Most of the slurry stores studied were earth lined and below ground. Only 4 of the articles were less than 10 year old and only 6 of the studies were conducted in Europe (the other studies were from the USA and Canada). Slurry storage legislation and drinking water standards may therefore not be comparable across studies. Outcomes measured No studies were found measuring FIOs or pathogens and only 3 studies measured pesticides Isofutole (n = 1); glyphosate (n = 1); metolachlor and atrazine (n = 1). One hundred and fourteen studies assessed the effec- tiveness of cover/catch crops for reducing N, mainly from subsurface (n = 77) measurements. Grass (n = 55), cereal (n = 44), brassica (n = 30) and legumes (n = 28) were the most commonly studied cover/catch crops (Table 4). Nitrate was the most commonly measured form of N. No studies were found measuring organic forms of N. • • Studies measuring survival rates of FIOs or patho- gens in slurry, the comparator being time (n = 11).f • • Field-based studies that measured the effect on water quality of timing and amounts of slurry application in winter following storage (n = 8). Fourteen studies measured P most commonly in sur- face or subsurface flow on a range of soil types. Grass (n  =  9) was the dominant cover/cover crop studied (Table 4). In some sediment studies, a crop cover of win- ter wheat rather than a traditional cover/catch crop was used. Most of the 19 cover/catch crops studies measuring sediment, studied grass (n = 8), winter wheat (n = 5) or other cereal (n = 6) on a loam soil type. The focus of a majority of the studies was erosion. Description of studies One hundred and thirty-two cover/catch crop studies were included in the database after studies with con- founding factors and articles reporting the same study were removed. Over two-thirds of studies were con- ducted in European countries (n = 87; UK n = 18) and the majority of the remainder in the USA (n = 33). Randall et al. Environ Evid (2015) 4:18 Page 10 of 15 sometimes used as controls, but in other cases used as crop covers. outside Europe where slurry storage construction legisla- tion may be different. For example, of the 23 studies that measured leakage of N, 17 were from the USA or Canada. Many studies (n = 20) also used earth lined stores which may not meet current legislation. The effectiveness of cover/crops in improving water quality was mainly measured from within field plots (n = 111). Only one study sampled river water [33]. Loam was the most commonly studied soil type (n = 71). Outcomes measuredh Thirty studies measured N. Whilst N was often detected under or near slurry storage many of the studies were not of the highest scientific rigour often lacking baseline pre and post slurry storage water quality data. Most stud- ies were conducted for less than 2  years, therefore the effect over time may not have been accurately assessed. Soil type has also been given as a reason for differences in slurry storage leakage. Description of studies Forty-two studies were included after studies with con- founding factors and articles reporting the same study twice were removed. Over half the evidence was from Only a small amount of evidence was found for P (n = 10) spread across the different study types. f Eighteen studies measured FIOs or pathogens with eleven studies measuring survival rates of FIOs or patho- gens in slurry. Table 4  Type and  number of  different cover/catch crops studied for reducing N, P and sediment Numbers are from full text studies without confounding factors. Cover/catch crop type N P Sediment Grass 55 9 8 Cereal 44 2 6 Crucifer 30 1 1 Legume 28 3 2 Other 3 0 1 Volunteer weeds 7 2 2 Winter wheat 12 2 5 Not clear 5 3 3 Table 4  Type and  number of  different cover/catch crops studied for reducing N, P and sediment Although timing of slurry application was not searched for, eight studies were found. However, experimental design and outcomes measured were variable. Woodland creation (excluding woodland buffers) Description of studies Twelve woodland creation studies were found all from Europe with the exception of one study from Canada. Outcomes measured Woodland creation studies most frequently measured N (n = 11), whereas P, sediment and FIOs were only meas- ured once. The variety of controls/comparators employed in woodland creation studies made it difficult to code outcomes. Some afforestation studies did not have a non-woodland control, but instead measured changes in water quality over different aged woodlands making it difficult to be certain if woodland had improved water quality compared to agricultural land [35, 36]. Some bio- mass studies did not have a non-woodland control, but instead used a non-fertilized treatment as a control [37]. Fewer studies were found investigating the effective- ness of slurry storage for improving water quality, the evidence often lacked rigour and many of the slurry stores used in the studies did not meet current EU legis- lation and the studies were from outside of the EU where legislation on storage may be different. The evidence for woodland creation and subsoiling was very limited and no evidence was found for the effectiveness of controlled trafficking on grassland for improving water quality. Modelling studies were excluded from the review. However, these studies are potentially useful as woodland experiments can take years to assess. The role of trees in pesticide drift reduction was not included in this review as pesticide was measured as a deposit rather than within water. Nisbet et al. recently reviewed the role of trees on water quality combining both woodland creation and buffer strip studies to provides a comprehensive review in this area [14]. Description of studies Only 5 studies were coded for subsoiling and these were all from outside of Europe (USA n = 4; Canada n = 1). Controlled traffic on grassland No studies were found for controlled traffic on grassland. Buffer strip studies were usually carried out on either loam or unknown soil types, which may not capture the effect of soil particle size on buffer strip performance. One multi-site study, with silt loam, and silt clay loam soils [39] noted that a wider buffer was needed for soils with a high clay content as soil particles were smaller and took longer to deposit in surface flow. Study designh The studies could be divided into three main categories: (1) Studies measuring water quality under afforested for- mer agricultural land, comparing results to cropped or Page 11 of 15 Randall et al. Environ Evid (2015) 4:18 reducing N followed by sediment, P, pesticides and FIOs or pathogens. forested land or measuring differences across different tree species. This included studies that reported findings from the AFFOREST project which measured the effect on water quality of afforestation on former agricultural soils in 3 different European countries [34]. (2) Studies measuring the effect on water quality over time of trees grown for biomass (3) Studies measuring the effect of water quality of trees intercropped with a cash crop. More than half of the cover/catch crops studies were carried out in Europe where studies focussed predomi- nantly on reducing N followed by fewer studies on sedi- ment, P and pesticides. No studies were found for FIOs or pathogens. The majority of the buffer strip and cover/catch crop studies were published in peer review journals and were of high rigour. However, much of this research is from field scale experiments and further research is needed to assess mitigation effectiveness at catchment scale. This research gap was also highlighted in a recent COST action knowledge exchange programme for buffer strips [38]. Studies of buffer strips and cover/catch crops were often of short duration and rarely recorded seasonal data. The majority of buffer strip studies included in this review were conducted only during the summer so the impact of rainfall events and mitigation effectiveness over time may not have been fully captured. Study design and outcomes measured l Buffer strip effectiveness was often assessed at field scale, which may not capture the effects of preferen- tial flow paths or buffer strip placement on buffer strip performance. A Defra commissioned buffer strip study at three sites representative of UK soil types [40] found no significant difference in levels of total-N, nitrate or molybdate reactive P in river samples taken from paired catchments (buffered and not). However, at the field site fans of sediment deposits were observed at the edge of the buffer strip and ground monitoring wells recorded reductions in nitrate and total N on buffer strip sites (not clear for P). One explanation given for the result was that phosphate could have been stored as sediment in the river and was acting as a source for sediment bound All the studies were manipulative and used a no-subsoil- ing control. Four out of the 5 subsoiling studies measured soil erosion and sediment loss from plots and one studied N and P. Major findingsh The majority of the evidence collated in this map inves- tigated the effectiveness of buffer strips and cover crops for improving water quality. More than half of the buffer strip studies originated from the northern states of the USA, with the remainder of the evidence from Europe. The majority of buffer strip studies focussed on Page 12 of 15 Page 12 of 15 Randall et al. Environ Evid (2015) 4:18 P which, until depleted, would mask any positive effects of buffer strip implementation. Another reason cited was that water flows may have bypassed the buffer strip either through underground drainage, or vertical movement into aquifers. Reductions in P measured at buffer strip plots not translating to reductions in stream samples have been observed in other studies [41]. The authors suggested that the study should have been longer than 2 years to observe the long term effectiveness of buffer strips. Variation in the effectiveness of buffer strips over time has also been noted in other studies [42, 43]. Dif- ferences between vegetation types such as grass and trees may only become apparent over time, as trees mature more slowly. Scientific rigour for slurry storage studies was variable with authors often suggesting results should be inter- preted with caution. For example, spraying of slurry on adjacent fields may have contaminated water sources, rather than the slurry storage unit. Furthermore, many studies had confounding factors as they were part of catchment studies, highlighting the difficulty in measur- ing water quality within river systems. In addition to the studies that examined leakage from slurry stores or reductions in FIOs in stored slurry, a few studies were also found that related to the timing of slurry application, even though this had not specifically been searched for. The timing of slurry application is an important consideration in current Nitrate Vulnerable Zone (NVZ) regulations in the UK, and a separate Rapid Evidence Assessment (REA) was subsequently commis- sioned and carried out regarding the alteration of timing to slurry application on water quality [50]. Variability in the hydrological landscape has been cited as an important factor for buffer strip effectiveness for leaching of N [44].f Seasonal differences in plant growth and nutrient uptake may impact on buffer strip effectiveness. Further analysis of the studies with data for all four seasons would be needed to identify any seasonal effect. Slurry storage Most of the evidence for slurry storage related to stud- ies measuring leakage from slurry stores, particularly for N and P. Many of the slurry storage studies were from outside of Europe, over 12 years old and used earth lined slurry stores and were therefore not relevant to current UK legislation. Limitations of the systematic map Articles lacking full text were coded on title and abstract which may result in the inclusion of some non-relevant studies. Only studies that demonstrated a direct effect of the intervention on water quality were included in the map, thereby excluding studies that measured indi- rect (but important) effects such as soil water infiltra- tion, crop yields, soil mineralization rates, and herbicide degradation. Furthermore, only overall outcomes were recorded for a study therefore differences in sampling location, mitigation, study site, and flow path were not captured. The only cover/catch crop study in the map that meas- ured water quality in stream/river samples was a long term catchment monitoring study (9–16  years) [49]. Cover/catch crop studies were often conducted on loam or unknown soil types, which may not capture differ- ences between soil types and nutrient leaching (e.g. sandy soils). Strengths and limitations of the review Limitations of the search Non English language search terms were excluded. However, over 100 articles in the map were assumed to be non-English language texts and included on titles/ abstracts. Their translation would extend the evidence base. For example, some woodland creation reports, writ- ten in French or German, were not coded on full text [51, 52]. Although web searches were conducted for a vari- ety of organisations, grey literature may be under-repre- sented, where it is not available online. Some included studies contained forms of the interventions that were not specifically searched for (e.g. winter wheat to provide a crop cover). These topics may be less comprehensively covered in the database. Cover crops Authors have suggested that a number of factors can impact on the effectiveness of cover/catch crops such as the amount of fertiliser applied, the crop rotation and crop or cover/catch crop type. Climatic data was often difficult to extract from stud- ies, however some studies reported year to year vari- ation in effectiveness depending upon the date when autumn rains started [12]. Only a quarter of the studies assessed effectiveness across all four seasons. However, a study reported in two articles cautioned that cover/catch crop effectiveness in reducing leaching of N should be assessed over the full crop succession [45, 46]. Although some studies were of long duration (up to 30 years), the impact of stopping cover/catch cropping was only inves- tigated in a few studies e.g. [47–49].h Cover/catch cropshf This systematic map can be used to help identify cases where there may be sufficient data on a specific question to justify a systematic review and identify evidence gaps for future primary research. • • The effect of time on the effectiveness of cover/catch crops.h • • The interaction between cover/catch crops and appli- cations of nitrogen and tillage.hf Knowledge gaps specific to the use of buffer strips • • The evidence base for slurry storage and effect on surrounding water quality is dated and may not relate to current/regional legislation. • • There was limited research investigating the effec- tiveness of buffer strips for reducing leaching of organic forms of N or P, or for pesticides that are cur- rently authorised for use/commonly used in UK agri- culture. •• Evidence for woodland creation, subsoiling and con- trolled trafficking on grassland, on water quality is limited or lacking. • • An evidence gap exists for the impact of cover/catch crops in reducing leaching of pesticides, FIOs and pathogens, and for organic forms of N and P. The systematic map provides a large database of research that can be used to filter information by miti- gation, water quality measurement or experimental fac- tor e.g. buffer width. This map could be of use to decision makers and delivery agencies to better facilitate catch- ment planning as required under the water framework directive [53–55].f Review conclusions • • There was very little research investigating the impact of subsoiling, controlled trafficking on grass- land or woodland creation on water quality.f Potential systematic review topics Evidence in the map often had a general inconsistency in approach that would make meta-analysis challenging for some interventions, for example slurry storage and woodland creation. The evidence base for cover/catch crop and buffer strips was more comprehensive and consistent and studies were of a higher scientific rigour compared to the other interventions investigated in this map and are more likely to be suitable for meta-analysis. Potential systematic review topics for these two interven- tions could include: Study outcomes for the effectiveness of some interven- tions for specific pollutants are recorded in the search- able database as outlined in the a priori protocol, but the findings are not synthesised in this systematic map. These are summarised in a report to Defra [56] and in an accompanying policy distillation [57]. Potential systematic review topics for these two interven- tions could include: Implications for policy/practiceh This systematic map documents and categorise all avail- able evidence on the effectiveness of buffer strips, cover/ catch crops, slurry storage, subsoiling, woodland crea- tion and controlled trafficking on grassland to improve water quality in terms of N, P, sediment, pesticides, FIOs and pathogens from faecal material. The map provides a useful output for decision makers to extract evidence on more specific areas of the topic. It can be used to high- light evidence gaps to direct future research funding and identify where there is sufficient evidence to answer spe- cific questions using systematic review (see implications for research).This map has specifically highlighted that: • • Few studies measured the effectiveness of interven- tions at catchment scale. • • Further, long term studies with controls, pre and post water quality measurements and multiple sampling points from both field and rivers would improve the evidence base. •• Further research investigating seasonal variations in the effectiveness of interventions, particularly buffer strips, woodland creation and cover/catch crops would also be useful. Priorities for primary research • • The effect of cover/catch crops compared to a cropped control (winter crop). Reporting of the primary research was variable, and improved reporting should be a priority for researchers. Standard reporting of statistics in water related stud- ies and submission of data with journal papers would increase the value of reported data, facilitate the reap- plication of data to subsequent analyses and ensure that Limitations in hierarchy of evidenceh The standard values that were applied to all studies may have excluded important water quality specific or experi- mental design factors that were not considered. Further- more, many related factors (such as the potential for Page 13 of 15 Page 13 of 15 Randall et al. Environ Evid (2015) 4:18 water quality data is not lost to science. In addition to this issue, the following key knowledge gaps were identified: pollution swapping) have not been considered by this work. pollution swapping) have not been considered by this work. Competing interests Th h d l h The authors declare that they have no competing interests. 22. Randall NP, James KL (2011) The effectiveness of integrated farm manage- ment, organic farming and agri-environment schemes for conserving biodiversity in temperate Europe-A Systematic Map. Environ Evid 1:4 22. Randall NP, James KL (2011) The effectiveness of integrated farm manage- ment, organic farming and agri-environment schemes for conserving biodiversity in temperate Europe-A Systematic Map. Environ Evid 1:4 ukga/1991/57/contents 21. Collaboration for Environmental Evidence (2010) Guidelines for systematic review in environmental management. Version 4.0. Environmental evidence 21. Collaboration for Environmental Evidence (2010) Guidelines for systematic review in environmental management. Version 4.0. Environmental evidence Authors’ contributions 11. Dabney S, Delgado J, Reeves D (2001) Using winter cover crops to improve soil and water quality. Commun Soil Sci Plant Anal 32:1221–1250f 11. Dabney S, Delgado J, Reeves D (2001) Using winter cover crops to improve soil and water quality. Commun Soil Sci Plant Anal 32:1221–1250 NPR, PLJ and LMD conceived and planned the systematic map. PLJ provided guidance on environmental quality and protection, and was a subject expert for buffer strips and slurry storage. LMD undertook the systematic map. NPR provided advice for systematic map activities. KJ and NPR wrote the manu- script. All authors read and approved the final manuscript. 12. Shepherd M (1999) The effectiveness of cover crops during eight years of a UK sandland rotation. Soil Use Manag 15:41–48 12. Shepherd M (1999) The effectiveness of cover crops during eight years of a UK sandland rotation. Soil Use Manag 15:41–48 13. Townsend LG (2010) What evidence is there that the utilization of catch crops can reduce nitrate leaching in temperate Europe? MSc thesis. Harper Adams University 13. Townsend LG (2010) What evidence is there that the utilization of catch crops can reduce nitrate leaching in temperate Europe? MSc thesis. Harper Adams University Additional file 1:  Spread sheet containing the exact search terms used. Additional file 2:  Coding categories used in the systematic map. Additional file 3:  Access database of coded review evidence searchable by category. 10. Arrus KM, Holley RA, Ominski KH, Tenuta M, Blank G (2006) Influence of temperature on Salmonella survival in hog manure slurry and seasonal temperature profiles in farm manure storage reservoirs. Livest Sci 102:226–236 Sources of support This Systematic Map was funded by the Natural Environment Research Coun- cil and the Department for Environment Food and Rural Affairs under work order WT0965. 23. Randall NP, Donnison LM, Lewis PJ (2012) How effective are slurry storage, cover or catch crops, woodland creation, controlled trafficking or break- up of compacted layers, and buffer strips as on-farm mitigation measures for delivering an improved water environment? Environ Evid 1:12 Received: 27 November 2014 Accepted: 29 July 2015 Received: 27 November 2014 Accepted: 29 July 2015 Received: 27 November 2014 Accepted: 29 July 2015 24. Tonitto C, David M, Drinkwater L (2006) Replacing bare fallows with cover crops in fertilizer-intensive cropping systems: a meta-analysis of crop yield and N dynamics. Agric Ecosyst Environ 112:58–72 25. Zhang X, Liu X, Zhang M, Dahlgren RA, Eitzel M (2010) A review of veg- etated buffers and a meta-analysis of their mitigation efficacy in reducing nonpoint source pollution. J Environ Qual 39:76–84 Acknowledgements 14. Nisbet T, Silgram M, Shah N, Morrow K, Broadmeadow S (2011) Woodland for water: woodland measures for meeting water framework directive objectives. Forest Research Monograph, 4. Forest Research, Surrey 14. Nisbet T, Silgram M, Shah N, Morrow K, Broadmeadow S (2011) Woodland for water: woodland measures for meeting water framework directive objectives. Forest Research Monograph, 4. Forest Research, Surrey The authors are grateful to the following subject experts from Harper Adams University for their comments and suggestions in the drafting of the protocol: Jim Waterson (Woodland creation), Nigel Hall (cover/catch crops) and Dick Godwin (loosening compacted soils, controlled trafficking and slurry storage). The authors would like to thank the librarians at Harper Adams University, and in particular Mathew Bryan for his help in ordering articles. The authors are also grateful to stakeholders Defra, The Natural Environment Research Council, the Environment Agency and the Forestry Commission for their input at review meetings. The authors also thank the anonymous reviewers for their comments on earlier drafts of the manuscript. 15. Donnison LM (2011) Review of the effects of farmland trees on erosion and pollution in the local farmed landscape. Report to the Woodland Trust by Harper Adams University. The Woodland Trust, UK 15. Donnison LM (2011) Review of the effects of farmland trees on erosion and pollution in the local farmed landscape. Report to the Woodland Trust by Harper Adams University. The Woodland Trust, UK 16. Hamza M, Anderson W (2005) Soil compaction in cropping systems: a review of the nature, causes and possible solutions. Soil Tillage Res 82:121–145 16. Hamza M, Anderson W (2005) Soil compaction in cropping systems: a review of the nature, causes and possible solutions. Soil Tillage Res 82:121–145 17. Jasa PJ, Dickey EC (1991) Subsoiling, contouring and tillage effects on erosion and runoff. Appl Eng Agric 7:81–85 17. Jasa PJ, Dickey EC (1991) Subsoiling, contouring and tillage effects on erosion and runoff. Appl Eng Agric 7:81–85 Whilst we consider that this work has been carried out in accordance with good industry practice, the University will not be liable for any use which may be made, reliance which may be placed, nor advice or information given, in connection with the results contained herein for commercial purposes. 18. Mayer PM, Reynolds SK, McCutchen MD, Canfield TJ (2007) Meta-analysis of nitrogen removal in riparian buffers. J Environ Qual 36:1172–1180f 18. Buffer stripsh Department for Environment, Food and Rural Affairs, UKf 8. Burke A (2011) Synthesis of diffuse pollution research in England and Wales funded by Department for Environment, Food and Rural Affairs and Environment Agency. Demonstrating Catchment Management, UK 8. Burke A (2011) Synthesis of diffuse pollution research in England and Wales funded by Department for Environment, Food and Rural Affairs and Environment Agency. Demonstrating Catchment Management, UK Buffer stripsh There are some pre-existing meta-analyses which meas- ured changes in levels of sediments, N, P and pesticides [17, 23, 32] as measured along the length of a buffer strip Page 14 of 15 Randall et al. Environ Evid (2015) 4:18 (comparing inflow/outflow) suggesting good potential for future systematic review. Other potential topics include: (comparing inflow/outflow) suggesting good potential for future systematic review. Other potential topics include: 2. Department for Environment (2009) Food and Rural Affairs (UK): safe- guarding our soils. Department for Environment, Food and Rural Affairs, UK 2. Department for Environment (2009) Food and Rural Affairs (UK): safe- guarding our soils. Department for Environment, Food and Rural Affairs, UK 3. Collins AL, Anthony SG, Hawley J, Turner T (2009) Predicting potential change in agricultural sediment inputs to rivers across England and Wales by 2015. Mar Freshw Res 60:626–637 3. Collins AL, Anthony SG, Hawley J, Turner T (2009) Predicting potential change in agricultural sediment inputs to rivers across England and Wales by 2015. Mar Freshw Res 60:626–637 • • The effect of time on the effectiveness of buffer strips.hff hfff • • The effect of pollutant solubility on mitigation effec- tiveness e.g. P. y 4. Edwards A, Withers P (2008) Transport and delivery of suspended solids, nitrogen and phosphorus from various sources to freshwaters in the UK. J Hydrol 350:144–153 y 4. Edwards A, Withers P (2008) Transport and delivery of suspended solids, nitrogen and phosphorus from various sources to freshwaters in the UK. J Hydrol 350:144–153 • • The effect of buffer strips compared to a cropped or bare ground control.f 5. Kay D, Crowther J, Fewtrell L, Francis C, Hopkins M, Kay C et al (2008) Quantification and control of microbial pollution from agriculture: a new policy challenge? Environ Sci Policy 11:171–184 • • Further research is also needed to capture the effects of preferential flow paths, buffer strip placement and buffer vegetation type on efficacy in order to maxim- ise the effectiveness of buffer strips. y g y 6. Agency Environment (2004) The State of soils in England and Wales. Environment Agency, UK 6. Agency Environment (2004) The State of soils in England and Wales. Environment Agency, UK 7. Department for Environment (2011) Food and Rural Affairs (UK): water for life white paper. Department for Environment, Food and Rural Affairs, UKf 7. Department for Environment (2011) Food and Rural Affairs (UK): water for life white paper. 1. Herzog F, Steiner B, Bailey D, Baudry J, Billeter R, Bukácek R et al (2006) Assessing the intensity of temperate European agriculture at the land- scape scale. Eur J Agron 24:165–181 • • Additional files Additional file 1:  Spread sheet containing the exact search terms used. Additional file 2:  Coding categories used in the systematic map. Additional file 3:  Access database of coded review evidence searchable by category. Additional file 1:  Spread sheet containing the exact search terms used. Additional file 2:  Coding categories used in the systematic map. Additional file 3:  Access database of coded review evidence searchable by category. Additional file 1:  Spread sheet containing the exact search terms used. Additional file 2:  Coding categories used in the systematic map. Additional file 3:  Access database of coded review evidence searchable by category. 9. Newell Price JP, Harris D, Taylor M, Williams JR, Anthony SG, Duethmann D et al (2011) An inventory of mitigation methods and guide to their effects on diffuse water pollution, greenhouse gas emissions and ammonia emissions from agriculture. Department for Environment, Food and Rural Affairs, UKl Acknowledgements Mayer PM, Reynolds SK, McCutchen MD, Canfield TJ (2007) Meta-analysis of nitrogen removal in riparian buffers. J Environ Qual 36:1172–1180 19. Muscutt A, Harris G, Bailey S, Davies D (1993) Buffer zones to improve water quality: a review of their potential use in UK agriculture. Agric Ecosyst Environ 45:59–77 19. Muscutt A, Harris G, Bailey S, Davies D (1993) Buffer zones to improve water quality: a review of their potential use in UK agriculture. Agric Ecosyst Environ 45:59–77 y 20. Water Resources Act 1991 http://www.legislation.gov.uk/ ukga/1991/57/contents y 20. Water Resources Act 1991 http://www.legislation.gov.uk/ ukga/1991/57/contents References Bechmann M, Deelstra J, Stalnacke P, Eggestad HO, Oygarden L, Pengerud A (2008) Monitoring catchment scale agricultural pollution in Norway: policy instruments, implementation of mitigation methods and trends in nutrient and sediment losses. Environ Sci Policy 11:102–114 y 50. Waterson A, Randall NP (2013) What impact does the alteration of timing to slurry applications have on leaching of nitrate, phosphate and bacte- rial pathogens? A Rapid Evidence Assessment. Water Security Knowledge Exchange Programme http://www.wskep.net/documents.php y 34. Rosenqvist L (2007) Afforestation of former arable land in North-Western Europe. PhD thesis. University Of Uppsala 51. Aureau F (2008) Afforestation, a benefit for water in Brittany (in french). RenDez Vous Techniques 22:44–46 35. Jussy J-H, Colin-Belgrand M, Ranger J (2000) Production and root uptake of mineral nitrogen in a chronosequence of Douglas-fir (Pseudotsuga menziesii) in the Beaujolais Mounts. For Ecol Manage 128:197–209 52. Anton K (1999) Afforestation of agricultural lands as a means of conserv- ing drinking water in the water conservation zone of Holdorf (in german). Forst und Holz 54:404–408 36. Van der Salm C, van der Gon HD, Wieggers R, Bleeker A, van der Toorn A (2006) The effect of afforestation on water recharge and nitrogen leach- ing in The Netherlands. For Ecol Manag 221(1–3):170–182 53. Fish R, Potschin M (2009) Catchment planning and the ecosystems approach: progress towards application LWEC pilot review. Contract Number: R8/H12/107. Nottingham University, UK 37. Jr Mortensen, Hauge Nielsen K, JØrgensen U (1998) Nitrate leaching dur- ing establishment of willow (Salix viminalis) on two soil types and at two fertilization levels. Biomass Bioenergy 15:457–466 54. Brown LE, Mitchell G, Holden J, Folkard A, Wright N, Beharry-Borg N et al (2012) Priority water research questions as determined by UK practition- ers and policy makers. Sci Total Environ 409:256–266 38. Stutter MI, Chardon WJ, Kronvang B (2012) Riparian buffer strips as a mul- tifunctional management tool in agricultural landscapes: introduction. J Environ Qual 41:297–303 55. McGonigle D, Harris R, McCamphill C, Kirk S, Dils R, Macdonald J et al (2012) Towards a more strategic approach to research to support catchment-based policy approaches to mitigate agricultural water pollu- tion: a UK case-study. Environmental Science & Policy 39. Syversen N, Borch H (2005) Retention of soil particle fractions and phos- phorus in cold-climate buffer zones. Ecol Eng 25:382–394 56. References 1. Herzog F, Steiner B, Bailey D, Baudry J, Billeter R, Bukácek R et al (2006) Assessing the intensity of temperate European agriculture at the land- scape scale. Eur J Agron 24:165–181 26. Hansen K (ed) (2002) Literature review for AFFOREST: planning afforesta- tion on previously managed arable land—influence on deposition, nitrate leaching, and carbon sequestration. http://www.fsl.dk/afforest 26. Hansen K (ed) (2002) Literature review for AFFOREST: planning afforesta- tion on previously managed arable land—influence on deposition, nitrate leaching, and carbon sequestration. http://www.fsl.dk/afforest Page 15 of 15 Page 15 of 15 Page 15 of 15 Randall et al. Environ Evid (2015) 4:18 27. Correll DL (2005) Principles of planning and establishment of buffer zones. Ecol Eng 24:433–439 45. Herrera JM, Liedgens M (2009) Leaching and utilization of nitrogen dur- ing a spring wheat catch crop succession. J Environ Qual 38:1410–1419 46. Herrera JM, Stamp P, Liedgens M (2005) Root development of catch crops and nitrate losses by leaching after spring wheat. Aspects Appl Biol 73:35–40 28. Stevens A, Milne R (1997) The effectiveness revolution and public health. In: Scalley G (ed) Progress in Public Health. Royal Society for Medicine Press, London, pp 197–225 47. Hansen EM, Djurhuus J, Kristensen K (2000) Nitrate leaching as affected by introduction or discontinuation of cover crop use. J Environ Qual 29:1110–1116 29. Pullin AS, Knight TM (2001) Effectiveness in conservation practice: point- ers from medicine and public health. Conserv Biol 15:50–54 30. Pullin AS, Knight TM (2003) Support for decision making in conservation practice: an evidence-based approach. J Nat Conserv 11:83–90 30. Pullin AS, Knight TM (2003) Support for decision making in co practice: an evidence-based approach. J Nat Conserv 11:83–90 48. Constantin J, Beaudoin N, Laurent F, Cohan J-P, Duyme F, Mary B (2011) Cumulative effects of catch crops on nitrogen uptake, leaching and net mineralization. Plant Soil 341:137–154 31. Landis JR, Koch GG (1977) The measurement of observer agreement for categorical data. Biometrics 33:159–174 32. Liu X, Zhang X, Zhang M (2008) Major factors influencing the efficacy of vegetated buffers on sediment trapping: a review and analysis. J Environ Qual 37:1667–1674 49. Constantin J, Mary B, Laurent F, Aubrion G, Fontaine A, Kerveillant P et al (2010) Effects of catch crops, no till and reduced nitrogen fertilization on nitrogen leaching and balance in three long-term experiments. Agric Ecosyst Environ 135:268–278 33. References Donnison LM, Lewis PJ, Smith B, Randall NP (2013) How effective are slurry storage, cover or catch crops, woodland creation, controlled trafficking or break-up of compacted layers, and buffer strips as on-farm mitigation measures for delivering an improved water environment? Final report for Defra, work order WT0965 40. Leeds-Harrison PB, Quinton JN, Walker MJ, Harrison KS, Gowing DJ, Tyrrel SF (1996) Report NT1101 buffer zones. Department for Environment, Food and Rural Affairs, UK f 41. Newbold JD, Herbert S, Sweeney BW, Kiry P, Alberts SJ (2010) Water qual- ity functions of a 15-Year-old riparian forest buffer system. J Am Water Resour Assoc 46:299–310 57. Randall NP, Donnison LV (2013) The value of on-farm interventions for improving water quality. What is the evidence? Policy distillation report for Defra, work order WT0965 57. Randall NP, Donnison LV (2013) The value of on-farm interventions for improving water quality. What is the evidence? Policy distillation report for Defra, work order WT0965 42. Barfield BJ, Blevins RL, Fogle AW, Madison CE, Inamdar S, Carey DI et al (1998) Water quality impacts of natural filter strips in karst areas. Trans Asae 41:371–381 43. Dillaha TA, Reneau RB, Mostaghimi S, Lee D (1989) Vegetative filter strips for agricultural nonpoint source pollution control. Trans Asae 32:513–519i 44. Angier JT, McCarty GW, Prestegaard KL (2005) Hydrology of a first-order riparian zone and stream, mid-Atlantic coastal plain, Maryland. J Hydrol 309:149–166 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit • Convenient online submission
W4226328916.txt
https://www.degruyter.com/document/doi/10.1515/bd-2022-0035/pdf
de
Die Bibliotheken, die Welt der Bücher und die heutigen Trends<b>: Haben traditionelle bibliothekarische Angebote überhaupt noch eine Zukunft?</b>
Bibliotheksdienst
2,022
cc-by
4,288
 Bibliotheksdienst 2022; 56(3–4): 227–239 Urs Bisig Die Bibliotheken, die Welt der Bücher und die heutigen Trends: Haben traditionelle bibliothekarische Angebote überhaupt noch eine Zukunft? Libraries, the World of Books, Current Trends – What Future is There for Traditional Library Services? https://doi.org/10.1515/bd-2022-0035 Zusammenfassung: Mit der fortschreitenden Digitalisierung wächst der Informationssektor immer schneller, die Orientierung darin wird immer schwieriger und anspruchsvoller, und der Komfort für die druckaffinen Leser nimmt ständig ab. Im Beitrag wird gezeigt, wie das Bibliothekswesen mithelfen kann, diese Entwicklung besser zu bewältigen. Der Autor ist der Meinung, das Ziel der Bibliotheken sollte nicht nur sein, möglichst bald Teil der E-Welt zu werden, sondern vor allem auch, diese intelligent zu ergänzen. Schlüsselwörter: Bibliothek, Digitalisierung, E-First-Politik, Kritik Abstract: Ongoing digitization sees the information sector growing faster and faster, the lack of orientation increases, and users interested in print products are more and more inconvenienced. The paper explores ways in which the library system could better cope with these developments. It is the author’s belief that the purpose of libraries is not just to quickly become part of the ever-expanding e-world, but to complement this world in intelligent ways. Keywords: Library, digitization, e-first policy, criticism Urs Bisig: urs.bi@bluewin.ch Open Access. © 2022 Urs Bisig, publiziert von De Gruyter. Dieses Werk ist ­lizensiert unter einer Creative Commons Namensnennung 4.0 International Lizenz. 228 Urs Bisig 1 Fragestellung Wollten Sie auch schon mal ein Buch ausleihen und dann gab es das gesuchte Werk in Ihrer Bibliothek „nur“ in E-Form und Sie haben sich darüber geärgert? Oder war es bei Ihnen gerade umgekehrt: Sie bekamen ein gedrucktes Buch, hätten aber die E-Form – aus welchen Gründen auch immer – bevorzugt? Das Erstere kommt immer häufiger vor, denn die Digitalisierung durchdringt die Bibliothekswelt immer mehr. Seit einiger Zeit herrscht eine eigentliche Digitalisierungseuphorie: E-Medien sind in, P-Medien (Printmedien) out. Das Ideal vieler bibliothekarischer Entscheidungsträger scheint es zu sein, möglichst bald möglichst viel digital im Netz anbieten zu können, wobei gedruckte Werke höchstens noch als Ergänzung vorgesehen sind.1 Wer die Begeisterung für das Digitale nicht teilt ist von gestern, nein von vorgestern. Diese Bibliothekare wollen nicht mehr viel mit Büchern zu tun haben. Selbst das Wort „Buch“ ist schon verdächtig und wird von einigen Bibliotheksverantwortlichen nur noch ungern verwendet. So schaffte es beispielsweise die Zentralbibliothek Zürich in ihrem letzten Flyer „WissensWelt“, in dem sie ihre verschiedenen Dienstleistungen anpries, das Wort „Buch“ kein einziges Mal zu erwähnen.2 Für die einseitige Bevorzugung des Digitalen bei der an und für sich begrüßenswerten Annäherung der Bibliotheken an die E-Welt gibt es u. a. folgende Gründe: – Im Gegensatz zu früher, als die historischen Wissenschaften die Bibliothekswelt prägten, sind es heute die empirischen Wissenschaften (STM-Bereich und Teile der Sozialwissenschaften), welche das Bibliothekswesen stark beeinflussen. Für diese Fächer ist Schnelligkeit ein ausschlaggebender Faktor, weshalb in diesem Bereich die E-Medien unverzichtbar sind. Die empirischen Wissenschaften gelten auch als die Disziplinen, deren Nutzen für die moderne Gesellschaft und Wirtschaft leicht nachvollzogen werden kann, weshalb deren Leitbildfunktion auch im Bibliothekswesen vielen Verantwortlichen als gerechtfertigt erscheint. – Der (fast) totale Medienwandel in vielen Wissenschaften erleichtert auch die Suche von zahlreichen Entscheidungsträgern nach einem zeitgemäßen Image für ihre Institution: Mit viel Elektronik und möglichst wenig Papier hoffen sie, ihre Bibliothek an den Mainstream andocken zu können und damit das Ansehen ihres Hauses in der Öffentlichkeit zu steigern. Keiner/ 1 Z. B. Bibliotheken: Weg damit! Interview von Michael Furger mit Rafael Ball. In: NZZ am Sonntag vom 07.02.2016, S. 25, https://nzzas.nzz.ch/hintergrund/bibliotheken-und-buecher-wegdamit-meint-rafael-ball-ld.147683 [Zugriff: 14.02.2022]. 2 WissensWelt. Die Zentralbibliothek Zürich entdecken. Zürich. Die Bibliotheken, die Welt der Bücher und die heutigen Trends – 229 keine von ihnen möchte die allerneuesten Trends verpassen und als altmodisch gelten. Die Kritik von einigen „ewiggestrigen“ Bücherfreunden3 ist für sie weit weniger gravierend als das Bild in der Öffentlichkeit von Bibliotheken als antiquierte, angestaubte Bücherlager. Schließlich seien noch die global tätigen Großverlage und Tech-Giganten erwähnt, die mit ihren digitalen Angeboten versuchen, die Informationswelt immer mehr zu dominieren und die Bibliotheken an sich zu binden. Ihnen ist es nicht gleichgültig, wie diese den Medienwandel angehen. Je verunsicherter die Bibliotheken sind, desto eher wird es den internationalen Großkonzernen gelingen, diese von sich abhängig zu machen. Leidtragende in diesem Powerplay sind neben den druckaffinen Nutzerinnen und Nutzern vor allem die kleinen und mittelgroßen Verlage, für die nicht mehr so viele finanzielle Ressourcen zur Verfügung stehen, nachdem viel Geld für all die E-Book-Pakete und anderen Spezialangebote der Großen ausgegeben worden ist. Heutzutage werden die Bibliotheken mit solchen Paketen regelrecht „geflutet“ und ihr Literaturangebot wird dadurch – und auch wegen der zunehmenden Auslagerung (von Teilen) der Erwerbung – immer uniformer. „Die Ware gleicht sich von Kiel bis Konstanz“4 und darüber hinaus, wobei der Anteil des Gedruckten an den neu zugänglichen Texten (fast) überall merklich zurückgeht. Eine weitere Folge dieser Entwicklung ist, dass in unserem Bereich heute auch in den allgemeinen Fachzeitschriften, nicht nur in den spezialisierten, vielen Fachbüchern und an den Tagungen in der Regel die Digitalisierung im Zentrum steht. Über klassische Themen wie Kataloge und gedruckte Bücher wird nicht mehr so häufig geschrieben und gesprochen, und wenn, dann manchmal gar abwertend. Ein Außenstehender könnte meinen, es gäbe für uns keine anderen Probleme mehr. Deshalb stellt sich schon die Frage: Ist diese Entwicklung gerechtfertigt? Haben die traditionellen bibliothekarischen Angebote ausgespielt und wirklich keine Zukunft mehr – oder vielleicht doch? 3 Vgl. z. B. Hagner, Michael: Zur Sache des Buches. 2. überarb. Aufl. Göttingen 2015 oder Yogeshwar, Ranga: Original oder digital. Schriftliches Kulturgut ist ein Schatz und eine Verpflichtung. In: BuB 67 (2015), S. 60–64, oder Knoche, Michael: Die Idee der Bibliothek und ihre Zukunft. Göttingen 2018. 4 Knoche (Anm. 3) S. 27. 230 Urs Bisig 2 A  uch Bücher sind und bleiben wichtig Bevor wir zu den eigentlichen Aufgaben der Bibliotheken kommen, vergegenwärtigen wir uns die heutige Lage im Informationssektor: Das Info-Meer wächst und wächst und zwar immer schneller; es wird immer unübersichtlicher und komplexer. Deshalb wird es für viele Menschen immer schwieriger und anspruchsvoller, sich darin kompetent – und nicht nur irgendwie – zurechtzufinden. Um den Überblick zu bewahren, sollten wir uns die klassische Grobeinteilung der Fachtexte in Erinnerung rufen. Die einfachste Unterteilung ist diejenige nach der Länge der Texte. Dabei können wir folgende drei Ebenen unterscheiden: Kürzesttexte Kurzinformationen, gefunden mithilfe einer einfachen Suche im Internet, bzw. aus Nachschlagewerken und Enzyklopädien, heute am bekanntesten die Wikipedia; für Schnellinformationen zu einem Thema. Kurztexte Aufsätze: Studien und Essays; geeignet für die Mitteilung von neuen Erkenntnissen und/oder die aktuelle Diskussion. Langtexte Bücher: Monografien und redigierte Textsammlungen; gerne benutzt für die Orientierung in einem Gebiet und/oder für die Erörterung eines Problems. Eine solche Einteilung hat natürlich immer auch etwas Willkürliches, denn die Textarten sind nicht starr, sondern fließen ineinander über. Dennoch kann die Unterteilung mithelfen, das Informationschaos besser in den Griff zu bekommen. Alle drei Ebenen sind wichtig und haben ihre spezifische Aufgabe. Im Folgenden soll der Fokus vor allem auf die Langtexte gelegt werden, für deren Zugänglichkeit und Auffindbarkeit traditionellerweise die Bibliotheken die Hauptverantwortung tragen. Die Ebene der Langtexte ist zentral und besonders wichtig für die vertiefte Auseinandersetzung mit einem Thema. Das Buch bietet in der Regel einen guten Überblick über einen Gegenstand und/oder beschäftigt sich ausführlich mit einer Frage, bzw. einem Fragekomplex in einem mehr oder weniger großen Bereich, wobei das auf ganz unterschiedlichem Anspruchsniveau sein kann. Auch die Glaubwürdigkeit einer Publikation ist von Bedeutung, die u. a. von der Seriosität des Verbreiters einer Information abhängt. Diese Kriterien werden immer wichtiger in Zeiten der zunehmenden Unübersichtlichkeit und der Fake News. Bei den Informationsträgern lassen sich zurzeit zwei widersprüchliche Tendenzen beobachten: Einerseits eine rasante Entwicklung der E-Welt, anderseits ein zähes Überleben der gedruckten Werke. Die P-Medien sind nämlich, vor allem im Bereich der Langtexte, noch lange nicht tot, sondern sie leben – wenn auch in etwas reduziertem Umfang – munter weiter. Der Medienwandel findet zwar statt, Die Bibliotheken, die Welt der Bücher und die heutigen Trends 231 aber nicht total wie letztes Mal vor über 500 Jahren beim Übergang vom handgeschriebenen zum gedruckten Buch, sondern er ist eher vergleichbar mit dem Aufkommen der Autos, die zwar der Eisenbahn immer größere Konkurrenz machten, diese jedoch (mindestens bei uns in Mitteleuropa) nie vollständig verdrängt und ersetzt haben; im Zusammenhang mit dem steigenden Umweltbewusstsein kommt es heute gar zu einem eigentlichen Revival der Bahnen. Ein Grund für die Beliebtheit von gedruckten Werken trotz der E-Welt und deren Möglichkeiten ist der größere Lesekomfort, den sie besonders beim vertieften Lesen bieten und den viele zu schätzen wissen. Ebenso mögen es zahlreiche Menschen, einmal nicht ständig auf einen Bildschirm schauen zu müssen, sondern zur Abwechslung einen Text auch mal physisch zu erleben. Und dann spricht die größere Vielfalt des Gedruckten die Leute an: Printmedien haben eine eigene Individualität, während die meisten Veröffentlichungen auf dem Bildschirm recht monoton daherkommen. Alles Gründe, weshalb selbst an der EDV interessierte Leserinnen und Leser ab und zu Gedrucktes vorziehen, was sich u. a. im breiten Printangebot im IT-Bereich zeigt. Sogar P-Books über E-Books sind keine Seltenheit.5 Ein weitere Ursache für das hartnäckige Überleben der analogen Texte ist die Aufwertung des Buches durch die digitalen Medien, „denn wenn in Zukunft etwas überhaupt gedruckt wird, so bedeutet das doch, dass es besonders wichtig sein muss.“6 Dies ist auch ein Grund, weshalb viele Autorinnen und Autoren von Langtexten ihr Werk nach der großen Arbeit nicht auf der gleichen Stufe sehen möchten wie die Texte im Internet. Das gedruckte Buch gilt nämlich in breiten Kreisen immer noch als das Premiumprodukt der Informationswelt. Printmedien werden uns deshalb noch lange begleiten. Den Durchschnittsleser, die Durchschnittsleserin gibt es jedoch nicht, die Nutzertypen sind sehr verschieden: Es gibt Menschen, die mehr technikaffin und andere, die mehr druckaffin sind. Für viele kommt es auch auf die Länge der Texte an; Langtextleser bevorzugen häufig P-Medien, Kurztextleser E-Medien.7 Am verbreitetsten sind wahrscheinlich die „Switcher“, also Menschen, die sich in beiden Welten mehr oder weniger wohl fühlen. Ihr präferiertes Medium ist abhängig vom 5 Z. B. Ward, Suzanne M.; Freeman, Robert S.; Nixon, Judith M. (eds.): Academic e-books. Publishers, librarians and users. West Lafayette (IN) 2016, oder Graf, Dorothee; Fadeeva, Yuliya; Falkenstein-Feldhoff, Katrin (Hrsg.): Bücher im Open Access. Ein Zukunftsmodell für die Geistesund Sozialwissenschaften? Opladen 2020. 6 Aussage von Katharina Blarer zitiert in Wirz, Tanja: Ein Buch, ein Thema. In: Journal. Die Zeitung der Universität Zürich 41 (2011/Februar) 1, S. 2. 7 Vgl. z. B. die Nutzer*innenbefragung 2020 der Deutschen Nationalbibliothek – Ergebnisbericht, https://www.dnb.de/SharedDocs/Downloads/DE/Benutzung/nutzerumfrage2020Ergebns ibericht.pdf? S. 4, Punkt 4. [Zugriff: 14.02.2022]. 232 Urs Bisig Problem, bzw. der Fragestellung. Ein gutes Beispiel dafür ist Hagner, der sich in seinem Werk „Zur Sache des Buches“ vehement für das gedruckte Buch einsetzt, dessen Anmerkungsapparat jedoch zahlreiche Hinweise auf Netzpublikationen enthält.8 Und als letztes sei noch auf die beiden Wissenswelten hingewiesen, auf den STM-Bereich (und die empirischen Sozialwissenschaften) einerseits und auf die Geistes- und Kulturwissenschaften anderseits, die sich in ihrem Denken und in ihrer Methodik grundlegend unterscheiden.9 Auch das Publikationsverhalten ist in beiden Bereichen verschieden: Während empirisch tätige Wissenschaftler möglichst viele Kurztexte in angesehenen Zeitschriften zu veröffentlichen suchen, möchten Geistes- und Kulturwissenschaftler ihre Erkenntnisse traditionellerweise gerne auch einmal in einem Buch über einen renommierten Verlag verbreiten. Ein Grund dafür ist, dass für diese Fachleute oft auch der Kontext wichtig ist und nicht nur die allerneuesten Forschungsergebnisse in einem klar abgrenzbaren, engen Bereich. Auch ist die Halbwertszeit ihres Wissens, bzw. ihrer Überlegungen in der Regel deutlich länger als die Erkenntnisse im Bereich ihrer empirisch arbeitenden Kolleginnen und Kollegen. Außerdem möchten die meisten Buchautorinnen und -autoren die Lesenden auch sprachlich ansprechen und nicht nur schnell Fakten mitteilen. Schließlich gibt es hier vermehrt noch die „Konkurrenz“ von Persönlichkeiten außerhalb der akademischen Welt, die ihre (zum Teil originellen) Überlegungen, wenn möglich, (auch) in gedruckter Form publizieren. Deshalb kann man sich schon fragen: Wie sollen die Bibliotheken auf diese widersprüchliche Entwicklung reagieren? Sollen sie sich möglichst schnell und umfassend in die E-Welt integrieren oder im Gegenteil versuchen, diese auch zu ergänzen? 3 Die Bibliotheken, das Buch und andere Herausforderungen Auch wenn die Bibliotheken heute nicht mehr die alleinigen Stützen der Informationswelt sind, so können sie sehr wohl ein Big Player dort bleiben, wenn es ihnen gelingt, den Nutzerinnen und Nutzern einen Mehrwert zu bieten. Besonders stark sind die Bibliotheken im Bereich der Orientierungshilfen und des Komforts, wo 8 Hagner (Anm. 3) S. 251–280. 9 Unter vielen anderen vgl. Snow, Charles P.: Die zwei Kulturen. Literarische und naturwissenschaftliche Intelligenz. Stuttgart 1967. Die Bibliotheken, die Welt der Bücher und die heutigen Trends 233 sie den Usern verschiedene Erleichterungen anbieten können. Aus der Sicht der Tradition lassen sich acht Hauptaufgaben unterscheiden: 3.1 A  ufgaben im Bereich der Orientierung 1. Eine wichtige Aufgabe von Bibliotheken ist ihre Rolle als Orientierungshelfer. Wie wir im letzten Abschnitt gesehen haben, sind für die Orientierung in der Wissenswelt, vor allem für die vertiefte Orientierung, Langtexte (Bücher) der Königsweg – für Schnellinformationen reicht in der Regel eine kurze Suche im Internet, oft genügt die Wikipedia. Auch sind Bücher in den Geistes- und Kulturwissenschaften – also in den Orientierungswissenschaften – weiterhin von großer Bedeutung, wenn auch ihre Stellung dort nicht mehr so zentral ist wie auch schon. 2. Für die Orientierung in der Welt der Langtexte braucht es gute Suchinstrumente. Diese gelten als gut, wenn bei einem Treffer sowohl der Verlust als auch der Ballast möglichst gering ist – und zwar bei allen Fragestellungen, nicht nur bei sehr engen (z. B. Mozarts Träume). Google-nahe Suchinstrumente haben zwar den Vorteil, dass sie „in“ sind und die Benutzer damit vertraut; nachteilig ist hier jedoch u. a., dass nicht immer fein zwischen einer Person als Urheber und einer Person als Gegenstand unterschieden werden kann, und bei umfassenderen Themen (z. B. Psychologie/Allgemeines) die Trefferzahlen riesig werden; aber gerade solche Übersichtsdarstellungen sind es, die normalerweise in Büchern veröffentlicht, bzw. dort gesucht werden. Außerdem ist die Resultatgewinnung mithilfe von Suchmaschinen oft schwer nachvollziehbar. Deshalb sollten die traditionellen Erschließungsinstrumente, bei denen Normierung, Strukturierung und sorgfältige Zuordnung zu einer Katalogstelle eine Rolle spielen, weiterhin gepflegt und nicht so schnell aufgegeben werden. 3. Auch die Unterstützung der Nutzerinnen und Nutzer beim Sich-Zurechtfinden in der immer komplexer werdenden Informationswelt ist Teil des Orientierungsangebotes. Im Idealfall kann das Bibliothekspersonal den Usern auf allen Stufen helfen – von den Laien oder Studienanfängern bis zu den Doktorierenden, die ihre Dissertation in OA-(Open Access-)Form zu veröffentlichen haben. Die Bibliotheken sind ja die Institutionen, von denen erwartet wird, dass sie Informationskompetenz haben und diese auch professionell weitergeben können. 4. Und dann ist da noch das Problem der steigenden Publikationsflut wegen der tiefer liegenden Veröffentlichungsschwelle im Internet. Walach meint dazu: „Heute kann scheinbar jeder alles veröffentlichen …, [weshalb die] Auswahl 234 Urs Bisig und [das] Filtern von Information und die Bewertung des Publizierten immer zentraler werden.“10 Das ist jedoch nicht immer ganz einfach. Denn wer entscheidet, ob eine freie Netzpublikation ein Qualitätsproblem hat und deshalb von der Bibliothekswelt nicht unbedingt voll berücksichtigt werden muss oder ob sie aus anderen Gründen nicht genehm ist? Dennoch werden sich die Bibliotheken – vor allem bei den Langtexten – vermehrt mit dieser Problematik auseinandersetzen müssen. 3.2 Aufgaben im Bereich des Komforts 1. Die Bibliotheken sind auch für den Komfort im Informationswesen zuständig. Dieser lässt sich nur selten mit einer einzigen Medienform abdecken, denn je nach Person und Fragestellung sind die Ansprüche an den Komfort verschieden: Manchmal sind es E-Medien, manchmal P-Medien, welche den Nutzern am besten dienen. Beide Medienarten haben in der Bibliothekswelt ihre Berechtigung, was eine immer weiter verbreitete E-First-Politik jedoch außer Acht lässt. Deshalb kann man sich schon fragen: Ist es sinnvoll, primär E-Medien anzubieten, selbst wenn sich immer wieder zeigt, dass bei Wahlfreiheit im Langtextbereich die meisten Nutzerinnen und Nutzer das gedruckte Werk in der Regel vorziehen?11 Viele Bibliotheksverantwortliche mögen dieses Benutzerverhalten zwar bedauern, ignorieren sollten sie es jedoch nicht, auch weil die Bibliotheken hier etwas bieten könnten, was das Internet so nicht geben kann. In diesem Zusammenhang haben sich die Entscheidungsträger auch stets bewusst zu sein, dass an den Diensten der Bibliotheken nicht nur die Forschungsfront der Naturwissenschaften und der empirischen Sozialwissenschaften interessiert ist, sondern dass der Nutzerkreis viel breiter ist. Auf Bibliotheken angewiesen sind neben interessierten Laien – diese sind übrigens auch Wähler und Steuerzahler – Menschen in der Aus- und Weiterbildung, Generalisten, interdisziplinär Arbeitende, Praktiker, Multiplikatoren (Lehrerinnen und Journalisten), Intellektuelle – und natürlich auch die in den Geisteswissenschaften tätigen Fachleute. Ideal wäre es deshalb, wenn die Nutzerinnen und Nutzer selber entscheiden könnten, welches Medium sie konsultieren möchten. 10 Walach, Harald: Psychologie. Wissenschaftstheorie, philosophische Grundlagen und Geschichte. Ein Lehrbuch. 3. überarb. und erw. Aufl. Stuttgart 2013, S. 66. 11 Vgl. Anm. 7. Die Bibliotheken, die Welt der Bücher und die heutigen Trends 2. 235 Zum Komfort gehören auch die Bibliotheken als Orte. Sie sind Orte des Lernens, der Begegnung und der Anregung, aber auch der Ruhe und des Denkens. Unsere hypernervöse Zeit braucht Rückzugsorte für die Geistesarbeitenden und nicht weitere Erlebniszentren für möglichst viele. Auch Sportanlagen wurden und werden in erster Linie für die Sporttreibenden gebaut und nicht für alle. Das Ziel der Bibliotheken muss es deshalb nicht sein, Orte für alle zu werden, jedoch sollten sie Wohlfühloasen für alle bleiben, die lesen und mit dem Kopf arbeiten wollen und das unabhängig von ihrem schulischen Background. 3.3 Aufgaben in anderen Bereichen 1. 2. Traditionellerweise sind die Bibliotheken auch dafür zuständig, dass der Zugang zum Wissen möglichst günstig, unkompliziert und vollständig ist. Leider gibt es damit bei vielen kostenpflichtigen elektronischen Publikationen immer wieder Probleme, denn sie können in vielen Fällen nicht so frei wie Druckwerke zur Verfügung gestellt werden. Oft sind sie nur für die jeweiligen Hochschulangehörigen einsehbar und/oder die Fernleihe ist erschwert bis unmöglich, was der Aufgabe von Bibliotheken widerspricht, für möglichst alle da zu sein. Last but not least ist eine weitere wichtige Funktion von Bibliotheken ihre Aufgabe als Archiv, sowohl von gedruckten wie auch von digitalen Medien. Als Gedächtnisinstitutionen sind die Bibliotheken in erster Linie für den immateriellen Teil des kulturellen Erbes zuständig, das heißt, dass einmal veröffentlichte Erkenntnisse, Ideen und Gedanken nicht verloren gehen. Dazu gehört indirekt aber auch die „Hardware“, also die konkreten Bücher und Zeitschriften. Ein Forscher, eine Forscherin, die sich beispielsweise mit einem Aspekt des 18. Jahrhunderts auseinandersetzt, sollte Texte aus dieser Zeit auch ab und zu mal sehen, ja spüren können und ihnen nicht nur am Bildschirm begegnen. Um die Zugänglichkeit und Sicherheit solcher Schätze zu erhöhen, wäre es wünschenswert, wenn diese möglichst dezentral zur Verfügung stünden. Wie die erwähnten Punkte in der Bibliothekswelt konkret umgesetzt werden könnten, wird im nächsten Abschnitt thematisiert. 236 Urs Bisig 4 Eine mögliche Entwicklung Das Ideal ist also nicht die digitale Bibliothek, sondern die Hybridbibliothek, eine Institution, welche die E-Welt und die P-Welt gleichermassen berücksichtigt. Das geht leider nicht flächendeckend, da eine solche Umsetzung zu teuer wäre. Eine Lösung dieses Problems könnte sein, die Bibliotheken als System zu sehen: Nicht jede Institution ist hybrid, jedoch das Bibliothekswesen als Ganzes. Mit anderen Worten: Es müsste zu einer weiteren Spezialisierung entlang der (Haupt-)Informationsträger kommen. Folgende Bibliothekstypen sind in diesem Zusammenhang denkbar: – Informationszentren, wo in erster Linie die E-Medien gepflegt werden und wo auch eine umfassende, professionelle Beratung und Schulung in diesem Bereich erwartet werden kann. Dort sind die User umgeben von Bildschirmen, weniger von Büchern. Diese Institutionen sehen sich als Teil der Forschungsinfrastruktur, vor allem im Bereich der STM-Fächer sowie der empirischen Sozialwissenschaften. Für diese Disziplinen sind Kurztexte in E-Form, Schnelligkeit und Hyperspezialisierung im Publikationsbereich typisch, was diese „Bibliotheken“ zu berücksichtigen haben. – Orientierungszentren, die sich primär den Langtexten (Büchern) widmen, zunächst einmal unabhängig vom Informationsträger. Ihr Schwerpunkt aber ist das traditionelle Buch. Sie sehen sich ähnlich wie Museen, Theater und Konzertsäle vor allem als Teil der kulturellen Infrastruktur, aber auch als Zentren für die Geistes- und Kulturwissenschaften sowie als Wohlfühloasen für die Wissensarbeitenden. Mit der E-First-Politik von vielen Institutionen werden Orte, wo das gedruckte Buch noch gepflegt wird, noch exklusiver. Das heißt aber noch lange nicht, dass es in der neuen klassischen Bibliothek aussieht, wie in einer solchen Institution vor fünfzig, sechzig Jahren. Selbstverständlich werden dort neben vielen Büchern auch Bildschirme zu sehen sein, denn diese braucht es unter anderem für die Suchinstrumente. Und zur Ergänzung des Bestandes sollten den Nutzerinnen und Nutzern „Ausflüge“ in die Welt der E-Medien ermöglicht werden, denn auch bei den Langtexten wird heute längst nicht mehr alles gedruckt, immer mehr Inhalte von älteren P-Books sind heute frei im Netz zugänglich und die Forschenden in den Geisteswissenschaften brauchen zunehmend auch Digitalisate. – Einige große Universal- und Spezialbibliotheken schaffen es, dem Ideal (Hybridbibliothek) nahe zu kommen und können in beiden Welten ein breites Angebot – und einen entsprechenden Service – zur Verfügung stellen. In der Regel sind das auch die Institutionen, die für die Langzeitarchivierung zuständig sind. Dieser Bibliothekstyp wird sich jedoch nicht so stark ausbreiten, weil die Kosten dafür enorm sind. Die Bibliotheken, die Welt der Bücher und die heutigen Trends – 237 Im Bereich der Öffentlichen Bibliotheken scheint sich ein immer größerer Teil von ihnen an keinem der vorgestellten Modelle zu orientieren, sondern entfernt sich zunehmend von der Bibliotheks-„Familie“. Diese „Bibliotheken“ sehen ihre Zukunft immer weniger als Orte von Büchern, sondern immer mehr als moderne Freizeit- und Bildungszentren, wo der (wenn möglich digitale) Informationssektor nur noch ein Teilbereich ist. Dafür gibt es dort Escape Rooms, Makerspaces, Gaming-Ecken, Workshopräume, Tonstudios, Lounges usw. Durch eine solche Arbeitsteilung wird die Bibliothekslandschaft noch vielfältiger und bunter werden. Als Folge davon sollten sich die Bibliotheken immer weniger als Einzelkämpfer sehen, sondern müssen sich immer mehr als Teil eines Ganzen verstehen.12 Wichtig ist nicht ein Kampf um Ressourcen, sondern Kooperation und Koordination unter den Bibliotheken sowie eine umfassende Information der User über die verschiedenen Angebote. 5 Schlussbemerkungen Obwohl die Bibliotheken heute nicht mehr die exklusiven Orte des Wissens sind wie früher, so gibt es für sie weiterhin eine Menge von Aufgaben. Einige der wichtigsten sind: – Das Bibliothekswesen bietet kostengünstige, niederschwellige Zugänge zur gesamten Informationswelt für alle – nicht nur für einen Teilbereich und nicht nur für einen ausgewählten Kreis. – Die Bibliotheken sind besonders verantwortlich für: – die Orientierung in der Info-Welt – durch Langtexte, Kataloge und Beratung, – den Komfort in der Info-Welt – durch Angebote von E- und P-Medien und – die Qualität in der Info-Welt – durch Auswahl. – Die Bibliotheken bieten ihre Institutionen auch als Orte an mit dem Ziel, dass sich bei ihnen möglichst viele Leute des Denkens wohl fühlen. Da heute in den meisten Fällen nicht mehr eine Institution allein das ganze Informationsspektrum gut abdecken kann, wird die Spezialisierung immer wichtiger: Statt dass viele Bibliotheken versuchen, ein möglichst breites Angebot zu präsen- 12 Vgl. z. B. Knoche (Anm. 3) S. 23 und S. 119. 238 Urs Bisig tieren, sehen sich die meisten nur noch für einen Teilbereich zuständig. Dieser wird dann aber besonders sorgfältig und professionell gepflegt. Was die Corona-Pandemie anbelangt, so haben wir dadurch zwar verstärkt einige Vorteile der Digitalisierung13 erfahren können, die Krise hat aber auch offenbart, wie verletzlich unsere moderne Gesellschaft ist. Möglicherweise ist es nicht ein biologisches Virus, welches eine nächste größere Beeinträchtigung im Bibliotheksbetrieb auslösen wird, sondern ein Computervirus, ein Hackerangriff, eine länger andauernde Störung im Netz oder gar ein Datenverlust wegen was auch immer. Für Digitalenthusiasten ist das zwar (fast) undenkbar, aber SuperGAUs in Kernkraftwerken waren auch unmöglich bis es sie gab. Im Falle eines Worst Case werden wahrscheinlich selbst die technikaffinsten Leute froh sein, dass viele Informationen dezentral auch noch auf Papier zur Verfügung stehen und nicht nur im Netz. Und noch etwas – eher Unerwartetes – hat uns die Coronazeit gezeigt: Die gedruckten Bücher haben gar nicht ein so schlechtes Image, wie uns viele technikaffine Vordenker immer wieder weismachen wollen. In Zeiten von Homeoffice und Videokonferenzen dienen vielen Leuten dekorative Bücherwände als Hintergrund für ihr neues „Büro“.14 Umgeben von Büchern fühlen sich diese Menschen wohl und/oder sie wollen sich so als gebildet und belesen zeigen. Eine solche Inszenierung wäre jedoch fatal, gälte der Bücherbesitz heute als ein Zeichen von hoffnungsloser Rückständigkeit. Auch das Treffen der beiden Präsidenten Biden und Putin in Genf vom 16. Juni 2021 fand in einem gediegenen Bibliotheksraum statt und eben gerade nicht in einem seelenlosen Saal voller Technik.15 Bücher und andere traditionelle bibliothekarische Angebote sind also noch lange nicht obsolet, sondern im Gegenteil: Sie sind geradezu ideale Instrumente zur Ergänzung der modernen Informationswelt. Dennoch wird sich die E-Welt, vor allem auch der Open-Access-Ansatz, wahrscheinlich immer mehr ausbreiten. Umso wichtiger ist es dann, dass es noch Institutionen gibt, welche auch gedruckte Medien berücksichtigen und die E- und P-Welt intelligent miteinander verknüpfen. Wir sollten die beiden Bereiche nicht als unversöhnliche Gegensätze sehen, sondern als Teile einer Welt, der Informationswelt: Gefragt ist mehr Sowohl-als-auch- und weniger Entweder-oder-Denken. Was Walach zum Wissenschaftsbetrieb meinte, kann ohne weiteres auch aufs Bibliothekswesen übertra13 Hinweise darauf u. a. in der Beitragssammlung COVID-19 trifft die Bibliothekswelt. In: b.i.t.online 23 (2020) 3, S. 239–276. 14 Klette, Kathrin: Die Bibliothek als Dekoration. In: Neue Zürcher Zeitung vom 05.02.2021, S. 16, https://www.nzz.ch/panorama/buecher-im-home-office-die-bibliothek-als-dekoration-nzzld.1599805 [Zugriff: 14.02.2022]. 15 https://de.wikipedia.org/wiki/Genfer_Gipfelkonferenz_(2021) [Zugriff: 14.02.2022]. Die Bibliotheken, die Welt der Bücher und die heutigen Trends 239 gen werden: „Monokulturen sind nicht nur im normalen Leben, sondern auch in der Wissenschaft hässlich und schädlich.“16 Die Bibliodiversität, also die Vielfalt in der Buchkultur, sollte deshalb von der Bibliothekswelt gepflegt und nicht ignoriert werden. Es lässt sich sogar behaupten: Je wichtiger die Ebene der Langtexte von den Bibliotheksverantwortlichen genommen wird, desto besser werden viele traditionelle Bibliotheksangebote sein. Die Bibliotheken müssen aufpassen, dass sie bei ihrer Annäherung an die E-Welt nicht zu abhängig von den internationalen Großverlagen, Tech-Giganten und weiteren gewinnorientierten Unternehmen werden. Eine gute Zusammenarbeit mit diesen Firmen ist zwar wichtig und hilfreich, die Bibliotheken müssen jedoch ebenso schauen, dass sie möglichst unabhängige Player bleiben und ihr eigenes Profil weiterentwickeln können. Dem Bibliothekswesen wäre mehr Selbstvertrauen zu wünschen, denn es kann auch einen wertvollen Beitrag zur Bewältigung von Orientierungsproblemen in der Wissenswelt, für den Komfort im Informationssektor und zur Bekämpfung von Fake-News leisten. Urs Bisig Eierbrechtstr. 11 8053 Zürich Schweiz E-Mail: urs.bi@bluewin.ch 16 Walach (Anm. 10) S. 313.
https://openalex.org/W4378951263
https://zenodo.org/records/7993910/files/DESAFIOS%20E%20AVAN%C3%87OS%20DA%20EDUCA%C3%87%C3%83O%20P%C3%9ABLICA%20%C3%80S%20CRIAN%C3%87AS%20AUTISTAS_%20UM%20OLHAR%20CR%C3%8DTICO%20%C3%80%20LUZ%20DA%20JUSTI%C3%87A%20DE%20INCLUS%C3%83O%20BRASILEIRA%20%E2%80%93%20ISSN%201678-0817%20Qualis%20B2.pdf
Portuguese
null
DESAFIOS E AVANÇOS DA EDUCAÇÃO PÚBLICA ÀS CRIANÇAS AUTISTAS: UM OLHAR CRÍTICO À LUZ DA JUSTIÇA DE INCLUSÃO BRASILEIRA
Zenodo (CERN European Organization for Nuclear Research)
2,023
cc-by
9,064
DESAFIOS E AVANÇOS DA EDUCAÇÃO PÚBLICA ÀS CRIANÇAS AUTISTAS: UM OLHAR CRÍTICO À LUZ DA JUSTIÇA DE INCLUSÃO BRASILEIRA Ciências da Saúde, Ciências Jurídicas, Ciências Sociais Aplicadas, Edição 123 JUN/23 / 01/06/2023 REGISTRO DOI: 10.5281/zenodo.7993910 Palavras-chave: Autismo. Educação. Inclusão. Legislação Abstract: This study deals with the right to education of autistic children, especially in the public sphere, analyzing from its historical context to its legal framework, highlighting the challenges faced during its implementation. The Convention on the Rights of Persons with Disabilities and brazilian national legislation establish the fundamental right of all children, including those with autism, to inclusive and quality education. However, reality demonstrates a gap between theory and practice. Thus, this work aims to examine the main obstacles that limit the full exercise of this right guaranteed by the Brazilian Abstract: This study deals with the right to education of autistic children, especially in the public sphere, analyzing from its historical context to its legal framework, highlighting the challenges faced during its implementation. The Convention on the Rights of Persons with Disabilities and brazilian national legislation establish the fundamental right of all children, including those with autism, to inclusive and quality education. However, reality demonstrates a gap between theory and practice. Thus, this work aims to examine the main obstacles that limit the full exercise of this right guaranteed by the Brazilian Federal Constitution*, raising the question of whether in fact public schools in Brazil are fulfilling their duty to create an inclusive school environment for children with Autistic Spectrum Disorder – ASD, in accordance with what is expressed by the Brazilian Inclusion and Education Laws. For this, a narrative bibliographical research was carried out, of an inductive nature, carried out in the period from 2022 to 2023. As a result of this review, it was possible to identify some problems, such as: the lack of adequate training of teaching professionals, the scarcity of financial and material resources, the lack of curricular adaptation and the discrimination faced by autistic children in the school environment. Keywords: Autism. Education. Social Inclusion. Legislation REGISTRO DOI: 10.5281/zenodo.7993910 Maria Eduarda Evelin de Oliveira Imily Pereira e Silva Wirna Maria Alves da Silva 1 2 3 Resumo: O presente estudo trata sobre o direito à educação das crianças autistas, especialmente no âmbito público, analisando desde seu contexto histórico até seu enquadramento legal, destacando os desafios enfrentados durante sua implementação. A Convenção sobre os Direitos das Pessoas com Deficiência e a legislação nacional estabelecem como direito fundamental de todas as crianças, incluindo aquelas com autismo, à educação inclusiva e de qualidade. No entanto, a realidade demonstra uma lacuna entre a teoria e a prática. Deste modo, esse trabalho tem como objetivo examinar os principais obstáculos que limitam o pleno exercício desse direito garantido pela Constituição, levantando o questionamento sobre se de fato as escolas públicas do Brasil estão cumprindo seu dever de criar um ambiente escolar inclusivo para as crianças com Transtorno do Espectro Autista- TEA, indo de acordo com o que é expresso pelas Leis de Inclusão e Educação Brasileiras. Para isso, foi realizada uma pesquisa bibliográfica narrativa, de caráter indutivo, realizada no período de Constituição, levantando o questionamento sobre se de fato as escolas públicas do Brasil estão cumprindo seu dever de criar um ambiente escolar inclusivo para as crianças com Transtorno do Espectro Autista- TEA, indo de acordo com o que é expresso pelas Leis de Inclusão e Educação Brasileiras. Para isso, foi realizada uma pesquisa bibliográfica narrativa, de caráter indutivo, realizada no período de 2022 a 2023. Como resultado dessa revisão, puderam-se identificar alguns problemas, como: a falta de capacitação adequada dos profissionais de ensino, a escassez de recursos financeiros e materiais, a falta de adaptação curricular e a discriminação enfrentada pelas crianças autistas no ambiente escolar. 1 INTRODUÇÃO Com os avanços da medicina e científicos, tem sido possível diagnosticar o Transtorno do Espectro Autista (TEA) de forma mais eficaz e precoce, e, com uma quantidade maior de diagnósticos, mais claro se torna ao Estado, as instituições de ensino e á população as dificuldades na inclusão e no auxílio dessas pessoas, especialmente à inclusão no sistema de educação, desde a básica ao ensino superior. O presente trabalho empenha-se a realizar estudos relacionados à garantia do direito à educação da criança com espectro autista e se esse direito tem sido de fato respeitado e seguido por parte do Estado e das instituições de ensino público, e quais seriam os empecilhos e dificuldades enfrentados durante o processo de cumprimentos destas leis? Tendo noção da importância dessa discussão e análise, são trazidos raciocínios partindo do mesmo pressuposto: a proteção integral e inclusiva da criança, que é um bem jurídico essencial e tutelado pelo Estado, cabendo ao mesmo o auxilio, amparo, proteção e principalmente a inclusão por meio de políticas publicas e sociais, que ofereçam as melhorias necessárias ao atendimento destas crianças portadoras do espectro autista, incluindo o atendimento pedagógico multidisciplinar, acompanhamentos com profissionais especializados como expresso na Lei Berenice Piana – Lei Federal n° 12.764 e a Lei Brasileira de Inclusão a Pessoa com Deficiência – Lei Federal n° 13.146. Deve ser prezado pelo cuidado á criança não apenas na área da educação, mas em seu desenvolvimento social, tendo o sistema de inclusão e auxílio como pontos primordiais para a ajuda no crescimento e na vida em sociedade dessas crianças. Este trabalho visa apontar as necessidades e o descaso do poder público á essa parte da população, focando naqueles que dependem do ensino fornecido pelo Estado. É importante ser pontuado que a educação é a base de todo e qualquer ser humano, e fornecer uma educação de qualidade e inclusiva é o mínimo que o Estado pode proporcionar, fazendo valer as garantias previstas e preservando a dignidade de cada criança. Tendo a importância da educação como base o estudo ira analisar a qualidade do ensino público fornecido ate a alfabetização, observando onde as leis citadas acima estão sendo cumpridas por meio de uma educação inclusiva de qualidade, aprimoramento do ensino e o aperfeiçoamento de profissionais qualificados para lidar com a situação. 1 INTRODUÇÃO Segundo pesquisa feita pelo censo escolar (INEP, 2019) houve um crescimento considerável de crianças autistas matriculadas em escolas comuns após a implantação das duas Leis que regulam essa inclusão no meio escolar, tanto no meio privado quanto no público, mas ainda é uma parcela relativamente pequena comparada com a quantidade de pessoas no Brasil que possuem TEA, o que nos leva a questionar os motivos destas crianças não estarem na escola, quais dificuldades enfrentadas por elas que fazem os pais desistirem de colocá- las em escolas comuns? Diante de todo o exposto, se torna visível a necessidade de uma quantidade maior de estudos críticos acerca do tema, para que seja cobrada das autoridades uma fiscalização de qualidade em cima de todas as leis vigentes. Seguindo a regra da proporcionalidade, quanto maior for à quantidade de informações acerca de determinado assunto, maior vai ser a comoção e mobilização gerada, sendo ela necessária para a responsabilização dos setores competentes pelas falhas na execução das leis. 2.1 A História do Autismo O ponto de partida para os estudos científicos sobre o TEA tem como marco na invenção do quadro diagnóstico criado por Leo Kanner, psiquiatra Austríaco, mais conhecido como Pai da psiquiatria. Todo o contexto histórico e científico do Autismotem como base principal a descrição e a observância de um compilado de comportamentos e características que somadas fizeram com que fosse criado um “quadro” diferente dos que já existiam. O diagnóstico psicopatológico do autismo se alicerça nas descrições e formulações diagnósticas de Kanner (2012, p. 168). Kanner se tornou o um dos maiores psiquiatras infantis, após a família Triplett buscar sua ajuda para avaliar e diagnosticar, Donald, o filho do casal em 1938. Antes da primeira consulta, Kanner solicitou que os Pais da criança contassem um pouco do que se tratava o caso, ao ler o relato do Pai, surge então à história do autismo, “[…] pouco a pouco aquela que viria a ser a descrição seminal de uma criança com autismo, palavra e diagnósticos ainda inexistentes” (Donvan e Zucker, 2017, p. 44). criança com autismo, palavra e diagnósticos ainda inexistentes” (Donvan e Zucker, 2017, p. 44). Ao passo que Kanner, através das cartas de Mary, Mãe de Donald, transcrevia os relatos do comportamento do seu filho, lia e se envolvia com o caso, ele se tornará um grande amigo. Em várias cartas, Kanner se demonstra um parceiro tranquilizador, que escuta o que ela tem a dizer, designando por ‘heroico’ (Donvan e Zucker, 2017). Em 1942, em uma das cartas trocadas com Mary ele afirma “[…] reconhecer pela primeira vez um distúrbio que, até agora, não foi descrito pela literatura psiquiátrica nem por nenhuma outra” (Donvan e Zucker, 2017, p. 50), reiterando ter encontrado outras crianças com problemas similares, de tal modo que “[…] se há um nome a ser aplicado ao problema de Don e das outras crianças” (p. 50), poderia designá-lo “[…] distúrbio autista do contato afetivo” (p. 50). Kanner, fez da história de Donald, o principal motivo dos estudos que temos até hoje sobre o comportamento de pessoas que naquela época já entendiam como um “distúrbio inato que impediam o contato destas crianças com outros” (Bialer, 2018). Anos foram passando, e a busca pelo motivo causador do transtorno foi aumentando, quando no ano de 1948, a revista Times, em uma das suas matérias de grande repercussão, rotulou as crianças portadoras do Espectro Autista como “[…] esquizoides de fralda […]” (p. 91) que ficam “[…] felicíssimos quando estavam sozinhos […]” (p. 91), remetidos a “[…] pais frios […]” (p. 91), citando que essas crianças “[…] ficavam simplesmente guardadas em uma geladeira que não descongelava” (91). Tal fato ocasionou uma grande onda de especulações e culpabilização social acerca da “mãe geladeira” causadora do autismo, remetendo à etiologia do autismo uma discussão sobre a culpa materna. Logo após essas publicações, Kanner se tornou famoso e considerado o pioneiro a respeito das descobertas cientificas e patológicas sobre o autismo e o mundo inteiro descobriu suas observações e indagações, o que levou os diagnósticos de autismo a se tornarem mais frequentes, repercutindo um movimento acusatório dos genitores, em especial as mães. Na época, o fato de Kanner ter culpabilizado os pais, mas principalmente as mães, pela “falha” no desenvolvimento social e afetivos das crianças gerou grande manifestações de pais ressentidos. Mas, para Kanner a falta de fraternidade materna e “não amar o suficiente” até aquele momento era o maior motivo da falta de afetividade por parte das crianças possuidoras do Espectro. criança com autismo, palavra e diagnósticos ainda inexistentes” (Donvan e Zucker, 2017, p. 44). Por muitos anos, perdurou no tratamento e na abordagem dos autistas uma leitura culpabilizadora dos pais, por este motivo os autistas eram tratados em instituições que consideravam seus métodos como “[…] maternidade terapêutica […]” (Donvan e Zucker, 2017, p.113) e prometiam tratar a falta de afeto por parte dos pais. Foi quando começaram a surgir pais de crianças autistas que iam de encontro aos métodos e ao modo em que a suposta “causa” do autismo teria surgido. A exemplo, Ruth Sulivan, mãe de sete crianças dentre as quais somente uma era autista, se dedicou a estudar desde que verificou a condição de seu filho e tornou-se pioneira e militante em defesa da então conhecidas como “mãe de geladeira”. Com o decurso do tempo, os estudos científicos foram se aprimorando, e uma figura de grande destaque neste período foi Bernard Rimland, cujo filho era portador do Espectro. Após observar o sofrimento de sua esposa, passou a sustentar uma narrativa que problematizava todo o material publicado a respeito das “mães de geladeira”, pois ao observar o cuidado de sua esposa com seu filho, chegou à conclusão de que, além de cruel, culpabilizar os pais pelo transtorno sofrido pelos filhos era sem fundamento. Rimland passou, então, a acreditar que o autismo estava relacionado a causas orgânicas que, segundo ele, precisavam de tratamentos relacionados, principalmente, às questões nutricionais, trazendo como principal mudança à alteração na de alimentação e um estilo de vida mais saudável. Desde a “descoberta” do autismo por Kenner, houve muitas ramificações de estudos foram acerca do tema e muitos cientistas do mundo inteiro passaram a estudar especificamente o causador e como poderiam ser revertidos os casos de autismo. Atualmente, a ramificação mais recorrente e aceita no mundo da medicina é a da “Terapias ABA”, criada por Lovaas de ABA, inovando que as pessoas portadoras do Espectro autista precisavam de terapias e acompanhamento comportamentais. Em 1987, uma publicação na revista New York Times divulgou a taxa de 47% de eficácia método Lovaas, tornando esse estudo conhecido no mundo inteiro, tendo sido aprimorado e melhorado com a inserção de profissionais para a melhora no desenvolvimento social e educacional de cada indivíduo. Ao logo dos anos o número de crianças diagnosticadas com o Espectro se tornou muito preocupante, chamando atenção não só das autoridades médicas, mas também das autoridades políticas de todo o mundo. criança com autismo, palavra e diagnósticos ainda inexistentes” (Donvan e Zucker, 2017, p. 44). Em especial do Brasil, foram criadas leis que asseguram e resguardam os direitos dos portadores do Espectro Autista. Embora seja uma condição “nova”, o tema já se tornou um assunto bastante discutido e motor, tanto de alterações na legislação pátria vigente, como de criações de novas leis. É indiscutível, portanto, a importância histórica e todo o contexto que o Espectro Autista teve que sofrer até os dias de hoje, mesmo que muitas vezes os métodos de tratamento não fossem eficazes, tornavam-se uma luz no fim do túnel para acalmar o coração de tantos pais aflitos, que muitas vezes eram culpabilizados pelas condições dos filhos. 2.2 A trajetória da educação autista: da exclusão à inclusão Por um longo período de tempo, pessoas possuidoras de necessidades especiais eram afastadas do convívio social e abandonadas por serem consideradas um “fardo”, uma espécie de punição á familia. Quando as primeiras ações voltadas a elas começaram a surgir, vieram na forma de exclusão, por meio de “casas de apoio” onde ficavam afastadas da sociedade ou escolas especificas para que tivessem seu ensino próprio e só convivessem com seus iguais. Instituições especializadas em educação especial começaram a surgir e, durante a ditatura militar, houve a criação da Lei n° 5.692/1971 de Diretrizes e Bases Educacionais do Brasil, com a proposta de uma reforma no ensino brasileiro e para aqueles com deficiência física ou mental a garantia de um tratamento especial. Desta forma, não promovendo a inclusão nas escolas regulares, e sim determinando que essas crianças continuassem em escolas especializadas em seus tratamentos, afastadas de todos os outros. Educacionais do Brasil, com a proposta de uma reforma no ensino brasileiro e para aqueles com deficiência física ou mental a garantia de um tratamento especial. Desta forma, não promovendo a inclusão nas escolas regulares, e sim determinando que essas crianças continuassem em escolas especializadas em seus tratamentos, afastadas de todos os outros. Em 1988, foi promulgada a Constituição Federal do Brasil, tendo como um de seus objetivos fundamentais o bem de todos, não podendo haver preconceitos de origem, raça, cor, idade e qualquer outra forma de discriminação (BRASIL, 88). Garantindo, ainda, em seu Art. 5° o direito a educação a todos, sem distinção, sendo assim entendível que crianças com necessidades especiais poderiam, e deveriam, ser integradas em ambientes escolares regulares. Porém, em 1989, logo após a criação da Carta Magna, outra Lei sobre educação inclusiva se tornou vigente, a Lei n° 7.853, e nela é abordada a integração social das pessoas com deficiência e a responsabilização o poder público pela inserção de escolas especias na forma pública e privada, mais uma vez determinando novamente a exclusão dessas crianças do ambiente escolar regular. 2.2 A trajetória da educação autista: da exclusão à inclusão Um ano após a ultima mudança, na Conferência Mundial sobre Educação para Todos sediada em Jomtiem e que teve o Brasil como participante, ocorreu a aprovação da “Declaração Mundial e Educação para Todos” que possuía como um dos objetivos a propagação da equidade no que diz respeito o acesso à educação, em especial aos portadores de alguma deficiência, devendo ser tomadas medidas que garantissem a igualdade dessas pessoas no sistema educativo (BRASIL, 1990). Influenciada pela referida Declaração Mundial e todas reflexões e discussões geradas por ela foi criada a Lei n° 8.069/90, uma das grandes conquistas do direito brasileiro no que se diz a respeito do ser criança como sujeito de direito, que dispõe sobre o Estatuto a Criança e do Adolescente (ECA), elaborado com leis que protegem os direitos e os deveres das crianças e adolescentes no Brasil e garantem um desenvolvimento mental, físico, moral, espiritual e social adequado. Sobre as crianças portadoras de alguma deficiência especificamente o Estatuto garante um atendimento educacional especializado em redes regulares de ensinos, tornando-se a primeira lei que encarreggada da inclusão na educação regular. Em 1994, a UNESCO participou da “Conferência Mundial sobre Necessidades Educacionais” que, por meio da “Declaração de Salamanca” e seus princípios, políticas e práticas na área as necessidades educativas especiais, construiu uma visão em que a escola deve se adaptar as especificidades do aluno e não o contrário como costuma acontecer (UNESCO). Esta declaração acabou por servir de inspiração na elaboração da Lei de Diretrizes e Bases da Educação Nacional (LDB) n° 9.394/1996, que garante um serviço de apoio especializado nas escolas regulares, para quando se fizer necessário, de capacitação dos funcionários para auxiliar na integração dos alunos e criação de métodos que atendam da melhor forma as necessidades das crianças especiais e as ajude a atingir seu melhor desenvolvimento educacional, pessoal e em sociedade A Lei consolida o direito à educação, já assegurado pelo Estatuto da Criança e do Adolescente e pela Constituição Federal, reafirmando as obrigações e deveres por parte do Estado em proporcionar educação pública de qualidade e inclusiva, abrangendo toda e quailquer necessidade especial de cada criança e adolescente, definindo também os papeis e responsabilidades da União, Distrito Federal, Estados e Municípios. Em seu texto versa que a pessoa com deficiência necessita ter atenção e amparo especial. Art. 58. 2.2 A trajetória da educação autista: da exclusão à inclusão Entende-se por educação especial, para os efeitos desta Lei, a modalidade de educação escolar oferecida preferencialmente na rede regular de ensino, para educandos com deficiência, transtornos globais do desenvolvimento e altas habilidades ou superdotação. A Convenção sobre os Direitos as Pessoas com Deficiência foi adotada pela Organização das Nações Unidas (ONU) em 2006, tendo o Brasil como signatário, cujo dispositivo estabelece que os Estados-Partes devem assegurar um sistema de educação inclusiva em todos os níveis de ensino, em ambientes que aumentem o desenvolvimento não apenas academico, mas também social (BRASIL, 2007b, p.04). Com base nela foram criados os decretos n° 6.094/2007 que visa a mobilização social pela melhoria na qualidade da educação básica, e o n° 7.611/2011 que . determina que as medidas de apoio adotadas sejam individualizadas de acordo com as necessidades individuais no sistema educacional e ressalta os pontos discutidos na Convenção. Antes da criação de uma Lei específica para o autismo, a obrigação do Estado com os direitos dessas pessoas era expressa por meio de outras leis que abordavam casos de Portadores de Deficiência no geral. Porém, mesmo com as pessoas portadoras do Transtorno de Espectro Autista (TEA) estando inclusas nessas garantias, por terem uma condição especial necessitam de uma regulamentação mais especifica para melhorar sua condição de vida e resguardar seus direitos que são constantemente violados. Por esse motivo, no final do ano de 2012, foi sancionada a Lei Berenice Piana, n°12.764/12 instituída de “Politica Nacional de Proteção dos Direitos da Pessoa com Transtorno do Espectro Autista”, cujo nome lhe foi dado em homenagem à mãe de uma criança portadora de autismo, ativista da luta pelos direitos dos autistas. Esta Lei, escrita coletivamente por um grupo de indivíduossensibilizados pela causa e pelas dificuldades encontradas pelas portadores de TEA, teve como principal intuito considerar o sujeito possuidor do espectro como alguém com deficiência na forma legal, tendo os mesmos direitos reconhecidos como outros portadores. Dessa maneira, todos as leis que asseguravam os direitos da pessoacom deficiência passaram a assegurar também, formalmente, pessoas com TEA. Por esse motivo, no final do ano de 2012, foi sancionada a Lei Berenice Piana, n°12.764/12 instituída de “Politica Nacional de Proteção dos Direitos da Pessoa com Transtorno do Espectro Autista”, cujo nome lhe foi dado em homenagem à mãe de uma criança portadora de autismo, ativista da luta pelos direitos dos autistas. 2.2 A trajetória da educação autista: da exclusão à inclusão É dever do Estado, da família, da comunidade escolar e da sociedade assegurar educação de qualidade à pessoa com deficiência, colocando-a a salvo de toda forma de violência, negligência e discriminação Este dispositivo reafirma o dever do Estado em garantir a educação, a integração e socialização da criança autista durante todo o período escolar, além de assegurar que o indivíduo deve ser protegido e amparado, também, de toda e qualquer discriminação que pode ocorrer dentro do ambiente educacional, fazendo com que a qualidade de vida de cada criança e adolescente portador de alguma necessidade especial seja respeitada. Este dispositivo reafirma o dever do Estado em garantir a educação, a integração e socialização da criança autista durante todo o período escolar, além de assegurar que o indivíduo deve ser protegido e amparado, também, de toda e qualquer discriminação que pode ocorrer dentro do ambiente educacional, fazendo com que a qualidade de vida de cada criança e adolescente portador de alguma necessidade especial seja respeitada. Em 08 de janeiro de 2020, foi implantada a lei nº 13.977, denominada Lei Romeo Mion, que facilitou o acesso das pessoas com autismo aos direitos já garantidos na Lei Berenice Piana, definindo o transtorno como deficiência. Tal regulametação tornou-se uma conquista histórica para a luta de tantos pais e responsáveis pelo reconhecimento e inclusão dos portadores de TEA, já que foi garantido a eles os direitos já alcançados por indivíduos com deficiência, gerando uma maior visibilidade para essa causa, muitas vezes, não identificável visivelmente, o que torna indispensável a identificação através de documentos. Em 08 de janeiro de 2020, foi implantada a lei nº 13.977, denominada Lei Romeo Mion, que facilitou o acesso das pessoas com autismo aos direitos já garantidos na Lei Berenice Piana, definindo o transtorno como deficiência. Tal A falta da carteira impunha aos indivíduos nessa condição a necessidade de andar com laudos médicos para fazer prova da sua deficiência o que criava uma série de obstáculos ao acesso a atendimentos e a serviços prioritários aos quais têm direito, como se utilizar de filas preferenciais ou da vaga reservada para pessoas com deficiência em estacionamentos. 2.2 A trajetória da educação autista: da exclusão à inclusão Esta Lei, escrita coletivamente por um grupo de indivíduossensibilizados pela causa e pelas dificuldades encontradas pelas portadores de TEA, teve como principal intuito considerar o sujeito possuidor do espectro como alguém com deficiência na forma legal, tendo os mesmos direitos reconhecidos como outros portadores. Dessa maneira, todos as leis que asseguravam os direitos da pessoacom deficiência passaram a assegurar também, formalmente, pessoas com TEA. Tal norma também determina o acesso a um diagnostico precoce, tratamentos, terapias e medicamentos disponibilizados pelo Sistema Único de Saúde – SUS, proteção social e visa proporcionar a igualdade entre as pessoas. Ademais, com o foco do presente estudo, garante que essa criança possa frequentar uma escola regular e, caso for necessário, tenha um acompanhamento especial a estes alunos. Entretanto, a possibilidade de conseguir matricular essas crianças é apenas o primeiro passo, as escolas devem fornecer profissionais de educação capacitados para lidar com qualquer questão que venham a ter, como alunos especiais com suas necessidades individuais. Além de terapeutas e outros profissionais da área comportamental e de saúde, que se fazem necessários para um acompanhamento mais especializado, conforme resguardado no art 3, inciso IV da Lei do Autismo. No ano de 2015, houve a instauração da Lei Brasileira de Inclusão da Pessoa com Deficiência, Lei nº 13.146, que possui um compilado de dispositivos que cujo objetivo é assegurar, resguardar e promover, de modo igualitário e inclusivo, o cumprimentos e asseguramento dos direitos e liberdades fundamentais individuais e coletivas por pessoas com as mais variadas tipos de deficiências, visando a sua inclusão social e cidadania. Em seu Art. 27 o legislador reforçou a importância de incluir e adaptar a educação as criança especiais: Art. 27. A educação constitui direito da pessoa com deficiência, assegurados sistema educacional inclusivo em todos os níveis e aprendizado ao longo de toda a vida, de forma a alcançar o máximo desenvolvimento possível de seus talentos e habilidades físicas, sensoriais, intelectuais e sociais, segundo suas características, interesses e necessidades de aprendizagem. Parágrafo único. 2.2 A trajetória da educação autista: da exclusão à inclusão A carteira é uma identificação pessoal que facilita o exercício dos direitos garantidos na legislação, sendo uma iniciativa altamente democrática da Lei, a sua emissão possibilita a existência de um cadastro de pessoas com TEA, facilitando o acesso da criança portadora do espectro aos programas de políticas públicas e participações em programas governamentais, principalmente no âmbito escolar e educacional. 3 EDUCAÇÃO COMO GARANTIA CONSTITUCIONAL 3 EDUCAÇÃO COMO GARANTIA CONSTITUCIONAL 3 EDUCAÇÃO COMO GARANTIA CONSTITUCIONAL 3.1 Direito a educação como uma garantia universal e a falha na execução das leis Até meados do século XVIII toda educação que um homem poderia ter era a religiosa, sendo Deus a explicação para tudo que ocorria ao redor. Com a Revolução Francesa instalada e os ideais iluministas sendo disseminados pelo mundo, houve-se a necessidade de valorizar o pensamento científico e aperfeiçoamento do indivíduo. Com as novas formas e modelos trazidos pela revolução, a escolarização ate então obrigatoriedade dos órgãos religiosos passou a ser responsabilidade do Estado. No Brasil, durante o período colonial, o processo educativo também era responsabilidade das redes jesuítas, tendo como foco os dogmas da Igreja Católica. O aristocrata Marquês de Pombal, com receio de toda a dominação exercida pela igreja, implantou uma espécie de reforma educacional no território brasileiro, alterando a educação ate então religiosa para algo com foco na ciência e no próprio desenvolvimento do ser humano. Sendo ele aquele que oficializou a profissão de professor, mesmo antes das primeiras escolas voltadas a profissionalização dos docentes surgir, o que só veio acontecer em 1835. A promulgação da Constituição Federal de 1988 veio em meio a um período repleto de discussões acerca de direitos humanas e dos ideais de dignidade humana, dessa forma, sendo influenciada a ter como sua base o respeito e a qualidade de vida do homem. A Educação ganhou um capítulo voltado especificamente a ela dentro da constituição, capítulo III, tornando-a assim um direito social, ou seja, sua garantia é fundamental na vida de qualquer ser humano inserido em uma sociedade. O direito à educação é considerado um Direito humano universal, reconhecido e protegido pela Declaração Universal dos direitos Humanos, no artigo 26°, sendo ele essencial para o pleno desenvolvimento dos indivíduos e para a construção de sociedades justas e equitativas. No entanto, apesar do seu reconhecimento, muitos países, principalmente o Brasil, enfrenta desafios na sua implementação efetiva. 3 EDUCAÇÃO COMO GARANTIA CONSTITUCIONAL Uma das principais falhas na execução das leis que garantem esse direito está relacionada à falta de investimento adequado na área educacional, a má distribuição das verbas educacionais, a precariedade na infraestrutura das escolas públicas, a falta de capacitação contínua de todo corpo docente, escassez de materiais didáticos adequados, e a qualidade de ensino insatisfatória, são apenas uns das muitas falha que pode ser percebido quando é analisado a qualidade do ensino brasileiro especialmente para os grupos mais vulneráveis, como crianças em situação de pobreza, minorias étnicas em especial as crianças com deficiência. O direito à educação implica a inclusão de todos os indivíduos, independentemente de suas origens étnicas, religiosas, socioeconômicas ou condições físicas e mentais. No entanto, o sistema educacional brasileiro ainda não consegue proporcionar uma educação inclusiva, falhando em atender às necessidades específicas como das crianças portadoras do Espectro, o que acaba ferindo um direito que já é garantido por lei, e dificulta a vida e o desenvolvimento de crianças que necessitam do amparo e acolhimento do Estado e todos os órgãos públicos. Para superar as falhas na execução das leis de garantia do direito à educação, é necessário um compromisso político firme e uma alocação adequada de recursos. O estado deve priorizar a educação em seus orçamentos, garantindo a infraestrutura adequada, a formação contínua de professores e a implementação de políticas inclusivas. Além disso, é importante promover a participação ativa da sociedade civil, incluindo pais, estudantes e organizações não governamentais, na formulação e monitoramento das políticas educacionais. Para que haja uma superação dos obstáculos que a falta de investimento na educação acarreta, é necessário um compromisso político sólido, recursos adequados e uma abordagem inclusiva, visando garantir a educação de qualidade para todos, sem exceção 4 INCLUSÃO A CRIANÇA AUTISTA: FORA E DENTRO DA SALA DE AULA 4 INCLUSÃO A CRIANÇA AUTISTA: FORA E DENTRO DA SALA DE AULA 4.1 Desafios no processo de escolarização da criança autista e os novos métodos de ensinar A escolarização de crianças autistas é um desafio complexo que requer abordagens educacionais diferenciadas e adaptadas às necessidades individuais de cada criança. Neste tópico, será discutido os desafios enfrentados no processo de escolarização da criança neurodivergente e a importância de novos métodos de ensino que promovam a inclusão e o desenvolvimento pleno desses estudantes. A escolarização de crianças autistas é um desafio complexo que requer abordagens educacionais diferenciadas e adaptadas às necessidades individuais de cada criança. Neste tópico, será discutido os desafios enfrentados no processo de escolarização da criança neurodivergente e a importância de novos métodos de ensino que promovam a inclusão e o desenvolvimento pleno desses estudantes. A criança autista pode enfrentar uma série de desafios no ambiente escolar, tanto sociais quanto acadêmicos. A dificuldade de interação social, a sensibilidade sensorial e a rigidez comportamental são características que demandam uma abordagem educacional especializada e inclusiva por parte do Estado, para que essas adversidades sejam amenizadas é necessário que cada escola, crie internamente novos métodos de ensinar, adaptando e ao passo que a medicina encontra novos meios de melhora para o ensino e desenvolvimento dessas crianças a escola desenvolva um meio eficaz de acompanhar essas atualizações e ponha em pratica cada uma delas, como diz escritor e psicólogo Gustavo Teixeira, em seu livro “Manual do Autismo ” (2016), é fundamental compreender as particularidades do espectro autista para melhor atender às necessidades dessas crianças, pois além das dificuldades de interação, nos deparamos com outra dificuldade, a falta de capacitação dos profissionais da educação em relação ao autismo, segundo (CAMARGO 2020), a falta de preparo dos educadores para lidar com as necessidades específicas da criança autista é uma das principais dificuldades enfrentadas, pois se torna impossível um professor que não tem especialização conseguir passar para o aluno portador do TEA, o tratamento e a educação necessária para seu melhor desenvolvimento. 4 INCLUSÃO A CRIANÇA AUTISTA: FORA E DENTRO DA SALA DE AULA (GOMES 2019), destaca a importância de promover uma formação continuada dos educadores, visando à compreensão das características do autismo e a adoção de estratégias pedagógicas adequadas, pois a falta de estrutura adequada nas escolas, como recursos pedagógicos adaptados e profissionais especializados, pode comprometer o processo de inclusão, a escritora (Oliveira, 2020), destaca a importância de uma abordagem individualizada, que considere as características e necessidades específicas de cada criança autista pois só com essa qualificação e preparo não só dos educadores, mas de todo o corpo docente A criança autista pode enfrentar uma série de desafios no ambiente escolar, tanto sociais quanto acadêmicos. A dificuldade de interação social, a sensibilidade sensorial e a rigidez comportamental são características que demandam uma abordagem educacional especializada e inclusiva por parte do Estado, para que essas adversidades sejam amenizadas é necessário que cada escola, crie internamente novos métodos de ensinar, adaptando e ao passo que a medicina encontra novos meios de melhora para o ensino e desenvolvimento dessas crianças a escola desenvolva um meio eficaz de acompanhar essas atualizações e ponha em pratica cada uma delas, como diz escritor e psicólogo Gustavo Teixeira, em seu livro “Manual do Autismo ” (2016), é fundamental compreender as particularidades do espectro autista para melhor atender às necessidades dessas crianças, pois além das dificuldades de interação, nos deparamos com outra dificuldade, a falta de capacitação dos profissionais da educação em relação ao autismo, segundo (CAMARGO 2020), a falta de preparo dos educadores para lidar com as necessidades específicas da criança autista é uma das principais dificuldades enfrentadas, pois se torna impossível um professor que não tem especialização conseguir passar para o aluno portador do TEA, o tratamento e a educação necessária para seu melhor desenvolvimento. (GOMES 2019), destaca a importância de promover uma formação continuada dos educadores, visando à compreensão das características do autismo e a adoção de estratégias pedagógicas adequadas, pois a falta de estrutura adequada nas escolas, como recursos pedagógicos adaptados e profissionais especializados, pode comprometer o processo de inclusão, a escritora (Oliveira, 2020), destaca a importância de uma abordagem individualizada, que considere as características e necessidades específicas de cada criança autista pois só com essa qualificação e preparo não só dos educadores, mas de todo o corpo docente da escola, o Estado poderá oferecer o que já é previsto em lei para todas crianças portadoras de alguma deficiência, em especial, as autistas. 4 INCLUSÃO A CRIANÇA AUTISTA: FORA E DENTRO DA SALA DE AULA Segundo Félix Díaz, organizador do livro “Educação Inclusiva, deficiência e contexto social” (2009), esse método utiliza estratégias visuais, organização do ambiente e sequenciamento de tarefas para facilitar a aprendizagem e a compreensão da criança autista, visando isso que os grandes cientistas e professores especializados na área, estão desenvolvendo métodos de ensino para adequar à educação escolar para TEACCH (Treatment and Education of Autistic and Related Communication- Handicapped Children) é um exemplo de abordagem que valoriza o aprendizado visual e a organização do ambiente, proporcionando maior clareza e estrutura ao aluno autista. Segundo a pesquisadora Anne Le Couteur, em seu estudo “Education and Autism: Putting the Pieces Together” (2021), o Método TEACCH tem demonstrado resultados positivos na promoção do desenvolvimento acadêmico e social das crianças autistas, fazendo com que a falsa crença que muitos possuem de que pessoas portadoras de alguma deficiência não sejam capazes de desenvolver plenamente seus potenciais, outro método que tem ganhado destaque é o ABA (Applied Behavior Analysis), que se baseia em técnicas de reforço e modelagem comportamental. A pedagoga Dailma da Silva Medeiros, em sua pesquisa, As contribuições da análise do comportamento (ABA) para a aprendizagem de pessoas com autismo, destaca a importância do ABA como uma ferramenta eficaz para o ensino de habilidades sociais, comunicativas e acadêmicas para crianças autistas. O processo de escolarização da criança autista apresenta desafios que exigem abordagens pedagógicas diferenciadas. A compreensão das necessidades individuais, a adoção de novos métodos de ensino e o respeito à legislação vigente são elementos cruciais para promover uma educação inclusiva e de qualidade, para enfrentar os desafios da escolarização da criança autista, novos métodos de ensino têm sido desenvolvidos como a abordagem do ensino estruturado, que tem se mostrado eficaz. Segundo Félix Díaz, organizador do livro “Educação Inclusiva, deficiência e contexto social” (2009), esse método utiliza estratégias visuais, organização do ambiente e sequenciamento de tarefas para facilitar a aprendizagem e a compreensão da criança autista, visando isso que os grandes cientistas e professores especializados na área, estão desenvolvendo métodos de ensino para adequar à educação escolar para TEACCH (Treatment and Education of Autistic and Related Communication- Handicapped Children) é um exemplo de abordagem que valoriza o aprendizado visual e a organização do ambiente, proporcionando maior clareza e estrutura ao aluno autista. 4 INCLUSÃO A CRIANÇA AUTISTA: FORA E DENTRO DA SALA DE AULA A participação e envolvimento da família são fundamentais para promover o acesso à educação das crianças autistas. O diálogo entre escola e família, o compartilhamento de informações e a construção de um trabalho conjunto são essenciais para o sucesso da inclusão educacional, a família deve ser incentivada a participar ativamente do processo educacional da criança, incluindo a participação em reuniões escolares, eventos, ajuda nas atividades escolares, e o ponto principal, a escola deve envolver a família no planejamento e implementação de estratégias de ensino específicas para atender às necessidades individuais da criança, ao estabelecer uma parceria sólida entre a escola e a família, é possível criar um ambiente educacional inclusivo, que valorize e atenda às necessidades individuais da criança autista. Essa colaboração fortalece a equipe de suporte da criança e maximiza seu potencial de aprendizado e crescimento, fazendo com que os desafios da educação e o novo método de ensinar, seja feito de forma mais fácil e rápida. Diante dos desafios apresentados, novos métodos de ensino têm sido desenvolvidos com o objetivo de promover a inclusão da criança autista no ambiente escolar. A abordagem do ensino estruturado, por exemplo, busca oferecer um ambiente organizado, com rotinas previsíveis e instruções claras, facilitando a compreensão e a adaptação da criança autista. Ronara Terra, em seu artigo, “A escola, o autismo e a inclusão” destaca a importância do uso de tecnologias assistidas, como aplicativos e jogos educacionais, que podem auxiliar no desenvolvimento de habilidades cognitivas, sociais e de comunicação das crianças autistas, o doutrinador Paulo Freire, em sua obra “Pedagogia da Autonomia” (1996), também retrata que a inclusão escolar da criança autista exige a adoção de práticas pedagógicas que considerem a diversidade e valorizem as potencialidades individuais, presando pelo respeito à individualidade, à autonomia e às características singulares de cada criança autista é essencial para garantir uma educação inclusiva e de qualidade. O processo de escolarização da criança autista apresenta desafios que exigem abordagens pedagógicas diferenciadas. A compreensão das necessidades individuais, a adoção de novos métodos de ensino e o respeito à legislação vigente são elementos cruciais para promover uma educação inclusiva e de qualidade, para enfrentar os desafios da escolarização da criança autista, novos métodos de ensino têm sido desenvolvidos como a abordagem do ensino estruturado, que tem se mostrado eficaz. 4 INCLUSÃO A CRIANÇA AUTISTA: FORA E DENTRO DA SALA DE AULA Segundo a pesquisadora Anne Le Couteur, em seu estudo “Education and Autism: Putting the Pieces Together” (2021), o Método TEACCH tem demonstrado resultados positivos na promoção do desenvolvimento acadêmico e social das crianças autistas, fazendo com que a falsa crença que muitos possuem de que pessoas portadoras de alguma deficiência não sejam capazes de desenvolver plenamente seus potenciais, outro método que tem ganhado destaque é o ABA (Applied Behavior Analysis), que se baseia em técnicas de reforço e modelagem comportamental. A pedagoga Dailma da Silva Medeiros, em sua pesquisa, As contribuições da análise do comportamento (ABA) para a aprendizagem de pessoas com autismo, destaca a importância do ABA como uma ferramenta eficaz para o ensino de habilidades sociais, comunicativas e acadêmicas para crianças autistas. Embora, a medicina e aos grandes nomes da educação no Brasil e no mundo já Embora, a medicina e aos grandes nomes da educação no Brasil e no mundo já tenham desenvolvido bons métodos para ajudar no processo de escolarização das crianças portadoras do Espectro, essa realidade ainda se torna distante para as crianças que necessitam do ensino público para estudar, principalmente quando nos deparamos com os grandes índices de analfabetismo (IBGE, 2020) no Brasil, somos inseridos num cenário de descaso com o ensino público, o que implica diretamente com as condições das crianças autistas em âmbito escolar, tornando mais difícil ainda o acesso a uma educação de qualidade, a falta de recursos e estrutura adequados, a falta de profissionais capacitados e a falta de conscientização sobre o autismo dificultam o acesso à educação para crianças autistas, o que acarreta grandes problemas no desenvolvimento e na vida inteira dos portadores do Espectro, tendo em vista que a educação é a base e desempenha um papel fundamental no desenvolvimento humano, na construção de sociedades justas e na promoção do progresso social, por isso é fundamental que haja investimentos contínuos em políticas públicas, formação de profissionais e adaptação das escolas, a fim de garantir um acesso efetivo, continuo e de qualidade à educação para todas as crianças autistas, capacitando cada um deles no novo modo de educar que o Espectro Autista demanda, além de fomentar a conscientização sobre o autismo em toda a comunidade escolar. Somente assim poderemos superar os desafios e garantir o acesso equitativo e inclusivo à educação para as crianças autistas. 4.1.1 A dificuldade na assistência escolar e inclusiva para a criança autista A inclusão educacional de crianças autistas tem sido um tema cada vez mais debatido na sociedade contemporânea. A busca por uma educação inclusiva, que atenda às necessidades específicas dessas crianças, tem encontrado diversos empecilhos ao longo do caminho. Sendo necessário abordar a dificuldade nessa assistência escolar e inclusiva para a criança autista, destacando a importância de uma abordagem integrada e respeitosa às suas necessidades individuais. Grande parte dos profissionais educacionais se sentem despreparados em lidar com a inclusão escolar e os desafios que vem junto a ela, possuindo dificuldade em adaptar os métodos de ensino ao estilo de aprendizado de cada aluno. No caso de crianças com TEA, por possuírem um desenvolvimento atípico com padrões de comportamento restritivos e estereotipados seu processo de aprendizagem é diferente, requerendo adaptações na forma de tratamento, nas atividades e no modo de ensinar, respeitando seu ritmo e suas necessidades. Quando se debate sobre o direito à educação da criança com autismo, não é apenas sobre a possibilidade de ingressar em uma sala de aula, ter a chance de poder se matricular no ensino regular é apenas o primeiro passo, mas sim sobre a permanência dessa criança no ambiente escolar. Sendo a escola a responsável por manter a integridade física e mental de seus alunos, não precisando necessariamente que eles estejam dentro de uma sala de aula para que ocorra uma intervenção, pois enquanto essa criança estiver dentro do ambiente acadêmico ela será responsabilidade da instituição de ensino. Em Março de 2023, ocorreu nas dependências de uma Escola Estadual do Rio de Janeiro o caso de bullying contra uma menina autista de 11 anos. O vídeo do momento da agressão circulou a internet e em inúmeros portais de notícia e nele mostrava a garota chorando enquanto era zombada e tinha seu cabelo puxado pelas agressoras. O Autor Carlos Roberto Gonçalves defende a ideia de que a responsabilidade pelas situações que ocorrem dentro do ambiente escolar deve recair sobre os funcionários daquele local, sendo essa vigilância ainda mais imprescindível em caso de alunos com necessidades especiais, que acabam se tornando mais vulneráveis á agressões verbais e físicas. “A ideia de vigilância é mais ampla do que a de educação, devendo entender-se que estas pessoas respondem pelos atos dos alunos e aprendizes, durante o tempo em que eles exercem vigilância e autoridade. 4.1.1 A dificuldade na assistência escolar e inclusiva para a criança autista Os danos por que respondem são, ordinariamente, os sofridos por terceiros, o que quer dizer que os danos sofridos pelo próprio aluno, ou aprendiz não possam acarretar a responsabilidade do mestre ou diretor do estabelecimento.” “A ideia de vigilância é mais ampla do que a de educação, devendo entender-se que estas pessoas respondem pelos atos dos alunos e aprendizes, durante o tempo em que eles exercem vigilância e autoridade. Os danos por que respondem são, ordinariamente, os sofridos por terceiros, o que quer dizer que os danos sofridos pelo próprio aluno, ou aprendiz não possam acarretar a responsabilidade do mestre ou diretor do estabelecimento.” As escolas que devem se adaptar, ou readaptar, ao aluno, estando preparadas para lidar com qualquer tipo de criança que venha a se matricular no local, devendo amparar e proteger todos aqueles presentes na instituição, não permitindo que situações como essa, dentro do ambiente escolar, se repitam. Quando se fala sobre incluir um aluno autista no ensino regular, não é sobre alterar o conteúdo, uma grade curricular ou a rotina da escola para que ele se encaixe, mas sim aprender a se comunicar, ensinar de uma forma mais abrangente, que consiga englobar todos os receptores da mensagem, é ter um zelo maior por esse menor que é mais suscetível a sofrer algum tipo de violência. Para que haja a inclusão na forma garantida pela lei é necessário um olhar crítico em relação às praticas pedagógicas, ao suporte escolar disponível, ao ambiente como um todo e em especial aos docentes que acompanham essas crianças na maior parte do tempo, devendo estes estar qualificados, possuindo o conhecimento e os recursos necessários para lidar com alunos possuidores do espectro, sendo essa capacitação responsabilidade do Estado, sendo sua obrigação incentivar e custear no caso das escolas públicas, para que essas crianças tenham um ensino paciente, criativo, estimulante e acolhedor. 5 CONCLUSÃO É possível notar que na última década ocorreu certa mobilização a cerca de diversos assuntos sociais, na tentativa de levar o olhar da sociedade para minorias e suas necessidades ignoradas, sendo uma das pautas recorrentes a inclusão no âmbito escolar. A educação deve ser desenvolvida e incentivada desde a primeira infância quando o ser humano se encontra nas suas primeiras fases de desenvolvimento. Campos (1986) fala que educação infantil se configura como uma das áreas educacionais que mais retribui à sociedade os recursos nela investidos, contribuindo para o funcionamento e desenvolvimento de toda uma sociedade e do próprio individuo, sendo assim, indispensável principalmente ás crianças com necessidades especiais por oferece-las uma maior autonomia e um treinamento para lidar melhor com suas próprias limitações. A educação na sua forma inclusiva não deve ser vista apenas pela possibilidade de uma criança com TEA frequentar o ensino regular, mas que todo aquele ambiente que a envolve, seja corpo docente, demais funcionários ou os materiais e as formas de ensino sejam adaptados da melhor forma para que suas necessidades sejam respeitadas. Diante de tudo que foi exposto, é notável a existência de muitos percalços a serem percorridos durante o processo de escolarização da criança autista, exigindo um trabalho coletivo de família, profissionais de ensino, da área da saúde e da sociedade como um todo por meio de ações integradoras que envolvam essas crianças, respeitando e compreendendo suas necessidades especificas. A fiscalização de leis que promovem essa inclusão é essencial para a garantia na qualidade de ensino e a promoção do plano desenvolvimento da criança com espectro autista. REFERÊNCIAS ALMEIDA, Hévelin Caroline Melo de. A eficácia da Lei nº 12.764/12, que resguarda os direitos dos portadores do Transtorno do Espectro Autista, no que tange a inclusão em escolas de ensino regular. Revista ÂmbitoJurídico, 1 de outubro de 2020. ANDRADE, E. F. D.. (2023). MARCHA E CONTRAMARCHA NA GARANTIA DO DIREITO À EDUCAÇÃO BÁSICA NO BRASIL. Educação Em Revista, 39, e35414. https://doi.org/10.1590/0102-469835414 American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorders. 5th ed. Washington (DC): American Psychiatric Association; 2013. BRAGA, Mariana; SCHUMACHER, Aluisio. Direito e inclusão da pessoa com deficiência: uma análise orientada pela teoria de reconhecimento social de Axel Honneth. Scielo, 2013. BRAGA, M.C.B. A Interação professor aluno em classe inclusiva: um estudo exploratório com criança autista. Dissertação (Mestrado em Educação) – Universidade Estadual Paulista, São Paulo, 2002 CUNHA, E. Autismo e inclusão: psicopedagogia práticas educativas na escola e na família.5ª ed. RJ: Wak Ed., 2014. CUNHA, E. Autismo e inclusão: psicopedagogia práticas educativas na escola e na família.5ª ed. RJ: Wak Ed., 2014. DOVANAN, Jhon; ZUCKER, Caren. Outra Sintonia: A história do autismo; 1° edição; Rio de Janeiro: Companhia das letras. 12 de junho 2017 Educação inclusiva, deficiência e contexto social: questões contemporâneas/ Féliz Díaz, Miguel Bordas, Nelma Galvão, Theresinha Miranda, organizadores; autores, Elias Souza dos Santos… [et al.]. – Salvador: EDUFBA, 2009. 354p. MEDEIROS, Dailma da Silva; AS CONTRIBUIÇÕES DA ANÁLISE DO COMPORTAMENTO (ABA) PARA A APRENDIZAGEM DE PESSOAS COM AUTISMO: UMA REVISÃO DA LITERATURA.2021 BRASIL. ConstituiçãodaRepúblicaFederativadoBrasil, Brasília, 1988. Brasil. Decreto n. 6.949, de 25 de agosto de 2009. Promulga a Convenção Internacional sobre os Direitos das Pessoas com Deficiência e seu Protocolo Facultativo, assinados em Nova York, em 30 de março de 2007. DiárioOficialdaUnião2009; 26 ago. BRASIL. Lei nº 12.764, de 27 de dezembro de 2012. Institui a Política Nacional de Proteção dos Direitos da Pessoa com Transtorno do Espectro Autista; e altera o § 3º do art. 98 da Lei nº 8.112, de 11 de dezembro de 1990. Brasil. Lei 13.146, de 6 de julho de 2015. Institui a Lei Brasileira de Inclusão das Pessoas com Deficiência (Estatuto da Pessoa com Deficiência). DiárioOficialdaUnião2015; 7 jul. BRASIL. Lei nº 13.977, de 8 de janeiro de 2021. Altera a Lei nº 12.764, de 27 de dezembro de 2012 (Lei Berenice Piana), e a Lei nº 9.265, de 12 de fevereiro de 1996, para instituir a Carteira de Identificação da Pessoa com Transtorno do Espectro Autista (Ciptea), e dá outras providências. BIALER, Marina Martins; JÚNIOR, Nelson Ernesto Coelho; APRESENTAÇAO PERCEPTIVA NO AUTISMO: ESTUDO DA AUTOBIOGRAFIA DE DONNA WILLIAMS 2018. CAMARGO, Siglia. Competência social, inclusão escolar e autismo: um estudo de caso comparativo CAMARGO, S. P. H., SILVA, G. L. D., CRESPO, R. O., OLIVEIRA, C. R. D., & MAGALHÃES, S. L.. (2020). DESAFIOS NO PROCESSO DE ESCOLARIZAÇÃO DE CRIANÇAS COM AUTISMO NO CONTEXTO INCLUSIVO: DIRETRIZES PARA FORMAÇÃO CONTINUADA NA PERSPECTIVA DOS PROFESSORES. Educação Em Revista, 36, e214220. https://doi.org/10.1590/0102-4698214220 MEDEIROS, Dailma da Silva; AS CONTRIBUIÇÕES DA ANÁLISE DO COMPORTAMENTO (ABA) PARA A APRENDIZAGEM DE PESSOAS COM AUTISMO: UMA REVISÃO DA LITERATURA.2021 Estatuto da criança e do adolescente (1990). Estatuto da criança e do adolescente: Lei n. 8.069, de 13 de julho de 1990, Lei n. 8.242, de 12 de outubro de 1990. 3. Ed. Brasília: Câmara dos Deputados, Coordenação de Publicações, 2012 EDUCAR EM REVISTA. Curitiba, PR: Ed. UFPR, n.23, 2004 Freire, Paulo. Pedagogia da autonomia: saberes necessários à prática educativa / Paulo Freire. – São Paulo: Paz e Terra, 1996. – (Coleção Leitura) FAUSTINO, C.R O aluno com deficiência e o Direito educacional na perspectiva inclusiva: Um estudo sobre o atendimento educacional especializado em uma escola de São Paulo. Gomes, C. G. S., & Mendes, E. G. Escolarização inclusiva de alunos com autismo na rede municipal de ensino de Belo Horizonte. Revista Brasileira De Educação Especial (2010). GOMES, Fernanda Bohnert. INCLUSÃO DE ALUNOS COM TRANSTORNO DO ESPECTRO AUTISTA (TEA) NO ENSINO DE CIÊNCIAS NAS ESCOLAS PÚBLICAS MUNICIPAIS DE DOM PEDRITO/RS. (2019) História Social™ e Autismo: uma Revisão de Literatura- LAZZARINI, Fernanda; ELIAS, Nassim. História Social™ e Autismo: uma Revisão de Literatura- LAZZARINI, Fernanda; ELIAS, Nassim. Inclusão de alunos autistas no ensino regular: concepções e práticas pedagógicas de professores regentes – MARTINS, M.R.R LAZZARINI, Fernanda; ELIAS, Nassim. História Social™ e Autismo: uma Revisão de Literatura Lei de Diretrizes e Bases da Educação Nacional – LDBEN (Leinº9.394/96, de 20 de dezembro de 1996) LIMA, Stefanie, Escolarização de Alunos com Autismo MACEDO, Eunice; MAFRA, Jason; SANTOS, Régia. Autismo na escola: da construção social estigmatizante ao reconhecimento como condição humana. Scielo, 2022. Disponível em: https://doi.org/10.24109/2176 6681.rbep.103i264.5108 Acesso em: outubro de 2022 MELO, S.C. Autismo e educação: a dialética da inclusão. Dissertação (Mestrado em Educação) -Universidade Estadual do Rio de Janeiro, Rio de Janeiro, 2004 RevistaFT A RevistaFT é uma Revista Científica Eletrônica Multidisciplinar Indexada de Alto Impacto e Qualis “B2” em 2023. Periodicidade mensal e de acesso livre. Leia gratuitamente todos os artigos e publique o seu também clicando aqui. MELO, S.C. Autismo e educação: a dialética da inclusão. Dissertação (Mestrado em Educação) -Universidade Estadual do Rio de Janeiro, Rio de Janeiro, 2004 MARTINS, M.R.R. Inclusão de alunos autistas no ensino regular: concepções e práticas pedagógicas de professores regentes. Dissertação (Mestrado em Psicologia) – Universidade Católica de Brasília, Brasília, 2007. Neves, D. T., & Giorgi, C. A. G. D.. (2022). De quem é a responsabilidade? Uma análise das perspectivas para a lei de responsabilidade educacional prevista no Plano Nacional de Educação. Ensaio: Avaliação E Políticas Públicas Em Educação, 30(114), 11–31. https://doi.org/10.1590/S0104-40362021002902469 Organização das Nações Unidas – ONU. (2006). Convenção sobre os direitos das pessoas com deficiência New York. OLIVEIRA, Francisco Lindoval: Autismo e inclusão escolar: os desafios da inclusão do aluno autista., 2020. SOCIEDADE BRASILEIRA DE PEDIATRIA. Triagem precoce para Autismo / Transtorno do Espectro Autista. Porto Alegre: SBP, 2017. SOCIEDADE BRASILEIRA DE FONOAUDIOLOGIA (SBFa). Parecer: Métodos Clínicos e Diretrizes Terapêuticas Ampliadas no Tratamento de Indivíduos com Transtorno do Espectro Do Autismo. São Paulo: SBFa, 2019. Teixeira, Gustavo Manual do autismo [recurso eletrônico] / Gustavo Teixeira. – 1. ed. – Rio de Janeiro : Best Seller, 2016 TERRA, Ronara de Oliveira: A escola, o autismo e a inclusão: uma revisão bibliográfica; 2017. UNESCO. Declaração de Salamanca. Sobre Princípios, Políticas e Práticas na Área das Necessidades Educativas Especiais. Disponível em: UNESCO. Declaração de Salamanca. Sobre Princípios, Políticas e Práticas na Área das Necessidades Educativas Especiais. Disponível em: http://portal.mec.gov.br/seesp/arquivos/pdf/salamanca.pdf Acesso em: outubro de 2022 Trezzi, C.. (2022). O acesso universal à Educação no Brasil: uma questão de justiça social. Ensaio: Avaliação E Políticas Públicas Em Educação, 30(117), 942–959. https://doi.org/10.1590/S0104-40362022003003552 . Dados da Educação Especial no Brasil Ministério da Educação, Brasília, 2007. Disponível em: http://portal.mec.gov.br/seesp/arquivos/pdf/brasil.pdf. Acesso em: outubro de 2022 1 Acadêmica do Curso de Bacharelado em Direito do Centro Universitário Santo Agostinho – UNIFSA . E-mail: dudaevelin@hotmail.com 2 Acadêmica do Curso de Bacharelado em Direito do Centro Universitário Santo Agostinho – UNIFSA . E-mail: imilysilva9@gmail.com 3 Professora do Curso de Bacharelado em Direito do Centro Universitário Santo Agostinho – UNIFSA. Doutora em Ciências Criminais pela Escola de Direito da PUC do Rio Grande do Sul (PUCRS). E-mail: wirnaalves@unifsa.com.br ← Post anterior ← Post anterior Conselho Editorial Editores Fundadores: Dr. Oston de Lacerda Mendes. Dr. João Marcelo Gigliotti. Editor Científico: Dr. Oston de Lacerda Mendes Orientadoras: Dra. Hevellyn Andrade Monteiro Dra. Chimene Kuhn Nobre Dra. Edna Cristina Dra. Tais Santos Rosa Revisores: Lista atualizada periodicamente em revistaft.com.br/expediente Venha fazer parte de nosso time de revisores também! Copyright © Editora Oston Ltda. 1996 - 2023 Rua José Linhares, 134 - Leblon | Rio de Janeiro-RJ | Brasil Conselho Editorial Contato Queremos te ouvir. WhatsApp: 11 98597-3405 e-Mail: contato@revistaft.com.br ISSN: 1678-0817 CNPJ: 48.728.404/0001-22 CAPES – Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), fundação do Ministério da Educação (MEC), desempenha papel fundamental na expansão e consolidação da pós-graduação stricto sensu (mestrado e doutorado) em todos os estados da Federação. expansão e consolidação da pós-graduação stricto sensu (mestrado e doutorado) em todos os estados da Federação. Conselho Editorial Copyright © Editora Oston Ltda. 1996 - 2023 Rua José Linhares, 134 - Leblon | Rio de Janeiro-RJ | Brasil
https://openalex.org/W4378907308
https://www.e3s-conferences.org/articles/e3sconf/pdf/2023/26/e3sconf_uesf2023_01074.pdf
English
null
Influence of process parameters on magnetic abrasive machining efficiency
E3S web of conferences
2,023
cc-by
2,242
Influence of process parameters on magnetic abrasive machining efficiency P.I. Romanov1* 1Peter the Great St. Petersburg Polytechnic University, Polytechnicheskaya, 29, 195251, St. Petersburg, Russia P.I. Romanov1* P.I. Romanov1* 1Peter the Great St. Petersburg Polytechnic University, Polytechnicheskaya, 29, 195251, St. Petersburg, Russia 1Peter the Great St. Petersburg Polytechnic University, Polytechnicheskaya, 29, 195251, St. Petersburg, Russia Abstract. The development of modern mechanical engineering forces us to constantly improve the technology of processing materials and look for ways not only to improve the quality of products but also to increase the efficiency of the processing process itself. This article discusses an alternative method to finish the product, which can replace the traditional method of grinding, magnetic abrasive treatment. This is a modern solution that allows you to obtain high surface quality of the processed product with minimal defects in their surface layer, and reduce the number of technological operations, which has a beneficial effect on the efficiency of the processing process. The efficiency of the magnetic abrasive machining process, which was determined by the amount of specific removal of the material, is influenced by many parameters, some of which were studied in this article: electromagnetic induction and processing time. The success of the application of magnetic abrasive treatment can also be judged by the topography of the surface of the product made using an electron microscope. The data obtained in the course of the study demonstrated the importance of individual input parameters of the magnetic abrasive treatment process in the formation of the surface layer of the product. According to the results of experimental studies, the dependences of the varied parameters were determined and recommendations were formed to improve the efficiency of the use of the magnetic abrasive process. * Corresponding author: romanov_pi@spbstu.ru E3S Web of Conferences 389, 01074 (2023) UESF-2023 E3S Web of Conferences 389, 01074 (2023) UESF-2023 https://doi.org/10.1051/e3sconf/202338901074 © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). 1 Introduction The development of modern mechanical engineering makes it necessary to constantly improve the technological methods of machining parts to ensure the geometric parameters of the machined product and achieve high surface quality. Traditional methods of finishing products that are currently used are based on the grinding process, which has a negative impact on the surface layer of the workpiece. The presence of high temperatures in the machining area during grinding leads to burn marks and grinding cracks on the surface of the workpiece, which subsequently affects the strength and structural changes. Heating the workpiece during grinding causes thermal deformation, which directly affects its geometrical accuracy, wear resistance and serviceability. Thus, ensuring quality machining of precision products is a complex technological task that requires modern methods of solution. 1.1 The study of the MAM process effectiveness The use of this method is a promising trend in mechanical engineering technology that can not only improve the performance properties and quality characteristics of the product, but can also significantly increase the efficiency of the machining process by reducing process operations. The use of this method is a promising direction in mechanical engineering technology, which can not only increase the operational properties and quality characteristics of the product but also significantly improve the efficiency of machining parts by reducing technological operations. In this paper, the efficiency of magnetic abrasive machining is investigated using the example of ceramic cutting inserts of the BOK-60. The specific material removal was analysed and the dependence of the process parameters on the MAM on the efficiency of the machining process was revealed. * Corresponding author: romanov_pi@spbstu.ru © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). E3S Web of Conferences 389, 01074 (2023) UESF-2023 https://doi.org/10.1051/e3sconf/202338901074 One possible way to solve the existing problem is to change the product's machining technology and replace the finishing operation with finishing turning using the magnetic abrasive machining (MAM) method [1,2]. This is a state-of-the-art machining approach that allows a high-quality machined product to be produced with minimal surface thermal defects. This method is widely used in the machining of linear and cylindrical parts made of aluminium, brass, steel and other materials [5,7,8]. 2 Research methodology Magnetic abrasive machining allows dimensionless machining to be realised. The material is removed from the surface of the workpiece by a kind of cutting tool made of magnetic abrasive powder, which is formed by the action of a magnetic field [3]. Magnetic abrasive tool (MAT) is characterised by a high degree of elasticity. The depth of penetration of the grit into the surface of the workpiece is the result of a settled equilibrium between the pressing forces pressing the grit against the workpiece and the resistance forces of the workpiece material against the penetration of the grit. The performance of magnetic blasting is determined by the mass of material removed, which is calculated by weighing the workpiece before and after blasting. The method of geometrically calculating the volume of workpiece material has a high degree of measurement error due to the small thickness of the removed layer and the temperature deformations of the workpiece. In order to assess the effectiveness of the process MAM the specific material removal value was applied q, g/cm2 с area units (1). 𝑞= 𝑄 𝑆 (1) 𝑞= 𝑄 𝑆 where Q is the mass of the material to be removed, determined by the difference in masses to 𝑄before and after 𝑄after MAM; where Q is the mass of the material to be removed, determined by the difference in masses to 𝑄before and after 𝑄after MAM; masses to 𝑄before and after 𝑄after MAM; 2 masses to 𝑄before and after 𝑄after MAM; 2 S is the surface area to be treated, cm2. Based on a literature review [1,4,6] and the capabilities of a dedicated magnetic blasting facility, two varying parameters were selected that have the greatest influence on the material removal process from the workpiece surface. These parameters include machining time and electromagnetic induction value (Table 1). 2 2 E3S Web of Conferences 389, 01074 (2023) UESF-2023 https://doi.org/10.1051/e3sconf/202338901074 Table 1. Varying parameters MAM. Varying parameters Level -2 -1 0 +1 +2 1. Magnitude of electromagnetic induction (Tesla) 0.35 0.5 0.65 0.8 0.95 2. Processing time (min) 4 6 8 10 12 Table 1. Varying parameters MAM. Table 1. Varying parameters MAM. Other process parameters such as workpiece speed, feed rate, composition MAT, the abrasive powder fraction, the amount of ferro-abrasive mass in the workspace, the clearance between the pole tips remained constant (Table 2). Table 2. Static parameters MAM. Table 2. Static parameters MAM. Fig. 1. General view of the experimental setup. 2 Research methodology Static parameters Rotation speed (min-1) 400 Feed rate (mm/min) 200 Ceramic plate Brand BOK-60 (12 mm х 12 mm х 4 mm) GOST 25003-81 Type of ferromagnetic abrasive powder (FAP) TiC-Fe Chemical composition FAP Ti-48.9%, 𝐶𝑡𝑖𝑒𝑑- 11.6%, 𝐶𝑓𝑟𝑒𝑒- 0.8%, Fe-38.9% Characteristics of the microstructure Flake carbide Granularity of the powder  160…315 μm Volume FAP 125 g Clearance between workpiece and pole tips 3 mm Lubricant-coolant ECOCOOL GLOBAL 10 UK Ceramic plates are clamped in place using a specially designed support fixture BOK-60 and feeding them into the working space. The tool is fed into the magnetic-abrasive powder at a predetermined speed of the machine spindle. The surface area of the ceramic plate in direct contact with the magnetic-abrasive tool S = 1.7 cm2. A diagram of the installation of the unit and the workpiece is shown in figure 1. Fig. 1. General view of the experimental setup. Fig. 1. General view of the experimental setup. 3 3 E3S Web of Conferences 389, 01074 (2023) UESF-2023 https://doi.org/10.1051/e3sconf/202338901074 3 Research results To investigate the effect of electromagnetic induction and time MAM on the specific material removal rate were plotted as shown in Figure 2. a) b) g. 2. Dependence of specific material removal q at MAM on: a) magnitude of electromagnetic duction, b) processing time a) b) D d f ifi t i l l t MAM ) it d f l t a) b) Fig. 2. Dependence of specific material removal q at MAM on: a) magnitude of electromagnetic induction, b) processing time 3.1 Effect of electromagnetic induction value on specific material removal q Figure 2 (a) shows that the dependence of the specific material removal q on the value of electromagnetic induction B is linear. The amount of material removal increases throughout the range of the varying parameter from 0.35 to 0.95 T. Minimal material removal 0.14 g/cm2 is observed at an electromagnetic induction value of 0.35 T, and the maximum 0.78 g/cm2 at 0.95 T. This is because with increasing electromagnetic induction the stiffness increases MAT, The increase in tool hardness also increases the temperature in the cutting zone, which has a negative effect on the quality of the machined surface. An increase in tool 4 E3S Web of Conferences 389, 01074 (2023) UESF-2023 https://doi.org/10.1051/e3sconf/202338901074 hardness also leads to an increase in temperature in the cutting zone, which has a negative effect on the quality of the machined surface. In this study, the range of electromagnetic induction has minimal negative impact on surface quality. Increasing this range can lead to material chipping from the workpiece surface and the emergence of stress concentrators, which have a negative effect on the wear resistance of the product, which requires additional research. 3.2 Influence of machining time on specific material removal q Figure 2 (b) shows that the dependence of the specific material removal q on the processing time t is linear, but less steep than the dependence on the electromagnetic induction. In the range from 2 to 12 minutes, the value of specific material removal varies from 0.05 till 0.4 g/cm2 respectively. This can be explained by the increased number of micro-cutting cycles of each of the grains individually, which are mixed throughout MAM. It can be assumed that over time MAT loses its original cutting capacity and the specific material removal q will then be relatively small or no longer increase at all. 4 Surface topography In assessing the effectiveness of the process MAM not only the specific material removal q was investigated, but also the quality of the machined surface of the workpiece. The results were evaluated with a scanning electron microscope MarVision MM320 (Fidure 3). b) a) a) b) c) d) Fig. 3. Surface microstructure of ceramic plates BOK-60 before machining: a - front surface, c - rear surface; after MAM at B = 0.65 T, t = 8 min: b - front surface, d - rear surface. Based on optical inspection it can be concluded that the process is effective MAM which b) a) b) d) ) c) d) c) Fig. 3. Surface microstructure of ceramic plates BOK-60 before machining: a - front surface, c - rear surface; after MAM at B = 0.65 T, t = 8 min: b - front surface, d - rear surface. Based on optical inspection, it can be concluded that the process is effective MAM, which can replace the traditional grinding method. The machining process smoothes out the cutting edge of the surface and forms a surface without visible defects: cracks, chips and burrs. However, it is possible to observe residual inheritance which cannot be levelled due to the small thickness of the layer removed from the surface of the workpiece. 5 E3S Web of Conferences 389, 01074 (2023) UESF-2023 https://doi.org/10.1051/e3sconf/202338901074 5 Conclusion Magnetic blasting is an effective surface treatment process which can be used as an alternative to grinding. The following conclusions can be drawn from the analysis and experimental studies: p 1. Selected varying parameters MAM (magnitude of electromagnetic induction, machining time) have a significant influence on the efficiency of the machining process; 2. Electromagnetic induction is the most important factor affecting the specific removal of material from the workpiece surface; 3. In the range of electromagnetic induction values from 0.35 to 0.95 T, there is a linear increase in material removal from the product surface; 4. The greatest amount of material removal is observed at an electromagnetic induction of 0.95 T and a treatment time of 12 min; 5. By varying the electromagnetic induction parameters and processing time, it is possible to achieve material removal rates of 0.05 till 0.78 g/cm2. References 1. V.V. Maksarov, М.A. Popov, V.P. Zakharova, Ferrous Metals 1, 67-73 (2023). DOI: 10.17580/chm.2023.01.10 1. V.V. Maksarov, М.A. Popov, V.P. Zakharova, Ferrous Metals 1, 67-73 (2023). DOI: 10.17580/chm.2023.01.10 2. A.D. Khalimonenko, E.G. Zlotnikov, I.V. Gorshkov, M.A. Popov, Materials Science Forum 1040, 21-27 (2021) doi.org/10.4028/www.scientific.net/MSF.1040.21 2. A.D. Khalimonenko, E.G. Zlotnikov, I.V. Gorshkov, M.A. Popov, Materials Science Forum 1040, 21-27 (2021) doi.org/10.4028/www.scientific.net/MSF.1040.21 3. М.A. Popov, M.V. Vaksarov, I.A. Brigadov, Introduction of Tula State University 9(872), 637-644 (2021) DOI: 10.24412/2071-6168-2021-9-637-644 3. М.A. Popov, M.V. Vaksarov, I.A. Brigadov, Introduction of Tula State University 9(872), 637-644 (2021) DOI: 10.24412/2071-6168-2021-9-637-644 4. D. Maksimov, R. Shcheglova, D. Klochkov, E3S Web Conf. International Conference of Innovations, Physical Studies and Digitalization in Mining Engineering 326(33) (2021). DOI 10.1051/e3sconf/202132600033 – Q3 5. D. Maksimov, R. Shcheglova, D. Klochkov, E3S Web Conf. International Conference of Innovations, Physical Studies and Digitalization in Mining Engineering 326(33) (2021). DOI 10.1051/e3sconf/202132600033 6. F. Panteleenko, G. Petrishin, V. Maksarov, D. Maksimov, Russian Engineering Research 10, 41, 944 – 947 (2021) 7. I.V. Gorshkov, M.A. Popov, Key Engineering Materials 854, 64-73 (2020) doi.org/10.4028/www.scientific.net/KEM.854.64A.D 8. E.I. Pryakhin, E.Yu. Troshina, Chernye metally 4, 74–80 (2023) DOI: 10.17580/chm.2023.04.12 6
https://openalex.org/W2130047542
https://www.scielo.br/j/csp/a/qtGcddDYSBSXqbsFWPSpSnj/?lang=pt&format=pdf
Portuguese
null
Pesquisa sobre as Condições de Saúde Bucal da População Brasileira (SB Brasil 2003): seus dados não produzem estimativas populacionais, mas há possibilidade de correção
Cadernos de Saúde Pública
2,009
cc-by
8,596
1 Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz, Rio de Janeiro, Brasil. 2 Escola Nacional de Ciências Estatísticas, Instituto Brasileiro de Geografia e Estatística, Rio de Janeiro, Brasil. Correspondência M. T. L. Vasconcellos Escola Nacional de Ciências Estatísticas, Instituto Brasileiro de Geografia e Estatística. Rua André Cavalcanti 106, Rio de Janeiro, RJ 20231-050, Brasil. mauricio.vasconcellos@ibge. gov.br Pesquisa sobre as Condições de Saúde Bucal da População Brasileira (SB Brasil 2003): seus dados não produzem estimativas populacionais, mas há possibilidade de correção Brazilian Oral Health Survey (SB Brazil 2003): data do not allow for population estimates, but correction is possible Abstract The Brazilian Oral Health Survey (SB Brazil 2003) was the most comprehensive study on oral health conditions ever conducted in Brazil. Probabilistic sampling methods were applied in order for the collected data to represent the population age groups selected in the 5 regions of the country. However, this was not possible be- cause the sampling process was never concluded, which would require estimation of the sample weights and identification of the sample’s other structural variables (selection strata and pri- mary sampling units). This paper describes the SB Brazil 2003 sample design, formulates the inclusion probabilities in the multiple selection stages, and proposes strategies for estimating the sample weights. The strategy to define the sample weights and identify the sample’s structural vari- ables hinges on retrieving data that should have been recorded in the reports produced during the survey, and in their absence, in information available from the Brazilian Institute of Geogra- phy and Statistics (IBGE) and Ministry of Educa- tion and Culture (MEC), as valid proxies. O processo de inserção da área de saúde bucal na agenda da saúde pública brasileira trouxe a necessidade de ampliar o conhecimento das condições de saúde bucal da população bra- sileira e de quantificar os eventos de interesse para apoiar o planejamento de ações na área. A partir da década de 80 do século passado, a área começou a acumular experiência em grandes inquéritos epidemiológicos que incluíam exa- mes clínicos bucais dos entrevistados, especial- mente escolares. O primeiro inquérito sobre a saúde bucal no Brasil foi realizado em 1986 1 pela Divisão Na- cional de Saúde Bucal (DNSB) do Ministério da Saúde, com o apoio do Instituto de Planejamento Econômico e Social da Secretaria de Planejamen- to da Presidência da República. As limitações or- çamentárias foram determinantes para definir a abrangência geográfica, a população alvo e a abrangência do exame bucal. Realizou-se um inquérito na área urbana de 16 capitais, distri- buídas entre as cinco macrorregiões do país, in- cluindo crianças (6 a 12 anos), adolescentes (15 a 19 anos) e adultos (35 a 44 anos e 50 a 59 anos). A justificativa para não considerar idosos entre 65 e 74, como proposto pela Organização Mundial da Saúde (OMS) e adotado em países desenvolvidos (65 a 74 anos), foi baseada no fato de a esperança de vida ao nascer ser, à época, de 65,5 anos no Brasil 2. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Rejane Christine de Sousa Queiroz 1 Margareth Crisóstomo Portela 1 Mauricio Teixeira Leite de Vasconcellos 2 Oral Health; Sampling Studies; Health Status; Population Pesquisa sobre as Condições de Saúde Bucal da População Brasileira (SB Brasil 2003): seus dados não produzem estimativas populacionais, mas há possibilidade de correção Brazilian Oral Health Survey (SB Brazil 2003): data do not allow for population estimates, but correction is possible Pesquisa sobre as Condições de Saúde Bucal da População Brasileira (SB Brasil 2003): seus dados não produzem estimativas populacionais, mas há possibilidade de correção 47 ARTIGO ARTICLE 47 ARTIGO ARTICLE 47 ARTIGO ARTICLE 47 ARTIGO ARTICLE Abstract Comparado aos inquéritos anteriores, o SB Brasil 2003 incluiu ainda uma gama maior de dados relativos à situação sócio-econômica, au- to-avaliação da saúde bucal e acesso aos servi- ços odontológicos. Além disso, foram realizados exames clínicos que, em princípio, permitiriam comparar a situação observada com as metas de saúde bucal preconizadas para o ano 2000 pela OMS 6. Embora restrito à área urbana de algumas ca- pitais, este primeiro inquérito representou um passo importante, na medida em que somou co- nhecimentos específicos das áreas de Odonto- logia e Saúde Pública aos métodos e técnicas do Instituto de Pesquisa Econômica Aplicada (IPEA) e do Instituto Brasileiro de Geografia e Estatística (IBGE). O planejamento deste inquérito considerou a seleção de uma amostra probabilística por con- glomerados, representativa de faixas etárias da população das cinco macrorregiões. O processo de amostragem, entretanto, não foi concluído. Faltou o cálculo dos pesos amostrais necessários para a produção de estimativas válidas para a po- pulação brasileira. Dez anos depois, em 1996, com o Sistema Único de Saúde (SUS) já implantado, foi realiza- do o segundo inquérito sob coordenação da Área Técnica de Saúde Bucal do Ministério da Saúde. O seu objetivo foi verificar as alterações ocorridas no perfil de saúde bucal da população brasileira, tomando como referência o inquérito de 1986, e obter um referencial para as ações preventivas do SUS 4. Apesar de ampliar a abrangência geo- gráfica (27 capitais das Unidades da Federação), o inquérito limitou-se ao exame de cárie na faixa etária de 6 a 12 anos, com crianças seleciona- das em escolas públicas e privadas 5. Ainda que houvesse dados disponíveis no Departamento de Informática do SUS (DATASUS; http://tabnet. datasus.gov.br/cgi/sbucal/sbdescr.htm, acessa- do em 12/Dez/2007), a pesquisa sequer gerou um relatório final. Em 2004, houve a publicação dos seus prin- cipais resultados pelo Ministério da Saúde 6. A disseminação do banco de dados associado à descrição de que a pesquisa “permitiu a produ- ção de inferências para cada uma das macrorre- giões brasileiras, por porte do município e para cada idade ou grupo etário” 6 (p. Abstract Ocorrência de cárie e doença periodon- Oral Health; Sampling Studies; Health Status; Population Queiroz RCS et al. 48 Queiroz RCS et al. 48 tal, necessidade e uso de prótese dentária foram observados, além de informações sobre acesso aos serviços de saúde. cientes para a construção de um quadro abran- gente sobre as condições de saúde bucal da po- pulação brasileira. Eles representavam as carac- terísticas dos subgrupos populacionais definidos pelas faixas etárias consideradas e residentes nas cidades pesquisadas; não eram representativos da população das regiões e, muito menos, da po- pulação brasileira. Nesse inquérito, a idade de 12 anos foi uti- lizada para fins de comparação internacional, obedecendo a critérios propostos pela Federação Dentária Internacional (FDI) e pela OMS, que também lançaram metas com índices a serem alcançados por volta do ano de 2000 3. O reconhecimento dessas limitações condu- ziu o Ministério da Saúde a conceber um projeto para enfrentar o desafio de realizar um diagnós- tico de saúde bucal do país, contemplando um número maior de eventos e de grupos etários. Nasce, então, o projeto SB Brasil 2003, com abrangência realmente nacional, incorporando, pela primeira vez, as faixas etárias de 18 a 36 meses e 65 a 74 anos, e a área rural. Apesar do título, esse terceiro inquérito foi realizado entre 2002 e 2003, tendo sido concebido para obter informações epidemiológicas capazes de sub- sidiar a elaboração das diretrizes para uma po- lítica nacional de saúde bucal, bem como ações para o fortalecimento da gestão dos serviços públicos de saúde bucal nas diferentes esferas de governo. O relatório desse inquérito aponta várias dificuldades enfrentadas para a sua efetivação, incluindo questões operacionais (dimensão e a diversidade regional do país) e a necessidade de obter dados mais acurados para o planejamento governamental, diante da magnitude da neces- sidade de tratamento odontológico 2 pela popu- lação. Em relação ao procedimento de amostra- gem, o relatório informa que foram utilizadas duas amostras, uma para a faixa etária entre 6 e 12 anos e outra para as demais faixas etárias. As amostras foram selecionadas em dois estágios. Para estudantes (6 a 12 anos) foram seleciona- dos estabelecimentos de ensino (1o e 2o graus) e, neles, os estudantes que foram entrevistados e examinados. Para as demais faixas etárias, fo- ram selecionados setores censitários; neles, os domicílios onde foram entrevistados e examina- dos todos os moradores das demais faixas etárias contempladas. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Por que as estimativas do SB Brasil 2003 não são válidas para a população? Desde que Kiar 11 apresentou suas idéias sobre o método representativo, hoje conhecido como amostragem probabilística, pesos amostrais são utilizados para a produção de inferências sobre a população de pesquisa ou população amostrada, como definiu Cochran 12. Na amos- tragem probabilística cada unidade da amostra tem uma probabilidade de inclusão na amostra, representada por πi, cujo inverso é o peso amos- tral (wi), ou seja, wi = πi-1. Infelizmente, os problemas da amostra do SB Brasil 2003 não se limitam à ausência dos pesos amostrais. Para estimar variâncias e de- mais estatísticas dela dependentes (erros pa- drão, intervalos de confiança, significância de parâmetros de modelos estatísticos, dentre ou- tras) é necessário, também, conhecer as outras informações estruturais do desenho da amos- tra: o estrato de seleção e a unidade primária de amostragem. Isto deriva do fato de o plano de amostragem do SB Brasil 2003 ser complexo 13, combinando estratificação, conglomeração e seleção com probabilidades desiguais. Ou seja, um plano amostral que não permite estimação baseada na inferência estatística clássica, visto não ser aplicável um de seus pressupostos bá- sicos: o de que todas as unidades da amostra derivam de observações independentes e igual- mente distribuídas. Como a base de dados do SB Brasil 2003 não tem pesos amostrais, os resultados desta pes- quisa são representativos do grupo de 108.921 pessoas examinadas, não podendo ser extrapo- lados para a população de pesquisa (população brasileira dos grupos etários estudados). É fato que alguns desenhos de amostra são concebidos de forma que as probabilidades de seleção sejam constantes para todas as unida- des da amostra (πi = π ∀ i ∈ S, onde S é a amos- tra). Esses planos, ditos autoponderados 12, têm como propriedade principal o fato de médias, razões, proporções e prevalências amostrais se- rem estimativas não enviesadas da população, o que não vale para totais. No entanto, este não é o caso da amostra do SB Brasil 2003, cujo dese- nho forçosamente conduz a pesos desiguais. Entretanto, partindo-se do pressuposto de que as instruções de coleta foram estritamente observadas por toda a equipe de campo, pode- se buscar informações exatas, previstas para constarem dos relatórios das equipes de campo, ou aproximações válidas derivadas de dados do IBGE e do Instituto Nacional de Estudos e Pes- quisas Educacionais Anísio Teixeira (INEP) do Ministério da Educação (MEC), para determinar os pesos amostrais. Abstract 13) talvez tenha induzido o uso dos dados do SB Brasil 2003 por diversos pesquisadores, que publicaram artigos científicos em revistas nacionais e internacionais 7,8,9,10, sem perceberem que os resultados não eram inferências estatisticamente válidas para os grupos etários da população de pesquisa (bra- sileira ou por macrorregião) porque eram referi- dos e limitados ao grupo estudado (estatísticas da amostra). Embora obtivessem dados em diferentes re- giões do país, os dois inquéritos não foram sufi- HÁ POSSIBILIDADE DE CORREÇÃO DO VIÉS DE ESTIMAÇÃO DO SB BRASIL 2003? 4 49 Este artigo tem por objetivo demonstrar que o procedimento de amostragem do SB Brasil 2003 não foi concluído, alertando seus poten- ciais usuários de que estimativas produzidas não são representativas da população brasilei- ra, além de apontar para as possibilidades de correção dos problemas identificados. O arti- go não avança na correção dos problemas in- dicados, pois ela depende do resgate de infor- mações constantes dos relatórios das equipes de campo, o que necessariamente passa pelo apoio do Ministério da Saúde e das coordena- ções estaduais e municipais do projeto SB Bra- sil 2003. 16.137,5 (35 a 44 anos, em municípios de mais de 100.000 habitantes do Sudeste). Para cada grupo etário, a variação dos pe- sos relativos é bem grande, conforme indica a Tabela 2. Mesmo para estimativas de um úni- co grupo etário e uma única macrorregião, a variação dos pesos relativos não é desprezível, fato que indica que as estatísticas da amostra constituem-se em estimativas enviesadas da população correspondente, com maior ou me- nor viés em função da variabilidade observada para cada variável. É certo que para variáveis com reduzida variação na amostra as estatísticas amostrais, embora enviesadas, não serão muito diferen- tes das estimativas obtidas de forma correta; esse é possivelmente o caso da variável cárie dentária, de elevada prevalência na maioria dos grupos estudados. Contudo, quanto maior for a variabilidade da variável estudada, maior será o viés da estatística amostral em relação à estimativa correspondente. Além deste proble- ma, há que se considerar que a falta dos pesos amostrais inviabiliza a estimação de totais po- pulacionais. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Por que as estimativas do SB Brasil 2003 não são válidas para a população? Na Tabela 1, são apresentados pesos relati- vos que, apesar de grosseiros, dão uma idéia de quantas pessoas da população cada pessoa en- trevistada representaria se o SB Brasil 2003 tives- se sido realizado em 2000 com um esquema de seleção equiprovável em cada estrato da amos- tra. Os pesos relativos, calculados dividindo a população observada no Censo Demográfico de 2000 (IBGE; http://www.ibge.gov.br) pelo tama- nho da amostra em cada estrato, variam entre 17,8 (12 anos, Centro-Oeste em municípios com população entre 50.001 a 100.000 habitantes) e Queiroz RCS et al. 50 Tabela 1 Tamanho da amostra, população em 2000 e peso relativo por faixas etárias, segundo as macrorregiões e estratos de porte populacional. Por que as estimativas do SB Brasil 2003 não são válidas para a população? Macrorregiões e Faixa etária estratos de porte 18-36 meses 5 anos 12 anos populacional Amostra População * Peso Amostra População * Peso Amostra População * Peso (habitantes) relativo ** relativo ** relativo ** Total 12.117 6.474.246 534,3 26.641 3.445.580 129,3 34.550 3.524.814 102,0 Norte 3.305 651.066 197,0 4.678 335.414 71,7 6.208 314.984 50,7 Até 5.000 329 20.765 63,1 300 10.724 35,7 515 10.285 20,0 5.001-10.000 526 45.113 85,8 557 23.379 42,0 957 22.031 23,0 10.001-50.000 861 254.101 295,1 1.034 132.369 128,0 1.338 123.084 92,0 50.001- 100.000 840 96.657 115,1 1.371 49.412 36,0 1.683 47.104 28,0 100.000 ou + 749 234.430 313,0 1.416 119.530 84,4 1.715 112.480 65,6 Nordeste 2.557 1.986.462 776,9 4.580 1.073.304 234,3 7.322 1.136.394 155,2 Até 5.000 533 45.604 85,6 476 25.850 54,3 680 27.948 41,1 5.001-10.000 473 133.965 283,2 601 73.882 122,9 1.075 78.420 72,9 10.001-50.000 425 895.366 2.106,7 660 491.400 744,5 1.451 516.364 355,9 50.001- 100.000 634 268.025 422,8 1.443 144.236 100,0 2.114 151.974 71,9 100.000 ou + 492 643.502 1.307,9 1.400 337.936 241,4 2.002 361.688 180,7 Sudeste 2.342 2.510.795 1.072,1 7.291 1.314.709 180,3 8.052 1.357.615 168,6 Até 5.000 447 55.179 123,4 457 30.882 67,6 604 33.195 55,0 5.001-10.000 545 105.913 194,3 586 58.906 100,5 1.030 62.855 61,0 10.001-50.000 660 472.142 715,4 1.334 257.414 193,0 1.871 271.194 144,9 50.001- 100.000 255 253.497 994,1 2.445 135.976 55,6 2.216 142.885 64,5 100.000 ou + 435 1.624.064 3.733,5 2.469 831.531 336,8 2.331 847.486 363,6 Sul 2.725 873.180 320,4 6.042 480.144 79,5 7.119 478.579 67,2 Até 5.000 449 44.116 98,3 442 26.026 58,9 456 27.227 59,7 5.001-10.000 545 74.531 136,8 583 43.186 74,1 699 43.491 62,2 10.001-50.000 652 263.888 404,7 1.105 147.911 133,9 1.592 146.777 92,2 50.001- 100.000 629 131.156 208,5 1.805 71.343 39,5 1.995 70.386 35,3 100.000 ou + 450 359.489 798,9 2.107 191.678 91,0 2.377 190.698 80,2 Centro-Oeste 1.188 452.743 381,1 4.050 242.009 59,8 5.849 237.242 40,6 Até 5.000 145 18.978 130,9 287 10.622 37,0 479 10.737 22,4 5.001-10.000 110 33.137 301,2 231 18.788 81,3 675 18.683 27,7 10.001-50.000 286 127.530 445,9 463 70.559 152,4 1.337 70.068 52,4 50.001- 100.000 387 58.300 150,6 1.414 30.096 21,3 1.516 26.984 17,8 100.000 ou + 260 214.798 826,1 1.655 111.944 67,6 1.842 110.770 60,1 (continua) mostra, população em 2000 e peso relativo por faixas etárias, segundo as macrorregiões e estratos de porte populacional. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais 50.001 a 100.000 habitantes; e mais de 100.000 habitantes). Em cada estrato de porte popula- cional foram, a princípio, selecionados 10 mu- nicípios e, em cada município foram escolhidos os seguintes conglomerados: setores censitários, quadras ou vilas, domicílios e estabelecimentos de ensino (escolas e creches). O SB Brasil 2003 utilizou uma amostra probabi- lística com estratificação e conglomeração. A po- pulação de referência é constituída pelas pessoas dos seguintes grupos etários: 18 a 36 meses; 5 anos; 12 anos; 15 a 19 anos; 35 a 44 anos; e 65 a 74 anos. A amostra foi estratificada por macrorre- gião e, em cada uma delas, por porte populacio- nal dos municípios (até 5.000 habitantes; 5.001 a 10.000 habitantes; 10.001 a 50.000 habitantes; No entanto, o manual do coordenador 14, ao descrever a seleção das unidades primárias de amostragem (UPA), indica que as capitais de to- das as Unidades da Federação foram incluídas na amostra, não fazendo parte do sorteio de municípios. Portanto, no estrato de maior por- Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 HÁ POSSIBILIDADE DE CORREÇÃO DO VIÉS DE ESTIMAÇÃO DO SB BRASIL 2003? 51 51 Tabela 1 (continuação) ç Tamanho da amostra, população em 2000 e peso relativo por faixas etárias, segundo as macrorregiões e estratos de porte populacional. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 p ** Razão entre o número de pessoas na população em 1o de setembro de 2000 e na amostra efetiva do SB Brasil 2003. Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais Macrorregiões e Faixa etária estratos de porte 15-19 anos 35-44 anos 65-74 anos populacional Amostra População * Peso Amostra População * Peso Amostra População * Peso (habitantes) relativo ** relativo ** relativo ** Total 16.833 17.939.815 1.065,8 13.431 22.808.223 1.698,2 5.349 6.323.408 1.182,2 Norte 3.877 1.524.484 393,2 2.486 1.420.318 571,3 746 303.763 407,2 Até 5.000 680 45.175 66,4 458 41.217 90,0 122 10.916 89,5 5.001-10.000 727 97.213 133,7 430 83.794 194,9 160 21.855 136,6 10.001-50.000 852 553.086 649,2 518 466.225 900,0 150 111.822 745,5 50.001- 100.000 838 220.581 263,2 575 191.049 332,3 166 42.286 254,7 100.000 ou + 780 608.429 780,0 505 638.033 1.263,4 148 116.884 789,8 Nordeste 3.998 5.571.708 1.393,6 3.370 5.513.073 1.635,9 1.446 1.667.384 1.153,1 Até 5.000 711 128.633 180,9 548 114.200 208,4 249 43.366 174,2 5.001-10.000 776 359.952 463,9 551 310.459 563,4 253 117.759 465,5 10.001-50.000 749 2.416.235 3.225,9 618 2.095.534 3.390,8 259 763.970 2.949,7 50.001- 100.000 969 742.010 765,7 881 722.961 820,6 416 217.288 522,3 100.000 ou + 793 1.924.878 2.427,3 772 2.269.919 2.940,3 269 525.001 1.951,7 Sudeste 2.981 7.155.091 2.400,2 2.340 10.580.065 4.521,4 1.052 2.994.665 2.846,6 Até 5.000 677 162.113 239,5 487 209.345 429,9 205 78.096 381,0 5.001-10.000 683 307.811 450,7 513 388.064 756,5 238 138.388 581,5 10.001-50.000 679 1.359.052 2.001,5 534 1.821.075 3.410,3 286 574.498 2.008,7 50.001- 100.000 391 736.638 1.884,0 366 1.061.078 2.899,1 158 288.090 1.823,4 100.000 ou + 551 4.589.477 8.329,4 440 7.100.503 16.137,5 165 1.915.593 11.609,7 Sul 3.841 2.451.895 638,3 3.189 3.702.974 1.161,2 1.374 1.025.853 746,6 Até 5.000 704 131.064 186,2 516 196.436 380,7 231 69.386 300,4 5.001-10.000 665 212.672 319,8 407 307.423 755,3 184 103.677 563,5 10.001-50.000 751 728.887 970,6 633 1.060.592 1.675,5 290 311.578 1.074,4 50.001- 100.000 896 363.703 405,9 769 553.576 719,9 291 144.581 496,8 100.000 ou + 825 1.015.569 1.231,0 864 1.584.947 1.834,4 378 396.631 1.049,3 Centro-Oeste 2.136 1.236.637 578,9 2.046 1.591.793 778,0 731 331.743 453,8 Até 5.000 298 52.154 175,0 278 68.127 245,1 103 18.432 179,0 5.001-10.000 273 88.919 325,7 269 114.003 423,8 95 28.674 301,8 10.001-50.000 506 343.407 678,7 555 439.176 791,3 195 105.315 540,1 50.001- 100.000 512 135.616 264,9 467 166.813 357,2 172 32.623 189,7 100.000 ou + 547 616.541 1.127,1 477 803.674 1.684,9 166 146.699 883,7 * População em 2000 (fonte: Censo Demográfi co de 2000); ** Razão entre o número de pessoas na população em 1o de setembro de 2000 e na amostra efetiva do SB Brasil 2003. * População em 2000 (fonte: Censo Demográfi co de 2000); Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais ra, população em 2000 e peso relativo por faixas etárias, segundo as macrorregiões e estratos de porte populacional. te populacional (mais de 100 mil habitantes) de cada macrorregião, os municípios de capital não correspondem à UPA e sim a estratos de seleção, uma vez que foram incluídos na amostra, sem procedimento de seleção aleatória. Além disso, no estrato de maior porte populacional da ma- crorregião Nordeste, as nove capitais foram in- cluídas, tendo sido selecionado apenas um mu- nicípio, o que viola um princípio básico de amos- tragem: o tamanho da amostra deve ser maior que um, para permitir o cálculo da variabilidade. Este problema pode ser corrigido pelo agrupa- mento dos municípios deste estrato (exceto os de capital) com os do estrato de porte populacional imediatamente inferior (50.001 a 100.000 habi- tantes). Dessa forma, são 51 os estratos da amos- tra, correspondentes a: 27 capitais; mais 20 estra- tos correspondentes à combinação das regiões Norte, Sudeste, Sul e Centro-Oeste com 5 estratos de porte populacionais; mais quatro estratos de porte populacional na Região Nordeste. Selecionados ou incluídos os municípios, os demais estágios de seleção foram diferenciados em dois grupos de estratos: (1) os de até 50 mil habitantes; e (2) os de mais de 50 mil habitan- tes. Dentro dos municípios com até 50 mil ha- Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Queiroz RCS et al. 52 52 Tabela 2 Variação, média e coefi ciente de variação do peso relativo, segundo os grupos etários. Grupos etários Variação do peso relativo Média Coeficiente de variação (CV) 18-36 meses 63,1 e 3.733,5 534,3 3.187% 5 anos 21,3 e 744,5 129,3 3.413% 12 anos 17,8 a 363,6 102,0 3.676% 15-19 anos 66,4 a 8.329,4 1.065,8 3.865% 35-44 anos 90,0 a 16.137,5 1.698,2 3.962% 65- 74 anos 89,5 a 11.609,7 1.182,2 2.520% Variação, média e coefi ciente de variação do peso relativo, segundo os grupos etários. de dados nos domicílios e nas escolas, indicando que esses relatórios seriam a fonte preferencial para obtenção dos dados requeridos. Porém, na ausência desses dados nos relatórios ou no caso dos relatórios não serem recuperados, a opção que resta é buscar informações em fontes exó- genas (IBGE e MEC) e pressupor que as regras do referido manual foram estritamente seguidas. Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais Além disso, como não foi prevista a inclusão de informações sobre as escolas selecionadas nem sobre os respectivos números de alunos nos re- latórios ou na folha do questionário de coleta de dados, só restam para esses casos os dados do Censo Escolar do MEC. bitantes, foram selecionadas quadras e escolas (segundo estágio). Dentro das quadras, foram selecionados os domicílios (terceiro estágio) em que foram entrevistados os indivíduos. E, dentro das escolas, foram selecionados, no terceiro está- gio, os alunos com a idade de interesse (12 anos). Para os municípios com mais de 50 mil habitan- tes foram selecionados setores, escolas e creches (primeiro estágio, nos municípios de capital, e segundo estágio nos demais). Dentro dos setores, foram selecionadas quadras (segundo ou terceiro estágio) e posteriormente domicílios (terceiro ou quarto estágio) em que foram entrevistados os indivíduos. E, dentro das escolas e creches, foram selecionados os alunos com as idades de interes- se (5 e 12 anos). A Tabela 3 fornece indicação dos estratos, das unidades elementares da amostra e dos distintos estágios de seleção para os diferen- tes grupos definidos no plano de amostragem. Para municípios, as informações necessárias para determinar as probabilidades de inclusão constam do manual do coordenador 14. No ca- so dos municípios de capital, a probabilidade de inclusão é igual a um, pois eles foram incluídos na amostra sem seleção. Para os municípios sele- cionados por processo aleatório, a probabilidade de inclusão é proporcional à estimativa popula- cional dos municípios para 1o de julho de 1999. As referidas probabilidades de inclusão constam das expressões 1 e 2 da Figura 1. Para o cálculo dos pesos amostrais, definidos como o inverso das probabilidades de inclusão das unidades elementares na amostra, é neces- sário identificar as unidades de seleção de cada estágio da amostra, a saber: municípios, esco- las, creches, setores, quadras/vilas, domicílios e alunos. Tais unidades, por sua vez, variam de acordo com o porte do município, com o fato de o município ser de capital ou não, e com o fato de o indivíduo ter sido entrevistado em domicí- lio ou estabelecimento de ensino, como indica a Tabela 3. No caso da seleção de setores, aplicável ape- nas a municípios com mais de 50 mil habitantes, embora não conste do manual do coordenador, foi feita uma seleção equiprovável e a probabili- dade de inclusão consta da expressão 3 da Figu- ra 1. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais Os dados da Base Operacional Geográfica (BOG) do IBGE são suficientes para determinar as probabilidades de cada setor selecionado. Os có- digos desses setores constam, exceto por alguns erros de digitação, da base de dados do SB Brasil 2003. A expectativa é que o relatório da equipe técnica responsável pela seleção da amostra per- mita corrigir esses poucos erros de digitação. Partindo do pressuposto de que a coleta de dados foi feita exatamente como o manual do coordenador 14 estabelece, é possível obter as informações necessárias para cálculo dos pesos amostrais. De fato, o referido manual previu um modelo de relatório de campo no qual deveriam constar informações que possibilitariam a recu- peração dos dados necessários ao cálculo das probabilidades de inclusão de quadras urbanas e vilas rurais sorteadas, além do número total de domicílios identificados, e do período de coleta A seleção de quadras/vilas foi feita com re- posição e eqüiprobabilidade, e as probabilida- des de inclusão constam das expressões 4 e 5 da Figura 1. Há pequena variação na forma de HÁ POSSIBILIDADE DE CORREÇÃO DO VIÉS DE ESTIMAÇÃO DO SB BRASIL 2003? 5 53 Tabela 3 Estratos, unidades elementares e estágios de seleção por porte do município, condição do município (capital ou não) e local da entrevista. Estratos, unidades Município até 50.000 habitantes Municípios com mais de 50.000 habitantes elementares e Capitais Outros municípios estágios de seleção Entrevista Entrevista Entrevista Entrevista Entrevista Entrevista domiciliar na escola domiciliar na escola domiciliar na escola Estratos de seleção Macrorregião e porte populacional Município Macrorregião e porte populacional Unidades elementares Moradores Alunos Moradores Alunos Moradores Alunos (12 anos) (5 ou 12 anos) (5 ou 12 anos) Unidade primária de amostragem Município Setor censitário Escola ou creche * Município Segundo estágio de seleção Quadra Escola Quadra Aluno Setor Escola ou vila (se mais de 20) ou vila censitário (se mais de 20) ou aluno ou aluno (se até 20 escolas) (se até 20 escolas) Terceiro estágio de seleção Domicílio ** Aluno Domicílio ** Quadra Aluno (se mais de 20 escolas) ou vila (se mais de 20 escolas) Quarto estágio de seleção Domicílio ** * Supõe-se que existam mais de 20 escolas e mais de 20 creches em todas as capitais, caso contrário valem as regras para outros municípios com mais de 50.000 habitantes. Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais Este é, em princípio, o único caso em que será necessário usar simulação estatística para avaliar o efeito de diferentes agrupamentos de * Supõe-se que existam mais de 20 escolas e mais de 20 creches em todas as capitais, caso contrário valem as regras para outros municípios com mais de 50.000 habitantes. Este é, em princípio, o único caso em que será necessário usar simulação estatística para avaliar o efeito de diferentes agrupamentos de * Supõe-se que existam mais de 20 escolas e mais de 20 creches em todas as capitais, caso contrário valem as regras para outros municípios com mais de 50.000 habitantes. Este é, em princípio, o único caso em que será necessário usar simulação estatística para avaliar o efeito de diferentes agrupamentos de alunos por escola ou creche; ** De fato, os domicílios não são sempre selecionados. Para alguns grupos etários, todos os moradores de todos os domicílios são entrevistados (os que se recusarem são considerados como não-resposta). No entanto, para certos grupos etários mais freqüentes na população foi estabelecida uma fração de amostragem, que implica a seleção de um a cada dois domicílios. Neste caso, há seleção de domicílios e são entrevistados todos os moradores do grupo etário, desde que dêem seu consentimento. pensar a diferença entre a amostra desejada e a efetiva (que, no caso, é a não-resposta). calcular numerador e denominador em função da maneira como é calculada a média de do- micílios por quadra. Nos municípios até 50 mil habitantes, onde não há seleção de setores, essa média parte do pressuposto de que o tamanho médio do domicílio é igual a quatro pessoas (ex- pressão 5.1 da Figura 1). Nos demais municí- pios, onde os setores foram selecionados, essa média parte do pressuposto de que os setores têm, em média, 300 domicílios (expressão 5.3 da Figura 1). Voltando à estratégia para calcular as pro- babilidades de inclusão, serão usados os dados dos relatórios das equipes de campo. Caso eles não existam, o número de quadras/vilas pode ser obtido nos mapas dos setores selecionados (mu- nicípios com mais de 50.000 habitantes) ou nos Mapas Municipais para fins Estatísticos (MME), disponíveis na BOG do IBGE. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais n é o número de municípios selecionados no estrato e, de um modo geral igual a 10, exceto para o maior estrato de porte populacional das macrorregiões Norte, Sudeste, Sul e Centro-Oeste, em que n é igual a 3, 6, 7 e 6, respectivamente, bem como para o quarto estrato de porte populacional (50.001 a 100.000 habitantes) do Nordeste onde n é igual 11 (pela inclusão do único município de mais de 100 mil habitantes selecionado). Probabilidade de inclusão de setores (apenas para municípios com mais de 50.000 habitantes): 10 (3) P(setor) = , se seleção com eqüiprobabilidade; ou número de setores do município 10 (3) P(setor) = , se seleção com eqüiprobabilidade; ou número de setores do município Desenho da amostra do SB Brasil 2003 e estratégias de cálculo dos pesos amostrais Para a seleção dos domicílios, deve-se con- siderar duas possibilidades que dependem do estrato da amostra e da faixa etária em conside- ração, como indica a expressão 6 da Figura 1: (1) não há seleção de domicílios e todas as unida- des domiciliares existentes na quadra entram na amostra (probabilidade de inclusão igual a um), ou (2) a seleção é realizada de forma sistemáti- ca, sendo pesquisado um a cada dois domicílios (probabilidade de inclusão igual a ½). As frações de amostragem de 100% ou 50% estão indicadas na Tabela 4. Também nesse estágio de seleção não é necessário contar os domicílios das qua- dras e vilas selecionados, bastando contá-los na base de dados do SB Brasil 2003 e aplicar a fração Há, também, uma certa imprecisão na de- finição do tamanho da amostra de domicílios (usados nas expressões 5.2 e 5.4 da Figura 1) no manual do coordenador 14, em função de ajus- tes feitos (correção de população finita, popula- ção média dos municípios em cada estrato etc.). Essa imprecisão fica evidente nos dados apre- sentados na Tabela 4. Mas isso não é importante para calcular as probabilidades de inclusão, pois será usado o tamanho efetivo da amostra, deter- minado com base do banco de dados do SB Bra- sil 2003. O uso do tamanho efetivo da amostra corresponde a uma correção de não-resposta, visto que irá alterar o peso amostral para com- Queiroz RCS et al. 54 Figura 1 Esquema probabilístico da amostra do SB Brasil 2003. A. Probabilidades de inclusão na amostra para pessoas entrevistadas nos domicílios Probabilidade de inclusão de municípios (1) P(município de capital = 1) n x população do município em 01/07/99 (2) P(outros municípios) = , onde população do estrato em 01/07/99 n é o número de municípios selecionados no estrato e, de um modo geral igual a 10, exceto para o maior estrato de porte populacional das macrorregiões Norte, Sudeste, Sul e Centro-Oeste, em que n é igual a 3, 6, 7 e 6, respectivamente, bem como para o quarto estrato de porte populacional (50.001 a 100.000 habitantes) do Nordeste onde n é igual 11 (pela inclusão do único município de mais de 100 mil habitantes selecionado). Probabilidade de inclusão de quadras ou vilas rurais: (4) P(quadras/vila) = 1, se o município tiver até 10 quadras; ou TAqv (5) P(quadras/vila) = , onde TPqv TAqv (5) P(quadras/vila) = , onde TPqv TAqv (5) P(quadras/vila) = , onde TPqv TAqv é o tamanho da amostra de quadras e vilas; e TPqv é o tamanho da população de quadras e vilas. TAqv é o tamanho da amostra de quadras e vilas; e TPqv é o tamanho da população de quadras e vilas. Para o cálculo de TAqv e TPqv as regras mudam de acordo com a população municipal. Para o cálculo de TAqv e TPqv as regras mudam de acordo com a população municipal. Para municípios com até 50 mil habitantes, não há seleção de setores e o tamanho da amostra depende do número médio de domicílios (MDQ) por quadra/vila, como indicado nas expressões 5.1 e 5.2: Para municípios com até 50 mil habitantes, não há seleção de setores e o tamanho da amostra depende do número médio de domicílios (MDQ) por quadra/vila, como indicado nas expressões 5.1 e 5.2: População total do município População total do município 4 4 (5.1) MDQ = Número de quadras/vilas do município 4 (5.1) MDQ = Número de quadras/vilas do município TAd (5.2) TAqv = onde TAd é o tamanho da amostra de domicílios indicado na Tabela 4. MDQ TAd (5.2) TAqv = onde TAd é o tamanho da amostra de domicílios indicado na Tabela 4. MDQ TAd (5.2) TAqv = onde TAd é o tamanho da amostra de domicílios indicado na Tabela 4. MDQ Para municípios com mais de 50 mil habitantes, há seleção de setores e o tamanho da amostra depende do número médio de domicílios (MDQ) por quadra/vila nos 10 setores selecionados, como indicado nas expressões 5.3 e 5.4: Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 300 x 10 Probabilidade de inclusão de escolas (para seleção de crianças de 12 anos) (7) P(escola) = 1, se o município tiver até 20 escolas de 1o grau freqüentadas por crianças de 12 anos; ou 20 x número de alunos da escola (8) P(escola) = Total de alunos de 12 nas escolas do município Probabilidade de inclusão de creches (crianças de 5 anos e municípios com população > 50.000) (9) P(creche) = 1, se o município tiver até 20 creches freqüentadas por crianças de 5 anos; ou Tamanho da amostra de alunos (10) P(creche) = Total de alunos de 5 anos nas creches do município Probabilidade de inclusão de crianças de 5 e 12 anos Tamanho de amostra de alunos (11) P(criança) = Total de alunos nas escolas selecionadas (ou incluídas) Probabilidade de inclusão das unidades elementares (alunos) A probabilidade de inclusão dos alunos corresponde ao produto das probabilidades de inclusão do município, da escola (ou creche) e da criança. A probabilidade de inclusão dos alunos corresponde ao produto das probabilidades de inclusão do município, da escola (ou creche) e da criança. A probabilidade de inclusão dos alunos corresponde ao produto das probabilidades de inclusão do município, da escola (ou creche) e da criança. de amostragem correspondente à idade das pes- soas entrevistadas. de estabelecimentos de ensino do município. A probabilidade de inclusão de um aluno consta da expressão 11 da Figura 1. Também, neste caso, será usado o tamanho da amostra efetiva como numerador (para correção das não-respostas) e o denominador, obtido da mesma fonte usada para o numerador das expressões 8 e 10 da Figura 1, corresponde à soma de alunos de 12 anos (ou 5 anos) das escolas (ou creches) selecionadas ou incluídas na amostra. As pessoas de 12 anos foram sempre selecio- nadas em escolas. As crianças de 5 anos foram se- lecionadas nos domicílios, se o município tinha até 50.000 habitantes, ou em creches no caso de municípios com mais de 50.000 habitantes. Pa- ra os estabelecimentos de ensino (escolas e cre- ches), as probabilidades de inclusão são dadas pelas expressões 7 a 10 da Figura 1. As expressões 7 e 9 aplicam-se aos municípios com até 20 es- colas ou creches, respectivamente, caso no qual os estabelecimentos de ensino são incluídos na amostra com probabilidade igual a um. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 300 x 10 300 x 10 (5.3) MDQ = Número de quadras/vilas nos 10 setores selecionados TAd (5.4) TAqv = onde TAd é o tamanho da amostra de domicílios indicado na Tabela 4. MDQ Probabilidade de inclusão dos domicílios (6) P(domicílio) = 1 ou ½, de acordo com as frações de amostragem indicadas na Tabela 4. Probabilidade de inclusão das unidades elementares (moradores) Como os moradores não são alvo de seleção, pois todos os residentes em domicílios selecionados que aceitaram a pesquisa foram examinados, sua proba- bilidade de inclusão corresponde ao produto das probabilidades de inclusão nos diferentes estágios de seleção. B. Probabilidades de inclusão para crianças selecionadas em estabelecimentos de ensino Probabilidade de inclusão de municípios São as mesmas definidas anteriormente nas expressões (1) e (2). Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 HÁ POSSIBILIDADE DE CORREÇÃO DO VIÉS DE ESTIMAÇÃO DO SB BRASIL 2003? 55 55 Figura 1 (continuação) Probabilidade de inclusão de escolas (para seleção de crianças de 12 anos) (7) P(escola) = 1, se o município tiver até 20 escolas de 1o grau freqüentadas por crianças de 12 anos; ou 20 x número de alunos da escola (8) P(escola) = Total de alunos de 12 nas escolas do município Probabilidade de inclusão de creches (crianças de 5 anos e municípios com população > 50.000) (9) P(creche) = 1, se o município tiver até 20 creches freqüentadas por crianças de 5 anos; ou Tamanho da amostra de alunos (10) P(creche) = Total de alunos de 5 anos nas creches do município Probabilidade de inclusão de crianças de 5 e 12 anos Tamanho de amostra de alunos (11) P(criança) = Total de alunos nas escolas selecionadas (ou incluídas) Probabilidade de inclusão das unidades elementares (alunos) A probabilidade de inclusão dos alunos corresponde ao produto das probabilidades de inclusão do município, da escola (ou creche) e da criança. 300 x 10 As ex- pressões 8 e 10 da Figura 1 indicam que, no caso de mais de 20 escolas e mais de 20 creches, elas foram selecionadas com probabilidade propor- cional ao seu número de alunos. Para determinar as probabilidades de inclusão das expressões 8 e 10 serão preferencialmente usadas as informa- ções dos relatórios das equipes de campo. Na au- sência dessas informações, a melhor aproxima- ção possível é usar o número de alunos de 6a série de cada escola (ou de cinco anos na pré-escola) selecionada e o número de alunos de 6a série de todas as escolas (ou de 5 anos de todas as pré-es- colas) do município, disponibilizados pelo Censo Escolar de 2003 (MEC). Enfim, a fonte de informações preferencial para o cálculo das probabilidades de inclusão são os relatórios das equipes de campo. Todavia, não sendo possível recuperar todas as informações necessárias nesses relatórios, o uso de informa- ções do IBGE e do INEP permitirá obter proba- bilidades de inclusão aproximadas em cada es- tágio de seleção. Com essas probabilidades de inclusão, exatas ou aproximadas, o peso amostral de cada unidade elementar da amostra pode ser obtido pelo inverso do produto das suas probabi- lidades de inclusão em cada estágio da amostra, como indicado na Figura 1. Após o cálculo exato ou aproximado dos pe- sos amostrais, um processo de calibração desses pesos deve ser aplicado para assegurar que as estimativas reproduzam os totais populacionais por sexo e faixa de idade, conhecidos para todos os estratos de seleção mediante a projeção da po- pulação feita pelo IBGE, para o ano de 2003. Para crianças de 5 e 12 anos selecionadas em estabelecimentos de ensino, foi adotado um procedimento de seleção sistemática com eqüi- probabilidade, independentemente do número Entretanto, calcular os pesos amostrais não é suficiente. É preciso identificar cada estrato de Queiroz RCS et al. 56 Queiroz RCS et al. 56 Tabela 4 Número previsto de domicílios na amostra, tamanho da amostra e fração de amostragem por faixa etária, segundo as macrorregiões e os estratos de porte populacional. Número previsto de domicílios na amostra, tamanho da amostra e fração de amostragem por faixa etária, segundo as macrorregiões e os estratos de porte populacional. Número previsto de domicílios na amostra, tamanho da amostra e fração de amostragem por faixa etária, segundo as macrorregiões e os estratos de porte populacional. 300 x 10 Macrorregiões e Número previsto Faixa etária porte populacional de domicílios 18-36 5 12 15-19 35-44 65-74 (habitantes) na amostra meses anos anos anos anos anos Tamanho da amostra Norte 100 145 178 91 51 15 Nordeste 100 163 240 97 59 23 Sudeste 100 329 297 76 43 14 Sul 100 218 238 92 40 13 Centro-Oeste 100 214 218 70 40 14 Fração de amostragem Norte Até 5.000 157 1 1 EE 1 1 1 5.001-10.000 175 1 1 EE 1 1 1 10.001-50.000 187 1 1 EE 1 1 1 Mais de 50.000 191 1 EE EE 1 1 1 Nordeste Até 5.000 172 1 1 EE 1 1 1 5.001-10.000 188 1 1 EE 1 1 1 10.001-50.000 201 1 1 EE 1 1 1 Mais de 50.000 208 1 EE EE 1 1 1 Sudeste Até 5.000 160 1 1 EE 1 1/2 1 5.001-10.000 172 1 1 EE 1 1/2 1 10.001-50.000 184 1 1 EE 1 1/2 1 Mais de 50.000 190 1 EE EE 1 1/2 1/2 Sul Até 5.000 184 1 1 EE 1 1/2 1 5.001-10.000 210 1 1 EE 1 1/2 1 10.001-50.000 230 1 1 EE 1 1/2 1 Mais de 50.000 236 1 EE EE 1 1/2 1/2 Centro-Oeste Até 5.000 144 1 1 EE 1 1/2 1 5.001-10.000 150 1 1 EE 1 1/2 1 10.001-50.000 155 1 1 EE 1 1 1 Mais de 50.000 160 1 EE EE 1 1/2 1 EE: pesquisado em estabelecimento de ensino. populacional na macrorregião Nordeste, onde o estrato de maior porte populacional foi agrupado ao estrato anterior, como indicado anteriormen- te). seleção e todas as suas unidades primárias de amostragem, a fim de viabilizar o cálculo de vari- âncias e demais estatísticas delas dependentes. O manual do coordenador 14 esclarece que a amostra tem 25 estratos (5 macrorregiões x 5 es- tratos de porte populacional). Já o texto anterior mostra que, na realidade, a amostra do SB Brasil 2003 foi selecionada de uma população estrati- ficada em 51 estratos (= 27 capitais + 20 estratos de porte populacional nas macrorregiões Norte, Sudeste, Sul e Centro-Oeste + 4 estratos de porte Também não são 250 UPA (10 municípios x 25 estratos), como está descrito no referido ma- nual, apesar de o número exato de UPA ainda não ter sido determinado. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Adaptado das Tabelas 2, 4 e 5 do manual do coordenador 14. EE: pesquisado em estabelecimento de ensino. 300 x 10 A Tabela 3 mostra que, de acordo com o estrato de seleção, a UPA pode ser o município, o setor censitário, a escola ou a creche. Como o número de escolas e creches ain- Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 HÁ POSSIBILIDADE DE CORREÇÃO DO VIÉS DE ESTIMAÇÃO DO SB BRASIL 2003? 57 HÁ POSSIBILIDADE DE CORREÇÃO DO VIÉS DE ESTIMAÇÃO DO SB BRASIL 2003? 57 57 da não foi determinado, o número total de UPA ainda é desconhecido. O problema, nesse caso, é que o arquivo de dados não identifica cada es- tabelecimento de ensino (escola ou creche) sele- cionado, apenas identificando o setor, a quadra ou vila e o tipo de estabelecimento de ensino. Os relatórios das equipes de campo constituem-se na fonte ideal para determinar quais estabeleci- mentos de ensino foram selecionados em cada município da amostra. Na sua ausência, a única alternativa é fazer uso da técnica de simulação estatística para identificar o impacto de diferen- tes agrupamentos (aleatoriamente construídos em cada município) de alunos por escola (ou creche) sobre as variâncias das principais vari- áveis da pesquisa. Se o impacto não for grande, pode-se determinar − e identificar no arquivo − os agrupamentos a serem utilizados no proces- so de estimação de variâncias e estatísticas delas derivadas. Conclusão da não foi determinado, o número total de UPA ainda é desconhecido. O problema, nesse caso, é que o arquivo de dados não identifica cada es- tabelecimento de ensino (escola ou creche) sele- cionado, apenas identificando o setor, a quadra ou vila e o tipo de estabelecimento de ensino. Os relatórios das equipes de campo constituem-se na fonte ideal para determinar quais estabeleci- mentos de ensino foram selecionados em cada município da amostra. Na sua ausência, a única alternativa é fazer uso da técnica de simulação estatística para identificar o impacto de diferen- tes agrupamentos (aleatoriamente construídos em cada município) de alunos por escola (ou creche) sobre as variâncias das principais vari- áveis da pesquisa. Se o impacto não for grande, pode-se determinar − e identificar no arquivo − os agrupamentos a serem utilizados no proces- so de estimação de variâncias e estatísticas delas derivadas. Trata-se, portanto, de uma proposta de correção trabalhosa, que requer a busca de informações junto às coordenações municipais, estaduais e regionais do SB Brasil 2003 e, para completar as informações faltantes, o processamento de bases de dados exógenas ao projeto. Contudo, o esforço de completar o processo de amostragem probabilística do SB Brasil 2003, calculando pesos amostrais e identificando as demais informações estruturais de sua amostra, justifica-se para ampliar o uso das informações coletadas, que, em última instância, é o que jus- tifica o custo de qualquer pesquisa. Além disso, inferências válidas para a popula- ção, então tecnicamente viáveis, permitirão que pesquisadores e gestores possam analisar as reais condições de saúde bucal no país e traçar metas para seu melhoramento. Resumo A Pesquisa sobre as Condições de Saúde Bucal da Po- pulação Brasileira (SB Brasil 2003) foi a mais abran- gente pesquisa sobre as condições de saúde bucal re- alizada no Brasil até hoje. Métodos de amostragem probabilística foram utilizados para assegurar que os dados coletados representariam a população das fai- xas etárias escolhidas nas cinco regiões do Brasil. No entanto, isso não foi possível, pois o processo de amos- tragem da pesquisa não foi finalizado. Sua conclusão pressupõe o cálculo dos pesos amostrais e a identifi- cação das demais variáveis estruturais da amostra (estratos de seleção e unidades primárias de amostra- gem). Este artigo descreve o desenho de amostra desta pesquisa, formula as probabilidades de inclusão nos diversos estágios de seleção e propõe estratégias de cál- culo dos pesos amostrais. A solução para determinar os pesos amostrais e identificar as demais variáveis estruturais da amostra, proposta no artigo, parte da recuperação de informações que deveriam ter sido re- gistradas nos relatórios produzidos durante a pesquisa de campo e, em sua ausência, em informações dispo- níveis no Instituto Brasileiro de Geografia e Estatística e no Ministério da Educação, como aproximações vá- lidas. M. C. Portela e R. C. S. Queiroz identificaram a falta de pesos amostrais no arquivo de dados do SB Brasil 2003. M. T. L. Vasconcellos orientou o estudo dos documentos existentes sobre o desenho da amostra e identificou os demais problemas do desenho. Os três autores planeja- ram os procedimentos de correção dos problemas iden- tificados. R. C. S. Queiroz escreveu a primeira versão do manuscrito, que foi revisado e aprovado pelos demais autores. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Saúde Bucal; Amostragem; Nível de Saúde; População Agradecimentos Os autores agradecem à Maria Cecília Goi Porto Alves as informações relativas ao teste piloto realizado em Diadema (São Paulo) e Canela (Rio Grande do Sul), e a Angelo Giuseppe Roncalli da Costa Oliveira os escla- recimentos prestados sobre a técnica de seleção dos setores, além de sua cooperação ao fornecer a listagem dos setores selecionados. Este trabalho recebeu apoio da Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ; processo E-26/171.336/2006), e os seus autores são bolsistas do Conselho Nacional de De- senvolvimento Científico e Tecnológico (CNPq). Saúde Bucal; Amostragem; Nível de Saúde; População Queiroz RCS et al. 58 Queiroz RCS et al. 58 58 Referências 1. Divisão Nacional de Saúde Bucal, Ministério da Saúde. Levantamento epidemiológico em saúde bucal: Brasil, zona urbana. Brasília: Ministério da Saúde; 1988. 1. Divisão Nacional de Saúde Bucal, Ministério da Saúde. Levantamento epidemiológico em saúde bucal: Brasil, zona urbana. Brasília: Ministério da Saúde; 1988. 7. Antunes JLF, Peres MA, Mello TRC, Waldman EA. Multilevel assessment of determinants of dental caries experience in Brazil. Community Dent Oral Epidemiol 2006; 34:146-52. 2. Oliveira AGR, Unfer B, Costa ICC, Arcieri RM, Guimaraes LOC, Saliba NA. Levantamentos epide- miológicos em saúde bucal: análise da metodolo- gia proposta pela Organização Mundial da Saúde. Rev Bras Epidemiol 1998; 1:177-89. 8. Antunes JLF, Peres MA, Mello TRC. Determinantes individuais e contextuais da necessidade de trata- mento odontológico na dentição decídua no Bra- sil. Ciênc Saúde Coletiva 2006; 11:79-87. 9. Matos DL, Lima-Costa MF. Auto-avaliação da saúde bucal entre adultos e idosos residentes na Região Sudeste: resultados do Projeto SB-Brasil, 2003. Cad Saúde Pública 2006; 22:1699-707. 3. Fédèration Dentaire Internationale. Global goals for oral health in the year 2000. Int Dent J 1982; 32:74-7. 3. Fédèration Dentaire Internationale. Global goals for oral health in the year 2000. Int Dent J 1982; 32:74-7. 4. Costa JFR, Chagas LD, Silvestre RM, organizado- res. A política nacional de saúde bucal do Brasil: registro de uma conquista histórica. Brasília: Or- ganização Pan-Americana da Saúde/Ministério da Saúde; 2006. (Serie Técnica Desenvolvimento de Sistemas e Serviços de Saúde, 11). 10. Narvai P, Frazão P, Roncalli A, Antunes J. Cárie den- tária no Brasil: declínio, polarização, iniqüidade e exclusão social. Rev Panam Salud Pública 2006; 19:385-93. 11. Kiar AN. Observations et expériences concernant les dénombrements représentatifs. Bulletin of the International Statistical Institute 1895; 9:176-205. 5. Cad. Saúde Pública, Rio de Janeiro, 25(1):47-58, jan, 2009 Recebido em 02/Jan/2008 Versão final reapresentada em 15/Mai/2008 Aprovado em 06/Jun/2008 Agradecimentos Coordenação de Saúde Bucal, Departamento de Assistência e Promoção à Saúde, Secretaria de As- sistência à Saúde, Ministério da Saúde. Levanta- mento epidemiológico em saúde bucal: 1a etapa – cárie dental. Brasília: Ministério da Saúde; 1996. 12. Cochran WG. Sampling techniques. 3rd Ed. New York: John Wiley and Sons; 1977. 13 13. Pessoa DGC, Silva PLN. Análise de dados amos- trais complexos. São Paulo: Associação Brasileira de Estatística; 1998. 6. Departamento de Atenção Básica, Secretaria de Atenção à Saúde, Ministério da Saúde. Projeto SB Brasil 2003. Condições de saúde bucal da popu- lação brasileira, 2002-2003: resultados principais. Brasília: Ministério da Saúde; 2004. 14. Área Técnica de Saúde Bucal, Departamento de Atenção Básica, Secretaria Políticas de Saúde, Mi- nistério da Saúde. Projeto SB2000. Condições de saúde bucal da população brasileira no ano 2000: manual do coordenador. Brasília: Ministério da Saúde; 2001. Recebido em 02/Jan/2008 Versão final reapresentada em 15/Mai/2008 Aprovado em 06/Jun/2008
https://openalex.org/W2026098917
https://escholarship.org/content/qt6cg9m5gt/qt6cg9m5gt.pdf?t=q6kliz
English
null
All Grown up and Ready to Rumble
Cell
2,001
cc-by
2,226
UC Irvine UC Irvine Previously Published Works Title All Grown up and Ready to Rumble Permalink https://escholarship.org/uc/item/6cg9m5gt Journal Cell, 107(3) ISSN 0092-8674 Author Goldstein, Steve AN Publication Date 2001-11-01 DOI 10.1016/s0092-8674(01)00555-4 Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed UC Irvine UC Irvine Previously Published Works Title All Grown up and Ready to Rumble Permalink https://escholarship.org/uc/item/6cg9m5gt Journal Cell, 107(3) ISSN 0092-8674 Author Goldstein, Steve AN Publication Date 2001-11-01 DOI 10.1016/s0092-8674(01)00555-4 Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Powered by the California Digital Library University of California eScholarship.org Cell All Grown up and Ready to Rumble Ion Channels of Excitable Membranes, 3rd Edition By Bertil Hille Sunderland, MA: Sinauer Associates (2001). 814 pp. $85.00 Potassium Channels in Cardiovascular Biology Edited by S.L. Archer and N.J. Rusch New York: Kluwer Academic/Plenum Publishers (2001). 899 pp. $125.00 “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) Induced overexpression of ion channel parts is now permitting isolation, crystallization, and high resolution imaging, for example, of a bacterial potassium channel pore (Doyle et al., Science 280, 69–77, 1998), catapulting discussions of the structural basis for function from the inferred to the determined. g Potassium channels are required contributors to the electrical impulses that maintain normal cardiac pump function and blood vessel tone. Because potassium ion efflux restores cells to negative resting potentials, po- tassium channel activity limits excitation, decreasing the frequency and duration of heart beats and promoting vasodilation. Still, one might well wonder about a book exclusively on potassium channels, and only those in the heart and vessels. The explanation lies in part with the vast diversity of potassium channels that allow for cardiovascular function: over 30 types of pore-forming potassium channel subunits (many of which manifest splice variation) that form a single pore by assembly in groups of four often nonidentical subunits. These com- plexes show uneven geographical and temporal expres- sion at both the cellular and subcellular levels. Moreover, normal ion channel function depends on interaction of pore-forming subunits with one or more regulatory sub- units that can be constant or occasional companions. More than 15 of these regulators are now known; some soluble, others integral membrane proteins—these too show geographic and temporal diversity. Variation in potassium channel function is also observed during healthy development and secondary to disease due to altered gene expression. Indeed, the functional diversity that results from inheritance of ion channel genes in wild-type, mutant, and polymorphic form reveals that normal cardiovascular physiology absolutely depends on the exquisite timing and normal activity level of potas- sium channels. Novel molecular approaches are rapidly moving the field beyond its biophysical origins. Now, the bevy of regulatory proteins that interact with individual ion chan- nels to determine their native behavior are being enu- merated. Levels of expression of ion channel genes in different tissues, during normal development and in response to stress, can now be quantified. High- throughput strategies to rapidly assess thousands of compounds are now regularly applied to discovery ef- forts seeking medications that alter ion channel function. “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) Later, a few channel types with rich natural sources, such as skeletal muscle or Torpedo electric organ, were purified and reconstituted in functional form, allowing both their study free of confounding cellular influences as well as direct biochemical assessment (see Ion Channel Recon- stitution, C. Miller, ed., New York: Plenum Press, 1986). even as the field evolves. As only a single author can, the book offers a coherent worldview, moving from action potentials and single-channel recordings to crystal structures and the mechanistic basis for disease. While not seeking to present at an “expert level,” the work bravely travels (with requisite forewarning) to the very edge of current knowledge; indeed, at this precipice where differing opinions still vie for dominance, some of us will disagree with the enunciated expectations— just as we should. The book is thus ideal for any student beyond a basic undergraduate science curriculum. It remains a resource for practitioners of the craft and the essential guide to the electrophysiological universe. Potassium Channels in Cardiovascular Biology, ed- ited by S.L. Archer and N.J. Rusch. The heart is a pump. Its every beat is orchestrated by wavelets of sodium, potassium, calcium, and chloride ions moving through ion channels to yield excitation, contraction, and relax- ation. The beats come faster as we run, slower when we sleep, and recur some 100,000 times a day. This activity maintains blood pressure and circulation through the vas- cular system, delivering oxygen and removing waste products from our organs. The brain is particularly sensi- tive to diminished blood flow. Pump failure for a few seconds causes loss of consciousness; after minutes, permanent brain damage and death result. The cloning of ion channel genes, first by classical molecular biological techniques and now via searches of genetic databases over the internet, has produced a deluge of revelatory information: hundreds of molecules that form ion channel pores and scores of proteins that regulate ion channels are now known. DNA sequences allow prediction of protein sequences, motifs that may participate in specific functions, and membrane topol- ogy. Genes expressed in cultured cells have allowed biophysical characterization of a plethora of ion chan- nels in wild type form and again after site-specific muta- tion to define relationships between sequence and func- tion. “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) Indeed, before they were even known to Book Reviews 275 Book Reviews 275 be proteins, the astounding efficiency of ion channels allowed their role in producing action potentials to be delineated using electrical recordings of the exceedingly large axons in squid (Curtis and Cole, J. Cell. Comp. Physiol. 19, 135–144, 1942; Hodgkin and Katz, J. Physiol. 108, 37–77, 1949). Thereafter, methods to record from tiny membrane patches made it routine to examine sin- gle ion channels operating in real time in native cells (see Single-Channel Recording, B. Sakmann and E. Neher, eds, New York: Plenum Press, 1995). Later, a few channel types with rich natural sources, such as skeletal muscle or Torpedo electric organ, were purified and reconstituted in functional form, allowing both their study free of confounding cellular influences as well as direct biochemical assessment (see Ion Channel Recon- stitution, C. Miller, ed., New York: Plenum Press, 1986). The cloning of ion channel genes, first by classical molecular biological techniques and now via searches of genetic databases over the internet, has produced a deluge of revelatory information: hundreds of molecules that form ion channel pores and scores of proteins that regulate ion channels are now known. DNA sequences allow prediction of protein sequences, motifs that may participate in specific functions, and membrane topol- ogy. Genes expressed in cultured cells have allowed biophysical characterization of a plethora of ion chan- nels in wild type form and again after site-specific muta- tion to define relationships between sequence and func- tion. Induced overexpression of ion channel parts is now permitting isolation, crystallization, and high resolution imaging, for example, of a bacterial potassium channel pore (Doyle et al., Science 280, 69–77, 1998), catapulting discussions of the structural basis for function from the inferred to the determined. be proteins, the astounding efficiency of ion channels allowed their role in producing action potentials to be delineated using electrical recordings of the exceedingly large axons in squid (Curtis and Cole, J. Cell. Comp. Physiol. 19, 135–144, 1942; Hodgkin and Katz, J. Physiol. 108, 37–77, 1949). Thereafter, methods to record from tiny membrane patches made it routine to examine sin- gle ion channels operating in real time in native cells (see Single-Channel Recording, B. Sakmann and E. Neher, eds, New York: Plenum Press, 1995). “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) Ion channels let us see, hear, feel, move, and dream. Impressive. But, is it really time for two more (big) books on the subject? The answer is a vigorous “Yes.” This reflects the awesome attributes that make these pro- teins pivotal to biology and how rapidly we have gained new knowledge since the advent of recombinant DNA methodologies. Ion channels form portals across cell membranes that open and close to allow specific ions to pass. It is essen- tial that they operate only at the correct time, location, and level of activity. Why the litany of restrictions? Be- cause they are extremely efficient. We exist out of equilibrium. Our cells are high in potas- sium and low in sodium, chloride, and calcium. This uneven arrangement is achieved by the slow steady labor of energy-driven pumps, carrier class transport proteins present in plasma membranes in great abun- dance. A cell can harbor 107 sodium-potassium pumps, each hydrolyzing one ATP molecule to power influx of two potassium ions and efflux of three sodium ions about 150 times a second. Pumps create ionic imbal- ance and electrical inequality ensues: the cell interior is negative relative to its surroundings primarily because small amounts of positively charged potassium ions leak out. These asymmetries are an immediately available reservoir of stored electrochemical energy, held in check, poised for abrupt release. Ion channels reside at this dynamic interface as the agents of excitability. Their activity is no serene affair. In response to a stimulus, such as a puff of neurotrans- mitter, ion channels undergo conformational changes, open water-filled pores that span the membrane, and allow specific ions to explode through the pores down their electrochemical gradients. With no direct link to input of metabolic energy or coupling between changes in protein structure and individual translocation events, as many as 108 potassium ions can rush through a single potassium-selective channel each second—no wonder some cells have just a dozen. Among classes of proteins, the functional attributes of ion channels are, perhaps, the best described. Bio- chemical methods were not the first source of this knowledge. “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) Molecular methods applied to the genetics of human disease are revealing a spectrum of ion channel dis- eases; in some cases, inherited mutations cause a disor- der (for example, cystic fibrosis), in other cases, our individual genetic differences (polymorphisms) predis- pose to a poor outcome when faced with a secondary challenge (for example, drug-induced cardiac arrhyth- mia). In response to this veritable onslaught of new infor- mation, the two new books seek to describe where we are, show how we got here, and speculate on where we are going. Ion Channels of Excitable Membranes, 3rd Edition, by Bertil Hille. Earlier versions were notable achieve- ments for their clarity and broad utility. The third edition is markedly expanded, remarkably up to date, and stands as an unmitigated tour de force. Complete and enthusiastic, but not overwhelming, the text builds intu- itively from the biophysical roots of the discipline. By maintaining a historical flavor, the reader is cleanly of- fered the essential ideas and concepts that remain key In this context, Archer and Rusch offer a rich tapestry in 899 pages of nonredundant, thoughtful prose. The compilation covers all the key areas and moves nicely from the general to the specific. Particularly wonderful is the half of the book that covers the vascular system. This is an important subject (indeed, the endothelial cells lining our blood vessels comprise the single largest organ in the body) that is not adequately considered Cell 276 elsewhere. Topics covered include the roles of potas- sium channels of endothelial cells, smooth muscle, and the vasculature of skeletal muscle, brain, heart, lungs, and kidneys in health and disease. In large measure, the general subject chapters and those on potassium channels in the heart are truly excellent; only a few do not meet the standard of existing resources or suffer from out-of-date or parochial viewpoints. Such limits do not detract from the overall strength of the compendium. It serves as intended as a comprehensive review of potassium channels for those who study the heart and vasculature. These two books herald the maturation of our disci- pline. Our inquires no longer focus solely on biophysics and a few great mysteries, for example, by what means can an ion current cross an insulating lipid membrane. The interrogations are now more subtle; we must seek to understand the specifics: how, when, why, and why not. “..we know what we are, but know not what we may be.” Ophelia in Hamlet, 4.5.43–44 (W. Shakespeare) These books make it clear that the work ahead is just as exciting as what has come before. Steve A.N. Goldstein Departments of Pediatrics and Cellular and Molecular Physiology Boyer Center for Molecular Medicine Yale University School of Medicine 295 Congress Avenue New Haven, Connecticut 06536 Steve A.N. Goldstein Departments of Pediatrics and Cellular and Molecular Physiology Boyer Center for Molecular Medicine Yale University School of Medicine 295 Congress Avenue New Haven, Connecticut 06536
https://openalex.org/W3172583753
https://www.frontiersin.org/articles/10.3389/fmars.2021.615773/pdf
English
null
Body Composition of Common Bottlenose Dolphins in Sarasota Bay, Florida
Frontiers in marine science
2,021
cc-by
10,555
Body Composition of Common Bottlenose Dolphins in Sarasota Bay, Florida Stephanie K. Adamczak1*, Rachel R. Holser1, Daniel P. Costa1, Elizabeth J. Berens McCabe2 and Randall S. Wells2 1 Ecology and Evolutionary Biology Department, University of California, Santa Cruz, Santa Cruz, CA, United States, 2 Chicago Zoological Society’s Sarasota Dolphin Research Program, c/o Mote Marine Laboratory, Sarasota, FL, United States 1 Ecology and Evolutionary Biology Department, University of California, Santa Cruz, Santa Cruz, CA, United States, 2 Chicago Zoological Society’s Sarasota Dolphin Research Program, c/o Mote Marine Laboratory, Sarasota, FL, United States Marine mammal body composition has been an important tool that is used as a proxy for the health and condition of individuals within a population. Common bottlenose dolphin (Tursiops truncatus) body composition is influenced by variations in blubber thickness resulting from changes in temperature, prey availability, health, and life-history traits. We examined how environmental, ontogenetic, and reproductive variables influenced the body composition of common bottlenose dolphins in Sarasota Bay using data collected from a long-term monitoring project by the Sarasota Dolphin Research Program (SDRP). We found that both sea surface temperature (SST) and catch per unit effort (CPUE), used as a proxy for prey availability, influenced body composition. There was a high degree of seasonality in body composition, with higher values occurring in winter when SST and CPUE were both low. Ontogeny also greatly influenced body composition, as younger dolphins typically had thicker blubber than mature individuals. Interestingly, young females allocated more energy to allometric growth than deposition of blubber for body composition when compared to young males. However, as females matured and their growth slowed, they invested more in body composition. We found no significant difference in body composition of females of varying reproductive states, providing further evidence of their status as true income breeders. Our work highlights that changes in body composition result from fluctuations in environmental variables and that energy allocation to body composition changes with ontogeny. Edited by: Adrian C. Gleiss, Murdoch University, Australia Reviewed by: Nahiid Stephens, Murdoch University, Australia Delphine Brigitte Hélène Chabanne, Murdoch University, Australia *Correspondence: Stephanie K. Adamczak sadamcza@ucsc.edu Reviewed by: Nahiid Stephens, Murdoch University, Australia Delphine Brigitte Hélène Chabanne, M d h U i i A li Murdoch University, Australia *Correspondence: Stephanie K. Adamczak sadamcza@ucsc.edu Specialty section: This article was submitted to Marine Megafauna, a section of the journal Frontiers in Marine Science Keywords: bottlenose dolphin, body composition, cetacean, long-term monitoring, ontogeny INTRODUCTION Received: 28 October 2020 Accepted: 27 April 2021 Published: 15 June 2021 Blubber aids marine mammals in thermoregulation, energy storage, and locomotion (Parry, 1949; Struntz et al., 2004; Noren and Wells, 2009). This vascularized tissue, composed of adipocytes and collagen, helps form the streamlined shape of marine mammals, while also providing insulation against the harsh aquatic environment; a particularly important function in cetaceans, as it is their primary insulative layer (Parry, 1949). Blubber is metabolically active and can be mobilized during nutritional stress when energy intake falls below energy output. As a result, body composition, defined as the ratio of blubber mass to either total body mass or lean mass, is representative of the health and condition of an individual (Lockyer et al., 1985; Lockyer, 1986; Read, 1990; Ryg et al., 1990; Koopman et al., 2002). ORIGINAL RESEARCH published: 15 June 2021 doi: 10.3389/fmars.2021.615773 ORIGINAL RESEARCH Data Collection Common bottlenose dolphin body condition, morphometrics, genetics, life history and health data were obtained through periodic capture-release research (Wells et al., 2004). The shallow, sheltered bay waters facilitated encircling small groups of selected dolphins with a 500 m long × 4 m deep seine net. Individual dolphins were placed in a sling, and lifted aboard a veterinary examination vessel. The animals were weighed, measured for standardized lengths and girths, and ultrasonic measurements of blubber thickness were made at standardized sites. Each individual was given a full examination, including ultrasonic imaging of internal organs, by experienced marine mammal veterinarians. Sex was confirmed through direct examination of the genital region. Upon completion of sampling, dolphins were marked with freeze brands or tagged as appropriate, photographed, and released. While the ages of most of the sampled dolphins were already known from observations since birth to identifiable mothers, occasionally ages needed to be determined from examination of growth layer groups in a tooth extracted under local anesthesia (Hohn et al., 1989). Reproductive status was evaluated through measurements of reproductive hormone concentrations, through ultrasonic examination of reproductive organs, and through observations of young calves believed to be the first calves of identifiable females. Blubber, and therefore body composition, is also heavily influenced by age, sex, and reproductive status (Koopman et al., 1996; McLellan et al., 2002; Dunkin et al., 2005; Beck et al., 2007; Koopman, 2007). The ontogeny of blubber development has been well studied in cetaceans (Struntz et al., 2004; Dunkin et al., 2005; Noren and Wells, 2009); however, less is known about how body composition changes throughout an animal’s lifetime (although see McLellan et al., 2002; Mallette et al., 2016). Cetacean calves and juveniles tend to have thicker blubber layers compared to adults as more energy is allocated to blubber deposition at this stage in life, likely due to an increased need for insulation as a result of their small body size and concurrently high surface area to volume ratio (McLellan et al., 2002; Struntz et al., 2004; Dunkin et al., 2005; Noren and Wells, 2009; Mallette et al., 2016). Understanding the ontogeny of blubber development has also provided insight into how male and female marine mammals differentially allocate resources to growth and body composition. Data Collection Male and female cetaceans and pinnipeds exhibit differences in fatty acid composition (West et al., 1979; Aguilar and Borrell, 1990; Samuel and Worthy, 2004) and blubber deposition and development (Ackman et al., 1975). However, the factors driving these differences in blubber composition, morphology, and deposition (i.e., the insulative properties of blubber) are often difficult to tease apart due to variability in feeding ecology, energy expenditure, and reproductive strategies between the sexes. Sex- specific differences in body composition can be particularly challenging to detect in income breeders who continue to feed during gestation and lactation, and thereby respond to the elevated metabolic costs of gestation and lactation by increasing food consumption (Reddy et al., 1991). Relative prey fish abundance of the 10 most common dolphin prey species (pinfish, Lagodon rhomboides; pigfish, Orthopristis chrysoptera; scaled sardine, Harengula jaguana; Atlantic threadfin herring, Opisthonema oglinum; ladyfish, Elops saurus; mullet, Mullet spp.; sheepshead, Archosargus probatocephalus; spot Leiostomus xanthurus; spotted seatrout, Cynoscion nebulosus; and Gulf toadfish, Opsanus beta; Wells et al., 2013) was estimated from systematic, seasonal purse seining in Sarasota Bay and vicinity (Gannon et al., 2009; Berens McCabe et al., 2010). Since 2004, multispecies fisheries surveys have been conducted during the months of June through September, and January through March, using a 183 × 6.7-m purse seine with 2.5-cm mesh, deployed from a 9-m flat-bottomed skiffin waters between 0.4 and 4 m in depth (Sarasota Bay’s maximum depth is 4 m), with the focus in recent years on sampling in seagrass meadows as an index habitat. The net sampled the entire water column, from surface to bottom. All fishes, cephalopods, and penaeid shrimps caught in each purse seine set were identified to species and counted, and a subsample was measured for length and weight. Catch per unit effort (CPUE), or the number of organisms caught per standardized purse seine set, was used as an index of prey availability. We aimed to assess the abiotic, biotic, and life history stage- related factors influencing the body composition of an income breeding cetacean, the common bottlenose dolphin (Turiops truncatus), using data from the Sarasota Dolphin Research Program (SDRP) long-term monitoring effort in Sarasota Bay, Florida. Data Collection This long-term monitoring effort began in 1970 and has resulted in detailed and high-quality data from the resident dolphin community, including information on age, sex, life history, and environmental features such as prey abundance and water temperature (Wells, 2009, 2014; Berens McCabe et al., 2010). Our objectives were to (1) assess the factors driving body composition for the resident common bottlenose dolphin community, (2) determine how these factors change based on age and sex, (3) examine whether the reproductive state alters Citation: Citation: Adamczak SK, Holser RR, Costa DP, Berens McCabe EJ and Wells RS (2021) Body Composition of Common Bottlenose Dolphins in Sarasota Bay, Florida. Front. Mar. Sci. 8:615773. doi: 10.3389/fmars.2021.615773 June 2021 | Volume 8 | Article 615773 1 Frontiers in Marine Science | www.frontiersin.org Adamczak et al. Bottlenose Dolphin Body Composition Temperature has been shown to impact the depth and thermal properties of blubber, as species in colder climates typically have more insulative blubber (Worthy and Edwards, 1990; Koopman, 2007; Montie et al., 2008). This aids in heat retention and lowers the cost of thermoregulation. Seasonal trends in blubber depth are also observed due to fluctuations in temperature (Williams and Friedl, 1990; Meagher et al., 2008). Temperature also has indirect effects on blubber due to the relationship between oceanographic features and prey availability. Features such as thermal fronts and cold, deep-water upwelling typically increase prey abundance via input of nutrients to the surface ocean, thereby influencing the body composition of marine mammals (Miller et al., 2011; Øigård et al., 2013; Braithwaite et al., 2015). body composition of sexually mature female dolphins, and (4) assess the long-term trends in body composition for this dolphin community. Truncated Cones Body composition was calculated as the proportion of blubber volume to total body volume following the truncated cones June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 2 Bottlenose Dolphin Body Composition Adamczak et al. URE 1 | Girth, blubber depth, and truncated cones methods for three time periods: pre-1989 (A), 1989–2012 (B), and post-2012 (C). Solid lines indicate ement of girth measurements, red circles indicate lateral placement of ultrasound measurements of blubber depth, blue circles indicate approximate location of al biopsy of blubber depth, and dotted lines represent the breakdown of cones used in the truncated cones calculations. For the pre-1989 measurements, the age of blubber depths a and b were averaged for cones 1 and 2, the average of blubber depths b and c were used for cone 3, and blubber depth d was used one 4. For the 1989–2012 measurements, blubber depth a was used for cone 1, the average of blubber depths a and b were used for cone 2, the average of ber depths b and c were used for cone 3, the average of blubber depths c and d were used for cone 4, and blubber depth d was used for cone 5. For the -2012 measurements blubber depth a was used for cone 1, the average of blubber depths a, b, and c were used for cones 2 and 3, and the average of blubber hs c, d, and e were used for cones 4 and 5. FIGURE 1 | Girth, blubber depth, and truncated cones methods for three time periods: pre-1989 (A), 1989–2012 (B), and post-2012 (C). Solid lines indicate placement of girth measurements, red circles indicate lateral placement of ultrasound measurements of blubber depth, blue circles indicate approximate location of caudal biopsy of blubber depth, and dotted lines represent the breakdown of cones used in the truncated cones calculations. For the pre-1989 measurements, the average of blubber depths a and b were averaged for cones 1 and 2, the average of blubber depths b and c were used for cone 3, and blubber depth d was used for cone 4. Truncated Cones For the 1989–2012 measurements, blubber depth a was used for cone 1, the average of blubber depths a and b were used for cone 2, the average of blubber depths b and c were used for cone 3, the average of blubber depths c and d were used for cone 4, and blubber depth d was used for cone 5. For the post-2012 measurements blubber depth a was used for cone 1, the average of blubber depths a, b, and c were used for cones 2 and 3, and the average of blubber depths c, d, and e were used for cones 4 and 5. method (Gales and Burton, 1987) using lengths, girths, and ultrasound measurements of blubber depth. Blubber density was assumed to be 1.039 kg/m3 (Dunkin et al., 2010). The morphometric measurements taken varied across the time series, with three distinct methods and time periods, detailed in Figure 1. for lengths and girths. Girth measurements were consistently taken at the blowhole, axillary, anus, and tail fluke insertion, and some individuals had measurements taken anterior to the pectoral flipper. To estimate anterior flipper girth in individuals missing this measurement, we used a linear regression to derive anterior flipper girth measurements from axillary girth based on individuals with both measurements (LM R2 = 0.9169, p < < 0.01). Ultrasound measurements were taken at four The first method of collected morphometrics occurred prior to 1989, when only four body segments were consistently measured June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 3 Bottlenose Dolphin Body Composition Adamczak et al. FIGURE 2 | Linear regression of pre-1989 (n = 108) and post-2012 (n = 80) body composition calculation methods against 1989–2012 method. Regression coefficients shown were used to correct data to body compositions from before and after the 1989 to 2012 time period to the 1989–2012 method. FIGURE 2 | Linear regression of pre-1989 (n = 108) and post-2012 (n = 80) body composition calculation methods against 1989–2012 method. Regression coefficients shown were used to correct data to body compositions from before and after the 1989 to 2012 time period to the 1989–2012 method. pre-1989 (n = 108) and post-2012 (n = 80) body composition calculation methods against 1989–2012 method. Regression correct data to body compositions from before and after the 1989 to 2012 time period to the 1989–2012 method. 1https://climatedataguide.ucar.edu/climate-data/sst-data-hadisst-v11 2https://github.com/rmendels/rerddapXtracto Truncated Cones 1989–2012 method to pre-1989 (LM R2 = 0.735, p << 0.01) and post-2012 (LM R2 = 0.780, p << 0.01) and used those coefficients to correct the pre-1989 and post-2012 calculations to the 1989–2012 method (Figure 2). non-corresponding locations along the animals’ length, so the average of two appropriate locations was used for each body segment (see Figure 1). Between 1989 and 2012, five body segments were consistently measured at the eye, blowhole, posterior pectoral flipper, axillary, anus, and tail fluke insertion. Blubber depth was measured at five lateral positions along the body with 4–11 measurements taken in a ring around the girth of the animal at 10 cm intervals at each position. These ring measurements were averaged to give a single value for each lateral position. For each body segment, either a single ring measurement or the mean of two corresponding lateral ring measurements was used as the blubber depth (see Figure 1). For the final method used after 2012, morphometric measurements were the same as those taken from 1989 to 2012, however, ultrasound measurements were only taken at 3–5 individual points along the length of the body. Again, mean ultrasound values at appropriate lateral locations were used to calculate blubber depth for each segment. Frontiers in Marine Science | www.frontiersin.org Statistics All statistical analyses were conducted using R version 4.03 (R Core Team, 2020) with packages “lme4” (Bates et al., 2015) and “MuMIn” (Barton, 2009). Sea surface temperature (SST) for models was extracted from the HadISST dataset1 using the “rerddapXtracto” package in R2. The data consisted of 534 individuals across varying ages and reproductive statuses (Supplementary Table 1), with 324 repeated measures of individuals occurring throughout the long-term study. To account for this, we repeated statistical analyses without repeated measures to ensure that our results were robust. We used generalized linear mixed models (GLMMs) to determine which factors influenced body composition (Breslow and Clayton, 1993). Fixed effects included standard body length, sex, season, year, monthly SST, and CPUE. Northern hemisphere seasons were used in these analyses and were defined as follows: Winter (December, January, and February), Spring (March, April, May), Summer (June, July, August), and Fall (September, October, November). Monthly SST data were extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution for 1986–2016. Repeated measures of individual dolphins and year were regarded as random effects. Age was not included in the model due to the significant correlation between age and body length (Kendall correlation; z = 24.12, p < 2.2 × 10−16; Supplementary Figure 1). Models The total body volume, which scales allometrically with mass, was calculated using the truncated cones method by summing the volume of each cone produced by an individual body segment. Lean body volume was then estimated using the truncated cones method and the girth measurements used here were estimated by subtracting blubber depth estimates from the estimated radius of the original girth measurement. Blubber volume was obtained by subtracting the lean volume from total body volume, and body composition was estimated as the ratio of blubber volume to total body volume. To ensure that data were comparable across the time series, we used measurements from the 1989 to 2012 methods to calculate body composition for the same individuals using the pre-1989 and post-2012 methods. We then calculated regression coefficients comparing the body composition calculated using the June 2021 | Volume 8 | Article 615773 4 Bottlenose Dolphin Body Composition Adamczak et al. Statistics were fit for all (n = 534), young (less than or equal to 10 years of age, n = 254; Read et al., 1993), and mature (n = 280) dolphins because of the importance of body length and age in determining body composition (see “Results”). GLMMs that incorporated CPUE as a predictor of body composition were evaluated on a subsample of the data from 2004 onward, when CPUE measurements first became available (n = 73 and n = 84 for young and mature individuals, respectively). Best fit models were selected using Akaike’s Information Criteria corrections (AICc) following suggestions to use this method unless sample sizes are exceedingly large (Burnham and Anderson, 1994; Wagenmakers and Farrell, 2004). Because the GLMMs conducted here were performed on subsamples of the full dataset the sample size for each analysis was reduced, necessitating the use of AICc. Models with delta AICc values less than two were compared and selected based upon biological accuracy. GLMMs In the GLMMs, monthly SST was an important determinant of body composition in both young and mature dolphins, as body composition decreased with increasing SST (Table 1, Supplementary Table 2, and Figure 3). For young individuals, body length, season, and sex also influenced body composition (Table 1). When applying the model to mature individuals, season and sex were important factors influencing body composition. However, body length was not a predictor of body composition (Table 1). Additionally, year impacted body composition of mature dolphins (Table 1). Models including CPUE (data from 2004 to 2016) mirrored the results stated above, however, CPUE was an important predictor of body composition for both young and mature dolphins (Table 2 and Figure 3). GLMMs indicated that season was a significant predictor of body composition in dolphins (see “Results”). We aimed to explore this relationship further using a Kruskal-Wallis one-way analysis of variance (ANOVA). The Kruskal-Wallis ANOVA is a non-parametric analysis that can test for variation amongst groups (McKnight and Najab, 2010). This test was applied to examine seasonal differences in body composition for both young and mature dolphins (n = 254 and n = 280, respectively). A Dunn post hoc analysis was performed to determine which groups significantly differed. Seasonal changes in body composition were further compared to monthly SST and seasonal CPUE from data collected in Sarasota Bay from 2006 to 2016 in winter (collected in January, February, and March) and summer (collected in June, July, August, and September). Seasonality There was a significant difference in body composition between seasons for both young and mature dolphins (X2 = 88.75, df = 3, p < 2.22 × 10−16 and X2 = 96.81, df = 3, p < 2.22 × 10−16, respectively). The greatest difference in body composition in both mature and young dolphins occurred between summer, when body composition was the lowest, and winter, when body composition was the highest (Figure 4). Body composition in spring was significantly higher than body composition in summer and significantly lower than body composition in winter (Figure 4). Based on visual analyses, trends in body composition appear to follow an inverse pattern with seasonal CPUE, as body composition was highest in winter when CPUE was lowest (Figure 4). However, seasonal CPUE were only available for winter and summer, potentially obscuring patterns and trends between body composition and CPUE. Lower seasonally averaged SST co-occurred with high body composition values We used generalized additive models (GAMs) to evaluate the non-linear relationship between dolphin age and body composition for all dolphins (Hastie and Tibshirani, 1987). GAMs were applied to data for both sexes (n = 260 and n = 274 for males and females, respectively) to account for sex-specific differences in body composition as found by GLMMs and the number of knots was restricted to 1 to avoid overfitting the model. TABLE 1 | Generalized linear mixed models predicting body composition of common bottlenose dolphin (Tursiops truncatus) in Sarasota Bay, Florida with fixed effects of body length (length), monthly SST (SST), season, sex, year from 1986 to 2016. Group Model df Log likelihood AICc 1 AIC All Body composition ∼ Length + SST + Season + Sex + Year + SST*Season 39 1487.52 −2890.70 0.00 Young Body composition ∼ Length + SST + Season + Sex 9 747.23 −1475.80 0.00 Mature Body composition ∼ SST + Season + Sex + Year 36 750.38 −1419.80 1.25 Models were run for all individuals (n = 534), young individuals (less than 10 years old; n = 254) and mature individuals (greater than 10 years old; n = 280) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. Frontiers in Marine Science | www.frontiersin.org Seasonality TABLE 1 | Generalized linear mixed models predicting body composition of common bottlenose dolphin (Tursiops truncatus) in Sarasota Bay, Florida with fixed effects of body length (length), monthly SST (SST), season, sex, year from 1986 to 2016. To examine how reproductive state influenced body composition in mature females of known reproductive status (n = 127), we used a Kruskal-Wallis ANOVA followed by the Dunn post hoc test. Analyses were run on mature females only, and then again with mature males (n = 150) to compare body composition in different reproductive states to body composition of males. To assess long-term trends in body composition we applied a GAM to the relationship between year and body composition for all dolphins, incorporating covariates of annually averaged SST and annually averaged CPUE. Due to a sampling bias in the months of data collection, with 76% of data collected in May and June, average body composition collected in June from 1986 to 2006 and May from 2007 to 2016 were used for this analysis. Models were fit from 1986 to 2016 (n = 409) and from 2004 to 2016 (when CPUE data were available; n = 125) to ensure that the best fit model was not impacted by limitations of CPUE. All combinations of monthly SST, seasonal CPUE, and the interaction between monthly SST and seasonal CPUE were run and the best fit model was selected using AICc. Models were run for all individuals (n = 534), young individuals (less than 10 years old; n = 254) and mature individuals (greater than 10 years old; n = 280) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. Models were run for all individuals (n = 534), young individuals (less than 10 years old; n = 254) and mature individuals (greater than 10 years old; n = 280) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 5 Bottlenose Dolphin Body Composition Adamczak et al. Age TABLE 2 | Generalized linear mixed models predicting body composition of common bottlenose dolphin (Tursiops truncatus) in Sarasota Bay, Florida with fixed effects of body length (length), monthly SST (SST), catch per unit effort (CPUE), season, sex, and year from 2004 to 2016. Seasonality FIGURE 3 | Comparison of the relationship between monthly sea surface temperature (SST) and body composition (n = 254 and n = 280 for young and mature individuals, respectively) (A) and CPUE and body composition (n = 73 and n = 84 for young and mature individuals, respectively) (B) for common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay. Monthly SST data were extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution. FIGURE 3 | Comparison of the relationship between monthly sea surface temperature (SST) and body composition (n = 254 and n = 280 for young and mature individuals, respectively) (A) and CPUE and body composition (n = 73 and n = 84 for young and mature individuals, respectively) (B) for common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay. Monthly SST data were extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution. annual averages of SST in Sarasota Bay have remained relatively constant (Supplementary Figure 2). and vice versa; however, there did not appear to be a direct inverse relationship between seasonally averaged SST and body composition (Figure 4). Reproductive State We found no significant difference in female’s body composition in different reproductive states (Figure 6). However, when re-running the Kruskal-Wallis test including reproductively active males, both pregnant and lactating mature females had significantly higher body composition than that of mature males, while mature females that were simultaneously pregnant and lactating had a body composition similar to that of males (Figure 6). Age Both male and female dolphins demonstrated statistically significant negative relationships between age and body composition; however, there was a better fit for males than females (GAMs: R2 = 0.39, p < 2.22 × 10−16 for males and R2 = 0.05, p = 0.001 for females; Figure 5). There was a high degree of variability in the relationship tested for females, resulting in poor model fit. Dolphin body composition is strongly linked to SST and seasonal changes in temperature. While the factors influencing body composition of common bottlenose dolphins varied by age and sex, SST was a significant variable in determining long-term trends in body composition and greatly affected the available Dolphin body composition is strongly linked to SST and seasonal changes in temperature. While the factors influencing body composition of common bottlenose dolphins varied by age and sex, SST was a significant variable in determining long-term trends in body composition and greatly affected the available TABLE 2 | Generalized linear mixed models predicting body composition of common bottlenose dolphin (Tursiops truncatus) in Sarasota Bay, Florida with fixed effects of body length (length), monthly SST (SST), catch per unit effort (CPUE), season, sex, and year from 2004 to 2016. Group Model df Log likelihood AICc 1 AIC All Body composition ∼ Length + CPUE + SST + Season + CPUE*Season + SST*Season 13 453.073 −877.6 0.00 Young Body composition ∼ Length + CPUE + SST + Season + Sex 9 221.487 −422.4 0.00 Mature Body composition ∼ CPUE + Season + Sex + Year 17 259.128 −474.1 0.00 Models were run for all individuals (n = 157), young individuals (less than 10 years old; n = 73) and mature individuals (greater than or equal to 10 years old; n = 84) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. TABLE 2 | Generalized linear mixed models predicting body composition of common bottlenose dolphin (Tursiops truncatus) in Sarasota Bay, Florida with fixed effects of body length (length), monthly SST (SST), catch per unit effort (CPUE), season, sex, and year from 2004 to 2016. Long-Term Trends The best fit GAM for long-term trends in body composition incorporated annually averaged SST and described 55.2% of the deviance in the data (Table 2 and Figure 7). Adding annually averaged CPUE to the 1986–2016 and 2004–2016 models did not affect their output. Peaks in body composition occurred during a brief period from 1997 to 1999, and during a sudden spike in 2005. Although body composition appeared to increase over time, this is likely not a result of increasing water temperatures, as Models were run for all individuals (n = 157), young individuals (less than 10 years old; n = 73) and mature individuals (greater than or equal to 10 years old; n = 84) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 6 Bottlenose Dolphin Body Composition Adamczak et al. FIGURE 4 | Body composition for young and mature (n = 254 and n = 280, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay by season from data collected from 1986 to 2016 (A) and monthly sea surface temperature (SST) and catch per unit effort (CPUE) in Sarasota Bay average by season during the years 1986–2016, 2003–2016, and 2006–2016, respectively (B). Significance from Dunn test results are indicated by one asterisk (p < 0.05), two asterisks (p < 0.001), and three asterisks (p < 0.0001). Monthly SST data were extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution. CPUE data were collected from seine trawls in winter and spring in Sarasota Bay. FIGURE 4 | Body composition for young and mature (n = 254 and n = 280, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay by season from data collected from 1986 to 2016 (A) and monthly sea surface temperature (SST) and catch per unit effort (CPUE) in Sarasota Bay average by season during the years 1986–2016, 2003–2016, and 2006–2016, respectively (B). Significance from Dunn test results are indicated by one asterisk (p < 0.05), two asterisks (p < 0.001), and three asterisks (p < 0.0001). Monthly SST data were extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution. CPUE data were collected from seine trawls in winter and spring in Sarasota Bay. Long-Term Trends Significance from Dunn test results are indicated by two asterisks (p < 0.001) and three asterisks (p < 0.0001). FIGURE 5 | Fitted generalized additive model with 95% confidence interval for the relationship between age and body composition for female and male (n = 260 and n = 274, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay. Number of knots was restricted to one. FIGURE 5 | Fitted generalized additive model with 95% confidence interval for the relationship between age and body composition for female and male (n = 260 and n = 274, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay. Number of knots was restricted to one. FIGURE 6 | Body composition of mature female (n = 127) common bottlenose dolphins (Tursiops truncatus) of differing reproductive status and adult male (n = 150) common bottlenose dolphins in Sarasota Bay. Significance from Dunn test results are indicated by two asterisks (p < 0.001) and three asterisks (p < 0.0001). FIGURE 6 | Body composition of mature female (n = 127) common bottlenose dolphins (Tursiops truncatus) of differing reproductive status and adult male (n = 150) common bottlenose dolphins in Sarasota Bay. Significance from Dunn test results are indicated by two asterisks (p < 0.001) and three asterisks (p < 0.0001). FIGURE 6 | Body composition of mature female (n = 127) common bottlenose dolphins (Tursiops truncatus) of differing reproductive status and adult male (n = 150) common bottlenose dolphins in Sarasota Bay. Significance from Dunn test results are indicated by two asterisks (p < 0.001) and three asterisks (p < 0.0001). FIGURE 6 | Body composition of mature female (n = 127) common bottlenose dolphins (Tursiops truncatus) of differing reproduc common bottlenose dolphins in Sarasota Bay. Significance from Dunn test results are indicated by two asterisks (p < 0.001) and of mature female (n = 127) common bottlenose dolphins (Tursiops truncatus) of differing reproductive status and adult male (n = 150 Sarasota Bay. Significance from Dunn test results are indicated by two asterisks (p < 0.001) and three asterisks (p < 0.0001). decreased the strength of the relationship between CPUE and body composition. Additionally, the effect of CPUE on body composition may be lagged and the seasonal CPUE averages used here may have been too coarse when paired with the fine-scale body composition values used in the GLMMs. more insulation when water temperatures decrease. Long-Term Trends environments often result in larger adipocyte size (Montie et al., 2008), thicker blubber layers (Noren and Wells, 2009), and greater stratification in blubber layers (Koopman, 2007). As such, dolphin body composition is strongly linked to SST and seasonal changes in temperature. energy stores of the Sarasota Bay common bottlenose dolphins. This pattern is common to marine mammals, as their blubber thickness and other characteristics typically vary in response to seasonal changes in water temperature (Beck et al., 1993; Kastelein et al., 1997; Meagher et al., 2008; Montie et al., 2008). Modifications in blubber thickness are especially important for dolphins as they rely solely on blubber for insulation, are unable to haul-out and use land as a thermal refuge, and are small in size when compared to other marine mammals, making them more vulnerable to heat loss in cold waters (Parry, 1949; Hokkanen, 1990; Ryg et al., 1993; Mauck et al., 2003; Beentjes, 2006). Water temperature impacts both the morphology and deposition of blubber in dolphins, as colder CPUE also influenced the body composition of both young and mature dolphins. Although GLMM results indicate that SST is an important factor driving changes in body composition for dolphins in Sarasota Bay, CPUE was still a significant predictor of body composition when included in GLMMs (Supplementary Table 3). CPUE influenced the body composition of both young and mature dolphins. In the present study, CPUE was only available for winter and summer, which may have June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org Frontiers in Marine Science | www.frontiersin.org 7 Bottlenose Dolphin Body Composition Adamczak et al. FIGURE 5 | Fitted generalized additive model with 95% confidence interval for the relationship between age and body composition for female and male (n = 260 and n = 274, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay. Number of knots was restricted to one. FIGURE 5 | Fitted generalized additive model with 95% confidence interval for the relationship between age and body composition for female and male (n = 260 and n = 274, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay. Number of knots was restricted to one. FIGURE 6 | Body composition of mature female (n = 127) common bottlenose dolphins (Tursiops truncatus) of differing reproductive status and adult male (n = 150) common bottlenose dolphins in Sarasota Bay. Frontiers in Marine Science | www.frontiersin.org Long-Term Trends Young females invest more heavily in growth than blubber deposition, yielding a trade- offbetween depositing blubber and reaching asymptotic length (Cockroft and Ross, 1990; Read et al., 1993; Calzada et al., 1997). The discrepancy in energy allocation early in life between females and males may result from the different reproductive strategies between the sexes and earlier sexual maturation of females (Read et al., 1993). Common bottlenose dolphins in Sarasota Bay typically have a promiscuous mating system, in which males prioritize competition for access to females and females prioritize maternal investment (Wells et al., 1987; Murphy et al., 2005). As a result of the differing strategies between the sexes, females prioritize early somatic growth and sexual maturation at the cost of body condition compared to males of the same age, allowing them to maximize fitness. p y p Long-term trends in body composition were strongly linked to annually averaged SST (Figure 7). Although body composition values appeared to increase throughout the study’s duration, this is likely due to changes over time from summer to spring health assessments, when water was cooler, rather than due to shifts in annual SST in Sarasota Bay. Although SST has been rising globally as a result of climate change, temperatures in the eastern Gulf of Mexico have slightly decreased within the past few decades (Lluch-Cota et al., 2013), and temperatures have remained static in Sarasota Bay (Supplementary Figure 2). As a result, body composition of dolphins in Sarasota Bay may not be impacted by climate change at the present moment. However, this dolphin community may be vulnerable to changes to oceanography resulting from climate change in the future as a result of their high site-fidelity, which may make it difficult for them to find a thermal refuge once water temperatures change (Wells, 2010). As a result, continued monitoring is critical to determine future SST trends in this region as body composition is heavily dependent upon this metric. Reproductive costs are exceedingly high for females across marine mammal species (Fedak and Anderson, 1982; Oftedal, 1985; Yasui and Gaskin, 1986; Reddy et al., 1991; West et al., 2007; Hückstädt et al., 2018), necessitating elevated body composition values. In marine mammals, the energy density of fetal tissues approximately doubles during gestation, increasing pregnancy costs through time (Lockyer, 1987), and female energy consumption increases by nearly 150% during lactation (Oftedal, 1985; Worthy, 2001). Long-Term Trends There may also be lagged effects of seasonal prey availability on the body composition of Sarasota Bay dolphins, as dolphins eat fish with higher lipid content in the fall, perhaps in preparation for the winter (Waples, 1995). Prey quality and availability control the fatty acid composition and thickness of blubber in marine mammals (Kirch et al., 2000; Bradshaw et al., 2003; Miller et al., 2011), so the higher fat content of fish consumed in the fall may heavily contribute to increased body composition in the following season. Our results mirror those of other studies that have documented the importance of seasonal trends in blubber quality, quantity, and composition (Aguilar and Borrell, 1990; Samuel and Worthy, 2004; Noren and Wells, 2009). Although our results suggest strong seasonal patterns in body composition, the seasonal bias in the present dataset toward spring and summer Body composition values were highest in winter, which coincided with low water temperatures but also low prey availability (Figure 4). Winter metabolic rates are typically lower for both male and female dolphins in Sarasota Bay when compared to other seasons, which may decrease the need to forage during the colder months of winter, decreasing the effects of low prey availability (Costa et al., 1993). This is particularly true for females, which decrease feeding rates in winter (Waples, 1995). The decreased metabolic rate of dolphins in winter may aid in maintaining high body composition, allowing for June 2021 | Volume 8 | Article 615773 8 Bottlenose Dolphin Body Composition Adamczak et al. FIGURE 7 | Fitted generalized additive model with 95% confidence interval for long term trends in body composition for male and female common bottlenose dolphins (Tursiops truncatus) of all ages in Sarasota Bay between 1986 and 2016 (n = 480). Analyses were restricted to May and June to reduce effects of seasonal sampling bias. FIGURE 7 | Fitted generalized additive model with 95% confidence interval for long term trends in body composition for male and female common bottlenose dolphins (Tursiops truncatus) of all ages in Sarasota Bay between 1986 and 2016 (n = 480). Analyses were restricted to May and June to reduce effects of seasonal sampling bias. warrants further inquiry into the association between season and seasonally driven features, such as SST and CPUE, on the body composition of Sarasota Bay common bottlenose dolphins. In Sarasota Bay, young dolphins demonstrated sex-specific differences in resource allocation. Frontiers in Marine Science | www.frontiersin.org SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmars.2021. 615773/full#supplementary-material ETHICS STATEMENT The Research in Sarasota Bay was conducted under a series of National Marine Fisheries Service Scientific Research Permits, and annual Mote Marine Laboratory Institutional Animal Care and Use Committee approvals. Supplementary Table 2 | Chi-squared value, degrees of freedom, and p-value for fitted generalized linear mixed models with fixed effects of body length (length), monthly SST (SST), season, sex, and year from 1986 to 2016. Models were run for all individuals (n = 534), young individuals (less than 10 years old, n = 254) and mature individuals (n = 280, greater than or equal to 10 years old) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. FUNDING Primary long-term funding for this research has come from Dolphin Quest, Inc., Earthwatch Institute, and the National Oceanic and Atmospheric Administration. SA was supported by the E&P Sound and Marine Life Joint Industry Project of the International Association of Oil and Gas Producers. Body composition was highly variable across all ages of mature females, but was not predicted by reproductive status. Female dolphins in Sarasota Bay limit competition by specializing on different prey types, decreasing overlap in diets between females (Rossman et al., 2015). This prey specialization may influence a female’s body composition and lead to high variability in these values, as prey selection can influence the composition and morphology of blubber in dolphins (Smith and Worthy, 2006). DATA AVAILABILITY STATEMENT Supplementary Figure 1 | Relationship between age and body length for female and male (n = 260 and n = 274, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay fitted with Gompertz growth curve from parameters determined by Read et al. (1993). Supplementary Figure 1 | Relationship between age and body length for female and male (n = 260 and n = 274, respectively) common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay fitted with Gompertz growth curve from parameters determined by Read et al. (1993). The data analyzed in this study is subject to the following licenses/restrictions: The data used in these analyses are a part of a long-term monitoring project conducted by the Chicago Zoological Society’s Sarasota Dolphin Research Program and, as such, are not publicly available. Requests to access these datasets should be directed to RW. Supplementary Figure 2 | Annual trends in sea surface temperature (SST) in Sarasota Bay from 1986 to 2016 and CPUE in Sarasota Bay from 2004 to 2016. Annual averages were calculated from monthly SST data extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution. Supplementary Figure 2 | Annual trends in sea surface temperature (SST) in Sarasota Bay from 1986 to 2016 and CPUE in Sarasota Bay from 2004 to 2016. Annual averages were calculated from monthly SST data extracted from HadISST Average Sea Surface Temperature at 1◦spatial resolution. Supplementary Table 1 | Sample sizes, mean values, and standard error for body composition, defined as the ratio of blubber volume to total body volume, of common bottlenose dolphins in Sarasota Bay, Florida. Supplementary Table 1 | Sample sizes, mean values, and standard error for body composition, defined as the ratio of blubber volume to total body volume, of common bottlenose dolphins in Sarasota Bay, Florida. CONCLUSION Our results suggest that both SST and CPUE influence the body composition of Sarasota Bay dolphins. There is evidence that CPUE is a strong predictor of body composition, however, the long-term monitoring effort of the SDRP must continue to further elucidate the significance of these relationships. Further, this study suggests that young female and male dolphins in Sarasota Bay differentially allocate resources to growth and body composition. We found that reproductive state did not alter the body composition of female dolphins, providing further evidence of their status as true income breeders. This research was made possible by long-term monitoring of the resident dolphin community in Sarasota Bay, and highlights the importance of continuing this effort to better understand the factors influencing body composition. ACKNOWLEDGMENTS We would like to thank the many individuals—staff, students, trained volunteers, collaborators–involved in data collection for the SDRP over the decades. Additionally, we would like to acknowledge Dr. Lisa Schwarz for her initial efforts on this body composition project and Drs. Roxanne Beltran and Pete Raimondi for providing feedback on the statistical methods. Research in Sarasota Bay is conducted under a series of National Marine Fisheries Service Scientific Research Permits, annual Mote Marine Laboratory Institutional Animal Care and Use Committee approvals (06-10-DG1 to 19-10-RW2), and Florida Fish and Wildlife Conservation Commission fishing permits (Special Activity License numbers 032r-809 to 19-0809-SR). Long-Term Trends Despite the elevated costs associated with reproduction, we found no significant difference in body composition between the reproductive states of females (Figure 6). As income breeders, female dolphins continue to forage while lactating, which may offset any potential declines in body composition. Both pregnant and lactating female dolphins increase prey intake and nutritional content of prey, which likely allows them to maintain body composition despite the high costs associated with these reproductive states Ontogenetic patterns in body composition revealed that mature dolphins were leaner, and body composition was less variable as dolphins reached maturity (Figure 5). Mature dolphins are larger with lower surface area to volume ratios, which would improve their capacity to retain heat, reducing the need for thicker blubber (Worthy and Edwards, 1990; Noren and Wells, 2009; Adamczak et al., 2020). In contrast, as young dolphins are smaller, they have thicker, more insulative blubber to compensate for the additional thermoregulatory costs of being small (McLellan et al., 2002; Dunkin et al., 2005; Koopman, 2007; Montie et al., 2008; Noren and Wells, 2009). June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org Frontiers in Marine Science | www.frontiersin.org 9 Bottlenose Dolphin Body Composition Adamczak et al. (Bernard and Hohn, 1989; Reddy et al., 1991; Malinowski and Herzing, 2015). REFERENCES Gannon, D. P., Berens, E. J., Camilleri, S. A., Gannon, J. G., Brueggen, M. K., Barleycorn, A., et al. (2009). Effects of Karenia brevis harmful algal blooms on nearshore fish communities in southwest Florida. Mar. Ecol. Prog. Ser. 378, 171–186. Ackman, R. G., Hingley, J. H., Eaton, C. A., Sipos, J. C., and Mitchell, E. D. Ackman, R. G., Hingley, J. H., Eaton, C. A., Sipos, J. C., and Mitchell, E. D. (1975). Blubber fat deposition in mysteceti whales. Can. J. Zool. 53, 1332–1339. doi: 10.1139/z75-158 (1975). Blubber fat deposition in mysteceti whales. Can. J. Zool. 53, 1332–1339. doi: 10.1139/z75-158 Hastie, T., and Tibshirani, R. (1987). Generalized additive models: some applications. J. Am. Stat. Assoc. 82, 371–386. doi: 10.1080/01621459.1987. 10478440 Adamczak, S. K., Pabst, D. A., McLellan, W. A., and Thorne, L. H. (2020). Do bigger bodies require bigger radiators? Insights into thermal ecology from closely related marine mammal species and implications for ecogeographic rules. J. Biogeogr. 47, 1193–1206. doi: 10.1111/jbi.13796 Hohn, A. A., Scott, M. D., Wells, R. S., Sweeney, J. C., and Irvine, A. B. (1989). Growth layers in teeth from known-age, free-ranging bottlenose dolphins. Marine Mamm. Sci. 5, 315–342. rules. J. Biogeogr. 47, 1193–1206. doi: 10.1111/jbi.13796 Aguilar, A., and Borrell, A. (1990). Patterns of lipid content and stratification in the blubber of fin whales (Balaenoptera physalus). J. Mammal. 71, 544–554. doi: 10.2307/1381793 Hokkanen, J. E. I. (1990). Temperature regulation of marine mammals. J. Theor. Biol. 145, 465–485. Barton, K. (2009). Mu-MIn: Multi-Model Inference. R Package Version 0.12.2. K. (2009). Mu-MIn: Multi-Model Inference. R Package Version 0. Hückstädt, L. A., Holser, R. R., Tift, M. S., and Costa, D. P. (2018). The extra burden of motherhood: reduced dive duration associated with pregnancy status in a deep-diving marine mammal, the northern elephant seal. Royal Soc. B 14:4. doi: 10.1098/rsbl.2017.0722 Bates, D., MäHoer, M., Bolker, B., and Walker, S. (2015). Fitting linear mixed-effects models using lme4. J. Stat. Softw. 67, 1–48. doi: 10.18637/jss.v067.i01 Beck, C., Rea, L., Iverson, S., Kennish, J., Pitcher, K., and Fadely, B. (2007). Blubber fatty acid profiles reveal regional, seasonal, age-class and sex differences in the diet of young Steller sea lions in Alaska. Mar. Ecol. Prog. Ser. 338, 269–280. doi: 10.3354/meps338269 Kastelein, R. A., van der Sijs, S. J., Staal, C., and Nieuwstraten, S. H. (1997). “Blubber thickness in harbour porpoises (Phocoena phocoena),” in The Biology of the Harbour Porpoise, eds A. J. Read, P. REFERENCES Body fat condition in Northeast Atlantic fin whales, Balaenoptera physalus, and its relationship with reproduction and food source. Can. J. Fish. Aquat. Sci. 43, 142–147. Breslow, N. E., and Clayton, D. G. (1993). Approximate inference in generalized linear mixed models. J. Am. Stat. Assoc. 88, 9–25. doi: 10.2307/2290687 Burnham, K. P., and Anderson, D. R. (1994). Model Selection and Multi-Model Inference: A Practical Information-Theoretic Approach. New York, NY: Springer Publishing. Lockyer, C. (1987). The relationship between body fat, food resource and reproductive energy costs in North Atlantic fin whales (Balaeoptera physalus). Symp. Zool. Soc. Lond. 57, 343–361. Lockyer, C. H., McConnell, L. C., and Waters, T. D. (1985). Body condition in terms of anatomical and biochemical assessment of body fat in North Atlantic fin and sei whales. Can. J. Zool. 63, 2328–2338. doi: 10.1139/z85-345 Calzada, N., Aguilar, A., Grau, E., and Lockyer, C. (1997). Patterns of growth and physical maturity in the western Mediterranean striped dolphin, Stenella coeruleoalba (Cetacea: Odontoceti). Can. J. Zool. 75, 632–637. Cockroft, V. G., and Ross, G. J. B. (1990). Age, growth and reproduction in bottlenose dolphins from the east coast of southern Africa. Fish. Bull. 88, 289–302. Malinowski, C. R., and Herzing, D. L. (2015). Prey use and nutritional differences between reproductive states and age classes in Atlantic spotted dolphins (Stenella frontalis) in the Bahamas. Mar. Mamm. Sci. 31, 1471–1493. doi: 10. 1111/mms.12238 Costa, D. P., Worthy, G. A. J., Wells, R. S., Read, A. J., Scott, M. D., Irvine, A. B., et al. (1993). “Seasonal changes in the field metabolic rate of bottlenose dolphins Tursiops truncatus,” in Proceedings of the Tenth Biennial Conference on the Biology of Marine Mammals, (Galveston, TX). Mallette, S. D., McLellan, W. A., Scharf, F. S., Koopman, H. N., Barco, S. G., Wells, R. S., et al. (2016). Ontogenetic allometry and body composition of the common bottlenose dolphin (Tursiops truncatus) from the U.S. mid-Atlantic. Mar. Mamm. Sci. 32, 86–121. doi: 10.1111/mms.12253 Dunkin, R. C., McLellan, W. A., Blum, J. E., and Pabst, D. A. (2005). The ontogenetic changes in the thermal properties of blubber from Atlantic bottlenose dolphin Tursiops truncatus. J. Exp. Biol. 208, 1469–1480. doi: 10. 1242/jeb.01559 Mauck, B., Bilgmann K, Jones DD, Eysel U, Dehnhardt G. (2003). Thermal windows on the trunk of hauled-out seals: hot spots for thermoregulatory evaporation? J. Exp. Biol. 206, 1727–1738. doi: 10.1242/jeb.00348 j Dunkin, R. C., McLellan, W. A., Blum, J. REFERENCES R. Wiepkema, and P. E. Nachtigall (Woerden: De Spil Publishers), 179–199. Beck, G. G., Smith, T. G., and Hammill, M. O. (1993). Evaluation of body condition in the Northwest Atlantic harp seal (Phoca groenlandica). Can. J. Fish. Aquat. Sci. 50, 1372–1381. doi: 10.1139/f93-157 Kirch, P. E., Iverson, S. J., and Bowen, W. D. (2000). Effect of a low-fat diet on body composition and blubber fatty acids of captive juvenile harp seals (Phoca groenlandica). Physiol. Biochem. Zool. 73, 45–58. doi: 10.1086/316723 Beentjes, M. P. (2006). Behavioral thermoregulation of the New Zealand Sea Lion (Phocarctos hookeri). Mar. Mamm. Sci. 22, 311–325. doi: 10.1111/j.1748-7692. 2006.00022.x Koopman, H. N. (2007). Phylogenetic, ecological, and ontogenetic factors influencing the biochemical structure of the blubber of odontocetes. Mar. Biol. 151, 277–291. doi: 10.1007/s00227-006-0489-8 Berens McCabe, E. J., Gannon, D. P., Barros, N. B., and Wells, R. S. (2010). Prey selection by resident common bottlenose dolphins (Tursiops truncatus) in Sarasota Bay, Florida. Mar. Biol. 157, 931–942. doi: 10.1007/s00227-009-1371- 2 Koopman, H. N., Iverson, S. J., and Gaskin, D. E. (1996). Stratification and age-related differences in blubber fatty acids of the male harbour porpoise (Phocoena phocoena). J. Comp. Physiol. B 165, 628–639. doi: 10.1007/ BF00301131 Bernard, H. J., and Hohn, A. A. (1989). Differences in feeding habits between pregnant and lactating spotted dolphins (Stenella attenuata). J. Mammal. 70, 211–215. doi: 10.2307/1381693 Koopman, H. N., Pabst, D. A., McLellan, W. A., Dillaman, R. M., and Read, A. J. (2002). Changes in blubber distribution and morphology associated with starvation in the harbor porpoise (Phocoena phocoena): evidence for regional references in blubber structure and function. Physiol. Biochem. Zool. 75, 498– 512. Bradshaw, C. J. A., Hindell, M. A., Best, N. J., Phillips, K. L., Wilson, G., and Nichols, P. D. (2003). You are what you eat: describing the foraging ecology of southern elephant seals (Mirounga leonina) using blubber fatty acids. Proc. R. Soc. Lond. B 270, 1283–1292. doi: 10.1098/rspb.2003.2371 Lluch-Cota, S. E., Tripp-Valdez, M., Lluch-Cota, D. B., Bautista-Romero, J. J., Lluch-Belda, D., Verbesselt, J., et al. (2013). Recent trends in sea surface temperature offMexico. Atmósfera 26, 537–546. doi: 10.1016/S0187-6236(13) 71094-4 Braithwaite, J. E., Meeuwig, J. J., Letessier, T. B., Jenner, K. C. S., and Brierley, A. S. (2015). From sea ice to blubber: linking whale condition to krill abundance using historical whaling records. Polar Biol. 38, 1195–1202. doi: 10.1007/ s00300-015-1685-0 Lockyer, C. (1986). AUTHOR CONTRIBUTIONS RW and DC conceived this study. RW and EB collected the morphometrics and blubber depth data used here. EB led the prey fish surveys. RH conceptualized and carried out the methodology necessary to convert the raw data to body composition. SA performed statistical analyses and drafted the manuscript. RW, DC, EB, and RH provided feedback on statistical analyses and manuscript drafts. All authors have read and approved the final manuscript. Supplementary Table 3 | Chi-squared value, degrees of freedom, and p-value for fitted generalized linear mixed models with fixed effects of body length (length), catch per unit effort (CPUE), monthly SST (SST), season, sex, and year from 2004 to 2016. Models were run for all individuals (n = 157), young individuals (n = 73, less than 10 years old) and mature individuals (n = 84, greater than or equal to 10 years old) and repeated measures for individuals were treated as random effects. Models were selected using Akaike’s Information Criterion correction. Monthly average sea surface temperature (SST) data were extracted using the HadISST Average Sea Surface Temperature dataset. June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 10 Bottlenose Dolphin Body Composition Adamczak et al. REFERENCES 68, 1962–1966. doi: 10.1139/z90-276 West, K. L., Oftedal, O. T., Carpenter, J. R., Krames, B. J., Campbell, M., and Sweeney, J. C. (2007). Effect of lactation stage and concurrent pregnancy on milk composition in the bottlenose dolphin. J. Zool. 273, 148–160. doi: 10.1111/ j.1469-7998.2007.00309.x Read, A. J., Wells, R. S., Hohn, A. A., and Scott, M. D. (1993). Patterns of growth in wild bottlenose dolphins, Tursiops truncatus. J. Zool. 231, 107–123. doi: 10.1111/j.1469-7998.1993.tb05356.x Williams, T. M., and Friedl, W. A. (1990). Heat flow properties of dolphin blubber: insulating warm bodies in cold water. Am. Zool. 30:191. Reddy, M., Kamolnick, T., Skaar, D., Curry, C., and Ridgway, S. (1991). “Bottlenose dolphins: Energy consumption during pregnancy, lactation, and growth,” in Proceedings of the International Marine Animal Trainer’s Association Conference Proceedings 1991, (San Diego, CA), 30–37. Worthy, G. A. J. (2001). “Nutrition and energetics,” in CRC Handbook of Marine Mammal Medicine, second Edn, eds L. A. Dierauf and F. M. D. Gulland (Boca Raton, FL: CRC Press), 791–827. Worthy, G. A. J., and Edwards, E. F. (1990). Morphometric and biochemical factors affecting heat loss in a small temperate cetacean (Phocoena phocoena) and a small tropical cetacean (Stenella attenuata). Physiol. Zool. 63, 432–442. Rossman, S., McCabe, E. B., Barros, N. B., Gandhi, H., Ostrom, P. H., Stricker, C. A., et al. (2015). Foraging habits in a generalist predator: sex and age influence habitat selection and resource use among bottlenose dolphins (Tursiops truncatus). Mar. Mamm. Sci. 31, 155–168. doi: 10.1111/mms.12143 Yasui, W. Y., and Gaskin, D. E. (1986). Energy budget of a small cetacean, the harbour porpoise, Phocoena phocoena (L.). Ophelia 25, 183–197. doi: 10.1080/ 00785326.1986.10429749 Ryg, M., Lydersen, C., Knutsen, L. Ø, Bjørge, A., Smith, T. G., and Øritsland, N. A. (1993). Scaling of insulation in seals and whales. J. Zool. 230, 193–206. doi: 10.1111/j.1469-7998.1993.tb02682.x Ryg, M., Smith, T. G., and Øritsland, N. A. (1990). Seasonal changes in body mass and body composition of ringed seals (Phoca hispida) on Svalbard. Can. J. Zool. 68, 470–475. doi: 10.1139/z90-069 Conflict of Interest: The authors declare that this study received funding from Dolphin Quest, Inc. The funder had the following involvement in the study: They provided some of the funding support for the health assessments, and one of the principals in the company, Dr. Jay Sweeney, as one of the most experienced marine mammal veterinarians in the world, collected some of the samples. Samuel, A. REFERENCES E., and Pabst, D. A. (2010). The buoyancy of the integument of Atlantic bottlenose dolphins (Tursiops truncatus): effects of growth, reproduction, and nutritional state. Mar. Mamm. Sci. 26, 573–587. doi: 10.1111/j.1748-7692.2009.00353.x McKnight, P. E., and Najab, J. (2010). “Kruskal-wallis test,” in The Corsini Encyclopedia of Psychology, eds W. E. Craighead and C. Nemeroff(Haboken: John Wiley & Sons), doi: 10.1002/9780470479216.corpsy0491 McLellan, W. A., Koopman, H. N., Rommel, S. A., Read, A. J., Potter, C. W., Nicolas, J. R., et al. (2002). Ontogenetic allometry and body composition of harbour porpoises (Phocoena phocoena, L.) from the western North Atlantic. J. Zool. 257, 457–471. doi: 10.1017/S0952836902001061 Fedak, M. A., and Anderson, S. S. (1982). The energetics of lactation: accurate measurements from a large wild mammal, the grey seal (Halichoerus grypus). J. Zool. 198, 473–479. doi: 10.1111/jzo.1982.198.4.473 Gales, N. J., and Burton, H. R. (1987). Ultrasonic measurement of blubber thickness of the southern elephant seal, Mirounga-Leonina (Linn). Austr. J. Zool. 35, 207–217. doi: 10.1071/zo9870207 Meagher, E. M., McLellan, W. A., Westgate, A. J., Wells, R. S., Blum, J. E., and Pabst, D. A. (2008). Seasonal patterns of heat loss in wild bottlenose dolphins (Tursiops truncatus). J. Comp. Physiol. B 178, 529–543. doi: 10.1007/s00360-007-0245-5 June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 11 Adamczak et al. Bottlenose Dolphin Body Composition Miller, C., Reeb, D., Best, P., Knowlton, A., Brown, M., and Moore, M. (2011). Blubber thickness in right whales Eubalaena glacialis and Eubalaena australis related with reproduction, life history status and prey abundance. Mar. Ecol. Ser. 438, 267–283. doi: 10.3354/meps09174 Waples, D. M. (1995). Activity Budgets of Free-Ranging Bottlenose Dolphins (Tursiops truncatus) in Sarasota Bay, Florida. Master’s of Science thesis. Santa Cruz: University of California. Wells, R. S. (2009). Learning from nature: bottlenose dolphin care and husbandry. Zoo Biol. 28, 635–651. Montie, E. W., Garvin, S. R., Fair, P. A., Bossart, G. D., Mitchum, G. B., McFee, W. E., et al. (2008). Blubber morphology in wild bottlenose dolphins (Tursiops truncatus) from the Southeastern United States: influence of geographic location, age class, and reproductive state. J. Morphol. 269, 496–511. doi: 10. 1002/jmor.10602 Wells, R. S. (2010). Feeling the heat – potential climate change impacts on bottlenose dolphins. Whalewatcher 39, 12–17. Wells, R. S. (2014). REFERENCES “Social structure and life history of common bottlenose dolphins near Sarasota Bay, Florida: Insights from four decades and five generations,” in Primates and Cetaceans: Field Research and Conservation of Complex Mammalian Societies, Primatology Monographs, eds J. Yamagiwa and L. Karczmarski (Tokyo: Springer), 149–172. doi: 10.1007/978-4-431-54523-1_8 Murphy, S., Collet, A., and Rogan, E. (2005). Mating strategy in the male common dolphin (Delphinus delphis): what gonadal analysis tells us. J. Mammal. 86, 1247–1258. Mammalian Societies, Primatology Monographs, eds J. Yamagiwa a Noren, S. R., and Wells, R. S. (2009). Blubber deposition during ontogeny in free-ranging bottlenose dolphins: balancing disparate roles of insulation and locomotion. J. Mammal. 90, 629–637. doi: 10.1644/08-MAMM-A-138R.1 Wells, R. S., McHugh, K. A., Douglas, D. C., Shippee, S., Berens McCabe, E. J., Barros, N. B., et al. (2013). Evaluation of potential protective factors against metabolic syndrome in bottlenose dolphins: feeding and activity patterns of dolphin in Sarasota Bay, Florida. Front. Endocrinol. 4:16. doi: 10.3389/fendo. 2013.00139 Oftedal, O. T. (1985). “Pregnancy and lactation,” in Bioenergetics of Wild Herbivores, eds R. J. Hudson and R. G. White (Boca Raton, FL: CRC Press), 215–238. Wells, R. S., Rhinehart, H. L., Hansen, L. J., Sweeney, J. C., Townsend, F. I., Stone, R., et al. (2004). Bottlenose dolphins as marine ecosystem sentinels: developing a health monitoring system. EcoHealth 1, 246–254. Øigård, T. A., Lindstrøm, U., Haug, T., Nilssen, K. T., and Smout, S. (2013). Functional relationship between harp seal body condition and available prey Øigård, T. A., Lindstrøm, U., Haug, T., Nilssen, K. T., and Smout, S. (2013). Functional relationship between harp seal body condition and available prey in the Barents Sea. Mar. Ecol. Prog. Ser. 484, 287–301. doi: 10.3354/meps10272 Wells, R. S., Scott, M. D., and Irvine, A. B. (1987). “The social structure of free- ranging bottlenose dolphins,” in Current Mammalogy, Vol. 1, ed. H. Genoways (New York, NY: Plenum Press), 247–305. Parry, D. A. (1949). The structure of whale blubber, and a discussion of its thermal properties. J. Cell Sci. 3, 13–25. West, G. C., Burns, J. J., and Modafferi, M. (1979). Fatty acid composition of blubber from the four species of Bering Sea phocid seals. Can. J. Zool. 57, 189–195. doi: 10.1139/z79-017 R Core Team (2020). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundatin for Statistical Computing. Available online at: http.//www.R-project.org/. Read, A. J. (1990). Estimation of body condition in harbour porpoises, Phocoena phocoena. Can. J. Zool. Frontiers in Marine Science | www.frontiersin.org REFERENCES M., and Worthy, G. A. J. (2004). Variability in fatty acid composition of bottlenose dolphin (Tursiops truncatus) blubber as a function of body site, season, and reproductive state. Can. J. Zool. 82, 1933–1942. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Smith, H. R., and Worthy, G. A. J. (2006). Stratification and intra- and inter- specific differences in fatty acid composition of common dolphin (Delphinus sp.) blubber: implications for dietary analysis. Comp. Biochem. Physiol. B 143, 486–499. Copyright © 2021 Adamczak, Holser, Costa, Berens McCabe and Wells. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Struntz, D. J., McLellan, W. A., Dillaman, R. M., Blum, J. E., Kucklick, J. R., and Pabst, D. A. (2004). Blubber development in bottlenose dolphins (Tursiops truncatus). J. Morphol. 259, 7–20. doi: 10.1002/jmor.10154 Wagenmakers, E. J., and Farrell, S. (2004). AIC model selection using Akaike weights. Psychon. Bull. Rev. 11, 192–196. doi: 10.3758/BF0320 6482 June 2021 | Volume 8 | Article 615773 Frontiers in Marine Science | www.frontiersin.org 12
https://openalex.org/W3207354914
https://link.springer.com/content/pdf/10.1140/epjc/s10052-022-10365-2.pdf
English
null
Tsallis holographic dark energy reconsidered
European physical journal. C, Particles and fields
2,022
cc-by
13,626
Tsallis holographic dark energy reconsidered M. Dheepikaa, Titus K Mathewb Department of Physics, Cochin University of Science and Technology, Kochi, Kerala 682022, India Received: 5 November 2021 / Accepted: 22 April 2022 / Published online: 3 May 2022 © The Author(s) 2022 Abstract We consider the interacting Tsallis Holographic Dark Energy (THDE), with the Granda–Oliveros (GO) scale as the infrared (IR) cutoff, as dynamical vacuum. We ana- lytically solved for the Hubble parameter, in a spatially flat FLRW universe with dark energy and matter as components, and the solution traces the evolutionary path from the prior deceleratedtothelateacceleratedepoch.Withoutinteraction, the model predicts a ΛCDM like behavior with an effective cosmological constant. We used Pantheon Supernovae type Ia, observational Hubble data (OHD), cosmic microwave background (CMB), and baryon acoustic oscillation (BAO) data to constrain the free parameters of the model. The esti- mated values of the cosmological parameters were consistent with observational results. We analyzed the behavior of the model using the statefinder and ω′ e −ωe plane where ωe and ω′ e corresponds to the effective equation of state and its evolu- tion, respectively. The model shows a quintessence behavior in general, and the model trajectory ends in a point that cor- responds to the de Sitter phase. We performed a dynamical analysis of the model, concluding that the prior decelerated and late accelerated phases are unstable and stable equilib- ria, respectively. We also investigated the thermodynamical nature of the model and found that the generalized second law remains valid in the dynamical vacuum treatment of the model. to unwind the mystery of the universe’s accelerated expan- sion. One of the simplest and best dark energy candidates is the cosmological constant, Λ as given in the most successful model,ΛCDM.However,thismodelsuffersmainlyfromtwo problems; one is the cosmological constant problem where there is a discrepancy in the observational and the theoret- ical value (10121 times larger) of the energy density of the cosmological constant. The other is the coincidence problem that is the unknown reason for the coincidence of matter den- sity and the cosmological constant during the current epoch, despite their different evolutionary nature. Dynamical dark energy models were suggested to mitigate these problems. Some representatives of this category are quintessence [3,4], k-essence [5], and phantom [6] models. One of the dynamical dark energy models is the holo- graphic dark energy (HDE) model, which pivots on the black hole thermodynamics, proposed mainly to allay the coinci- dence problem. Tsallis holographic dark energy reconsidered The underlying idea of these models is the holographic principle [7,8] which states that the entropy of a gravitating system is related to its surface area, not its vol- ume, which implies that the maximum entropy of any region of space should be less than or equal to the entropy of a black hole of similar size. The limit set by this principle implies a relation between the short distance (ultraviolet-UV) cutoff and the long distance (infrared-IR) cutoff [9]. This idea rec- onciles the breakdown of the quantum field theory to describe a black hole [9,10]. The initial proposal of the holographic principle considers the entropy of the cosmological horizon as the Bekenstein-Hawking entropy and the dark energy den- sity scales as the square of the Hubble parameter. However, it could not explain the current accelerated expansion of the universe [11,12]. HDE model with particle horizon [13,14] as IR cutoff also could not explain the present acceleration. Taking future event horizon [14] as IR cutoff successfully explained the present accelerated expansion, but it suffered from causality problem. Since any known symmetry does not dodge the interaction between the dark sectors [15], mod- els considering such interactions were proposed, and they a e-mail: mdheepika@cusat.ac.in b e-mail: titus@cusat.ac.in (corresponding author) Eur. Phys. J. C (2022) 82:399 https://doi.org/10.1140/epjc/s10052-022-10365-2 Regular Article - Theoretical Physics 1 Introduction Comparison of THDE model with other HDE models using statefinder analysis was reviewed in [72]. The evolution of cosmological perturbations in THDE models with Hubble horizon and future event horizon as IR cutoff and Bayesian model comparison with ΛCDM as reference model has been scrutinized in [73–75]. In [76–93], the THDE model with dif- ferent IR cutoffs within various modified gravity theories and scalar field theories has also been explored. The growth rate of clustering for different IR cutoffs for the THDE model in the FRW universe can be found in [94]. Cosmological implications through non linear interactions between THDE with Hubble horizon as IR cutoff and cold dark matter in the framework of loop quantum cosmology has been discussed in [95]. Investigations on the THDE model with Hubble hori- zon as IR cutoff in the higher derivative theory of gravity in [96] show that it is not compatible with late time acceleration as it could not acquire the required value of the equation of state parameter. Due to the quantified non extensivity in Tsal- lis entropy, studies on modification of Friedmann equation and gravity theory, including emergence proposal of gravity [97–102], also flourish in this field. All these works were car- ried out considering the equation of state parameter of dark energy as varying with the expansion of the universe. Holographic Ricci dark energy model with Ricci scalar curvature as IR cutoff introduced in [20] is a phenomenolog- ically viable model which avoids causality and coincidence problems. Inspired by this model and on a pure dimensional basis, Granda and Oliveros proposed a new IR cutoff [12], a combined function of Hubble parameter and its time deriva- tive. The resulting model also successfully avoids the coin- cidence problem, causality problem and explains the current accelerated expansion. Further explorations in HDE models with different IR cutoffs can be found in [21–25]. In order to incorporate the quantum corrections [26–35], conventional HDE required modifications. q Tsallis and Cirto [36] introduced a generalized non addi- tive entropy, popularly known as Tsallis entropy, to solve the thermodynamic inconsistencies in non standard systems like a black hole. The pioneer works on the analysis of dark energy models with Tsallis non extensive statistical formu- lation can be found in [37] and further possibilities in cos- mology are probed in [38]. 1 Introduction After the discovery of the accelerated expansion of the uni- verse [1,2], the desire to understand the universe spired, resulting in intensive works in recent literature. The nature and origin of the accelerated expansion of the universe are a mystery even now. Postulating dark energy models and modified theories of gravity are the two approaches that try 12 3 3 Eur. Phys. J. C (2022) 82 :399 399 Page 2 of 14 399 Page 2 of 14 cutoff in Rastall framework and on Randall–Sundrum brane has been considered in [51,52]. Dynamical system studies on interacting and non interacting THDE in a fractal uni- verse with Hubble radius and apparent horizon as IR cutoff can be found in [53–56]. The equivalence between Tsallis entropic dark energy and generalized HDE with cutoffs in terms of particle horizon, future horizon, and its derivatives are established in [57]. Cosmological analysis of the THDE with Hubble horizon as IR cutoff in the axially symmetric Bianchi-I universe within the framework of general relativ- ity has been explained in [58,59]. Sign changeable mutual interactions between dark sectors are also considered to study the effects of anisotropy in the Bianchi universe [60]. Similar analysis of THDE with Hubble horizon and GO scale as IR cutoff in Bianchi-III universe has been discussed in [61,62]. Investigations on dynamics of THDE with Hubble horizon as IR cutoff, by assuming power law-exponential form for the scale factor have been studied in [63]. Geometrical evolu- tionary studies in THDE models with Hubble horizon, future event horizon, and GO scale as IR cutoffs corresponding to different interactions have been explored in [64–70]. Investi- gations on THDE with GO scale as IR cutoff [71], presuming that this energy density is responsible for inflation, show the potentiality of the model in explaining the early universe. Comparison of THDE model with other HDE models using statefinder analysis was reviewed in [72]. The evolution of cosmological perturbations in THDE models with Hubble horizon and future event horizon as IR cutoff and Bayesian model comparison with ΛCDM as reference model has been scrutinized in [73–75]. In [76–93], the THDE model with dif- ferent IR cutoffs within various modified gravity theories and scalar field theories has also been explored. The growth rate of clustering for different IR cutoffs for the THDE model in the FRW universe can be found in [94]. 1 Introduction Cosmological implications through non linear interactions between THDE with Hubble horizon as IR cutoff and cold dark matter in the framework of loop quantum cosmology has been discussed in [95]. Investigations on the THDE model with Hubble hori- zon as IR cutoff in the higher derivative theory of gravity in [96] show that it is not compatible with late time acceleration as it could not acquire the required value of the equation of state parameter. Due to the quantified non extensivity in Tsal- lis entropy, studies on modification of Friedmann equation and gravity theory, including emergence proposal of gravity [97–102], also flourish in this field. All these works were car- ried out considering the equation of state parameter of dark energy as varying with the expansion of the universe. It is to be noted that, in formulating the HDE density, one has to compare the UV cutoff, corresponding to the vac- ith th IR t ff ti th l l th yield better consistency with the cosmic observations than non interacting models [16–19]. cutoff in Rastall framework and on Randall–Sundrum brane has been considered in [51,52]. Dynamical system studies on interacting and non interacting THDE in a fractal uni- verse with Hubble radius and apparent horizon as IR cutoff can be found in [53–56]. The equivalence between Tsallis entropic dark energy and generalized HDE with cutoffs in terms of particle horizon, future horizon, and its derivatives are established in [57]. Cosmological analysis of the THDE with Hubble horizon as IR cutoff in the axially symmetric Bianchi-I universe within the framework of general relativ- ity has been explained in [58,59]. Sign changeable mutual interactions between dark sectors are also considered to study the effects of anisotropy in the Bianchi universe [60]. Similar analysis of THDE with Hubble horizon and GO scale as IR cutoff in Bianchi-III universe has been discussed in [61,62]. Investigations on dynamics of THDE with Hubble horizon as IR cutoff, by assuming power law-exponential form for the scale factor have been studied in [63]. Geometrical evolu- tionary studies in THDE models with Hubble horizon, future event horizon, and GO scale as IR cutoffs corresponding to different interactions have been explored in [64–70]. Investi- gations on THDE with GO scale as IR cutoff [71], presuming that this energy density is responsible for inflation, show the potentiality of the model in explaining the early universe. 1 Introduction As a result, reconsidering the THDE as the dynamical vacuum in analyzing the evolutional history of the universe is of practical significance. We are considering the THDE as the dynamical vacuum in the present work. We study the evolution of the universe by considering the inter- action, between dark matter and dark energy in conformity with the total conservation of energy, in the THDE model with the GO scale as IR cutoff. Our analysis shows that the model predicts a transition to the late accelerating universe. This work also involves the geometrical and dynamical anal- ysis and thermodynamical study of the model to check the feasibility of explaining the accelerating universe. defined as, ρde = CL2δ−4, (3) (3) ρde = CL2δ−4, where the constant, C = γ (4π)δ with dimension [L2−2δ] (in units of 8πG = ¯h = c = 1). The simplest choice for scale is L = H−1, the Hubble horizon. In [40], authors have estab- lished that the corresponding model of the universe, with non interacting dark sectors, can show a transition into the late time accelerated epoch. This contrasts with the conventional HDE model, which failed to predict a transition into the late accelerated epoch if cosmic components are non interacting. In the present study we adopt the GO scale as IR cutoff, which was originally proposed in reference [12] to study the conventional HDE model and is given by, Thestructureofthispaperisasfollows.Inthenextsection, we present an interacting THDE as a dynamical vacuum. We analytically solve for the Hubble parameter and investigate its evolutionary behavior. In Sect. 3, we constrain the param- eters with observational data and discuss its cosmological implications. Along with that, we also analyze the evolution- ary trajectory of the model in geometrical plane r −s and phase plane ω′ e−ωe plane. In Sect. 4, we perform the dynam- ical analysis on the interacting THDE model. In Sect. 5, we study the thermodynamical properties of the model. In the last section, we summarize the conclusions of the work. L−2 = (αH2 + β ˙H), (4) (4) where α and β are unknown dimensionless constants and ˙H, is the derivative of Hubble parameter with respect to the cosmic time. 1 Introduction Using (4) in (3), THDE density can be written as ρde = 3(α′H2 + β′ ˙H)2−δ, (5) where α′ = α 3 C 1 2−δ and β′ = β 3 C 1 2−δ have dimension [L 2−2δ 2−δ ]. ρde = 3(α′H2 + β′ ˙H)2−δ, (5) 1 1 (5) The Friedmann equation for the flat FRW universe is given by by 3H2 = ρm + ρde. (6) (6) 2 Interacting THDE model as dynamical vacuum Since the THDE is considered as a dynamical vacuum, its equation of state is Pde = −ρde and the matter is considered as pressureless. Considering the above assumptions, the equations in (7) reduces to S = γ Aδ, (1) (1) S = γ Aδ, where A is the horizon area of the black hole, γ is a positive [97] constant, and δ is the positive non additive parameter [36]. This will reduces to the Bekenstein entropy for γ = 1 4L2p 2 where A is the horizon area of the black hole, γ is a positive [97] constant, and δ is the positive non additive parameter [36] This will reduces to the Bekenstein entropy for γ 1 where A is the horizon area of the black hole, γ is a positive [97] constant, and δ is the positive non additive parameter [36]. This will reduces to the Bekenstein entropy for γ = 1 4L2p and δ = 1, with L2 p as the Planck length. Following the holographic principle, Cohen et al. [9] have found a relation between the entropy, IR cutoff (L), and the UV cutoff (Λ) as, where A is the horizon area of the black hole, γ is a positive [97] constant, and δ is the positive non additive parameter [36]. This will reduces to the Bekenstein entropy for γ = 1 4L2p and δ = 1, with L2 p as the Planck length. Following the holographic principle, Cohen et al. [9] have found a relation between the entropy, IR cutoff (L), and the UV cutoff (Λ) as, p p [36]. This will reduces to the Bekenstein entropy for γ = 1 4L2p and δ = 1, with L2 p as the Planck length. Following the holographic principle, Cohen et al. [9] have found a relation between the entropy, IR cutoff (L), and the UV cutoff (Λ) as, L3Λ3 ≤S3/4. (2) L3Λ3 ≤S3/4. (2) (2) Following the Tsallis entropy in (1) and substituting for area, A = 4π L2, leads to the relation Λ4 ≤γ (4π)δL2δ−4, and it gives a measure of the vacuum energy density. Taking con- sideration of the equality in this relation, a modified energy density, known as the Tsallis HDE (THDE) density, can be ˙ρde = −3bHρm, ˙ρm = −3(1 −b)Hρm. 1 Introduction This kind of entropy agrees well with the Friedmann equations and Padmanabhan’s proposal of the emergence of space time [39]. Like in the conventional HDE model, it is possible to construct dark energy models using Tsallis entropy, and as a result, Tsallis holographic dark energy (THDE) with Hubble horizon as IR cutoff was intro- duced in [40]. Taking inspiration from the aforementioned study, dynamics of FRW universe having dark matter and THDE with the apparent horizon, the particle horizon, the Ricci scalar curvature scale, and the Granda–Oliveros (GO) scale as IR cutoffs was studied considering non interacting and interacting scenarios [41–44]. It is found that the THDE model with particle horizon as IR cutoff explains the current accelerated expansion of the universe, unlike the correspond- ingconventionalHDEmodel.Theresultsfrom[41]showthat the THDE model is not always stable for the GO scale and the Ricci scalar cutoffs in both interacting and non interacting cases. Whereas in [42] THDE model with the GO scale as IR cutoff shows stability in (n + 1) dimensional FRW universe. Thermodynamical stability studies of THDE with the appar- ent horizon as IR cutoff in [45] shows that the model does not satisfy the stability conditions in both interacting and non interacting cases. The investigations on the evolution of the THDE with Hubble horizon as IR cutoff, by considering time varying deceleration parameter in FRW universe is discussed in [46], in Brans–Dicke cosmology is discussed in [47,48]. Geometrical diagnosis of THDE model of the universe with the apparent horizon as IR cutoff, considering the interaction between dark sectors of the universe, was made in [49]. Cos- mological model in higher dimensional Kaluza–Klien the- ory, having THDE with Hubble horizon as IR cutoff, and with Generalized Chaplygin Gas (GCG) as cosmic compo- nents are studied in [50]. THDE with Hubble horizon as IR It is to be noted that, in formulating the HDE density, one has to compare the UV cutoff, corresponding to the vac- uum energy, with the IR cutoff, representing the large length scale of the universe. There is a broad consensus that the cosmological constant in the standard ΛCDM model can be the vacuum energy, having an equation of state param- 123 123 Page 3 of 14 399 Eur. Phys. J. C (2022) 82 :399 defined as, eter, −1. Hence, the UV cutoff involved in the HDE models indicates that the corresponding dark energy density is of dynamical vacuum. 2 Interacting THDE model as dynamical vacuum where ρm and ρde is the dark matter and dark energy den- sity respectively. The conservation equations including the interaction between THDE and dark matter are given by where ρm and ρde is the dark matter and dark energy den- sity respectively. The conservation equations including the interaction between THDE and dark matter are given by A generalization of the Boltzmann–Gibbs (BG) theory, now known as the non extensive statistical mechanics, was pro- posed [32] to address the complexities in non standard sys- tems. For large scale systems, the thermodynamical entropy must be modified to non additive entropy [36]. According to Tsallis and Cirto [36], the quantum correction modified the entropy area relation as, ˙ρde + 3H(ρde + Pde) = −Q, ˙ρm + 3H(ρm + Pm) = Q, (7) (7) where ˙ρde and ˙ρm are the derivatives of dark energy and dark matter densities with respect to the cosmic time, Pde and Pm are the pressure of dark energy and matter respec- tively and Q represents the interaction, which determines the rate of exchange of energy between the dark sectors. From Eq. (7) it is clear that Q has to be a function of energy den- sity and inverse of time. We are adopting a simple function, Q = 3bHρm where b is the coupling constant. None of the previous works has considered this form of interaction in combination with the GO scale in studying the evolution of the FLRW universe. Since the THDE is considered as a dynamical vacuum, its equation of state is Pde = −ρde and the matter is considered as pressureless. Considering the above assumptions, the equations in (7) reduces to where ˙ρde and ˙ρm are the derivatives of dark energy and dark matter densities with respect to the cosmic time, Pde and Pm are the pressure of dark energy and matter respec- tively and Q represents the interaction, which determines the rate of exchange of energy between the dark sectors. From Eq. (7) it is clear that Q has to be a function of energy den- sity and inverse of time. We are adopting a simple function, Q = 3bHρm where b is the coupling constant. None of the previous works has considered this form of interaction in combination with the GO scale in studying the evolution of the FLRW universe. Hubble parameter and x = ln a as Hubble parameter and x = ln a as dΩde dx = −3bΩm, (9) dΩm dx = −3(1 −b)Ωm. (10) (9) ˜Ωm0 = 2 3β′ ⎡ ⎣α′ −  1 −Ωm0 H2(1−δ) 0  1 2−δ ⎤ ⎦, ˜Ωde0 = 1 −˜Ωm0. (17) (10) (17) The solution of Eq. (10) is Ωm = Ωm0e−3(1−b)x where Ωm0 = ρm0 3H2 0 is the present matter density parameter. In sim- ilar manner, Ωde0 = ρde0 3H2 0 is the present dark energy density parameter. The Eq. (6) reduces to, 1 = Ωm0 + Ωde0, for H = H0. Considering this result and using Eqs. (10), (9) and (6), a second order differential equation can be formu- lated as below This shows that, in the absence of interaction, the present model is similar to that of the standard ΛCDM, with an effective cosmological constant corresponding to the mass density parameter ˜Ωde0. Even though the dark energy den- sity in Eq. (5) is varying with H, the behavior of it as effective vacuum energy with equation of state, ωde = −1 is the pri- mary reason for this ΛCDM like behavior, in the absence of the interaction. Further, the coupled conservation equations in (7) become independent conservation, in which matter and dark energy are separately conserved. Under such a condi- tion, the dark energy density, equivalent to the vacuum energy density, will effectively become a constant. In contrast, the previous works on the non interacting THDE models [40– 42,44–46,66,69,72,73,105–107] shows quintessence like, phantom like, or phantom divide crossing behavior according to the value of model parameters. d2h2 dx2 + 3dh2 dx + 9bΩm0e−3(1−b)x = 0. (11) (11) where h2 = H2/H2 0 . The solution of the above differential equation in terms of scale factor is [103], h2 = Ωm0 1 −ba−3(1−b) + 1 3c′a−3 + c′′, (12) (12) which tells us how the Hubble parameter evolves with respect to the scale factor of the universe. In the previous works of THDE model with GO scale as IR cutoff [41,68,94], such an exact solution has not been obtained. Initial conditions to evaluate the constants c′ and c′′ are h2|a=1 = 1, dh2 dx  a=1 = 2 β′ ⎡ ⎢⎢⎣  Ωde0 H2(1−δ) 0  1 2 −δ −α′ ⎤ ⎥⎥⎦. (13) 3.1 The model parameter estimation ⎦ (13) (13) In this section, we estimate the constant model parameters, δ, α′, β′, b, H0 and Ωm0 by contrasting the model with cosmological observational data. The dataset consists of the type Ia supernovae data [109], observational Hubble data (OHD) [110,111], cosmic microwave background (CMB) data [112], and baryon acoustic oscillation (BAO) data [113]. In this section, we estimate the constant model parameters, δ, α′, β′, b, H0 and Ωm0 by contrasting the model with cosmological observational data. The dataset consists of the type Ia supernovae data [109], observational Hubble data (OHD) [110,111], cosmic microwave background (CMB) data [112], and baryon acoustic oscillation (BAO) data [113]. We have applied the Markov chain Monte Carlo (MCMC) method by employing the emcee python package [114] using the lmfit python library [115] to constrain the model param- eters. To perform the analysis, we assume uniform priors for parameters specific to the model as follows: δ ∈(0.1, 5.0), b ∈(−0.1, 0.5), α′ ∈(0.7, 1.2), and β′ ∈(0.001, 0.6), considering the recent literature [40–42,44–46,49,55,64– 66,68,69,72,73,75,94,105–107,116]. We have considered uniform priors from the recent literature suitably for other general parameters like H0, Ωm0, and M. In this section, we estimate the constant model parameters, δ, α′, β′, b, H0 and Ωm0 by contrasting the model with cosmological observational data. The dataset consists of the type Ia supernovae data [109], observational Hubble data The constants are then obtained as The constants are then obtained as The constants are then obtained as c′ = 2 β′ ⎡ ⎣α′ −  Ωde0 H2(1−δ) 0  1 2−δ ⎤ ⎦−3Ωm0, (14) c′′ = 1 −bΩm0 1 −b −2 3β′ ⎡ ⎣α′ −  Ωde0 H2(1−δ) 0  1 2−δ ⎤ ⎦. (15) (14) We have applied the Markov chain Monte Carlo (MCMC) method by employing the emcee python package [114] using the lmfit python library [115] to constrain the model param- eters. To perform the analysis, we assume uniform priors for parameters specific to the model as follows: δ ∈(0.1, 5.0), b ∈(−0.1, 0.5), α′ ∈(0.7, 1.2), and β′ ∈(0.001, 0.6), considering the recent literature [40–42,44–46,49,55,64– 66,68,69,72,73,75,94,105–107,116]. We have considered uniform priors from the recent literature suitably for other general parameters like H0, Ωm0, and M. 2 Interacting THDE model as dynamical vacuum (8) (8) The above equations (8) can be rewritten in terms of density parameter Ω = ρ 3H2 0 where H0 denotes the present value of 12 3 399 Page 4 of 14 Eur. Phys. J. C (2022) 82 :399 399 P 399 Hubble parameter and x = ln a as where ˜Ωm0 and ˜Ωde0 are the mass density parameters [104] which have the form Hubble parameter and x = ln a as dΩ where ˜Ωm0 and ˜Ωde0 are the mass density parameters [104] which have the form Hubble parameter and x = ln a as dΩ where ˜Ωm0 and ˜Ωde0 are the mass density parameters [104] which have the form where ˜Ωm0 and ˜Ωde0 are the mass density parameters [104] which have the form Hubble parameter and x = ln a as w 3.1 The model parameter estimation (15) In the asymptotic limit a →0, the constant c′′ can be neglected due to domination of the first two terms in the Hubble parameter equation (12) for b ≪1, consequently the resulting solution represents the decelerated expansion. In the future limit, a →∞, the constant term in the Hubble parameter will dominate over the rest of the terms, indicating an end de Sitter phase. Hence the model predicts a transition into a late accelerating epoch in the evolution of the universe. For b = 0 Eq. (12) will reduce to In the asymptotic limit a →0, the constant c′′ can be neglected due to domination of the first two terms in the Hubble parameter equation (12) for b ≪1, consequently the resulting solution represents the decelerated expansion. In the future limit, a →∞, the constant term in the Hubble parameter will dominate over the rest of the terms, indicating an end de Sitter phase. Hence the model predicts a transition into a late accelerating epoch in the evolution of the universe. For b = 0 Eq. (12) will reduce to We use the latest supernovae data, the Pantheon Sample, which comprises 1048 SNe Ia data points in the redshift range of 0.01 < z < 2.3. The luminosity distance of SN Ia can be obtained using the relation, h2 = ˜Ωm0a−3 + ˜Ωde0, (16) h2 = ˜Ωm0a−3 + ˜Ωde0, (16) Page 5 of 14 399 Eur. Phys. J. C (2022) 82 :399 Table 1 The best estimated values and 68.3% confidence limit for interacting THDE model parameters Data δ b α′ β′ Ωm0 H0 M SNIa 1.021+0.017 −0.020 −0.027+0.062 −0.054 0.963+0.139 −0.121 0.309+0.201 −0.185 0.300+0.035 −0.034 69.977+3.514 −3.436 −19.359+0.107 −0.109 OHD 1.021+0.021 −0.024 0.024+0.100 −0.053 0.962+0.153 −0.150 0.283+0.220 −0.202 0.297+0.037 −0.034 69.789+2.609 −2.533 −19.054+0.735 −0.672 SNIa+OHD 1.020+0.016 −0.021 0.023+0.033 −0.026 0.960+0.138 −0.125 0.337+0.184 −0.207 0.296+0.038 −0.033 69.241+1.230 −1.229 −19.381+0.037 −0.037 SNIa+OHD+CMB 1.021+0.012 −0.014 0.014+0.016 −0.022 0.970+0.129 −0.094 0.256+0.227 −0.158 0.271+0.022 −0.015 68.867+1.199 −1.159 −19.393+0.035 −0.034 SNIa+OHD+BAO 1.020+0.016 −0.022 0.028+0.033 −0.029 0.974+0.142 −0.132 0.352+0.172 −0.205 0.294+0.020 −0.019 69.267+1.209 −1.204 −19.379+0.036 −0.037 SNIa+OHD+CMB+BAO 1.021+0.012 −0.015 0.005+0.016 −0.017 0.981+0.128 −0.099 0.308+0.201 −0.181 0.281+0.017 −0.015 68.672+1.196 −1.141 −19.399+0.034 −0.034 (a) (b) Fig. 3.1 The model parameter estimation 1 The two dimensional posterior contours with 68% and 95% confidence limits and one dimensional posterior distribution (using pygtc open python package [108]) from the (a) SNIa+OHD+CMB and (b) SNIa+OHD+CMB+BAO datasets for the free parameters of THDE model dL (δ, α′, β′, b, H0, Ωm0, zi) ′ where μi = m −M is the observational distance modulus of SN I d M th t d th b l t it d Table 1 The best estimated values and 68.3% confidence limit for interacting THDE model parameters Data δ b α′ β′ Ωm0 H0 M SNIa 1.021+0.017 −0.020 −0.027+0.062 −0.054 0.963+0.139 −0.121 0.309+0.201 −0.185 0.300+0.035 −0.034 69.977+3.514 −3.436 −19.359+0.107 −0.109 OHD 1.021+0.021 −0.024 0.024+0.100 −0.053 0.962+0.153 −0.150 0.283+0.220 −0.202 0.297+0.037 −0.034 69.789+2.609 −2.533 −19.054+0.735 −0.672 SNIa+OHD 1.020+0.016 −0.021 0.023+0.033 −0.026 0.960+0.138 −0.125 0.337+0.184 −0.207 0.296+0.038 −0.033 69.241+1.230 −1.229 −19.381+0.037 −0.037 SNIa+OHD+CMB 1.021+0.012 −0.014 0.014+0.016 −0.022 0.970+0.129 −0.094 0.256+0.227 −0.158 0.271+0.022 −0.015 68.867+1.199 −1.159 −19.393+0.035 −0.034 SNIa+OHD+BAO 1.020+0.016 −0.022 0.028+0.033 −0.029 0.974+0.142 −0.132 0.352+0.172 −0.205 0.294+0.020 −0.019 69.267+1.209 −1.204 −19.379+0.036 −0.037 SNIa+OHD+CMB+BAO 1.021+0.012 −0.015 0.005+0.016 −0.017 0.981+0.128 −0.099 0.308+0.201 −0.181 0.281+0.017 −0.015 68.672+1.196 −1.141 −19.399+0.034 −0.034 (a) (b) Fig. 1 The two dimensional posterior contours with 68% and 95% confidence limits and one dimensional posterior distribution (using pygtc open python package [108]) from the (a) SNIa+OHD+CMB and (b) SNIa+OHD+CMB+BAO datasets for the free parameters of THDE model (b) (a) (b) Fig. 1 The two dimensional posterior contours with 68% and 95% confidence limits and one dimensional posterior distribution (using pygtc open python package [108]) from the (a) SNIa+OHD+CMB and (b) SNIa+OHD+CMB+BAO datasets for the free parameters of THDE model (b) (a) Fig. 1 The two dimensional posterior contours with 68% and 95% confidence limits and one dimensional posterior distribution (using pygtc open python package [108]) from the (a) SNIa+OHD+CMB and (b) SNIa+OHD+CMB+BAO datasets for the free parameters of THDE model dL (δ, α′, β′, b, H0, Ωm0, zi) = c(1 + zi) zi 0 dz′ H(δ, α′, β′, b, H0, Ωm0, z′), (18) where μi = m −M is the observational distance modulus of SN Ia, m and M are the apparent and the absolute magnitude of the SN Ia, n = 1048, the total number of data points and σ 2 i is the variance of ith measurement. We treat M as a nuisance parameter throughout this analysis. 3.1 The model parameter estimation The estimated values of the model parameters using the SNIa data are given in the first row of Table 1. The minimum χ2 is obtained as 1035.485. The χ2 dof = χ2 min n−n p is the minimum χ2 function per degrees of freedom where n is the number of data points, and n p is the number of model parameters. The obtained value of χ2 goodness of fit, χ2 d.o. f. = 0.995, i.e., around unity. The estimated value of the coupling constant is obtained to be negative, contributing a negative interaction term. It conveys the possibility of energy transfer from dark matter to THDE. where μi = m −M is the observational distance modulus of SN Ia, m and M are the apparent and the absolute magnitude of the SN Ia, n = 1048, the total number of data points and σ 2 i is the variance of ith measurement. We treat M as a nuisance parameter throughout this analysis. The estimated values of the model parameters using the SNIa data are given in the first row of Table 1. The minimum χ2 is obtained as 1035.485. The χ2 dof = χ2 min n−n p is the minimum χ2 function per degrees of freedom where n is the number of data points, and n p is the number of model parameters. The obtained value of χ2 goodness of fit, χ2 d.o. f. = 0.995, i.e., around unity. The estimated value of the coupling constant is obtained to be negative, contributing a negative interaction term. It conveys the possibility of energy transfer from dark matter to THDE. (18) where zi is the redshift of the SN Ia, c is the speed of light and H(δ, α′, β′, b, H0, Ωm0, z′) is the Hubble parameter in terms of model parameters and redshift. The theoretical distance modulus of SN Ia is given by μth (δ, α′, β′, b, H0, Ωm0, zi) = 5 log10 dL(δ, α′, β′, b, H0, Ωm0, zi) Mpc + 25. 3.1 The model parameter estimation The obtained value of χ2 goodness of fit, χ2 d.o. f. = 0.990, i.e., around unity. f Combining the type Ia supernovae data and the OHD, the total χ2 function takes the form χ2 f = χ2(δ, α′, β′, b, H0, Ωm0, M)SN Ia + χ2(δ, α′, β′, b, H0, Ωm0)O H D. (22) (22) f Combining the type Ia supernovae data, the OHD, the CMB, and the BAO data, the total χ2 function takes the form χ2 s = χ2(δ, α′, β′, b, H0, Ωm0, M)SN Ia + χ2(δ, α′, β′, b, H0, Ωm0)O H D + χ2(δ, α′, β′, b, H0, Ωm0, M)C M B + χ2(δ, α′, β′, b, H0, Ωm0, M)B AO. (27) The estimated values of the model parameters using the SNIa data and OHD are given in the third row of Table 1. The minimum χ2 f is obtained as 1071.352. The obtained value of χ2 goodness of fit, χ2 d.o. f. = 0.995, i.e., around unity. (27) d.o. f. The distance prior, shift parameter, R, which influences the CMB temperature power spectrum, is defined in terms of THDE model parameters as follows The estimated values of the model parameters using the type Ia supernovae data, the OHD, the CMB, and the BAO data are given in the sixth row of Table 1. The minimum χ2 s is obtained as 1174.015. The obtained value of χ2 goodness of fit, χ2 d.o. f. = 1.088, i.e., around unity. The estimated value of the coupling constant is obtained to be positive and is smaller than the value obtained from all the other datasets except SNIa for the present THDE model and the previ- ous THDE models [116] with Hubble horizon and future event horizon as IR cutoff. The positive value of the cou- pling constant contributes to a positive interaction term, con- sequently the energy exchange is from THDE to dark mat- ter and thereby satisfying the Le Chatelier–Braun princi- ple [55,119]. The uncertainty on b is relatively high com- pared to the value at which the maximum likelihood function peaks for all datasets. Evidence for such plausibility using SNIa and OHD can be observed in the work of C. P. Singh [120] on Holographic dark energy. The best fit value of δ obtainedusingallthedatasetsissimilartothepreviousresults from THDE models with future event horizon as IR cutoff [105,116], which are stable against the background pertur- bations. 3.1 The model parameter estimation (19) (19) The χ2 function of SN Ia data can be expressed as The χ2 function of SN Ia data can be expressed as Another independent observable we use is the Hubble dataset consisting of 36 H(z) data points in the redshift range of 0.07 ≤z ≤2.36, out of which 31 data points are determined using the cosmic chronometric technique, 3 χ2 (δ, α′, β′, b, H0, Ωm0, M)SN Ia χ (δ, α , β , b, H0, Ωm0, M)SN Ia = n  i=1 [μth(δ, α′, β′, b, H0, Ωm0, zi) −μi]2 σ 2 i , (20) 123 12 3 399 Page 6 of 14 399 Page 6 of 14 Eur. Phys. J. C (2022) 82 :399 399 Pa is defined in terms of THDE model parameters as follows data points from the radial BAO signal in the galaxy distri- bution and 2 data points from the BAO signal in the Lyman α forest distribution alone or cross correlated with quasi stellar objects [110]. The following χ2 function is minimized for this Hubble parameter measurement A(δ, α′, β′, b, H0, Ωm0) = √Ωm0 h(z A)1/3  1 z A z A 0 dz h(δ, α′, β′, b, H0, Ωm0, z)  2 3 , (25) (25) χ2 (δ, α′, β′, b, H0, Ωm0)O H D = n  i=1 [H(δ, α′, β′, b, H0, Ωm0, zi) −Hi]2 σ 2 i , (21) where z A is the redshift of the acoustic peak parameter. We adopt the value of Aobs = 0.484 ± 0.016 at the redshift z A = 0.35 from the SDSS-BAO distance data [118]. The corresponding χ2 from the BAO data is (21) where Hi is the observational Hubble parameter measure- ment, and σ 2 i is the variance of ith measurement. The esti- mated values of the model parameters using the OHD data are given in the second row of Table 1. The minimum χ2 is obtained as 22.034. The obtained value of χ2 goodness of fit, χ2 d.o. f. = 0.760. χ2(δ, α′, β′, b, H0, Ωm0, M)B AO = [A(δ, α′, β′, b, H0, Ωm0) −Aobs]2 σ 2 A , (26) (26) where σA is the variance of the Aobs measurement. The esti- mated values of the model parameters using the SNIa+OHD+BAO are given in the fifth row of Table 1. The minimum χ2 is obtained as 1067.240. 123 3.1 The model parameter estimation The evolution of the deceleration parameter q(z), as a function of z, for the best estimated model param- eters from the fourth and sixth datasets is plotted in Fig. 4. It is lucid from the figure that the model explains the current accelerated universe and also the transition from the prior decelerated phase to the present accelerated phase. The inset figure in Fig. 4 shows the transition in small scale to reveal precisely the difference in the transition redshift for different datasets. The transition redshift zt (i.e., q(zt) = 0) is found to be 0.799 using the fourth dataset and 0.763 using the sixth dataset, which is slightly greater than the transition redshift obtained for interacting THDE model with Hubble horizon and future event horizon as IR cutoff (zt = 0.634+0.051 −0.045 and zt = 0.649+0.010 −0.025 respectively) [116]. The current value of deceleration parameter (q0 = q(z)|z=0) obtained for the esti- mated values of the model parameters is q0 = −0.602 using the fourth dataset and −0.594 using the sixth dataset, is closer to the observational value q0 = −0.63 ± 0.12 from ΛCDM based CMB priors [123]. Our estimated cosmological param- eter values are in good concordance with previous results from model independent methods, parametric reconstruction techniques as well as other dark energy models mentioned in [124–127]. 3.1 The model parameter estimation 3 The evolution of apparent magnitude with redshift for the best estimated values of the model parameters using the fourth and sixth datasets, compared with type Ia SNe data points q = −1 − ˙H H2 = −1 + 3Ωm0(1 + z)3(1−b) + c′(1 + z)3 2Ωm0 1 −b (1 + z)3(1−b) + 2 3c′(1 + z)3 + 2c′′ . (28) the best fit δ value (less than 1) obtained from the study on the growth of matter fluctuations of the THDE model [73] with future event horizon as IR cutoff using the Gold- 2017 dataset of 18 uncorrelated σ8 measurements, SNIa data, and OHD. (28) In the limit z →−1, the deceleration parameter q →−1 and in the limit z →∞, the deceleration parameter will tend to a positive value since second term in Eq. (28) domi- nates in that case. The evolution of the deceleration parameter q(z), as a function of z, for the best estimated model param- eters from the fourth and sixth datasets is plotted in Fig. 4. It is lucid from the figure that the model explains the current accelerated universe and also the transition from the prior decelerated phase to the present accelerated phase. The inset figure in Fig. 4 shows the transition in small scale to reveal precisely the difference in the transition redshift for different datasets. The transition redshift zt (i.e., q(zt) = 0) is found to be 0.799 using the fourth dataset and 0.763 using the sixth dataset, which is slightly greater than the transition redshift obtained for interacting THDE model with Hubble horizon and future event horizon as IR cutoff (zt = 0.634+0.051 −0.045 and zt = 0.649+0.010 −0.025 respectively) [116]. The current value of deceleration parameter (q0 = q(z)|z=0) obtained for the esti- mated values of the model parameters is q0 = −0.602 using the fourth dataset and −0.594 using the sixth dataset, is closer to the observational value q0 = −0.63 ± 0.12 from ΛCDM based CMB priors [123]. Our estimated cosmological param- eter values are in good concordance with previous results from model independent methods, parametric reconstruction techniques as well as other dark energy models mentioned in [124–127]. In the limit z →−1, the deceleration parameter q →−1 and in the limit z →∞, the deceleration parameter will tend to a positive value since second term in Eq. (28) domi- nates in that case. 3.1 The model parameter estimation Concurrently, it is smaller than the best fit value (δ > 2) obtained from the studies on the THDE model with Hubble horizon as IR cutoff [116], which is unstable against the background perturbation. Furthermore, it is greater than R(δ, α′, β′, b, H0, Ωm0) =  Ωm0 z∗ 0 dz h(δ, α′, β′, b, H0, Ωm0, z), (23) (23) where z∗is the redshift at the photon decoupling epoch. We adopt the distance prior measurement value Robs = 1.7502± 0.0046 at the redshift z∗= 1089.92 from the Planck 2018 observations [112]. The corresponding χ2 [117] from the CMB data is χ2(δ, α′, β′, b, H0, Ωm0, M)C M B = [R(δ, α′, β′, b, H0, Ωm0) −Robs]2 σ 2 R , (24) (24) where σR is the variance of the Robs measurement. The esti- mated values of the model parameters using the SNIa+OHD+CMB are given in the fourth row of Table 1. The minimum χ2 is obtained as 1203.630. The obtained value of χ2 goodness of fit, χ2 d.o. f. = 1.117, i.e., around unity. The distance–redshift relation determined by a BAO mea- surement is given by the acoustic peak parameter, A, which 123 Eur. Phys. J. C (2022) 82 :399 Page 7 of 14 399 Fig. 2 The evolution of Hubble parameter with redshift for the best estimated values of the model parameters using the fourth and sixth datasets, compared with H(z) data points Fig. 3 The evolution of apparent magnitude with redshift for the best estimated values of the model parameters using the fourth and sixth datasets, compared with type Ia SNe data points Fig. 2 The evolution of Hubble parameter with redshift for the best estimated values of the model parameters using the fourth and sixth datasets, compared with H(z) data points Fig. 3 The evolution of apparent magnitude with redshift for the best estimated values of the model parameters using the fourth and sixth datasets, compared with type Ia SNe data points Fig. 2 The evolution of Hubble parameter with redshift for the best estimated values of the model parameters using the fourth and sixth datasets, compared with H(z) data points Fig. 3.3 The age of the universe 3.3 The age of the universe The age of the universe can be estimated using the present cosmological observational data, even though systematic and statistical uncertainties will arise during observation and estimation. Theoretically, considering the interacting THDE model, the age of the universe can be calculated by slightly rearranging the Eq. (12) and the resultant equation takes the form H = ˙a a = H0  Ωm0 1 −ba−3(1−b) + 1 3c′a−3 + c′′  1 2 . (29) (29) Above equation can be solved by considering the value of scale factor a = 0 for the big bang time tb and a = 1 for the present time t0 in the following way Fig. 4 The evolution of deceleration parameter with redshift for the best estimated values of the model parameters using the fourth and sixth datasets. The inset figure clearly distinguishes the continuous and dot-dashed lines t0 tb dt = H−1 0 1 0 a−1  Ωm0 1 −ba−3(1−b) + 1 3c′a−3 + c′′  −1 2 da. Using the best estimated values of the parameters, the age of the universe is evaluated as 14.327 Gyrs using the fourth dataset and 14.190 Gyrs using the sixth dataset, which is closer to the age calculated for interacting THDE model with future event horizon as IR cutoff (14.20+0.18 −0.32 Gyrs) and slightly greater than the age calculated for interacting THDE model with Hubble horizon as IR cutoff (13.71+0.24 −0.41 Gyrs, upper bond of 13.43 and lower bond of 13.04 Gyrs) [107, 116]. Our results are closer to the standard value of age 13.8± 0.02 Gyrs obtained from Planck mission and 13.72 ± 0.12 GyrsfromWMAP+BAO+SNdataassumingΛCDMmodel [121,122] and 13.5+0.16 −0.14(stat.) ±0.23(0.33)(sys.) Gyrs from the oldest globular cluster [129]. Fig. 5 Evolution of matter density and THDE density with scale factor of the universe in logarithmic scale using the fourth and sixth datasets 3.2 Evolution of cosmological parameters Estimating luminosity and redshift from the cosmological observables contributes to determining the present Hub- ble parameter of the universe. Here, we studied the back- ground evolution of the Hubble parameter using the inter- acting THDE solution. The predicted evolution of the Hub- ble parameter with redshift using this model has been compared with the 36 H(z) data points and is shown in Fig. 2. The Hubble parameter decreases as the cosmic time evolves and the present value of the Hubble param- eter is obtained as 68.867+1.199 −1.159 km s−1 Mpc−1 (fourth dataset) and 68.672+1.196 −1.141 km s−1 Mpc−1 (sixth dataset) which is slightly lower than the observational value H0 = 70.5 ± 1.3 km s−1 Mpc−1 from WMAP + BAO + SN data [121] and slightly higher than observational value H0 = 67.37 ± 0.54 km s−1 Mpc−1 from Planck data [122] assum- ing ΛCDM model. The estimated present value of the Hubble parameter in our model is close to the best estimated values of Hubble parameter in interacting and non interacting THDE models with Hubble horizon and future event horizon as IR cutoff [73,105,116]. The theoretical evolution of apparent magnitude with respect to the redshift using this model is compared with the 1048 SNe Ia data points as shown in Fig. 3. Both the error bar plots show proximity between theoretical and observational results. A juxtaposition of the evolution of matter density and THDE density in logarithmic scale is shown in Fig. 5. The figure very well implies that THDE will dominate the uni- The deceleration parameter which measures the rate of cosmic expansion is obtained in the following form 12 123 3 Eur. Phys. J. C (2022) 82 :399 399 Page 8 of 14 Fig. 4 The evolution of deceleration parameter with redshift for the best estimated values of the model parameters using the fourth and sixth datasets. The inset figure clearly distinguishes the continuous and dot-dashed lines 3.4 Evolution in r −s plane and ω ′ e −ωe plane Fig. 5 Evolution of matter density and THDE density with scale factor of the universe in logarithmic scale using the fourth and sixth datasets To check further, the reliability of THDE model being a gen- eralized model of dark energy in contrast to the present obser- vational data we studied the evolution using the geometrical pair called the statefinder pair {r, s} constructed from the scale factor and its derivatives, first defined by Sahni et al. [130,131]. The relationship connecting the statefinder pair and the scale factor of the universe are given by verse in the future. The evolution of the densities of matter and dark energy are comparable with each other for the high redshift values even though the matter density is dominating in the early universe and is diminishing in the late phase of the universe, that solves the coincidence problem as the acceler- ation began at low redshifts and this is well demonstrated in the Fig. 5. The current value of the matter density parameter estimated using the observational constraints is slightly lower than the observational value Ωmo = 0.3147 ± 0.0074 from Planck data [122] and the derived value Ωmo = 0.29 ± 0.07 from WMAP results [128]. Our estimation of the current val- ues of the matter density parameter are similar to the current value obtained for past THDE models [105,116] and slightly higher than the value for the THDE model with future event horizon as IR cutoff [73]. r = ...a aH3 , s = r −1 3(q −1 2) . (30) (30) It is lucid from (30), r is third order derivative of ‘a’ and s is related to ‘r’ and ‘q’ linearly. The statefinder pair {r, s} can also be expressed in terms of Hubble parameter and its derivatives as r = 1 2h2 d2h2 dx2 + 3 2h2 dh2 dx + 1, s = − 1 2h2 d2h2 dx2 + 3 2h2 dh2 dx 3 2h2 dh2 dx + 9 2 . (31) (31) 12 123 123 Eur. Phys. J. C (2022) 82 :399 Page 9 of 14 399 Fig. 6 The statefinder evolutionary trajectory for interacting THDE in the r −s plane for the best estimated values of the model parameters using the fourth and sixth datasets contrast to these results, the evolution of the present model shows a quintessence behavior that finally approaches the ΛCDMbehaviorwithoutshowing,Chaplygingaslikebehav- ior. In the absence of interaction, the present model shows a behavior almost similar to ΛCDM. The effect of interaction (i.e., with respect to the value of coupling constant) will only increase or decrease the distance of approach to the ΛCDM from the present. In general, for the estimated value of δ, the present THDE model shows quintessence behavior and will finally approach ΛCDM for the form of the interaction considered here (devoid of the coupling constant value) and shows ΛCDM like behavior in case of zero interaction. In addition, we have obtained the effective equation of state parameter using the relation, Fig. 6 The statefinder evolutionary trajectory for interacting THDE in the r −s plane for the best estimated values of the model parameters using the fourth and sixth datasets ωe = P ρ = −1 −2 ˙H 3H2 , (33) (33) Using Eq. (12) in the above equations will results in the fol- lowing two expressions in terms of model parameters where P and ρ are the effective pressure and energy density respectively. Using the Eq. (12) in (33) the effective equation of state parameter for this model takes the following form r = 1 − 9bΩm0a−3(1−b) 2( Ωm0 1−b a−3(1−b) + 1 3c′a−3 + c′′) , s = −bΩm0a−3(1−b) c′′ + b 1−bΩm0a−3(1−b) . (32) ωe = −1 + 3Ωm0e−3(1−b)x + c′e−3x 3Ωm0 1−b e−3(1−b)x + c′e−3x + 3c′′ . (34) (34) (32) The asymptotic limit of this is as follows. As x →−∞, the effective equation of state, ωe →0 and as x →+∞, ωe → The asymptotic limit of this is as follows. As x →−∞, the effective equation of state, ωe →0 and as x →+∞, ωe → −1. In the former case, the decelerated epoch, the matter component could be the dominant component, while in the latter case, the accelerated epoch, the dominant component is the dark energy. The derivative of ωe with respect to x is obtained to be The parametric plot of {r, s} is obtained as shown in Fig. 6. 123 Using this diagnostics THDE model can be differentiated fromthe ΛCDMmodel. Thestatefinder pair{r, s} for ΛCDM is {1, 0}. Applying the limit a →∞, shows that the {r, s} pair for the model approaches {1, 0}. Thus this model approaches ΛCDM asymptotically at late times. The present value of {r, s} pair from the evolutionary trajectory of the model is obtained as {0.983, 0.005}and {0.994, 0.002} for the best fit parameter values using the fourth and sixth datasets, respec- tively. For the quintessence models the {r, s} pair lies in r < 1 and s > 0. The trajectory in its early phase and in the present shows aforesaid behavior. Comparing the evolu- tionary trajectories in Fig. 6 for the best fit parameter values from the fourth and sixth datasets tells us that the distance to the ΛCDM point from the present is larger for the larger val- ues of coupling constant and the distance to the ΛCDM point from the present is small for the smaller values of coupling constant. −1. In the former case, the decelerated epoch, the matter component could be the dominant component, while in the latter case, the accelerated epoch, the dominant component is the dark energy. The derivative of ωe with respect to x is obtained to be ω′ e = 3 x3 x1 2 −x2 x1 , (35) (35) where where x1 = 3Ωm0 1 −b e−3(1−b)x + c′e−3x + 3c′′, x2 = 9Ωm0(1 −b)e−3(1−b)x + 3c′e−3x, and x3 = 3Ωm0e−3(1−b)x + c′e−3x. x3 = 3Ωm0e−3(1−b)x + c′e−3x. The evolutionary trajectory of the model in the phase plane of ω′ e −ωe, the effective equation of state parameter and its derivative, is depicted in Fig. 7. It complements the results from statefinder analysis by showing that the trajectory is initially confined in the quintessence region ωe > −1, ω′ e < 0 and approaches de Sitter phase in the late time as ωe →−1, ω′ e →0. The late time cosmic evolution of non interacting THDE model with the Hubble horizon as IR cutoff in flat FRW universe [65,66,106] using the ωde −ω′ de plane, where ωde is the equation of state parameter of dark energy and ω′ de is its evolution, shows that for 1 < δ < 2, the model shows quintessence behavior. 123 In the presence of interaction [49,65], The statefinder evolutionary trajectory of the non interact- ing THDE model with the Hubble horizon as IR cutoff in the flat FRW universe in the past works [65,66,106] shows that for 1 < δ < 2, the model is quintessenc like and approaches ΛCDM in the future. The interacting THDE model with the Hubble horizon as IR cutoff in flat FRW universe in past works [49,65] shows Chaplygin gas like behavior for 1 < δ < 2 and the distance to the ΛCDM point from the present time will be larger or smaller depending on the cor- responding larger or smaller value of coupling constant. In 12 399 Page 10 of 14 Eur. Phys. J. C (2022) 82 :399 Fig. 7 The evolutionary trajectory for interacting THDE in the ω′ e−ωe plane for the best estimated values of the model parameters using the fourth and sixth datasets in which ξ and η are small compared to ˜u and ˜v. Linearizing the set of Eqs. (37) and (38) with respect to ξ and η, will result in a matrix equation  ξ′ η′  = ⎡ ⎢⎣  ∂u′ ∂u  (˜u,˜v)  ∂u′ ∂v  (˜u,˜v)  ∂v′ ∂u  (˜u,˜v)  ∂v′ ∂v  (˜u,˜v) ⎤ ⎥⎦ ξ η , (40) (40) where 2 × 2 matrix on the right hand side of the above equa- tion is the Jacobian at the critical points corresponding to the autonomous system and the partial derivatives are calculated about the critical points (˜u, ˜v). The eigenvalues of the 2 × 2 Jacobian matrix can be obtained by diagonalizing the matrix, and its nature decides the asymptotic stability nature of critical points. The critical points (equilibrium points) can be source point or past attrac- tor if the eigenvalues were positive values, or a saddle point if at least one of the eigenvalues were of a different sign, and future attractor if all the eigenvalues were negative values [132]. Fig. 7 The evolutionary trajectory for interacting THDE in the ω′ e−ωe plane for the best estimated values of the model parameters using the fourth and sixth datasets the model shows completely phantom behavior with higher coupling constant values. Whereas the present THDE model shows quintessence behavior in general and approaches de Sitter phase in the late time in the ω′ e −ωe plane as well. 123 The eigenvalues corresponding to the Jacobian matrix are evaluated and are given in Table 2. The critical points (0.958, 0) and (0.985, 0) for which the matter is the domi- nant component, are unstable since their eigenvalues are pos- itive. All trajectories diverge from (0.958, 0) and (0.985, 0), which is clear from Figs. 8 and 9. The critical point (0, 1), for which dark energy is the dominant component, is stable since its eigenvalues are negative. This represents a future stable point. The convergence of all trajectories to the future attractor at (0, 1) is well depicted in the phase space diagram shown in Figs. 8 and 9. Universe undergoes a de Sitter expan- sion at this point. Our analysis shows that the system emerges from a decelerated expansion and ends on a de Sitter epoch. The previous works [64,107] in the dynamical system anal- ysis of interacting and non interacting THDE with Hubble horizon as IR cutoff also show a stable point corresponding to the de Sitter phase in the future. 4 Dynamical analysis of interacting THDE model The dynamical system analysis enables us to see whether the model shows the stable evolution consistent with the obser- vations. For the phase space analysis of the model, a set of dimensionless variables is introduced as follows: u = ρm 3H2 , v = ρde 3H2 . (36) (36) The autonomous equations in terms of phase space vari- ables can be obtained using the Friedmann equation as, u′ = u2 −2uv + (3b −1)u, (37) (37) v′ = (2 −δ)[2(α′ −β′)(v2 −v) −uv((α′ −β′) + 3β′ 2 (3b −1))] α′ + β′(v −u 2 −1) −2v2 + uv + 2v, (38) 5 Thermodynamics of interacting THDE model (38) −2v2 + uv + 2v, Any isolated macroscopic system advances to the maximum entropy state in compliance with the constraints of the gen- where prime (′) denotes differentiation with respect to x = ln a. Both the equations are functions of u and v. By equating u′ = 0 and v′ = 0, the real and physically meaningful critical points obtained are (˜u, ˜v) = (0, 1), (1 −3b, 0) and they corresponds to the de Sitter phase and the matter dominated phase respectively. Table 2 Critical points and stability of the corresponding eigenvalues calculated using the best estimated model parameter values Data Critical points Eigenvalues Nature SNIa+OHD+ (0.958, 0) (0.958, 0.045) Unstable CMB (0, 1) (−2.958, −0.559) Stable SNIa+OHD+ (0.985, 0) (0.985, 0.054) Unstable CMB+BAO (0, 1) (−2.985, −0.657) Stable Table 2 Critical points and stability of the corresponding eigenvalues calculated using the best estimated model parameter values Table 2 Critical points and stability of the corresponding eigenvalues calculated using the best estimated model parameter values The behavior of the trajectory near the critical points can be analyzed by considering the small neighbourhood of crit- ical points as u = ˜u + ξ, v = ˜v + η. (39) u = ˜u + ξ, v = ˜v + η. (39) 123 123 Eur. Phys. J. C (2022) 82 :399 Page 11 of 14 399 Fig. 8 The evolution of the phase space trajectories in the u −v plane for the best estimated model parameters using SNIa+OHD+CMB dataset Fig. 10 The evolution of SH γ with scale factor for the best estimated values of the model parameters using the fourth and sixth datasets Fig. 10 The evolution of SH γ with scale factor for the best estimated values of the model parameters using the fourth and sixth datasets Fig. 8 The evolution of the phase space trajectories in the u −v plane for the best estimated model parameters using SNIa+OHD+CMB dataset Fig. 11 The evolution of S′ H γ with scale factor for the best estimated values of the model parameters using the fourth and sixth datasets Fig. 9 The evolution of the phase space trajectories in the u −v plane for the best estimated model parameters using SNIa+OHD+CMB+BAO dataset Fig. 11 The evolution of S′ H γ with scale factor for the best estimated values of the model parameters using the fourth and sixth datasets Fig. 5 Thermodynamics of interacting THDE model 12 The evolution of S′′ H γ with scale factor for the best estimated values of the model parameters using the fourth and sixth datasets Fig. 9 The evolution of the phase space trajectories in the u −v plane for the best estimated model parameters using SNIa+OHD+CMB+BAO dataset eralized second law (GSL) of thermodynamics [133]. From the aforesaid statement, it can be deduced that the entropy S, of isolated systems, cannot decrease, i.e., S′ ≥0, where the prime means derivative with respect to the scale factor a and it must be a convex function of the scale factor, S′′ < 0, at least at the last juncture of the evolution so that the system attains a stable equilibrium state. Fig. 12 The evolution of S′′ H γ with scale factor for the best estimated values of the model parameters using the fourth and sixth datasets In the present work, we consider Tsallis entropy, given in Eq. (1) as the horizon entropy in place of the standard Bekenstein-Hawking entropy. Then the horizon entropy of the apparent horizon of a flat universe, SH becomes to scale factor is given by to scale factor is given by to scale factor is given by S′ H = γ  4πc2δ −2δH′ H2δ+1  . (42) (42) SH = γ 4πc2 H2 δ kB, (41) (41) The second derivative of horizon entropy with respect to the scale factor is given by The second derivative of horizon entropy with respect to the scale factor is given by wherec isthespeedoflightandkB istheBoltzmannconstant. The evolution of SH γ with respect to the scale factor is shown in Fig. 10. The rate of change of horizon entropy with respect S′′ H = 2γ δ  4πc2δ  (2δ + 1)  H′ Hδ+1 2 − H′′ H2δ+1  . (43) (43) 123 399 Page 12 of 14 399 Page 12 of 14 Eur. Phys. J. C (2022) 82 :399 The numerical simulations and analysis of the observa- tional data on Hubble parameter shows that H′ < 0 and H′′ > 0 [134–137]. Figure 11 clearly shows that the hori- zon entropy satisfies S′ H γ ≥0, always, which in turn implies that the horizon entropy is always increasing. Finally, in the asymptotic limit a →∞, the entropy S′ H approaches zero, consequently horizon entropy attains a constant value. 5 Thermodynamics of interacting THDE model That is, the horizon entropy asymptotically reaches constant at the end de Sitter epoch. The behavior of S′′ H γ with scale fac- tor in Fig. 12 guarantees the convexity, thereby ensuring that entropy does not grow unboundedly. An exact solution for the Hubble parameter in terms of scale factor was obtained by analytically solving the Friedmann equation with the energy conservation equa- tion. The solution successfully explains the prior decel- erated expansion and the late time accelerated expansion, hence effectively elucidating the transition. We have imposed the observational constraints on the cosmological param- eters of the interacting THDE model using the Pantheon sample, the OHD, the CMB and the BAO data. The esti- mated current values of the Hubble parameter (H0 = 68.867+1.199 −1.159 km s−1 Mpc−1 from the fourth dataset and H0 = 68.672+1.196 −1.141 km s−1 Mpc−1 from the sixth dataset) and the matter density parameter (Ωm0 = 0.271+0.022 −0.015 from the fourth dataset and Ωm0 = 0.281+0.017 −0.015 from the sixth dataset) are consistent with the observational results. The value of the coupling constant is positive, and hence it obeys the Le Chatelier–Braun principle. The GSL of thermodynamics stipulates that the entropy of the horizon together with the entropy of the matter inside the horizon must always increase with time [138]. Since it is clear the matter entropy is much less than the horizon entropy (smaller in the order of 35) [139,140], the total entropy of the universe can approximately be taken as the horizon entropy [141]. Since the horizon entropy is an increasing function as shown in Figs. 10 and 11, the GSL is considered to be sat- isfied. Certain past studies [55] of interacting THDE model with Hubble horizon as IR cutoff with the form of interaction Q = 3H(b2 1ρm + b2 2ρde), where b2 1,2 are the coupling con- stants, and with the varying equation of state of DE, shows the plausibility of violation of GSL depending on the evolution of the universe. Unlike those studies, the present analysis by considering DE as dynamical vacuum guarantees the validity of GSL throughout the evolution of the universe. 5 Thermodynamics of interacting THDE model The cosmological evolution of the deceleration parameter shows that the end phase will be a de Sitter one, q →−1 as z →−1 and the transition from prior matter dominated epoch to the late THDE dominated epoch is found to occur at a redshift around zt ∼0.8. The studies on the expansion profile of the THDE density and the matter density explains the dominating and diminishing nature of the matter density in the early phase and late phase, respectively, and along with that, similar nature of the evolution of THDE in the early phase removes the coincidence problem. The universe’s age that can be intuited from the interacting THDE model is around 14 Gyrs from the best estimated values of model parameters. Recent observations do subsistence the results. As evident from the Eq. (16), when the interaction param- eter b = 0, the present model reduces to a model, like ΛCDM with an effective cosmological constant corresponding to ˜Ωde0, and the value of which is around 0.735 and 0.729 for the best estimated parameter values using the fourth and sixth datasets, respectively. Results from various studies, like in [142,143], support that a ΛCDM like model would sat- isfy the GSL. Following this, it can be concluded that the present model with b = 0 will satisfy the GSL throughout the evolution. The diagnosis of the interacting THDE model is carried out using the statefinder analysis and the ω′ e −ωe pair. The evolutionary trajectory of the interacting THDE model shows that the model is distinguishably different from the standard ΛCDM and shows a quintessence behavior in the early phase and approaches ΛCDM in the far future. The trajectories of the ω′ e−ωe affirm the results from the statefinder diagnostics. The critical points obtained from the phase space analysis of the model shows that the de Sitter phase is a stable equilib- rium, and matter dominated phase is an unstable equilibrium. Thus the dynamical system analysis of the interacting THDE model portrays a consistent background evolution of the uni- verse from the prior decelerated phase to the late accelerated phase. 6 Conclusion Recently much interest has been arisen in the new holo- graphic dark energy based on the non extensive entropy, Tsal- lisentropy,andholographicprinciple.Inthepresentwork,we have analyzed interacting Tsallis Holographic Dark Energy (THDE) as dynamical vacuum to explain the recent accel- erated cosmic expansion. We have considered the Granda– Oliveros (GO) scale, a function of Hubble parameter and its time derivative, as IRcutoff. The interaction between the dark sectors has been accounted, by a simple form of interaction term Q = 3bHρm. The thermodynamical analysis of the THDE model shows that the horizon entropy increases with cosmic time and achieve the maximization condition during the late epoch. It can be concluded that the generalized second law of thermo- dynamics remains valid in the dynamical vacuum treatment of the model by considering the dominant contribution of the horizon entropy to the total entropy of the universe. It is to be noted that, for zero interaction parameter, the model will 123 123 Page 13 of 14 399 Eur. Phys. J. C (2022) 82 :399 reduce to a ΛCDM like one, where the validity of the laws of thermodynamics is guaranteed. reduce to a ΛCDM like one, where the validity of the laws of thermodynamics is guaranteed. 22. L.P. Chimento, M.G. Richarte, Phys. Rev. D 84, 123507 (2011) 23. L.P. Chimento, M.G. Richarte, Phys. Rev. D 85(12), 127301 (2012) 24. L.P. Chimento, M.G. Richarte, Eur. Phys. J. C 73, 2352 (2013) Acknowledgements We are thankful to the referee for the valuable and insightful comments which helped in bringing improvements in the current manuscript. We are thankful to Dr. Jerin Mohan N D and Hassan Basari V T for their valuable suggestions in data analysis. M Dheepika is thankful to CSIR for the financial support through JRF. 25. S. Nojiri, S.D. Odintsov, Eur. Phys. J. C 77(8), 528 (2017) 26. S. Das, P. Majumdar, R.K. Bhaduri, Class. Quantum Gravity 19(9), 2355 (2002) 27. S. Das, S. Shankaranarayanan, S. Sur, Black Hole Entropy from Entanglement: A Review, Vol. 268 (2009) 28. S. Das, S. Shankaranarayanan, S. Sur, Phys. Rev. D 77(6), 064013 (2008) Data Availability Statement This manuscript has no associated data or the data will not be deposited. [Authors’ comment: The Pantheon data used in this article is available at https://github.com/dscolnic/Pantheon and also available in reference [109]. The observational Hubble data (OHD) is available in references [110] and [111]. The CMB data is available in [112] and the BAO data is available in references [113] and [118].] 29. S. Das, S. Shankaranarayanan, S. Sur, in 12th Marcel Grossmann Meeting on General Relativity, pp. 1138–1141 (2010) 30. N. Radicella, D. Pavón, Phys. Lett. B 691(3), 121 (2010) 31. S. Das, S. Shankaranarayanan, S. Sur, in The Twelfth Marcel Grossmann Meeting: On Recent Developments in Theoretical and Experimental General Relativity, Astrophysics and Relativistic Field Theories (In 3 Volumes) (World Scientific, 2012), pp. 1138– 1141 Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, pro- vide a link to the Creative Commons licence, and indicate if changes were made. Neto, arXiv:1403.5706 (2014) 38. R.C. Nunes, J. Barboza, Edésio M., E.M.C. Abreu, J. Ananias Neto, JCAP 2016(8), 051 (2016) 39. H. Moradpour, Int. J. Theor. Phys. 55(9), 4176 (2016) 40. M. Tavayef, A. Sheykhi, K. Bamba, H. Moradpour, Phys. Lett. B 781, 195 (2018) g y Funded by SCOAP3. 41. M.A. Zadeh, A. Sheykhi, H. Moradpour, K. Bamba, Eur. Phys. J. C 78(11), 1 (2018) 42. A.A. Aly, Adv. Astron. 2019, 8138067 (2019) 123 The images or other third party material in this article are included in the article’s Creative Commons licence, unless indi- cated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permit- ted use, you will need to obtain permission directly from the copy- right holder. To view a copy of this licence, visit http://creativecomm ons.org/licenses/by/4.0/. 32. C. Tsallis, J. Stat. Phys. 52, 479 (1988) 32. C. Tsallis, J. Stat. Phys. 52, 479 (1988) 33. M.L. Lyra, C. Tsallis, Phys. Rev. Lett. 80, 53 (1998) 34. C. Tsallis, R. Mendes, A.R. Plastino, Phys. A Stat. Mech. Appl. 261(3–4), 534 (1998) 35. G. Wilk, Z. Włodarczyk, Phys. Rev. Lett. 84(13), 2770 (2000 36. C. Tsallis, L.J.L. Cirto, Eur. Phys. J. C 73, 2487 (201 37. R.d.C. Nunes, J. Barboza, Edésio M., E.M.C. Abreu, J. Ananias Neto, arXiv:1403.5706 (2014) References 43. V. Srivastava, U.K. Sharma, Int. J. Geom. Meth. Mod. Phys. 17(11), 2050144 (2020) 1. A.G. Riess et al., Astron. J. 116, 1009 (1998) 44. U.K. Sharma, V. Srivastava, New Astron. 84, 101519 (2021) 2. S. Perlmutter et al., Astrophys. J. 517, 565 (1999) 45. M.A. Zadeh, A. Sheykhi, H. Moradpour, Gen. Relativ. Gravit. 51, 1 (2019) 3. R.R. Caldwell, R. Dave, P.J. Steinhardt, Phys. Rev. Lett. 80, 1582 (1998) 46. A. Dixit, U.K. Sharma, A. Pradhan, New Astron. 73, 101281 (2019) 4. I. Zlatev, L. Wang, P.J. Steinhardt, Phys. Rev. Lett. 82(5), 896 (1999) 47. Y. Aditya, S. Mandal, P. Sahoo, D. Reddy, Eur. Phys. J. C 79(12), 1 (2019) 5. C. Armendariz-Picon, V.F. Mukhanov, P.J. Steinhardt, Phys. Rev. Lett. 85, 4438 (2000) 48. A.K. Yadav, Eur. Phys. J. C 81(1), 8 (2021) 49. U.K. Sharma, V.C. Dubey, A. Pradhan, Int. J. Geom. Meth. Mod. Phys. 17(02), 2050032 (2020) 6. R.R. Caldwell, Phys. Lett. B 545, 23 (2002) 79. A. Shaikh, arXiv:2105.04411 (2021) 79. A. Shaikh, arXiv:2105.04411 (2021) 80. P. Ens, A. Santos, EPL 131(4), 40007 (2020) 122. N. Aghanim et al., Astron. Astrophys. 641, A6 (2020 81. A. Jawad, S. Hussain, S. Rani, S. Qummer, Int. J. Geom. Meth. Mod. Phys. 17(08), 2050124 (2020) 123. U. Alam, V. Sahni, A.A. Starobinsky, JCAP 2004(06), 008 (2004) 82. A. Jawad, S. Hussain, Int. J. Geom. Meth. Mod. Phys. (2020) 124. A. Al Mamon, K. Bamba, Eur. Phys. J. C 78(10), 862 (2018) 125. A.A. Mamon, K. Bamba, S. Das, Eur. Phys. J. C 77(1), 29 (2017) 83. M.V. Santhi, Y. Sobhanbabu, Eur. Phys. J. C 80(12), 1 (2020) , , , y ( ), ( ) 126. A.A. Mamon, S. Das, Eur. Phys. J. C 77(7), 495 (2017) 6. A.A. Mamon, S. Das, Eur. Phys. J. C 77(7), 495 (2017) 84. A.A. Aly, M.A. Elrashied, M.M. Selim, Int. J. Mod. Phys. D 29(03), 2050023 (2020) 127. O. Farooq, F.R. Madiyar, S. Crandall, B. Ratra, Astrophys. J. 835(1), 26 (2017) 85. S. Rani, A. Jawad, K. Bamba, I.U. Malik, Symmetry 11(4), 509 (2019) 128. D.N. Spergel, L. Verde, H.V. Peiris, E. Komatsu, M.R. Nolta, C.L. Bennett, M. Halpern, G. Hinshaw, N. Jarosik, A. Kogut, M.Limon,S.S.Meyer,L.Page,G.S.Tucker,J.L.Weiland,E.Wol- lack, E.L. Wright, Astrophys. J. Suppl. Ser. 148(1), 175 (2003) 86. S.H. Shekh, P.H.R.S. Moraes, P.K. Sahoo, Universe 7(3), 67 (2021) 87. G. Varshney, U.K. Sharma, A. Pradhan, Eur. Phys. J. Plus 135(7), 541 (2020) 129. D. Valcin, R. Jimenez, L. Verde, J.L. Bernal, B.D. Wandelt, JCAP 08, 017 (2021) 88. U.K. Sharma, arXiv:2005.03979 (2020) 130. V. Sahni, T.D. Saini, A.A. Starobinsky, U. Alam, JETP Lett. 77, 201 (2003) 89. G. Varshney, U.K. Sharma, A. Pradhan, N. Kumar, Chin. J. Phys. 73, 1474 (2021) 90. M. Vijaya Santhi, Y. Sobhanbabu, New Astron. 89, 101648 (2021) 131. U. Alam, V. Sahni, T.D. Saini, A.A. Starobinsky, Mon. Not. R. Astron. Soc. 344, 1057 (2003) 91. S. Maity, U. Debnath, Int. J. Geom. Meth. Mod. Phys. 17(11), 2050170 (2020) 132. R.D. Gregory, Classical Mechanics (Cambridge University Press, 2006) 92. Y. Liu, Eur. Phys. J. Plus 136(5), 579 (2021) 133. D. Pavon, N. Radicella, Gen. Relativ. Gravit. 45, 63 (2013) 93. M. Zubair, L.R. Durrani, Chin. J. Phys. 69, 153 (2021) 94. S. Bhattacharjee, Eur. Phys. J. C 81(3), 1 (2021) 134. S.M. Crawford, A.L. Ratsimbazafy, C.M. Cress, E.A. Olivier, S.L. Blyth, K.J. van der Heyden, Mon. Not. R. Astron. Soc. y Phys. 17(02), 2050032 (2020) 7. G. ’t Hooft, Conf. Proc. C 930308, 284 (1993) 50. A. Saha, S. Ghose, Astrophys. Space Sci. 365(6), 98 (2020) 8. L. Susskind, J. Math. Phys. 36, 6377 (1995) 51. S. Ghaffari, A. Mamon, H. Moradpour, A. Ziaie, Mod. Phys. Lett. A 35(33), 2050276 (2020) 9. A.G. Cohen, D.B. Kaplan, A.E. Nelson, Phys. Rev. Lett. 82, 4971 (1999) 52. A.V. Astashenok, A.S. Tepliakov, Int. J. Mod. Phys. D 29(01), 1950176 (2020) 10. Y.S. Myung, Phys. Lett. B 610(1–2), 18 (2005) y g y 11. D. Pavón, J. Phys. A Math. Theor. 40(25), 6865 (2007) 53. S. Ghaffari, E. Sadri, A.H. Ziaie, Mod. Phys. Lett. A 35(14), 2050107 (2020) 12. L.N. Granda, A. Oliveros, Phys. Lett. B 669, 275 (2008 13. W. Fischler, L. Susskind, arXiv:hep-th/9806039 (1998) 13. W. Fischler, L. Susskind, arXiv:hep-th/9 54. A. Al Mamon, Mod. Phys. Lett. A 35(30), 2050251 (2020) 14. M. Li, Phys. Lett. B 603(1–2), 1 (2004) 15. C. Wetterich, Astron. Astrophys. 301, 321 (1995) 55. A. Al Mamon, A.H. Ziaie, K. Bamba, Eur. Phys. J. C 80(10), 1 (2020) 16. O. Bertolami, F.G. Pedro, M. Le Delliou, Phys. Lett. B 654(5–6), 165 (2007) ( ) 56. A. Jawad, A.M. Sultan, Adv. High Energy Phys. 2021 (2021) 56. A. Jawad, A.M. Sultan, Adv. High Energy Phys. 2 17. G. Olivares, F. Atrio-Barandela, D. Pavon, Phys. Rev. D 71(6), 063523 (2005) 57. S. Nojiri, S.D. Odintsov, T. Paul, Symmetry 13 58. V.C. Dubey, S. Srivastava, U.K. Sharma, A. Pradhan, Pramana 93(5), 1 (2019) 18. D. Pavon, W. Zimdahl, Phys. Lett. B 628(3–4), 206 (2005) 59. V. Chandra Dubey, A. Kumar Mishra, S. Srivastava, U. Kumar Sharma, Int. J. Geom. Meth. Mod. Phys. 17(01), 2050011 (2020) 19. W. Zimdahl, D. Pavón, Class. Quantum Gravity 24(22), 5461 (2007) 20. C. Gao, F. Wu, X. Chen, Y.G. Shen, Phys. Rev. D 79(4), 043511 (2009) 60. M. Abdollahi Zadeh, A. Sheykhi, K. Bamba, H. Moradpour, Mod. Phys. Lett. A 35(09), 2050053 (2019) 21. S. Nojiri, S.D. Odintsov, Gen. Relativ. Gravit. 38, 1285 (2006) S. Nojiri, S.D. Odintsov, Gen. Relativ. Gravit. 38, 1285 (20 12 3 3 3 399 Page 14 of 14 Eur. Phys. J. C (2022) 82 :399 106. V.C. Dubey, U.K. Sharma, A. Beesham, Int. J. Mod. Phys. D 28(15), 1950164 (2019) 61. U.K. Sharma, S. Srivastava, A. Beesham, Int. J. Geom. Meth. Mod. Phys. 17(07), 2050098 (2020) y 62. M. Korunur, Mod. Phys. Lett. y Phys. 17(02), 2050032 (2020) A 34(37), 1950310 (2019) 107. E. Ebrahimi, Astrophys. Space Sci. 365, 1 (2020) 63. S. Bhattacharjee, Astrophys. Space Sci. 365(6) (2020) 108. S. Bocquet, F.W. Carter, Int. J. Open Source Softw. Process. 1(6) (2016) 64. Q. Huang, H. Huang, J. Chen, L. Zhang, F. Tu, Class. Quantum Gravity 36(17), 175001 (2019) 109. D.M. Scolnic et al., Astrophys. J. 859(2), 101 (2018) 65. G.Varshney,U.K.Sharma,A.Pradhan,NewAstron.70,36(2019) 110. H. Yu, B. Ratra, F.Y. Wang, Astrophys. J. 856(1), 3 (2018) 111. H. Amirhashchi, S. Amirhashchi, Phys. Dark Univ. 29, 100557 (2020) 66. U.K. Sharma, A. Pradhan, Mod. Phys. Lett. A 34(13), 1950101 (2019) 112. L. Chen, Q.G. Huang, K. Wang, JCAP 02, 028 (2019) 67. A. Jawad, S. Rani, N. Azhar, Mod. Phys. Lett. A 34(07n08), 1950055 (2019) 113. C. Blake et al., Mon. Not. R. Astron. Soc. 418, 1707 (2011) 68. N. Zhang, Y.B. Wu, J.N. Chi, Z. Yu, D.F. Xu, Mod. Phys. Lett. A 35(08), 2050044 (2020) 114. D. Foreman-Mackey, D.W. Hogg, D. Lang, J. Goodman, Publ. Astron. Soc. Pac. 125(925), 306 (2013) 69. V. Srivastava, U.K. Sharma, New Astron. 78, 101380 (2020) 115. M. Newville, R. Otten, A. Nelson, A. Ingargiola, T. Stensitzki, D. Allan, A. Fox, F. Carter, P.D. Michał, et al. Lmfit/lmfit-py (2018) 69. V. S vastava, U. . S a a, New st o . 8, 0 380 ( 0 0) 70. A. Iqbal, A. Jawad, Phys. Dark Univ. 26, 100349 (2019) 70. A. Iqbal, A. Jawad, Phys. Dark Univ. 26, 100349 (2019) 71. A. Mohammadi, T. Golanbari, K. Bamba, I.P. Lobo, Phys. Rev. D 103(8), 083505 (2021) 116. E. Sadri, Eur. Phys. J. C 79(9), 1 (2019) 72. V.C. Dubey, U.K. Sharma, New Astron. 86, 101586 (2021) 117. Z.Y. Yin, H. Wei, Eur. Phys. J. C 79(8), 698 (2019) 73. R. D’Agostino, Phys. Rev. D 99(10), 103524 (2019) 118. C. Blake, E.A. Kazin, F. Beutler, T.M. Davis, D. Parkinson, S. Brough, M. Colless, C. Contreras, W. Couch, S. Croom et al., Mon. Not. R. Astron. Soc. 418(3), 1707 (2011) 74. M. Younas, A. Jawad, S. Qummer, H. Moradpour, S. Rani, Adv. High Energy Phys. 2019 (2019) Mon. Not. R. Astron. Soc. 418(3), 1707 (2011) 75. W. da Silva, R. Silva, Eur. Phys. J. Plus 136(5), 1 (2021) 119. D. Pavón, B. Wang, Gen. Relativ. Gravit. 41(1), 1 (2009) 76. A.A. Aly, Eur. Phys. J. Plus 134(7), 1 (2019) 120. C.P. Singh, A. Kumar, Phys. Rev. y Phys. 17(02), 2050032 (2020) D 102, 123537 (2020) 77. M. Sharif, S. Saba, Symmetry 11(1) (2019) 121. G. Hinshaw, J. Weiland, R. Hill, N. Odegard, D. Larson, C. Ben- nett, J. Dunkley, B. Gold, M. Greason, N. Jarosik, E. Komatsu, M.Nolta,L.Page,D.Spergel,E.J.Wollack,M.Halpern,A.Kogut, M. Limon, S. Meyer, G. Tucker, E. Wright, Astrophys. J. Suppl. Ser. 180, 225 (2008) 78. A. Jawad, A. Aslam, S. Rani, Int. J. Mod. Phys. D 28(11), 1950146 (2019) 79. A. Shaikh, arXiv:2105.04411 (2021) 406, 2569 (2010) 95. A. Jawad, K. Bamba, M. Younas, S. Qummer, S. Rani, Symmetry 10(11) (2018) 135. J.C. Carvalho, J.S. Alcaniz, Mon. Not. R. Astron. Soc. 418, 1873 (2011) 96. A. Pradhan, A. Dixit, New Astron., 101636 (2021) 97. A. Lymperis, E.N. Saridakis, Eur. Phys. J. C 78(12), 136. J. Simon, L. Verde, R. Jimenez, Phys. Rev. D 71, 123001 (2005) 98. A. Sheykhi, Phys. Lett. B 785, 118 (2018) 98. A. Sheykhi, Phys. Lett. B 785, 118 (2018) J. Simon, L. Verde, R. Jimenez, Phys. Rev. D 71, 123001 99. S. Nojiri, S.D. Odintsov, E.N. Saridakis, Eur. Phys. J. C 79(3), 1 (2019) 137. D. Stern, R. Jimenez, L. Verde, M. Kamionkowski, S.A. Stanford, JCAP 02, 008 (2010) 138. B. Wang, Y. Gong, E. Abdalla, Phys. Rev. D 74, 083520 (2006) 100. S. Nojiri, S.D. Odintsov, E.N. Saridakis, R. Myrzakulov, Nucl. Phys. B. 950, 114850 (2020) 139. C.A. Egan, C.H. Lineweaver, Astrophys. J. 710(2) 101. K. Abbasi, S. Gharaati, arXiv:2006.01763 (2020) 140. P.B. Krishna, T.K. Mathew, arXiv:2002.02121 141. G.W. Gibbons, S.W. Hawking, Phys. Rev. D 15, 2738 (1977) 102. M. Asghari, A. Sheykhi, arXiv:2106.15551 (2021) 103. P. George, V.M. Shareef, T.K. Mathew, Int. J. Mod. Phys. D 28(04), 1950060 (2018) 142. E.N. Saridakis, S. Basilakos, Eur. Phys. J. C 81(7), 644 (2021) 143. T.K. Mathew, R. Aiswarya, V.K. Soman, Eur. Phys. J. C 73, 2619 (2013) 104. A.A. Starobinsky, JETP Lett. 68, 757 (1998) 105. E.N.Saridakis,K.Bamba,R.Myrzakulov,F.K.A 105. E.N.Saridakis,K.Bamba,R.Myrzakulov,F.K.Anagnostopoulos, JCAP 2018(12) 012 (2018) 105. E.N.Saridakis,K.Bamba,R.Myrzakulov,F.K.Anagnostopoulos, JCAP 2018(12), 012 (2018) 12 3
https://openalex.org/W4242322617
https://www.qeios.com/read/9XNK6M/pdf
English
null
DNA Recombination Process
Definitions
2,020
cc-by
109
Qeios · Definition, February 2, 2020 Open Peer Review on Qeios Open Peer Review on Qeios DNA Recombination Process National Cancer Institute National Cancer Institute Qeios ID: 9XNK6M · https://doi.org/10.32388/9XNK6M Source National Cancer Institute. DNA Recombination Process. NCI Thesaurus. Code C17082. National Cancer Institute. DNA Recombination Process. NCI Thesaurus. Code C17082. Any type of DNA sequence rearrangement involving possible exchange and frequent loss of segments of DNA by either homologous recombination or non-homologous end- joining of broken DNA strands. Recombination is involved in assorting genes during reproduction, in repairing broken DNA, and in maintaining cell viability at the expense of long-term genomic stability. Qeios ID: 9XNK6M · https://doi.org/10.32388/9XNK6M 1/1
https://openalex.org/W2058633911
https://europepmc.org/articles/pmc3787565?pdf=render
English
null
Association between Angiotensin II Type 1 Receptor Polymorphism and Sudden Cardiac Death in Myocardial Infarction
Disease markers
2,013
cc-by
5,563
Hindawi Publishing Corporation Disease Markers Volume 35 (2013), Issue 5, Pages 287–293 http://dx.doi.org/10.1155/2013/731609 Hindawi Publishing Corporation Disease Markers Volume 35 (2013), Issue 5, Pages 287–293 http://dx.doi.org/10.1155/2013/731609 Hindawi Publishing Corporation Disease Markers Volume 35 (2013), Issue 5, Pages 287–293 http://dx.doi.org/10.1155/2013/731609 Correspondence should be addressed to Peter Kruzliak; peter.kruzliak@savba.sk Correspondence should be addressed to Peter Kruzliak; peter.kruzliak@savba.sk Received 17 June 2013; Accepted 21 August 2013 Received 17 June 2013; Accepted 21 August 2013 Academic Editor: Grant Izmirlian Academic Editor: Grant Izmirlian Copyright © 2013 Peter Kruzliak et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Objective. The renin-angiotensin system is involved in the pathogenesis of coronary artery disease and myocardial infarction (MI). Angiotensin II (Ang II) has many adverse effects such as vasoconstriction and vascular remodeling, and these actions are mediated by the angiotensin II type 1 receptor (AT1R). Patients and Methods. A total of 1376 patients were recruited from January 2010 to April 2012. The study group consisted of 749 patients with ACS (317 females and 432 males) and of 627 healthy controls. Results. The ACS patients demonstrated a lower proportion of AA genotypes and AC genotypes but higher proportions of CC genotypes than the control population. The AT1R CC genotype conferred a 2.76-fold higher risk of MI compared with the genotype AC and AA. In addition, the CC genotype was also associated with a 4.08 times higher risk of left anterior descending artery infarction and a 3.07 times higher risk of anterior wall infarction. We also found that the CC genotype was independently associated with sudden cardiac death. In Summary. This study demonstrated that the AT1R CC genotype is an independent risk factor for ACS incidence, and this genotype is associated with a greater ACS severity and greater risk of sudden cardiac death. Peter Kruzliak,1,2 Gabriela Kovacova,3 Olga Pechanova,1 and Stefan Balogh 1 Institute of Normal and Pathological Physiology and Centre of Excellence for Regulatory Role of Nitric Oxide in Civilisation Diseases, Slovak Academy of Sciences, Sienkiewiczova 1, 813 71 Bratislava, Slovakia y f 2 Department of Cardiovascular Diseases, International Clinical Research Center, St. Anne’s Faculty Hospital and Masaryk University, Pekarska 53, 602 00 Brno, Czech Republic 3 5th Department of Internal Medicine, University Hospital and Medical Faculty, Comenius University, Ruzinovska 6, 826 06 Bratislava, Slovakia 2. Methods The study was designed as a case-control study. Aim of this study was to determine the prevalence of polymorphisms of the gene for angiotensin II type 1 receptor and compare it with the healthy population. the presence or assess the extension of coronary artery dis- ease, and patients with obstructive lesions greater than 50% were selected for the study. Total cholesterol and triglyceride plasma levels were performed on automated analyzer Hitachi 917 Germany, using commercial kits supplied by Roche diag- nostics (Mannheim Germany). Demographic and clinical characteristics are in Table 1. the presence or assess the extension of coronary artery dis- ease, and patients with obstructive lesions greater than 50% were selected for the study. Total cholesterol and triglyceride plasma levels were performed on automated analyzer Hitachi 917 Germany, using commercial kits supplied by Roche diag- nostics (Mannheim Germany). Demographic and clinical characteristics are in Table 1. A total of 1376 patients were recruited from January 2010 to April 2012 in the 5th Department of Internal Medicine, University Hospital Bratislava, Slovak Republic. The study conforms with the Declaration of Helsinki and Institutional Ethics Committee of the University Hospital Bratislava, and written informed consent was obtained from all the subjects.h 2.1. Genotyping Methods. Blood was collected in ethylene- diaminetetraacetic acid. DNA was isolated from leukocytes according to standard procedures using proteinase K. DNA segments were amplified by polymerase chain reaction (PCR) in a total volume of 15 𝜇L containing 0.1 𝜇L Taq, 1.5 𝜇L buffer, 2.5 𝜇L MgCl2, 0.5 𝜇L dNTP, 0.5 𝜇L of each primer, and 7.4 𝜇L H2O. The primers used were 5󸀠-AGAAGCCTGCACCAT- GTTTT-3󸀠(sense) and 5󸀠-TGTGGCTTTGCTTTGTCTTG- 3󸀠(antisense). The reaction conditions were as follows: initial denaturation 95∘C 5 min, then 33 cycles of denaturation 94∘C for 30 s, annealing 53∘C for 25 s, elongation 72∘C for 25 s, final elongation 72∘C for 10 min, and cooling 10∘C for 10 min. The PCR product (233 bp) was digested by 5 U of Dde I for 12 h. The fragments were separated by electrophoresis using 3% agarose gel at 85 V and visualized by ethidium bromide staining under UV light. The AA variant was detected as one fragment (233 bp), AC as three fragments (233, 118, 115 bp), and CC as two fragments (118 and 115 bp). 1. Introduction in the pathogenesis of coronary artery disease and MI [2– 5]. Angiotensin II (Ang II) is a key component of the renin- angiotensin system formed by action of the angiotensin- converting enzyme on the precursor molecule angiotensin I. It has many adverse effects, such as vasoconstriction, cellular proliferation, vascular remodeling, and aldosterone secre- tion, and it contributes to endothelial dysfunction and athe- rosclerosis by promotion of oxidative stress [6, 7]. All these actions are mediated by the angiotensin II type 1 receptor (AT1R).h Acute myocardial infarction (MI) is defined as death or necrosis of myocardial cells due to an inadequate amount of oxygen supply to the heart. The classical symptoms of MI are shortness of breath, anxiety, chest pain typically radiating to the left arm or left side of the neck, vomiting, and palpitations. Important risk factors include previous history of vascular disease, such as atherosclerosis, angina, previous heart attack or stroke, and age—especially men over 40 and women over 50 years [1].h The AT1R gene is located on chromosome 3q21-q25, its length is >55 kb, and it is composed of five exons and four introns. It belongs to the superfamily of G-protein-coupled The renin-angiotensin system plays an important role in blood pressure regulation and homeostasis, and it is involved 288 Disease Markers Table 1: Demographic and clinical characteristics of ACS patients and healthy controls. Table 1: Demographic and clinical characteristics of ACS patients and healthy controls. receptors. The gene polymorphism in the present study was located at the untranslated 3󸀠region (A1166C), corresponding to an A →C transversion in the position of nucleotide 1166 of the mRNA sequence, resulting in an A/C polymorphism [8, 9]. Table 1: Demographic and clinical characteristics of ACS patients and healthy controls. 1. Introduction ACS patients (𝑛= 749) Healthy controls (𝑛= 627) Mean age (years) 61.4 ± 9.7 54 ± 10.3 Sex (male/female) 432/317 354/273 BMI (kg/m2) 28.4 ± 4.6 27.3 ± 5.4 Smokers 𝑛(%) 328 (43.79%) 163 (25.99%) Fasting glucose (mmol/L) 6.4 ± 1.4 4.9 ± 1.2 Diabetes mellitus (%) 329 (43.65%) 0 (0%) HbA1c (%) 7.3 ± 2.4 3.4 ± 1.5 Arterial hypertension 𝑛(%) 377 (50.33%) 0 (0%) Systolic blood pressure (mmHg) 154 ± 16.4 124 ± 8.6 Diastolic blood pressure (mmHg) 93.6 ± 10.4 82.3 ± 6.7 Dyslipidemia (%) 213 (28.43%) 49 (7.81%) Total cholesterol (mmol/L) 8.2 ± 1.7 4.3 ± 0.9 HDL-cholesterol (mmol/L) 1.5 ± 0.2 1.3 ± 0.3 LDL-cholesterol (mmol/L) 5.9 ± 1.6 2.7 ± 0.9 Triglycerides (mmol/L) 2.1 ± 0.4 1.3 ± 0.2 Studies have been carried out to find an association between the A/C polymorphism of AT1R and MI. Some stud- ies have revealed that the CC genotype is not associated with MI [10, 11], while others have observed that the CC homozygous genotype is associated with MI and coronary heart disease [12–14]. Tiret et al. found significant association between angiotensin type II receptor polymorphism and acute coronary syndrome. On the other hand, in the study by Gardemann et al. There was no detected association between angiotensin II type 1 receptor A1166C gene polymorphism and coronary artery disease. In view of these controversial reports, the present study was aimed at assessing the associ- ation of the A1166C polymorphism of the AT1 receptor gene with MI and sudden cardiac death. 3. Results Comparisons of the distributions of demographic and rele- vant clinical risk factors between ACS patients and healthy controls and the estimated OR for each risk factor are listed in Table 2. There was no significant difference in either the age or gender distribution between the two groups included. The prevalence of arterial hypertension (AH), dyslipidemia, and diabetes mellitus (DM) was significantly increased in acute MI patients as compared to healthy controls. Diastolic blood pressure (DBP) and systolic blood pressure (SBP), total cholesterol, and LDL-cholesterol were also higher in MI patients than in healthy controls. In the univariate logistic regression model, older age, higher systolic blood pressure or diastolic blood pressure and the presence of diabetes mellitus were found to be significant risk factors for ACS.h The AT1R is a major component of the renin-angiotensin system. It mediates most of the classical and biological func- tions of angiotensin II [20]. It belongs to the G-protein- coupled receptor superfamily and is a peptide containing 360 amino acids. Binding of angiotensin II to the AT1R in vascular smooth muscle cells results in phospholipase C-mediated generation of inositol 1,4,5-trisphosphate and diacylglycerol. Inositol 1,4,5-trisphosphate mobilizes calcium from the endo- plasmic reticulum, and calcium and diacylglycerol activate enzymes, including protein kinase C and calcium/calmodu- lin-activated kinases. Ang II acts as a mitogen in vascular smooth muscle cells by activating several signaling pathways, such as that of phospholipase C, phospholipase A2, and phos- pholipase D, as well as activating a large number of kinases, such as tyrosine kinases, mitogen-activated protein kinases (MAPKs), c-src kinase, Janus-associated tyrosine kinase, and receptors with tyrosine-kinase activity. Ang II also stimulates transcription factors, such as the activating protein, signal transduction and transcription activators (STATs), and the nuclear factor kappa B (Nf𝜅B) [6, 20–22]. The AT1R has been shown to be involved in the process of vascular hypertrophy, sodium retention, cardiac remodeling, hypertension and fibrinogenesis [23–25]. The AT2 receptor plays a counter- regulatory role [6, 26, 27]. i The distribution of the AT1R genotypes was in agreement with the Hardy-Weinberg equilibrium. The frequency of the A1166C polymorphism was significantly different between patients and controls (𝑃= 0.024). Table 3 illustrates that ACS patients demonstrated a lower proportion of AA genotypes (32.17%) but higher proportions of CC genotypes (23.79%) than the control population (AA 50.39%, AC 42.90%, CC 6.69%). 3. Results The AT1R CC genotype conferred a 2.76-fold risk of ACS compared with the genotype AC and AA (95% CI: 1.07–6.49, 𝑃= 0.004, logistic regression). After controlling for other risk factors, the CC genotype was still significantly associated with ACS, conferring a 4 times higher risk (OR = 4.295; 95% CI: 1.436–12.851). To further evaluate the etiologic effects of ATR1 polymorphisms in ACS, we analyzed the association between AT1R genotypes and different ACS manifestations (UAP and acute MI), relevant clinical risk fac- tors (hypertension, diabetes mellitus, hypercholesterolemia, systolic blood pressure, diastolic blood pressure, and body mass index), clinical history (heart failure, CAD, MI, and family history), and history of smoking. As described by Duncan et al. (2001), many polymor- phisms of the AT1R gene have been identified, but the A1166C polymorphism has been the most extensively studied. The physiological significance of this polymorphism is uncertain because of its location in the 3󸀠-untranslated region of the gene [28].h There are controversial reports regarding the role of AT1R gene A/C polymorphism as a risk factor for MI. Some studies have reported a positive association [12–14], while others could not find any association [10, 11, 28, 29]. i y In the study by Berge et al. (1997), 247 survivors of myocardial infarction (MI) were compared to 384 controls with respect to distribution of genotypes and gene frequen- cies in the A1166C polymorphism at the AT1R locus. No differences in allele frequencies or genotype distribution were observed when all patients were compared with all controls. When they compared CC homozygotes with the combined group of CA heterozygotes and AA homozygotes (CA/AA), a difference in borderline significance between the MI group and controls was observed (𝑃= 0.05). In males alone, this difference was much more pronounced because of the larger proportion of males with the CC genotype in MI cases than in male controls (𝑃= 0.01). No significant differences were observed between female cases and controls. When was the subjects subdivided into a “low-risk” and a “high-risk” group, based on levels of apolipoprotein B (apoB) and body mass index (BMI), and whether or not the person used lipid- lowering drugs, the frequency of CC homozygotes in male cases of the “low-risk” group differed significantly compared to the frequency in male controls of the “low-risk” group As shown in Table 4, the CC genotype conferred a signif- icant 3.35-fold risk of acute MI (Table 2. 2. Methods All statistical tests were two sided, and the level of significance was set at 0.05. 2. Methods j The study group consisted of 749 patients with acute coronary syndrome (ACS patients—317 females and 432 males) with a mean age of 61.4 ± 9.7 years. All patients were admitted with acute coronary syndrome, and on the basis of typical ECG changes, elevated cardiac markers, clinical history, and coronary angiography (50% stenosis affected at least one major coronary vessel) the diagnosis was confirmed as MI. Blood samples were collected from patients after 12 to 14 hours of fasting. Simultaneously, blood samples were collected from 627 healthy, age- and sex- matched controls (male : female = 354 : 273) with mean age 54 ± 10.3. All controls were nonhypertensive and nondiabetic. Information on height, weight, body mass index, cigarette smoking, hypertension, diabetes, family history of coronary artery disease, and diabetes was collected using a structured questionnaire. Acute coronary syndrome was diagnosed on the basis of recommendations of European Society of Cardi- ology (ESC) [15, 16]. Hypertension was defined according to ESC guidelines as systolic blood pressure >140 mmHg and/or diastolic blood pressure >90 mmHg, based on the average of two blood pressure measurements, or a patient’s self-reported history of hypertension [17]. Diabetes was diagnosed if fasting plasma glucose was >110 mg/100 mL or the patient was on antidiabetic medications [18]. The reference value for cholesterol was 200 mg/dL and for triglycerides 130– 150 mg/dL. Dyslipidemia was diagnosed according to ESC guidelines [19]. Coronary angiography was used to confirm 2.2. Statistical Methods. Patient’s age at diagnosis was record- ed as a continuous variable and shown as mean ± standard deviation. The odds ratios (ORs) and corresponding 95% confidence intervals (CIs) for assessing the effect of the AT1R A/C genotype distribution and allele frequencies on ACS were calculated by logistic regression analysis with adjustment for relevant significant variables (Tables 4 and 5). All the statistical analyses were performed using SPSS 16.0 289 Disease Markers patients with ACS (Table 5, OR = 6.48, 95% CI: 1.04–40.60, 𝑃= 0.036, logistic regression). patients with ACS (Table 5, OR = 6.48, 95% CI: 1.04–40.60, 𝑃= 0.036, logistic regression). patients with ACS (Table 5, OR = 6.48, 95% CI: 1.04–40.60, 𝑃= 0.036, logistic regression). patients with ACS (Table 5, OR = 6.48, 95% CI: 1.04–40.60, 𝑃= 0.036, logistic regression). and Excel 2007. All statistical tests were two sided, and the level of significance was set at 0.05. and Excel 2007. 3. Results ATR1 A1166C Polymorphism ACS patients (𝑛= 749) Healthy controls (𝑛= 627) OR CC (%) 179 (23.89%) 42 (6.69%) OR = 2.76 (1.07–6.49), P = 0.004 AC (%) 329 (43.92%) 269 (42.90%) OR = 1.17 (0.79–2.1), P = 0.603 AA (%) 241 (32.17%) 316 (50.39%) OR = 0.43 (0.23–0.76), P = 0.015 C (%) 508 (67.82%) 311 (49.60%) OR = 1.68 (1.17–2.41), P = 0.004 A (%) 570 (76.10%) 585 (93.30%) OR = 0.59 (0.41–0.85), P = 0.004 AT1R genes and coronary artery disease in 2244 Caucasian men [31]. Table 3: Distribution of AT1R A/C gene polymorphisms. Table 3: Distribution of AT1R A/C gene polymorphisms In this study no association was detected between angiot- ensin II type 1 receptor A1166C gene polymorphism and coro- nary artery disease. Similarly, there was no link to myocardial infarction. Rice et al. (1999) demonstrated the absence of a synergic association with myocardial infarction, although they found a weak association between some genotypes (AC/II and CC/DD) and coronary stenosis [32]. y de Ara´ujo et al. (2004) found no association between A1166C polymorphism of the AT1R gene and the severity of the coronary lesions, when comparing the genotypes CC versus AC versus AA [11]. It was a prospective cross-sec- tional study with 110 consecutive patients (66.6% of the male sex and 36.4% of the female sex, mean age = 61.82 ± 10.81 years) diagnosed with acute myocardial infarction based on clinical, electrocardiographic, and enzymatic data and confirmed through coronary angiography. The control group comprised 104 patients (57.7% of the female sex and 42.3% of the male sex, mean age = 56.69 ± 11.52 years) who were renal transplant donors or had valvular heart disease or atypical chest pain but had no lesions in the coronary arteries on coronary angiography. The genotypic frequency in the infarcted patients was as follows: AA = 54.5%; AC = 35.5%; and CC = 10%, which was similar and nonsignificant in regard to that in the control group without myocardial infarction (𝑃= 0.83). No risk increase occurred for acute myocardial infarction when comparing the genotypes as follows: CC versus AA (OR = 1.35; 95% CI = 0.50–3.59); AC versus AA (OR = 1.03; 95% CI = 0.58–1.84); and AA + AC versus AA (OR = 1.33; 95% CI = 0.51–3.45). (𝑃< 0.001). 3. Results 95% CI: 1.111– 10.115, 𝑃= 0.032, logistic regression); however, there was no significant association between A1166C polymorphism and any other risk factor among the recruited ACS group. Subsequent analyses evaluated the possible influence of A1166C polymorphism on final ACS clinical outcome. The association between AT1R A/C polymorphism and ACS severity and mortality was determined. Interestingly, ACS patients with the CC genotype had a significantly higher risk of presenting with a greater number of stenosed vessels, with a 3.87 times increased risk of stenosis in all three coronary arteries (Table 5, 95% CI: 1.09–13.81, 𝑃= 0.037, logistic regression). In addition, the CC genotype was also associated with a 4.08-fold risk of left anterior descending artery infarction (Table 5, 95% CI: 1.04–9.12, 𝑃= 0.042) and a 3.07-fold risk of anterior wall infarction (Table 5, 95% CI: 1.04–9.12, 𝑃= 0.024). The CC genotype was found to be independently associated with sudden cardiac death in 290 Disease Markers Table 2: Distributions of demographic and relevant clinical risk factors between ACS patients and healthy controls and the estimated OR for each risk factor. Table 2: Distributions of demographic and relevant clinical risk factors between ACS patients and healthy controls and the estimated OR for each risk factor. Subject characteristics ACS patients (𝑛= 749) Controls (𝑛= 627) Univariate analysis P Multivariate analysis P OR (95% CI) OR (95% CI) AT1R-genotypes AA 241 316 1 (ref.) 1 (ref.) AC 329 269 1.17 (0.79–2.1) 0.603 1.284 (0.595–2.773) 0.524 CC 179 42 2.76 (1.07–6.49) 0.004 4.295 (1.436–12.851) 0.009 Mean age 61.4 ± 9.7 54 ± 10.3 0.75 (0.24–2.76) 0.98 1.25 (0.46–2.96) 0.425 Sex—male 432 354 1.53 (0.89–2.62) 0.123 — — Systolic blood pressure ≥140 mmHg 327 97 9.73 (3.44–17.16) <0.001 6.97 (2.53–14.37) 0.003 Diastolic blood pressure ≥90 mmHg 94 23 5.92 (1.47–9.81) 0.002 3.65 (1.45–10.40) 0.004 Diabetes mellitus 329 0 14.22 (4.98–25.26) <0.001 12.75 (3.41–23.58) 0.001 Table 3: Distribution of AT1R A/C gene polymorphisms. 3. Results No interaction between the insertion/deletion (I/D) polymorphism at the angiotensin I-converting enzyme (ACE) locus and the polymorphism at the AT1R locus was detected [13]. Tiret et al. (1994) described a synergic association of DD (I/D) homozygotes of the ACE and CC gene of the A1166C polymorphism of the AT1R gene. For these authors, this gene interaction of risk for myocardial infarction might suggest a possible epistatic effect of the two genes, assuming that the C allele of AT1R is associated with a modified response of the angiotensin II receptor, thus modulating the possible risk conferred by the D allele of the ACE gene [14]. ´Alvarez et al. (1998) in a young Spanish population developed a case- control study to determine the I/D genotypes of the ACE gene and A1166C of the gene of receptor 1 of angiotensin II. Separately, no association was observed; but the ACE-DD and AT1R-CC genotypes interacting synergically were associated with coronary artery disease (OR = 5.32; 95% CI, 1.45–19.51) [30]. Based on these discrepancies, we decided to study the association between AT1R A/C polymorphism and myocar- dial infarction in the Slovak population. To our knowledge, no study dealing with the relationship between the AT1R A/C polymorphism and sudden cardiac death has as yet been published. For this reason we intended to determine whether AT1R A/C polymorphism was an independent risk factor for sudden cardiac death.h This study demonstrated that the CC genotype acted as an independent risk factor for ACS and in particular for In contrast, Gardemann et al. (1998) found no synergic interaction between the polymorphic variants of the ACE and 291 Disease Markers Table 4: Association between AT1R A/C polymorphism and ACS risk factors. 3. Results ATR1 genotypes AA (𝑛= 241) n AC (𝑛= 329) n Univariate P CC (𝑛= 179) n Univariate P ACS risk factors OR (95% CI) OR (95% CI) UAP 145 214 1.37 (0.31–2.18) 0.273 66 1.24 (0.73–2.98) 0.148 AMI 96 115 1.21 (0.51–2.834) 0.670 113 3.35 (1.11–0.12) 0.032 Age (mean) 65.7 ± 11.1 61.2 ± 13.5 0.98 (0.95–1.01) 0.232 60.3 ± 14.1 0.97 (0.94–1.01) 0.149 Sex—male 137 146 1.35 (0.43–7.54) 0.450 149 1.18 (0.38–3.72) 0.776 Hypertension 131 143 0.57 (0.23–1.44) 0.236 103 0.40 (0.14–1.21) 0.104 Diabetes mellitus 117 137 0.88 (0.36–2.16) 0.780 75 1.36 (0.43–4.33) 0.603 Hypercholesterolemia 54 107 0.63 (0.21–1.87) 0.403 52 0.92 (0.23–3.68) 0.905 BMI (kg/m2) ≥27 17 26 2.16 (0.65–9.33) 0.151 13 1.67 (0.22–9.73) 0.622 Systolic blood pressure ≥140 mmHg 109 131 0.50 (0.20–1.25) 0.137 87 1.13 (0.39–3.24) 0.822 Diastolic blood pressure ≥90 mmHg 28 40 0.36 (0.12–1.10) 0.143 26 1.18 (0.38–3.72) 0.776 Heart failure history 21 32 0.88 (0.31–2.51) 0.817 16 1.59 (0.49–5.18) 0.445 CAD history 143 204 1.67 (0.37–2.07) 0.755 107 0.95 (0.17–3.62) 0.257 MI history 36 42 1.23 (0.28–5.51) 0.785 27 1.08 (0.17–7.00) 0.936 Family history of ACS 12 17 0.35 (0.03–4.01) 0.399 9 0.80 (0.07–9.30) 0.855 Family history of diabetes 21 33 0.47 (0.07–2.03) 0.576 17 0.97 (0.25–2.46) 0.767 Smoking 105 136 0.89 (0.38–2.09) 0.793 87 1.28 (0.45–3.64) 0.639 UAP: unstable angina pectoris, AMI: acute myocardial infarction, BMI: Body mass index, and CAD: coronary artery disease. Table 4: Association between AT1R A/C polymorphism and ACS risk factors. ion between AT1R A/C polymorphism and ACS severity and myocardial infarct related sudden cardiac death. Table 5: Association between AT1R A/C polymorphism and ACS severity and myocardial infarct related sudden cardiac death. 3. Results ACE genotypes AA (𝑛= 241) n AC (𝑛= 329) n Univariate P CC (𝑛= 179) n Univariate P Clinical symptoms OR (95% CI) OR (95% CI) Stenosis numbers 0 or 1 vessel 97 149 1.384 (0.679–3.243) 0.665 51 2.154 (0.821–4027) 0.294 2 vessels 76 107 1.267 (0.435–3.686) 0.665 49 2.006 (0.729–5.515) 0.178 3 vessels 68 73 2.111 (0.530–8.407) 0.289 79 3.870 (1.085–13.812) 0.037 Infarcted left anterior descending artery 109 154 2.166 (0.912–5.144) 0.080 117 4.080 (1.041–9.118) 0.024 Infarcted left circumflex 93 124 1.477 (0.630–3.460) 0.370 76 1.744 (0.292–2.436) 0.753 Infarcted right coronary 98 137 1.174 (0.498–2.764) 0.714 74 0.779 (0.275–2.211) 0.639 Anterior infarction wall 113 157 1.727 (0.279–3.893) 0.514 119 3.073 (1.039–9.091) 0.043 Inferior infarction wall 97 126 2.429 (0.609–9.679) 0.208 81 2.386 (0.482–11.803) 0.286 Sudden cardiac death 4 6 1.489 (0.258–8.595) 0.656 13 6.484 (1.036–40.598) 0.036 Table 5: Association between AT1R A/C polymorphism and ACS severity and myocardial infarct related su genetic variants, affecting the structure or expression of the AT1R.f acute MI (Table 2). In addition, the CC genotype was also found to be associated with greater ACS severity, including more stenosed vessels, greater occlusion of the left anterior descending branch, and a greater risk of anterior wall myocar- dial infarction (Table 5). Finally, we focused this study on the effect of AT1R A/C polymorphisms on sudden cardiac death within 24 hours of ACS patients registered in the emergency room. AT1R CC genotype carriers had a higher mortality risk than AC or AA carriers. Genetic polymorphisms in AT1R-related genes have been independently associated with higher risk of MI-related sudden cardiac arrest. Other studies reported that CC genotype is to be associ- ated with atherosclerosis [25], arterial hypertension [27], and coronary artery disease [12, 33]. Positive associations between A1166C polymorphism and disease may be the result of linkage disequilibrium with an- other polymorphism of functional importance, either within the AT1R gene or within a nearby one [34]. The A/C transver- sion per se does not characterize any functional diversity. Although there is no evidence to support this hypothesis, this polymorphism can be considered as a possible marker, in linkage disequilibrium with other functionally relevant References [17] G. Mancia, G. de Backer, A. Dominiczak, R. Cifkova, R. Fagard, and G. Germano, “Guidelines for the management of arterial hypertension: the task force for the management of arterial hypertension of the european society of hypertension and of the european society of cardiology,” Journal of Hypertension, vol. 25, pp. 1105–1187, 2007. [1] E. Braunwald, Ed., Heart Disease: A Textbook of Cardiovascular Medicine, WB Saunders, Philadelphia, Pa, USA, 1997. [2] E. Ambrosioni, S. Bacchelli, D. D. Esposti, and C. Borghi, “ACE- inhibitors and atherosclerosis,” European Journal of Epidemiol- ogy, vol. 8, no. 1, pp. 129–133, 1992.h [3] T. W. Kurtz, “The ACE of hearts,” Nature, vol. 359, no. 6396, pp. 588–589, 1992. [18] L. Ryd´en, E. Standl, B. Małgorzata et al., “Guidelines on diabetes, pre-diabetes, and cardiovascular diseases: executive summary. The task force on diabetes and cardiovascular diseases of the european society of cardiology (ESC) and of the european association for the study of diabetes (EASD),” European Heart Journal, vol. 28, no. 1, pp. 88–136, 2007. [4] D. R. J. Singer, C. G. Missouris, and S. Jeffery, “Angiotensin- converting enzyme gene polymorphism: what to do about all the confusion?” Circulation, vol. 94, no. 3, pp. 236–239, 1996. [5] M. M. Morales-Su´arez-Varela, M. L. Mansego, A. M. Vicedo- Cabrera et al., “Inefficient arterial hypertension control in pa- tients with metabolic syndrome and its link to renin-angiotens- in-aldosterone system polymorphisms,” Hypertension Research, vol. 34, no. 6, pp. 758–766, 2011. [19] Z. Reiner, A. L. Catapano, G. de Backer et al., “ESC/EAS Guide- lines for the management of dyslipidaemias. The task force for the management of dyslipidaemias of the european society of cardiology (ESC) and the european atherosclerosis society (EAS),” European Heart Journal, vol. 32, no. 4, pp. 1769–1818, 2011. [6] I. A. Reid, “The renin-angiotensin system: physiology, patho- physiology, and pharmacology,” The American Journal of Physi- ology, vol. 275, no. 6, pp. S236–S245, 1998. [20] D. F. Guo, Y. L. Sun, P. Hamet, and T. Inagami, “The angiot- ensin II type 1 receptor and receptor-associated proteins,” Cell Research, vol. 11, no. 3, pp. 165–180, 2001. [7] M. G. Nicholls, A. M. Richards, and M. Agarwal, “The impor- tance of the renin-angiotensin system in cardiovascular dis- ease,” Journal of Human Hypertension, vol. 12, no. 5, pp. 295–299, 1998. [21] W. C. Paxton, Ruglssi, C. Horulst et al., “Imunohistochemical localization of rat angiotensin II at 1 receptor,” The American Journal of Physiology, vol. Disclosure [14] L. Tiret, A. Bonnardeaux, O. Poirier et al., “Synergistic effects of angiotensin-converting enzyme and angiotensin-II type 1 receptor gene polymorphisms on risk of myocardial infarction,” The Lancet, vol. 344, no. 8927, pp. 910–913, 1994. This study was elaborated within the projects. The authors declare that they have no conflict of interests. The authors declare that they have no conflict of interests. Acknowledgment This work was supported by Grant of European Region- al Development Fund-Project FNUSA ICRC (no. CZ.1.05/ 1.1.00/02.0123) and VEGA 2/0190/11. This work was supported by Grant of European Region- al Development Fund-Project FNUSA ICRC (no. CZ.1.05/ 1.1.00/02.0123) and VEGA 2/0190/11. [16] P. G. Steg, S. K. James, D. Atar et al., “ESC Guidelines for the management of acute myocardial infarction in patients present- ing with ST-segment elevation,” European Heart Journal, vol. 33, no. 20, pp. 2569–2619, 2012. 5. Conclusions In summary, this case-control study showed that the AT1R CC genotype was an independent risk factor for ACS inci- dence and especially for acute MI. In addition, the AT1R CC genotype was associated with greater ACS severity and 292 Disease Markers a greater risk of sudden cardiac death. This experience of patients requiring urgent ACS treatment suggests that AT1R CC genotype carriers should be followed up and treated especially carefully because of the higher relative risk of more severe ACS and sudden cardiac death. Further studies are necessary to evaluate the personal pharmacogenomic and pharmacogenetic effects of AT1R A/C polymorphisms to improve ACS therapeutic strategies and prognosis. In future, examination of this polymorphism could become a part of the risk stratification of patients with acute coronary syndromes and predictor of sudden cardiac death. [11] M. A. de Ara´ujo, B. S. Menezes, C. Lourenc¸o, E. R. Cordeiro, R. Rispoli Gatti, and L. R. Goulart, “The A1166C polymorphism of the angiotensin II type-1 receptor in acute myocardial infarc- tion,” Arquivos Brasileiros de Cardiologia, vol. 83, no. 5, pp. 404– 408, 2004. [12] M. Buraczy´nska, Z. Pijanowski, D. Spasiewicz et al., “Renin- angiotensin system gene polymorphisms: assessment of the risk of coronary heart disease,” Kardiologia Polska, vol. 58, no. 1, pp. 1–9, 2003. [13] K. E. Berge, A. Bakken, M. Bøhn, J. Erikssen, and K. Berg, “A DNA polymorphism at the angiotensin II type 1 receptor (AT1R) locus and myocardial infarction,” Clinical Genetics, vol. 52, no. 2, pp. 71–76, 1997.f Conflict of Interests h [15] C. W. Hamm, J. P. Bassand, S. Agewall et al., “ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the task force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the european society of cardiology (ESC),” European Heart Journal, vol. 32, no. 23, pp. 2999–3054, 2011. References 264, pp. 989–995, 1993. [8] A. Bonnardeaux, E. Davies, X. Jeunemaitre et al., “Angiotensin II type 1 receptor gene polymorphisms in human essential hypertension,” Hypertension, vol. 24, no. 1, pp. 63–69, 1994. [22] R. Ross, “Atherosclerosis: an inflammatory disease,” The New England Journal of Medicine, vol. 340, no. 2, pp. 115–126, 1999. [9] A. T. Chiu, W. F. Herblin, D. E. McCall et al., “Identification of angiotensin II receptor subtypes,” Biochemical and Biophysical Research Communications, vol. 165, no. 1, pp. 196–203, 1989. [23] Z. N. Jiang, W. S. Zhao, F. Yu, and G. Xu, “Association of angiot- ensin II type 1 receptor gene polymorphism with essential hypertension,” Chinese Medical Journal, vol. 114, no. 12, pp. 1249–1251, 2001. [10] N. Ranjith, R. J. Pegoraro, L. Rom, P. A. Lanning, and D. P. Naidoo, “Renin-angiotensin system and associated gene poly- morphisms in myocardial infarction in young South African Indians,” Cardiovascular Journal of South Africa, vol. 15, no. 1, pp. 22–26, 2004. [24] Y. Tanaka, K. Tamura, Y. Koide et al., “The novel angiotensin II type 1 receptor (AT1R)-associated protein ATRAP downregu- lates AT1R and ameliorates cardiomyocyte hypertrophy,” FEBS Letters, vol. 579, no. 7, pp. 1579–1586, 2005. 293 Disease Markers [25] J. Yu, G. N. Prado, L. Taylor, R. Pal-Ghosh, and P. Polgar, “Hybrid formation between the intracellular faces of the brad- ykinin B2 and angiotensin II AT1 receptors and signal transduc- tion,” International Immunopharmacology, vol. 2, no. 13-14, pp. 1807–1822, 2002. [26] T. Mizuno, M. Makino, Y. Fujiwara et al., “Renin-angiotensin system gene polymorphism in Japanese stroke patients,” Inter- national Congress Series, vol. 1252, pp. 83–90, 2003. [27] A. Jones, S. S. Dhamrait, J. R. Payne et al., “Genetic variants of angiotensin II receptors and cardiovascular risk in hyperten- sion,” Hypertension, vol. 42, no. 4, pp. 500–506, 2003. [28] J. A. Duncan, J. W. Scholey, and J. A. Miller, “Angiotensin II type 1 receptor gene polymorphisms in humans: physiology and pathophysiology of the genotypes,” Current Opinion in Nephrology and Hypertension, vol. 10, no. 1, pp. 111–116, 2001. [29] M. S. Ulgen, O. Ozturk, M. Yazici et al., “Association between A/C1166 gene polymorphism of the angiotensin II type 1 receptor and biventricular functions in patients with acute myocardial infarction,” Circulation Journal, vol. 70, no. 10, pp. 1275–1279, 2006. [30] R. ´Alvarez, J. R. Reguero, A. References Batalla et al., “Angiotensin-convert- ing enzyme and angiotensin II receptor 1 polymorphisms: asso- ciation with early coronary disease,” Cardiovascular Research, vol. 40, no. 2, pp. 375–379, 1998. [31] A. Gardemann, Q. D. Nguyen, J. Humme et al., “Angiotensin II type 1 receptor A1166C gene polymorphism. Absence of an asso- ciation with the risk of coronary artery disease and myocardial infarction and of a synergistic effect with angiotensin-convert- ing enzyme gene polymorphism an the risk of these diseases,” European Heart Journal, vol. 19, no. 11, pp. 1657–1665, 1998. [32] G. I. Rice, C. A. Foy, and P. J. Grant, “Angiotensin converting enzyme and angiotensin II type 1-receptor gene polymorphisms and risk of ischaemic heart disease,” Cardiovascular Research, vol. 41, no. 3, pp. 746–753, 1999. [33] J. E. Eichner, V. J. Christiansen, W. E. Moore, S. T. Dunn, and E. Schechter, “Angiotensin-converting enzyme gene polymor- phism in a cohort of coronary angiography patients,” Atheroscle- rosis, vol. 154, no. 3, pp. 673–679, 2001. [34] M. Castellano, M. L. Muiesan, M. Beschi et al., “Angiotensin II type 1 receptor A/C1166 polymorphism: relationships with blood pressure and cardiovascular structure,” Hypertension, vol. 28, no. 6, pp. 1076–1080, 1996.
https://openalex.org/W3126433918
https://phsreda.com/e-articles/10243/Action10243-97594.pdf
Russian
null
Nravstvennye determinanty lichnosti studentov v situatsii moral'nogo vybora
Pedagogy, Psychology, Society
2,021
cc-by
1,682
Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house Sreda Левкова Татьяна Валериевна канд. пед. наук, доцент, заведующая кафедрой ФГБОУ ВО «Приамурский государственный университет имени Шолом-Алейхема» г. Биробиджан, Еврейская автономная область DOI 10.31483/r-97594 НРАВСТВЕННЫЕ ДЕТЕРМИНАНТЫ ЛИЧНОСТИ СТУДЕНТОВ В СИТУАЦИИ МОРАЛЬНОГО ВЫБОРА DOI 10.31483/r-97594 Publishing house "Sreda" НРАВСТВЕННЫЕ ДЕТЕРМИНАНТЫ ЛИЧНОСТИ СТУДЕНТОВ В СИТУАЦИИ МОРАЛЬНОГО ВЫБОРА Аннотация: в статье рассмотрены понятия «моральный выбор», «нрав- ственная сфера», систематизированы основные возрастные задачи студенче- ства. Автор акцентирует внимание на том, что нравственная сфера является важной детерминантой морального выбора в студенчестве. В эксперименталь- ном исследовании была использована методика «Друг-советчик-2» Е.К. Весело- вой и С.А. Черняевой. Выборку составили студенты ФГБОУ ВО ПГУ имени Шо- лом-Алейхема. В ситуации морального выбора у испытуемых проявляется зна- чимость индекса нравственности, внешние и внутренние нравственные мо- тивы, а также тенденция ответственности в этом выборе. Ключевые слова: ситуация морального выбора, нравственные детерми- нанты, студенчество. Студенчество, как период личностного, нравственного, ценностного само- определения, ставит перед молодыми людьми важную задачу в отношении прин- ципов, действий и решений относительно своей жизни. Сама по себе духовно- практическая ситуация самоопределения личности это и есть ситуация мораль- ного выбора, направленная осознанное решение человека выбрать тот или иной вариант поведения, соответствующий личным или общественным моральными правилам. Моральный выбор выражает степень нравственной свободы человека и об- щества и может быть выполнен только при важных условиях: 1 Издательский дом «Среда» − при отсутствии внешнего давления и запрета, а также при отсутствии вли- яния воспринятых освоенных моральных установок, принципов, привычек; − при рефлексивности и осознанности, а также осознании возможных вари- антов предстоящего выбора; − при рефлексивности и осознанности, а также осознании возможных вари- антов предстоящего выбора; − при присутствии ценностных опор свободы воли, а также средств, способ- ствующих осуществлению задуманного, и при наличии основных способов осу- ществления деятельности [1]. − при присутствии ценностных опор свободы воли, а также средств, способ- ствующих осуществлению задуманного, и при наличии основных способов осу- ществления деятельности [1]. Говоря о нравственном выборе человека, мы все-таки можем отметить, что не только моральный, но и любой другой зависит от целей и задач, которые че- ловек определяет перед собой, кстати это касается и группы, общества. Учитывая тот факт, что любой моральный выбор человека сцеплен с его ответственностью, поэтому переживая ситуации морального выбора каждый поднимается на более высокую ступень своего нравственного развития. Качественно проанализировать ситуацию морального выбора, понять и объ- яснить сущность тех или иных поступков участников ситуации, осуществить анализ, сделать выводы и стимулировать свое поведение могут лишь те, кто об- ладает достаточно высоким уровнем развития нравственной сферы и развитым этическим мышлением. Для высокого уровня развития нравственной сферы определяющим явля- ется ясное представление о нравственных нормах и устойчивость их проявления в нравственном поведении. Средний уровень развитости нравственной сферы определяется знанием нравственных норм, которые не всегда являются мотива- торами поведения человека. ttps://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) НРАВСТВЕННЫЕ ДЕТЕРМИНАНТЫ ЛИЧНОСТИ СТУДЕНТОВ В СИТУАЦИИ МОРАЛЬНОГО ВЫБОРА Низкий уровень развития нравственной сферы че- ловека ориентирован на внешние формы поведения. В этих условиях у личности проявляется конформизм, опора в принятии решения на мнение других [1]. Принимая во внимание, что нравственная сфера человека – это интегриро- ванное сообщность нравственного сознания, ценностей, установок и норм, об- разцов, нравственных отношений и чувств, нравственного поведения, проявляю- щихся как в поступках, так и в любой деятельности личности [2]. Нравственное развитие личности является определяющим не только в направленности, 2 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" содержании и формах выражения ценностных ориентиров, но и в целях, сред- ствах, используемых в принятии решения в ситуации морального выбора. содержании и формах выражения ценностных ориентиров, но и в целях, сред- ствах, используемых в принятии решения в ситуации морального выбора. Индивидуализация нравственных представлений мировоззренческого уровня – представлений о счастье, о человеке как высшей ценности, понимание ценностно-нравственной составляющей смысла жизни в ранней юности и как ре- зультат – человек становится способным самостоятельно осуществлять нрав- ственный выбор [3]. Наше экспериментальное исследование было направлено на изучение важ- ной детерминанты принятия решения в ситуации морального выбора личности – нравственной сферы человека. Исследование проводилось в 2019–2020 годах, выборку составили студенты ФГБОУ ВО «Приамурского государственного уни- верситета имени Шолом-Алейхема». Выборка составила 70 человек. С целью исследования сформированности характеристики нравственной сферы студентов нами была использована методика «погружения в ситуацию мо- рального выбора» «Друг-советчик-2» Е.К. Веселовой и С.А. Черняевой [4]. Суть этого метода психодиагностики – психологическое погружение испытуемого в ситуацию морального выбора, опираясь на проективный характер заданий и си- туаций. Психодиагностическая ситуация подразумевает множественность отве- тов, поэтому позволяет испытуемым анализировать психологические аспекты ситуации, тем самым осознавая собственное понимание морального выбора и ос- нований собственного нравственного поведения. Результаты после диагностического исследования испытуемых представ- лены в табл. 1. Таблица 1 ц Средние показатели состояния нравственной сферы студентов курса (по методике Е.К. Веселовой и С.А. Черняевой) Параметры (шкалы методики Друг-советчик-2) Средние показатели состояния нравственной сферы студентов I (индекс нравственности) 39,36 P (количество позитивных ответов) 15,44 Средние показатели состояния нравственной сферы студентов курса (по методике Е.К. Веселовой и С.А. Черняевой) Параметры (шкалы методики Друг-советчик-2) Средние показатели состояния нравственной сферы студентов I (индекс нравственности) 39,36 P (количество позитивных ответов) 15,44 едние показатели состояния нравственной сферы студентов курса (по методике Е.К. Веселовой и С.А. 4 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) 4 https://phsreda.com 4 https://phsreda.com Содержимое до НРАВСТВЕННЫЕ ДЕТЕРМИНАНТЫ ЛИЧНОСТИ СТУДЕНТОВ В СИТУАЦИИ МОРАЛЬНОГО ВЫБОРА Черняевой) 3 Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Издательский дом «Среда» U (количество уклончивых ответов) 11,06 O (количество отрицательных ответов) 8,04 V (волевой компонент, конструктивность решений) 1,30 VM (внешние мотивы: социальные нормы, долг, последствия, закон) 2,03 VNM (внутренние мотивы: совесть, сердце, чувство вины) 2,07 NR (нравственный релятивизм) 1,46 RAC (рационализация) - A- (отрицательные эмоции) 2,40 T (нежелание влиять на поступки героя ситуации, отчуждение) 1,40 F (общий совет, отговорка) 2,06 E (совет сделать позитивный выбор при условии, если есть возможность) 1,43 M (совет устранить чувства, эмоции, изоляция) - A+ (положительные эмоции, одобрение, похвала) 1,51 Таблица 1 показывает, что средний показатель индекса нравственности (I) у студентов составляет 39,36. По условию методики «Друг-советчик-2» показатель индекса нравственности меньше 37 баллов означает снижение роли нравствен- ных нормативов в принятии решений, но по результатам нашего исследования средний показатель I несколько выше 37, как вывод – испытуемые ориентиру- ются на нравственные нормы, но, так как показатель не высок, то для респонден- тов нравственные ориентиры не постоянно являются основой для их морального выбора и не всегда актуализируются в личностно значимых ситуациях. Средний показатель U у испытуемых составляет 11,06, что отражает откры- тость респондентов при стремлении раскрыть свое отношение к моральным нор- мам. Средний показатель отрицательных ответов (О) составляет 8,04, что отра- жает отсутствие жесткой позиции неприятия моральных норм. Категория V, в нашем случае данная категория выражена 1,30, что свиде- тельствует о возможности принятия конструктивных решений, тенденции давать конкретные советы в ситуациях морального выбора, но данная категория выра- жена не ярко. Помимо четырех основных показателей состояния нравственной сферы (I, P, U, O), в методике «Друг-советчик-2» определены частные показатели, Publishing house "Sreda" характеризующих нравственную сферу личности. Произведя подсчёт выражен- ности этих категории, мы определили средние значения [1]. Категория VM составляет 2,03, что отражает выраженность внешних моти- вов в процессе принятия решения. Студенты осознают необходимость соблю- дать этические нормы и думают о последствиях своего выбора. В их ответах ча- сто встречаются слова «долг», «последствия», «закон». VNM составляет 2,07, что отражает такие критерии принятия решений как совесть, чувство стыда / вины. Для респондентов важными в принятии решений являются внутренние нравственные убеждения и ценности, многие их обследуе- мых говорили – «надо представить, что с тобой поступят также». Наличие высказываний типа NR (1,46 – это выраженный средний показа- тель) свидетельствует о том, что оценивание поведения другого человека в ситу- ации морального выбора не выражено, что может определяться обособленно- стью или завистливостью. Категория F составляет 2,06. Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) НРАВСТВЕННЫЕ ДЕТЕРМИНАНТЫ ЛИЧНОСТИ СТУДЕНТОВ В СИТУАЦИИ МОРАЛЬНОГО ВЫБОРА У обследуемых эта категория в ответах встре- чалась часто, что отражает присутствие импульсивности в принятии решения, а также наличии чувства спонтанности. Особого внимания заслуживает категория А, которая выражена в среднем значении – 2,40. Это достаточно выраженное средний показатель, обозначающее отрицательные эмоции в реакции респондента на ситуацию. Отрицательными эмоциями могут выступать злость, раздражение, гнев, ярость, осуждение. Такая выраженность этой категории свидетельствует о высокой потребности к аффи- лиации испытуемых [4]. Средний показатель U составляет 11, 06, что говорит о выраженной закры- тости и нежелании раскрывать свое отношение к моральным нормам. Средний показатель категории T у студентов составляет 1,40. Эта категория определяет самодостаточность респондентов, уравновешенность в принятии ре- шения, а также свидетельствует об отсутствии тревожности и недостатке эмпа- тии [4]. Остальные показатели в представленной выборке не выражены вообще (RAC и М) или выражены значительно ниже нормы (Р, Е, А+). 5 Издательский дом «Среда» Издательский дом «Среда» Таким образом, мы видим по результатам исследования, что состояние нравственной сферы личности является существенной детерминантой, опреде- ляющей принятие решения в ситуации морального выбора. Степень сформиро- ванности нравственной сферы испытуемых в выборки определяется в средних значениях с тенденцией к ответственности при принятии решения, а также от- крытости моральным принципам, хотя обнаруживается весьма интересная тен- денция – доля представленности влияния на выбор внешних и внутренних фак- тором очень близка, что может отражать проживание внутреннего конфликта в ситуации морального выбора респондентами. 6 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Список литературы 1. Аминева Н.А. Характеристики нравственной сферы личности как детер- минанты принятия решения в ситуации морального выбора / Н.А. Аминева, Т.В. Левкова // Постулат. – 2018. – №1 [Электронный ресурс]. – Режим доступа: http://e-postulat.ru/index.php/Postulat/article/view/1130 1. Аминева Н.А. Характеристики нравственной сферы личности как детер- минанты принятия решения в ситуации морального выбора / Н.А. Аминева, Т.В. Левкова // Постулат. – 2018. – №1 [Электронный ресурс]. – Режим доступа: http://e-postulat.ru/index.php/Postulat/article/view/1130 тусь Б.С. Психология. Нравственность. Культура. – М., 1994. – 60 с. 2. Братусь Б.С. Психология. Нравственность. Культура. – М., 1994. – 60 с. 3. Психология человека от рождения до смерти / под ред. А.А. Реан. – СПб., 2002. – С. 324–455. 4. Веселова Е.К. Метод исследования нравственной сферы личности // Диа- гностика здоровья. Психологический практикум. – СПб., 2007. – С. 359–374. 4. Веселова Е.К. Метод исследования нравственной сферы личности // Диа- гностика здоровья. Психологический практикум. – СПб., 2007. – С. 359–374. 6 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0)
https://openalex.org/W4309690646
https://www.researchsquare.com/article/rs-2288815/latest.pdf
English
null
Sophomoric sex ed: Sexism through definition in collegiate sexuality education textbooks
Research Square (Research Square)
2,022
cc-by
11,424
Sophomoric sex ed: Sexism through definition in collegiate sexuality education textbooks Chloe Kearin Willeford  George Washington University Caroline J. Etgen  George Washington University Carly N. Jordan  (  cnjordan@gwu.edu ) George Washington University Sex education in the United States Almost every adolescent in the United States takes part in formal sex and sexuality education. Ninety-six percent of female and ninety-seven percent of male teenagers received formal sexuality education before the age of 18 (Martinez et al. 2010). 80% of youth receive formal education on STIs (sexually transmitted infections), STDs (sexually transmitted diseases), and consent, and 55-60% of youth receive formal education on birth control (Lindberg 2016). In thirty states and the District of Columbia, public schools mandate sex education programs (NCSL 2020). Because of the prevalence of sexuality education, it is extremely important that the information presented therein is inclusive and accurate. Students themselves want to see more inclusive sexuality education (Coll et al. 2018). Women’s experience of sexism directly links to their felt sadness and mental health (Chevallereau et al. 2021). And, there is a direct link between comprehensive sexual education (CSE) and sexual health outcomes (Jozkowski and Crawford 2016). In this paper, we focus on the disparity in treatment of men and women in sexual education textbooks and the real consequences for physical and mental health that that disparity has. Research Article Keywords: sex education, gender roles, curricula Posted Date: November 22nd, 2022 DOI: https://doi.org/10.21203/rs.3.rs-2288815/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/23 Page 1/23 Abstract The majority of adolescents in the United States receive formal sexuality and sexual health education, making it important that information presented therein is accurate and unbiased. Sexual education presents a pattern of describing men as active and sexually driven and women as passive victims of men’s sexual urges. Our research found that this pattern manifests in sexual health textbooks, when female anatomy is described only in terms of male anatomy. We read and analyzed anatomy chapters in ten of the most common sexual education textbooks used in American high schools and colleges to determine how bodies of each sex were defined and found a statistically significant correlation with a large effect size between the sex of anatomy and the incidence of description of that anatomy from the perspective of another sex. This analysis shows that sexual health textbooks continue to depict women as sexual tools for men, contributing to the sexism in American society. Sexism has negative psychological effects on women from an early age. Our research highlights the frequency with which women receive sexist messages in education, compounding these negative psychological effects. Literature review: gender in sexual education textbooks This equation of consensual sexual activity and rape is extremely harmful This constructed binary between active men and passive women extends into discussion of the ultimate demonstration of men’s supposedly uncontrollable sexual drive: rape. A recent survey of 16 American sexual education textbooks at the high-school level found that some textbooks include language that associates  girls’ desire to participate in sex at all with an increased chance that they will be raped (Clonan-Roy et al. 2021). For example, some books recommend “avoidance strategies,” like staying away entirely from parties and bars, to avoid sexual assault. Abstinence is also frequently recommended, and some books suggest that, by having sex at all, you are opening yourself up to others wanting to have sex with you and therefore to rape. This equation of consensual sexual activity and rape is extremely harmful — experiences of these two types of sexual encounters are different, and suggesting that they are not removes sexual agency from girls and women. Activities like Q&As and role-playing sessions where participants act out sexually-pressured scenarios suggest that, by planning ahead, girls will be able to avoid sexual assualt. This further associates women and girls’ behavior and choices with their victimhood. Sharon Lamb et al. (2011) found that textbooks present girls as victims of assault or rape much more than they do boys. Some textbooks imply that, since men have such strong sexual urges, women should dress conservatively so as not to provoke men, and they should also set clear “standards” for themselves before entering a relationship so that they might be able to predict a sexual assault (Lamb et al. 2011). Again, girls and women’s sexual agency only extends so far as to counter men’s. Michelle Fine (1988) observed sexual health classes and spoke with students in public high schools, finding that this sexual education omitted discussion of desire or pleasure. Fine argues that teaching girls a full picture of sex will empower them to make educated choices and remove them from the female victimization narrative of sex education that doesn’t discuss pleasure. — experiences of these two types of sexual encounters are different, and suggesting that they are not removes sexual agency from girls and women. Activities like Q&As and role-playing sessions where participants act out sexually-pressured scenarios suggest that, by planning ahead, girls will be able to avoid sexual assualt. Literature review: gender in sexual education textbooks In many sexual education textbooks, men are presented as active and sexually driven, while women are presented as passive recipients of male desire. For example, Christine Beyer et al. (1996) found that definitions of the penis, the vagina, intercourse, and fertilization were male-centric, quoting a textbook which defined intercourse as when “the penis of the male is slipped inside the vagina of the female.” Similarly, Stevi Jackson (1999) notes the tendency of textbooks to describe intercourse as “the ‘penetration’ or ‘insertion’ of the penis into the vagina.” This pattern doesn’t apply only to penetration itself. In her review of literature about sexual education textbooks, Laina Bay-Cheng (2003) wrote that Page 2/23 many studies found that these materials cast boys as sexual initiators and girls as sexual gatekeepers in sexual encounter scenarios. It is important to note that this discourse of men as active and women as passive in sex extends beyond textbooks: Renée DePalma and Diébédo Francis (2014) found through interviews with sex education teachers that these educators themselves had “a strong tendency to portray the girls as sexual victims and boys as sexual predators in any kind of intimate relationship.” And, Lorena García (2009) discovered in interviews with Latina adolescent girls that sexual education teachers scolded boys for not taking sex ed seriously and being disruptive, while girls were instead scolded for being active, constructive participators. Through this pattern, these teachers were suggesting that boys are inherently sexual but girls are not sexual beings and should not take any interest or agency in their own sexual practices. This constructed binary between active men and passive women extends into discussion of the ultimate demonstration of men’s supposedly uncontrollable sexual drive: rape. A recent survey of 16 American sexual education textbooks at the high-school level found that some textbooks include language that associates  girls’ desire to participate in sex at all with an increased chance that they will be raped (Clonan-Roy et al. 2021). For example, some books recommend “avoidance strategies,” like staying away entirely from parties and bars, to avoid sexual assault. Abstinence is also frequently recommended, and some books suggest that, by having sex at all, you are opening yourself up to others wanting to have sex with you and therefore to rape. Literature review: gender in sexual education textbooks This further associates women and girls’ behavior and choices with their victimhood. Sharon Lamb et al. (2011) found that textbooks present girls as victims of assault or rape much more than they do boys. Some textbooks imply that, since men have such strong sexual urges, women should dress conservatively so as not to provoke men, and they should also set clear “standards” for themselves before entering a relationship so that they might be able to predict a sexual assault (Lamb et al. 2011). Again, girls and women’s sexual agency only extends so far as to counter men’s. Michelle Fine (1988) observed sexual health classes and spoke with students in public high schools, finding that this sexual education omitted discussion of desire or pleasure. Fine argues that teaching girls a full picture of sex will empower them to make educated choices and remove them from the female victimization narrative of sex education that doesn’t discuss pleasure. Sexual education curricula have little to no discussion of female pleasure and present female sexuality as reproductive only (Whatley 1988). Louisa Allen and Moira Carmody (2012) theorize that eliminating the discussion of female pleasure points the discussion to reproduction only, which is inherently misogynistic. One survey of college students before and after a course in sexual education showed that one of the most important elements of the course for students was that the discussion of pleasure made them more comfortable with their own bodies, sexualities, and sex lives (Goldfarb 2005). However, K-12 sex ed rarely discusses pleasure, instead framing sexual activity as dangerous and risky (Clonan-Roy et al. 2021). While teenage girls might be initially interested in sexual activity, sexual education’s lack of Page 3/23 discussion of pleasure teaches girls that sex can only cause negative outcomes like assault, disease, and unwanted pregnancies (Fine 1998). Rachael Eastham and Ali Hanbury (2021) caution that, while it is important to discuss pleasure in relation to sex, pleasure should not be constructed as mandatory, expected, or inevitable, as not all people can easily achieve sexual pleasure. discussion of pleasure teaches girls that sex can only cause negative outcomes like assault, disease, and unwanted pregnancies (Fine 1998). Rachael Eastham and Ali Hanbury (2021) caution that, while it is important to discuss pleasure in relation to sex, pleasure should not be constructed as mandatory, expected, or inevitable, as not all people can easily achieve sexual pleasure. Literature review: gender in sexual education textbooks Sexual education that omits discussion of female bodies and pleasure teaches students to continue to view women as objects (Allen 2004). It can also lead to acceptance of other gender stereotypes about sex, like that women lack sexual agency or inherent sexual desire (Allen 2007). DePalma and Francis (2014) found that many sexual education teachers tend to portray men as predatory and girls as passive sexual victims, potentially passing these stereotypes on to students. If women are uneducated about their sexuality, they will feel shame around feeling pleasure or discussing sex (Tolman 1994). Catherine Connell and Sinikka Elliot’s (2009) literature review found that, in general, school sexuality education reinforces gender inequalities. Theoretical Framing We used two pieces of theory to ground our work on biases in sexual health education textbooks and relate it to student impact. Barbara Fredrickson and Tomi-Ann Roberts (1997) offer objectification theory as a way to understand being female in a society that objectifies women, writing that external objectification can lead to self- objectification, which can lead to poor mental health outcomes. They note that “the common thread running through all forms of sexual objectification is the experience of being treated as a body (or collection of body parts) valued predominantly for its use to (or consumption by) others” (Fredrickson & Roberts 1997). This definition of a person as a body primarily for consumption by others is something we examined in our study of sexual health textbooks. Adrienne Rich (1980)’s foundational paper Compulsory Heterosexuality and Lesbian Existence theorizes that men use sexuality and sexual desire as an expression of power over women. She writes that Western tradition “asserts that women are inevitably, even if rashly and tragically, drawn to men” (Rich 1980). This belief removes women’s agency from the equation of heterosexual relationships and erases (the possibility of) queer relationships entirely. The sexual health textbooks we studied often did not give women agency in heterosexual relationships. While our paper does not address LGBTQ+ representation in textbooks, it is important to keep in mind how this problematic representation of women could affect queer-identifying people as well. Methods The DSM-5 defines sex as “the biological indicators of male and female (understood in the context of reproductive capacity), such as in sex chromosomes, gonads, sex hormones, and nonambiguous internal and external genitalia” and gender as “the public (and usually legally recognized) lived role as boy or girl, man or woman” (American Psychiatric Association 2013). We acknowledge that this paper addresses sex as a binary, which is incorrect and excludes intersex people. In this paper, we also use the term “biological sex” to describe the sex assigned to a person at birth based on physical characteristics such as genitals. We understand that this characterization of sex does not fully recognize the wide variety of sex characteristics and therefore does not represent all experiences. However, understanding of sex as biological and genital-based is the societal understanding and that used in these textbooks. We also recognize that this paper equates female and woman, and male and man, which conflates sex and gender and excludes transgender and nonbinary individuals. Because the textbooks themselves address sex as a binary and conflate sex and gender, we had to apply this framework of thinking and language to effectively study gender and sex discrimination in them. These texts’ anatomy chapters are also rife with heteronormativity, “the assumption that heterosexuality is the standard for defining normal sexual behavior and that male–female differences and gender roles are the natural and immutable essentials in normal human relations” (American Psychiatric Association 2020). Our analysis therefore includes heteronormativity in our discussions of our coded incidents. Because the textbooks themselves reflect the incorrect and problematic assumption that heterosexuality is the standard of normal behavior, our paper necessarily includes analyses with this problematic frame. To begin, we chose ten textbooks commonly used in American sexual health and reproduction high school and college courses. These books are intended and marketed for undergraduate sexuality education courses. They have been published in many editions, the most recent of which are between 2014-2021 and were successful in sales. Because students and instructors are a major source of textbook sales, and these books’ marketing targets the education community, we can reasonably assume that these texts are commonly used in American undergraduate sexuality courses. For efficiency and concision, we refer to these books by labels (e.g. “Book A”) instead of their full titles. A list of books and their codes follows in Table. 1. Our work The prevalence and impact of sexism in sexual education has been established. We sought to uncover how these sexist attitudes manifest in sexuality education textbooks, specifically in how authors define sexual anatomy. To do this, we searched for incidents of sexist language in sexual anatomy chapters of sexual health and reproduction textbooks used in United States high school and college curricula. Page 4/23 Specifically, we used qualitative codes to note instances where anatomy was defined in terms of the other sex’s anatomy, physical pleasure, or values. We then analyzed the data and found a statistically significant correlation with a large effect size between the sex of anatomy and the incidence of description of that anatomy from the perspective of another sex. Methods We chose to focus on the sexual anatomy chapters of each textbook because our research examined how sexual anatomy was defined, and these chapters were most likely to include definitions. To evaluate incidence of sexism in sexual health textbooks, we used qualitative codes to note instances where anatomy was defined in terms of the other sex’s anatomy, physical pleasure, or values. We then Page 5/23 Page 5/23 coded each incident “F” or “M” depending on whose anatomy was being defined. An example of an “F” code is the following passage from Book G: "the vagina is the tube-shaped organ into which the penis is inserted during coitus; it also receives the ejaculate.” In this example, vagina is not defined in terms of its function, structure, significance to female pleasure, or sexual well being — it is only defined in terms of male anatomy. An example of an “M” code is this sentence from Book D, "it (the penis) serves to deposit sperm in a woman's vagina.” Again, the penis is not described in terms of its own function, structure, or pleasure, but only in terms of the purpose it serves in relation to female anatomy. At the beginning of the research process, all three members of the team met to code a chapter as a group, discuss our results, and come to a consensus. This was important to do at the beginning to familiarize ourselves with the coding definitions and increase our intercoder reliability. After this first chapter, each member of the three person research team coded all chapters individually. When one sentence contained multiple incidents of the same type, we coded it only once. Take, for example, this passage from Book I, “The vagina is a thin-walled tube extending from the cervix of the uterus to the external genitalia; it serves as the female organ for penetration, a passageway for sperm, and a canal through which menstrual fluid and babies can pass from the uterus,” we would only code that once. Though there are multiple instances of defining one sex’s anatomy in terms of another sex, we did not code those individually because they are consistent and in the same sentence. We coded all content on each page of the textbook, including picture captions and discussion boxes, not just main body content. We also coded “review” or “summary” sections at the end of chapters. Methods After our individual analysis, we came together to discuss our results and find consensus to create one final set of codes and incidents fo each chapter. In cases of uncertainty, we reviewed the quote as a group and referenced our “F” and “M” code definitions to come to a conclusion. We implemented this process in an effort to safeguard against individual biases, avoid errors in the creation of data sets, and increase intercoder reliability. Results Our analysis shows that sex had a stastically significant impact on the number of incidents where anatomy was defined in terms of the other sex’s anatomy. Instances where female anatomy was described only in terms of male anatomy (F) were far more common than instances where male anatomy was described in terms of female anatomy (M) (Fig. 1). We performed Welch’s two-sample t-test and found a statistically significant difference between instances of F and M (t(14) = 2.384, p = .032). The application of both Cohen’s d (1.066) and Glass’s delta (0.863) tests demonstrated a large effect size. The mean number of F incidents per book was 3.2 (SD = 2.781), and the mean number of M incidents per book was 0.8 (SD = 1.549). In nine out of ten textbooks, F was more common than M. Textbook D had the largest disparity with nine instances of F and only one of M. Out of 40 total incidents, 32 were coded F; meaning 80% of the instances we coded defined female anatomy based on male anatomy, physical pleasure, or values. On average, there were fewer than one M code per book, and only four out of the ten books analyzed had any M incidents at all. Page 6/23 We collected data regarding page counts of chapters and sections to evaluate whether the number of pages had an impact on the number of instances (Table 2). Multiple linear regression was performed and no relationship was observed. On average, there was an incidence of F about every 13.7 pages, and there was an incidence of M about every 29.4 pages. Longer chapters did not necessarily have more codes than shorter chapters. The mean of pages per anatomy chapter about female anatomy was 23.5 (SD =  11.78), and the mean for pages describing male anatomy was 16.7 (SD = 7.76). However, page counts cannot be considered completely representative of the overall amount of material, as the content of pages varied. Pages sometimes included images or were formatted differently than others, and fonts and words per page were not uniform across textbooks. We collected data regarding page counts of chapters and sections to evaluate whether the number of pages had an impact on the number of instances (Table 2). Multiple linear regression was performed and no relationship was observed. Results On average, there was an incidence of F about every 13.7 pages, and there was an incidence of M about every 29.4 pages. Longer chapters did not necessarily have more codes than shorter chapters. The mean of pages per anatomy chapter about female anatomy was 23.5 (SD =  11.78), and the mean for pages describing male anatomy was 16.7 (SD = 7.76). However, page counts cannot be considered completely representative of the overall amount of material, as the content of pages varied. Pages sometimes included images or were formatted differently than others, and fonts and words per page were not uniform across textbooks. Page 7/23 Table 1 Book codes, titles, editions, and authors of the texts examined in this study. Book Code Book Title & Chapter(s) Year, Edition Author(s) Book A* Human Sexuality: Diversity in Contemporary Society; Chap. 3- Female Sexual Anatomy, Physiology and Response; Chap. 4: Male Sexual Anatomy, Physiology and Response 2019, Tenth Edition William Yarber and Barbara Sayad Book B Human Sexuality: Self, Society and Culture; Chap. 4- Sexual Anatomy and Physiology 2014, First Edition Gilbert Herdt and Nicole Polen-Petit Book C Human Sexuality: A Contemporary Introduction; Chap. 4- Sexual Anatomy: The Parts, the Pieces and How They Respond 2017, Second Edition Edited by Caroline F. Pukall; ch. 4 by Caroline F. Pukall and Richard J. Wasserburg Book D Human Sexuality Today; Chap. 2- Our Sexual and Reproductive Anatomy 2019, Ninth Edition Bruce King, Pamela Regan Book E The Psychology of Human Sexuality; Chap. 3- Human Sexual Anatomy 2018, Second Edition Justin J. Lehmiller Book F* Discovering Human Sexuality; Chap. 2- Women’s Bodies; Chap. 3- Men’s Bodies 2018, Fourth Edition Simon LeVay, Janice Baldwin, John Baldwin Book G Understanding Human Sexuality; Chap. 4- Sexual Anatomy 2020, Fourteenth Edition Janet Hyde and John Delamater Book H* Our Sexuality; Chap. 3-Female Sexuality and Physiology; Chap. 4 - Male Sexuality and Physiology 2021, Fourteenth Edition Robert L. Crooks, Karla Baur, Laura Widman Book I* Sexuality Now: Embracing Diversity; Chap. 5 - Female Sexual Anatomy and Physiology; Chap. 6 - Male Sexual Anatomy and Physiology 2019, Sixth Edition Janell L. Carroll Book J Human Sexuality in a Changing World; Chap. 2 - Female Sexual Anatomy and Physiology; Chap. 3 - Male Sexual Anatomy and Physiology 2018, Tenth Edition Spencer A. Rathus, Jeffrey S. Table 2 Incidents of description of anatomy from the perspective of another sex in ten sexuality and reproduction textbooks commonly used in American high school and college courses. We will apply both Cohen’s d and Glass’s delta effect size tests. Number of incidents Pages devoted to anatomy   F M Total F M Non- specific A* 2 1 51 31 20 0 B 2 0 34 19 10 5 C 0 1 30 10 8 12 D 9 1 23 12 6 5 E 1 0 33 15 11 7 F* 2 0 63 35 28 0 G 3 0 32 9 15 8 H* 4 0 59 37 22 0 I* 2 0 49 28 21 0 J 7 5 65 39 26 0 Totals 32 8 439 23.5 16.7 37 Mean 3.2 0.8 43.9 23.5 16.7 3.7 p < 0.05 (Welch’s: 0.032, Student’s: 0.028) * in this book, the anatomy content was spread over two chapters Number of incidents Pages devoted to anatomy In this section, we provide some quotes from the textbooks studied to show the prevalence of these incidents and to better illustrate how we defined the F and M codes. Problematic quotes tended to either depict female people as passive sex objects, ascribe importance to societal perceptions of certain pieces of anatomy, or introduce unnecessary opinions from another sex. Many books define female genitals in terms of how male genitals interact with them or the way that male people can derive pleasure from them, rather than the purpose they serve for female pleasure: Results Nevid, Lois Fichner-Rathus * in this book, the anatomy content was spread over two chapters Table 1 Page 8/23 Page 8/23 Page 8/23 Table 2 The vagina serves as a depository for the man's sperm. (Book D, p. 36) Not only is the vagina used for sexual intercourse, but it also receives the sperm and serves as the passageway through which the baby travels during childbirth. The penis not only produces sexual Page 9/23 pleasure but also is responsible for ejaculation and impregnation. (Book G, p. 69) The vagina is the tube-shaped organ into which the penis is inserted during coitus; it also receives the ejaculate. (Book G, p. 74) The vagina is the tube-shaped organ into which the penis is inserted during coitus; it also receives the ejaculate. (Book G, p. 74) Menstrual flow and babies pass from the uterus to the outer world through the vagina. During coitus, the vagina contains the penis. (Book J, Chap. 2 Review) Menstrual flow and babies pass from the uterus to the outer world through the vagina. During coitus, the vagina contains the penis. (Book J, Chap. 2 Review) We defined these quotes as F because they define the vagina primarily as something that receives the penis. All language surrounding the female anatomy is passive: the vagina is “used.” It “receives” and “serves,” while the penis “produces” and “is responsible.” The vagina is where “the penis is inserted,” while the penis “produces sexual pleasure.” All these phrases orient female anatomy as inherently passive and submissive, and only describe it in terms of its “service” to male anatomy, which is active and dominant. Instead, descriptions of the vagina should mention other functions like female pleasure, childbirth, and expelling menstrual fluid. They should not relate to the penis, other male anatomy, or male values and opinions. The mechanism of erection functions in both clitorises and penises (Puppo 2013). However, some books only discuss this mechanism in relation to the penis: The mechanism of erection functions in both clitorises and penises (Puppo 2013). However, some books only discuss this mechanism in relation to the penis: The mechanism of [clitoris] erection is described in more detail in connection with the penis. (Book F, p. 27) 27) The corpora cavernosum and the mechanism of erection will be considered in more detail in the discussion of the male sexual organs. (Book G, p. 68) The corpora cavernosum and the mechanism of erection will be considered in more detail in the discussion of the male sexual organs. (Book G, p. The vagina serves as a depository for the man's sperm. (Book D, p. 36) 68) We coded these quotes as F because they place the importance of erection and therefore pleasure in the sections on the male anatomy and penises, necessarily defining a biological response common to both female and male anatomy only in terms of male anatomy and male sexual pleasure. This therefore fulfills our code definition of defining female anatomy in terms of male anatomy and implies that pleasure is not relevant to the female experience of sex, which further represents female people as passive. Description of the clitoris should include and explain the process of erection without relating it to the penis or postponing it until the penis section as seen in Book F. In general, sections on the clitoris tended to be problematic. Because of the unique role the clitoris plays in female sexual pleasure, it has become one of the most controversial aspects of female genitalia. (Book E, p. 69) Because of the unique role the clitoris plays in female sexual pleasure, it has become one of the most controversial aspects of female genitalia. (Book E, p. 69) Authors struggled to define a body part whose function is sexual pleasure for female people and which does not play a part in reproduction (or pleasure for sexual partners). Book E’s choice to call the clitoris “controversial” is problematic. A better term might have been “much-discussed” or something along those lines. Calling the clitoris itself controversial might cause readers to feel shame about their clitoris or Page 10/23 Page 10/23 pleasure they experience from it. Also, the authors should have written that the function of the clitoris is oft-discussed, rather than the clitoris itself. Female people should not have to feel that their anatomy itself is being scrutinized. pleasure they experience from it. Also, the authors should have written that the function of the clitoris is oft-discussed, rather than the clitoris itself. Female people should not have to feel that their anatomy itself is being scrutinized. The vagina serves as a depository for the man's sperm. (Book D, p. 36) The section defining the hymen, which has historically but inaccurately been used as a measure of female virginity, is also rife with opportunity for male opinion: The section defining the hymen, which has historically but inaccurately been used as a measure of female virginity, is also rife with opportunity for male opinion: For men, would you want your intended spouse or lifetime partner to be examined to see if her hymen was intact to prove her virginity? (Book B, p. 103) For men, if a woman's hymen was broken, would you want it to be reconstructed? Why or why not? (Book B, p. 103) Many American men nevertheless expect their partner to bleed during first intercourse, so much so that some sexually experienced women elect to have a hymen surgically reconstructed before marriage. (Book D, p. 33) Many American men nevertheless expect their partner to bleed during first intercourse, so much so that some sexually experienced women elect to have a hymen surgically reconstructed before marriage. (Book D, p. 33) We considered these quotes to be F because they introduce men’s opinions in a section on female anatomy, therefore defining female anatomy in terms of its utility and desirability to men. The hymen should be described only as a feature of anatomy and should not be related to either the constructed concept of virginity or male pleasure. To be clear, the hymen’s presence or lack thereof does not correlate with sexual experience. The hymen can be broken through everyday activities like sports, through the insertion of tampons, or through accidents or injuries unrelated to sexual activity (Cook and Dickens 2009). Some female people don’t have a hymen, or the hymen is very small (Kimberley et al. 2012). Even medical professionals believe in the hymen as an indicator of having had sexual intercourse: one study of British midwives found that two thirds of respondents believed that the hymen breaks at first sexual intercourse (Christianson and Erikkson 2013). John McCann et al. (2007) found that hymenal lacerations, after healing, may or may not leave evidence of the previous injury. Descriptions of female anatomy should stand alone and should not be dependent on interjections placing emphasis only on the value that anatomy has to men. Female anatomy has value to female people themselves, but many textbooks do not recognize this. The vagina serves as a depository for the man's sperm. (Book D, p. 36) Textbooks emphasize men’s opinions of female bodies: For women only: What is your response to looking at your genitals? For men only: What is your response to viewing photos of women’s genitals? Why is it that women are discouraged from touching or looking at their genitals? (Book A, p. 89) Men who persuade their partners to have labiaplasty typically want their partners’ labia to conform to the vulva appearance that appeals to them most when watching pornography. (Book H, p.50) Men who persuade their partners to have labiaplasty typically want their partners’ labia to conform to the vulva appearance that appeals to them most when watching pornography. (Book H, p.50) Page 11/23 It might be interesting to discuss women’s opinions on why they are discouraged from interest in their genitals, but this is cast aside in favor of hearing male voices and thoughts. This emphasis on men’s Page 11/23 opinions of female bodies reinforces the cultural belief that female bodies are solely for the purpose of male pleasure. opinions of female bodies reinforces the cultural belief that female bodies are solely for the purpose of male pleasure. mphasis on male opinions is also seen in these books’ discussion of breast This emphasis on male opinions is also seen in these books’ discussion of breasts: This emphasis on male opinions is also seen in these books’ discussion of breasts: Although breasts are not part of a woman's reproductive system, they are considered to be highly erotic by most men in Western societies, and therefore we must consider them part of a woman's sexual anatomy. (Book D, p. 33) Many American men develop a very powerful interest in, and attraction to, women's breasts. (Book G, p. 77) 77) We might instead ask whether female people themselves consider their breasts to be sexual. Breasts were a subject that was addressed problematically in the sexual anatomy sections of many books. As the authors of Book D explained above, many people mistakenly consider breasts to be a sexual organ, and the authors rationalize this belief by noting how men find breasts attractive. This fits the F code because it necessarily defines female anatomy based on how men perceive it, how it relates to male anatomy and sexual pleasure. While these authors’ conclusion is misguided, their base assumption is correct: men are more likely to consider breasts as serving a sexual function rather than a function of lactation (Ward et al. 2006). Information on breasts should explain that, in fact, breasts do not biologically serve a reproductive or sexual function but a childrearing function. Breasts’ primary biological function is to nourish offspring through lactation (Howard and Gusterson 2000). Women are also, albeit less often, asked their opinions of male bodies: Studies examining how women in California, New Zealand, and Cameroon, for example, rate the attractiveness of non-erect penises found that women in all of these regions rated penises of intermediate length as most attractive. (Book C, p. Men who persuade their partners to have labiaplasty typically want their partners’ labia to conform to the vulva appearance that appeals to them most when watching pornography. (Book H, p.50) 84) ... the vagina, which is the canal that extends from the introitus to the uterus. (Book E, p. 72) These quotes were able to accurately describe the shape and location of the vagina without discussing the penis in conjunction. These quotes were able to accurately describe the shape and location of the vagina without discussing the penis in conjunction. It is also important for texts to discuss female pleasure in relation to sexual anatomy, as does this quote: It is also important for texts to discuss female pleasure in relation to sexua Beneath the vulva are several important structures, including the pelvic floor muscles, a complex, crisscrossed set of muscles that contract involuntarily when a woman reaches orgasm. (Book E, p. 72) Beneath the vulva are several important structures, including the pelvic floor muscles, a complex, crisscrossed set of muscles that contract involuntarily when a woman reaches orgasm. (Book E, p. 72) While sections on the penis usually mentioned pleasure and/or orgasm, sections on female anatomy like the vagina, vulva, and clitoris usually did not always do so. Any discussion of pleasure for female people was often in relation to their sexual partners (usually assumed to be male people). It is important to include pleasure in discussion of all sexes in order to counteract the narrative that female people are passive in sex while male people are active. Because female anatomy is no less capable of sexual pleasure than male anatomy, attention to correcting this bias is essential in producing an accurate text. As discussed above, description of the hymen often was problematic in its valuation of the societally constructed concept of virginity and emphasis on male opinion in defining worth of female anatomy. Here is an example that does not include that concept: [The hymen] is usually present at birth, and it generally remains intact until first intercourse, when it is commonly torn. Some women, however, are born without a hymen, and some women who are born with one may unknowingly tear it during sporting or other non-sexual activities. (Book C, p. 82–83) [The hymen] is usually present at birth, and it generally remains intact until first intercourse, when it is commonly torn. Some women, however, are born without a hymen, and some women who are born with one may unknowingly tear it during sporting or other non-sexual activities. (Book C, p. Men who persuade their partners to have labiaplasty typically want their partners’ labia to conform to the vulva appearance that appeals to them most when watching pornography. (Book H, p.50) 92) Studies examining how women in California, New Zealand, and Cameroon, for example, rate the attractiveness of non-erect penises found that women in all of these regions rated penises of intermediate length as most attractive. (Book C, p. 92) While the textbooks we read discuss female anatomy in terms of its applicability to male anatomy and sexual desire in multiple ways — like labia appearance, breast appearance, and function of the vagina — female opinions on the value of male bodies are much less common and are virtually all in relation to penis size. In the same way that male opinions were invoked in discussion of female bodies, this quote defines penises based on what value they hold for female sexual pleasure. Penises should be described without evaluation of their appearance or “attractiveness.” However, female opinions of male bodies are not included nearly as often as male opinions of female bodies as our above results demonstrate. When Book C includes women’s opinions of penises, they quickly follow up with male people’s thoughts (Book C, p.92). However, books including men’s opinions of breasts do not follow up with female people’s ideas. Penis size, as well as breast appearance, is discussed in relation to male opinion. When authors do introduce a female opinion, they quickly add a male opinion to counter it. Good examples Page 12/23 Page 12/23 Page 12/23 In this section, we present some examples of passages that we felt succeeded in describing anatomy without playing into gendered stereotypes. These did not fit into our stated codes because they do not define one sex’s anatomy in terms of how it serves the other sex. In this section, we present some examples of passages that we felt succeeded in describing anatomy without playing into gendered stereotypes. These did not fit into our stated codes because they do not define one sex’s anatomy in terms of how it serves the other sex. It is important to describe the vagina in anatomical terms without referencing how other people interact with the vagina: It is important to describe the vagina in anatomical terms without referencing how other people interact with the vagina: The vagina… is a flexible, muscular structure that extends 3–5 inches back and upward from the vaginal opening. (Book A, p. 65) The vagina is a canal that starts at the vaginal opening and extends into the body. (Book C, p. Men who persuade their partners to have labiaplasty typically want their partners’ labia to conform to the vulva appearance that appeals to them most when watching pornography. (Book H, p.50) 82–83) Here, the hymen is discussed without referencing the concept of virginity, or discussion of how some anatomy is more desirable to men. This passage appropriately does not reference anyone’s opinions or desires related to hymens, and it explains how the hymen’s presence or lack thereof does not always indicate virginity status. It presents factual, accurate information regarding female anatomy without situating it in service to men’s desires, expectations or pleasure. Efforts to change the narrative around anatomy have had success. Some activists have introduced new terms for the hymen to challenge the patriarchal terms about virginity associated with it. Swedish researchers found a lower incidence of Page 13/23 Page 13/23 association with sexual experience for the term “vaginal corona” rather than the traditional Swedish term for the hymen, which translates to “virginity membrane” (Milles et al. 2018). association with sexual experience for the term “vaginal corona” rather than the traditional Swedish term for the hymen, which translates to “virginity membrane” (Milles et al. 2018). Discussion of the penis should not comment on ideal size or import certain appearances over others: The penis… serves important functions in sexual pleasure, reproducion, and elimination of body wastes by urination. It is a tubular organ with an end or tip called the glans. (Book G, p. 77) Appropriately, this section on the penis describes the anatomical features of the penis and the three purposes it serves, ascribing equal importance to each. It does not mention size, shape, or appearance, factors which are often used to value some penises over others and define male anatomy in terms of what women (the assumed male partner) may or may not find desirable. Findings We found, through our statistical analysis and textual investigation, that sexism is rampant in sexuality education literature. The data show a statistically significant correlation with a large effect size between sex and definition of that sex’s anatomy in terms of other people’s values and beliefs. Our data, coupled with a literature analysis, demonstrates that sexuality education often treats women and girls as passive participants in sex, emphasizes men’s opinions about female bodies, and describes the females body as a tool for male pleasure. This is an extremely harmful trend that does not align with recognized attributes of positive, comprehensive sexual education and that is associated with poorer physical and mental health outcomes for girls and women. Importance of comprehensive, inclusive sexual education Good sexual education should be positive, research-based, comprehensive, and inclusive of people of all genders, sexes, and sexualities. Esther Miedema et al. (2020) analyzed a variety of guidelines and standards for comprehensive sexuality education (CSE) and found that the four common goals and themes of CSE are rights, participation, and agency; sexual and reproductive health; gender equality and power; and positive sexualities and respectful relationships. In their paper on the benefits of and challenges to CSE, Doortje Braeken and Melissa Cardinal (2008) argue that CSE gives students the ability to make informed, healthy choices about their sexual decisions. Joan Helmich (2009) incorporates lessons from psychology and studies of youth development groups about how young people learn into sexual education, arguing that good sexual health education for adolescents focuses on and is tailored to the audience. Instructors should also incorporate discussions of pleasure into their teaching materials and lessons. A literature review of arguments for the discussion of pleasure in sex education found that discussing pleasure is thought to improve sexual health outcomes for a variety of underrepresented groups (Koepsel 2016). Jessie Ford et al. (2019) point out that, since pleasure is such a large driver of sexual behavior worldwide, curriculae should frame discussion in pleasure to better relate good sexual health practices to sexual behavior. Focusing sexual education on reproduction centers a heterosexual framework and introduces negativity and shame around sex that is not reproduction-focused (Ingham 2005). To keep sexual education positive and inclusive of all genders, it is important that textbooks do not include the type of sexist language found in our study. Comprehensive, inclusive sex education also has tangible links to improved health behaviors. A 2021 edition of the CDC’s Morbidity and Mortality Weekly Report reported that 30.3% of women aged 18–49 determined to be of ongoing or potential need of contraceptive services did not use any contraception at last sexual encounter (Zapata et al. 2021). This behavior could result in unintended pregnancy or the spread of STI’s and STD’s, and improved education on sexual health could reduce this risky behavior. Risky sexual health behavior is also found in adolescents. Rose Grose et al. (2014)’s survey of eighth graders enrolled in a community sexual health program found that traditional attitudes towards women and agreement with traditional masculinity ideology predicted less safe beliefs about contraception. Impact of sexism on women In general, experiencing sexism leads to anger, depression, and psychological distress in women. Both Jennifer Bosson et al. (2009) and Rebecca Hurst and Denise Beesley (2013) surveyed students at Midwestern universities and found that women who had experienced sexism reported experiencing anger and symptoms of depression, as well as psychological distress, in response to the incident. Janet Swim et al. (2001) conducted three daily diary studies of college students’ experiences of sexism and saw that women’s experiences of sexism increased their feelings of anger, depression, and low self-esteem. Men in the study had similar psychological responses to sexism, but they experienced less sexism, which suggests a smaller overall impact (Swim et al. 2001). Sarah McKenney and Rebecca Bigler (2016) showed that internalized sexualization in adolescent girls is associated with body shame and body surveillance, while Emily Impett et al. (2011) observed that girls with high body objectification have more depressive symptoms and lower self esteem. In three survey studies, Chevallereau et al. (2021) found that, when a woman’s perception that she is only valued for her body increases, so do her felt anger and sadness. Because we studied textbooks used in high school and college sexual health classes, it is Page 14/23 doubly important to recognize the heightened impact of sexism on adolescents and prevent sexist content in sexual education books. Importance of comprehensive, inclusive sexual education High school students notice the bias and prejudice in their sex education, and this might mean that they entirely dismiss their sexual education as out-of-date and miss valuable information (Pound et al. 2016). Because of this, it is critical that textbooks teaching sexual health are inclusive and accurate. College students also want comprehensive sexuality education that covers the whole spectrum of sexual experience, including LGBTQ + people and student experiences that reflect their own lives rather than life in an idealized abstinence world. Emily Gardner (2015)’s interview study of students at two Southern universities showed that students want sex education that understands and recognizes their sex lives and does not assume abstinence. They also wanted thorough information about safe sex and contraceptive measures, as well as information on how to obtain contraception and services like STD testing (Gardner 2015). Another focus group study of college students highlighted a need for up-to-date content that reflects students’ situations, and students also wanted information on safe LGBTQ + sex (Astle et al. 2020). Similar to the first study, Astle et al. (2020)’s group wanted more information on safe sex and contraception. Importance of comprehensive, inclusive sexual education Jozkowski and Crawford (2016) conducted a statistical survey of public health data across the 50 states and found that teens in Alabama, Arkansas, Louisiana, Oklahoma, and Texas reported higher rates of sex without condoms. In these states, education mandates do not require sexuality education but stipulate that, if sexuality education is provided, it must be abstinence-based (Jozkowski and Crawford 2016). These states also have the highest teen pregnancy, teen birth, and STD rates in the country, establishing a correlation between sex education and improved health outcomes (Jozkowski and Crawford, 2016). Page 15/23 Adolescent students recognize the bias in textbooks and see the need for inclusive sexuality education. Coll et al. (2018) surveyed teenage students at a secondary school, who expressed a desire to see more LGBTQ + representation and content. Students also noted that making sexuality education resources and materials more representative of their daily lives and experiences would make it more useful and relevant to them. Allen (2008) found that secondary school students want to be treated as if they have agency in their sexual decisions, and they want to assert their right to express their sexuality positively. We found that sexuality textbooks often treat women as passive, but all students want to be recognized as active participants in sex. High school students notice the bias and prejudice in their sex education, and this might mean that they entirely dismiss their sexual education as out-of-date and miss valuable information (Pound et al. 2016). Because of this, it is critical that textbooks teaching sexual health are inclusive and accurate. Adolescent students recognize the bias in textbooks and see the need for inclusive sexuality education. Coll et al. (2018) surveyed teenage students at a secondary school, who expressed a desire to see more LGBTQ + representation and content. Students also noted that making sexuality education resources and materials more representative of their daily lives and experiences would make it more useful and relevant to them. Allen (2008) found that secondary school students want to be treated as if they have agency in their sexual decisions, and they want to assert their right to express their sexuality positively. We found that sexuality textbooks often treat women as passive, but all students want to be recognized as active participants in sex. Limitations and further research Ultimately, a better understanding of how sexism is expressed in sex education textbooks will allow researchers, writers, and educators to best combat the issue. Curricular recommendations and considerations To address the issue of female anatomy being referred to from the perspective of male ideas and pleasure, we will discuss the major problem topic areas for these incidents and provide the following recommendations for educators. As discussed in our methods section, we worked hard to avoid conflating sex and gender in this paper. However, some conflation was unavoidable due to the books’ pattern of conflation. Because in these texts, definitions and descriptions of female and male anatomy are so tied to cultural expectations of women and men, it was difficult to describe sexist and heteronormative incidents without conflating sex and gender. Texts should clearly indicate the distinctions between sex and gender, and educators should not allow the conflation of the two in their classroom, as it is factually incorrect and extremely offensive to many students. These books were also not written in a vacuum, and the line between scientific, unbiased definitions of anatomy and sexual behavior and the authors’ cultural ideas and beliefs was blurred. Cultural practices around sex and anatomy are important to understand and include, but they should be clearly separated from scientific information so that students understand one as fact and the other as belief. Two areas impacted by this issue were the discussions of breasts and hymens, as mentioned earlier in the “examples” sections of this paper. These two body parts were often explained in the context of male desire and sexual pleasure.We recommend that educators discuss breasts within the sexual anatomy section from a perspective of female sexual pleasure rather than male. Description of hymens should focus on a scientific definition, including the facts, as mentioned earlier, that hymens can be broken through activities other than sex, and that not all female people are born with hymens. Revising the sections on breasts and hymens alone would greatly reduce the amount of sexist language in these textbooks. It is also important to discuss pleasure within sexual health education. Often in these textbooks, male pleasure was emphasized and female pleasure was ignored, even in sections on female anatomy. Limitations and further research We analyzed ten textbooks in wide circulation in human sexuality courses in the United States. However, this is not an all-inclusive or comprehensive review. Therefore, the trends that we discovered are only those that appeared in these books. Furthermore, our analysis does not account for how the textbook content is taught or how teachers handle this material — we focus only on the text itself. Whether teachers address and point out the biases present in these textbooks or enforce them (either passively or actively) likely impacts how students are affected by these textbooks and would have implications for the results of our study. Esther Ohito and Celia Oyler (2017) argue that not enough teachers are themselves educated in implicit biases and creating an inclusive classroom. This implies that teachers are then less likely to point out biases present in curricular materials as well. In future, we hope to highlight other ways that sexism is expressed through sexuality and reproduction textbooks. While reading these texts, we noticed two additional factors present in anatomy chapters: comparisons for purposes of definition of a body part of one sex to a body part of another sex, and value judgements of one sex’s body or behavior. Preliminary results show a similar skew against female bodies in both of these codes, but we want to further analyze these factors. We have already seen that sexism in sex education is a multifaceted, complex issue, and exploring comparisons and value judgements in In future, we hope to highlight other ways that sexism is expressed through sexuality and reproduction textbooks. While reading these texts, we noticed two additional factors present in anatomy chapters: comparisons for purposes of definition of a body part of one sex to a body part of another sex, and value judgements of one sex’s body or behavior. Preliminary results show a similar skew against female bodies in both of these codes, but we want to further analyze these factors. We have already seen that sexism in sex education is a multifaceted, complex issue, and exploring comparisons and value judgements in Page 16/23 Page 16/23 these texts would provide a better picture of the issue. Ultimately, a better understanding of how sexism is expressed in sex education textbooks will allow researchers, writers, and educators to best combat the issue. these texts would provide a better picture of the issue. Limitations and further research Educators and text authors should discuss pleasure for both sexes in order to avoid enforcing the cultural idea introduced earlier that men are active sexual beings and women are passive sexual objects. Inclusive sexuality education necessitates a multi-pronged approach. CSE is one important way to reframe and restructure school-based sexuality education. This more inclusive sex education starts with better research on female sexual pleasure and anatomy, since many incidents we uncovered were based on incorrect information. It is further necessary to create comprehensive curricular materials as well as inclusivity training for educators using these materials to ensure that all students will feel valued, heard, and acknowledged in their sexual health education. Page 17/23 Page 17/23 Declaration of interests The authors report there are no competing interests to declare. The authors report there are no competing interests to declare. Data availability statement The data that support the findings of this study are available from the corresponding author, [CJ], upon request. The data that support the findings of this study are available from the corresponding author, [CJ], upon request. Acknowledgements Several collaborators deserve thanks for their support of this work. Thanks to members of the Research for Equality in Sexuality Education and Texts (RESET) team at The George Washington University- Gaia Norman, Tim Steves, Grace Hallam, Hailey Messina, Madelyn Bedard, Samantha Goffe, Ashlyn Cano, and Sofia Schuchner- for their ideas, discussion, and feedback at all stages of this project. Thank you also to Dr. Elizabeth A. Flaherty (Purdue University) for providing guidance on statistical analysis University) for providing guidance on statistical analysis. References 1. Allen, L. (2004). Beyond the birds and the bees: Constituting a discourse of erotics in sexuality education. Gender and Education, 16(2), 151–167. https://doi.org/10.1080/09540250310001690555. 2. Allen, L. (2007). Denying the sexual subject: Schools' regulation of student sexuality. British Educational Research Journal, 33(2), 221–234. https://www.jstor.org/stable/30032742. 3. Allen, L. (2008). 'They think you shouldn't be having sex anyway': Young people's suggestions for improving sexuality education content. Sexualities, 11(5), 573–594. https://doi.org/10.1177/1363460708089425. 4. Allen, L., & Carmody, M. (2012). ‘Pleasure has no passport’: Re-visiting the potential of pleasure in sexuality education. Sex Education, 12(4), 455–468. https://doi.org/10.1080/14681811.2012.677208. 5. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596 5. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596 6. American Psychiatric Association (2020). APA dictionary of psychology: Heteronormativity. https://dictionary.apa.org/heteronormativity 6. American Psychiatric Association (2020). APA dictionary of psychology: Heteronormativity. https://dictionary.apa.org/heteronormativity 7. Astle, S., McAllister, P., Emanuels, S., Rogers, J., Toews, M., & Yazedjian, A. (2021). College students’ suggestions for improving sex education in schools beyond ‘blah blah blah condoms and STDs’. Sex Education, 21(1), 91–105. https://doi.org/10.1080/14681811.2020.1749044. Page 18/23 Page 18/23 8. Bay-Cheng, L. Y. (2003). The trouble of teen sex: The construction of adolescent sexuality through school-based sexuality education. Sex Education, 3(1), 61–74. https://doi.org/10.1080/1468181032000052162. 9. Beyer, C. E., Ogletree, R. J., Ritzel, D. O., Drolet, J. C., Gilbert, S. L., & Brown, D. (1996). Gender representation in illustrations, text, and topic areas in sexuality education curricula. Journal of School Health, 66(10), 361–364. https://doi.org/10.1111/j.1746-1561.1996.tb03393.x. 10. Bosson, J. K., Pinel, E. C., & Vandello, J. A. (2009). The emotional impact of ambivalent sexism: Forecasts versus real experiences. Sex Roles, 62(7–8), 520–531. https://doi.org/10.1007/s11199- 009-9664-y. 11. Braeken, D., & Cardinal, M. (2008). Comprehensive sexuality education as a means of promoting sexual health. International Journal of Sexual Health, 20(1/2), 50–62. https://doi.org/10.1080/19317610802157051. 12. Chevallereau, T., Stinglhamber, F., Maurage, P., & Demoulin, S. (2021). My physical appearance at the center of others’ concerns: What are the consequences for women’s metadehumanization and emotions? Psychologica Belgica, 61(1), 116–130. https://doi.org/10.5334/pb.558. 13. Christianson, M., & Eriksson, C. (2013). Myths and misconceptions: Midwives' perception of the vaginal opening or hymen and virginity. British Journal of Midwifery, 21(2), 108–115. https://doi.org/10.12968/bjom.2013.21.2.108. 14. Clonan-Roy, K., Goncy, E. A., Naser, S. C., Fuller, K. A., DeBoard, A., Williams, A., & Hall, A. (2021). Preserving abstinence and preventing rape: How sex education textbooks contribute to rape culture. Archives of Sexual Behavior, 50, 231–245. References American Journal of Sexuality Education, 10(2), 125–139. https://doi.org/10.1080/15546128.2015.1015760. 25. Goldfarb, E. S. (2005). What is comprehensive sexuality education really all about? American Journal of Sexuality Education, 1(1), 85–102. https://doi.org/10.1300/J455v01n01_07. 25. Goldfarb, E. S. (2005). What is comprehensive sexuality education really all about? American Journal of Sexuality Education, 1(1), 85–102. https://doi.org/10.1300/J455v01n01_07. 26. Grose, R. G., Grabe, S., & Kohfeldt, D. (2014). Sexual education, gender ideology, and youth sexual empowerment. The Journal of Sex Research, 51(7), 742–753. https://doi.org/10.1080/00224499.2013.809511. 27. Helmich, J. (2009). What is comprehensive sexuality education? Going WAAAAAY beyond abstinence and condoms. American Journal of Sexuality Education, 4(1), 10–15. https://doi.org/10.1080/15546120902870315. 27. Helmich, J. (2009). What is comprehensive sexuality education? Going WAAAAAY beyond abstinence and condoms. American Journal of Sexuality Education, 4(1), 10–15. https://doi org/10 1080/15546120902870315 28. Hurst, R., & Beesley, D. (2013). Perceived sexism, self-silencing, and psychological distress in college women. Sex Roles, 68(5), 311–320. https://doi.org/10.1007/s11199-012-0253-0. 28. Hurst, R., & Beesley, D. (2013). Perceived sexism, self-silencing, and psychological distress in college women. Sex Roles, 68(5), 311–320. https://doi.org/10.1007/s11199-012-0253-0. 29. Howard, B. A., & Gusterson, B. A. (2000). Human breast development. Journal of Mammary Gland Biology and Neoplasia, 5(2), 119–137. https://doi.org/10.1023/a:1026487120779. 29. Howard, B. A., & Gusterson, B. A. (2000). Human breast development. Journal of Mammary Gland Biology and Neoplasia, 5(2), 119–137. https://doi.org/10.1023/a:1026487120779. 30. Impett, E. A., Henson, J. M., Breines, J. G., Schooler, D., & Tolman, D. L. (2011). Embodiment feels better: Girls’ body objectification and well-being across adolescence. Psychology of Women Quarterly, 35(1), 46–58. https://doi.org/10.1177/0361684310391641. 30. Impett, E. A., Henson, J. M., Breines, J. G., Schooler, D., & Tolman, D. L. (2011). Embodiment feels better: Girls’ body objectification and well-being across adolescence. Psychology of Women Quarterly, 35(1), 46–58. https://doi.org/10.1177/0361684310391641. 31. Ingham, R. (2005). ‘We didn't cover that at school’: Education against pleasure or education for pleasure? Sex Education, 5(4), 375–388. https://doi.org/10.1080/14681810500278451. 31. Ingham, R. (2005). ‘We didn't cover that at school’: Education against pleasure or education for pleasure? Sex Education, 5(4), 375–388. https://doi.org/10.1080/14681810500278451. 32. Jackson, S. (Ed.). (1999). How to make babies: Sexism in sex educationHeterosexuality In Question (pp. 57–72). SAGE Publications Ltd. https://doi.org/10.4135/9781446217382.n4. 32. Jackson, S. (Ed.). (1999). How to make babies: Sexism in sex educationHeterosexuality In Question (pp. 57–72). SAGE Publications Ltd. https://doi.org/10.4135/9781446217382.n4. 33. Jozkowski, K. N., & Crawford, B. L. (2016). The status of reproductive and sexual health in southern USA: Policy recommendations for improving health outcomes. Sexuality Research and Social Policy, 13(3), 252–262. https://doi.org/10.1007/s13178-015-0208-7. 33. Jozkowski, K. References 15. Coll, L., O'Sullivan, M., & Enright, E. (2018). 'The trouble with normal': (Re)imagining sexuality education with young people. Sex Education, 18(2), 157–171. https://doi.org/10.1080/14681811.2017.1410699. 16. Connell, C., & Elliott, S. (2009). Beyond the birds and the bees: Learning inequality through sexuality education. American Journal of Sexuality Education, 4(2), 83–102. https://doi.org/10.1080/15546120903001332. 16. Connell, C., & Elliott, S. (2009). Beyond the birds and the bees: Learning inequality through sexuality education. American Journal of Sexuality Education, 4(2), 83–102. https://doi.org/10.1080/15546120903001332. 17. Cook, R. J., & Dickens, B. M. (2009). Hymen reconstruction: Ethical and legal issues. International Journal of Gynecology & Obstetrics, 107(3), 266–269. https://doi.org/10.1016/j.ijgo.2009.07.032. 17. Cook, R. J., & Dickens, B. M. (2009). Hymen reconstruction: Ethical and legal issues. International Journal of Gynecology & Obstetrics, 107(3), 266–269. https://doi.org/10.1016/j.ijgo.2009.07.032. 18. DePalma, R., & Francis, D. A. (2014). The gendered nature of South African teachers’ discourse on sex education. Health Education Research, 29(4), 624–632. https://doi.org/10.1093/her/cyt117. 18. DePalma, R., & Francis, D. A. (2014). The gendered nature of South African teachers’ discourse on sex education. Health Education Research, 29(4), 624–632. https://doi.org/10.1093/her/cyt117. 19. Eastham, R., & Hanbury, A. (2021). Education from sexual pleasure workshops with self-defining women: A commentary. Sex Education, 21(3), 319–330. https://doi.org/10.1080/14681811.2020.1801408. 19. Eastham, R., & Hanbury, A. (2021). Education from sexual pleasure workshops with self-defining women: A commentary. Sex Education, 21(3), 319–330. https://doi.org/10.1080/14681811.2020.1801408. 20. Fine, M. (1988). Sexuality, schooling, and adolescent females: The missing discourse of desire. Harvard Educational Review, 58(1), 29–54. https://doi.org/10.17763/haer.58.1.u0468k1v2n2n8242. 20. Fine, M. (1988). Sexuality, schooling, and adolescent females: The missing discourse of desire. Harvard Educational Review, 58(1), 29–54. https://doi.org/10.17763/haer.58.1.u0468k1v2n2n8242 21. Ford, J. V., Vargas, C., Finotelli, E. Jr., Fortenberry, I., Kismödi, J. D., Philpott, E., Rubio-Aurioles, A., E., & Coleman, E. (2019). Why pleasure matters: Its global relevance for sexual health, sexual rights and Page 19/23 wellbeing. International Journal of Sexual Health, 31(3), 217–230. https://doi.org/10.1080/19317611.2019.1654587. wellbeing. International Journal of Sexual Health, 31(3), 217–230. https://doi.org/10.1080/19317611.2019.1654587. 22. Fredrickson, B. L., & Roberts, T. (1997). Objectification theory. Psychology of Women Quarterly, 21(2), 173–206. https://doi.org/10.1111/j.1471-6402.1997.tb00108.x. 23. García, L. (2009). “Now why do you want to know about that?”: Heteronormativity, sexism, and racism in the sexual (mis)education of Latina youth. Gender & Society, 23(4), 520–541. https://doi.org/10.1177/0891243209339498. 24. Gardner, E. A. (2015). Abstinence-only sex education: College students’ evaluations and responses. American Journal of Sexuality Education, 10(2), 125–139. https://doi.org/10.1080/15546128.2015.1015760. 24. Gardner, E. A. (2015). Abstinence-only sex education: College students’ evaluations and responses. References N., & Crawford, B. L. (2016). The status of reproductive and sexual health in southern USA: Policy recommendations for improving health outcomes. Sexuality Research and Social Policy, 13(3), 252–262. https://doi.org/10.1007/s13178-015-0208-7. 34. Kimberley, N., Hutson, J. M., Southwell, B. R., & Grover, S. R. (2012). Vaginal agenesis, the hymen, and associated anomalies. Journal of Pediatric & Adolescent Gynecology, 25(1), 54–58. https://doi.org/10.1016/j.jpag.2011.08.003. 34. Kimberley, N., Hutson, J. M., Southwell, B. R., & Grover, S. R. (2012). Vaginal agenesis, the hymen, and associated anomalies. Journal of Pediatric & Adolescent Gynecology, 25(1), 54–58. https://doi.org/10.1016/j.jpag.2011.08.003. 35. Koepsel, E. R. (2016). The power in pleasure: Practical implementation of pleasure in sex education classrooms. American Journal of Sexuality Education, 11(3), 205–265. https://doi.org/10.1080/15546128.2016.1209451. 35. Koepsel, E. R. (2016). The power in pleasure: Practical implementation of pleasure in sex education classrooms. American Journal of Sexuality Education, 11(3), 205–265. https://doi.org/10.1080/15546128.2016.1209451. Page 20/23 36. Lamb, S., Graling, K., & Lustig, K. (2011). Stereotypes in four current AOUM sexuality education curricula: Good girls, good boys, and the new gender equality. American Journal of Sexuality Education, 6(4), 360–380. https://doi.org/10.1080/15546128.2011.624477. 37. Lindberg, L. D., Maddow-Zimet, I., & Boonstra, H. (2016). Changes in adolescents' receipt of sex education, 2006–2013. Journal of Adolescent Health, 58(6), 621–627. https://doi.org/10.1016/j.jadohealth.2016.02.004. 38. Martinez, G., Abma, J., & Copen, C. (2010). Educating teenagers about sex in the United States. NCHS Data Brief, (44), 1–8. https://www.cdc.gov/nchs/data/databriefs/db44.pdf 39. McCann, J., Miyamoto, S., Boyle, K., & Rogers, K. (2007). Healing of hymenal injuries in prepubertal and adolescent girls: A descriptive study. Pediatrics, 119(5), e1094–e1106. https://doi.org/10.1542/peds.2006-0964. 40. McKenney, S. J., & Bigler, R. S. (2016). Internalized sexualization and its relation to sexualized appearance, body surveillance, and body shame among early adolescent girls. The Journal of Early Adolescence, 36(2), 171–197. https://doi.org/10.1177/0272431614556889. 40. McKenney, S. J., & Bigler, R. S. (2016). Internalized sexualization and its relation to sexualized appearance, body surveillance, and body shame among early adolescent girls. The Journal of Early Adolescence, 36(2), 171–197. https://doi.org/10.1177/0272431614556889. 41. Miedema, E., Mat, L., Marielle, L. J., & Hague, F. (2020). But is it comprehensive? Unpacking the ‘comprehensive’ in comprehensive sexuality education. Health Education Journal, 79(7), 747–762. https://doi.org/10.1177/0017896920915960. 41. Miedema, E., Mat, L., Marielle, L. J., & Hague, F. (2020). But is it comprehensive? Unpacking the ‘comprehensive’ in comprehensive sexuality education. Health Education Journal, 79(7), 747–762. https://doi.org/10.1177/0017896920915960. 42. Milles, K., Holmdahl, J., Melander, I., & Fugl-Meyer, K. (2018). References ‘Something that stretches during sex’: Replacing the word 'hymen' with 'vaginal corona' to challenge patriarchal views on virginity. Gender and Language, 12(3), 294–317. https://doi.org/10.1558/genl.31894. 43. Muehlenhard, C. L., & Peterson, Z. D. (2011). Distinguishing between sex and gender: History, current conceptualizations, and implications. Sex Roles, 64(11–12), 791–803. https://doi.org/10.1007/s11199-011-9932-5. 44. National Conference of State Legislatures (2020, October 10). State policies on sex education in schools. https://www.ncsl.org/research/health/state-policies-on-sex-education-in-schools.aspx 45. Ohito, E. O., & Oyler, C. (2017). Feeling our way toward inclusive counter-hegemonic 46. pedagogies in teacher education. In L. Florian & N. Pantić (Eds.), Teacher education for the changing demographics of schooling: Issues for research and practice (pp.183–198).Springer Cham. 47. Pillard, C. T. L. (2007). Our other reproductive choices: Equality in sex education, contraceptive access, and work-family policy.Georgetown Law Faculty Publications and Other Works.188. https://scholarship.law.georgetown.edu/facpub/188 47. Pillard, C. T. L. (2007). Our other reproductive choices: Equality in sex education, contraceptive access, and work-family policy.Georgetown Law Faculty Publications and Other Works.188. https://scholarship.law.georgetown.edu/facpub/188 48. Pound, P., Langford, R., & Campbell, R. (2016). What do young people think about their school-based sex and relationship education? A qualitative synthesis of young people's views and experiences. British Medical Journal Open, 6(9), e011329. https://doi.org/10.1136/bmjopen-2016-011329. 48. Pound, P., Langford, R., & Campbell, R. (2016). What do young people think about their school-based sex and relationship education? A qualitative synthesis of young people's views and experiences. British Medical Journal Open, 6(9), e011329. https://doi.org/10.1136/bmjopen-2016-011329. 49. Puppo, V. (2013). Anatomy and physiology of the clitoris, vestibular bulbs, and labia minora with a review of the female orgasm and the prevention of female sexual dysfunction. Clinical Anatomy, 26(1), 134–152. https://doi.org/10.1002/ca.22177. 49. Puppo, V. (2013). Anatomy and physiology of the clitoris, vestibular bulbs, and labia minora with a review of the female orgasm and the prevention of female sexual dysfunction. Clinical Anatomy, 26(1), 134–152. https://doi.org/10.1002/ca.22177. Page 21/23 Page 21/23 50. Rich, A. (1980). Compulsory heterosexuality and lesbian existence. Signs, 5(4), 631–660. https://www.jstor.org/stable/3173834. 51. Schudson, Z. C., Beischel, W. J., & van Anders, S. M. (2019). Individual variation in gender/sex category definitions. Psychology of Sexual Orientation and Gender Diversity, 6(4), 448–460. https://doi.org/10.1037/sgd0000346. 52. Swim, J. K., Hyers, L. L., Cohen, L. L., & Ferguson, M. J. (2001). Everyday sexism: Evidence for its incidence, nature, and psychological impact from three daily diary studies. Journal of Social Issues, 57(1), 31–53. https://doi.org/10.1111/0022-4537.00200. 53. Tolman, D. (1994). Doing desire: Adolescent girls' struggles for/with sexuality. Gender and Society, 8(3), 324–342. https://www.jstor.org/stable/189709. References 54. Ward, L. M., Merriwether, A., & Caruthers, A. (2006). Breasts are for men: Media, masculinity ideologies, and men’s beliefs about women’s bodies. Sex Roles, 55(9), 703–714. https://doi.org/10.1007/s11199-006-9125-9. 55. Whatley, M. H. (1988). Raging hormones and powerful cars: The construction of men's sexuality in school sex education and popular adolescent films. The Journal of Education, 170(3), 100–121. https://www.jstor.org/stable/42748650. 55. Whatley, M. H. (1988). Raging hormones and powerful cars: The construction of men's sexuality in school sex education and popular adolescent films. The Journal of Education, 170(3), 100–121. https://www.jstor.org/stable/42748650. 56. Zapata, L. B., Pazol, K., Curtis, K. M., Kane, D. J., (Eds.),, E. M. Cox, S., &amp 56. Zapata, L. B., Pazol, K., Curtis, K. M., Kane, D. J., (Eds.),, E. M. Cox, S., &amp 57. Whiteman, M. K. (2021). Need for contraceptive services among women of reproductive age — 45 jurisdictions, United States, 2017–2019. Morbidity and Mortality Weekly Report, 70, 910–915. https://doi.org/10.15585/mmwr.mm7025a2. 57. Whiteman, M. K. (2021). Need for contraceptive services among women of reproductive age — 45 jurisdictions, United States, 2017–2019. Morbidity and Mortality Weekly Report, 70, 910–915. https://doi.org/10.15585/mmwr.mm7025a2. Figures Page 22/23 Figure 1 Incidence of description of male (m) and female (f) anatomy from the perspective of another sex in a study of the anatomy chapters of ten sexuality and reproduction textbooks commonly used in American high school and college courses. Figure 1 Figure 1 Incidence of description of male (m) and female (f) anatomy from the perspective of another sex in a study of the anatomy chapters of ten sexuality and reproduction textbooks commonly used in American high school and college courses. Page 23/23
https://openalex.org/W2895878639
https://europepmc.org/articles/pmc6174569?pdf=render
English
null
Alpha-1 antitrypsin deficiency as a common treatable mechanism in chronic respiratory disorders and for conditions different from pulmonary emphysema? A commentary on the new European Respiratory Society statement
Multidisciplinary respiratory medicine
2,018
cc-by
9,482
COMMENTARY Open Access Alpha-1 antitrypsin deficiency as a common treatable mechanism in chronic respiratory disorders and for conditions different from pulmonary emphysema? A commentary on the new European Respiratory Society statement Andrea Gramegna1, Stefano Aliberti1* , Marco Confalonieri2, Angelo Corsico3, Luca Richeldi4, Carlo Vancheri5 and Francesco Blasi1 Andrea Gramegna1, Stefano Aliberti1* , Marco Confalonieri2, Angelo Corsico3, Luca Richeldi4, Carlo Vancheri5 and Francesco Blasi1 Andrea Gramegna1, Stefano Aliberti1* , Marco Confalonieri2, Angelo Corsico3, Luca Richeldi4, Carlo Vancheri5 and Francesco Blasi1 * Correspondence: stefano.aliberti@unimi.it * Correspondence: stefano.aliberti@unimi.it p 1Department of Pathophysiology and Transplantation, University of Milano; Internal Medicine Department, Respiratory Unit and Adult Cystic Fibrosis Center, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy y Full list of author information is available at the end of the article Abstract Background: The European Respiratory Society recently published an important statement reviewing available evidence on diagnosis and treatment of lung disease associated to alpha-1 antitrypsin deficiency (AATD). Several issues on this topic still remain unresolved and subject of interpretation according to different standard procedures and healthcare systems worldwide. The purpose of this commentary is to offer a critical contribution to most of these controversial issues in light of an Italian perspective for the management of this disease. Main body: The clinical spectrum of AATD lung disease might include different manifestations and the traditional paradigm of a younger emphysematous patient has been revealing insufficient. Targeting with appropriate testing only COPD patients might be considered a limited approach leading to underestimation of the real prevalence of the disease. Several reports have suggested the association between AATD and other chronic respiratory conditions, as asthma and bronchiectasis. A deeper evaluation of clinical, radiological, microbiological and functional variables is, therefore, needed in order to investigate different phenotypes in AATD patients. In addition, a new line of translational research in AATD might focus on the development of personalized therapeutic regimens taking into account the patient clinical profile and needs. Conclusions: Over the past years, AATD has been interpreted as a common mechanism of inflammatory disequilibrium and tissue damage across different conditions. Future research is gradually pointing toward this new paradigm by expanding the evidence of the role of AAT as a potent immunomodulatory and anti-inflammatory drug in conditions different from pulmonary emphysema. Keywords: Alpha-1 antitrypsin deficiency, Bronchiectasis, COPD, ERS statement * Correspondence: stefano.aliberti@unimi.it 1Department of Pathophysiology and Transplantation, University of Milano; Internal Medicine Department, Respiratory Unit and Adult Cystic Fibrosis Center, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waive (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise state Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 https://doi.org/10.1186/s40248-018-0153-4 Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 https://doi.org/10.1186/s40248-018-0153-4 Background heterogeneity to pulmonary involvement in AATD. However, in their considerations authors relied on the classical phenotype of a patient with onset of COPD at a younger age and more likely to have basal panlobular emphysema than usual COPD patients [10]. This is, in our view, a crucial point. g Alpha-1 Antitrypsin Deficiency (AATD) is the most common hereditary disorder in adults [1]. There are sev- eral mutations of the SERPINA1 gene, encoding for AAT which is the most prevalent protease inhibitor in serum, that can cause disease. Despite its name, the major biological role of AAT is to inhibit neutrophil elastase (NE), a proteolytic enzyme targeting elastin and other basal membrane and matrix components [2]. AATD is inherited as an autosomal codominant disease and more than 100 genetic variants have been described so far and the homozigous genotype for Z allele (PI*ZZ) accounts for the vast majority of clinically recognized se- vere deficiency [3]. Severe AATD is a multi-organ dis- order that might be associated with a large spectrum of lung disease with resulting FEV1 decline and impairment in gas transfer. Although the clinical course is not fully explained, patients with AATD have increased mortality, especially in the presence of smoking and severe FEV1 impairment [4–6]. Over the past years, a few random- ized clinical trials have reported augmentation therapy with intravenous purified human AAT concentrate to improve AAT blood levels and reduce FEV1 or, as re- cently described, lung density decline in severe AATD patients [1, 7–9]. AATD pulmonary involvement might include emphy- sema with severe functional and radiological impairment as well as different pulmonary pictures or, even, normal chest CT scan [11]. In a series of PI*ZZ patients, Gishen analyzed 165 chest radiographs and reported emphy- sema pattern only in 20% of patients, while 15% ap- peared completely normal [12]. DeMeo and colleagues demonstrated in a population of 378 patients with severe AATD (PI*ZZ) that variability in FEV1 was huge and did not relate to smoking status or younger age. Thus, some smokers had preserved lung function, while some non-smokers had lower FEV1 [13]. These data suggest that neither FEV1 nor chest radiology are sufficient to rise clinical suspicion of AATD in daily practice. As a consequence, most of severe AATD patients might not be still identified and case finding only among COPD patients is not sufficient. © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Page 2 of 10 Page 2 of 10 Page 2 of 10 Background In addition, the unusual associ- ation between most common deficitary alleles in AATD and idiopathic pulmonary fibrosis (IPF), or combined pulmonary fibrosis and emphysema (CPFE) have also been described [14, 15]. p In 2017 the European Respiratory Society (ERS) pub- lished a renewed statement on diagnosis and treatment of pulmonary disease in AATD [10]. This document was developed by a task force including experts in AATD pa- tient management, basic and clinical researchers and one methodologist. The previous 2003 ATS/ERS state- ment was used as a starting point [1]. The panelists ad- dressed the most relevant issues on diagnosis, clinical management and treatment of pulmonary disease in AATD patients by the formulation of a priori questions and answered through a comprehensive search of litera- ture. The search was limited to severe AATD cohorts (i.e. homozygous Z or null genotype and Z/null geno- type). A GRADE approach to degree recommendations according to the quality of evidences was not adopted. A summary of ERS statements is reported in Table 1. Targeting with appropriate testing only those patients with classic AATD clinical presentation might be con- sidered a limited approach leading to underestimation of the real prevalence of the disease. In view of the extreme variety of this condition, the paradigm of pure emphyse- matous patients has gradually changed by the under- standing that the clinical spectrum of AATD lung disease might include different phenotypes (Fig. 1). AATD and asthma Laboratory diagnosis and hierarchy of testing • The quantitative determination of AAT levels in blood is a crucial first test to identify AATD. Quantitative deficiency must be supported by qualitative tests to identify the genetic mutation(s) causing AATD. • Protein phenotyping by isoelectric focusing identifies variants where AAT is present in the sample including the rarer variants F, I and P etc. • Genotyping allows a rapid and precise identification/exclusion of S and Z alleles and other variants, where specific primers are available. • Gene sequencing remains necessary for those cases where a null variant or a deficient variant other than Z or S is suspected. • Testing of relatives of identified patients should be considered after appropriate counselling. • Genetic testing should be carried out only after informed consent is given and in accordance with the relevant guidelines and legislation. Lung disease progression in AATD • Annual measurement of lung function including post-bronchodilator FEV1 and gas transfer provides information about disease progression. • Lung densitometry, as performed in observational cohort studies and randomised clinical trials is the most sensitive measure of emphysema progression. • The correlation between change in lung density and any short-term change in measures of pulmonary function is weak. However, in the longer term, CT lung density decline correlates with reduction in FEV1 and health status. • The role of CT in the follow-up of patients in routine clinical practice requires further validation. The risk of lung disease in heterozygotes • Never-smoking PiMZ subjects do not have an increased risk for COPD. • Smoking PiMZ and PiSZ subjects have an increased risk of COPD compared to smoking PiMM subjects. • The role of other heterozygotes remains unknown due to their rarity and potential ascertainment bias from measuring AAT in unusual cases of lung or liver disease. • Most screening studies have been biased as they did not involve random population samples. • Population-based screening studies provide less biased prevalence estimates of specific AATD protein and clinical phenotypes as well as valuable insights into the natural history of AATD. • Neonatal screening has been shown to be effective in reducing the smoking rates for 18–20-year-olds compared to age-matched individuals. • Screening may have negative psychological effects on parents and on mother–child bonding. However, these negative effects can be addressed by comprehensive genetic counselling and care provision at centres of excellence for AATD. AATD and asthma • Gene sequencing remains necessary f Z or S is suspected. • Testing of relatives of identified patien • Genetic testing should be carried out relevant guidelines and legislation. Lung disease progression in AATD • Annual measurement of lung function information about disease progression • Lung densitometry, as performed in o sensitive measure of emphysema prog • The correlation between change in lu function is weak. However, in the long and health status. • The role of CT in the follow-up of pat The risk of lung disease in heterozygotes • Never-smoking PiMZ subjects do not • Smoking PiMZ and PiSZ subjects have • The role of other heterozygotes rema measuring AAT in unusual cases of lu Role and benefits of screening • Most screening studies have been bia • Population-based screening studies pr clinical phenotypes as well as valuable • Neonatal screening has been shown t compared to age-matched individuals • Screening may have negative psychol these negative effects can be address centres of excellence for AATD. Augmentation therapy for AATD • Several randomised clinical trials in se the progression of emphysema as ass • There is no evidence to support effica any protein phenotype. • Clinical trials have used fixed doses of based on trough levels for each patie Lung volume reduction surgery in AATD • Surgical volume reduction and EBV pl further studies are needed to confirm • The optimal results of these technique performed by a multidisciplinary team Lung transplantation for emphysema associated with AATD • The survival benefit of lung transplant • In general, patients with AATD have im • Referral timing, rate of decline in lung patient, and will have an influence on • The role of post-transplant augmentat • The clinical impact of AATD is highly variable. Heterogeneity in lung disease is only partly explained by exposure to known risk factors, such as cigarette smoke. • Lung disease in AATD generally presents at a younger age than “usual” COPD and may be misdiagnosed as asthma. • Although the patients’ clinical phenotype may vary they are more likely to have basal emphysema than patients with usual COPD. • The WHO recommends all patients with a diagnosis of COPD or adult-onset asthma should be tested for AATD. • The WHO recommends all patients with a diagnosis of COPD or adult-onset asthma should be tested for AATD. AATD and asthma A review explored the association and overlap between AATD and asthma [16]. Subsequent to the first descrip- tions of AATD patients, Makino and Black reported case series of emphysematous patients with clinical feature of airway reactivity and wheezing [17, 18]. In the next years, several reports ranged the prevalence of asthma in AATD population from 4 to 34% [19–22]. In the National Heart, Lung and Blood Institute’s Registry of AAT Deficient Individuals (NHLBI), a network of 37 centers across USA and Canada for a total of 1219 sub- jects, 55% of patients reported a significant response to bronchodilation, while concomitant asthma and respira- tory allergy were reported in 31% and 23% of cases, re- spectively [23, 24]. On this basis, in 1997 the WHO recommended that all patients with COPD and adult-onset asthma should be screened for AATD [25]. A recent study conducted through the Alpha-1 The purpose of this commentary is to offer a contribu- tion to most controversial issues regarding pulmonary disease associated with AATD. In presenting these argu- ments, three main topics will be addressed, including 1) the limitation of the current definition of AATD lung disease; 2) the need of broader testing and improvement of AATD awareness among physicians taking care of chronic respiratory diseases; 3) new paradigm and per- spectives in AATD future research. Heterogeneity of AATD associated lung disease In the attempt to give a definition of associated lung dis- ease, the task force attributed a wide degree of g y g In the attempt to give a definition of associated lung dis- ease, the task force attributed a wide degree of Page 3 of 10 Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Table 1 Summary of European Respiratory Society statements on diagnosis, clinical management and treatment of pulmonary disease in α1-antitrypsin deficiency [10] AATD and lung disease • The clinical impact of AATD is highly variable. Heterogeneity in lung disease is only partly explained by exposure to known risk factors, such as cigarette smoke. • Lung disease in AATD generally presents at a younger age than “usual” COPD and may be misdiagnose as asthma. • Although the patients’ clinical phenotype may vary they are more likely to have basal emphysema than patients with usual COPD. • The WHO recommends all patients with a diagnosis of COPD or adult-onset asthma should be tested fo AATD. AATD and asthma Laboratory diagnosis and hierarchy of testing • The quantitative determination of AAT levels in blood is a crucial first test to identify AATD. Quantitative deficiency must be supported by qualitative tests to identify the genetic mutation(s) causing AATD. • Protein phenotyping by isoelectric focusing identifies variants where AAT is present in the sample including the rarer variants F, I and P etc. • Genotyping allows a rapid and precise identification/exclusion of S and Z alleles and other variants, where specific primers are available. • Gene sequencing remains necessary for those cases where a null variant or a deficient variant other tha Z or S is suspected. • Testing of relatives of identified patients should be considered after appropriate counselling. • Genetic testing should be carried out only after informed consent is given and in accordance with the relevant guidelines and legislation. Lung disease progression in AATD • Annual measurement of lung function including post-bronchodilator FEV1 and gas transfer provides information about disease progression. • Lung densitometry, as performed in observational cohort studies and randomised clinical trials is the mo sensitive measure of emphysema progression. • The correlation between change in lung density and any short-term change in measures of pulmonary function is weak. However, in the longer term, CT lung density decline correlates with reduction in FEV1 and health status. • The role of CT in the follow-up of patients in routine clinical practice requires further validation. The risk of lung disease in heterozygotes • Never-smoking PiMZ subjects do not have an increased risk for COPD. • Smoking PiMZ and PiSZ subjects have an increased risk of COPD compared to smoking PiMM subjects. • The role of other heterozygotes remains unknown due to their rarity and potential ascertainment bias fr measuring AAT in unusual cases of lung or liver disease. Role and benefits of screening • Most screening studies have been biased as they did not involve random population samples. • Population-based screening studies provide less biased prevalence estimates of specific AATD protein an clinical phenotypes as well as valuable insights into the natural history of AATD. • Neonatal screening has been shown to be effective in reducing the smoking rates for 18–20-year-olds compared to age-matched individuals. • Screening may have negative psychological effects on parents and on mother–child bonding. However, these negative effects can be addressed by comprehensive genetic counselling and care provision at centres of excellence for AATD. Lung disease progression in AATD Laboratory diagnosis and hierarchy of testing Lung transplantation for emphysema associated with AATD AATD and asthma Augmentation therapy for AATD • Several randomised clinical trials in severe AATD have shown intravenous augmentation therapy to redu the progression of emphysema as assessed by CT densitometry. • There is no evidence to support efficacy of AAT augmentation therapy in PiSZ, PiMZ or current smokers any protein phenotype. • Clinical trials have used fixed doses of AAT determined by body weight. Whether individualising dosage based on trough levels for each patient has any benefit requires confirmation. Lung volume reduction surgery in AATD • Surgical volume reduction and EBV placement may be considered in selected patients with AATD, but further studies are needed to confirm the role of such therapies. • The optimal results of these techniques are obtained when a careful appraisal of risks and benefits are performed by a multidisciplinary team experienced in LVR and AATD. Lung transplantation for emphysema associated with AATD • The survival benefit of lung transplant in AATD patients is not clear. • In general, patients with AATD have improved quality of life following lung transplantation. • Referral timing, rate of decline in lung function, health status and social support differ from patient to patient, and will have an influence on the evaluation for transplant. • The role of post-transplant augmentation therapy in particular needs to be explored. Table 1 Summary of European Respiratory Society statements on diagnosis, clinical management and treatment of pulmonary disease in α1-antitrypsin deficiency [10] Table 1 Summary of European Resp disease in α1-antitrypsin deficiency [1 AATD and lung disease Laboratory diagnosis and hierarchy of testing Lung disease progression in AATD The risk of lung disease in heterozygotes Role and benefits of screening Augmentation therapy for AATD Lung volume reduction surgery in AATD Lung transplantation for emphysema associated with AATD Table 1 Summary of European Respiratory Society statements on diagn disease in α1-antitrypsin deficiency [10] AATD and lung disease • The clinical impact of AATD is highly v exposure to known risk factors, such a • Lung disease in AATD generally prese as asthma. • Although the patients’ clinical phenot patients with usual COPD. • The WHO recommends all patients w AATD. Laboratory diagnosis and hierarchy of testing • The quantitative determination of AAT deficiency must be supported by qua • Protein phenotyping by isoelectric foc including the rarer variants F, I and P • Genotyping allows a rapid and precise where specific primers are available. AATD and bronchiectasis Several reports have suggested the association between AATD and bronchiectasis [21, 31, 32]. In order to address the prevalence and clinical impact of bronchiectasis in this population, Parr and Stockley retrospectively studied 74 subjects with severe AATD who had received a HRCT scan in the time period 1995–2002. They reported an un- expectedly high prevalence of bronchiectatic changes in almost all patients (70/74 subjects, 94.6%) without regard to clinical manifestations. If criteria for clinically signifi- cant bronchiectasis were applied, i.e. chronic cough with sputum production in association with bronchial dilation in at least four bronchial segments, bronchiectasis oc- curred in 27% of study population [33]. Although current literature is controversial about the prevalence of AAT deficiency alleles in asthmatic popu- lation, AATD was reported in 2.4% of patients with poorly controlled asthma with 10.5% being a carrier of a deficiency mutation [27, 28]. This contributed to the in- terpretation that AATD itself might predispose to airway hyper-reactivity and participate to asthma pathophysi- ology. While the underlying biological mechanism is not understood, data from basic research have in fact sug- gested that AAT has also immunomodulatory functions and might affect eosinophilic cells [29]. The concept of association between bronchiectasis and AATD is also supported by data coming from a large targeted screening program for AATD in Germany. Greulich and colleagues analyzed 18,638 testing kits and identified 1835 patients with severe AATD (9.82% of the tested population). Along with emphysema and COPD, bronchiectasis demonstrated to be a strong predictor for PI*ZZ genotype suggesting the implementation of screening in this emerging population [34]. Finally, AATD patients with concomitant asthma have worse prognosis. Bronchodilator response has been asso- ciated with greater FEV1 decline and poorer clinical out- comes [13, 24, 30]. From the clinician’s point of view, improving diagnosis in this population is fundamental to optimize clinical management and further clinical trials are needed to investigate patients’ response to available treatments. From a different perspective, among physicians taking care of bronchiectasis AATD has been recognized as a cause of bronchiectasis and a causal link has been hy- pothesized. The common origin of airway disease in AATD might be explained by the disequilibrium be- tween neutrophilic inflammation and protease inhibi- tors, thus resulting in bronchial connective tissue damage [35]. AATD and asthma • The systematic collection of data concerning clinical characteristics and natural history of patients with AATD in national and international registries will enhance knowledge about the evolution of this disease and its optimal management. • The systematic collection of data concerning clinical characteristics and natural history of patients with AATD in national and international registries will enhance knowledge about the evolution of this disease and its optimal management. • For many AATD individuals a respiratory service is the first point of diagnosis. The operational pathway includes varying assessments and follow-up depending on personalising the patients’ risk and defining the respiratory phenotype. Links to multidisciplinary teams will ensure the best quality of care. Foundation Research Registry, involving 500 patients with severe deficiency, confirmed that clinical manifesta- tions of uncontrolled asthma or asthma with fixed ob- struction are frequent in AATD; however, among the 34% participants undergoing allergological evaluation, only 5% were diagnosed with AATD [26]. AATD and asthma • Several randomised clinical trials in severe AATD have shown intravenous augmentation therapy to reduce the progression of emphysema as assessed by CT densitometry. • There is no evidence to support efficacy of AAT augmentation therapy in PiSZ, PiMZ or current smokers of any protein phenotype. • Clinical trials have used fixed doses of AAT determined by body weight. Whether individualising dosage based on trough levels for each patient has any benefit requires confirmation. • Surgical volume reduction and EBV placement may be considered in selected patients with AATD, but further studies are needed to confirm the role of such therapies. • The optimal results of these techniques are obtained when a careful appraisal of risks and benefits are performed by a multidisciplinary team experienced in LVR and AATD. • The survival benefit of lung transplant in AATD patients is not clear. • In general, patients with AATD have improved quality of life following lung transplantation. • Referral timing, rate of decline in lung function, health status and social support differ from patient to patient, and will have an influence on the evaluation for transplant. • The role of post-transplant augmentation therapy in particular needs to be explored. Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Page 4 of 10 Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Table 1 Summary of European Respiratory Society statements on diagnosis, clinical management and treatment of pulmonary disease in α1-antitrypsin deficiency [10] (Continued) New lines of research in AATD • According to the European Council, management of patients with AATD should be supervised by reference centres of excellence at a national or regional level. • The systematic collection of data concerning clinical characteristics and natural history of patients with AATD in national and international registries will enhance knowledge about the evolution of this disease and its optimal management. • For many AATD individuals a respiratory service is the first point of diagnosis. The operational pathway includes varying assessments and follow-up depending on personalising the patients’ risk and defining the respiratory phenotype. Links to multidisciplinary teams will ensure the best quality of care. Table 1 Summary of European Respiratory Society statements on diagnosis, clinical management and treatment of pulmonary disease in α1-antitrypsin deficiency [10] (Continued) • According to the European Council, management of patients with AATD should be supervised by reference centres of excellence at a national or regional level. Clinical implications: The role and benefits of screening g Since a majority of AATD patients is still unidentified and delays in making the correct diagnosis are common, structured programs of AAT testing have been dis- cussed. ERS statement discussed the role and benefits of screening for AATD and summarized current evidence (Table 1). Most of population screening of adults in the past has been limited to specific groups, as blood do- nors, with resulting esteem of prevalence biased by se- lection [10]. A more accurate evaluation came from programs including random population samples [41, 42]. A large population-based screening was performed in Ireland on 3000 subjects and identified 42 ZZ, 44 SZ and 430 MZ individuals, a prevalence estimation greater than previously thought [47]. Another experience came from Poland where 859 adult subjects (mean age 49.5 years, range 20–90) were screened with resulting fre- quencies of PI*S and PI*Z equal to 17.5 per 1000 and 10.5 per 1000, respectively. Therefore, the estimated prevalence of AATD was high enough to consider implementation of a large-scale screening program [48]. However, the ERS statement reported that no RCT investigated efficacy and effectiveness of AATD screening programs and evidence for this approach is currently poor [10]. Although this amount of data documented the role of AATD as an underlying etiology of bronchiectasis, it is possible to presume that a majority of patients never re- ceived AAT testing. In 2014 the Italian Respiratory Soci- ety (IRS/SIP) conducted a national audit on adult patients with bronchiectasis attending secondary care clinics in Italy [41]. The audit analysed adherence among physicians to quality standards as suggested by the Brit- ish Thoracic Society. To date, this represents the only testing of quality standards for bronchiectasis in Italy. In regard to etiological screening of bronchiectasis, only 8.2% of patients were tested for AATD. As a result, it is likely to hypothesize that estimation of AATD preva- lence in bronchiectasis population is poor. Notably, both in the European analysis as well as in the Italian report COPD was among the three most prevalent cause of bronchiectasis with 15% and 7.8% of population, respectively [37] http://www.world-bronchiectasis-confer- ence.org/wp-content/uploads/2017/07/AbstractBook_WE B.pdf. The association between COPD and bronchiectasis might represent a distinct phenotype in patients charac- terized by chronic respiratory infections, more severe air- way inflammatory pattern and frequent pulmonary exacerbations [42]. AATD and bronchiectasis In addition, AAT has been demonstrated to act as an anti-inflammatory mediator by the inhib- ition of TNF-α and IL-1β levels in mononuclear cell cul- tures [36]. However, whether bronchiectasis comes from a primary mechanism of the disease or is the result of recurrent respiratory infection is still matter of debate. Fig. 1 AATD might be interpreted as a common mechanism with different clinical manifestations and frequent overlap among chronic respiratory disorders In 2015 Lonni and co-authors conducted an extensive analysis of seven cohorts of adult outpatients with bron- chiectasis prospectively enrolled between 2009 and 2013 in different countries across Europe [37]. All patients underwent the same etiological workup as suggested by BTS guidelines, including quantitative analysis of AAT in plasma in those subjects with radiological Fig. 1 AATD might be interpreted as a common mechanism with different clinical manifestations and frequent overlap among chronic respiratory disorders Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Page 5 of 10 Page 5 of 10 emphysematous changes. AATD was identified as a cause of bronchiectasis in a total of 8 (0.6%) of popula- tion. This result is consistent with those reported by a comprehensive review performed to identify papers reporting the underlying etiologies of published cohorts; AATD was identified as a cause of bronchiectasis in 36 (0.4%) out of 8608 patients [38]. The vast majority of those patients came from Europe (n = 23) and Northern America (n = 12), suggesting that only a portion of stud- ies performed in developed countries had provided some sort of testing on AATD. reports demonstrated an increasing role of NTM in pa- tients with underlying respiratory disorders, as bronchi- ectasis and COPD [43–45]. However, susceptibility to the development of NTM pulmonary disease is different among patients and AATD has been postulated as a po- tential predisposing factor. In a cohort of 100 patients with NTM pulmonary disease, AAT genetic variants were reported in 27% of cases, which is 1.6 times than estimated prevalence in general population. Most of those patients showed Pi*MS and Pi*MZ heterozigosity and no sign of pulmonary emphysema. In addition, AAT has also been demonstrated to impair M. abscessus phagocytosis by human macrophages, thus forbidding NTM the ideal intracellular milieu [46]. Thus, in consid- eration of its pleiotrophic anti-inflammatory and immu- nomodulatory effects, AATD might increase patient vulnerability to NTM both directly and through bron- chiectasis development. AATD and bronchiectasis Over the past decade, different national and international registries have been set up to evaluate the real-life charac- teristics of patients with different respiratory diseases, in- cluding bronchiectasis, severe asthma and NTM [39] https://www.registroirene.it/it/user [40]. Looking into the Italian situation, the first data from the Italian Registry on Adult Bronchiectasis (IRIDE) have been presented at the 2nd World Bronchiectasis Conference http://www.world -bronchiectasis-conference.org/wp-content/uploads/2017/ 07/AbstractBook_WEB.pdf. Among 522 adults with bron- chiectasis (to May 2017) the diagnosis of AATD was less than 1% of population. Clinical implications: The role and benefits of screening However, the question if COPD and bronchiectasis co-exist as two indipendent diseases or one is the cause of the other still remains unanswered. This bundle of evidence has been evaluated differently across guidelines regarding chronic respiratory condition that might be related to AATD. Although AATD is strongly associated to the development of COPD, The Global Initiative for Chronic Obstructive Lung Disease (GOLD) document edited in 2018 mentioned AATD in only two occasions. First, as a known genetic risk factor for COPD and an example of gene-environment inter- action; second, as a step in COPD assessment that WHO guidelines recommended to screen once in all AATD and NTM Nontuberculous mycobacteria (NTM) are a large num- ber of pathogenic and non-pathogenic mycobacterial species different than m. tubercolosis complex. Recent Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Page 6 of 10 priorities. In this respect, AATD patients should be pro- vided with proper genetic counseling and support. patients with a diagnosis of COPD, especially in areas with high AATD prevalence [49]. On the contrary, Glo- bal Initiative for Asthma Management and Prevention (GINA) document has only one mention of AATD as a possible differential diagnosis in patient with suspected asthma in presence of family history of emphysema and aged between 12 and 39 years. Notably, no indication for AATD testing in asthmatic patients can be found in this document [50]. Different perspectives of AATD definition and indications for augmentation therapy The definition of AATD might rise some controver- sial among physicians in respiratory care. Three dif- ferent scenarios are possible: 1) deficiency as the condition of lower-than-normal AAT levels in blood; 2) deficiency as the result of mutations affecting SER- PINA gene; 3) deficiency as the indication for aug- mentation therapy for AATD. In regard to bronchiectasis, the scenario is more com- plex since international guidelines did not exist until re- cently. A few national guidelines did not recommend routine AATD testing unless the radiological investiga- tions suggest basal emphysema [51, 52]. Otherwise, guide- lines from Pulmonology Portuguese Society included AAT quantitative test in the minimum bundle of etiological screening for all patients with bronchiectasis [53]. In 2017 the ERS published the first international standards of care for bronchiectasis in adult patients [54]. This document suggests providing AATD testing only in the presence of basal emphysema or early onset airway obstruction. A comprehensive summary of recommendations for AATD testing in different documents is provided in Table 2. AAT deficiency as the condition of lower AAT levels in blood. This definition is focused on the quantita- tive determination of blood levels and represent the widest interpretation of deficient status. Currently nephelometry and immunoturbidimetric assay are the most widely employed techniques; in this regard, ERS statement reported a comprehensive list of laboratory methodologies used in central laboratories across Europe [10]. g p This lack of common ground in testing AAT across re- spiratory diseases needs further considerations. Identify- ing AATD is fundamental in view of clinical, social and psychological consequences in target population. First, AATD is a systemic disease with extra-respiratory clin- ical manifestations that might benefit from a specific management with positive impact on clinical outcomes and patients’ quality of life. AATD patients should have access to reference centers, that are able to provide the best standard of care and contribute to clinical data net- working. Furthermore, patients affected by a rare disease might benefit from free care or proper health insurance issues across different jurisdictions. From a psychological perspective, AATD is both a genetic disease with in- creased risk of affected children and a chronic condition with possible impact on family planning and patients’ life AAT deficiency as the result of a genetic disease AAT deficiency as the result of a genetic disease As suggested by ERS statement, abnormal quantitative measurements of AAT levels in blood must be sup- ported by a qualitative test detecting SERPINA gene mu- tations on both copies of the gene [7]. The normal allele, designated as M, is present in 85–90% of population, while the prevalent deficient alleles are designated as S and Z [1]. As a consequence, the most common geno- types result from the combination of these allelic forms: the homozygous genotype Pi*MM (wild-type), the het- erozygous Pi*MZ and Pi*MS and the semi-deficient or deficient genotypes Pi*SZ, Pi*SS and Pi*ZZ. Besides defi- cient alleles S and Z, about 50 other uncommon muta- tions have been described, including (null) allele and other rare variants [55]. Table 2 A comprehensive summary of indications for AATD testing across documents for the clinical management of chronic respiratory disease Document Disease Indications for AATD testing Global strategy for prevention, diagnosis and management of COPD, 2018 [49] COPD All patients with a diagnosis of COPD, especially in areas with high AATD prevalence Global strategy for asthma management and prevention, 2018 [50] Asthma No indication for AATD testing European Respiratory Society, 2017 [54] Bronchiectasis Patients with basal emphysema or early onset airflow obstruction Pulmonology Portuguese Society Bronchiectasis Study Group, 2018 [53] Bronchiectasis All patients with a diagnosis of BE Thoracic Society of Australia and New Zealand, 2015 [62] Bronchiectasis No indication for AATD testing British Thoracic Society, 2010 [51] Bronchiectasis Patients with basal emphysema. Normativa SEPAR, 2008 [52] Bronchiectasis Patients with emphysema and /or liver disease Page 7 of 10 Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 AAT deficiency as the indication for augmentation therapy Fig. 2 The protease/anti-protease balance in healthy subjects (a), AATD patients (b) and bronchiectasis patients (c). Definitions. AAT = alpha-1 antitrypsin deficiency; NE = neutrophil elastase; LLN = lower limit of normality Intravenous infusion of alpha1-proteinase inhibitor, prepared from pooled human plasma of normal donors, has the bio- chemical goal to raise and maintain serum AAT levels above the protective threshold and clinical goals to slow the pro- gression of emphysema, reduce number of exacerbations and enhance the duration and quality of life. [56, 57]. Up to now augmentation therapy has been approved only for commercial use in adults with severe AAT deficiency-related pulmonary emphysema and a number of AAT preparations are currently available in various markets [56]. Evidence that augmentation therapy brings benefits (eg, slowed rate of FEV1, decline and decreased mortality) is stronger for individuals with moderate obstructive pulmonary impairment (eg, FEV1 35– 60% predicted) than for those with severe airflow obstruction [57]. Implications for future research In consideration of the bundle of evidence showing a causative role of AATD in bronchiectasis, a deeper evaluation of clinical, radiological, microbiological and functional variables should be applied in order to in- vestigate different phenotypes in AATD patients. The identification of a distinct bronchiectasis phenotype might be useful for a better understanding of AATD pathophysiology, as well as to improve personalized approach towards sub-groups of patients that might benefit from a focused treatment. A new generation of RCTs is, therefore, needed to test efficacy of intra- venous AAT augmentation therapy in patients with combined bronchiectasis and severe AATD. At the same time, new end-points for monitoring progres- sion of lung disease and response to therapy in this population ask to be explored. Multidisciplinary Respiratory Medicine (2018) 13:39 Page 8 of 10 Table 3 A list of last decade clinical trials testing AAT safety and efficacy in conditions different from severe deficiency AAT di t li i lt i l l t 15 j l 2018 Conclusions During the last years AATD has been interpreted as a diseases and extra-respiratory conditions. The final clinical picture would, therefore, result of a complex and mostly unknown interaction between genetic pattern and envir- onmental factors. Future research is gradually pointing The final clinical picture would, therefore, result of a complex and mostly Table 3 A list of last decade clinical trials testing AAT safety and efficacy in conditions different from severe deficiency AAT according to clinicaltrials.gov, last access 15 july 2018 Study Title Condition Phase Start Date Status A proof-of-concept pilot trial of AAT for pre-emption of steroid refractory acute GVDH GVDH 2 2018 Recruiting Multicenter trial of the effect of AAT on Islet Transplant engraftment and durability after Renal Transplant Diabetes Mellitus, Type 1 2 2017 Recruiting Anti-inflammatory therapy to improve outcomes in patients with Chronic Pancreatitis undergoing Total Pancreatectomy Islet Autotransplantation Chronic Pancreatitis; Diabetes Mellitus, Type 1 4 2016 Recruiting AAT enhances islet autograft survival Chronic Pancreatitis 1–2 2016 Recruiting A phase 2/3 clinical study to evaluate the safety and efficacy of AAT as an add-on biopharmacotherapy to conventional steroid treatment in subjects with acute GVDH with lower gastrointestinal involvement GVDH 2–3 2016 Terminated A proof-of-concept study evaluating the safety and efficacy of human AAT treatment in first Lung Transplantation Transplantation, Lung Rejection 2 2015 Active, not recruiting Improving single donor success rate in clinical islet using AAT Diabetes Mellitus, Type 1 1–2 2015 Active, not recruiting A multicenter, randomized, placebo-controlled study to evaluate the safety and efficacy of a human plasma-derived AAT in subjects with new-onset Type 1 Diabetes Mellitus Diabetes Mellitus, Type 1 2 2014 Terminated Single dose administration of AAT for the amelioration of organ injury and post-operative bleeding in patients undergoing cardiac surgery with cariopulmonary bypass Post Cardiac Surgery Systemic Inflammatory Response 1–2 2014 Unknown Pilot study of AAT: a novel treatment to mitigate Neuromyelitis Optica attacks Neuromyelitis Optica 1 2014 Unknown Long-term treatment safety, tolerability and efficacy of AAT in Type 1 Diabetes Diabetes Mellitus, Type 1 2 2013 Completed AAT to quench the acute inflammatory response in ST-segment elevation Acute Myocardial Infarction Acute Myocardial Infarction 1–2 2013 Completed Phase 2 study to evaluate the efficacy and safety of human AAT in the treatment of new onset Type 1 Diabetes Diabetes Mellitus, Type 1 2 2013 Completed Effects of Prolastin aerosol therapy on bacterial density in the airways of patients with CF Cystic Fibrosis 2 2013 Terminated AAT in treating patients with acute GVDH GVDH 1–2 2012 Completed Safety and tolerability of inhaled AAT once-a-day in patients with Cystic Fibrosis Cystic Fibrosis 2 2012 Completed Safety and efficacy of AAT in HIV Disease HIV Disease 2–3 2012 Terminated A pilot study of AAT in steroid refractory acute GVDH GVDH 2 2012 Active, not recruiting The effects of AAT on the progression of Type-1 Diabetes in subjects with detectable C-peptide Diabetes Mellitus, Type 1 1 2011 Completed Proof-of-concept study of the safety, tolerability and efficacy of intravenous AAT in Type 1 Diabetes Diabetes Mellitus, Type 1 1–2 2011 Completed Safety and efficacy of inhaled AAT in preventing Bronchiolitis Obliterable Syndrome in Lung Transplant recipients Transplantation, Lung Rejection 2 2011 Unknown Safety and tolerability study of AAT inhalation solution in patients with Cystic Fibrosis Cystic Fibrosis 1 2011 Completed Effects of intravenous AAT on preserving beta-cell function in new-onset Type 1 Diabetes Mellitus Diabetes Mellitus, Type 1 2 2010 Withdrawn AAT alpha-1 antitrypsin, CF cystic fibrosis GVDH graft-versus-host disease, HIV human immunodeficiency virus AAT alpha-1 antitrypsin, CF cystic fibrosis GVDH graft-versus-host disease, HIV human immunodeficiency virus diseases and extra-respiratory conditions. Fig. 2 The protease/anti-protease balance in healthy subjects (a), AATD patients (b) and bronchiectasis patients (c). Definitions. AAT = alpha-1 antitrypsin deficiency; NE = neutrophil elastase; LLN = lower limit of normality Fig. 2 The protease/anti-protease balance in healthy subjects (a), AATD patients (b) and bronchiectasis patients (c). Definitions. AAT = alpha-1 antitrypsin deficiency; NE = neutrophil elastase; LLN = lower limit of normality approach to augmentation therapy based, not only on the genotype of patient, but also on a careful evaluation of ex- pected benefits, risks and sanitary costs. The use of AAT treatment in patients other-than-PI*ZZ dates back decades and has been already explored in cystic fibrosis (CF), a genetic sys- temic disease characterized by massive neutrophilic in- flammation on the lung epithelial surface. In 2007 Griese and colleagues performed a prospective uncon- trolled study of 4-week inhaled AAT in a cohort of 52 patients with CF. The primary outcome of this study was a reduction in NE activity, along with change in neutrophils and pro-inflammatory cytokines on spu- tum. Although results on pulmonary function were controversial, the authors found a significant decrease in all inflammatory markers measured during treatment [59]. Similarly, two other studies showed reduction in elastase activity in CF subjects following inhaled treat- ment of AAT [60, 61]. However, a correlation with clin- ical outcomes would require long-term multicentre studies. In addition, decisions for starting the chronic augmenta- tion therapy need to be made at this point on a risk/bene- fit basis, as the lung destruction in emphysema is irreversible and a new line of translational research in AATD might focus on the development of personalized therapeutic regimens taking into account the patient clin- ical profile and needs [58]. In this light, investigating a new role for augmentation therapy in patients with het- erozygosity (PI*MZ, PI*MS) is urgently needed, especially in the presence of other chronic respiratory conditions as asthma or bronchiectasis. The rationale is that the anti-protease activity provided by AAT, although present in normal amounts, is finally compromised in patients with chronic inflammatory disease, such as bronchiectasis, due to the overwhelming NE inflammatory burden [2]. The condition would result in functional deficiency of AAT (Fig. 2). This supports the idea of an individualized Gramegna et al. Competing interests 11. Silverman EK, Sandhaus RA. Clinical practice. Alpha1-antitrypsin deficiency. N Engl J Med. 2009;360(26):2749–57. AG reports personal fees from Boehringer Ingelheim, Chiesi Farmaceutici, Vertex Spa. SA reports grant support from AstraZeneca, Bayer Healthcare, Grifols, Pfizer Ltd. and GlaxoSmithKline. AC has received payments for board membership, income for lectures, or support for research in respiratory disease from the following organizations: Astra Zeneca SPA, Boehringer Ingelheim Italia SPA, GlaxoSmithKline S.p.A., Grifols Italia S.P.A., Meda Pharma, Novartis Farma S.p.A., Stallergenes Italia S.r.l, CSL Behring SpA, A. Menarini Industrie Farmaceutiche Riunite Srl. LR reports personal fees from Medimmune, Biogen, Sanofi-Aventis, Takeda, ImmuneWorks, Shionogi, Cipla, Pliants Therapeutics, Boehringer Ingelheim, Roche. CV participated as a consultant, speaker and investigator for Actelion, Boehringer Ingelheim, Gilead, InterMune Inc. and Roche. FB reports research grants from Boehringer Ingelheim, Chiesi, Zambon, and Pfizer, congress lecture fees from Boehringer Ingelheim, Guidotti-Malesci, Menarini, GSK, Chiesi, Pfizer and Novartis, and consultancy fees from AstraZeneca, Menarini, Mundipharma, Novartis, GSK, Teva and Pfizer. FB is Editor-in-Chief of Multidisciplinary Respiratory Medicine, whereas SA, AC and LR are Associate Editors of Multidisciplinary Respiratory Medicine. 12. Gishen P, Saunders AJ, Tobin MJ, Hutchison DC. Alpha 1-antitrypsin deficiency: the radiological features of pulmonary emphysema in subjects of pi type Z and pi type SZ: a survey by the British thoracic association. Clin Radiol. 1982;33(4):371–7. 13. Demeo DL, Sandhaus RA, Barker AF, Brantly ML, Eden E, McElvaney NG, et al. Determinants of airflow obstruction in severe alpha-1-antitrypsin deficiency. Thorax. 2007;62(9):806–13. 14. Calabrò A, Torricelli E, Rosi E, et al. Idiopathic pulmonary fibrosis associated with ALpha-1 antitrypsin deficiency: concomitant finding or real association? SM J Case Rep. 2017;3(8):1075. 15. Cottin V, Nunes H, Brillet PY, Delaval P, Devouassoux G, Tillie-Leblond I, et al. Combined pulmonary fibrosis and emphysema: a distinct underrecognised entity. Eur Respir J. 2005;4:586–93. 16. Siri D, Farah H, Hogarth DK. Distinguishing alpha1-antitrypsin deficiency from asthma. Ann Allergy Asthma Immunol. 2013;111(6):458–64. 17. Makino S, Chosy L, Valdivia E, Reed CE. Emphysema with hereditary alpha-1 antitrypsin deficiency masquerading as asthma. J Allergy. 1970l; 46(1):40–8. Consent for publication Not applicable. Consent for publication Not applicable. 10. Miravitlles M, Dirksen A, Ferrarotti I, Koblizek V, Lange P, Mahadeva R, et al. European Respiratory Society statement: diagnosis and treatment of pulmonary disease in α1-antitrypsin deficiency. Eur Respir J. 2017;50:1700610. Publisher’s Note 18. Black LF, Hyatt RE, Stubbs SE. Mechanism of expiratory airflow limitation in chronic obstructive pulmonary disease associated with 1 -antitrypsin deficiency. Am Rev Respir Dis. 1972;105(6):891–9. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 19. Tobin MJ, Cook PJL, Hutchison DCS. Alpha1- antitrypsin deficiency: the clinical and physiological features of pulmonary emphysema in subjects homozygous for pi type Z. Br J Dis Chest. 1983;77:14–27 34. Abbreviations AATD: alpha-1 antitrypsin deficiency; COPD: Chronic Obstructive Pulmonary Disease; CPFE: combined pulmonary fibrosis and emphysema; IRIDE: Italian Registry on Adult Bronchiectasis; NE: neutrophil elastase; NTM: Nontuberculous mycobacteria 7. Dirksen A, Piitulainen E, Parr DG, Deng C, Wencker M, Shaker SB, et al. Exploring the role of CT densitometry: a randomised study of augmentation therapy in alpha-1 antitrypsin deficiency. Eur Respir J. 2009;33:1345–53. Author details 1 1Department of Pathophysiology and Transplantation, University of Milano; Internal Medicine Department, Respiratory Unit and Adult Cystic Fibrosis Center, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy. 2Department of Medical Sciences, Respiratory Diseases Unit, University Hospital of Cattinara, Trieste, Italy. 3Department of Internal Medicine and therapeutics, Division of Respiratory Diseases, IRCCS Policlinico San Matteo Foundation, University of Pavia, Pavia, Italy. 4Università Cattolica del Sacro Cuore, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy. 5Regional Referral Centre for Rare Lung Diseases, Department of Clinical and Experimental Medicine, University of Catania, University–Hospital Policlinico “G. Rodolico”, Catania, Italy. 20. Janus ED, Phillips NT, Carrell RW. Smoking lung function, and alpha-i- antitrypsin deficiency. Lancet. 1985;1:152–4. 20. Janus ED, Phillips NT, Carrell RW. Smoking lung antitrypsin deficiency. Lancet. 1985;1:152–4. 21. Larsson C. Natural history and life expectancy in severe alpha,-antitrypsin deficiency. PiZ Acta Med Scand. 1978;204:345–51. 22. Eriksson S. Studies in alpha-i-antitrypsin deficiency. Acta Med Scand. 1965;177:1–85. 23. McElvaney NG, Stoller JK, Buist AS, Prakash UB, Brantly ML, Schluchter MD, et al. Baseline characteristics of enrollees in the National Heart, lung and blood institute registry of alpha 1-antitrypsin deficiency. Alpha 1-antitrypsin deficiency registry study group. Chest. 1997;111(2):394–403. 24. [No authors listed]. Survival and FEV1 decline in individuals with severe deficiency of alpha1-antitrypsin. The Alpha-1-Antitrypsin Deficiency Registry Study Group. Am J Respir Crit Care Med. 1998;158(1):49–59. Received: 30 July 2018 Accepted: 25 September 2018 25. [No authors listed]. Alpha 1-antitrypsin deficiency: memorandum from a WHO meeting B ll World Health Organ 1997 75(5) 397 415 25. [No authors listed]. Alpha 1-antitrypsin deficiency: memorandu 25. [No authors listed]. Alpha 1-antitrypsin deficiency: memorandum from WHO meeting. Bull World Health Organ 1997;75(5):397–415. WHO meeting. Bull World Health Organ 1997;75(5):397–415. Conclusions During the last years AATD has been interpreted as a common mechanism of inflammatory disequilibrium and tissue damage among both chronic respiratory Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 Page 9 of 10 Page 9 of 10 toward this new paradigm by expanding the evidence of the role of AAT augmentation treatment as a potent im- munomodulatory and anti-inflammatory drug for condi- tions different from pulmonary emphysema (Table 3). 4. Seersholm N, Dirksen A, Kok-Jensen A. Airways obstruction and two-year survival in patients with severe alpha1 antitrypsin deficiency. Eur Respir J. 1994;7:1985–7. 5. Wu MC, Eriksson S. Lung function, smoking and survival in severe alpha-1 antitrypsin deficiency, pi*ZZ. J Clin Epidemiol. 1988;41:1157–65. 5. Wu MC, Eriksson S. Lung function, smoking and survival in severe alpha-1 antitrypsin deficiency, pi*ZZ. J Clin Epidemiol. 1988;41:1157–65. 6. Evald T, Dirksen A, Keittelmann S, , Viskum K, Kok-Jensen A. Decline in pulmonary function in patients with alpha 1-antitrypsin deficiency. Lung 1990;168 Suppl:579–585. Authors’ contributions C d d 8. Stockley R, Parr DG, Piitulainen E, Stolk J, Stoel BC, Dirksen A, et al. Therapeutic efficacy of alpha-1 antitrypsin augmentation therapy on the loss of lung tissue: an integrated analysis of 2 randomised clinical trials using computed tomography densitometry. Respir Res. 2010;11(136). Conception and design: AG, SA, FB. Drafting the manuscript: AG, SA. All authors participated in writing and revising the article prior to submission. All authors read and approved the final manuscript. 9. Chapman KR, Burdon JG, Piitulainen E, Sandhaus RA, Seersholm N, Stocks JM, et al. Intravenous augmentation treatment and lung density in severe α1 antitrypsin deficiency (RAPID): a randomised, double-blind, placebo- controlled trial. Lancet. 2015;386:360–8. Ethics approval and consent to participate Not applicable. Ethics approval and consent to participate Not applicable. References Long-term experience in the treatment of α1-antitrypsin deficiency: 25 years of augmentation therapy. Eur Respir Rev. 2015;24(135): 46–51. 33. Parr DG, Guest PG, Reynolds JH, Dowson LJ, Stockley RA. Prevalence and impact of bronchiectasis in alpha-1 antitrypsin deficiency. Am J Respir Crit Care Med. 2007;176:1215–21. 58. Stockley RA, Miravitlles M, Vogelmeier C, Alpha One International Registry (A.I.R.). Augmentation therapy for alpha-1 antitrypsin deficiency: towards a personalised approach. Orphanet Journal of Rare Disease. 2013;8:149. 34. Greulich T, Nell C, Herr C, Vogelmeier C, Kotke V, Wiedmann S, et al. Results from a large targeted screening program for alpha-1-antitrypsin deficiency: 2003–2015. Orphanet J Rare Dis. 2016;11(1):75. 59. Griese M, Latzin P, Kappler M, Weckerle K, Heinzlmaier T, Bernhardt T, et al. O1-antitrypsin inhalation reduces airway inflammation in cystic fibrosis patients. Eur Respir J. 2007;29:240–50. 35. Strange C. Airway disease in alpha-1 antitrypsin deficiency. COPD. 2013; 10(Suppl 1):68–73. 60. McElvaney N, Hubbard R, Birrer P, Chernick MS, Caplan DB, Frank MM, et al. Aerosol alpha-1 antitrypsin treatment for cystic fibrosis. Lancet. 1991;337: 392–4. 36. Aggarwal N, Delgado BM, Salipalli S, Matamala N, Rademacher J, Schwerk N,, et al. Why do some adults with PiMZ α1-antitrypsin develop bronchiectasis? [corrected]. ERJ Open Res. 2016;2(2). 61. Martin SL, Downey D, Bilton D, Keogan MT, Edgar J, Elborn JS, et al. Safety and efficacy of recombinant alpha(1)-antitrypsin therapy in cystic fibrosis. Pediatric Pulmonol. 2006;41:177–83. 37. Lonni S, Chalmers JD, Goeminne PC, McDonnell MJ, Dimakou K, De Soyza A, et al. Etiology of non-cystic fibrosis bronchiectasis in adults and its correlation to disease severity. Ann Am Thorac Soc. 2015;12(12):1764–70. 62. Chang AB, Bell SC, Torzillo PJ, King PT, Maguire GP, Byrnes CA, et al. Chronic suppurative lung disease and bronchiectasis in children and adults in Australia and new Zealand Thoracic Society of Australia and new Zealand guidelines. Med J Aust. 2015;202(3):130. 38. Gao YH, Guan WJ, Liu SX, Wang L, Cui JJ, Chen RC, et al. Aetiology of bronchiectasis in adults: a systematic literature review. Respirology. 2016; 21(8):1376–83. 39. Chalmers JD, Aliberti S, Polverino E, Vendrell M, Crichton M, Loebinger M, et al. The EMBARC European bronchiectasis registry: protocol for an international observational study. ERJ Open Research. 2016;2(1):00081–2015. y 40. Rademacher J, de Roux A, Ringshausen FC. PROGNOSIS - the PROspective German NOn-CF BronchiectaSIS patient registry. Pneumologie. 2015;69(7):391–3. 41. Aliberti S, Hill AT, Mantero M, Battaglia S, Centanni S, Lo Cicero S, et al. References 26. Kelbel T, Morris D, Walker D, Henao MP, Craig T. The Allergist’s role in detection of severe Alpha-1 antitrypsin deficiency. J Allergy Clin Immunol Pract. 2017;5(5):1302–6. 1. American Thoracic Society; European Respiratory Society. American Thoracic Society/European Respiratory Society statement: standards for the diagnosis and management of individuals with alpha-1 antitrypsin deficiency. Am J Respir Crit Care Med. 2003;168(7):818–900. 27. Miravitlles M, Vilà S, Torrella M, Balcells E, Rodríguez-Frías F, de la Roza C, et al. Influence of deficient alpha1-anti-trypsin phenotypes on clinical characteristics and severity of asthma in adults. Respir Med. 2002;96(3):186–92. 27. Miravitlles M, Vilà S, Torrella M, Balcells E, Rodríguez-Frías F, de la Roza C, et al. Influence of deficient alpha1-anti-trypsin phenotypes on clinical characteristics and severity of asthma in adults. Respir Med. 2002;96(3):186–92. 28. Eden E. Asthma and COPD in alpha-1 antitrypsin deficiency. Evidence for the Dutch hypothesis COPD. 2010;7(5):366–74. 28. Eden E. Asthma and COPD in alpha-1 antitrypsin deficiency. Evidence for the Dutch hypothesis COPD. 2010;7(5):366–74. 3. Stoller JK, Aboussouan LS. A review of α1-antitrypsin deficiency. Am J Respir Crit Care Med. 2012;185(3):246–59. Page 10 of 10 Page 10 of 10 Page 10 of 10 Gramegna et al. Multidisciplinary Respiratory Medicine (2018) 13:39 29. Cosio MG, Bazzan E, Rigobello C, Tinè M, Turato G, Baraldo S, et al. Alpha-1 antitrypsin deficiency: beyond the protease/Antiprotease paradigm. Ann Am Thorac Soc. 2016;13(Suppl 4):S305–10. 53. Pulmonology Portuguese Society Bronchiectasis Study Group. Recommendations for aetiological diagnosis of bronchiectasis. Rev Port Pneumol (2006). 2016;22(4):222–35. 54. Polverino E, Goeminne PC, McDonnell MJ, Aliberti S, Marshall SE, Loebinger MR, et al. European Respiratory Society guidelines for the management of adult bronchiectasis. Eur Respir J. 2017;50:1700629. 30. Dowson LJ, Guest PJ, Stockley RA. Longitudinal changes in physiological, radiological, and health status measurements in alpha(1)-antitrypsin deficiency and factors associated with decline. Am J Respir Crit Care Med. 2001;164(10 Pt 1):1805–9. 55. Blanco I, Bueno P, Diego I, Pérez-Holanda S, Casas-Maldonado F, Esquinas C, et al. Alpha-1 antitrypsin pi*Z gene frequency and pi*ZZ genotype numbers worldwide: an update. Int J Chron Obstruct Pulmon Dis. 2017;12:561–9. 31. Fallat RJ. Reactive airways disease and alpha-1 antitrypsin deficiency. In: Crystal RG, editor. Alpha 1-antitrypsin deficiency: biology, pathogenesis, clinical manifestations, therapy. New York: Marcel Dekker; 1996. p. 259–79. 56. Hatipoğlu U, Stoller JK. α1-antitrypsin deficiency. Clin Chest Med. 2016;37(3): 487–504. 32. Stockley RA. The multiple facets of alpha-1-antitrypsin. Ann Transl Med. 2015;3(10):130. 57. Teschler H. References SIP bronchiectasis audit working group. Quality standards for the management of bronchiectasis in Italy: a national audit. Eur Respir J. 2016;48(1):244–8. 42. Blasi F, Chalmers J, Aliberti S. COPD and bronchiectasis: phenotype, Endotype or co-morbidity? COPD. 2014;6:603–4. 43. Faverio P, Stainer A, Bonaiti G, Zucchetti SC, Simonetta E, Lapadula G, et al. Characterizing Non-Tuberculous Mycobacteria Infection in Bronchiectasis. Int J Mol Sci. 2016;17(11). 44. Mirsaeidi M, Farshidpour M, Ebrahimi G, Aliberti S, Falkinham JO 3rd. Management of nontuberculous mycobacterial infection in the elderly. Eur J Intern Med. 2014;25(4):356–63. 45. Bonaiti G, Pesci A, Marruchella A, Lapadula G, Gori A, Aliberti S. Nontuberculous mycobacteria in noncystic fibrosis bronchiectasis. Biomed Res Int. 2015;2015:197950. 45. Bonaiti G, Pesci A, Marruchella A, Lapadula G, Gori A, Aliberti S. Nontuberculous mycobacteria in noncystic fibrosis bronchiectasis. Biomed Res Int. 2015;2015:197950. 46. Chan ED, Kaminska AM, Gill W, Chmura K, Feldman NE, Bai X, et al. Alpha-1- antitrypsin (AAT) anomalies are associated with lung disease due to rapidly growing mycobacteria and AAT inhibits Mycobacterium abscessus infection of macrophages. Scand J Infect Dis. 2007;39(8):690–6. 47. Carroll TP, O’Connor CA, Floyd O, McPartlin J, Kelleher DP, O'Brien G, et al. The prevalence of alpha-1 antitrypsin deficiency in Ireland. Respir Res. 2011; 12:91. 48. Kaczor MP, Sanak M, Libura-Twardowska M, Szczeklik A. The prevalence of alpha1-antitrypsin deficiency in a representative population sample from Poland. Respir Med. 2007;101:2520–5. 49. Global strategy for prevention, diagnosis and management of COPD, 2018 report; https://goldcopd.org/wp-content/uploads/2018/02/WMS-GOLD- 2018-Feb-Final-to-print-v2.pdf 49. Global strategy for prevention, diagnosis and management of COPD, 2018 report; https://goldcopd.org/wp-content/uploads/2018/02/WMS-GOLD- 2018-Feb-Final-to-print-v2.pdf 50. Global strategy for asthma management and prevention, 2018 update; https://ginasthma.org/2018-gina-report-global-strategy-for-asthma- management-and-prevention/ 51. Pasteur MC, Bilton D. Hill AT on behalf of the British Thoracic Society bronchiectasis (non-CF) guideline group British Thoracic Society guideline for non-CF bronchiectasis. Thorax. 2010;65:i1–i58. 51. Pasteur MC, Bilton D. Hill AT on behalf of the British Thoracic Society bronchiectasis (non-CF) guideline group British Thoracic Society guideline for non-CF bronchiectasis. Thorax. 2010;65:i1–i58. 52. Vendrell M, de Gracia J, Olveira C, Martínez MA, Girón R, Máiz L, et al. Diagnosis and treatment of bronchiectasis. Spanish Society of Pneumology and Thoracic Surgery. Arch Bronconeumol. 2008;44(11):629–40.
https://openalex.org/W2982586857
https://edoc.unibas.ch/72746/1/20191114140552_5dcd513092d2b.pdf
English
null
Mass campaigns combining antimalarial drugs and anti-infective vaccines as seasonal interventions for malaria control, elimination and prevention of resurgence: a modelling study
BMC infectious diseases
2,019
cc-by
14,131
Camponovo et al. BMC Infectious Diseases (2019) 19:920 https://doi.org/10.1186/s12879-019-4467-4 Camponovo et al. BMC Infectious Diseases (2019) 19:920 https://doi.org/10.1186/s12879-019-4467-4 (2019) 19:920 Camponovo et al. BMC Infectious Diseases https://doi.org/10.1186/s12879-019-4467-4 Open Access Mass campaigns combining antimalarial drugs and anti-infective vaccines as seasonal interventions for malaria control, elimination and prevention of resurgence: a modelling study Flavia Camponovo1,2, Chris F. Ockenhouse3, Cynthia Lee3 and Melissa A. Penny1,2* © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. * Correspondence: melissa.penny@unibas.ch 1Swiss Tropical and Public Health Institute, Basel, Switzerland 2University of Basel, Basel, Switzerland Full list of author information is available at the end of the article Abstract Background: The only licensed malaria vaccine, RTS,S/AS01, has been developed for morbidity-control in young children. The potential impact on transmission of deploying such anti-infective vaccines to wider age ranges, possibly with co-administration of antimalarial treatment, is unknown. Combinations of existing malaria interventions is becoming increasingly important as evidence mounts that progress on reducing malaria incidence is stalling and threatened by resistance. Methods: Malaria transmission and intervention dynamics were simulated using OpenMalaria, an individual-based simulation model of malaria transmission, by considering a seasonal transmission setting and by varying epidemiological and setting parameters such as transmission intensity, case management, intervention types and intervention coverages. Chemopreventive drugs and anti-infective vaccine efficacy profiles were based on previous studies in which model parameters were fitted to clinical trial data. These intervention properties were used to evaluate the potential of seasonal mass applications of preventative anti-infective malaria vaccines, alone or in combination with chemoprevention, to reduce malaria transmission, prevent resurgence, and/or reach transmission interruption. Results: Deploying a vaccine to all ages on its own is a less effective intervention strategy compared to chemoprevention alone. However, vaccines combined with drugs are likely to achieve dramatic prevalence reductions and in few settings, transmission interruption. The combined mass intervention will result in lower prevalence following the intervention compared to chemoprevention alone and will increase chances of interruption of transmission resulting from a synergistic effect between both interventions. The combination of vaccine and drug increases the time before transmission resurges after mass interventions cease compared to mass treatment alone. Deploying vaccines and drugs together requires fewer rounds of mass intervention and fewer years of intervention to achieve the same public health impact as chemoprevention alone. Conclusions: Through simulations we identified a previously unidentified value of deploying vaccines with drugs, namely the greatest benefit will be in preventing and delaying transmission resurgence for longer periods than with other human targeted interventions. This is suggesting a potential role for deploying vaccines alongside drugs in transmission foci as part of surveillance-response strategies. Keywords: Vaccine, Mass intervention, Resurgence, Seasonal transmission, Malaria, Simulation * Correspondence: melissa.penny@unibas.ch 1Swiss Tropical and Public Health Institute, Basel, Switzerland 2University of Basel, Basel, Switzerland Full list of author information is available at the end of the article * Correspondence: melissa.penny@unibas.ch 1Swiss Tropical and Public Health Institute, Basel, Switzerland 2University of Basel, Basel, Switzerland Full list of author information is available at the end of the article Background protection over time, routine immunization of children with RTS,S is likely to have a positive public health im- pact [13]. Given that duration of protection from RTS,S immunization is shorter than expected, but longer than chemoprevention, alternative uses of the vaccine as a yearly seasonal intervention targeting children is cur- rently being investigated and compared to seasonal mal- aria chemoprevention (SMC) in clinical studies in Burkina Faso and Mali [14]. Seasonal transmission of malaria due to rainfall patterns is common in the sub- Sahelian region [15], and novel seasonal preventive tools could be of great value in many endemic countries. g In 2015 WHO set new goals of reducing malaria inci- dence by 90% and eliminating malaria in 35 countries by 2030 [1]. This followed dramatic reductions in malaria prevalence from 2000 to 2015 [1] due to scale-up of vec- tor control and artemisinin combination therapy [2]. However, recent estimates indicate that the improve- ments have stalled, with increasing incidence in several countries [3], and current interventions and implemen- tation strategies may not be sufficient. New tools such as novel insecticides, drugs, and vaccines, are likely to play a role in interruption of malaria transmission [4] and continue to be developed. As we wait for novel tools to be tested and approved, it is important to use existing tools as effectively as possible to help facilitate the efforts of 20 countries in achieving their goal of malaria elimin- ation in the next 7 years as documented by WHO [1]. Increased funding and greater access to interventions, such as effective case management and vector control tools, are essential; it is also increasingly important to innovate the deployment of combinations of existing interventions. To reach and maintain pre-elimination prevalence levels, the entire population of all ages needs to be pro- tected from infection, rather than children only. Mass Drug Administration (MDA) can accelerate prevalence reduction by clearing infections over a short time period, but the chemo-preventative effect is relatively short- lived. MDA is also unlikely to interrupt malaria parasite transmission, since prevalence reduction is transient and likely to be followed by malaria resurgence [16–18]. Background MDA is currently implemented in Zambia alongside re- active case detection [19], and being considered in the Strategy for Elimination in the Greater Mekong Sub- region (2015–2030) to combat drug resistance [20], and to address migrant and mobile populations such as for- est workers in regions of Cambodia [21]. There is just one malaria vaccine that has successfully completed Phase 3 clinical studies, RTS,S/AS01, a pre- erythrocytic or anti-infective vaccine (AIV), targeting sporozoite stages of the Plasmodium falciparum life cycle. RTS,S/AS01 received a positive opinion from the European Medicines Agency in 2015 [5], and pilot im- plementation via routine childhood immunization in re- gions of Ghana, Kenya and Malawi have started earlier this year [6]. From extensive analysis of Phase 3 trials the vaccine was estimated to have a high initial efficacy, with protection waning over time [7, 8]. Recently pub- lished data showed that an altered regimen of RTS,S in which the third dose is delayed and given at a fraction of the original dosage (delayed, fractional dose RTS,S) re- sults in a higher efficacy in human challenge trial [9] and thus may result in a longer duration of protection, although this still needs to be confirmed in the field. There are many vaccines in development targeting sporozoite, liver, blood, or mosquito stages of the mal- aria parasite, including AIVs such as the delayed frac- tional dose RTS,S, the whole Plasmodium falciparum sporozoite based PfSPZ Vaccine [10, 11], and R21 [12]. A vaccine which can confer many years of protection is desirable but in addition to technical and regulatory challenges there may be biological hurdles that cannot be easily overcome. An alternative to SMC or MDA with drugs alone is to combine them with vaccines in seasonal settings. To date little has been done to estimate the impact of add- ing RTS,S or a similar AIV to existing mass treatment strategies. Previous modelling predicted that a long dur- ation of protection is critical for sustained reduction of malaria transmission via mass vaccination [22], so com- bination with other interventions is expected to be crit- ical for effective mass vaccination with the existing or future AIVs which have limited duration of protection. There has been no clinical trial combining MDA with an AIV, although drug and vaccine efficacy profiles are known from field trials. Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 2 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 2 of 15 Model and setting details Malaria transmission and intervention dynamics were simulated using OpenMalaria [23], an individual-based simulation model of malaria transmission detailed in [24] and [25]. Briefly, OpenMalaria combines an individual-based model of malaria in humans [24] with a deterministic model of malaria in mosquitoes [26]. The simulation model includes sub-models of infection of humans [27], blood-stage parasite densities [28], infec- tiousness to mosquitoes [29], incidence of morbidity and mortality [30, 31]; and immunity is separated in pre- erythrocytic and blood-stage immunity. y y g y A range of transmission levels were investigated by varying the initial entomological inoculation rate (EIR) from 0.5 to 5 which approximates PfPR2–10 of 1–20%, depending on the underlying case management (Add- itional file 1: Figure S10). This is intended to correspond to the residual EIR conditional on vector control mea- sures already in place, which are assumed to remain constant throughout the duration of the simulations. The yearly transmission pattern was chosen as an arche- typal highly seasonal setting with a transmission season of 3 months (roughly equating to Senegal where SMC is ongoing [14]). Routine case-management levels were modelled as the 14 day effective treatment coverage (E14) with a range of 15–80% (typifying both current levels of case management in the vast majority of African coun- tries [32], as well as optimistic scale up of health system improvements). If not otherwise specified, results in the main analysis correspond to E14 = 45%. Three levels of imported infections were investigated when analyzing re- surgence (0, 2 and 20 infections per 1000 population). A population of 10,000 individuals was assumed, and 10 stochastic realizations of the model were simulated for each strategy and setting combination (total of 847,000 simulations). Initial parasite prevalence levels in 2–10 year olds (PfPR2–10), levels of effective access to care (E14), and timing and coverage levels of the interventions are summarized in Additional file 1: Tables S1 and S2. ε ¼ ε0 exp −log2 ð Þ 1 kt Lk ! ; where ε0 is the initial value of efficacy against infec- tion, L is the half-life of decay (time till efficacy against infection is half of the initial value), and parameter k in- forms the shape of the decay (k = 1 is exponential). Bayesian MCMC was used to compare simulated inci- dence and Phase 3 trial incidence in order to find the most appropriate vaccine properties in OpenMalaria de- fined as efficacy against infection. Background Investigating novel applications of existing interventions requires understanding the likely impact over a large number of possible strategies and delivery paths, and it is not feasible to test all com- binations in the field. Modelling and simulation lend themselves to providing evidence before moving to dem- onstration field studies [16]. In this study, we used an existing individual based model of malaria dynamics to understand the role, and potential public health benefits, of an AIV delivered through mass administration alone or in combination with drugs, in a seasonal setting. We estimated the benefits of mass vaccination with respect to elimination, transmission reduction, resurgence, and prevention, for a range of prevalence and health system settings. These modeling outcomes were compared to Clinical development of RTS,S focused on reducing morbidity and mortality, especially in young children, the rationale being to add additional protection to vector control for the vulnerable population experiencing the highest burden of disease. Modeling studies informed by clinical trial results estimated that, despite waning Page 3 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 3 of 15 Fig. 1. The underlying parameterization of the vaccine was informed by previous model fitting to RTS,S/AS01 Phase 3 trial results [13], and should be considered to encompass vaccine parameterizations of other AIV in development. The clinical trials reported an observed efficacy against clinical cases in children receiving four doses (5–17 months of age at first dose) of 43.9% (95% CI 39.7–47.8%) over 32 months of follow-up, and with a decline over time [33], which was the basis for fitting the vaccine efficacy for our microsimulations [13]. The action of an AIV in our microsimulation is to prevent infections, which has a con- sequential effect on preventing clinical disease. The under- lying vaccine efficacy against infection in time (including likely decay of effect), defined as the proportionate reduc- tion in force of infection assuming beta distributed vari- ation in efficacy, was parameterized via fitting of our microsimulation to extensive Phase 3 clinical trial data of RTS,S/AS01 and has been described elsewhere [7]. Briefly, AIVs are modelled in Openmalaria as preventing new in- fections via stochastic process with the probability of pre- venting an infection referred to as the efficacy against infection (also the proportion of blood-stage infections averted). Background The time course of efficacy was described as a weibull function capturing possible exponential, biphasic and step-like decays. This decay function is described via the following equation: those produced by modeling use of MDA alone, to esti- mate the potential for vaccination in addition to treat- ment to accelerate prevalence reduction or interruption of transmission, and to estimate the effect of mass vac- cination on resurgence rates. In order to consider likely next generation AIVs, we considered two AIVs with dif- ferent protection profiles, a vaccine with equivalent pro- tection profile to RTS,S/AS01 and a vaccine with a longer duration of protection compared to RTS,S/AS01. those produced by modeling use of MDA alone, to esti- mate the potential for vaccination in addition to treat- ment to accelerate prevalence reduction or interruption of transmission, and to estimate the effect of mass vac- cination on resurgence rates. In order to consider likely next generation AIVs, we considered two AIVs with dif- ferent protection profiles, a vaccine with equivalent pro- tection profile to RTS,S/AS01 and a vaccine with a longer duration of protection compared to RTS,S/AS01. Model and setting details Models were simultan- eously fitted to both control and vaccinated incidence from the trial, and estimates of vaccine properties, site- specific access to care, and the extent of within-site vari- ability were obtained. This model fitting resulted in an estimated efficacy against infection to be high following immunization, 91% [8], and this efficacy wanes quickly over time result- ing in limited duration of protection with the half-life against infection, L, of less than 1 year (efficacy of 45.5% at 7.3 months, and k = 0.69 describing the shape of effi- cacy decay) [7, 8]. This half-life refers to time in which protection against infection has decayed to half the value of the initial level. We note decay of this protection is bi- phasic so that protection continues after this time and decays more slowly than the first 7 months (Fig. 1). This Fig. 1 Main simulated strategies and drug and vaccine efficacy profiles. a Illustrated strategies: MDA alone, vaccine alone, or MDA with vaccine. MDA application alone is 3 rounds coinciding with the pattern of seasonal transmission, with 2–3 years of 3 rounds or 3 rounds for only the first year followed by 1–2 years of 1 round at the beginning of the transmission season; RTS,S-like-duration vaccine or longer duration vaccine application alone is 3 rounds coinciding with the pattern of seasonal transmission with 1–2 years of 1 dose at the beginning of the season or as 3 rounds before the pattern of seasonal transmission, with 1–2 years of 1 dose at the beginning of the season; and strategies combining MDA with RTS,S-like-duration vaccine or longer duration vaccine are a combination of all MDA and vaccine implementations combined together. b Efficacy against infection profiles of 2 years of intervention with MDA, RTS,S like vaccine, vaccine with longer duration of protection, and vaccine with lower initial efficacy. x-axis represents the time in months, and efficacy against infection is indicated on the y-axis for MDA given at 1 months interval during 3 months the first year and 1 (yellow) or 3 (orange) rounds the second year, and for mass vaccination with a RTS,S like vaccine (purple), a longer duration protection vaccine (pink), or vaccines with lower initial efficacy against infection (dashed lines). Mass vaccination is delivered either before the peak of transmission the first year (bottom plot) or during the transmission season (middle plot). The first year 3 doses of vaccination are administered, but only the 3rd dose of vaccination is modeled, and the second year only one dose is administered. Initial efficacy against infection for both vaccines are 91% or 50% and the half-life, which is the time in which protection against infection has decayed to half the value of the initial level, is 7.3 months for the RTS,S like vaccine and 18 months for the vaccine with extended duration of protection, indicated with arrows on the diagram. Period of peak transmission are indicated by black boxes (minimum 9 months old at 3rd dose) with 3 monthly spaced doses and yearly single dose vaccination follow- ing first year for 2 or 3 years; 2) deployment of MDA of up to 3 monthly rounds per year delivered to ages from 6 months old for 2 or 3 years; or 3) the combination of both interventions (Additional file 1: Table S2). MDA was assumed deployed at the beginning of the transmis- sion season, with an optimistic strategy of 2 additional rounds deployed at 1 month intervals during the trans- mission season. Either all 3 MDA rounds were imple- mented every year of the intervention up to 3 years, or only the first round was kept for the following year(s). The vaccine was delivered either prior (with the 3rd dose completed at the beginning of the season) or dur- ing the transmission season, the latter possibly favored for logistical reasons as a fewer number of intervention rounds in the combination strategies are required as- suming drug and vaccination are delivered at the same time. Vaccination coverage represents the proportion of population receiving all three doses, with no efficacy as- sumed if receiving less than 3 doses in the first year. The efficacy parametrization is in alignment with the esti- mates of other modelling groups [13]. In addition to this vaccine parametrization, we have investigated a vaccine with longer duration of protection with a half-life of 1.5 years, and in a subset of settings a vaccine with lower ef- ficacy, reducing initial efficacy against infection from 91 to 50% (Additional file 1: Table S1, and Fig. 1). The fourth dose of the vaccine, and the fifth dose in the 3 years deployment strategies, is assumed to have the same efficacy profile and coverage as the third dose. Strategies including MDA were implemented in the model with Dihydroartemisinin piperaquine (DHAp), which has slightly lower treatment failure rates than arthemeter-lumefantrine [34] and was the main mass drug used in southern Zambia field studies assessing the short term impact of MDA [35]. Intervention properties Intervention properties for both vaccination and MDA, are summarized in the Additional file 1: Table S1, and Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 4 of 15 Impact on time course of prevalence Use of a vaccine with equivalent protection profile to RTS,S/AS01 (with high initial efficacy, waning over time reaching half of initial efficacy after 7.3 months) fitted to the Phase 3 clinical trial data [33], or an AIV with longer duration of protection (same initial efficacy but with ex- tended duration of protection, reaching half of initial ef- ficacy after 18 months), or an AIV with lower initial efficacy against infection of 50%, was compared to MDA as single mass interventions alone or in combination, over 2 to 3 years of intervention, as schematized in Fig. 1a, with efficacy profiles illustrated in Fig. 1b. Example time courses of predicted all age prevalence before and after 2 years of mass intervention are illustrated for one simulation from each strategy and across all simulations as average yearly prevalence (Fig. 2, variability shown by minimum and maximum range) (additional strategies shown in Additional file 1: Figures S1–S2). The differ- ence between intervention strategies is most distinguish- able at the annual peak of prevalence. MDA and/or mass vaccination strategies are found to effect preva- lence in one of two ways, either the strategy successfully interrupts transmission, or the strategy leads to a rapid drop in prevalence followed by malaria resurgence back to initial prevalence levels. Relative prevalence reduction The impact of different intervention strategies was esti- mated by comparing the relative reduction in prevalence achieved each year, defined as the yearly prevalence re- duction measured for each intervention strategy relative to the prevalence in the control strategy (i.e. no inter- vention deployed). MDA alone was predicted to be more effective than mass vaccination alone in reducing preva- lence, and thus also avoiding resurgence, (Fig. 3a), in set- tings with different initial prevalence levels (EIR) and case management levels (E14) (Additional file 1: Figure Deployment strategies The strategies simulated were 1) deployment of mass vaccination of an initial 3 doses in the first year delivered to population older than 6 months of age at first dose Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 5 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 proportion of the population receiving a drug or vaccine (full 3 dose regimen) was randomly allocated for each round and intervention given coverage in the default set- ting, and in a subset of simulations, the vaccine and MDA, when delivered simultaneously, was administered to the same population for a given coverage. A range of coverages from 40 to 100% were considered. sigmoidal curve and was defined as Rt ¼ PcðtÞ−PsðtÞ PcðtÞ−Pmin ∝ ð1 þ xb λ50 Þ −1 , where Pmin is the minimum prevalence reached by strategy s, λ50 is the estimated half-life of re- surgence or time till 50% resurgence representing the years after maximum prevalence reduction was reached in which prevalence resurges by 50% of the reduction, and b is the Hill’s slope, representing the steepness of the logistic curve. Similarly, we estimated λ10 is the time till 10% resurgence representing the years after max- imum prevalence reduction was reached in which preva- lence resurges by 10% of the reduction. Resurgence parameters were estimated for strategies with 60% cover- age for 2 years of deployment. Outcome measured as interruption of transmission and estimated synergistic behavior of the combined intervention To estimate the chances of an intervention strategy to achieve elimination in our model we calculated the proportion of simulations achieving interruption of transmission over all stochastic realizations for each setting. The synergistic behavior of MDA combined with mass vaccination in chances to reach elimination was evaluated. The combined mass vaccination with MDA interventions were considered synergistic if the chance to interrupt transmission was greater than the sum of the separate interventions and was defined consistent with previous disease intervention analyses [36]. The level of synergy for differing coverages and transmission settings was calculated as σ ¼ ð Φcombined−Φcontrol ðΦMDA−ΦcontrolÞþðΦvaccine−ΦcontrolÞÞ−1, where ϕ represents the probability of interrupting transmission in a simu- lation set denoted in the subscript, namely with no intervention (control), MDA alone (MDA), mass vac- cination alone (vaccine), or the combined MDA and vaccination (combined). Values of σ greater than 0 represent synergism so that the estimated effect of the two combined interventions is greater than additive, values of 0 imply the combined interventions are not more than additive, and values less than 0 imply less than additive or maximum level was reached by one or both single interventions. Outcome measured as prevalence reduction We estimated the relative reduction in prevalence achieved using different intervention deployment as the prevalence reduction predicted to be reached each year in a given setting, namely Dt ¼ 100ð1−PsðtÞ PcðtÞÞ, where Dt is the relative reduction of the mean prevalence at year t of strategy s; Ps(t) the prevalence at time t of strategy s and Pc(t) the prevalence at time t of the control strategy. The relative maximum prevalence reduction, directly follow- ing the intervention deployment, is simply max(Dt). Outcome measured as the rate of resurgence a estimated all age prevalence following mass vaccination (purple) or mass vaccination with longer duration vaccine (pink), b estimated all age prevalence following full rounds of MDA alone (orange) or in combination with mass vaccination (green) or mass vaccination with longer duration vaccine (blue). Intervention coverage of 60% was assumed, with an initial yearly average PfPR2–10 = 4% with peak PfPR2–10 ≈10–15% (corresponding to an initial EIR of 2 and effective access to care E14 = 45%). Full description of the different strategies can be found in Additional file 1: Table S2 S4). Vaccination only strategies in which three doses of vaccine were given prior to seasonal transmission in the first year resulted in larger reductions in prevalence compared to when delivered during the first season (Fig. 3a and Additional file 1: Figure S4). For combined inter- vention strategies of mass vaccination added to MDA, the predicted relative maximum prevalence reduction achieved was higher compared to strategies deploying MDA alone (Fig. 3b-e), with a predicted higher preva- lence reduction when the drug and vaccine was deliv- ered independently from each other (random coverage, Fig. 3b-c) compared to when given simultaneously to the same people (same coverage, Fig. 3d-e). Delivering the 3 doses of vaccine prior to seasonal transmission the first year instead of during the seasonal transmission had lit- tle value (Fig. 3b-c)). MDA alone achieved large reduc- tions in prevalence, especially at higher coverage levels and low transmission levels, such that adding mass vac- cination had only a limited relative impact on prevalence reduction at these higher coverage levels (Additional file 1: Figure S5). S4). Vaccination only strategies in which three doses of vaccine were given prior to seasonal transmission in the first year resulted in larger reductions in prevalence compared to when delivered during the first season (Fig. 3a and Additional file 1: Figure S4). For combined inter- vention strategies of mass vaccination added to MDA, the predicted relative maximum prevalence reduction achieved was higher compared to strategies deploying MDA alone (Fig. 3b-e), with a predicted higher preva- lence reduction when the drug and vaccine was deliv- ered independently from each other (random coverage, Fig. 3b-c) compared to when given simultaneously to the same people (same coverage, Fig. 3d-e). Delivering the 3 doses of vaccine prior to seasonal transmission the first year instead of during the seasonal transmission had lit- tle value (Fig. 3b-c)). Outcome measured as the rate of resurgence MDA alone achieved large reduc- tions in prevalence, especially at higher coverage levels and low transmission levels, such that adding mass vac- cination had only a limited relative impact on prevalence reduction at these higher coverage levels (Additional file 1: Figure S5). Outcome measured as the rate of resurgence Where no elimination occurred, we estimated the time it takes after the mass interventions cease for prevalence to return to initial levels, by estimating resurgence rates. The rate of resurgence was assumed to follow a Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 6 of 15 Fig. 2 Example time course of predicted all age prevalence following intervention. Single simulation examples (chosen at random) of estimated continuous (upper plots) and median and range of estimated yearly average (lower plots) all age prevalence following different intervention strategies deployed during 2 years. a estimated all age prevalence following mass vaccination (purple) or mass vaccination with longer duration vaccine (pink), b estimated all age prevalence following full rounds of MDA alone (orange) or in combination with mass vaccination (green) or mass vaccination with longer duration vaccine (blue). Intervention coverage of 60% was assumed, with an initial yearly average PfPR2–10 = 4% with peak PfPR2–10 ≈10–15% (corresponding to an initial EIR of 2 and effective access to care E14 = 45%). Full description of the different strategies can be found in Additional file 1: Table S2 Fig. 2 Example time course of predicted all age prevalence following intervention. Single simulation examples (chosen at random) of estimated continuous (upper plots) and median and range of estimated yearly average (lower plots) all age prevalence following different intervention strategies deployed during 2 years. a estimated all age prevalence following mass vaccination (purple) or mass vaccination with longer duration vaccine (pink), b estimated all age prevalence following full rounds of MDA alone (orange) or in combination with mass vaccination (green) or mass vaccination with longer duration vaccine (blue). Intervention coverage of 60% was assumed, with an initial yearly average PfPR2–10 = 4% with peak PfPR2–10 ≈10–15% (corresponding to an initial EIR of 2 and effective access to care E14 = 45%). Full description of the different strategies can be found in Additional file 1: Table S2 Fig. 2 Example time course of predicted all age prevalence following intervention. Single simulation examples (chosen at random) of estimated continuous (upper plots) and median and range of estimated yearly average (lower plots) all age prevalence following different intervention strategies deployed during 2 years. Interruption of transmission Estimated maximum prevalence reduction achieved (proportion, where 0 indicates no prevalence reduction and 1 interruption of transmission) for coverage levels 40–100% (x-axis) by a single interventions, b-d combined interventions with drugs and vaccination given independently, and d-e combined interventions with drugs and vaccination given to the same population when administered simultaneously. a Compares MDA or mass vaccination alone: mass vaccination (purple) before (dashed lines) or during (solid lines) transmission season and full rounds of MDA (orange) or reduced rounds of MDA (yellow); b and d Compare reduced rounds of MDA (yellow) with combined strategies of mass vaccination before (dashed lines) or during (solid lines) the transmission season together with full rounds of MDA (brown); and c and e Compare reduced rounds of MDA (orange) with combined strategies of mass vaccination before (dashed lines) or during (solid lines) the transmission season together with full rounds of MDA (green). Each intervention is represented by the median and minimum-maximum range across 10 simulations per a strategy. Initial average PfPR2–10 = 4% with peak PfPR2–10 ≈10–15% (corresponding to an initial EIR = 2 and effective access to care E14 = 45%) Fig. 3 Relative maximum prevalence reduction achieved immediately following 2 years of intervention for different intervention coverages. Estimated maximum prevalence reduction achieved (proportion, where 0 indicates no prevalence reduction and 1 interruption of transmission) for coverage levels 40–100% (x-axis) by a single interventions, b-d combined interventions with drugs and vaccination given independently, and d-e combined interventions with drugs and vaccination given to the same population when administered simultaneously. a Compares MDA or mass vaccination alone: mass vaccination (purple) before (dashed lines) or during (solid lines) transmission season and full rounds of MDA (orange) or reduced rounds of MDA (yellow); b and d Compare reduced rounds of MDA (yellow) with combined strategies of mass vaccination before (dashed lines) or during (solid lines) the transmission season together with full rounds of MDA (brown); and c and e Compare reduced rounds of MDA (orange) with combined strategies of mass vaccination before (dashed lines) or during (solid lines) the transmission season together with full rounds of MDA (green). Each intervention is represented by the median and minimum-maximum range across 10 simulations per a strategy. Interruption of transmission The proportion of simulations achieving interruption of transmission for each strategy was calculated over all stochastic realizations for each setting. Assuming 60% coverage and 2 years of intervention, interruption of transmission occurred in 10% of the control simulations with initial PfPR2–10 less than 1%, highlighting possible stochastic extinction in the models at very low preva- lence [17]. Interruption of transmission for all strategies was dependent on initial PfPR2–10 with proportion of simulations in which transmission was interrupted de- creasing with increasing initial PfPR2–10 (Fig. 4a-b), and thus interruption of transmission was more likely in a setting where vector control is already at high coverage (expressed in this study as low initial prevalence). Inter- ruption of transmission was less likely to occur with lower case management (Additional file 1: Table S5 and Figure S6), with MDA, mass vaccination or combined strategies more likely to interrupt transmission if effect- ive access to care is higher. Excluding simulations with initial PfPR2–10 under 1% where stochastic extinction Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 7 of 15 Fig. 3 Relative maximum prevalence reduction achieved immediately following 2 years of intervention for different intervention coverages. Estimated maximum prevalence reduction achieved (proportion, where 0 indicates no prevalence reduction and 1 interruption of transmission) for coverage levels 40–100% (x-axis) by a single interventions, b-d combined interventions with drugs and vaccination given independently, and d-e combined interventions with drugs and vaccination given to the same population when administered simultaneously. a Compares MDA or mass vaccination alone: mass vaccination (purple) before (dashed lines) or during (solid lines) transmission season and full rounds of MDA (orange) or reduced rounds of MDA (yellow); b and d Compare reduced rounds of MDA (yellow) with combined strategies of mass vaccination before (dashed lines) or during (solid lines) the transmission season together with full rounds of MDA (brown); and c and e Compare reduced rounds of MDA (orange) with combined strategies of mass vaccination before (dashed lines) or during (solid lines) the transmission season together with full rounds of MDA (green). Each intervention is represented by the median and minimum-maximum range across 10 simulations per a strategy. Initial average PfPR2–10 = 4% with peak PfPR2–10 ≈10–15% (corresponding to an initial EIR = 2 and effective access to care E14 = 45%) Fig. 3 Relative maximum prevalence reduction achieved immediately following 2 years of intervention for different intervention coverages. Interruption of transmission Initial PfPR2–10 (%) levels are shown on the x-axis, and proportion of the simulations falling into each category are shown on the y-axis. All interventions were deployed for 2 years at a coverage of 60%. Categories of simulations are i) interruption of transmission occurred with no intervention at all, due to very low initial prevalence (black), ii) interruption of transmission occurred with single interventions, namely with mass vaccination with RTS,S like vaccine (purple) or longer duration vaccine (pink), or with MDA (orange), iii) interruption of transmission occurred only adding mass vaccination to MDA (green and blue using with RTS,S like vaccine or longer duration vaccine respectively), and iv) resurgence occurred and no interruption of transmission was achieved (grey).c Estimated synergy coefficient (σ) of the combined mass vaccination and MDA intervention in regards probability to interrupt transmission. The x-axis indicates coverage levels of MDA, and the y-axis coverage levels of mass vaccination, and the level of synergy between the two intervention strategies are indicated in colour. Blue represents synergistic behavior (> 0) in the combined MDA and mass vaccination, light green represents values of 0 which imply the combined interventions are not more than additive, and yellow to red colours represent values less than 0 which imply less than additive or maximum level was reached by one or both single interventions. Grey areas represent settings where resurgence occurred in all simulation, thus no synergy could be calculated. The synergy coefficient are shown for the combination of MDA with RTS,S like vaccine (upper row) and MDA with the longer duration vaccine (bottom row), and for different levels of initial PfPR2–10 (%) (columns) Fig. 4 Interruption of transmission and synergism for different intervention strategies. a-b Proportion of simulations in which interruption of transmission is estimated to be achieved with mass vaccination (a) or MDA combined with mass vaccination compared to MDA only (b). Initial PfPR2–10 (%) levels are shown on the x-axis, and proportion of the simulations falling into each category are shown on the y-axis. All interventions were deployed for 2 years at a coverage of 60%. Interruption of transmission Categories of simulations are i) interruption of transmission occurred with no intervention at all, due to very low initial prevalence (black), ii) interruption of transmission occurred with single interventions, namely with mass vaccination with RTS,S like vaccine (purple) or longer duration vaccine (pink), or with MDA (orange), iii) interruption of transmission occurred only adding mass vaccination to MDA (green and blue using with RTS,S like vaccine or longer duration vaccine respectively), and iv) resurgence occurred and no interruption of transmission was achieved (grey).c Estimated synergy coefficient (σ) of the combined mass vaccination and MDA intervention in regards probability to interrupt transmission. The x-axis indicates coverage levels of MDA, and the y-axis coverage levels of mass vaccination, and the level of synergy between the two intervention strategies are indicated in colour. Blue represents synergistic behavior (> 0) in the combined MDA and mass vaccination, light green represents values of 0 which imply the combined interventions are not more than additive, and yellow to red colours represent values less than 0 which imply less than additive or maximum level was reached by one or both single interventions. Grey areas represent settings where resurgence occurred in all simulation, thus no synergy could be calculated. The synergy coefficient are shown for the combination of MDA with RTS,S like vaccine (upper row) and MDA with the longer duration vaccine (bottom row), and for different levels of initial PfPR2–10 (%) (columns) Fig. 4 Interruption of transmission and synergism for different intervention strategies. a-b Proportion of simulations in which interruption of transmission is estimated to be achieved with mass vaccination (a) or MDA combined with mass vaccination compared to MDA only (b). Initial PfPR2–10 (%) levels are shown on the x-axis, and proportion of the simulations falling into each category are shown on the y-axis. All interventions were deployed for 2 years at a coverage of 60%. Interruption of transmission Initial average PfPR2–10 = 4% with peak PfPR2–10 ≈10–15% (corresponding to an initial EIR = 2 and effective access to care E14 = 45%) can occur, mass vaccination alone did not lead to trans- mission interruption when using RTS,S-like-duration vaccine and led to transmission interruption in 20% of the simulations in settings of initial PfPR2–10 of 2% when using the vaccine with longer duration of protection (Fig. 4a). With a vaccine of lower vaccine efficacy (50%), no interruption of transmission for PfPR2–10 higher than 1% was observed. Similarly, at initial PfPR2–10 of 2% the proportion of simulations with transmission interruption was 20% for strategies deploying MDA alone. For initial PfPR2–10 greater than 4%, transmission interruption was less likely for both vaccination alone and MDA alone (Fig. 4a-b). Combining both interventions at 60% cover- age, the highest proportion of simulations in which interruption of transmission was achieved for 2 years of intervention was estimated for initial PfPR2–10 range 2– 4%, with interruption of transmission estimated at 75% for strategies deploying MDA combined with immunization with a vaccine with equivalent protection properties to RTS,S/AS01, and of 80% when combined with the vaccine with longer duration of protection (Fig. 4b). Including a 3rd year of intervention in these settings increased the proportion to 95–100%, even assuming a coverage of 60% (Additional file 1: Table S3). Combining MDA with a vaccine with lower efficacy resulted in lower chances to interrupt transmission compared to combined strategies with a vaccine with higher efficacy profile, however, for a PfPR2–10 of 2%, this lower efficacy Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 8 of 15 Fig. 4 Interruption of transmission and synergism for different intervention strategies. a-b Proportion of simulations in which interruption of transmission is estimated to be achieved with mass vaccination (a) or MDA combined with mass vaccination compared to MDA only (b). Initial PfPR2–10 (%) levels are shown on the x-axis, and proportion of the simulations falling into each category are shown on the y-axis. All interventions were deployed for 2 years at a coverage of 60%. Categories of simulations are i) interruption of transmission occurred with no intervention at all, Fig. 4 Interruption of transmission and synergism for different intervention strategies. a-b Proportion of simulations in which interruption of transmission is estimated to be achieved with mass vaccination (a) or MDA combined with mass vaccination compared to MDA only (b). Interruption of transmission Categories of simulations are i) interruption of transmission occurred with no intervention at all, due to very low initial prevalence (black), ii) interruption of transmission occurred with single interventions, namely with mass vaccination with RTS,S like vaccine (purple) or longer duration vaccine (pink), or with MDA (orange), iii) interruption of transmission occurred only adding mass vaccination to MDA (green and blue using with RTS,S like vaccine or longer duration vaccine respectively), and iv) resurgence occurred and no interruption of transmission was achieved (grey).c Estimated synergy coefficient (σ) of the combined mass vaccination and MDA intervention in regards probability to interrupt transmission. The x-axis indicates coverage levels of MDA, and the y-axis coverage levels of mass vaccination, and the level of synergy between the two intervention strategies are indicated in colour. Blue represents synergistic behavior (> 0) in the combined MDA and mass vaccination, light green represents values of 0 which imply the combined interventions are not more than additive, and yellow to red colours represent values less than 0 which imply less than additive or maximum level was reached by one or both single interventions. Grey areas represent settings where resurgence occurred in all simulation, thus no synergy could be calculated. The synergy coefficient are shown for the combination of MDA with RTS,S like vaccine (upper row) and MDA with the longer duration vaccine (bottom row), and for different levels of initial PfPR2–10 (%) (columns) vaccine when combined with MDA increased the chances of interrupting transmission from 20 to 60% and from 20 to 70% with longer duration vaccine com- pared with using MDA only (Additional file 1: Figure S7). Administering the vaccine and MDA simultan- eously to the same population for a given coverage level resulted in slightly lower predictions of chances to interrupt transmission than with random coverage between the two interventions (Additional file 1: Fig- ure S6). Proportions of simulations with transmission interruption were similar for strategies vaccinating the population before the transmission season or during the transmission for mass vaccination or combined interventions, even if slightly higher when vaccinating Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 9 of 15 before the season (Additional file 1: Table S3). Fur- ther results can be found in Additional file 1: Tables S3-S5. before the season (Additional file 1: Table S3). Fur- ther results can be found in Additional file 1: Tables S3-S5. Interruption of transmission independently from each other or simultaneously to the same target population (Fig. 4c and Additional file 1: Figure S7). Synergism, defined as observing a greater propor- tion of simulations with interrupted transmission in the combined interventions than in the sum of the single interventions (see methods), was calculated for a range of initial PfPR2–10 and intervention coverage levels (Fig. 4c and Additional file 1: Figure S7 and Figure S9). The lower the initial PfPR2–10, the less likely synergism was observed, as the chances to interrupt transmission using MDA as a single inter- vention was higher. Similarly, increasing case manage- ment or increasing MDA coverage decreased synergism between the two interventions. Increasing vaccine coverage or duration of protection increased the chances to interrupt transmission but did not have an obvious effect on synergism. The lower initial vaccine efficacy decreased the number of settings where synergism was predicted (Additional file 1: Fig- ure S7). Synergism was predicted for both scenarios of when the interventions were delivered Resurgence Resurgence occurs in each simulation in which transmis- sion was not interrupted with prevalence eventually returning to initial levels. Additional to predicting the chances to achieve transmission interruption (or risk of resurgence), we estimated the time to resurgence after deployment of the intervention cease. This time reflects a period where prevalence remains at very low levels and other interventions such as reactive case detection (not simulated here) could potentially be effective to inter- rupt transmission. In settings of initial PfPR2–10 = 9% re- surgence occurred in all simulations, and the estimated resurgence half-life, λ50, was lowest for the MDA strat- egy (median λ50 = 1.83 years), whereas for the strategy combining MDA with mass vaccination with RTS,S-like- duration vaccine λ50 was more than 2.5 years longer than MDA alone (median λ50 = 3.68 years) (Fig. 5 and Add- itional file 1: Table S6). The estimated Hill’s slope Fig. 5 Resurgence following 2 years of deployment of different interventions at PfPR2–10 = 4 and 9%. a Predicted relative resurgence at initial PfPR2–10 = 9% (EIR = 2 and E14 = 45%) with 95% confidence intervals estimated from fitting a 4-parameter logistic regression to the pooled simulations for strategy MDA alone (orange), mass vaccination alone (purple), mass vaccination with a longer duration vaccine (pink), MDA combined with mass vaccination (green) and MDA combined with a longer duration vaccine (blue). b The average minimum prevalence in the total population reached directly following each intervention. c-d Summarized parameters describing resurgence (boxplots with median and 95% confidence intervals) estimated from the logistic regressions to each simulations for each strategy are the estimated half-life or time till 50% resurgence, λ50, representing the years after maximum prevalence reduction was reached in which prevalence resurges by 50% of the reduction and the time till 10% resurgence, λ10, representing the years after maximum prevalence reduction was reached where prevalence resurges 10% of the reduction. The estimates are shown for initial PfPR2–10 = 9% in c and for initial PfPR2–10 = 4% in d. The corresponding number of simulations with resurgence per strategy and setting are indicated at the top of each summary boxplot Fig. 5 Resurgence following 2 years of deployment of different interventions at PfPR2–10 = 4 and 9%. Vaccination to reduce years of mass deployment and reduce number of MDA rounds For initial PfPR2–10 of 4%, resurgence parameters were calculated for 80% of the simulations in which resur- gence occurred. In the strategy combining MDA with mass vaccination using a vaccine with equivalent protec- tion profile to RTS,S/AS01, λ50 was estimated to be more than 2 years longer than MDA alone (median λ50 = 6.78 years compared to median λ50 = 4.63 years) (Fig. 5 and Additional file 1: Table S6), and the time after maximum prevalence reduction was reached in which prevalence resurges to 10% of the reduction in- creased by 1.48 years (median λ10 = 3.46 against median λ10 = 1.98 for MDA only). Adding mass vaccination to MDA strategies of 2 years resulted in a similar or higher proportion of simulations with interruption of transmission in a shorter time period of intervention than compared to MDA deployed for 3 years. This was true when MDA and vaccination was administered simultaneously to the same people and when given independently from each other (random coverage between both interventions) (Table 1). Resurgence BMC Infectious Diseases (2019) 19:920 Page 10 of 15 Page 10 of 15 rates for both prevalence levels investigated (Additional file 1: Table S6). Using a vaccine with lower initial effi- cacy in the combined strategies had less impact in delay- ing resurgence but nevertheless the resurgence half-life was estimated to be increased by almost a year com- pared to using MDA only (Additional file 1: Table S6). Simulations that assumed increased levels of imported infections resulted in decreased estimates of resurgence half-life, however the estimated resurgence rates remained lower for the combined strategies compared to MDA alone (Additional file 1: Table S6). Increasing levels of case management decreased proportion of sim- ulations where resurgence occurred (Additional file 1: Table S6). (median 3.55) was also higher, as was λ10 (increased by 1.28 years), defined as the time after maximum preva- lence reduction was reached in which prevalence re- surges to 10% of the reduction, indicating that resurgence was delayed in the combined strategy and prevalence remained lower for a longer time period of time compared to MDA alone. Half-life of resurgence were similar in the strategies where mass vaccination was performed before the transmission season, with no evident increase (Additional file 1: Table S6). For strat- egies with MDA combined with mass vaccination using a vaccine with equivalent protection profiles to RTS,S/ AS01, the half-life of resurgence was increased by 1.28 years if the interventions were deployed for 3 years in- stead of 2 years. Resurgence a Predicted relative resurgence at initial PfPR2–10 = 9% (EIR = 2 and E14 = 45%) with 95% confidence intervals estimated from fitting a 4-parameter logistic regression to the pooled simulations for strategy MDA alone (orange), mass vaccination alone (purple), mass vaccination with a longer duration vaccine (pink), MDA combined with mass vaccination (green) and MDA combined with a longer duration vaccine (blue). b The average minimum prevalence in the total population reached directly following each intervention. c-d Summarized parameters describing resurgence (boxplots with median and 95% confidence intervals) estimated from the logistic regressions to each simulations for each strategy are the estimated half-life or time till 50% resurgence, λ50, representing the years after maximum prevalence reduction was reached in which prevalence resurges by 50% of the reduction and the time till 10% resurgence, λ10, representing the years after maximum prevalence reduction was reached where prevalence resurges 10% of the reduction. The estimates are shown for initial PfPR2–10 = 9% in c and for initial PfPR2–10 = 4% in d. The corresponding number of simulations with resurgence per strategy and setting are indicated at the top of each summary boxplot Fig. 5 Resurgence following 2 years of deployment of different interventions at PfPR2–10 = 4 and 9%. a Predicted relative resurgence at initial PfPR2–10 = 9% (EIR = 2 and E14 = 45%) with 95% confidence intervals estimated from fitting a 4-parameter logistic regression to the pooled simulations for strategy MDA alone (orange), mass vaccination alone (purple), mass vaccination with a longer duration vaccine (pink), MDA combined with mass vaccination (green) and MDA combined with a longer duration vaccine (blue). b The average minimum prevalence in the total population reached directly following each intervention. c-d Summarized parameters describing resurgence (boxplots with median and 95% confidence intervals) estimated from the logistic regressions to each simulations for each strategy are the estimated half-life or time till 50% resurgence, λ50, representing the years after maximum prevalence reduction was reached in which prevalence resurges by 50% of the reduction and the time till 10% resurgence, λ10, representing the years after maximum prevalence reduction was reached where prevalence resurges 10% of the reduction. The estimates are shown for initial PfPR2–10 = 9% in c and for initial PfPR2–10 = 4% in d. The corresponding number of simulations with resurgence per strategy and setting are indicated at the top of each summary boxplot Camponovo et al. Percentage of resurgence in the simulations are shown for interventions (rows) at 60% coverage, for 2- and 3- years of full rounds of MDA, and 2 years of full or reduced rounds of MDA combined with mass vaccination with RTS,S like vaccine (initial efficacy against infection 91%, half-life 0.61 years), longer duration of protection vaccine (half-life 1.5 years), or lower initial efficacy (50%). Results are shown for different levels of initial PfPR2–10 (%) (columns), and results where the vaccine is given to same proportion of the population than the simultaneous MDA are indicated with A in the simulations are shown for interventions (rows) at 60% coverage, for 2- and 3- years of full rounds of MDA, and 2 years of full or bi d ith i ti ith RTS S lik i (i iti l ffi i t i f ti 91% h lf lif 0 61 ) l d ti f Vaccination to reduce years of mass deployment and reduce number of MDA rounds For PfPR2–10 = 2–4%, the proportion of simulations in which interruption of transmission occurred was 30% higher in strategies combining RTS,S-like-duration mass vaccin- ation with full rounds of MDA for 2 years (65% Overall, delivering the vaccine to the same people as receiving MDA produced similar estimates of resurgence Table 1 Percentage of simulations in which interruption of transmission occurred for different strategies and deploy times Strategies Interrupted transmission [%] PfPR2–10 [%] 2 4 5 7 3 years full MDA rounds 80 20 0 0 2 years full MDA rounds + vaccine 90 40 0 0 100A 40A 0A 0A 2 years full MDA rounds + longer duration vaccine 100 60 30 0 80A 40A 30A 0A 2 years full MDA rounds + vaccine lower efficacy 60 0 0 0 30A 0A 0A 0A 2 years full MDA rounds 20 0 0 0 2 years reduced MDA rounds + vaccine 70 10 0 0 70A 0A 0A 0A 2 years reduced MDA rounds + longer duration vaccine 90 70 0 0 80A 10A 0A 0A 2 years reduced MDA rounds + vaccine lower efficacy 30 0 0 0 30A 0A 0A 0A Percentage of resurgence in the simulations are shown for interventions (rows) at 60% coverage, for 2- and 3- years of full rounds of MDA, and 2 years of full or reduced rounds of MDA combined with mass vaccination with RTS,S like vaccine (initial efficacy against infection 91%, half-life 0.61 years), longer duration of protection vaccine (half-life 1.5 years), or lower initial efficacy (50%). Results are shown for different levels of initial PfPR2–10 (%) (columns), and results where the vaccine is given to same proportion of the population than the simultaneous MDA are indicated with A of simulations in which interruption of transmission occurred for different strategies and deploy times entage of simulations in which interruption of transmission occurred for different strategies and deploy times Interrupted transmission [%] Percentage of resurgence in the simulations are shown for interventions (rows) at 60% coverage, for 2- and 3- years of full rounds of MDA, and 2 years of full or reduced rounds of MDA combined with mass vaccination with RTS,S like vaccine (initial efficacy against infection 91%, half-life 0.61 years), longer duration of protection vaccine (half-life 1.5 years), or lower initial efficacy (50%). Vaccination to reduce years of mass deployment and reduce number of MDA rounds For PfPR2–10 = 2% the proportion of simulations with interrupted transmission was 70% if 2 years of reduced rounds of MDA combined with mass vaccination was deployed (namely 3 rounds of MDA plus vaccination the first year and only 1 round of MDA plus vaccine the second year) compared to an interruption of transmission of 20% in strategies of 2 years of 3 rounds of MDA only (Table 1). Using a longer duration vaccine, 2 years of reduced rounds of MDA combined with mass vaccination at PfPR2–10 = 2% led to a proportion of simulations with interrupted transmis- sion up to 90% (Table 1). A vaccine with a lower initial efficacy of 50% could replace 2 rounds of MDA in the second year, but in both scenarios (vaccine and MDA with reduced rounds or full round of MDA) chances to interrupt transmission remained low (Table 1). It will be important to consider whether the ap- proaches modelled here are operationally feasible and what level of health system and program strength is re- quired to coordinate vaccination and drug campaigns. For this reason, we investigated a range of coverages and our main results were presented for coverages of 60%, assuming any higher was too optimistic. In those condi- tions, with initial PfPR2–10 = 4%, we found that for the simulations where transmission is not interrupted, the rate of resurgence is halved compared to MDA alone, extending the resurgence half-life up to two and a half years. Alternatively, we investigated a higher MDA coverage compared to vaccination and we found that adding mass vaccination even at a lower coverage (40%) remained beneficial in delaying resurgence and reducing MDA rounds. We also found limited differences in im- pact between the combined strategies with vaccination before or during the transmission season, arguing for a preference of during the season with 3 visits as opposed to more visits. Simulations where MDA coverage was 60% but vac- cine coverage was 40% in the combined interventions were additionally compared. In settings of PfPR2–10 = 1– 4% the proportion of simulations with transmission interruption was lower for vaccine coverage of 40% com- pared to 60%, thus increasing proportion with resur- gence (Additional file 1: Table S3 and S4). Vaccination to reduce years of mass deployment and reduce number of MDA rounds Results are shown for different levels of initial PfPR2–10 (%) (columns), and results where the vaccine is given to same proportion of the population than the simultaneous MDA are indicated with A Percentage of resurgence in the simulations are shown for interventions (rows) at 60% coverage, for 2- and 3- years of full rounds of MDA, and 2 years of full or reduced rounds of MDA combined with mass vaccination with RTS,S like vaccine (initial efficacy against infection 91%, half-life 0.61 years), longer duration of protection vaccine (half-life 1.5 years), or lower initial efficacy (50%). Results are shown for different levels of initial PfPR2–10 (%) (columns), and results where the vaccine is given to same proportion of the population than the simultaneous MDA are indicated with A Page 11 of 15 Page 11 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Camponovo et al. BMC Infectious Diseases (2019) 19:920 longer than the duration of the transmission season (even with waning protective immune responses restored with yearly immunizations) resulted in higher maximum prevalence reduction, increased chance of elimination, and importantly, delayed resurgence compared to inter- ventions without the vaccine. Our simulations suggest that combining MDA with an AIV has a synergistic ef- fect in regards to chances to interrupt transmission, and synergism increased with greater levels of initial preva- lence and lower MDA coverage or case-management. Synergism can be the result of deploying two interven- tions which increases overall coverage of the population covered by at least one intervention, but as synergism was also predicted when the vaccine and the drug were given to the same people given coverage, although in a lesser extent, it is also likely a consequence of the inter- ventions targeting different parasite life stages. transmission interruption) compared to 3 years of full MDA rounds only (50% transmission interruption) (Table 1). For MDA combined with longer duration vac- cines, we estimated 80% of simulations interrupted transmission (Table 1). For MDA combined with a vac- cine of lower initial efficacy (50%), we predicted 3 years of MDA to have greater chances to interrupt transmission. Adding mass vaccination to MDA could replace 2 rounds of MDA in the second year for similar or im- proved chances of interruption of transmission, espe- cially at lower prevalence levels. Vaccination to reduce years of mass deployment and reduce number of MDA rounds However, even a strategy with a lower coverage of vaccination at 40%, combination strategies (mass vaccination and MDA) deployed during 2 years remained more favorable compared to using 3 years of MDA alone (Additional file 1: Table S3 and S4). In addition, we estimated that vaccines of longer dur- ation of protection can further delay resurgence, arguing that research into the feasibility of increasing longevity of protection of a vaccine has importance for the elimin- ation strategies investigated here. Discussion context of mass vaccination, as the efficacy and pro- tection might be different for different ages and previ- ous exposure, extension of the protection profile and vaccine parameterization from children to the entire population is a necessary simplification based on cur- rently available data. A full sensitivity analysis around vaccine efficacy profiles has not been performed in this study, but simulations where initial efficacy against infection for children and adults was reduced to 50% suggest that this would considerably decrease the prevalence and case management settings where the combination strategies would reach very low transmission levels or interruption of transmission. Thirdly, we did not model potential increases in vector control or case management levels over the time of intervention, either due to vector control campaigns or active efforts to improve access to treatment or due to a better efficiency in the health system as fewer cases need to be treated. Both would further reduce prevalence levels following mass intervention deployment, and in some settings, would likely enhance the effect of mass vaccination, especially during the resurgence period. Conversely, we did not model deceasing levels of vector control which may decrease the benefits of the mass intervention strategies or increase rates of resurgence. Furthermore, interruption of transmission with mass interventions is likely only possible for scenarios of low initial prevalence (low prevalence as a result of vector control or other interventions), with a high level of case management and an increase in other intervention tools, which was recently highlighted by the WHO Malaria Policy Advisory Committee meeting with regards to the use of MDA [38]. Our simulations also indicated that, chances to interrupt transmission increase with higher case management. This implies that such mass interven- tions should not be investigated independently of case management, and that together with vector control or reactive tools, will play a decisive role in the success of the interventions to reach elimination and avoid resurgence. g g We chose a 3-month seasonal malaria transmission setting of sub-Sahel African countries approximating the pattern in Senegal where SMC is ongoing, and investi- gated the impact on average transmission, not account- ing for heterogeneities in transmission. Further studies could be undertaken to tailor the analysis specifically to a country or region, with corresponding transmission pattern, heterogeneity, and other underlying settings, such as case management or adapted timing and dur- ation of deployment. Discussion Given the enormous multiplicity of combinations of in- terventions that might be considered when targeting the whole population with combination strategies, and given the expense of mass deployment, it is critical to carry out in silico investigations of the potential of novel de- ployment options in order to avoid wasteful field trials of fruitless combinations [37]. The addition of mass vaccination to MDA, compared with MDA alone, has the potential to reduce the number of MDA rounds needed, as well as the number of years that interventions may be needed, to achieve similar public health benefits of prevalence reduction and chances to interrupt transmission, and thus reduce the length of deployment of interventions from 3 to 2 years. Reducing the length of overall intervention deployment has the potential to reduce the number of mass inter- vention visits, and thus costs of visits. Moreover, al- though an economic analysis was outside the scope of this study, a shorter period of deployment could be To date little has been done to estimate the impact of adding a similar AIV to existing mass treatment strat- egies. By combining mass vaccination to treatment, the infectious reservoir is reduced by clearing infections in a large proportion of the population via MDA, and adding an intervention with ability to prevent infections for Page 12 of 15 Page 12 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Camponovo et al. BMC Infectious Diseases (2019) 19:920 beneficial to national malaria control and elimination strategies facing budgetary constraints. Although this work did not explicitly explore mobility, we did model importations of infections and highly mobile populations will limit operational coverage of the vaccines and MDA. Limited duration mass interventions are not intended to tackle importation. As with MDA, importation is likely to have little effect on the estimated prevalence reduc- tion unless initial prevalence is very low, and in these settings where interruption of transmission is achieved a high level of case management will be important to maintain zero prevalence [16]. In line with this, we found importation of infections had little effect on prevalence reduction achieved but that time to resur- gence decreased as importation increased. Discussion In more realistic settings the proportion of the population covered would neither be completely independent nor would it be exactly the same between deployments, and thus our as- sumptions might be slightly optimistic so that the nom- inal coverages we simulate reflect higher values in the field. alone, there may be a role in malaria outbreaks or in fra- gile contexts where health systems are weakened by other disease outbreaks [44] or conflict. However, the number of doses required to reach high efficacy may limit this application. There may also be a role in help- ing mitigate spread of drug resistance in the Mekong or for targeting forest workers [37], or for areas in which residual or outdoor transmission renders indoor residual spraying less effective, but further investigations are required. Given that MDA and SMC are currently imple- mented in several settings [19, 40], and that the com- bination of RTS,S and SMC is being tested in field trials [14], we are in a position to consider adapting these combination of interventions to different ages. Technical challenges however remain. It is envisioned that RTS,S may be delivered in much the same way as meningococcal A conjugate vaccine is being de- ployed in sub-Saharan Africa in 1–29 year old, ini- tially in intense mass vaccination campaigns, and then followed by introduction into existing Extended Immunization Programmes in the individual countries [41]. Any vaccines used in malaria elimination cam- paigns will need to be WHO pre-qualified ensuring that characteristics are aligned with cold-chain re- quirements available in the countries where it is de- ployed [42]. For an AIV to be used as a mass intervention in Africa or Asia, several important un- knowns must be addressed with respect to immuno- genicity in adults and immunogenicity in populations outside Africa. Measurement of immune responses and protection in adults have been tested in the Me- kong, where RTS,S was administered alone or in combination to drugs [43] as are many other CHMI (controlled human malaria infection) studies of RTS,S, R21, and PfSPZ. These studies should provide add- itional evidence that could be used to model their po- tential role in mass vaccination scenarios. Other challenges include safety and efficacy testing when vaccines are administered with drugs and regulatory hurdles for use of any of the AIVs for children and adults, which may be more difficult with new anti- gens. Discussion The impact of interventions may be sensitive to timing and the seasonal pattern chosen for malaria transmission, and that the heterogeneity of transmission across a given region will imply a need for focal planning. This work was confined to a limited number of model-based scenarios of mass interventions and as such there are several limitations. Firstly, several aspects not captured might enhance or diminish the impact of mass vaccination combined with MDA. On a biological level, vaccination with an AIV with drugs has not been suffi- ciently investigated, and it is unclear how vaccine effi- cacy, immune dynamics in response to the vaccine, the length of protection, or the rate of vaccine failure would be influenced. Such effects are unknown, however a lower rate of vaccine failure and longer duration of pro- tection would delay resurgence and extend the pre- elimination period, and conversely higher failure rates would decrease benefits. Additionally, we did not ad- dress any potential evolutionary effects such as vaccine insensitivity, nor antigenic sin. However, we note deploying with treatment may protect against evolution of vaccine insensitivity. Previous modeling has shown the impact of MDA is very sensitive to population size, being best suited for small populations and with total impact sensitive to the proportion of the population receiving no treatment from any of the deployed rounds [39]. In our current work, individual rounds of MDA were given to a ran- dom proportion of the population for a given coverage, mass vaccination was deployed either independently to MDA or in some settings simultaneously to MDA, and the 3 vaccine doses independently from the fourth dose. As expected, scenarios with total correlation between those covered by MDA and mass vaccination predicted a slightly reduced impact, but the key determinants probably remains in the random coverage between MDA rounds similar to results found for multiple Secondly, the underlying vaccine assumptions in the models, including the effect of a forth vaccine dose, may influence results. Vaccine properties were based on analysis of the results of RTS,S Phase 3 clinical trials [13], which targeted infants aged 6 to 12 weeks and young children aged 5 to 17 months. In the Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 13 of 15 Page 13 of 15 Page 13 of 15 rounds of MDA [16], and total correlation between rounds has not been investigated here. Discussion From an operational perspective, the challenges may be significant, ensuring sufficient coverage of the first three or required initial number of doses of the vaccine. And lastly, to make use of the potential benefit in delaying resurgence, settings must be pre- pared with additional interventions during the low prevalence period following 2 or 3 years of mass intervention, and as with MDA, it will be important to include surveillance and response strategies com- bined with strong health systems to address import- ation of infection and preserve this delay in resurgence. Gi en combining mass accination ith treatment has Our modelling study provides evidence of potential use of an AIV in malaria elimination with characteris- tics similar to the most advanced vaccines: high initial efficacy and with limited duration of protection in an area with highly seasonal malaria transmission. As- sumptions in regards to demographics, past history of malaria, transmission profile, vector species, and health systems strengths were not geographic specific. An important next step will be to understand which settings and countries these mass interventions strat- egies are feasible for and what range of settings with pockets of focal transmission. This would include an assessment of cost-effectiveness. Conclusion This work reports modelling outputs of vaccination with existing and potential forthcoming AIVs combined with MDA - indicating such a combination could be a pos- sible strategy to rapidly decrease prevalence and keep prevalence levels lower than would MDA alone for sev- eral years before resurgence, thus challenging thinking on the role of malaria tools already at our disposal. A deep understanding of current malaria burden and level of case management, with a cautious investigation of the timing and operational feasibility of the mass interven- tions would nevertheless be of paramount importance before implementing such a strategy for malaria elimin- ation, as the risk of resurgence and thus failure of the mass campaigns will remain a threat. Acknowledging that malaria elimination will not be achieved with a sin- gle intervention, we should continue to investigate com- bination intervention strategies targeting different parts of the lifecycle that will help achieve our goals. Supplementary information pp y Supplementary information accompanies this paper at https://doi.org/10. 1186/s12879-019-4467-4. Additional file 1: Table S1. Overview of the input parameter values for the simulations and the direct outputs measured. Table S2. Overview of the main and supplementary simulated strategies. Figure S1. Single simulation examples of estimated continuous all age prevalence following different intervention. Figure S2. Median and range of estimated yearly average all age prevalence following different intervention strategies. Figure S3. Relative impact of combined strategies Additional file 1: Table S1. Overview of the input parameter values for the simulations and the direct outputs measured. Table S2. Overview of the main and supplementary simulated strategies. Figure S1. Single simulation examples of estimated continuous all age prevalence following different intervention. Figure S2. Median and range of estimated yearly average all age prevalence following different intervention strategies. Figure S3. Relative impact of combined strategies Additional file 1: Table S1. Overview of the input parameter values for the simulations and the direct outputs measured. Table S2. Overview of the main and supplementary simulated strategies. Figure S1. Single simulation examples of estimated continuous all age prevalence following different intervention. Figure S2. Median and range of estimated yearly average all age prevalence following different intervention strategies. Figure S3. Relative impact of combined strategies Given combining mass vaccination with treatment has potential to delay resurgence for longer than treatment Page 14 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Page 14 of 15 Page 14 of 15 Received: 1 February 2019 Accepted: 12 September 2019 References 1. World Health Organization (WHO). Global technical strategy for malaria 2016–2030. WHO Geneva. 2015. 1–35 p. 2. Bhatt S, Weiss DJ, Cameron E, Bisanzio D, Mappin B, Dalrymple U, et al. The effect of malaria control on Plasmodium falciparum in Africa between 2000 and. Nature. 2015;2015:1–9. 3. World Health Organization (WHO). World Malaria Report. 2017:2017. 3. World Health Organization (WHO). World Malaria Report. 2017:2017. 4. Greenwood BM. Control to elimination: implications for malaria research. Trends Parasitol. 2008;24(10):449–54. 4. Greenwood BM. Control to elimination: implications for malaria research. Trends Parasitol. 2008;24(10):449–54. 5. European Medicines Agency. First malaria vaccine receives positive scientific opinion from EMA. Press Release. 2015;44(July):1–3. 5. European Medicines Agency. First malaria vaccine receives positive scient opinion from EMA. Press Release. 2015;44(July):1–3. 6. Global Advisory Committee on Vaccine Safety. Safety monitoring of the RTS,S vaccine pilot implementation programme [Internet]. 2017 [cited 2017 Nov 1]. Available from: https://www.who.int/vaccine_safety/committee/ topics/malaria_vaccine/June_2017/en/ 7. Penny MA, Pemberton-Ross P, Smith TA. The time-course of protection of the RTS,S vaccine against malaria infections and clinical disease. Malar J 2015;14(1):437. 8. White MT, Verity R, Griffi JT, Asante KP, Owusu-agyei S, Greenwood B, et al. Immunogenicity of the RTS,S/AS01 malaria vaccine and implications for duration of vaccine effi cacy : secondary analysis of data from a phase 3 randomised controlled trial. Lancet Infect Dis. 2015;15:1450–8. 8. White MT, Verity R, Griffi JT, Asante KP, Owusu-agyei S, Greenwood B, et al. Immunogenicity of the RTS,S/AS01 malaria vaccine and implications for duration of vaccine effi cacy : secondary analysis of data from a phase 3 randomised controlled trial. Lancet Infect Dis. 2015;15:1450–8. Funding The work was funded by the Swiss National Science Foundation through SNSF Professorship of MAP (PP00P3_170702). 17. Pemberton-Ross P, Chitnis N, Pothin E, Smith TA. A stochastic model for the probability of malaria extinction by mass drug administration. Malar J. 2017; 16(1):1–9. 17. Pemberton-Ross P, Chitnis N, Pothin E, Smith TA. A stochastic model for the probability of malaria extinction by mass drug administration. Malar J. 2017; 16(1):1–9. Abbreviations AIV A i I f i 9. Regules JA, Cicatelli SB, Bennett JW, Paolino KM, Twomey PS, Moon JE, et al. Fractional third and fourth dose of RTS,S/AS01 malaria candidate vaccine: A phase 2a controlled human malaria parasite infection and immunogenicity study. J Infect Dis 2016;214(5):762–771. AIV: Anti Infective Vaccine; E14: 14 day effective treatment coverage; EIR: Entomological Inoculation Rate; MDA: Mass Drug Administration; PfPR2– 10: Initial parasite prevalence levels in 2–10 year olds; SMC: Seasonal Malaria Chemoprevention AIV: Anti Infective Vaccine; E14: 14 day effective treatment coverage; EIR: Entomological Inoculation Rate; MDA: Mass Drug Administration; PfPR2– 10: Initial parasite prevalence levels in 2–10 year olds; SMC: Seasonal Malaria Chemoprevention 10. Sanaria Inc. Sanaria® PfSPZ vaccine against malaria receives FDA fast track designation [internet]. 2016. Available from: http://www.sanaria.com Authors’ contributions MAP, FC, and CL conceived the study, MAP and FC designed the simulations, FC run the simulations, FC and MAP performed the analysis, MAP and FC wrote the first draft of the manuscript. All authors have reviewed and approved the manuscript. 15. Cairns M, Roca-Feltrer A, Garske T, Wilson AL, Diallo D, Milligan PJ, et al. Estimating the potential public health impact of seasonal malaria chemoprevention in African children. Nat Commun. 2012;3:881. 15. Cairns M, Roca-Feltrer A, Garske T, Wilson AL, Diallo D, Milligan PJ, et al. Estimating the potential public health impact of seasonal malaria chemoprevention in African children. Nat Commun. 2012;3:881. 16. Brady OJ, Slater HC, Pemberton-Ross P, Wenger E, Maude RJ, Ghani AC, et al. Role of mass drug administration in elimination of plasmodium falciparum malaria: a consensus modelling study. Lancet Glob Health. 2017; 5(7):e680–7. Received: 1 February 2019 Accepted: 12 September 2019 for different coverage levels of each intervention. Figure S4. Relative maximum prevalence reached depending on intervention coverage. Figure S5. Impact of mass vaccination compared to MDA with same coverage levels comparing strategies 3 and 4 with strategy 1 and 2. Table S3. Risk of resurgence [%] for different deployment strategies. Table S4. Risk of resurgence [%] for different deployment strategies, with reduced coverage of mass vaccination. Table S5. Predicted risk of resurgence [%] for different strategies at low prevalence levels (PfPR2–10 1 to 5%) for different levels of case management and intervention coverage. Figure S6. Interruption of transmission for different strategies with lower case management. Figure S7. Interruption of transmission and synergism for different combined strategies with simultaneous MDA and mass vaccination interventions delivered to the same proportion of the population given coverage. Figure S8. Interruption of transmission for mass vaccination and combined strategies with lower vaccine efficacy. Figure S9. Synergy coefficient between mass vaccination and mass drug administration in the probability to interrupt transmission. Table S6. Estimated resurgence parameters for 2 years deployment of MDA, mass vaccination, or combination of both MDA and mass vaccination, including 2 rates of imported infections. Figure S10. Relationship between entomological inoculation rate, EIR, and effective access to care, E14, with PfPR2–10 and prevalence in all population. Acknowledgements 11. Ishizuka AS, Lyke KE, DeZure A, Berry AA, Richie TL, Mendoza FH, et al. Protection against malaria at 1 year and immune correlates following PfSPZ vaccination. Nat Med. 2016;22(6):614–23. We acknowledge and thank Tom Smith and our colleagues in the Infectious Disease Modeling unit of the Swiss Tropical and Public Health Institute for fruitful discussions that led to improvement of this work, and also acknowledge the helpful discussions with PATH’s Malaria Vaccine Initiative (PATH/MVI) and the Bill & Melinda Gates Foundation. Calculations were performed at sciCORE (http://scicore.unibas.ch/) scientific computing core facility at University of Basel. We thank Philip A. Bejon and Thomas Eisele for reviewing the manuscript and their insightful comments. 12. Collins KA, Snaith R, Cottingham MG, Gilbert SC, Hill AVS. Enhancing protective immunity to malaria with a highly immunogenic virus-like particle vaccine. Sci Rep. 2017;7(March):1–15. 13. Penny MA, Verity R, Bever CA, Sauboin C, Galactionova K, Flasche S, et al. Public health impact and cost-effectiveness of the RTS,S/AS01 malaria vaccine: a systematic comparison of predictions from four mathematical models. Lancet 2016;6736(15):1–9. 14. Greenwood B, Dicko A, Sagara I, Zongo I, Tinto H, Cairns M, et al. Seasonal vaccination against malaria: a potential use for an imperfect malaria vaccine Malar J. 2017;16(1):182. 14. Greenwood B, Dicko A, Sagara I, Zongo I, Tinto H, Cairns M, et al. Seasonal vaccination against malaria: a potential use for an imperfect malaria vaccine. Malar J. 2017;16(1):182. Consent for publication Not applicable Not applicable. 21. Canavati S, Chea N, Guyant P, Roca-feltrer A, Yeung S. Strategy to Address Migrant and Mobile Populations for Malaria Elimination in Cambodia. National Center for Parasitology, Entomology and Malaria Control (CNM). Phnom Penh. 2013. Availability of data and materials h l d d The simulation datasets generated and analyzed during the current study are available from the corresponding author on request. 18. Poirot E, Skarbinski J, Sinclair D, Kachur SP, Slutsker L, Hwang J. Mass drug administration for malaria. Cochrane Database Syst Rev. 2013;12(12): CD008846. 18. Poirot E, Skarbinski J, Sinclair D, Kachur SP, Slutsker L, Hwang J. Mass drug administration for malaria. Cochrane Database Syst Rev. 2013;12(12): CD008846. Ethics approval and consent to participate Not applicable. 19. Zambia National Malaria Elimination Centre. National Malaria Elimination Strategic Plan 2017-2021 Moving from accelerated burden reduction to malaria elimination in Zambia 2017. 19. Zambia National Malaria Elimination Centre. National Malaria Elimination Strategic Plan 2017-2021 Moving from accelerated burden reduction to malaria elimination in Zambia 2017. 20. World Health Organization. Strategy for malaria elimination in the greater Mekong subregion (2015-2030). J Chem Inf Model. 2015;53(9):1689–99. 20. World Health Organization. Strategy for malaria elimination in the greater Mekong subregion (2015-2030). J Chem Inf Model. 2015;53(9):1689–99. Publisher’s Note 24. Smith T, Killeen GF, Maire N, Ross A, Molineaux L, Tediosi F, et al. Mathematical modeling of the impact of malaria vaccines on the Clinical epidemiology and natural history of plasmodium falciparum malaria : overview. Am J Trop Med Hyg. 2006;75(Suppl 2):1–10. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. overview. Am J Trop Med Hyg. 2006;75(Suppl 2):1–10. 25. Smith T, Ross A, Maire N, Chitnis N, Studer A, Hardy D, et al. Ensemble modeling of the likely public health impact of a pre-erythrocytic malaria vaccine. PLoS Med. 2012;9(1):e1001157. https://doi.org/10.1371/journal. pmed.1001157. 26. Chitnis N, Hardy D, Smith T. A periodically-forced mathematical model for the seasonal dynamics of malaria in mosquitoes. Bull Math Biol. 2012;74(5):1098–124. 27. Smith T, Maire N, Dietz K, Killeen GF, Vounatsou P, Molineaux L, et al. Relationship between the entomologic inoculation rate and the. Trop Med. 2006;75(Suppl 2):11–8. 28. Maire N, Smith T, Ross A, Owusu-Agyei S, Dietz K, Molineaux L. A model for natural immunity to asexual blood stages of plasmodium falciparum malaria in endemic areas. Am J Trop Med Hyg. 2006;75(SUPPL. 2):19–31. 29. Ross A, Killeen G, Smith T. Relationships between host infectivity to mosquitoes and asexual parasite density in. Trop Med. 2006;75(Suppl 2):32–7. 30. Ross A, Maire N, Molineaux L, Smith T. An epidemiologic model of severe morbidity and mortality caused by plasmodium falciparum. Am J Trop Med Hyg. 2006;75(2 Suppl):63–73. y 31. Smith T, Ross A, Maire N, Rogier C, Molineaux L. An epidemiologic model of the incidence of acute illness in. Trop Med. 2006;75(Suppl 2):56–62. 32. Galactionova K, Tediosi F, De Savigny D, Smith T, Tanner M. Effective coverage and systems effectiveness for malaria case management in sub- Saharan African countries. PLoS One. 2015;10(5):1–21. 33. RTSS Clinical Trials Partnership. Efficacy and safety of RTS,S/AS01 malaria vaccine with or without a booster dose in infants and children in Africa: Final results of a phase 3, individually randomised, controlled trial. Lancet. 2015;386(9988):31–45. 34. Zani B, Gathu M, Donegan S, Olliaro PL, Sinclair D. Dihydroartemisinin- piperaquine for treating uncomplicated plasmodium falciparum malaria. Cochrane Database Syst Rev. 2014;1:CD010927. 35. Eisele TP, Bennett A, Silumbe K, Finn TP, Chalwe V, Kamuliwo M, et al. Short- term impact of mass drug administration with dihydroartemisinin plus piperaquine on malaria in southern province Zambia: a cluster-randomized controlled trial. J Infect Dis. 2016;214(12):1831–9. 36. Dodd PJ, White PJ, Garnett GP. Competing interests 22. Penny MA, Maire N, Studer A, Schapira A, Smith TA. What should vaccine developers ask? Simulation of the effectiveness of malaria vaccines. PLoS ONE. 2008;3(9):e3193. https://doi.org/10.1371/journal.pone.0003193. 22. Penny MA, Maire N, Studer A, Schapira A, Smith TA. What should vaccine developers ask? Simulation of the effectiveness of malaria vaccines. PLoS ONE. 2008;3(9):e3193. https://doi.org/10.1371/journal.pone.0003193. Author details 1Swiss Tropical and Public Health Institute, Basel, Switzerland. 2University of Basel, Basel, Switzerland. 3PATH’s Malaria Vaccine Initiative, Washington, DC, USA. 23. OpenMalaria Microsimulation [Internet]. [cited 2016 Jul 1]. Available from: https://github.com/SwissTPH/openmalaria/wiki 23. OpenMalaria Microsimulation [Internet]. [cited 2016 Jul 1]. Available from: https://github.com/SwissTPH/openmalaria/wiki Page 15 of 15 Page 15 of 15 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Camponovo et al. BMC Infectious Diseases (2019) 19:920 Publisher’s Note Notions of synergy for combinations of interventions against infectious diseases in heterogeneously mixing populations. Math Biosci. 2010;227(2):94–104. 37. Gosling R, von Seidlein L. The Future of the RTS,S/AS01 Malaria Vaccine: An Alternative Development Plan. PLoS Med. 2016;13(4):1–6. 37. Gosling R, von Seidlein L. The Future of the RTS,S/AS01 Malaria Vaccine: An Alternative Development Plan. PLoS Med. 2016;13(4):1–6. 38. Glopbal Malaria Programme. WHO Malaria Policy Advisory Committee (MPAC) meeting. World Health Organisation; 2019 Apr. 38. Glopbal Malaria Programme. WHO Malaria Policy Advisory Committee (MPAC) meeting. World Health Organisation; 2019 Apr. 39. Okell LC, Pemberton-Ross P, Wenger EA, Maude RJ, Brady OJ, Slater HC, et al. Consensus modelling evidence to support the design of mass drug administration programmes. World Health Organization. 2015. 39. Okell LC, Pemberton-Ross P, Wenger EA, Maude RJ, Brady OJ, Slater HC, et al. Consensus modelling evidence to support the design of mass drug administration programmes. World Health Organization. 2015. 40. York A. Seasonal malaria chemoprevention in the Sahel. Lancet Infect Dis. 2017;17(6):588. 40. York A. Seasonal malaria chemoprevention in the Sahel. Lancet Infect Dis. 2017;17(6):588. 41. Isaïe Medah. Proud to Protect Burkinabè from Meningitis | | Blogs | CDC [Internet]. [cited 2019 Jul 17]. Available from: https://blogs.cdc.gov/global/2 018/04/20/proud-to-protect-burkinabe-from-meningitis/ 41. Isaïe Medah. Proud to Protect Burkinabè from Meningitis | | Blogs | CDC [Internet]. [cited 2019 Jul 17]. Available from: https://blogs.cdc.gov/global/2 018/04/20/proud-to-protect-burkinabe-from-meningitis/ 42. Department of immunization vaccines and biologicals. Assessing the programmatic suitability of vaccine candidates for WHO prequalification [Internet]. World Health Organization; 2015 Jan [cited 2019 Jul 17]. Available from: https://apps.who.int/iris/bitstream/handle/10665/148168/WHO_ IVB_14.10_eng.pdf?sequence=1 43. von Seidlein L, Hanboonkunupakarn B, Jittamala P, Pongsuwan P, Chotivanich K, Tarning J, Hoglund RM, Winterberg M, Mukaka M, Peerawaranun P, Sirithiranont P, Doran Z, Ockenhouse CF, Ivinson K, Lee C, Birkett AJ, Kaslow DC, Singhasivanon P, Day NPJ, Dondorp AM, White NJ, Pukrittayakamee S. Combining antimalarial drugs and vaccine for malaria elimination campaigns: a randomized safety and immunogenicity trial of RTS,S/AS01 administered with dihydroartemisinin, piperaquine, and primaquine in healthy Thai adult volunteers, Human Vaccines & Immunotherapeutics. 2019. https://doi.org/10.1080/21645515.2019.1643675. 44. Kuehne A, Tiffany A, Lasry E, Janssens M, Besse C, Okonta C, et al. Impact and lessons learned from mass drug administrations of malaria chemoprevention during the ebola outbreak in Monrovia, Liberia, 2014. PLoS One. 2016;11(8):1–17.
https://openalex.org/W2776049597
https://ejournal.upi.edu/index.php/mimbar/article/download/8411/pdf
Indonesian
null
Pengembangan Perangkat Pembelajaran IPA Berorientasi pada Taxonomy for Science Education di Sekolah Dasar
Mimbar Sekolah Dasar/Mimbar sekolah dasar
2,017
cc-by-sa
4,707
Imanuel Sairo Awang1 & Andri2 This research is aim to develop learning kits of natural science that taxonomy for science education oriented in elementary school. The kits consist of domain I (imagine and understanding), domain II (science process skills), domain III (imagine and creativity), domain IV (attitude and value), and domain V (conection and applying). This is a research and development method, with subjects were both of fourth grade elementary school students of SD negeri 20 Mambok and SD Negeri 5 Sintang. The results of evaluation of learning science experts to examine the validity of the learning kits stated valid. Then, the trial results showed that the learning kits was developed to meet the criteria of effective to use in elementary school. It concludes that the learning kits of natural science that taxonomy for science education oriented is able to applied in elementary school. Keywords: learning kits; taxonomy for science education; elementary school Selama ini pembelajaran hanya berfokus pada kognitif siswa. Sementara siswa harus menerima materi secara komprehensif. Penelitian ini bertujuan untuk mengembangkan perangkat pembelajaran IPA yang berorientasi pada taxonomy for science education. Di mana taxonomy for science education meliputi lima ranah hasil belajar yaitu Ranah I (Knowing and Understanding); Ranah II (Science Process Skills); Ranah III (Imagine and Creativity); Ranah IV (Attitude and Value); dan Ranah V (Conection and Applying). Penelitian ini dilaksanakan dengan metode research and development, dengan siswa kelas IV SD Negeri 20 Mambok dan siswa kelas IV SD Negeri 5 Sintang sebagai subjek ujicoba. Hasil evaluasi oleh ahli menunjukkan bahwa perangkat pebelajaran berorientasi pada taxonomy for science education valid untuk digunakan. Sedangkan hasil ujicoba menunjukkan bahwa perangkat pebelajaran berorientasi pada taxonomy for science education efektif digunakan di sekolah dasar. Simpulan dari penelitian ini adalah perangkat pembelajaran ipa berorientasi pada taxonomy for science education layak digunakan di sekolah dasar. Kata Kunci: perangkat pembelajaran; taxonomy for science education; sekolah dasar How to Cite: Awang, I., & Andri, A. (2017). Pengembangan Perangkat Pembelajaran IPA Berorientasi pada Taxonomy for Science Education di Sekolah Dasar. Mimbar Sekolah Dasar, 4(3), 192–202. http://doi.org/10.17509/mimbar-sd.v4i3.8411. ABSTRACT The learning method has been focused on student cognitive. In the mean time, student should accept comprehensively. This research is aim to develop learning kits of natural science that taxonomy for science education oriented in elementary school. p-ISSN 2355-5343 e-ISSN 2502-4795 http://ejournal.upi.edu/index.php/mimbar p-ISSN 2355-5343 e-ISSN 2502-4795 http://ejournal.upi.edu/index.php/mimbar p-ISSN 2355-5343 e-ISSN 2502-4795 http://ejournal.upi.ed p-ISSN 2355-5343 e-ISSN 2502-4795 htt // j l Article Received: 09/10/2017; Accepted: 28/12/2017 Mimbar Sekolah Dasar, Vol 4(3) 2017, 192-202 DOI: 10.17509/mimbar-sd.v4i3.8411 http://ejournal.upi.edu/index.php/mimbar How to Cite: Awang, I., & Andri, A. (2017). Pengembangan Perangkat Pembelajaran IPA Berorientasi pada Taxonomy for Science Education di Sekolah Dasar. Mimbar Sekolah Dasar, 4(3), 192–202. http://doi.org/10.17509/mimbar-sd.v4i3.8411. Pengembangan Perangkat Pembelajaran IPA Berorientasi pada Taxonomy for Science Education di Sekolah Dasar Imanuel Sairo Awang1 & Andri2 1,2STKIP Persada Khatulistiwa Sintang 1,2Jl. Pertamina-Sengkuang, Sintang 1Email: imanuel.s.a@stkippersada.ac.id 2Email: andri@stkippersada.ac.id Imanuel Sairo Awang1 & Andri2 The kits consist of domain I (imagine and understanding), domain II (science process skills), domain III (imagine and creativity), domain IV (attitude and value), and domain V (conection and applying). This is a research and development method, with subjects were both of fourth grade elementary school students of SD negeri 20 Mambok and SD Negeri 5 Sintang. The results of evaluation of learning science experts to examine the validity of the learning kits stated valid. Then, the trial results showed that the learning kits was developed to meet the criteria of effective to use in elementary school. It concludes that the learning kits of natural science that taxonomy for science education oriented is able to applied in elementary school. ABSTRAK Selama ini pembelajaran hanya berfokus pada kognitif siswa. Sementara siswa harus menerima materi secara komprehensif. Penelitian ini bertujuan untuk mengembangkan perangkat pembelajaran IPA yang berorientasi pada taxonomy for science education. Di mana taxonomy for science education meliputi lima ranah hasil belajar yaitu Ranah I (Knowing and Understanding); Ranah II (Science Process Skills); Ranah III (Imagine and Creativity); Ranah IV (Attitude and Value); dan Ranah V (Conection and Applying). Penelitian ini dilaksanakan dengan metode research and development, dengan siswa kelas IV SD Negeri 20 Mambok dan siswa kelas IV SD Negeri 5 Sintang sebagai subjek ujicoba. Hasil evaluasi oleh ahli menunjukkan bahwa perangkat pebelajaran berorientasi pada taxonomy for science education valid untuk digunakan. Sedangkan hasil ujicoba menunjukkan bahwa perangkat pebelajaran berorientasi pada taxonomy for science education efektif digunakan di sekolah dasar. Simpulan dari penelitian ini adalah perangkat pembelajaran ipa berorientasi pada taxonomy for science education layak digunakan di sekolah dasar. Imanuel Sairo Awang1 & Andri2 1,2STKIP Persada Khatulistiwa Sintang 1,2Jl. Pertamina-Sengkuang, Sintang 1Email: imanuel.s.a@stkippersada.ac.id 2Email: andri@stkippersada.ac.id ABSTRACT ABSTRAK The learning method has been focused on student cognitive. In the mean time, student should accept comprehensively. This research is aim to develop learning kits of natural science that taxonomy for science education oriented in elementary school. The kits consist of domain I (imagine and understanding), domain II (science process skills), domain III (imagine and creativity), domain IV (attitude and value), and domain V (conection and applying). This is a research and development method, with subjects were both of fourth grade elementary school students of SD negeri 20 Mambok and SD Negeri 5 Sintang. The results of evaluation of learning science experts to examine the validity of the learning kits stated valid. Then, the trial results showed that the learning kits was developed to meet the criteria of effective to use in elementary school. It concludes that the learning kits of natural science that taxonomy for science education oriented is able to applied in elementary school. Keywords: learning kits; taxonomy for science education; elementary school Selama ini pembelajaran hanya berfokus pada kognitif siswa. Sementara siswa harus menerima materi secara komprehensif. Penelitian ini bertujuan untuk mengembangkan perangkat pembelajaran IPA yang berorientasi pada taxonomy for science education. Di mana taxonomy for science education meliputi lima ranah hasil belajar yaitu Ranah I (Knowing and Understanding); Ranah II (Science Process Skills); Ranah III (Imagine and Creativity); Ranah IV (Attitude and Value); dan Ranah V (Conection and Applying). Penelitian ini dilaksanakan dengan metode research and development, dengan siswa kelas IV SD Negeri 20 Mambok dan siswa kelas IV SD Negeri 5 Sintang sebagai subjek ujicoba. Hasil evaluasi oleh ahli menunjukkan bahwa perangkat pebelajaran berorientasi pada taxonomy for science education valid untuk digunakan. Sedangkan hasil ujicoba menunjukkan bahwa perangkat pebelajaran berorientasi pada taxonomy for science education efektif digunakan di sekolah dasar. Simpulan dari penelitian ini adalah perangkat pembelajaran ipa berorientasi pada taxonomy for science education layak digunakan di sekolah dasar. Kata Kunci: perangkat pembelajaran; taxonomy for science education; sekolah dasar How to Cite: Awang, I., & Andri, A. (2017). Pengembangan Perangkat Pembelajaran IPA Berorientasi pada Taxonomy for Science Education di Sekolah Dasar. Mimbar Sekolah Dasar, 4(3), 192–202. http://doi.org/10.17509/mimbar-sd.v4i3.8411. ABSTRACT ABSTRAK The learning method has been focused on student cognitive. In the mean time, student should accept comprehensively. ABSTRACT Oleh karena itu, pembelajaran yang mengedepankan hasil yang seimbang dan holistik mutlak [193] Mimbar Sekolah Dasar, Volume 4 Nomor 3 Desember 2017 Menurut Sumaji, dkk. (Awang, 2012), “Pembelajaran IPA di SD hendaknya membuka kesempatan untuk memupuk rasa ingin tahu peserta didik secara ilmiah. Dengan demikian, pembelajaran IPA yang disampaikan harus dirancang sedemikian rupa sehingga semua potensi yang terdapat dalam diri siswa dapat terakomodasi. Salah satu pembelajaran yang dapat mengakomodasi seluruh potensi siswa adalah pembelajaran yang berorientasi pada taxonomy for science education. pendidikan sains, yang mampu meningkatkan aktivitas pembelajaran IPA di kelas serta dapat meningkatkan minat siswa terhadap mata pelajaran IPA. Taxonomy for science education (Prasetyo, 1998) meliputi lima ranah hasil belajar yakni Ranah I (Knowing and Understanding); Ranah II (Science Process Skills); Ranah III (Imagine and Creativity); Ranah IV (Attitude and Value); dan Ranah V (Conection and Applying). Ranah I merupakan acuan hasil belajar yang berfokus pada pengetahuan. Pendidikan IPA mempunyai fakta, konsep, prinsip, hukum, dan teori yang harus dikuasai oleh siswa. Melalui penekanan pada ranah ini, siswa diharapkan dapat mengetahui serta memahami pengetahuan IPA. Selanjutnya, ranah II merupakan acuan hasil belajar yang berfokus pada keterampilan proses IPA. Pengetahuan yang berkaitan dengan fakta, konsep, prinsip, hukum, dan teori dalam pendidikan IPA diperoleh melalui suatu proses yakni dengan metode ilmiah. Penerapan metode ilmiah dalam memperoleh pengetahuan dilaksanakan dengan melakukan berbagai kegiatan diantaranya pengamatan, pengukuran, pengelompokkan, pengkomunikasian, sampai pada keterampilan melakukan eksperimen. Melalui penekanan pada ranah ini, siswa diharapkan dapat menguasai keterampilan proses IPA tingkat dasar. Ranah I merupakan acuan hasil belajar yang berfokus pada pengetahuan. Pendidikan IPA mempunyai fakta, konsep, prinsip, hukum, dan teori yang harus dikuasai oleh siswa. Melalui penekanan pada ranah ini, siswa diharapkan dapat mengetahui serta memahami pengetahuan IPA. Selanjutnya, ranah II merupakan acuan hasil belajar yang berfokus pada keterampilan proses IPA. Pengetahuan yang berkaitan dengan fakta, konsep, prinsip, hukum, dan teori dalam pendidikan IPA diperoleh melalui suatu proses yakni dengan metode ilmiah. Penerapan metode ilmiah dalam memperoleh pengetahuan dilaksanakan dengan melakukan berbagai kegiatan diantaranya pengamatan, pengukuran, pengelompokkan, pengkomunikasian, sampai pada keterampilan melakukan eksperimen. Melalui penekanan pada ranah ini, siswa diharapkan dapat menguasai keterampilan proses IPA tingkat dasar. Taxonomy for science education adalah sebuah acuan pembelajaran dalam pendidikan sains, yang mampu meningkatkan aktivitas pembelajaran IPA di kelas dan mengembangkan sikap positif terhadap mata pelajaran itu (Loucks- Horsley, et al., 1990). Taxonomy for science education diyakini merupakan perluasan dari ketiga ranah hasil belajar Bloom. ABSTRACT Keywords: learning kits; taxonomy for science education; elementary school Kata Kunci: perangkat pembelajaran; taxonomy for science education; sekolah dasar peradapan suatu bangsa. Oleh karena itu, usaha untuk menjadikan pendidikan sebagai penyangga utama perkembangan suatu bangsa perlu terus ditingkatkan. PENDAHULUAN ~ Pendidikan adalah unsur penting bagi pengembangan potensi manusia agar siap menghadapi segala permasalahan dalam kehidupannya. Pendidikan mendapat perhatian utama berkaitan dengan pengembangan [192] Imanuel Sairo Awang & Andri, Pengembangan Perangkat Pembelajaran IPA… Peningkatan kualitas pendidikan dapat dilakukan melalui berbagai komponen dalam pembelajaran. Berbagai komponen dalam pembelajaran yang dapat tingkatkan kualitasnya adalah penggunaan media pembelajaran dan sumber belajar yang bervariasi. Selain itu, peningkatan kualitas pembelajaran juga dapat dilakukan dengan cara mengatur suasana pembelajaran yang meliputi pemilihan strategi, pendekatan, metode, teknik, serta taktik yang sesuai agar pembelajaran menjadi lebih efektif dan menyenangkan. diperlukan, dimana tidak hanya ranah kognitif yang dikembangkan, namun juga ranah afektif, dan psikomotorik. Dalam pelaksanaannya, pembelajaran berbasis ranah Bloom masih belum seimbang dan holistik yaitu umumnya hanya menitikberatkan pada tujuan ranah kognitif dan menghindari tujuan ranah afektif (Collete-Chiapetta, dalam Prasetyo, 2008, p. 21). Sebagai akibatnya, pembelajaran menjadi monoton dan tidak memberi peluang siswa berperan aktif. Perluasan dari pembelajaran berbasis ranah Bloom diperlukan untuk mengeksplorasi potensi siswa dimana tidak hanya kemampuan kognitif tetapi juga ranah yang lain, terutama pada mata pelajaran IPA. Pembelajaran yang efektif dan menyenangkan terutama pada jenjang sekolah dasar sangat diperlukan. Hal ini dikarenakan sesuai dengan tingkat perkembangan mental serta emosional siswa. Siswa usia sekolah dasar berada pada tahap perkembangan operasi konkret. Implikasinya adalah guru harus mampu memberikan pengalaman belajar yang nyata agar dapat diterima dengan baik oleh siswa. Ilmu Pengetahuan Alam (IPA) atau science (bahasa Indonesia: sains) diambil dari kata Latin Scientia yang memiliki arti harfiah pengetahuan. Menurut Oxford English Dictionary (Buxton & Provenzo, 2007, p. 6) sains didefinisikan sebagai, “Those branches of study that related to the phenomena of the material universe and their laws”. Pendapat tersebut menjelaskan bahwa IPA merupakan cabang ilmu yang berhubungan dengan fenomena-fenomena yang terdapat di alam serta hukum-hukum yang berlaku padanya. Dengan demikian, pembelajaran IPA di sampaikan di sekolah agar siswa memahami secara baik bagaimana alam semesta berlaku. Pembelajaran yang hanya memfokuskan pada ranah kognitif hanya akan menjauhkan pengetahuan dengan siswa. Siswa dipaksa untuk memahami pengetahuan yang diberikan oleh guru sebagai pemenuhan dari standar kompetensi yang ditetapkan. Sebagai akibatnya pengetahuan yang diterima oleh siswa bersifat sementara. ABSTRACT Taxonomy for science education meliputi lima ranah hasil belajar yakni Ranah I (Knowing and Understanding); Ranah II (Science Process Skills); Ranah III (Imagine and Creativity); Ranah IV (Attitude and Value); dan Ranah V (Conection and Applying) (MacCormack & Yager, dalam Prasetyo, 2008, p. 18). Sesuai dengan uraian tersebut, maka salah satu acuan pembelajaran yang patut dikembangkan adalah pembelajaran yang berorientasi pada taxonomy for science education. Taxonomy for science education adalah sebuah acuan pembelajaran dalam [194] Imanuel Sairo Awang & Andri, Pengembangan Perangkat Pembelajaran IPA… pokok/pembelajaran, kegiatan pembelajaran, indikator pencapaian kompetensi untuk penilaian, alokasi waktu dan sumber belajar (Depdiknas, 2007b, p. 4). Tahapan pengembangan silabus adalah: (1) mengkaji dan menentukan standar kompetensi; (2) mengkaji dan menentukan kompetensi dasar; (3) mengidentifikasi materi pokok pembelajaran; (4) mengembangkan kegiatan pembelajaran; (5) merumuskan indikator pencapaian kompetensi; (6) menentukan jenis penilaian; serta (7) menentukan alokasi waktu, dan (8) menentukan sumber belajar. pokok/pembelajaran, kegiatan pembelajaran, indikator pencapaian kompetensi untuk penilaian, alokasi waktu dan sumber belajar (Depdiknas, 2007b, p. 4). Tahapan pengembangan silabus adalah: (1) mengkaji dan menentukan standar kompetensi; (2) mengkaji dan menentukan kompetensi dasar; (3) mengidentifikasi materi pokok pembelajaran; (4) mengembangkan kegiatan pembelajaran; (5) merumuskan indikator pencapaian kompetensi; (6) menentukan jenis penilaian; serta (7) menentukan alokasi waktu, dan (8) menentukan sumber belajar. Ranah III merupakan acuan hasil belajar yang berfokus pada kemampuan menginterpretasi produk pengetahuan khususnya pengetahuan IPA. Keterampilan ini harus dikuasai agar dapat memahami secara utuh pengetahuan IPA. Sementara ranah IV merupakan acuan hasil belajar yang berfokus pada pengembangan nilai dan sikap dalam belajar IPA. Pengetahuan diperoleh dengan sikap jujur dan teliti. Setelah mengetahui produk IPA, siswa juga diharapkan dapat mengaktualisasi nilai- nilai yang terkandung di dalamnya. Keterampilan ini harus dikuasai agar siswa mampu bersikap positif serta menerapkan nilai-nilai kehidupan dengan baik pula. Kemudian ranah V dari taxonomy for science education adalah acuan hasil belajar yang berfokus pada penerapan dari pengetahuan yang telah diperoleh. Muara dari produk IPA adalah teknologi. Sehingga diharapkan dengan diacunya ranah ini dalam pembelajaran IPA siswa terbiasa untuk menghasilkan sebuah karya sebagai bentuk pemahaman dari pengetahuan IPA. Rencana pelaksanaan pembelajaran (RPP) dijabarkan dari silabus untuk mengarahkan kegiatan belajar peserta didik dalam upaya mencapai kompetensi dasar (Depdiknas, 2007b, p. 4). Komponen substansial RPP meliputi: standar kompetensi, kompetensi dasar, indikator, alokasi waktu, model/strategi/metode pengajaran, kegiatan pembelajaran (disusun tiap pertemuan), alat/bahan, penilaian, dan sumber belajar. RPP disusun untuk setiap kompetensi dasar. Bagian penting lainnya dalam penyusunan RPP adalah indikator. ABSTRACT Indikator merupakan penanda pencapaian kompetensi dasar yang ditandai oleh perubahan perilaku yang dapat diukur sesuai dengan lima domain Ranah I (knowing and understanding); Ranah II (science process skills); Ranah III (imagine and creativity); Ranah IV (attitude and value); dan Ranah V (conection and applying). Perangkat pembelajaran berorientasi taxonomy for science education yang dikembangkan dalam penelitian ini meliputi silabus, rencana pelaksanaan pembelajaran (RPP), lembar kerja siswa (LKS) dan lembar penilaian. Silabus merupakan rencana pembelajaran pada suatu/pokok mata pelajaran tertentu yang mencakup standar kompetensi, kompetensi dasar, materi [195] Mimbar Sekolah Dasar, Volume 4 Nomor 3 Desember 2017 Selanjutnya penyusunan lembar kerja siswa (LKS) biasanya berupa petunjuk, langkah-langkah untuk menyelesaikan suatu tugas sesuai dengan kompetensi dasar. LKS adalah materi ajar yang dikemas secara integrasi sehingga memungkinkan peserta didik mempelajari materi tersebut secara mandiri. LKS digunakan untuk mengoptimalkan hasil belajar peserta didik dan meningkatkan kegiatan belajar peserta didik di sekolah maupun di rumah. pembelajaran yang berorientasi pada taxonomy for science education peneliti ingin melakukan penelitian dengan tujuan untuk mengembangkan perangkat pembelajaran IPA berorientasi pada taxonomy for science education di sekolah dasar. METODE Penelitian ini dilakukan dengan model penelitian dan pengembangan (research and development). Sukmadinata (2006, p. 164) menyatakan bahwa penelitian dan pengembangan adalah suatu proses atau langkah-langkah untuk mengembangkan produk baru atau menyempurnakan produk yang telah ada, yang dapat dipertanggungjawabkan. Langkah- langkah dalam proses penelitian dan pengembangan menunjukkan suatu siklus, yang diawali dengan adanya suatu kebutuhan, sampai permasalahan yang membutuhkan pemecahan dengan menggunakan suatu produk tertentu. Bagian terakhir dari perangkat pembelajaran adalah lembar penilaian. Penilaian adalah proses pengumpulan dan pengolahan informasi untuk menentukan pencapaian hasil belajar peserta didik (Depdiknas, 2007a, p. 3). Tujuan dan fungsi penilaian salah satunya adalah sebagai pengukur keberhasilan. Pada penilaian ini dimaksudkan untuk mengetahui sejauh mana suatu program berhasil diterapkan. Mengingat pendidikan IPA mempunyai kekhususan tersendiri, maka munculnya taxonomy for science education diharapkan dapat mengakomodasi keunikan dari pembelajaran pendidikan IPA. Selain itu, kelima ranah pada taxonomy for science education diharapkan dapat meningkatkan aktivitas dan minat, serta kemampuan kognitif siswa. Terkait dengan masalah di atas, maka untuk mengkaji lebih dalam tentang kelayakan dan keefektifan perangkat pembelajaran setelah melaksakan Langkah-langkah dalam penelitian ini mengikuti langkah dari Borg, Gall, & Gall (2003) yang terdiri dari sembilan langkah dan dimodifikasi dengan model 4-D dari Thiagarajan (Thiagarajan, Semmel, & Semmel, 1974) yakni: (1) studi pendahuluan, (2) perencanaan, (3) mengembangkan produk awal, (4) uji coba awal, (5) revisi produk utama, (6) melakukan uji coba lapangan, (7) revisi terhadap produk operasional, (8) uji coba lapangan produk operasional, dan (9) revisi produk final. [196] Imanuel Sairo Awang & Andri, Pengembangan Perangkat Pembelajaran IPA… Subjek uji coba dalam penelitian ini adalah siswa kelas IV SD Negeri 20 Mambok, Sintang. Sedangkan subjek penelitian uji coba lapangan adalah siswa kelas IV SD Negeri 5 Sintang. Adapun intrumen yang digunakan yakni lembar validasi, tes hasil belajar, lembar pengamatan, dan portofolio. Lembar validasi digunakan untuk menilai kelayakan perangkat pembelajaran yang dikembangkan. Sedangkan lembar tes, lembar pengamatan, dan portofolio siswa digunakan untuk menilai tingkat keefektifan penerapan perangkat pembelajaran yang dikembangkan. lembar pengamatan, dan portofolio. Kriteria yang ditentukan adalah, apabila terdapat paling sedikit 80% ketuntasan maka, dinyatakan efektif. HASIL Pada tahap awal, dilakukan perancangan perangkat pembelajaran yang terdiri dari silabus, rencana pelaksanaan pembelajaran (RPP), lembar kerja siswa (LKS) dan lembar penilaian. Semua unsur perangkat pembelajaran ini dirancang dengan berorientasi pada taxonomy for science education. Setelah dirancang, langkah selanjutnya adalah melakukan tahapan validasi yang dilakukan oleh ahli pembelajaran IPA berorientasi taxonomy for science education. Penilaian dilakukan dengan cara mengisi lembar penilaian pada masing-masing komponen. Hasil rekapitulasi penilaian dari ahli pembelajaran IPA secara lengkap dapat dilihat pada Tabel 1. Analisis data untuk mengetahui kevalidan dilakukan dengan melakukan analisis kelayakan oleh ahli pembelajaran IPA tentang taxonomy for science education. Proses ini dilaksanakan dengan meminta ahli mengisi lembar penilaian yang telah disiapkan. Setelah itu dilakukan analisis keefektifan yang dilakukan menggunakan statistik deskriptif terhadap tes hasil belajar, Tabel 1. Data Hasil Penilaian Kevalidan Perangkat Pembelajaran oleh Ahli Pembelajaran IPA No. Komponen Perangkat Pembelajaran Total Skor Aktual (X) Kategori 1. Silabus 41 Baik 2. RPP 71 Baik 3. LKS dan Kunci LKS 36 Baik 4. Lembar Penilaian dan Kunci Valid dengan revisi Baik Tabel 1. Data Hasil Penilaian Kevalidan Perangkat Pembelajaran oleh Ahli Pembelajaran IPA No. Komponen Perangkat Pembelajaran Total Skor Aktual (X) Kategori 1. Silabus 41 Baik 2. RPP 71 Baik 3. LKS dan Kunci LKS 36 Baik 4. Lembar Penilaian dan Kunci Valid dengan revisi Baik Penilaian Kevalidan Perangkat Pembelajaran oleh Ahli Pembelajaran IPA Tabel 1. Data Hasil Penilaian Kevalidan Perangkat Pembelajaran oleh Ahli Pembelajaran IPA Berdasarkan penilaian dari ahli pembelajaran IPA, beberapa aspek perlu direvisi. Aspek tersebut meliputi konstruk pada LKS dan lembar penilaian. Selain itu, pada teknik pelaksanaan pembelajaran, perlu ditekankan pada proses pembimbingan yang intensif agar siswa memahami perintah di LKS dengan baik. Setelah direvisi selanjutnya dilakukan ujicoba terbatas di Sekolah Dasar Negeri No. 20 Mambok. Ujicoba dilakukan untuk melihat keefektifan pelaksanaan [197] Mimbar Sekolah Dasar, Volume 4 Nomor 3 Desember 2017 berorientasi taxonomy for science education pada ujicoba terbatas dapat dilihat pada Tabel 2. pembelajaran berorientasi taxonomy for science education. Hasil penilaian keefektifan perangkat pembelajaran Tabel 2. Rekapitulasi Hasil Penilaian Keefektifan Perangkat Pembelajaran pada Ujicoba Terbatas No. Komponen Keefektifan Rerata Total Skor Aktual Ketuntasan (%) Kategori 1. Hasil Tes Siswa 87,64 100 Sangat Baik 2. Lembar Pengamatan 63,18 92,92 Sangat Baik 3. Penilaian Portofolio 78,8 48,15 Kurang abel 2. Rekapitulasi Hasil Penilaian Keefektifan Perangkat Pembelajaran pada Ujicoba Terbatas berarti. Revisi yang dilakukan yakni, pada rencana pelaksanaan pembelajaran (RPP), pemilihan contoh benda yang mempunyai tiga wujud dilakukan dengan lebih banyak. HASIL Kelima ranah tersebut yakni ranah I (knowing and understanding); ranah II (science process skills); ranah III (imagine and creativity); ranah IV (attitude and value); dan ranah V (conection and applying). Perangkat pembelajaran merupakan hal penting dan mendasar yang harus disiapkan oleh seorang guru ketika akan melaksanakan kegiatan pembelajaran. Perangkat pembelajaran yang baik adalah perangkat pembelajaran yang disusun berdasarkan kondisi serta potensi yang dimiliki oleh lingkungan pembelajaran. Selain itu, perangkat pembelajaran yang baik juga dapat dilihat dari kesesuaian antara perangkat yang satu dengan perangkat dukungan yang lain. Dalam penelitian ini telah dirancang perangkat pembelajaran yakni silabus, rencana pelaksanaan pembelajaran (RPP), lembar kerja siswa (LKS), dan lembar penilaian, di mana semua perangkat tersebut berorientasi pada taxonomy for science education. Taksonomi pembelajaran sains ini diyakini merupakan pengembangan dari tiga ranah taksonomi pembelajaran Bloom. Kelima ranah hasil belajar sains ini mencoba untuk mengembangkan berbagai keterampilan yang harus dikuasai oleh seseorang yang belajar sains. Sehingga, apabila seseorang atau siswa yang belajar sains, tidak hanya befokus pada pencapaian tujuan kognitifnya, melainkan juga menguasai hal-hal yang berkaitan dengan pengusaan kognitif tersebut. HASIL berarti. Revisi yang dilakukan yakni, pada rencana pelaksanaan pembelajaran (RPP), pemilihan contoh benda yang mempunyai tiga wujud dilakukan dengan lebih banyak. Hasil penilaian pada Tabel 2, menunjukkan bahwa pada komponen tes siswa, rerata hasil tes siswa sudah melampaui kriteria yang ditentukan yakni kriteria ketuntasan minimal sebesar 80. Walaupun rata-rata siswa sudah mencapai kriteria yang ditentukan, namun masih terdapat beberapa siswa saat dilaksanakan pengamatan pada saat kegiatan pembelajaran, masih belum optimal dalam melaksanakan tugas yang diberikan guru. Hasil pengamatan terhadap aktivitas siswa secara berkelompok memperlihatkan persentase sebesar 92,92%. Artinya secara umum siswa mampu menunjukkan sikap dan prilaku yang mendukung pelaksanaan pembelajaran taxonomy for science education. Pada materi tentang “benda dapat melarutkan benda lain”, guru diminta untuk dapat memberikan lebih banyak contoh dan bimbingan yang intensif kepada siswa. Dengan demikian diharapkan mereka mampu menyelesaikan tugas di LKS dengan baik. Secara teknik penerapannya juga dilakukan penjelasan dengan lambat. Sehingga siswa dapat mengikuti dan memahami dengan baik. Pada lembar kerja siswa (LKS), hanya direvisi pada tahap penerapannya. Siswa harus dibimbing secara pelan-pelan agar mereka dapat mengikuti arahan dengan baik. Mengingat masih terdapat berbagai kendala dalam pelaksanaan pembelajarannya, maka dilakukan revisi pada beberapa bagian terhadap perangkat pembelajarannya. Selain itu, pada saat penerapannya di kelas juga harus diperhatikan teknik penyampaian agar siswa tidak menemui kendala yang Setelah dilakukan revisi, maka dilanjutkan dengan melaksanakan ujicoba terakhir yakni ujicoba lapangan. Ujicoba lapangan untuk mengetahui keefektifan perangkat pembelajaran berorientasi [198] Imanuel Sairo Awang & Andri, Pengembangan Perangkat Pembelajaran IPA… taxonomy for science education ini dilaksanakan di Sekolah Dasar Negeri 5 Sintang. Hasil ujicoba lapangan perangkat pembelajaran berorientasi taxonomy for science education dapat dilihat pada Tabel 3. Tabel 3. Rekapitulasi Hasil Penilaian Keefektifan Perangkat Pembelajaran pada Ujicoba Lapangan No. Komponen Keefektifan Rerata Skor Aktual Ketuntasan (%) Kategori 1. Hasil Tes Siswa 88,34 100 Sangat baik 2. Lembar Pengamatan 61,75 90,81 Sangat baik 3. Penilaian Portofolio 81,70 81,25 Sangat baik Rekapitulasi Hasil Penilaian Keefektifan Perangkat Pembelajaran pada Ujicoba Lapangan Tabel 3. Rekapitulasi Hasil Penilaian Keefektifan Perangkat Pembelajaran pada Ujicoba La terutama pada mata pelajaran Ilmu Pengetahuan Alam (IPA) menjadi lima bagian (ranah). Kelima ranah tersebut yakni ranah I (knowing and understanding); ranah II (science process skills); ranah III (imagine and creativity); ranah IV (attitude and value); dan ranah V (conection and applying). terutama pada mata pelajaran Ilmu Pengetahuan Alam (IPA) menjadi lima bagian (ranah). PEMBAHASAN Taxonomy for science education pertama kali dikembangkan oleh MacCormack & Yager (Prasetyo, 2008, p. 18), yang merupakan pengembangan hasil belajar dari taksonomi Bloom. Taxonomy for science education merupakan sebuah konsep pengelompokkan hasil belajar Penelitian ini bertujuan untuk mengetahui karateristik pembelajaran berorientasi pada taxonomy for science education. Selain itu, penelitian ini juga ingin mengetahui tingkat kelayakan dan [199] Mimbar Sekolah Dasar, Volume 4 Nomor 3 Desember 2017 Pada pembelajaran yang berorientasi pada taxonomy for science education, selain mengembangkan kemampuan berpikir (aplikasi ranah I), siswa juga diajak untuk mengeksplorasi materi melalui kegiatan dengan melibatkan keterampilan proses sains (aplikasi ranah II). Keterampilan proses sains merupakan keterampilan yang harus dimiliki agar mereka dapat melaksanakan aktivitas keilmuan melalui metode ilmiah. Melalui kegiatan pengamatan, pengelompokkan, diskusi (berkomunikasi), menyimpulkan, dan presentasi, siswa mampu membangun konsep pemahaman sesuai dengan pengalaman belajar yang ia miliki. keefektifan pembelajaran dengan menggunakan perangkat pembelajaran berorientasi pada taxonomy for science education. Pada langkah awal, dirancang perangkat pembelajaran berorientasi pada taksonomi for science education dengan mengintegrasikan silabus, RPP, LKS, dan lembar penilaian. Berdasarkan hasil validasi dari ahli pembelajaran IPA dapat diketahui bahwa rata-rata skor setiap komponen perangkat pembelajaran berada pada kategori “baik” dengan rincian hasil penilaian silabus berkategori “baik”; hasil penilaian RPP berkategori “baik”; hasil penilaian LKS berkategori “baik”; dan hasil penilaian lembar penilaian berkategori “baik”. Hal ini mengindikasikan bahwa meskipun terdapat berbagai revisi berdasarkan saran dan masukan dari validator, produk awal perangkat pembelajaran berorientasi pada taxonomy for science education tersebut sudah layak digunakan pada ujicoba. Hasil pengamatan juga menunjukkan bahwa ranah IV taxonomy For Science Education yakni attitude and value berkembang dengan baik. Ini dapat dilihat dari pelaksanaan pembelajaran baik secara pribadi maupun berkelompok mereka menunjukkan sikap rasa ingin tahu, sikap berpikir kritis, dan sikap berpikiran terbuka dan kerjasama. Perilaku yang menunjukkan sikap rasa ingin tahu, sikap berpikir kritis, dan sikap berpikiran terbuka dan kerjasama tersebut terlihat dari sikap antosias siswa saat kegiatan pembelajaran berlangsung, selain itu, semua siswa menerima pembagian kelompok dan menerima tugas dengan sikap positif dengan rasa ingin tahu nyang tinggi. Hal ini karena masing-masing anggota kelompok bertanggung jawab untuk menyelesaikan tugas yang diberikan Keefektifan perangkat pembelajaran IPA berorientasi pada dilihat dari hasil tes, hasil pengamatan, dan hasil portofolio siswa. Hasil tes diberikan setelah melaksanakan kegiatan pembelajaran, kemudian hasil pengamatan dilaksanakan selama penerapan RPP yakni sebanyak empat kali pertemuan. Sedangkan hasil portofolio siswa diberikan pada akhir pelaksanaan pembelajaran pertemuan keempat. PEMBAHASAN [200] Imanuel Sairo Awang & Andri, Pengembangan Perangkat Pembelajaran IPA… penelitian ini adalah: (1) perangkat pembelajaran IPA berorientasi pada taxonomy for science education mempunyai karakteristik yakni perangkat pembelajaran IPA yang dikembangkan terdiri dari silabus, rencana pelaksanaan pembelajaran (RPP), lembar kerja siswa (LKS), dan lembar penilaian; (2) Perangkat pembelajaran IPA berorientasi pada Taxonomy For Science Education dengan lima ranah hasil belajar yaitu Ranah I (knowing and understanding); Ranah II (science process skills); Ranah III (imagine and creativity); Ranah IV (attitude and value); dan Ranah V (conection and applying); dan (3) Perangkat pembelajaran IPA berorientasi pada taxonomy for science education efektif diterapkan pada pembelajaran IPA di sekolah dasar. dalam LKS. Dalam penelitian ini, sebagaimana yang dilaporkan Khotimah & Suliyanah (2017), optimalisasi LKS akan turut memberikan pengaruh yang besar dalam meningkatkan kemampuan belajar IPA siswa. Pada hasil penilaian portofolio siswa menunjukkan bahwa, ranah III dan ranah V yakni (imagine and creativity) dan (conection and applying) berkembang dengan baik. Siswa mampu mengembangkan daya khayalnya (imagine) dalam membuat tiruan model yang diminta. Dari khayalannya, tercipta sebuah kreativitas (creativity). Kemudian, dari kreativitas tersebut, mereka mampu menghubungkan materi yang diperlajari (conection) serta dapat menerapkan konsep-konsep yang telah dipelajarinya (applying). REFERENSI Secara umum berdasarkan hasil analisis dan pembahasan yang telah diuraikan, maka dapat diambil kesimpulan bahwa perangkat pembelajaran IPA berorientasi pada taxonomy for science education yang telah dikembangkan efektif digunakan pada proses pembelajaran IPA di sekolah dasar, terlebih karena perangkat pembelajaran yang dikembangkan lebih berorientasi pada pemecahan masalah (Sulistiono & Rahayu, 2014). Awang, I.S. (2012). Pengembangan Subject Specific Pedagogy (SSP) Mata Pelajaran IPA Untuk Mengembangkan Karakter Peserta Didik Sekolah Dasar. Jurnal Pendidikan Vox Edukasi. 3 (1). 1- 10. Borg, W.R., Gall, M.D., & Gall. J.P. (2003). Educational research: an introduction. (7th Ed). New York. Pearson Education. Inc. Buxton, C.A. & Provenzo, E.F. (2007). Teaching science in elementary and middle school: a cognitive and cultural approach. Los Angeles. Sage Publications, Inc. Depdiknas. (2007a). Peraturan Menteri Pendidikan Nasional Republik Indonesia Nomor 20, Tahun 2007, tentang Standar Penilaian Pendidikan. SIMPULAN Berdasarkan hasil analisis data dan pembahasan, maka kesimpulan dari [201] Mimbar Sekolah Dasar, Volume 4 Nomor 3 Desember 2017 Depdiknas. (2007b). Peraturan Menteri Pendidikan Nasional Republik Indonesia Nomor 41, Tahun 2007, tentang Standar Proses untuk Satuan Pendidikan Dasar dan Menengah. Khotimah, P.C. & Suliyanah. (2017). Pengembangan lembar kerja siswa (LKS) untuk melatihkan keterampilan berpikir kritis peserta didik SMAN 4 Sidoarjo pada materi kalor. Inovasi Pendidikan Fisika (JIPF), 6(03), 295-300. Loucks-Horsley, S., et al. (1990). Elementary school science for the ’90’s. Andover, MA. Network. Prasetyo, Z. K. (2008). Kontribusi pendidikan sains dalam pengembangan moral peserta didik. Pidato Pengukuhan Guru Besar dalam Bidang Ilmu Pendidikan IPA pada Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Negeri Yogyakarta. 5 Januari 2008, Yogyakarta, Indonesia. Prasetyo, Z.K. (1998). Taksonomi untuk pendidikan fisika (sains) Yogyakarta: Cakrawala Pendidikan Majalah Ilmiah Kependidikan. Edisi Khusus Dies. 146- 151. Sukmadinata, N. S. (2006). Metodologi penelitian pendidikan. Bandung. PT Remaja Rosdakarya. Sulistiono, E., & Rahayu, Y. S. (2014). PENINGKATAN KEMAMPUAN BERPIKIR KRITIS SISWA MENGGUNAKAN PERANGKAT PEMBELAJARAN IPA SMP BERORIENTASI PENYELESAIAN MASALAH. Jurnal Pena Sains, 1(2), 46-55. Thiagarajan, S., Semmel, D.S., & Semmel, M.I. (1974). Instructional development for training teachers of exceptional children A Sourcebook. Bloomington, Indiana. Indiana University. [202]
https://openalex.org/W2903385063
https://www.intechopen.com/citation-pdf-url/62071
English
null
Housing Policy in the Slovak Republic
IntechOpen eBooks
2,018
cc-by
7,994
Daniela Spirkova Daniela Spirkova Additional information is available at the end of the chapter Additional information is available at the end of the chapter Additional information is available at the end of the chapter http://dx.doi.org/10.5772/intechopen.78611 Additional information is available at the end of the chapter Additional information is available at the end of the chapter Additional information is available at the end of the chapter http://dx.doi.org/10.5772/intechopen.78611 http://dx.doi.org/10.5772/intechopen.78611 Abstract Housing is an important source of employment. Consistency between housing and employment in several developed economies has become an essential part of state public policy and local governments. Low labour mobility in Slovakia is also related to the struc- ture of use of the housing stock, which is significantly dominated by owner-occupied housing. The mobility of housing is closely linked to the desired and undesired labour mobility. The demand-determining factor on the side of the housing market is the demo- graphic development of the population from which the need for housing is objectively derived. The real demand for housing is mainly dependent on the social-economic status of potential applicants for housing and their current and expected financial situation. The chapter discusses the historical context of housing and housing policy in Slovakia, “a massive privatisation of flats” and the housing stock in Slovakia at the beginning of 1990s, the problems and the causal relationship between the housing market, labour mobility and housing finance in Slovakia. Keywords: housing policy, labour mobility, rental housing, economic tools, Slovakia © 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons ttribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, istribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, Selection of our books indexed in the Book Citation Index in Web of Science™ Core Collection (BKCI) Interested in publishing with us? Contact book.department@intechopen.com Numbers displayed above are based on latest data collected. For more information visit www.intechopen.com Open access books available Countries delivered to Contributors from top 500 universities International authors and editors Our authors are among the most cited scientists Downloads We are IntechOpen, the world’s leading publisher of Open Access books Built by scientists, for scientists 14% 191,000 210M TOP 1% 154 7,200 Chapter 2 © 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ttribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, stribution, and reproduction in any medium, provided the original work is properly cited. Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided the original work is properly cited 1. Introduction Housing policy is one of the most important areas of social policy. It is a mirror of social defi- ciencies and changes. Housing is bound to a specific location in space. In this context, when considering whether to invest, households are influenced by several factors: the availability of jobs, the income of households, interest rate of housing credits, the labour market status and the amount of indebtedness. The Slovak Republic is one of the Central European countries whose economies were cen- trally planned until 1989. The main characteristic of transition countries is the transition state of their economies—transitive stage—from a centrally controlled system to market economy. © 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Housing 22 Housing 22 Until 1989, housing policy in the Slovak Republic was centralised. Moreover, the state pro- vided housing for all citizens. The state provided building construction centrally, variedly subsidised all forms of housing (mainly subsidising the price of services related to housing, as prices of energies and water), ran housing stock, especially state and cooperative flats, and centralised the administrative system of the redistribution of flats. Therefore, political changes in the Slovak Republic after 1989 required several fundamental changes in the field of institu- tional arrangement of the public administration, as well as legislative changes. Similarly, as in other fields of economics, also in the sphere of housing, transformation from the rationing system of housing to the system based on the respected market economy was initiated. In 1993 the independent Slovak Republic was established, and therefore other inevitable changes occurred in the field of housing and housing policy. One of the important decisions was the division of tasks in the frame of housing policy to three levels (central, regional and local). The state stayed a valuable player, who, however, stopped being involved in hous- ing construction but took over the task of a creator of frame documents, via which the state tried to define basic principles and rules, in which the housing policy of Slovakia further moved. 1. Introduction A significant impact of the development of housing in the years 1992–2006 was done by a process of massive and inappropriate privatisation of the housing stock, which caused a subsequent decrease in rental flats in Slovakia. Overall, between the years 1992 and 2006 according to OECD in Slovakia, approximately 340ths of council flats and 270ths of coopera- tive flats were privatised. Decentralisation was another distinct impact, the part of which was also the transfer of com- petence in the field of housing, territorial planning and local development to the lower man- agement level in the years 2000–2003. It resulted from the assumption that towns and villages will manage to reflect the needs on the local level better. A key player stayed a citizen, who should do his best to provide housing for himself and his family. Quickly this imagination was shown as illusory especially from the view of possibilities and the ability of citizens and especially of specific groups to provide housing exclusively with their own contribution. According to first conceptions, the role of the state in the field of housing should have been limited just to the creation of conditions for origination of the real estate market. In regard to vigorous withdrawal of the state from financial participation in new housing construction at the beginning of the 1990s (the period of transformation process), the creation of conditions for the implementation of standard tools in the financial market was necessary to help households gain financial sources for the construction or purchase of a house/flat. It is mainly the imple- mentation of the benefit for housing, building society accounts (1992) and mortgages (1999). 2. Basic types of housing policy Housing policy of the EU countries has a lot in common. A closer view on the housing market and housing policy shows each country having its form, which has been developing for long decades under the influence of local economic, political and geographical conditions, demo- graphic development, process of urbanisation and predominantly under the influence of local traditions and approaches. Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 23 The difference between housing policies of particular countries is in many cases specific with the overall distinct acting of a state in the housing market. Such a predominantly “geographic border” dividing different housing systems goes between the north and south of Europe. The Northern EU member states (Denmark, Germany) are characterised by a long tradition of complex, actively intervening housing policy. On the other hand, there are the southern states (Portugal, Greece), whose housing policy from the view of the degree of state intervention into housing is more likely less developed. The difference between housing policies of particular countries is in many cases specific with the overall distinct acting of a state in the housing market. Such a predominantly “geographic border” dividing different housing systems goes between the north and south of Europe. The Northern EU member states (Denmark, Germany) are characterised by a long tradition of complex, actively intervening housing policy. On the other hand, there are the southern states (Portugal, Greece), whose housing policy from the view of the degree of state intervention into housing is more likely less developed. A significant criterion, from which other “borders” develop, is the range of state intervention in the housing market, i.e., the complexity of housing policy. In this spirit we can define two basic types of housing policies: The first of them is labelled as additional and is characterised by an intense relying on the market mechanism, while the state effort is oriented mainly on personalised help for low income and other endangered group of inhabitants; the example of such a country following this type of housing policy is mainly Great Britain. The second is labelled as complex and is characterised by quite extensive state interventions in the housing market, which is oriented, more or less, to all layers of the society. Housing policy of this type prevails in the EU member states; mainly the approach of Germany, France, Netherlands and northern states represents it [1]. 2. Basic types of housing policy In a more detailed way, the type of social policy in the broadest sense of the word, i.e., the type of welfare state differentiates the European housing policy [2]. Nowadays, the already classical typology of regimens of the welfare state allows us to derive four basic theoretical types of the European housing policy, the carriers of which, in the prac- tice, are the EU countries [3]. The first type of housing policy is derived from the social-democratic welfare state, which does not understand a human being just as a workforce and is based on the utilization of the citizen principle at providing social services. From the view of housing policy, it is a regimen, which supports the same access of all citizens to housing. The state actively intervenes in the housing market and supports mainly rental and cooperative housing. From the economic view, it is the ineffective model and in the long term not sustainable. Representatives of this approach are mainly the northern states. The second type of housing policy is derived from so-called corporatist or work and perfor- mance-related model of a social state, whose social policy derives mainly from the work activity of an individual in the labour market and emphasizes on the traditional social differentiation of the society. For housing policy of this type of a welfare state, there is a typical and significant reliance on the market mechanism; the state intervention into the housing market is also quite active and focused on all forms of housing in the housing market. Germany, Austria, France and the Netherlands represent a housing policy of this type. The mentioned model might be considered as the closest model of housing policy in the Slovak Republic. The third type of housing policy is derived from the liberal state of public and social services, whose main characteristic feature is the emphasis on the activity of an individual and state Housing 24 Housing 24 interest to provide the help just for those who need it and cannot solve their difficult situ- ation themselves. In the field of housing, the state help is oriented just to low-income and handicapped groups. Financing of housing is through private sources, and public ones are provided partially. In the EU the examples are mainly England and Ireland. 2. Basic types of housing policy The fourth, the theoretically determined type of housing policy, is derived from the so-called rudimentary welfare state, whose social policy in the broadest sense of the word is not devel- oped, so the state provides for citizens just a rescue social net. This approach is usually also related to the conception of housing policy, which means that any care for housing is left on a citizen and the state is involved minimally. The examples of this approach are more likely agrarian countries, such as Portugal or Greece. 3. Housing affordability The harmony between housing (the real estate market) and working opportunities (the labour market) became an essential part of public policy of the state and self-government. There are several reasons why the availability of housing is still related to the availability of working opportunities: • the improvement in the quality of life for a better approach to work, education and services, • the improvement in the quality of life for a better approach to work, education and services • request by companies for workers accommodated in close surroundings, • solution of problems with transport (high travel fares), • protection of environment, • more effective utilisation of land resources and so on. As Cervero [4] presents, the response to this discrepancy might be so-called the job and housing balance policy. He proceeds from a general condition that the ratio between work opportunities and flats leads to a higher efficacy, equality, quality of life and environmental sustainability. However, the measures of housing and work balance policy do not have appar- ently a housing character but vice versa; they function in cooperation with other policies, especially of social protection, labour market and transport policy. According to Krcmar and Rychtarik [5], the analysis of housing affordability in Slovakia can be done on the background of economic and financial cycle. Even if the time series for Slovakia is too short, the available data do suggest a positive correlation between the changes in the real GDP and the flow of loans. On this background housing affordability appears to move against the cycle. In the booming times of 2007 until the first half of 2008, when the GDP growth as well as the lending market reached its historical maxima, average housing became unaffordable for average families. This can be explained by soaring property prices that out- performed growing disposable incomes of households. As the growth of income levels could not keep the pace of the one of the housing market, the lending activity accelerated even more Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 25 to counterbalance the lack of savings generally needed for house purchases. A different situ- ation is observed after 2010, when a combination of stable development on property market and the modest growth of disposable income, accompanied by falling interest rates, led to constant improvements in the affordability of housing [5]. 4. Institutions and economic tools to support housing The present condition of housing in Slovakia is the result of complicated historical develop- ment. Before 1989 the whole system of housing construction development unwound from the system of planned management of the national economy, which meant the state share in com- plex housing construction is almost in the whole range. After the year 1990, the responsibility for the financing of housing moved from the state to citizens in Slovakia. In the Slovak Republic, the system of tools for the support of housing was realized in the form of direct or indirect support. 1. The direct state support: 1. The direct state support: 1. The direct state support: • subsidy for the procurement of rental flats, technical facilities and removal of systemic defects in blocks of flats in the Act No 443/2010 Coll. on subsidies for the development of housing and the Act No. 134/2013 Coll. (MDaV SR) as amended, • the contribution for housing (the Ministry of Labour, Social Affairs and Renewal of Housing Stock, hereinafter MPSVaR SR), • the contribution for insulation of a family house (MDaV SR), • premium loans for procurement of rental housing and renewal of the housing stock (SFRB), • subsidies for procurement of alternative rental flats, technical facilities and building plot (the Ministry of Transport and Construction of the Slovak Republic, hereinafter MDaV SR). . Indirect state support: • subsidies for procurement of alternative rental flats, technical facilities and building plot (the Ministry of Transport and Construction of the Slovak Republic, hereinafter MDaV SR). 2. Indirect state support: • bank guarantee for the renewal of blocks of flats for the loan provided by the state hous- ing development fund (hereinafter SFRB), • state subsidy for young, the height of which is determined by the state each year in the Act on the state budget—the Ministry of Finance of the Slovak Republic (hereinafter MF SR), • state premium to building savings, the height of which is determined by MF SR; the process of determination of percentage height of the state premium is automated and unwinds from the formula, which is defined in the Act on Building Savings (Figure 1). It might be said that the state created a particular (some) system of supportive tools for the development of housing differentiated according to the income structure of households. It might be said that the state created a particular (some) system of supportive tools for the development of housing differentiated according to the income structure of households. Housing 26 Housing 26 Figure 1. Institutions providing public and private sources of financing in Slovakia. Figure 1. Institutions providing public and private sources of financing in Slovakia. The particular category from the view of financing is rental flats determined for social housing, which present flats procured with public sources. They are determined for adequate housing of natural entities, who cannot gain housing by their own efforts. 1. The direct state support: The applicants for such a flat have to fulfil the conditions, which are defined by the Act on Subsidies for the Development of Housing and Social Housing No 443/2010 Coll. as amended. Public sources for the support of social housing might be provided mainly as: • subsidies for procurement of housing, • advantageous loans for procurement of housing, • irrecoverable contributions lowering the cost related to housing, • irrecoverable contributions for operation of a building, in which social housing is provided. • irrecoverable contributions for operation of a building, in which social housing is provided. 4.1. Credits for housing Important factors which distinctly contribute to the dynamisation of the real estate market in Slovakia are accessibility of credit sources and low interest rate. The volume of new credits for housing presents the in long term a crucial part of the overall volume of drawn credits by households. But the aim of new credits for housing is not a priority investment to a purchase of new real estate but also renewal and modernization including the insulation of existing residential buildings as well as the refinancing of older and unfavourable credits. The highest Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 27 sum for the financing of housing is repaid by the Bratislava people. The average height of provided mortgage to inhabitants in the Bratislava region (in 2017) was 95,874 Euros and the average height of their monthly instalment is 403 Euros [6]. It is necessary to emphasize that the key factor of development of the real estate market from the view of development of new financial tools in the field of legal entities as also in a retail area was the entrance of foreign investors into the Slovak financial market. The priority financ- ing by commercial banks becomes investment credits oriented mainly to flat construction. A new form of financing of housing gradually became also project financing, investment financ- ing of legal entities in combination with mortgages and purpose-tailored building credits. The trend of housing credits in Slovakia is presented in Figure 2. gure 2. The trend of housing credits in Slovakia. Source: own processing by the National Bank of Slovakia [7]. 4.2. Taxation policy Taxes and subsidies are mutually closely related and might induce the demand for housing. The developed countries consider housing and real estate a significant source of income for the state budget (especially the real estate tax), which is also one of the reasons why they support the development of housing. Tax relief plays a vital role in the development of housing construction in the world, which encourages the participation of the private sector in the development of housing construction, and so they replace direct state intervention. The provision of tax relief is easier and simpler to be managed as a subsidy, because their implementation does not require the creation of new institutions, as the tax offices already exist. Besides that, the tax relief is not too frequently re-evaluated and changed. It makes their acting more stable and more long term than the utilization of other tools of housing policy. A unique possibility from the view of tax relief in Slovakia is the exemption from the property sale tax after 5 years since the day of acquisition or its elimination from commercial property Housing 28 Housing 28 in the wording of valid legislature (the Act on Income as amended). The relief is related just to the first owner or his close person according to the Civil Code. The tax reform in Slovakia implemented equal tax since 2005, lowering of tax burden and simplification of the tax system, which also stimulated the entry of foreign investors. This reform prefers provision of subsidies, the effect of which seems to be predictable for reform creators. The reform cancelled almost all types of tax relief and at the same time provided minimum space for their sufficient implementation, because it would mean the lowering of the income side of the state budget. Therefore, it limits also the possibilities of implementation of tax relief as a supportive tool of housing policy. A controversial point of tax reform in Slovakia was the implementation of the unified VAT rate [8], which besides other important commodities also overprices flats, subsequently lead- ing to the lowered accessibility of housing for lower-income groups of inhabitants. However, the majority of the EU countries have kept lower VAT (see Table 1). Source: VAT rates applied in the member states of the European Union, 2016—taxud.c.1, taken from Sutovska [9]. Table 1. Overview of VAT in EU member states for selected categories. Table 1. Overview of VAT in EU member states for selected categories. 4.2. Taxation policy In Table 1, we show the overview of rates in the EU member states in two categories, where “ex” means exemption from taxation and “N/A” means not applied lowered rates. From Table 1, we understand that lowered VAT rates for social housing and reconstruction and repair of private flats are applied just in the national legislative of some member states. Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 29 Lowered rates for social housing are applied in Belgium, the Czech Republic, Spain, France, Ireland, Italy, Hungary (since 1 Jan 2016), Poland, Portugal, Romania, Slovenia and Great Britain. Lowered rates for social housing are applied in Belgium, the Czech Republic, Spain, France, Ireland, Italy, Hungary (since 1 Jan 2016), Poland, Portugal, Romania, Slovenia and Great Britain. As Sutovska [9] presents, these lowered rates in the member states are applied just in case of fulfilment of other specific conditions determined in the domestic regulations. In Hungary the lowered VAT rate of 5% is valid just for social housing, where the floor area is lower than 150 m2. Otherwise the basic VAT rate of 27% is applied. Even the European Union requires from the member states the implementation of housing policy to the higher VAT rate; some countries rank housing construction for socially weak layers into a lower VAT rate. These countries realise that based on the lower or zero VAT rate for the construction of small-area flats of common standard, the availability of housing might be increased. The stated possibility is not allowed by the present legislative in the Slovak Republic. 5. Housing prices in Slovakia The trend of housing prices in Slovakia: residential property prices by the types of flats and houses. Source: own processing by the National Bank of Slovakia [7]. Figure 3. The trend of housing prices in Slovakia: residential property prices by the types of flats and houses. Source: own processing by the National Bank of Slovakia [7]. transactions in the residential market are flats. The average price of a square meter of a flat is kept during the last years minimally 10% over the reported overall price of real estate for housing in the Slovak Republic. Figure 3 presents residential property prices by the types of flats and houses. transactions in the residential market are flats. The average price of a square meter of a flat is kept during the last years minimally 10% over the reported overall price of real estate for housing in the Slovak Republic. Figure 3 presents residential property prices by the types of flats and houses. 5. Housing prices in Slovakia Much more attention is paid to the development of real estate prices, namely due to the need to investigate their influence on price stability in the Eurozone. Recently as a distinctly dif- ferentiated development, there might be considered the development of prices of particular types of real estate determined for living not only in Slovakia, but also in the particular EU member states. The development of real estate prices determined for living from the view of currency stability has a substantial impact on the economic activity and changes of prices as a whole. The growth of flat prices contributes to the increase of economic activity not only via the growth of investments into housing but also by increase in household consumption through effects of ownership related to assessing real estate determined for housing. But the intensity of these effects differs distinctly in various countries and depends on the extent of private ownership of houses and flats. On the other hand a possible decline of prices of real estate prices determined for living presents a risk for the stability of the bank system with serious macroeconomic impacts as from the view of financial stability, the immovable asset is one of the main forms of collaterals of credit resources. Demand for housing is significantly higher in those areas of Slovakia which provide more job opportunities, which can be based on economic and social parameters, considered as more advanced. For example, due to the currently implemented mega-investment Jaguar Land Rover in Slovakia, the pressure on prices of flats in the region of Nitra has increased. The arrival of the new investor makes the surrounding of Nitra a lucrative locality which greatly contributed to an increase of apartment prices to about 10–15%, in some cases up to 20% in comparison, towards the end of 2015. Average prices of real estate for housing in the Slovak Republic are in the long term deter- mined mainly by flat prices, as in the market with housing just the offer of flats represents two-thirds from the overall offer of real estate for housing. More than 80% from all realised Housing 30 Housing 30 Figure 3. The trend of housing prices in Slovakia: residential property prices by the types of flats and houses. Source: own processing by the National Bank of Slovakia [7]. Figure 3. 6. Public rental housing which cannot exceed triple of life subsistence by 31 Dec of the previous calendar year, and at the same time the applicant cannot be an owner of any real estate). A severe argument for the lack of rental flats is: A severe argument for the lack of rental flats is: • Contrary to western countries, in Slovakia, at the beginning of 1990s, “a massive privatisa- tion of flats” took place. Consequently within a few years, originally state flats or coopera- tive flats turned into flats under private ownership (as their original “tenants” bought them for residual value). • This situation did not happen in western countries, and the real estate market went its natural way; this is also the reason why the share of rental housing in Slovakia is signifi- cantly lower than in several countries of the European Union. The difference between Slovakia and countries, where rental housing is presented with a high percentage, is that: The difference between Slovakia and countries, where rental housing is presented with a high percentage, is that: • In the Slovak Republic, it is prevailingly assessed as “social housing” for socially weaker layers of inhabitants. • In countries, where housing is presented with a high percentage, rental housing is deter- mined for all layers of the society and not only for that one with a lower social standard. • A majority of inhabitants solve the question of housing mainly with the purchase of a flat into private ownership, while it is a life-long investment with mortgage burden for many years. Housing in rental flats is under the conditions of the Slovak Republic, contrary to foreign countries, where it is still understood as “the assessment of a social status” of a dweller and is prevailingly also assessed as “social housing” for the socially weaker class of inhabitants. On the other hand, in the countries where rental housing is represented by high-volume flats, they are available for all social classes [10]. 6. Public rental housing Availability of tenement, contrary to home ownership, is limited in Slovakia. The reason, besides its lack, is also high rent. Thus, this fact might be considered as one of the important reasons as to why the share of rental housing in Slovakia is significantly lower than in several countries of the European Union. On the other hand, we might assume that those who finance housing through mortgages are not to be able to repay them and they will lose their existing housing. In a horizon of several years a new group of people will originate, who will have problems with the availability of housing not only due to difficulties to get mortgage but also due to quite high rents in bigger towns (e.g., Bratislava, Košice, Banská Bystrica and Žilina, where the prices of rents remain high for a long time, at present at the level of repayments). Rental flats in Slovakia are possible to be divided into two groups: • Flats built before 2000 which were initially state rental flats and in the 1990s they moved to the ownership of towns and villages. • Flats built before 2000 which were initially state rental flats and in the 1990s they moved to the ownership of towns and villages. • Flats whose construction since 2000 has been covered by villages and towns and they are financed from the Programme for Housing Development. These flats have to keep the rental character for at least 30 years and their tenants can be households with income to “a certain fixed height” (to the general conditions there belongs the income of an applicant • Flats whose construction since 2000 has been covered by villages and towns and they are financed from the Programme for Housing Development. These flats have to keep the rental character for at least 30 years and their tenants can be households with income to “a certain fixed height” (to the general conditions there belongs the income of an applicant Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 31 which cannot exceed triple of life subsistence by 31 Dec of the previous calendar year, and at the same time the applicant cannot be an owner of any real estate). which cannot exceed triple of life subsistence by 31 Dec of the previous calendar year, and at the same time the applicant cannot be an owner of any real estate). 6.2. Donation and loan policy in the rental housing Public funding sources, which are currently provided for the development of the construction of rental apartments, have two forms—subsidies and “cheap” loans from the State Housing Development Fund. It should be noted that in terms of economic theory, direct subsidies are considered as one of the least efficient housing policy instruments and should therefore be used only where the market is unable to provide the needs of certain groups in the construc- tion or reconstruction of housing fund, which is a source of serious problems. Subsidy (donation): they are subsidies for rental flats and the construction of technical infra- structure. According to the legislation, applicants can be only municipalities, respectively boroughs. This means that applicants for grants cannot be housing organisations that would be interested in rental housing for low-income groups. Government loans: they are provided by the State Housing Development Fund and are consid- ered as “cheap loans” because their interest rate is at 1.0%, with a repayment within 40 years, and the maximum loan amount per dwelling is 60,000 EUR [11]. However, the reality is that for the loan from the State Housing Development Fund as well as different types of subsidies for the development of housing services provided by the Ministry of Transport and Construction of the Slovak Republic, there is no legal entitlement and its provision and purposive fullness is set by strict conditions in the relevant law. Donations for housing are one of the claims in the provision of assistance in the material need to cover housing costs. The height of the contribution for housing is [12]: • 55.80 € monthly, if it is a citizen in material poverty and • 89.20 € monthly, if it is a citizen in material poverty and natural entities, who are assessed together with the citizen in material poverty. • 89.20 € monthly, if it is a citizen in material poverty and natural entities, who are assessed together with the citizen in material poverty. 6.1. Sources of funding for municipal rental housing in Slovakia Currently there is only one type of rental housing supported by the state for the socially disadvantaged. Subsidies for rental housing, subsidies for the construction of technical infrastructure and loans from the State Housing Development Fund are public sources of funding which are currently the most utilised and also the most preferred. Other funding opportunities are own resources of towns, villages and commercial loans. Their availabil- ity for municipal entities is limited in terms of higher interest rates and limiting factors for spending commercial loans in terms of the Financial Regulation, which municipal entities must comply with. It should be noted (based on the Law of the Financial Regulation of towns and villages) that the loans from the State Housing Fund are not added to loan indebtedness of the given municipal subject (unlike commercial loans, which are added to debt). Housing 32 Housing 32 7. The relationship between the housing and labour mobility Consistency between housing (housing market) and jobs (labour market) has become an important part of the public policy of the state and local governments. The working places are increasingly dependent on the availability of housing (or rental housing). There are many reasons for this—for example, a higher quality of life, such as the demand for businesses for employees living in the neighbourhood or the solution of traffic situation (saving money on transport or saving time in traffic jams). Housing is therefore an important source of employ- ment. The functioning of the housing market is connected with the mobility of housing, which is closely linked with the desired and undesired labour mobility. Many people have become “prisoners” of their homes, whether rented or own housing, due to lack of offers or for economic reasons. Requirement of ensuring increased mobility of housing is a prerequisite for greater labour mobility and thus better opportunities and success in the labour market. In order to ensure the competitiveness of regions and foster job mobility, it is necessary to Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 33 diversify the supply of housing—not only at the level of different types of ownership and usage but also diversity in the supply cost to provide housing. The reason is particularly the availability of quality housing, which will proportionally correspond with the financial resources of households. Also, instant availability of housing for employees is an essential condition for the development of a dynamic labour market—the development of employment is blocked by the lack of housing. In Slovakia, the real estate market is characterised by a high proportion of owner-occupied real estates and almost non-existent rental housing. Moving for work is therefore complicated because of the lack of availability of rental housing. The current structure of the housing market is largely due to historical development —both in the development before 1989 and in the lack of adaptation of the housing sector to the new conditions in recent decades. After the change of the political system in 1990, the transformation process began, which also included the privatisation process. Part of the privatisation process in 1992 was the introduction of the right to redeem the property for tenants in state apartments at very reasonable prices (apartments were redeemed into private property at prices that correspond to 5% of their market value), resulting in a change in the ownership structure. 7. The relationship between the housing and labour mobility Migration is usually associated Housing 34 Housing 34 with a change in the residence of a worker (in economic literature, a migrant is considered a person who stays in the destination region for more than 1 year). Commuting occurs as a rule without a change in the worker’s residence (i.e., a worker’s place of work and place of residence are located in two different regions) [14]. From Figure 4 we can see the negative trend in the development of rental flats in Slovakia. Important is also the fact that social rental housing depends on income. This means that in case of the entitlement of rented apartments, the motivation to earn more money is lower. The reason is the fear that the tenant loses his entitlement to a rented flat. The result of such a situation is reduced incentives to work and thereby a decrease in the mobility of labour. On the other side, the second extreme in Slovakia is a high proportion of persons who have apartments in their ownership [16]. Purchase, respectively, rental of residential property is generally not just a matter of rational economic calculation, but to some extent it is also a reflection of national usage. Within the internationally comparable survey on income and living conditions (Statistics of Income and Living Conditions—SILC) in European countries, it is proven that not only in Slovakia but also in most countries the ownership of the house or apartment is preferred more than rent. In the EU, 28.70% of the population subscribes to the ownership of residential property. However, between countries there are obvious differences. Slovakia clearly belongs to the category of countries where the term ownership of real estate considerably resonates (up nearly 90%). On the other hand, the most significant relationship to rental housing in Europe is shown by the people of Germany and Austria. Available cross-country comparisons show that mobility flows in Slovakia are low in interna- tional terms. This applies in particular to internal migration, which means to flows connected with a change in the place of residence. The internal migration rate for Slovakia, measured as gross regional outflow as a percentage of working age population is significantly lower than in most EU countries and other advanced economies as well as neighbouring Visegrad countries (see e.g. [17]). According to national statistics, 1.6% of the productive age population Figure 4. 7. The relationship between the housing and labour mobility A large proportion of public rental housing was subsequently purchased by tenants into the private sector. For example, in Bratislava, the share of private housing increased from 12% in 1992 to 70% in 1998. The high occupation of flats, the rise of prices and rents, geographic differentiation of wages and so on lead to the reduced labour mobility [13]. While the historical development has helped to shape the current structure of housing, housing support policy helps to preserve this structure and thus contributes to the slower development of rental housing. At present, most public expenditure spent on housing support is used for the subsidy of owner-occupied housing. Cheaper financing of owner-occupied housing has a direct impact on the entire housing market. Subsidised owner-occupied housing is financially more attractive and more accessible than rental housing. This promotes the demand for owner-occupied housing at the expense of rental housing. The market reacts to a lower demand for rental housing with lower investments for the creation of new rental housing, which leads to a slower development of this segment of market. Flexible labour market enables workers in the better choice of employment and the efficient allocation of labour among companies. In Slovakia, one of the obstacles of labour mobility in the labour market is low regional mobility. Comparing internationally, Slovaks move for work much less than people in neighbouring EU countries. One of the factors that may nega- tively influence this condition of the labour market is the housing market. Low mobility on the labour market has a negative impact on economic growth. Regional unemployment is rising while jobs remain unoccupied. The low labour mobility also reflects regional differ- ences in unemployment rates. It leads to inefficient use and allocation of human resources. People are forced to refuse job opportunities in other regions, where they could generate more economic value and thus earn more money. Higher unemployment and limiting productivity growth are associated with lower GDP and lower living standards. Labour mobility involves changes in the physical location of workers (geographical mobility) or their movement across jobs and sectors (occupational mobility). Two main types of internal labour mobility are in the centre of attention—internal migration and commuting. 7. The relationship between the housing and labour mobility The evolution of the share of rental apartments in Slovakia among the total number of completed apartments (in percentage). Source: own processing according to data from the Ministry of Transport and Construction of the Slovak Republic [15]. Figure 4. The evolution of the share of rental apartments in Slovakia among the total number of completed apartments (in percentage). Source: own processing according to data from the Ministry of Transport and Construction of the Slovak Republic [15]. Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 35 (15–64 years) changed their place of residence in 2011, of which 46% moved between munici- palities within districts (i.e., short-distance migration), 29% between districts within a region and 25% between regions. These figures are remarkably stable over time (there are only minor variations in data between 2000 and 2013), implying that the overall internal migration capacity and main flows are relatively unresponsive to external factors such as international migration or policy interventions. No more than 3–4% of residential migrants (3.3% in 2011) indicate employment-related motives as the reason for relocation. “Housing reasons” and “following a family member” account for almost two-thirds of officially declared migration reasons. The share of labour migration increases with the spatial distance of relocation and is the highest in migrations between regions (8.5% of all inter-regional migrations in 2011; in comparison, employment reasons were stated in merely 1.1% of relocations between munici- palities). Jurcova and Vano [18] points out that this observation is consistent with data on the educational structure of migrants and the general assumption that the propensity to move for work increases along with educational attainment. 8. Conclusion In this chapter, we tried to discuss the development of housing policy in Slovakia after 1990, some selected issues related to the development of housing in Slovakia and the availability of housing and labour mobility. An indispensable factor in housing development is a well- developed institutional environment regarding efficient economic instruments (particularly credit and tax policy) as well as the establishment of the organisations in the housing market. Finally, we can say that focusing on supporting housing housed by owners with existing tools causes a particular disruption to the real estate market. There are possibilities to help adjust the current situation and avoid future complications. One of them is the redirection of the state aid for housing housed by the owner for financial contributions to housing (housing allowance). This housing policy tool would allow the recipient to choose between a rental and a housed owner. It means that financial contributions could be more tied to income and other social criteria and thus to achieve the more significant resurgence of resources. The strength- ening of this instrument in Slovakia is also recommended by the OECD. The high occupation of flats, the rise of prices and rents and geographic differentiation of wages lead to reduced labour mobility. In Slovakia, one of the obstacles of labour mobility in the labour market is low regional mobility. Comparing internationally, Slovaks move for work much less than people in the neighbouring EU countries. One of the factors that have negatively influenced this condition of the labour market is the low share in the offer of rent housing for various income groups of inhabitants. In the future there is a need to focus on the simplification of administration schemes of sup- portive state financing in the field of rental housing, to prepare new financial mechanisms to improve the development of the rental sector in the Slovak Republic. Housing 36 Housing 36 Author details Daniela Spirkova Address all correspondence to: daniela.spirkova@stuba.sk Institute of Management, Slovak University of Technology in Bratislava, Bratislava, Slovak Republic Address all correspondence to: daniela.spirkova@stuba.sk Institute of Management, Slovak University of Technology in Bratislava, Bratislava, Slovak Republic References [1] Heidenheimer AJ, Heclo H, Adams CT. Comparative Public Policy of Social Choice in American, Europe and Japan. New York: St. Martin’s Press; l990 2] Balchin P, editor. Housing Policy in Europe. London and New York: Routledge; 19 [3] Donner C. Housing Policies in the European Union–Theory and Practice. Vienna: Christian Donner; 2000 [4] Cervero R. Jobs-housing balancing and regional mobility. Journal of the American Planning Association. 1989;55(2):136-150 [5] Krcmar M, Rychtarik S. Dynamics of housing affordability in Slovakia and its regions on the background of macroeconomic development. Biatec. 2013;21(9):18-21 [6] CSOB. Financovanie bývania (Housing finance). Available from: https://touchit.sk/ priemerna-vyska-hypoteky-sa-zvysila-o-15-najviac-v-bratislavskom-a-kosickom-kraji/ [Accessed 20 April 2018] [7] National Bank of Slovakia. Review of Housing Loans. 2018. Available from: https:// www.nbs.sk/en/financial-market-supervision-practical-info/publications-data/selected- data/review-of-housing-loans [Accessed: 15 April 2018] [8] VAT Rates Apllied in the Member States of the European Union, 2016—taxud.c.1; 2016 [9] Sutovska A. Impact of value added tax at the corporate economy in the construction sector [Dissertation thesis]. Bratislava: Slovak University of Technology; 2016 [10] Spirkova D, Babelova J. Financing of rental housing in Slovakia. In: Proceedings of 28th International-Business-Information-Management-Association Sevilla, Spain. Innovaton Management and Sustainable Economic Competitive Advantage: From Regional Devel opemnt to global Growth. Vols. I–VI. 2016 [11] Act No. 607/2003 on State Housing Development Fund, Collection of Laws, Slovak Republic [12] Act no. 601/2003 Coll. on Subsistence Minimum and on the Amendment and Supplements to Certain Acts, as amended. The Ministry of Labour, Social Affairs and Family of the Slovak Republic Housing Policy in the Slovak Republic http://dx.doi.org/10.5772/intechopen.78611 37 37 [13] Vagac L. Housing Construction, Mobility, Employment. Bratislava, Slovakia: Centre for Economic Development; 2003 [14] Vagac L. Internal Labour Mobility in Slovakia. European Employment Observatory; 2013 [15] Information on housing construction in the Slovak Republic 2003-2017. Ministry of transport andconstructions of the Slovak Republic [16] Spirkova D, Fulopova C. The causal relationship between the housing and labor mobility. In: Proceedings of 27th International-Business-Information-Management-Association Conference; Milan, Italy. Vols. I–VI. 2016 [17] Palacin J, Shelburne R. The Private Housing Market in Eastern Europe and CIS. Geneva, Switzerland: United Nations Economic Commission for Europe; 2005 [18] Jurcova D, Vano B. PopulačnývývojnaSlovensku v kontextepopulačnéhovývoja v EÚ. In: DemografickáperspektívaEÚ a Európy pre 21. Storočie a národnostnéietnickémenšiny v krajináchEÚ. Bratislava: SSZPaÚradvlády SR; 2011
https://openalex.org/W4311661954
https://sjtrem.biomedcentral.com/counter/pdf/10.1186/s13049-022-01050-0
English
null
Moderate and severe traumatic brain injury in general hospitals: a ten-year population-based retrospective cohort study in central Norway
Scandinavian journal of trauma, resuscitation and emergency medicine
2,022
cc-by
10,633
Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 https://doi.org/10.1186/s13049-022-01050-0 Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 https://doi.org/10.1186/s13049-022-01050-0 Open Access ORIGINAL RESEARCH Moderate and severe traumatic brain injury in general hospitals: a ten‑year population‑based retrospective cohort study in central Norway Moderate and severe traumatic brain injury in general hospitals: a ten‑year population‑based retrospective cohort study in central Norway Shavin Rahim1, Eivor Alette Laugsand2,3,4, Even Hovig Fyllingen5,6, Vidar Rao1,7, Rabea Iris Pantelatos1, Tomm Brostrup Müller7, Anne Vik1,7 and Toril Skandsen1,8* Shavin Rahim1, Eivor Alette Laugsand2,3,4, Even Hovig Fyllingen5,6, Vidar Rao1,7, Rabea Iris Pantelatos1, Tomm Brostrup Müller7, Anne Vik1,7 and Toril Skandsen1,8* © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Abstract Background:  Patients with moderate and severe traumatic brain injury (TBI) are admitted to general hospitals (GHs) without neurosurgical services, but few studies have addressed the management of these patients. This study aimed to describe these patients, the rate of and reasons for managing patients entirely at the GH, and differences between patients managed entirely at the GH (GH group) and patients transferred to the regional trauma centre (RTC group). We specifically examined the characteristics of elderly patients. Methods:  Patients with moderate (Glasgow Coma Scale score 9–13) and severe (score ≤ 8) TBIs who were admitted to one of the seven GHs without neurosurgical services in central Norway between 01.10.2004 and 01.10.2014 were retrospectively identified. Demographic, injury-related and outcome data were collected from medical records. Head CT scans were reviewed. Results:  Among 274 patients admitted to GHs, 137 (50%) were in the GH group. The transferral rate was 58% for severe TBI and 40% for moderate TBI. Compared to the RTC group, patients in the GH group were older (median age: 78 years vs. 54 years, p < 0.001), more often had a preinjury disability (50% vs. 39%, p = 0.037), and more often had moderate TBI (52% vs. 35%, p = 0.005). The six-month case fatality rate was low (8%) in the GH group when transfer- ral was considered unnecessary due to a low risk of further deterioration and high (90%, median age: 87 years) when neurosurgical intervention was considered nonbeneficial. Only 16% of patients ≥ 80 years old were transferred to the RTC. For this age group, the in-hospital case fatality rate was 67% in the GH group and 36% in the RTC group and 84% and 73%, respectively, at 6 months. Conclusions:  Half of the patients were managed entirely at a GH, and these were mainly patients considered to have a low risk of further deterioration, patients with moderate TBI, and elderly patients. Less than two of ten patients ≥ 80 years old were transferred, and survival was poor regardless of the transferral status. Keywords:  Traumatic brain injuries, Craniocerebral trauma, General hospitals, Trauma centers, Tertiary care centers, Referral and consultation, Mortality, Aged, 80 and over *Correspondence: toril.skandsen@ntnu.no Traumatic brain injury (TBI) is frequent, with an esti- mated 4  000 hospital admissions per year in Norway [1]. In low-income countries, the incidence of TBI has 1 Department of Neuromedicine and Movement Science, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway Full list of author information is available at the end of the article 1 Department of Neuromedicine and Movement Science, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway Full list of author information is available at the end of the article 1 Department of Neuromedicine and Movement Science, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 Page 2 of 11 (the GH group). We investigated the rate of and reason for managing patients entirely at the GH and explored differences between patients managed entirely at the GH and patients transferred to the RTC for further treat- ment, including differences in survival. Furthermore, we aimed to specifically describe the transferral rate and sur- vival in elderly patients. increased due to more traffic accidents, while in high- income countries, more elderly people sustain TBIs due to falls [2]. *Correspondence: toril.skandsen@ntnu.no Approximately 15% of all hospital-referred TBIs are moderate or severe [1], with outcomes varying from complete restitution to severe disability or death. Several guidelines for the initial management of TBI in Scandinavia have been published in recent decades [3–5]. In 2008, the Scandinavian Neurotrauma Commit- tee (SNC) published summarised guidelines for the pre- hospital management of severe TBI based on previously developed guidelines by the Brain Trauma Foundation (BTF) [4, 6]. These guidelines recommend that all cases of suspected severe TBI (defined by a Glasgow Coma Scale (GCS) score of 3–8) should be stabilised and trans- ported promptly to the nearest neurosurgical department for specialised neurointensive management. Notably, the SNC and BTF have not addressed the management of elderly patients with TBI specifically, even though these patients present with a greater burden of comorbid con- ditions and a higher risk of death and may not benefit from aggressive neurosurgical treatment [7]. Study setting and study participantsh The study comprised patients admitted primarily to GHs without neurosurgical services in central Norway, one of the four health regions in the country. Central Norway covered a population of approximately 670,000 inhabit- ants in the study period, and the distance between the GHs and the RTC ranged from 42 to 348 km by car. At the time of the study, the region held two levels of hos- pitals [11]: seven GHs meeting the requirements for level III trauma centres (for details see Additional file 1: Table S1) and one level I/II trauma centre, St. Olavs Hos- pital, with a neurosurgical department, denoted the RTC. The RTC also serves as a GH for approximately 45% of the population in the region, and most patients admit- ted there with TBI in need of hospitalisation for more than 24 h are admitted to the neurosurgical department, regardless of TBI severity. Patients belonging to the catchment area of St. Olavs Hospital were therefore not included in this study. As the RTC and the GHs lie within the same health region in central Norway, the standards of the prehospital care were the same for all patients and were in line with the guidelines applicable at the time they were admitted [3–5]. For patients with a TBI not considered severe, the deci- sion on primary admission to the regional trauma centre (RTC) or the closest local trauma centre (general hospi- tal (GH)) is based on evaluations performed by the pre- hospital emergency personnel, taking a range of factors into consideration. The benefit of direct transportation has been extensively studied, with conflicting results, and a systematic review did not find a clear benefit of direct transportation to an RTC in this patient group [8]. Hence, the Norwegian National Trauma Plan states that if trans- portation time to the RTC is 45 min or more, transpor- tation to the closest trauma centre should be considered [9].h Study procedures and inclusion and exclusion criteria Patients with a GCS score of 13 were included as moderate TBI because of the higher frequency of intracranial lesions and risk of adverse outcomes [15, 16], in accordance with the Head Injury Severity Scale, commonly used in Scandinavia [17]. The acute phase was defined as the time from admission to the GH to discharge from any ward responsible for the management of the trauma or related complications. If the patient was transferred to the RTC, discharge from the RTC denoted the end of the acute phase. A subse- quent stay at a GH or rehabilitative institution was not considered part of the acute phase. to capture the entire burden of moderate and severe TBI managed at the GHs. TBI was classified as mild (GCS scores 14–15), moderate (GCS scores 9–13), or severe (GCS scores 3–8) [14]. Patients with a GCS score of 13 were included as moderate TBI because of the higher frequency of intracranial lesions and risk of adverse outcomes [15, 16], in accordance with the Head Injury Severity Scale, commonly used in Scandinavia [17]. The acute phase was defined as the time from admission to the GH to discharge from any ward responsible for the management of the trauma or related complications. If the patient was transferred to the RTC, discharge from the RTC denoted the end of the acute phase. A subse- quent stay at a GH or rehabilitative institution was not considered part of the acute phase. and admission to the intensive care unit (ICU) were also registered. For patients transferred to the RTC, any neu- rosurgical procedure performed, including intracranial pressure (ICP) monitoring, was recorded.i Head CT scans for the entire acute phase (first and later CT scans) were reviewed by a resident in radiology (EHF), who was blinded to the patients’ medical records except for the CT referral information. CT findings were recorded dichotomously as yes or no for the variables epi- dural haematoma (EDH), subdural haematoma (SDH), traumatic subarachnoid haemorrhage (tSAH), single con- tusion/haematoma, multiple contusions/haematomas, intraventricular haemorrhage, general oedema (one or both hemispheres, or cerebellum), skull fracture with or without impression, and intracranial air. Whether the consultant neurosurgeon or resident in neurosurgery on call at the RTC was consulted while the patient was in the ED, or was consulted later, was regis- tered. Study procedures and inclusion and exclusion criteria Study p ocedu es a d c us o a d e c us o c te a Patients admitted to one of the seven GHs of the region between 01.10.2004 and 01.10.2014, with a main or sec- ondary diagnosis registered as ICD-10 code(s) S06.1- S06.9 or S09.7-S09.9 (International Classification of Diseases, 10th revision), were screened for inclusion by review of their medical records. The ICD-10 coding sys- tem does not have unique codes for the severity of TBI, but moderate or severe TBI would most likely be coded as one or several of the abovementioned codes [12]. The Trondheim moderate and severe (ms)TBI study served as an additional source of patient identification. This is a prospective cohort study of all patients with moderate or severe TBI admitted to St. Olavs Hospital, including those transferred from GHs, from 01.10.2004 onwards [13]. The initial in-hospital management of minimal, mild, and moderate TBI in adults is described in guidelines that were developed by the SNC in 2000 and revised in 2013 [3, 5]. According to the guidelines applied in the period of the current study, all patients with signs of moderate TBI should have an early head CT scan and be admitted to the hospital for further observation for ≥ 12 h. In cases with abnormal head CT scans, consultation with a neu- rosurgical department is recommended. However, the guidelines do not specify which patients should be man- aged at the GH and which patients should be managed in a neurosurgical department. Most studies of treatment and outcome of moderate and severe TBI originate from university hospitals with neurosurgical services [10]; hence, there is a shortage of knowledge about patients managed entirely at GHs.h We included patients with moderate and severe TBIs upon admission to the emergency department (ED), as well as patients with mild TBI in the ED who deterio- rated to moderate or severe TBI during the acute phase This study aimed to describe adult patients with mod- erate and severe TBI admitted to GHs in a defined health region, focusing on patients who were never transferred Page 3 of 11 Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 to capture the entire burden of moderate and severe TBI managed at the GHs. TBI was classified as mild (GCS scores 14–15), moderate (GCS scores 9–13), or severe (GCS scores 3–8) [14]. Study procedures and inclusion and exclusion criteria When the patient was not transferred to the RTC, the main reason for keeping the patient at the GH after stabilisation in the ED was derived from the patient’s medical records and assigned to one of four predefined categories: unsalvageable was used for patients with a low GCS score, bilateral fixed semidilated pupils and/or a head CT scan demonstrating signs of severe herniation. Neurosurgical intervention considered nonbeneficial was selected when the risk associated with a neurosurgical procedure or intensive care was considered to outweigh the potential benefits associated with these interventions. Presumed low risk of further deterioration was selected in cases where the patient had a relatively high/stable GCS score upon arrival and modest head CT findings or in cases where alcohol intoxication was suspected to influ- ence the GCS score [18]. Another reason was selected when the reason for keeping the patient at the GH did not fall into the other specified categories, for instance, severe trauma to other organ systems requiring imme- diate surgery at the GH or cases suggestive of isolated traumatic axonal injury (TAI) assuming no need for ICP measurement. Intoxicated patients with an initially low GCS score who regained normal consciousness within 12  h were considered to have mild TBI and were excluded [18]. Patients who did not reside in Norway were excluded. Children < 17  years old were also excluded, as we have previously shown that most of these patients are trans- ferred to the RTC [19]. Patients with a chronic subdural haematoma and patients where a diagnosis of TBI was considered unlikely or uncertain were also excluded (e.g., patients with stroke), as well as patients who presented to the ED > 72 h after the trauma. Medical records and head CT scans for the entire acute phase were reviewed, and patients were categorised into patients managed entirely at the GH (GH group) and patients transferred to the RTC (RTC group). Study variables Preinjury disability was classified as yes if the patient had any known concurrent condition that affected daily func- tioning at the time of injury and was further categorised into neurological condition (including dementia), alcohol abuse, cardiopulmonary disease, psychiatric disorder, substance abuse, cancer, developmental disorder, other, or several. Preinjury antithrombotic medication was recorded as yes if the patient was using platelet inhibitors or anticoagulants (including heparins or direct-acting oral anticoagulants). Outcome variables were the length of the acute hospi- tal stay, discharge destination (nursing home, own home, specialised rehabilitation, other rehabilitative institu- tion, or another hospital), death during the acute phase (i.e., in-hospital fatality), and death within 6 months and 12  months postinjury (6- and 12-month fatality). The cause of in-hospital fatality was categorised as intrac- ranial hypertension when the clinical and radiological characteristics indicated increased ICP resulting in tam- ponade. Complications of TBI was selected when the patient died from complications or severe neurological impairments secondary to TBI. Other trauma, or unre- lated to the injury, was selected when death was caused Alcohol intoxication upon admission was recorded as yes if clinical suspicion was described in the medical records or if the patient had a measured serum etha- nol ≥ 2.2 mmol/L [20]. Injury-related variables were the cause of injury (fall, motor vehicle accident, or other cause including violence), admission GCS score (the low- est recorded GCS score in the ED) and severity of TBI at admission (mild, moderate, or severe). In cases with prehospital intubation, the last recorded GCS before sedation was used. Trauma team activation (TTA) at admission, extracranial injuries, extracranial surgery, Page 4 of 11 Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 by trauma to a different organ system or was deemed unrelated to the TBI (e.g., a myocardial ischaemic infarction). and were more often injured in falls (Table 1). In addi- tion, patients in the GH group more often had preinjury disability than did patients in the RTC group (50% vs. 39%, p = 0.037) and more often received antithrombotic medication (48% vs. 28%, p < 0.001). Platelet inhibitors were the most common antithrombotic medication for both groups (Additional file 1: Table S2). Anticoagulants were used by 14% of patients in the GH group and 9% in the RTC group (p = 0.106) (Table  1). In the GH group, fewer patients were admitted with TTA (35% vs. Included patients In total, 1307 index hospital stays at the seven GHs were reviewed. Of these, 253 patients met the inclusion criteria for moderate or severe TBI (including deterioration from mild TBI in the acute phase). In addition, 19 patients were included from the Trondheim msTBI study’s data- base (Fig. 1). These were not in the retrieved lists of ICD- 10 codes, for instance, because very short stays at the GH had not been coded or because other ICD-10 codes had been used, for instance, I61 intracerebral haemorrhage. Hence, the total study population was 274 patients. Reasons for managing patients at the GH and relation to fatality A consultant neurosurgeon or resident in neurosurgery on call at the RTC was consulted from the ED in 110 cases (80%) in the GH group. In addition, 12 cases (8%) were discussed later in the acute phase. All patients cat- egorised as unsalvageable died during the acute phase (Table 4). Among patients for whom neurosurgical inter- vention was considered nonbeneficial, 17% presented with severe TBI at admission, and 90% died within 6 months. Among patients with a presumed low risk of further deterioration, four patients (8%) died within 6  months (age range: 71–94  years). These four patients died dur- ing the acute phase or were discharged to palliative care. One was later diagnosed with a severe concurrent somatic disease, while the remaining three patients were > 70  years old, none had preinjury disability, all had a mild TBI in the ED phase, and all were on antithrombotic medications. The overall transferral rate was 50%, leaving 137 in the GH group and 137 in the RTC group. There were no major differences between the seven GHs in transferral rate (Additional file 1: Table S1). The transferral rate was 58% for patients with severe TBI (n = 125) and 40% for patients with moderate TBI (n = 119). The patients trans- ferred to the RTC had a median length of stay of 2.8 h (IQR 1.9–5.2 h) at the GH before transferral. Study variables 57%, p < 0.001), received extracranial surgery (13% vs. 32%, p < 0.001), or were admitted to the ICU (80% vs. 93%, p < 0.001). Statistical analyses Descriptive data are reported as the mean with stand- ard deviation when normally distributed and the median with interquartile range (IQR) when nonnormally distrib- uted. Comparisons between the GH group and the RTC group were performed using the independent samples t test for normally distributed continuous variables and the Mann‒Whitney U test for ordinal and nonnormally distributed continuous data. Proportions were compared using the Pearson χ2 test, and the exact z-pooled test when the expected count was less than 5 for any cell, as recommended [21]. Cases with missing values were not included in the analyses. Proportions are reported as a percentage of the total included cases, regardless of the missing status, except for reported CT findings. All CT variables were missing in seven patients, and propor- tions of all cases with nonmissing status were reported. Missing data are noted in all tables, except when the frequency of missing values was < 5%. The significance level was set to α = 0.05. All statistical analyses were per- formed using IBM SPSS Statistics version 25. EDH, multiple contusions, and general oedema were more frequent in the RTC group (p < 0.001, p = 0.026 and p < 0.001, respectively Table 2). SDH and tSAH were frequent in both groups. In the RTC group, 94 patients (69%) had a neurosurgical procedure (Table 1). More patients died in the acute phase in the GH group than in the RTC group (44% vs. 16%, p < 0.001) (Table 3) but less often from intracranial hypertension (70% vs. 95%, p = 0.015). The six-month case fatality rate (CFR) was higher in the GH group than in the RTC group (54% vs. 25%, p < 0.001). Patients in the GH group were discharged to a variety of settings, including specialised rehabilitation (23%) and nursing homes (32%). In the RTC group, most (91%) were discharged to their respective GHs. Elderly patientsh There were more patients with moderate TBI (52% vs. 35%, p = 0.005) in the GH group. The patients in the GH group were significantly older than the patients in the RTC group (median age: 78 years vs. 54 years, p < 0.001) The percentage of patients transferred to the RTC decreased with increasing age, from 64% of patients < 65 years old to 16% of patients ≥ 80 years old Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 Page 5 of 11 Index hospital stay n = 1 307 Non-resident n = 11 Moderate or severe TBI, age ≥ 17 years n = 253 Transferred to the RTC, identified retrospectively and also included in the Trondheim msTBI study n = 98 TBI, age ≥ 17 years n = 831 Negative or doubtful TBI n = 118 Chronic subdural hematoma n = 240 Age < 17 years n = 105 Managed entirely at the GH n = 137 Transferred to the RTC, only identified through the Trondheim msTBI study n = 21 Transferred to the RTC n = 137 Minimal or mild TBI n = 482 Borderline mild-to-moderate TBI† n = 96 > 72 h from trauma to admission n = 2 Transferred to the RTC, only identified retrospectively n = 18 Fig. 1  Flowchart of the inclusion process. Index hospital stay refers to any admission to a general hospital with ICD-10 code S06.1-S06.9, S09.7-S09.9 during the study period. GH General hospital, msTBI moderate and severe TBI, RTC​ Regional trauma centre, TBI Traumatic brain injury. †Categorised as borderline mild-to-moderate due to the inability to accurately classify these patients’ TBI severity. Elderly patientsh For instance, intoxicated patients with an initially low GCS score who regained normal consciousness within 12 h, and severely demented patients with an already reduced GCS score preinjury Index hospital stay n = 1 307 > 72 h from trauma to admission n = 2 TBI, age ≥ 17 years n = 831 Minimal or mild TBI n = 482 Borderline mild-to-moderate TBI† n = 96 Moderate or severe TBI, age ≥ 17 years n = 253 Transferred to the RTC, only identified through the Trondheim msTBI study n = 21 Transferred to the RTC, identified retrospectively and also included in the Trondheim msTBI study n = 98 Transferred to the RTC, only identified retrospectively n = 18 Managed entirely at the GH n = 137 Transferred to the RTC n = 137 Fig. 1  Flowchart of the inclusion process. Index hospital stay refers to any admission to a general hospital with ICD-10 code S06.1-S06.9, S09.7-S09.9 during the study period. GH General hospital, msTBI moderate and severe TBI, RTC​ Regional trauma centre, TBI Traumatic brain injury. †Categorised as borderline mild-to-moderate due to the inability to accurately classify these patients’ TBI severity. For instance, intoxicated patients with an initially low GCS score who regained normal consciousness within 12 h, and severely demented patients with an already reduced GCS score preinjury In the category unsalvageable, 75% of patients were ≥ 65 years old, and 44% of patients were ≥ 80 years old. In the category neurosurgical intervention consid- ered nonbeneficial, 95% of patients were ≥ 65 years old, and 83% of patients were ≥ 80 years old (Table 4). (Table  5). Patients ≥ 80  years old constituted 25% of the total study population: 42% of the GH group versus 8% of the RTC group (p < 0.001) (Table  1). In the GH group, 63% of the patients ≥ 80 years old had a prein- jury functional disability, most commonly a neurologic condition. Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 Page 6 of 11 In patients ≥ 65  years old, the in-hospital CFR was higher in the GH group than in the RTC group (62% vs. 24%, p < 0.001), as was the 6-month CFR (76% vs. 43%, p < 0.001) (Table 5). Elderly patientsh In patients ≥ 80 years old, the in-hospital CFR seemed to be higher in the GH group than in the RTC group, although not significantly dif- ferent (67% and 36%, p = 0.062), whereas the CFR at 6 h i il (84% d 73% 0 420) Table 1  Patients characteristics in the GH group and the RTC group (n = 274) Number of missing cases not stated when frequency < 5% GCS Glasgow coma scale, GH General hospital, IQR Interquartile range, N/A Not applicable, RTC​ Regional trauma centre, SD Standard deviation, TBI Traumatic brain injury Bold font indicates statistical significance a Two patients were transferred to a different RTC than St. Elderly patientsh Olavs Hospital due to extraordinary circumstances b Statistical analyses were not conducted due to non-normality in the two groups GH group RTC ­groupa p-value n = 137 n = 137 Age in years, median [IQR] 78 [55, 87] 54 [33, 68] < 0.001 Age in years, mean (SD)b 68 (23) 52 (21) N/A Patients ≥ 80 years old, n (%) 57 (42) 11 (8) < 0.001 Male, n (%) 90 (66) 108 (79) 0.015 Preinjury functional disability, n (%) yes (%) 69 (50) 54 (39) 0.037  Missing 8 (6) 4 (3) N/A Antithrombotic medication, n (%) 66 (48) 38 (28) < 0.001  Missing 11 (8) 1 (1) N/A Anticoagulant use, n (%) 19 (14) 12 (9) 0.106  Missing 13 (9) 1 (1) N/A Alcohol intoxication, n (%) 25 (18) 44 (32) 0.009 Cause of injury, n (%)  Fall 111 (81) 75 (55) < 0.001  Motor vehicle accident 17 (12) 41 (30) < 0.001  Other cause 7 (5) 17 (12) N/A Admission GCS score, median [IQR] 10 [5, 12] 8 [5, 13] 0.827  Missing, n (%) 30 (22%) 4 (3%) N/A Admission GCS score of 13, n (%) 13 (9%) 17 (12%) 0.883 Severity of TBI at admission, n (%)  Severe TBI 53 (39) 72 (53) 0.021  Moderate TBI 71 (52) 48 (35) 0.005  Mild TBI 13 (9) 17 (12) 0.439 Trauma team activation, n (%) 48 (35) 78 (57) < 0.001 Extracranial injury, n (%) 58 (42) 68 (50) 0.224 Extracranial surgery, n (%) 18 (13) 44 (32) < 0.001 ICU stay, n (%) 109 (80) 128 (93) < 0.001 Neurosurgical procedure, n (%) – 94 (69) – Table 2  Main findings at head CT scans in the GH group and the RTC group CT Computed tomography, GH General hospital, RTC​ Regional trauma centre Bold font indicates statistical significance a There was missing data for 6 cases in the GH group, for all variables except epidural haematoma of which there were 7 missing cases b There was missing data for 1 case in the RTC group, for all variables GH group RTC group p-value n = ­131a n = ­136b Normal first CT, n (%) 2 (1) 1 (1) 0.540 Epidural haematoma, n (%) 6 (4) 26 (19) < 0.001 Subdural haematoma, n (%) 92 (67) 93 (68) 0.744 Subarachnoid haemorrhage, n (%) 92 (67) 90 (66) 0.477 Single contusion/haematoma, n (%) 33 (24) 18 (13) 0.013 Multiple contusions/haematomas, n (%) 45 (33) 65 (47) 0.026 Intraventricular haemorrhage, n (%) 32 (23) 36 (26) 0.702 General oedema, n (%) 4 (3) 48 (35) < 0.001 Skull fracture without impression, n (%) 53 (39) 72 (53) 0.041 Skull fracture with impression, n (%) 5 (4) 10 (7) 0.210 Intracranial air, n (%) 18 (13) 39 (28) 0.003 Table 3  Fatality and discharge destinations in the GH group and the RTC group GH General hospital, IQR Interquartile range, N/A Not applicable, RTC​ Regional trauma centre, TBI Traumatic brain injury Bold font indicates statistical significance GH group RTC group p-value n = 137 n = 137 Length of stay in days, median [IQR] 5 [2, 12] 6 [3, 12] 0.169 In-hospital fatality, n (%) 60 (44) 22 (16) < 0.001 Cause of in-hospital fatality, n (%) 60 (44) 22 (16)  Intracranial hypertension 42 (70) 21 (95) 0.015  Complications of TBI 17 (28) 1 (5) 0.021  Unrelated to the injury 1 (2) 0 (0) N/A  Other trauma 0 (0) 0 (0) N/A Discharge destination, n (%) 77 (56) 115 (84)  Nursing home 25 (32) 0 (0) < 0.001  Home 22 (29) 2 (2) < 0.001  Specialised rehabilitation 18 (23) 8 (7) 0.001  Other rehabilitative institution 6 (8) 0 (0) N/A  Other hospital 6 (8) 105 (91) < 0.001 6-month fatality, n (%) 74 (54) 34 (25) < 0.001  Missing 0 (0) 8 (6) N/A 12-month fatality, n (%) 79 (58) 38 (28) < 0.001  Missing 0 (0) 8 (6) N/A Table 1  Patients characteristics in the GH group and the RTC group (n = 274) Table 2  Main findings at head CT scans in the GH group and the RTC group Number of missing cases not stated when frequency < 5% GCS Glasgow coma scale, GH General hospital, IQR Interquartile range, N/A Not applicable, RTC​ Regional trauma centre, SD Standard deviation, TBI Traumatic brain injury Bold font indicates statistical significance a Two patients were transferred to a different RTC than St. Elderly patientsh Olavs Hospital due to extraordinary circumstances b Statistical analyses were not conducted due to non-normality in the two groups In patients ≥ 65  years old, the in-hospital CFR was higher in the GH group than in the RTC group (62% vs. 24%, p < 0.001), as was the 6-month CFR (76% vs. 43%, p < 0.001) (Table 5). In patients ≥ 80 years old, the in-hospital CFR seemed to be higher in the GH group than in the RTC group, although not significantly dif- ferent (67% and 36%, p = 0.062), whereas the CFR at 6 months was similar (84% and 73%, p = 0.420). Rahim et al. ED Emergency department, GCS Glasgow coma scale, GH General hospital, RTC​ Regional trauma centre, TBI Traumatic brain injury *p < 0.05, **p < 0.01, ***p < 0.001 in statistical analyses between the GH group and the RTC group for this age category ED Emergency department, GCS Glasgow coma scale, GH General hospital, RTC​ Regional trauma centre, TBI Traumatic brain injury 70 years of age were discharged to palliative care and died, thus fatal outcome was observed in the acute phase in 6% of patients in this ED Emergency department, GCS Glasgow coma scale, GH General hospital, RTC​ Regional trauma centre, TBI Traumatic brain injury *p < 0.05, **p < 0.01, ***p < 0.001 in statistical analyses between the GH group and the RTC group for this age category Discussionh This is the first Scandinavian study describing adult patients with moderate and severe TBIs admitted to GHs, focusing on patients never transferred to RTCs. Of all patients initially admitted to a GH, 60% of patients with moderate and 42% of patients with severe TBI were man- aged entirely at the GH. Most (94%) of the patients pre- sumed to have a low risk of further deterioration remained stable and survived the acute phase. In addition to a large proportion of patients considered unsalvageable (23%), we found that the status neurosurgical intervention con- sidered nonbeneficial, especially among elderly individu- als with preinjury disability, was a common cause of not transferring the patient. More patients died in the acute phase in the GH group than in the RTC group, poten- tially reflecting the higher number of elderly patients in the GH group. Overall, half of the patients were managed entirely at the GH. The decision regarding the appropri- ate level of care for the patient was mostly made in coop- eration with a neurosurgeon. Most patients with EDH, multiple contusions or general oedema were transferred. Most of the transferred patients received neurosurgical procedures, indicating that patients evaluated as likely to benefit from neurosurgery were transferred.i The second group included elderly patients considered either unsalvageable or unlikely to benefit from neuro- surgical intervention. Patients managed entirely at the GH were older than patients transferred to the RTC, and only 16% of patients ≥ 80  years old were transferred to the RTC. Even among patients with severe TBI, almost half were never transferred to the RTC, despite a high degree of conferring with the RTC and current guide- lines recommending transfer [4, 28]. The most common rationale for not transferring patients with severe TBI was that the patients were considered unsalvageable, and most (75%) of these were ≥ 65 years old. It is likely that younger patients, even those with no hope of survival, were admitted directly to or rapidly transferred to the RTC, for instance, when involved in road traffic accidents [29]. Nevertheless, even when excluding patients catego- rised as unsalvageable, one-fourth of patients with severe TBI were managed entirely at the GH, many of whom were elderly patients where neurosurgical intervention was considered nonbeneficial. Elderly patients in the GH group had a high rate of preinjury disability, which increased with increasing age. Discussionh Age and comorbidity have been associated with less aggressive management [7, 30, 31] and lower rates of transferral [32–34] in previous studies, consistent with our findings. i We identified two main categories of patients managed entirely at the GHs: patients considered safe to manage at the GH level and elderly patients considered either unsalvageable or unlikely to benefit from neurosurgical intervention. i The choice of management level for elderly patients has been insufficiently studied since many studies exclude elderly patients with preinjury disability or significant comorbidity [35]. One study did not find an associa- tion between transfer to a level I or II trauma centre and long-term functional outcomes in older patients with all-severity TBI and concluded that routine transfer is not warranted [36]. Considering the increasingly older population in high-income countries and a demonstrated increase in TBIs in elderly individuals, this is an impor- tant topic that needs further investigation [7]. First, few patients with a presumed low risk of further deterioration died during the acute phase, including patients with a GCS score ≤ 8 at admission, indicating that communication between the managing physician at the GH and the on-call neurosurgeon, with shared access to the CT scan, may provide an accurate estimate of the patient’s prognosis; this finding is in line with other stud- ies [22, 23]. The four patients in this category who dete- riorated and died or entered a palliative setting within six months after the injury were all elderly and receiv- ing antithrombotic medication or were later diagnosed with a severe concurrent disease. It remains uncertain whether the four deaths could have been prevented by transferral to the RTC. This study extends the current knowledge regarding patients with moderate TBI [24], as we have demonstrated that 60% of the patients admit- ted to GHs were managed entirely at the GH. Several studies have investigated the opportunities for the man- agement of patients with TBI without transfer to a neuro- surgical department, but most of these studies have only included patients with mild TBI or have health care sys- tems not readily comparable to the Scandinavian health care model [25–27]. Elderly patientsh Scand J Trauma Resusc Emerg Med (2022) 30:68 Page 8 of 11 recommendations for the management of moderate and severe TBI at the GH level. Elderly patientsh Scand J Trauma Resusc Emerg Med (2022) 30:68 Page 7 of 11 Page 7 of 11 Table 4  Reasons for managing patients entirely at the GH (n = 137), with corresponding severity of TBI and fatality ED Emergency department, GH General hospital, IQR Interquartile range, TBI Traumatic brain injury a Two additional patients > 70 years of age were discharged to palliative care and died, thus fatal outcome was observed in the acute phase in 6% of patients in this category Unsalvageable Neurosurgical intervention considered non-beneficial Presumed low risk of further deterioration Another reason Unknown reason n = 32 (23%) n = 42 (31%) n = 49 (36%) n = 8 (6%) n = 6 (4%) Age, median [IQR] 79 [69, 85] 87 [82, 90] 58 [41, 71] 48 [37, 65] 43 [25, 78] Patients ≥ 65 years old, n (%) 24 (75) 40 (95) 16 (33) 2 (25) 2 (33) Patients ≥ 80 years old, n (%) 14 (44) 35 (83) 7 (14) 0 (0) 1 (17) Severe TBI at admission, n (%) 31 (97) 7 (17) 8 (16) 4 (50) 3 (50) Moderate TBI at admission, n (%) 1 (3) 31 (74) 32 (65) 4 (50) 3 (50) Mild TBI at admission, n (%) 0 (0) 4 (10) 9 (18) 0 (0) 0 (0) In-hospital fatality, n (%) 32 (100) 27 (64) 1a (2) 0 (0) 0 (0) 6-month fatality, n (%) 32 (100) 38 (90) 4 (8) 0 (0) 0 (0) Table 4  Reasons for managing patients entirely at the GH (n = 137), with corresponding severity of TBI and fatality Table 4  Reasons for managing patients entirely at the GH (n = 137), with corresponding severity of TBI and fatality Table 5  Characteristics of elderly patients in the GH group and the RTC group Table 5  Characteristics of elderly patients in the GH group and the RTC group Number of missing cases not stated when frequency < 5%  < 65 years old  ≥ 65 years old  ≥ 80 years old GH group RTC group GH group RTC group GH group RTC group n = 53 n = 95 n = 84 n = 42 n = 57 n = 11 Preinjury functional disability, n (%) 20 (38) 34 (36) 49 (58) 20 (48) 36 (63) 4 (36)  Missing 4 (8) 4 (4) 4 (5) 0 (0) 2 (4) 0 (0) Antithrombotic medication, n (%) 5 (9) 8 (8) 61 (73) 30 (71) 41 (72) 9 (82)  Missing 6 (11) 1 (1) 5 (6) 0 (0) 3 (5) 0 (0) Cause of injury, n (%)  Fall 35 (66)* 43 (45) 76 (90)* 32 (76) 53 (93)* 7 (64)  Motor vehicle accident 11 (21) 32 (34) 6 (7)* 9 (21) 3 (5)** 4 (36)  Other cause 5 (9) 16 (17) 2 (2) 1 (2) 1 (2) 0 (0) Severity of TBI at admission, n (%)  Severe TBI 20 (38)* 53 (56) 33 (39) 19 (45) 20 (35) 7 (64)  Moderate TBI 32 (60)** 33 (35) 39 (46) 15 (36) 31 (54)*** 0 (0)  Mild TBI 1 (2) 9 (9) 12 (14) 8 (19) 6 (11) 4 (36) Trauma team activation, (%) 24 (45)* 59 (62) 24 (29) 19 (45) 12 (21) 4 (36) Reason for managing patient at the GH, n (%)  Unsalvageable 8 (15) – 24 (29) – 14 (25) –  Neurosurgical intervention considered non-beneficial 2 (4) – 40 (48) – 35 (61) –  Presumed low risk of further deterioration 33 (62) – 16 (19) – 7 (12) –  Another reason 6 (11) – 2 (2) – 0 (0) –  Unknown reason 4 (8) – 2 (2) – 1 (2) – Neurosurgical procedure, n (%) – 70 (74) – 24 (57) – 6 (55) In-hospital fatality, n (%) 8 (15) 12 (13) 52 (62)*** 10 (24) 38 (67) 4 (36) 6-month fatality, n (%) 10 (19) 16 (17) 64 (76)*** 18 (43) 48 (84) 8 (73)  Missing 0 (0) 7 (7) 0 (0) 1 (2) 0 (0) 0 (0) 12-month fatality, n (%) 11 (21) 18 (19) 68 (81)*** 20 (48) 49 (86) 8 (73)  Missing 0 (0) 7 (7) 0 (0) 1 (2) 0 (0) 0 (0) Rahim et al. Discussionh On both the national and interna- tional levels, there is a lack of guidelines offering nuanced g CFRs were high in the elderly patients in the GH group for both patients≥ 65 years and ≥ 80 years old but also in patients ≥ 80  years old in the RTC group. Studies regarding the benefits of neurosurgical intervention in the elderly are conflicting [37–42]. Moreover, these stud- ies were conducted at neurosurgical referral centres and were likely biased towards cases considered suitable for surgical intervention. The in-hospital CFR was higher for patients ≥ 80 years old in the GH group versus the RTC group; however, the 6-month CFR was very high in both groups. This increase in fatality rate during the first few months demonstrates that studies on elderly patients must also consider post-discharge status when investigat- ing outcomes. Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 Page 9 of 11 Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 Whether the high CFR among elderly patients indi- cates a nihilistic approach or well-considered treatment- limiting decisions was not investigated in this study. The former possibility has been explored in several previous studies, which found that less aggressive management in elderly patients might increase mortality rates [31, 43]. In the latter study, increasing age was associated with reduced management intensity, irrespective of head injury severity, and an association between low manage- ment intensity and increased risk of 30-day mortality was found. However, in some cases, palliative care would be in the best interest of the patient [44], and our study shows that treatment-limiting decisions are common for elderly patients with severe TBI in our region, consist- ent with another Norwegian study [45]. Interestingly, a German and Austrian study found that even if neurosur- geons were willing to perform an emergency operation on an elderly patient with a life-threatening TBI [46], the elderly patients themselves might not wish for a life-pro- longing intervention if it would involve severe disability [47]. the case for the cause of death; the results must therefore be interpreted with caution. An additional limitation is that the present study used the HISS classification of moderate TBI (GCS scores of 9–13), instead of the range of 9–12 used in many current studies. Last, patients residing in the area where the RTC also serves as the GH were not included in this study. Abbreviations Abbreviations BTF: Brain Trauma Foundation; CFR: Case fatality rate; CT: Computed tomography; ED: Emergency department; EDH: Epidural haematoma; GCS: Glasgow Coma Scale; GH: General hospital; ICD-10: International Classification of Diseases, 10th revision; ICP: Intracranial pressure; ICU: Intensive care unit; msTBI: Moderate and severe TBI; RTC​: Regional trauma centre; tSAH: Traumatic subarachnoid haemorrhage; SDH: Subdural haematoma; SNC: Scandinavian Neurotrauma Committee; TAI: Traumatic axonal injury; TBI: Traumatic brain injury; TTA​: Trauma team activation. Conclusions GH GHs manage many patients with moderate or severe TBI, and there is a high degree of communication between the GHs and the RTCs about these patients. The patients managed entirely at the GHs differed from the patients transferred to the RTC by being older and having more preinjury disabilities. Hence, studies conducted at hos- pitals with a neurosurgical department will not be rep- resentative of all patients with moderate or severe TBI. Two main groups of patients were identified in the GH group: patients considered to have a low risk of further deterioration—a mostly accurate prediction, and elderly patients dying upon arrival or presenting with severe comorbidities considered not to benefit from neurosurgi- cal intervention due to assumed poor prognosis. Notably, few patients ≥ 80 years old were transferred to the RTC, and the 6-month CFR in this age group was high regard- less of transferral. Transferral to the RTC may therefore not be in the best interest of older patients with high premorbid frailty. Future guidelines should describe the management of moderate and severe TBI in the elderly population in greater detail, especially regarding the treatment level and benefit of transferral. As in geriatric medicine, the focus on frailty in the field of TBI has increased. Frailty reflects the individual’s physiological vulnerability beyond that of chronological age, and a TBI-specific frailty index was recently devel- oped and validated [48]. Management guidelines for TBI could probably be improved by incorporating the assess- ment of frailty [49]. Discussionh Consequently, the present study did not include all cases of moderate and severe TBI in central Norway; instead, it attempted to investigate the situation from the perspec- tive of GHs without neurosurgical services. Limitations Although the population-based design, covering all the GHs without neurosurgical services in the region over a ten-year period, was a strength of this study, the retro- spective nature of this study entails some limitations of the data material. First, the long inclusion period and the time passed since the last included patients, imply that the study results might not generalize to the current situ- ation at general hospitals. Second, ICD-10 codes to be reviewed were limited as described in the methods sec- tion, thereby excluding cases coded with S06.0 (concus- sion). We might therefore have missed a few patients with a normal head CT scan who clinically suffered a moder- ate or severe TBI due to isolated TAI. We might also have missed cases due to incorrect coding [12]. Third, the GCS score was in some cases not documented or could not be validly obtained from the medical records, for instance, due to concomitant alcohol intoxication, extracranial severe injuries affecting respiratory or circulatory func- tion, dementia, or other neurologic comorbidities. More- over, reasons for managing patients entirely at the GH were not always accurately described and can be espe- cially difficult to derive retrospectively, which was also Abbreviations BTF: Brain Trauma Foundation; CFR: Case fatality rate; CT: Computed tomography; ED: Emergency department; EDH: Epidural haematoma; GCS: Glasgow Coma Scale; GH: General hospital; ICD-10: International Classification of Diseases, 10th revision; ICP: Intracranial pressure; ICU: Intensive care unit; msTBI: Moderate and severe TBI; RTC​: Regional trauma centre; tSAH: Traumatic subarachnoid haemorrhage; SDH: Subdural haematoma; SNC: Scandinavian Neurotrauma Committee; TAI: Traumatic axonal injury; TBI: Traumatic brain injury; TTA​: Trauma team activation. Declarations 9. Norwegian National Advisory Unit on Trauma. Identifying the severely injured patient. In: National Trauma Plan. 2016. 9. Norwegian National Advisory Unit on Trauma. Identifying the severely injured patient. In: National Trauma Plan. 2016. Availability of data and materials h d d d l d d The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. 8. Sewalt CA, Gravesteijn BY, Menon D, Lingsma HF, Maas AIR, Stocchetti N, Venema E, Lecky FE. Primary versus early secondary referral to a specialized neurotrauma center in patients with moderate/severe traumatic brain injury: a CENTER TBI study. Scand J Trauma Resusc Emerg Med. 2021;29(1):113. Funding 6. Gabriel EJ, Ghajar J, Jagoda A, Pons PT, Scalea T, Walters BC. Guidelines for prehospital management of traumatic brain injury. J Neurotrauma. 2002;19(1):111–74. 6. Gabriel EJ, Ghajar J, Jagoda A, Pons PT, Scalea T, Walters BC. Guidelines for prehospital management of traumatic brain injury. J Neurotrauma. 2002;19(1):111–74. Open access funding provided by Norwegian University of Science and Technology. 7. Gardner RC, Dams-O’Connor K, Morrissey MR, Manley GT. Geriatric trau- matic brain injury: epidemiology, outcomes, knowledge gaps, and future directions. J Neurotrauma. 2018;35(7):889–906. 7. Gardner RC, Dams-O’Connor K, Morrissey MR, Manley GT. Geriatric trau- matic brain injury: epidemiology, outcomes, knowledge gaps, and future directions. J Neurotrauma. 2018;35(7):889–906. Supplementary Information The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s13049-​022-​01050-0. The online version contains supplem org/​10.​1186/​s13049-​022-​01050-0. Additional file 1: Table S1 Patients with moderate or severe TBI in the GHs of central Norway 01.10.2004–01.10.2014. Table S2 Preinjury func- tional disability and antithrombotic medication usage in the GH group versus the RTC group. Page 10 of 11 Page 10 of 11 Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 Acknowledgements 3. Unden J, Ingebrigtsen T, Romner B, Scandinavian Neurotrauma C. Scandi- navian guidelines for initial management of minimal, mild and moderate head injuries in adults: an evidence and consensus-based update. BMC Med. 2013;11:50. The authors would like to thank Beate Mærk Voll, Torun Elsie Wanvik Farnes, Stine Borgen Lund, and Oddrun Sandrød for management of the Trondheim msTBI study database. 4. Sollid S, Sundstrom T, Kock-Jensen C, Juul N, Eskesen V, Bellander BM, Wester K, Romner B. Scandinavian guidelines for prehospital man- agement of severe traumatic brain injury. Tidsskr Nor Laegeforen. 2008;128(13):1524–7. Consent for publication 13. Bjarko VV, Skandsen T, Moen KG, Gulati S, Helseth E, Nilsen TIL, Vik A. Time of injury and relation to alcohol intoxication in moderate-to-severe trau- matic brain injury: a decade-long prospective study. World Neurosurg. 2019;122:e684–9. 1. Andelic N, Sigurdardottir S, Brunborg C, Roe C. Incidence of hospital- treated traumatic brain injury in the Oslo population. Neuroepidemiol- ogy. 2008;30(2):120–8. Author details 1 Department of Neuromedicine and Movement Science, Faculty of Medi- cine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway. 2 Department of Public Health and Nursing, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway. 3 Department of Surgery, Levanger Hospital, Nord-Trøndelag Hospital Trust, 7600 Levanger, Norway. 4 Department of Surgery, St. Olavs Hospital, Trondheim University Hospital, 7006 Trondheim, Norway. 5 Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, 7491 Trondheim, Nor- way. 6 Department of Circulation and Medical Imaging, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7006 Trondheim, Norway. 7 Department of Neurosurgery, St. Olavs Hospital, Trondheim University Hospital, 7006 Trondheim, Norway. 8 Clinic of Physical Medicine and Rehabilitation, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway. 15. Stein SC, Ross SE. Moderate head injury: a guide to initial management. J Neurosurg. 1992;77(4):562–4. 16. Mena JH, Sanchez AI, Rubiano AM, Peitzman AB, Sperry JL, Gutierrez MI, Puyana JC. Effect of the modified Glasgow Coma Scale score criteria for mild traumatic brain injury on mortality prediction: comparing classic and modified Glasgow Coma Scale score model scores of 13. J Trauma. 2011;71(5):1185–92. 17. Stein SC, Spettell C. The Head Injury Severity Scale (HISS): a practical clas- sification of closed-head injury. Brain Inj. 1995;9(5):437–44. 18. Rundhaug NP, Moen KG, Skandsen T, Schirmer-Mikalsen K, Lund SB, Hara S, Vik A. Moderate and severe traumatic brain injury: effect of blood alcohol concentration on Glasgow Coma Scale score and relation to computed tomography findings. J Neurosurg. 2015;122(1):211–8. 19. Olsen M, Vik A, Lund Nilsen TI, Uleberg O, Moen KG, Fredriksli O, Lien E, Finnanger TG, Skandsen T. Incidence and mortality of moderate and severe traumatic brain injury in children: a ten year population-based cohort study in Norway. Eur J Paediatr Neurol. 2019;23(3):500–6. Received: 30 April 2022 Accepted: 22 November 2022 Received: 30 April 2022 Accepted: 22 November 2022 Received: 30 April 2022 Accepted: 22 November 2022 20. Department of medical biochemistry SOH. Ethanol (ethyl alcohol) in plasma and serum. In: Lab analyses and tests. 2020. 21. Lydersen S, Fagerland MW, Laake P. Recommended tests for association in 2×2 tables. Stat Med. 2009;28(7):1159–75. p g The authors declare that they have no competing interests. 14. Maas AI, Stocchetti N, Bullock R. Moderate and severe traumatic brain injury in adults. Lancet Neurol. 2008;7(8):728–41. Author contributions SR and TS contributed to the design of the study, participated in data collec- tion and analysis, and drafted the manuscript. EL and AV contributed to the design of the study. EHF read the CT scans. RP participated in data collection. All authors edited the manuscript and approved the final version. 5. Ingebrigtsen T, Romner B, Kock-Jensen C. Scandinavian guidelines for initial management of minimal, mild, and moderate head injuries. Scand Neurotrauma Committee J Trauma. 2000;48(4):760–6. 5. Ingebrigtsen T, Romner B, Kock-Jensen C. Scandinavian guidelines for initial management of minimal, mild, and moderate head injuries. Scand Neurotrauma Committee J Trauma. 2000;48(4):760–6. 2. Maas AIR, Menon DK, Adelson PD, Andelic N, Bell MJ, Belli A, Bragge P, Brazinova A, Buki A, Chesnut RM, et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurol. 2017;16(12):987–1048. Ethics approval and consent to participate The Regional Committee for Medical Research Ethics approved the Trondheim msTBI study (reference number 2009/2328), with the present study approved as an extension. The study was approved without the requirement of patient consent for the patients in the GH group. For the Trondheim msTBI study, written informed consent was obtained from the surviving patients or their next of kin for incapacitated patients. Permission was obtained from the Norwegian Directorate of Health to use data from deceased patients without consent from their next of kin. All data access committees of the individual GHs have approved the access to medical journal records (reference numbers 2017/1523, 17336/2017, 2017/1033-8, and 13/1373). 10. Steyerberg EW, Wiegers E, Sewalt C, Buki A, Citerio G, De Keyser V, Ercole A, Kunzmann K, Lanyon L, Lecky F, et al. Case-mix, care pathways, and outcomes in patients with traumatic brain injury in CENTER-TBI: a Euro- pean prospective, multicentre, longitudinal, cohort study. Lancet Neurol. 2019;18(10):923–34. 11. Dehli T, Gaarder T, Christensen BJ, Vinjevoll OP, Wisborg T. Implementa- tion of a trauma system in Norway: a national survey. Acta Anaesthesiol Scand. 2015;59(3):384–91. 12. Organization WH: Chapter XIX Injury, poisoning and certain other consequences of external causes (S00-T98) In: International statistical classification of diseases and related health problems 10th revision—ICD- 10 Version: 2019. 2020. Competing interests The authors declare that they have no competing interests. p g The authors declare that they have no competing interests. 1. Andelic N, Sigurdardottir S, Brunborg C, Roe C. Incidence of hospital- treated traumatic brain injury in the Oslo population. Neuroepidemiol- ogy. 2008;30(2):120–8. 2. Maas AIR, Menon DK, Adelson PD, Andelic N, Bell MJ, Belli A, Bragge P, Brazinova A, Buki A, Chesnut RM, et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurol. 2017;16(12):987–1048. References 22. Ashkenazi I, Haspel J, Alfici R, Kessel B, Khashan T, Oren M. Effect of teleradiology upon pattern of transfer of head injured patients from a rural general hospital to a neurosurgical referral centre. Emerg Med J. 2007;24(8):550–2. 23. Ashkenazi I, Zeina AR, Kessel B, Peleg K, Givon A, Khashan T, Dudkiewicz M, Oren M, Alfici R, Olsha O. Effect of teleradiology upon pattern of trans- fer of head injured patients from a rural general hospital to a neurosurgi- cal referral centre: follow-up study. Emerg Med J. 2015;32(12):946–50. Page 11 of 11 Page 11 of 11 Rahim et al. Scand J Trauma Resusc Emerg Med (2022) 30:68 24. Einarsen CE, van der Naalt J, Jacobs B, Follestad T, Moen KG, Vik A, Haberg AK, Skandsen T. Moderate traumatic brain injury: clinical characteris- tics and a prognostic model of 12-month outcome. World Neurosurg. 2018;114:e1199–210. 43. Kirkman MA, Jenks T, Bouamra O, Edwards A, Yates D, Wilson MH. Increased mortality associated with cerebral contusions following trauma in the elderly: bad patients or bad management? J Neurotrauma. 2013;30(16):1385–90. 44. Hwang F, Pentakota SR, Glass NE, Berlin A, Livingston DH, Mosenthal AC. Older patients with severe traumatic brain injury: national variability in palliative care. J Surg Res. 2020;246:224–30. 25. Capron GK, Voights MB, Moore HR 3rd, Wall DB. Not every trauma patient with a radiographic head injury requires transfer for neurosurgical evalua- tion: application of the brain injury guidelines to patients transferred to a level 1 trauma center. Am J Surg. 2017;214(6):1182–5. 45. Robertsen A, Forde R, Skaga NO, Helseth E. Treatment-limiting decisions in patients with severe traumatic brain injury in a Norwegian regional trauma center. Scand J Trauma Resusc Emerg Med. 2017;25(1):44. 26. Yun BJ, White BA, Benjamin Harvey H, Prabhakar AM, Sonis JD, Glover M, Vallillo E, Choi S, Borczuk P, Raja AS. Opportunity to reduce transfer of patients with mild traumatic brain injury and intracranial hemorrhage to a Level 1 trauma center. Am J Emerg Med. 2017;35(9):1281–4. 46. Unterhofer C, Hartmann S, Freyschlag CF, Thome C, Ortler M. Severe head injury in very old patients: to treat or not to treat? Results of an online questionnaire for neurosurgeons. Neurosurg Rev. 2018;41(1):183–7. 27. Joseph B, Friese RS, Sadoun M, Aziz H, Kulvatunyou N, Pandit V, Wynne J, Tang A, O’Keeffe T, Rhee P. The BIG (brain injury guidelines) project: defin- ing the management of traumatic brain injury by acute care surgeons. References J Trauma Acute Care Surg. 2014;76(4):965–9. 47. Unterhofer C, Ho WM, Wittlinger K, Thome C, Ortler M. “I am not afraid of death”—a survey on preferences concerning neurosurgical interventions among patients over 75 years. Acta Neurochir (Wien). 2017;159(8):1547–52. 28. Carney N, Totten AM, O’Reilly C, Ullman JS, Hawryluk GWJ, Bell MJ, Bratton SL, Chesnut R, Harris OA, Kissoon N, et al. Guidelines for the manage- ment of severe traumatic brain injury, fourth edition. Neurosurgery. 2017;80(1):6–15. 48. Galimberti S, Graziano F, Maas AIR, Isernia G, Lecky F, Jain S, Sun X, Gard- ner RC, Taylor SR, Markowitz AJ, et al. Effect of frailty on 6-month outcome after traumatic brain injury: a multicentre cohort study with external validation. Lancet Neurol. 2022;21(2):153–62. 29. Moen KG, Klepstad P, Skandsen T, Fredriksli OA, Vik A. Direct transport ver- sus interhospital transfer of patients with severe head injury in Norway. Eur J Emerg Med. 2008;15(5):249–55. 49. Newcombe VFJ, Chow A. The features of the typical traumatic brain injury patient in the ICU are changing: what will this mean for the intensivist? Curr Opin Crit Care. 2021;27(2):80–6. 30. Roe C, Skandsen T, Anke A, Ader T, Vik A, Lund SB, Mannskow U, Sollid S, Sundstrom T, Hestnes M, et al. Severe traumatic brain injury in Norway: impact of age on outcome. J Rehabil Med. 2013;45(8):734–40. Publisher’s Note S i N i 31. Skaansar O, Tverdal C, Ronning PA, Skogen K, Brommeland T, Roise O, Aarhus M, Andelic N, Helseth E. Traumatic brain injury-the effects of patient age on treatment intensity and mortality. BMC Neurol. 2020;20(1):376. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. 32. Sharma S, Gomez D, de Mestral C, Hsiao M, Rutka J, Nathens AB. Emer- gency access to neurosurgical care for patients with traumatic brain injury. J Am Coll Surg. 2014;218(1):51–7. 33. Visca A, Faccani G, Massaro F, Bosio D, Ducati A, Cogoni M, Kraus J, Servadei F. Clinical and neuroimaging features of severely brain-injured patients treated in a neurosurgical unit compared with patients treated in peripheral non-neurosurgical hospitals. Br J Neurosurg. 2006;20(2):82–6. 34. Munro PT, Smith RD, Parke TR. Effect of patients’ age on management of acute intracranial haematoma: prospective national study. BMJ. 2002;325(7371):1001. 35. Thompson HJ, Dikmen S, Temkin N. Prevalence of comorbidity and its association with traumatic brain injury and outcomes in older adults. Res Gerontol Nurs. 2012;5(1):17–24. 36. Nishijima DK, Gaona SD, Faul M, Tancredi DJ, Waechter T, Maloney R, Bair T, Blitz A, Elms AR, Farrales RD, et al. The association of trauma center transport and long-term functional outcomes in head-injured older adults transported by emergency medical services. Acad Emerg Med. 2020;27(3):207–16. 37. Herou E, Romner B, Tomasevic G. Acute traumatic brain injury: mortality in the elderly. World Neurosurg. 2015;83(6):996–1001. 38. Dang Q, Simon J, Catino J, Puente I, Habib F, Zucker L, Bukur M. More fateful than fruitful? Intracranial pressure monitoring in elderly patients with traumatic brain injury is associated with worse outcomes. J Surg Res. 2015;198(2):482–8. 39. Trevisi G, Sturiale CL, Scerrati A, Rustemi O, Ricciardi L, Raneri F, Tomatis A, Piazza A, Auricchio AM, Stifano V, et al. Acute subdural hematoma in the elderly: outcome analysis in a retrospective multicentric series of 213 patients. Neurosurg Focus. 2020;49(4):E21. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: Publisher’s Note S i N i Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 40. Wutzler S, Lefering R, Wafaisade A, Maegele M, Lustenberger T, Walcher F, Marzi I, Laurer H, TraumaRegister DGU. Aggressive operative treatment of isolated blunt traumatic brain injury in the elderly is associated with favourable outcome. Injury. 2015;46(9):1706–11. 41. Won SY, Dubinski D, Brawanski N, Strzelczyk A, Seifert V, Freiman TM, Konczalla J. Significant increase in acute subdural hematoma in octo- and nonagenarians: surgical treatment, functional outcome, and predictors in this patient cohort. Neurosurg Focus. 2017;43(5):E10. 42. Taussky P, Hidalgo ET, Landolt H, Fandino J. Age and salvageability: analysis of outcome of patients older than 65 years undergoing craniotomy for acute traumatic subdural hematoma. World Neurosurg. 2012;78(3–4):306–11. 42. Taussky P, Hidalgo ET, Landolt H, Fandino J. Age and salvageability: analysis of outcome of patients older than 65 years undergoing craniotomy for acute traumatic subdural hematoma. World Neurosurg. 2012;78(3–4):306–11.
https://openalex.org/W2586001590
https://jurnal.ar-raniry.ac.id/index.php/Share/article/download/%23907/714
English
null
EFFECTIVE CRISIS MANAGEMENT FOR ISLAMIC FINANCIAL INDUSTRY AND THE INSTITUTION OF HISBAH: LESSONS FROM GLOBAL FINANCIAL CRISIS
Share/Share: Jurnal Ekonomi dan Keuangan Islam
2,016
cc-by-sa
8,904
Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ Kata Kunci: Hukum Islam, Krisis Keuangan Global, Kebijakan Publik dan Sadd al-Dharai INTRODUCTION Whether inspired by the sincere desire to fulfill the religious obligations of the Muslim population or an ingenious marketing tactic, Islamic finance has successfully emerged as a viable alternative or, at minimum, a complement to conventional finance. Regardless of its true intention in inception, the development of Islamic finance is crucial indeed. It evidences the fact that there is no separation between spiritual and mundane in Islam. The Shariah, i.e. Islamic Law, is a code of conduct that, apart from its catering for the spiritual needs, encompasses all social, political, legal, and economic aspects of the Muslims’ lives. Hence, they are required to adhere to the Shariah commands in their economic endeavors, which include, but are not limited to, refraining from the major prohibited activities of riba (interest), gharar (grave uncertainty), maysir (high speculation), and jahl (ignorance). Profit and loss sharing, profit based on readiness to take risk, and all transactions backed by an asset are also among the unanimously accepted features of Islamic finance. Although Islamic banks have come out of the sub-prime and credit crunch in better shape than many of their conventional counterparts, banks do fail from time to time, and eventually an Islamic bank will too. This prospect raises some important issues because the bulk of Islamic banking is a voyage into the realm of unknown, including the consequences of a downfall. However, the recent financial crisis has brought this issue to the full front. As country after country in the advance economy was hit by it, the stakeholders in Islamic finance were automatically alarmed about any such scenario faced by Islamic finance in the near future. There are many questions that need to be answered. At the front of these questions is the possible response of the regulator in times of such crisis. Since Islamic finance is concerned with Shariah compliance, this kind of compliance has to be maintained in times of policy making and policy decision related to crisis management. However, since Islamic finance has thus far been safe from any severe crisis, there is a lack of certainty as to how would such a crisis be managed in the future. But conventional financial system has been through many such crises; it has leant some good lessons from its past experiences and has forgotten many others. EFFECTIVE CRISIS MANAGEMENT FOR ISLAMIC FINANCIAL INDUSTRY AND THE INSTITUTION OF HISBAH: LESSONS FROM GLOBAL FINANCIAL CRISIS Najeeb Zada1 Ahcene Lahsasna2 Muhammad Yusuf Saleem3 1INCEIF, Malaysia and Islamia College Peshawar 2,3 INCEIF, Malaysia 1*Email: najeebzada@icp.edu.pk ABSTRACT - The recent financial crisis resulted destructive effects on finance industry. Islamic financial industry (IFI) is still naïve and largely untested in the face of a major financial turmoil. Major issues and uncertainties of the insolvency of IFI include the issue of moral hazard, government bailouts, excessive risk taking and deposit insurance. This paper addresses the issue of crisis management in IFI from the perspective of al-Siyasah al-Shar’iyyah and attempt to derive public policy guidelines that are useful in developing a timely and efficient crises management framework for Islamic finance industry. By using qualitative methods, the study found that the global financial crisis resulted in great destruction of financial institution. Although Islamic finance was quite immune to the global crisis as compared to its conventional peer, concerns still exist. It is time that Islamic finance industry learns from the financial woes of the rest of the world. Key Words: Islamic Law, Global Financial Crisis, Public Policy, Sadd al-Dharai’ ABSTRAK - Krisis keuangan baru-baru ini mengakibatkan efek destruktif pada industri keuangan. Industri keuangan Islam (IKI) masih naif dan sebagian besar belum teruji dalam menghadapi gejolak keuangan besar. Isu utama dan ketidakpastian dari kebangkrutan IKI meliputi moral hazard, dana talangan pemerintah, pengambilan risiko yang berlebihan dan asuransi deposito. Makalah ini membahas isu manajemen krisis dalam IKI dari perspektif al-Siyasah al-Shar'iyyah dan berusaha mendapatkan pedoman kebijakan publik yang bermanfaat dalam mengembangkan kerangka kerja manajemen krisis yang tepat waktu dan efisien bagi IKI. Dengan menggunakan metode kualitatif, studi ini menemukan bahwa krisis keuangan global mengakibatkan kehancuran besar bagi industri keuangan. Meskipun keuangan Islam cukup kebal terhadap krisis global dibandingkan dengan keuangan konvensional, kekhawatiran masih ada. Sudah saatnya industri keuangan Islam belajar dari krisis keuangan dari seluruh dunia. SHARE | Volume 5 | Number 1| January - June 2016 SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 2 INTRODUCTION Thus, Islamic finance has the chance to benefit from all the good that can be offered by conventional and avoid what had made conventional finance suffer most. The objective of this paper is to draw such lesson as a roadmap for Islamic finance in terms of effective crises management. In addition, it is argued that one classical Islamic institution known as al-hisbah can also help SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 3 in making Islamic finance well prepared for any disastrous situation in the future. In detail the objective of the paper are: (1) To determine the most important facts and episodes of global financial crisis, (2) To highlight the causes of the crisis and the factors responsible on it, (3) To see the interconnectedness of al- hisbah with crisis related regulations in Islamic finance, (4) To suggest how Islamic finance can handle better with such situation in the future and (5) To recommend future directions to the policy makers. The structure of this paper consists of three section. Section one discusses the global financial crisis in 2007-2008. The causes and consequences as well as the damages of this event are briefly highlighted. The next section is an overview of the classical Islamic institution of al-hisbah and its connectedness and importance in the sphere of finance and business. This section briefly introduces the principles of sadd al-dharai, masalihmursalah, and legal maxims of Shariah. Section three outlines the important lessons that can be learnt by Islamic finance from the global financial crisis. These lessons can help regulators and policy makers adopt the most suitable approach to crisis management. The paper ends with a brief conclusion. METHODOLOGY OF THE STUDY In line with these objectives, the study follows qualitative methodology of research which suits the nature of the topic. There is no quantitative data required for it since the directive and objective is to help in policy and decision making. The relevant literature to global financial crisis is analyzed and the most important lessons, causes and future course of action are detected. The institution of al-hisbah is also explored as it matches regulatory and public policy essence. On the basis of these facts, the possible public policy direction for Islamic finance is arrived at. BACKGROUND TO GLOBAL FINANCIAL CRISIS Literature has sufficiently analyzed the panicky financial episodes of the past; but the bulk of it is related to the global financial crisis in 2007-2008 which has analyzed almost each aspect of it. The stage for financial crisis was set by large complex financial institutions (LCFIs) through “(1) originating high risk loans… and (2) pooling these loans to create securities that could be sold to SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 4 investors” (Wilmarth, 2011). But this was not the end of the story; “with the blessing of regulators”, LCFIs were allowed securitization through which they reduced their apparent risk and, consequently, lower their capital requirements by creating structured finance securities. These securities consisted of a variety of products matching the demand of customers according to their risk and return appetite, including: asset backed securities (ABS), residential mortgage- backed securities (RMBS) and commercial mortgage-backed securities (CMBS). This was the first level of securitization which was followed by the second layer of securitization including: collateralized debt obligations (CDOs) and (corporate) leveraged buyouts (LBOs). There was a third layer of securitization in the form of “CDOs-squared”. The risk of securitization was intensified through over the counter (OTC) credit derivatives called credit default swaps (CDS) which was equal to insurance in the event of default. The financial system’s risk exposure was further enhanced through synthetic CDOs (Wilmarth, 2011). According to an estimate by IMF, private-sector financial institutions issued about $15 trillion of ABS, MBS, and CDOs in global markets between 2000 and 2007, including $9 trillion issued in the United States and, as per a second study, $11 trillion of structured-finance securities were outstanding in the U.S. market in 2008 (Wilmarth, 2011). This sway of securitization motivated lenders to provide more credit to the borrowers without looking into their credit history (Leicht & Fitzgerald, 2007). What added to the gravity of the situation was the government’s decision not to regulate the securities’ market. However, concerns were raised regarding the substantial risk that was involved in securities and which was difficult to evaluate. The financial crisis was born from the gradual development of institutions that encouraged opportunities for risk taking without enough institutional checks against such behavior (J. L. Campbell, 2011). BACKGROUND TO GLOBAL FINANCIAL CRISIS Nevertheless, the champions of deregulation dominated the scene and even if efforts were made to control the situation, they were to no avail. However, once there was a decline in the housing prices starting from late 2006 as a result of subprime mortgages, customers began to default on their loans and the effect spread across the finance industry. Large mortgage companies in US like Fannie Mae and Freddie Mac were taken by government. The collateral damage caused by both mortgage backed and asset-backed securities was unprecedented. The bankruptcy of Lehman Brother added fuel to the fire. Among the great giants that succumbed or almost succumbed to this situation was American International Group which almost collapsed. Since its SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 5 failure would cause other institutions fall, the government provided an $85 billion loan in exchange for a 79.9% equity stake in the company—a bailout that eventually totaled $182 billion by March 2009 (J. L. Campbell, 2011). Financial Liberalization An important episode before the global financial crisis was a sway of deregulation of the market and one of the most important changes before the financial crisis was the repeal of the Glass-Steagall Act of 1933 which was introduced then to mitigate opportunism as well as excessive risk taking by the banking industry. These two were the considered the causes of the Great Depression. However, it was replaced by the Financial Services Modernization Act- a politically driven legislation- in 1999 in order to enable banks to compete efficiently in the international market (Stiglitz, 2003). This act freed many markets from the regulation of government. Especially, it terminated regulatory oversight from credit default swap which had a market of $60 trillion world-wide by late 2008 (Morgan, 2008). Financial services mergers occurred at a massive scale (J. L. Campbell, 2011) and ultimately “shadow banking system” grew rapidly. This banking system did not accept deposits like commercial banks do. Therefore, they were not bound by the same safety regulations as other traditional banks were, like the requirement for capitalization (Wolf, 2009). Further deregulation followed in 2004 when capital requirements were further eased, making credit availability much easy. It was allowed now to shift capital from safer to riskier and more profitable investments. Previously, a fixed percentage of each dollar of capital had to be reserved by securities’ firms to ensure safeguard in case of collapse. However, now these firms were allowed to use non-cash assets like asset- backed securities to avoid this risk. This made more money available for use as collateral to borrow money. The leverage ratio before this change was 12-1 which reached a peak of 33-1 after this change in policy (Blinder, 2009). Several other reforms in securities markets assisted each other to create further incentives for more risky investment behavior. All this meant that that credit flowed easily and excessively than before especially to the housing market. The opponents of regulations succeeded in their argument that they hinder the efficiency of these new and lucrative markets (J. L. Campbell, 2011). The availability and ease of loans is indeed the tool which was utilized and which led to the gloomy financial landscape of 2007-2008. Financial Liberalization This was made possible due to the financial innovation which was much boosted upon before the crisis but which was nothing more than a second name for issuing more and SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 6 more debt. As John Kenneth Galbraith expresses it: “financial innovation is simply another way to issue debt” (Samwick, 2009). In the case Northern Rack bank of UK, the result of this financial innovation was that customers could borrow 125 percent of their property value and up to six times their annual income: the boring days when customers could borrow only 75 percent of their property value and a maximum of three times their income were over (Dowd, 2009). The modern risk management tools and practices used by risk management practitioners are based on unreal assumptions: “they assume that financial risks follow Gaussian distributions (and so ignore the “fat tails” which really matter); they assume that correlations are constant (and ignore the fact that correlations tend to radicalize in crises and so destroy the portfolio diversification on which a risk management strategy might be predicated) and they make assumptions about market liquidity that break down when they are most needed (Dowd, 2009). Accordingly, the risk involved in the era of financial innovation were hidden, or even misunderstood until it was late to return and fix the matters. Taleb (2010) blames the unrealistic theories, models and tools used by the economic establishment cheatingly. He said that it is surprisingly the economics establishment should have been aware of the use the wrong tools and complete fiasco in the theories, but they kept pushing the warnings under the rug, or hiding their responses. The inadequacy of economic theories to explain the causes of run is expressed by Ayotte (2009) that economic theories can explain the consequences of panic runs when they occur, but they do not explain what triggers them in the first place. The dangers of macroeconomic models are explained by Leijonhfvud (2014) that macromodels that ignore problems of instability are dangerous to the health and welfare of untold millions of people. The same writers further hints towards the inadequacy of another model by concluding that my personal conclusion is that Walrasian equilibrium models are hopelessly inadequate for dealing with financial crises and their aftermaths. SHARE | Volume 5| Number 1| January - June 2016 Financial Liberalization To conclude, there were numerous factors responsible for the downfall of economies in 2007-2008. The factors responsible for crisis include regulators, corporate governance and risk management, excessive borrowing, government failure, ethical failure, mortgage-lending and securitization, derivatives and rating agencies. However, if there was one factor generating all these sources of failure, it was ‘moral hazard’ alone. A moral hazard is where one party is responsible for the interests of another, but has an incentive to put his or her SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ own interests first: the standard example is a worker with an incentive to shirk on the job (Dowd, 2009). The principal incentive problem is moral hazard. Moral hazard is the familiar concern that someone who is protected against the consequences of a risk has less incentive to take precautions against the risk (Ayotte & Skeel., 2009). Moral hazard in finance world can take place in many shapes such as a person may sell another person a financial product, like a mortgage, with the knowledge that it is against the interest of the buyer. He may pay himself huge bonuses from the funds that he manages on behalf of others. He may take risks the consequences of which are born by others. According to Dowd (2009), the most fundamental reason for the failure of risk management is economic in nature, there is strong incentive to take risk which leads to risk taking ultimately. In such scenario, risk management is like “war on drugs” which helps curing the disease but it can never win conclusively. No matter how prudent risk managers are, they are prone to orders from senior management, pressurized to take short cuts, produce lower risk number to decrease capital requirements and are, thus, not willing to rock the boat. In addition, if senior management itself is fed on remuneration package that induce excessive risk-taking, the result is not difficult to guess. As elaborated above, it is moral hazard that laid at the foundation of all that happened before and during the financial crisis are: excessive risk taking, debt creation in the name of financial engineering, multi-layers of securitization, financial liberalization and deregulation and billions of bailouts and assistance from the taxpayers’ money. Financial Liberalization However, the question worth pondering is that what causes this moral hazard which is the root of all problems in this case? According to Dowd (2009) The root problem is limited liability, which allows investors and executives the full upside benefit of their risk-taking, while limiting their downside exposure. Since modern corporations are founded on the notion of legal personality and limited liability, the parties involved in the financial market are aware of the mechanism of reaping its fruit. Campbell and Griffin (2006) said that rightly opine on this situation that one has to stretch the point to say that the executives of large public companies are exposed to the economic risks of failure in any significant way, and certainly they are more or less completely cocooned from the most fundamental market pressure, fear of personal bankruptcy. The root of the problem of limited liability is highlighted by Ho and Price (2011:25) in these words, the surprisingly tenacious mythology in Salomon v Salomon, that a company is a separate person whose liabilities are not necessarily shouldered by its proprietors, while promoting investment in SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 8 situations where there might be none, is also the prime cause of moral hazard in the global financial system. However, what the writers attribute here to limited liability is sensed long ago by Adam Smith when he criticized corporations. What the writers mentioned above argue about can be summarized with the help of a simple tree diagram as below: Diagram 1. An Anatomy of the Global Financial Crises The diagram is clear in essence. Although the sequencing may be changed if viewed from a different perspective, it is generally agreed that the problem starts when liability of parties involved in any transaction or market is limited. It directly leads to the issue of moral hazard that is further exhibited in different forms or branches. However, on the tops of all is the issue of deposit insurance. Ample of literature has elaborated how moral hazard is magnified by deposit insurance. It has been argued that financial institutions, especially banks, have unique prudential regulations like explicit insurance deposit guarantees. Whereas these guarantees are to make sure the stability of the system and avoid bank runs, they induce moral hazard problem in ex ante sense, i.e. Financial Liberalization before the failure of bank. Due to such implicit guarantee by governments, moral hazard is also created in the ex post sense after the banks have become or near to insolvency (Marinc & Vlahu, 2012). It hardly needs any argument that it is deposit insurance that makes depositors indifferent towards the performance and stability of the banks, since they are sure of the safety of their deposits. In addition, banks are also induced to take as much risk Diagram 1. An Anatomy of the Global Financial Crises The diagram is clear in essence. Although the sequencing may be changed if viewed from a different perspective, it is generally agreed that the problem starts when liability of parties involved in any transaction or market is limited. It directly leads to the issue of moral hazard that is further exhibited in different forms or branches. However, on the tops of all is the issue of deposit insurance. Ample of literature has elaborated how moral hazard is magnified by deposit insurance. It has been argued that financial institutions, especially banks, have unique prudential regulations like explicit insurance deposit guarantees. Whereas these guarantees are to make sure the stability of the system and avoid bank runs, they induce moral hazard problem in ex ante sense, i.e. before the failure of bank. Due to such implicit guarantee by governments, moral hazard is also created in the ex post sense after the banks have become or near to insolvency (Marinc & Vlahu, 2012). It hardly needs any argument that it is deposit insurance that makes depositors indifferent towards the performance and stability of the banks, since they are sure of the safety of their deposits. In addition, banks are also induced to take as much risk The diagram is clear in essence. Although the sequencing may be changed if viewed from a different perspective, it is generally agreed that the problem starts when liability of parties involved in any transaction or market is limited. It directly leads to the issue of moral hazard that is further exhibited in different forms or branches. However, on the tops of all is the issue of deposit insurance. Ample of literature has elaborated how moral hazard is magnified by deposit insurance. It has been argued that financial institutions, especially banks, have unique prudential regulations like explicit insurance deposit guarantees. SHARE | Volume 5| Number 1| January - June 2016 Financial Liberalization Whereas these guarantees are to make sure the stability of the system and avoid bank runs, they induce moral hazard problem in ex ante sense, i.e. before the failure of bank. Due to such implicit guarantee by governments, moral hazard is also created in the ex post sense after the banks have become or near to insolvency (Marinc & Vlahu, 2012). It hardly needs any argument that it is deposit insurance that makes depositors indifferent towards the performance and stability of the banks, since they are sure of the safety of their deposits. In addition, banks are also induced to take as much risk SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 9 as they can in order to increase value for their shareholders and ensure their personal bonuses etc. Indeed, deposit insurance is one of the major reasons that contributed to the collapse in 2007-2008. Market Regulation in the Earliest Islamic Era Even though small and simple, market was regulated in the early Islamic period by the Prophet peace be upon him and the four pious Caliphs, may Allah be pleased with them all. Numerous instances can be quoted in this regard. Theoretically, the Prophet peace be upon him stressed “individual” accountability in these words: “Beware that every one of you is a shepherd and everyone is answerable with regard to his folk. The caliph is a shepherd over the people and shall be questioned about them. A woman is a guardian over the house hold of her husband and his children and has to be questioned about them. A slave is a shepherd over the property of his master and shall be questioned about it”. Then he, may peace be upon him, adds: “beware that every one of you is a guardian and every one of you shall be questioned with regard to his trust”. The words of this hadith explicitly make each individual responsible for his/her subordinates. Thus, it is a manifestation of both individual and “institutional” or “organizational” responsibility in which those in charge of some task or some persons are responsible for their duties towards them accordingly. Dost (undated) argues that the institution of “hisbah” that is to be found in the later Islamic era was inspired by the practice of the Prophet peace be upon him, as he used to visit the market personally and check on the accuracy of measures. The Prophet also assigned specific places to different mongers and prohibited different market practices like meeting the producer-farmer at the entrance of the city before they reached the market, fearing they may sell at lower price. The Prophet peace be upon him was also opposed to price fixation at the time of scarcity when he replied those who complained high prices and requested to fix it: “the seller and the buyer, the one who provides and who fixes the prices are nobody but the God. I don’t wish to die whilst people having demands of life and property from me”. This conduct of the Prophet peace be upon him was taken as a role model and followed by the four Caliphs after him. Market Regulation in the Earliest Islamic Era In this regard, the first Caliph of Islam, Abu Bakkar, may Allah be pleased with him used to keep a check on his SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 10 collectors when they leave for their task or return from it. Thus, he asked Muaz ibn Jabal may Allah be pleased with him, to give account of what he had done. He replied, are we subject to two accounts, one towards Allah and the other towards you? However, he did as he was directed by the Caliph. Usmani (Usmani, 1414 H) narrates an incident of the practical supervision of Umar may Allah be pleased with him. Once he was on a visit to the market where he found that a person was selling something at a lower than the market price. Watching this, he ordered that person: “Either you increase in price, or go away from our market.” It has also been narrated that it was the habit of Umar may Allah be pleased with him to patrol in the market of Medina with his durrah in his hand and would guide those who needed correction (Shahatah, 1999). What the above instances show is that keeping a proper check and balance on the market of that time was considered an important task by the Prophet and his followers. Their personal interest in the issue reveal that disciplining the prevalent business practices of their respective time, though these may be of a small scale, held serious importance in their opinion. This consequently led to the establishment of the institution of hisbah in the Islamic realm. This institution and position of muhtasib will be discussed below. The Institution of Hisbah and Duties of Muhtasib Keeping in view the aforementioned examples from the earliest Islamic era, a separate institution for market supervision was established later which was called hisbah. This institution was formed based on the Islamic concept of “al amr bil maroof wal nahyu anil munkar” i.e. (the task of) enjoining what is right and forbidding what is evil. This is an essential feature of a given Islamic society and Quran stresses upon the establishment of this institution, either by the state or by the community members themselves. In fact Quran declares this task to be the sole purpose and duty of the Muslim Ummah: “You are the best Ummah ever raised for mankind. You bid the Fair and forbid the Unfair, and you believe in Allah.” Based on these Quranic injunctions, hisbah has been defined as a religious institution under the authority of the state that appoint people to carry out the responsibility of enjoining what is right, whenever people start to neglect it and forbidding what is wrong, whenever people start to engage in it (Yaacob & Donglah, 2012). This definition is broader enough to include both financial and non-financial obligation of a muhtasib, the person in charge of performing this duty. SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 11 In its strict financial and business sense, the duty of muhtasib was to ensure that the daily business transactions are done in a manner that is not harming the society (Rahim & al, 2012). The major responsibility of muhtasib was to ensure that the business practices prevalent in the market are in accordance with Islamic injunctions and each party is getting its due rights as well as performing its duties. It is not an exaggeration to state that the current day “regulator” has the same task to perform as was done by the classical muhtasib, although a regulators role is restricted to the extent of monitoring that everything in the financial markets is according to the regulations whereas muhtasib concentrated on the physical inspection of the market. This change is the direct outcome of the changes that have taken place in the business structure during the course of the centuries. The classical Muslim writer, Al-Mawardi (450 AH), has elaborated the duties of muhtasib in his famous book al-Ahkam al-Sultaniyya. SHARE | Volume 5 | Number 1| January - June 2016 The Institution of Hisbah and Duties of Muhtasib In fact he has devoted a full chapter of his book for this purpose which he titled as al-hisbah. The writer starts with outlining in detail the nine differences between a muhtasib and a person who performs the task of “enjoining the good and forbidding evil.” Regarding the conditions of muhtasib, Al-Mawardi (450 AH) opines that he should be free, just, of sound judgment, firm in religion, and be aware of what evil is.. He then elaborates how this institution lies in between two other institutions: the judiciary and the court of grievances. This is followed by a detailed description of the duties of muhtasib which the writer categorizes into two broader categories first: commanding the good and forbidding the evil. Commanding the good is further classified by him into three types and forbidding the evil into three types alike, each with its detailed rulings and ample examples. A muhtasib would hear cases related to different aspects. However, the financial matters dealt by him included: (1) Weights and measures, (2) Adulteration, (3) Supervision of certain sensitive professions and (4) Checking public places for proper maintenance or removing obstructions in the market (Niazi, 1991). Some duties that the writer assigns to the muhtasib include the prevention of contract or practices that is totally prohibited by Shariah, e.g. usury etc. He must also denounce any fraudulent practices in the market that harm any of the bargaining parties. Of supreme importance is the duty of muhtasib is to prevent the practice of giving short measures. To do this, he should test and control the market weights and measures appropriately and may even use a stamp of his own to ensure a single standard practice in the whole market. Al-Mawardi (450 AH) argues that if the jurisdiction of inspection is SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 12 large enough, the muhtasib can even appoint inspectors of investigation who would help him in performing his duty. These inspectors must be trustworthy person, should be paid from the Bait al-Maal and be immediately dismissed if it is known that they are involved in any malpractice themselves. The Institution of Hisbah and Duties of Muhtasib Thus, it can be concluded that accountability in general and financial accountability in particular has held great importance in both Islamic literature and the practice of Islamic state at the time of its inception. Whereas the two major sources of Islamic Shariah, namely Quran and Sunnah, provided a general framework as well as theoretical foundation for financial accountability, it was practically demonstrated by the Prophet peace be upon him and his followers since the early Islamic era. Both these steps led to the establishment of the institution of hisbah or market supervision in the later years. The rulings about hisbah and the duties of muhtasib, the market supervisor, were elaborated by Muslim jurists and scholars as we have seen Al- Mawardi (450 AH) as an example of it. This shows the deep concern of Islam with what can be termed in today’s world as “market regulation”. However, we have also observed that regulation in Islamic terminology is wider than its meaning in conventional term; the former encompassing one’s actions, thoughts, and even intentions. In his case study, Stilt (2008) reviews the application of law by muhtasib in the era of Mamaleek and arrives at some interesting conclusions. He defines muhtasib as: “an inspector of the markets and public spaces in general, was a legal official charged with “commanding right and forbidding wrong,” and was tasked with patrolling public streets, especially in the marketplaces, and enforcing laws as he understood them whenever he encountered a violation. According to the writer the general population also associated the muhtasib with availability and cost of food and “when prices rose and food was unavailable, the people often held the muhtasib responsible. In one important instance, Stilt (2008) shows how the higher authorities have to intervene in the market if the calamity is big enough. This was a large-scale crisis, encompassing both major cities and requiring grain to be brought from other parts of the sultanate, so it is not surprising that the price setting came down from the highest level. An important conclusion one can draw from this case study is that the rulings in fiqh may change according to circumstances of the people and their sufferings in a specific period. In other words, the law was applied in the context of different factors working from behind the screen. The Institution of Hisbah and Duties of Muhtasib Thus, the writer firmly concludes that we can see that the muhtasib’s actions resulted from a combination of social, SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 13 economic, and political factors; the potentially applicable laws; and the particular personality of the muhtasib. In a nutshell, the institution of hisbah has played a vital role in the past in the Muslim societies especially in the economic domain. This institution was effective in dealing with the crises’ situations and people looked at it in times of food scarcity and other times of distress. Masalih Mursalah Maslaha in simple terms means the benefit that Shariah has given consideration for the sake of people’s interest. One specific type of maslaha is where Shariah has left a specific matter to be decided by people according to their best interest and in line with Shariah guidelines; this type of maslaha is called maslahamursalah. Famous examples include the compilation of Quran and the order of taking qisas from a group of people for the murder of a single person (Al-Aamidi, 1983). There is no doubt that the bulk of modern day financial matters can fall under the category of maslahamursalah unless they do not violate the spirit of Shariah. The very existence of financial institutions like banks etc constitutes great interest for the masses. Therefore, their existence is justified on the basis of public interest even if they are not directly found in or derived from the sources of Shariah. In the context of crisis management, one can see the potential role that might be played by maslaha, especially when it is coupled with the legal maxims of Shariah. Consider, for instance, the case of deposit insurance or bailouts for financial institutions in times of crisis. Based on the notion of public interest, it is, at least apparently, in the best interest of masses at large to be protected through such mechanisms. Hence, providing these facilities for the sake of public confidence and their fund protection seems in line with the notion of maslaha. However, there is another angle of this too. As reviewed in the first section of this paper, deposit insurance and government bailouts lead to moral hazard as well. They can induce bankers take risks beyond their capacity and then make the general masses or taxpayers pay for it. To solve this issue, one needs to turn towards legal maxims for guidance. One of the maxims in this regard states that: averting harm takes precedence over achieving benefit (Laldin & al, 2013). With Looked at in a combination with these types of legal maxims, the important lesson from the global financial crisis seems to be that the issues of deposit insurance and governments’ bailouts needs careful deliberation. This is due to the fact that maslaha does not exist on the positive side only; on the other side, it is the second name for averting and avoiding the harm too. SHARE | Volume 5 | Number 1| January - June 2016 Sadd al-Dharai One unique feature of Islamic law is that it proactive in nature; it takes precautionary measures to stop a bad or evil from happening. Unlike other legal systems where such mechanism may not necessarily be found, Islamic law uproots the evil from the scratch and blocks its way from the beginning. Take the example of fornication or zina. Islam has prohibited this act and has also banned the ways that might lead to the commitment of this act, like free mixing up of the two genders and other practices that might ultimately lead to fornication. In fact Islamic law has gone one step further by disallowing acts that might even be permissible in the first place but they might lead to an evil or impermissible result in the long run. This notion is technically known as sadd al-dharai or simple blocking the permissible means to prevent unlawful things. One example given by scholars is selling weapons to people at the time of fitnah or selling wine to someone who is known to be manufacturing from it (al-Zuhaily, 1986). In contrast to the notion of maslaha or benefit, sadd al-dharai can be utilized as a handy tool for policy making in crisis management by not allowing something the result of which are known to be detrimental in the long run. This is especially true in the context of the global financial crisis. As explained in section 1, excessive lending led to credit boom which later on turned into one of the worst financial woos for humanity. Based on the principle of sadd al- dhariah, policy makers in Islamic finance needs to have an eye on this principle while deciding and mapping for the long run. If credit creation is allowed for Islamic banks, should there be a limit for it or not? Similar questions can be asked about derivatives too. No doubt that the existence of moral hazard will definitely hamper any reactive measures when times are bad. Therefore, sadd al-dhariah has to be utilized as a proactive measure in Islamic finance policy making. Masalih Mursalah SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 14 Excessive Debt Creation It was seen above that excessive lending was the founding step towards the financial destruction in 2008. It was in the form of securitization and derivatives etc. However, Islamic commercial law has a unique mechanism to deal with this issue. This mechanism consists of not only reactive measures but is also proactive in nature. In the first place, Islam wants debt not to be taken except in the case of real need. Secondly, there is a security measure in the form of rahn to ensure safe repayment of debt created. The third layer is that of insolvency law which further restricts the options of escape of avoidance of debt obligations. In fact the uniqueness of Islamic stance on the issue of debt can be easily understood from the fact that the longest verse of the Holy Quran deals with securing debt. Lastly, non-payment of debt is announced as one of the gravest sins, making the borrower morally and religiously obliged to pay it at any cost. Thus, these layers after layers provide a natural safety mechanism against excessive risk taking that was evident in global financial crisis. The above facts and the West’s experience with debt during the global financial crisis should be an indication for Islamic finance policy makers to reconsider issues like fractional banking reserve system that sometimes makes debt creation possible to unlimited extent. It also calls for an insolvency regime that clearly distinguished the rights and obligations of the parties to financial markets in the case of insolvency, bankruptcy and financial distress. Such a regime, however, is missing currently and it is not clear what will happen in case the Islamic finance industry faces a crisis at mass level. RECOMMENDATIONS FOR POLICY FORMATION IN ISLAMIC FINANCE The discussion in above two sections leads to a few conclusions that can be put into suggestions and recommendation for regulators and policy makers in Islamic finance. However, it can be seen that the following points are not clearly distinguishable and all of them are closely interconnected. The reason to put them separate is to show that they are individually important in terms of crisis management and need special focus of the policy makers. SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 15 Legal Personality and Limited Liability As explained above, it is the modern principle of limited liability that lays at the heart of this whole debate since this principle leads to all other moral hazard problems. Once again, Islam has addressed the issue explicitly, for instance, in the form of legal maxims like: “Benefit goes with liability and Liability accompanies gain” (Laldin & al, 2013). Apparently, the maxims indicate the liability cannot be limited if the gain is unlimited. This is also asserted by Usmani who accepted the notion of legal personality and limited liability for large public corporations only due to necessity. However, many of the contemporary scholars defend the acceptance of these principles on different grounds that sound reasonable. SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 16 But one cannot forget the role played by moral hazard during the crisis which was the direct outcome of limiting liability. It was elaborated above that legal personality and limited liability have been questioned in the West since the time of Adam Smith. It is also accepted that both these principles are required for modern day corporations to work. But it does not mean that precautionary steps are not available to solve, or at least minimize, the negative impact of these. Consider some suggestions by Schluter (2000) to address this issue. Firstly, the shareholders may be made liable for ten percent of the company’s debt if insolvency happens. Under this principle, the shareholders would be liable according to their respective shares in the company and not according to their personal wealth; i.e. the largest and not the wealthiest should be liable. Secondly, company law could also be changed to move away from this principle. Risk Taking and Moral Hazard If risk is taken with the knowledge that its consequences will be born by other parties, it has no limits as its destruction can be unlimited. Many of these moral hazards involve increased risk-taking: if I can take risks that you have to bear, then I may as well take them; but if I have to bear the consequences of my own risky actions, I will act more responsibly. Thus, inadequate control of moral hazards often leads to socially excessive risk-taking and excessive risk-taking is certainly a recurring theme in the current financial crisis (Dowd, 2009). Risk taking can come in numerous forms and risk that is natural to the business and finance environment is acceptable, rather essential, to be taken and even mitigated. But risk creation in the form of toxic instruments is highly questionable. These toxic instruments may produce good results in the short- run, but their long term damaging effect is visible even today after years of the global financial crisis. Considering the legal maxim, averting harm takes precedence over achieving benefit (Laldin & al, 2013), toxic risky instruments need to be reconsidered. Shariah compliant derivatives etc. need to be scrutinized and controlled before they turn into something uncontrollable. Perhaps the small harm caused by the absence of these instruments is far better than the damages it caused during the global financial crisis. It is only a matter of time to realize it. An excellent example of risk taking is provided by Al-Mawardi (450 AH) states that the muhtasib must prevent ship- owners from carrying a load which exceeds the capacity of the ship, lest it should lead to its sinking; likewise, he should stop them sailing when strong winds are blowing. The writer understood that the ship owners, in their quest SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 17 for more profit, may put the money of other people at stake and, therefore, it is the responsibility of muhtasib, the regulator, to control such behavior. Governments’ Bailouts Government bailout to the troubled institutions is a very serious issue in crisis management. It is often argued that bailing out large financial institutions in crises is something not avoidable; if not provided, the whole system can collapse. However, this may not be the case and it is sometimes wise not to make taxpayer suffer for the sake of too big to fail. Ayotte and Skeel (2009) provide the example Lehman Brothers collapse and AIG bailout to prove that bailing out and letting fall may produce the same results which shows that insolvency is not always a bad option. Table 1. Comparative Statistics Index between Lehman Brother and AIG Index Lehman Brothers AIG 12 Sep 15 Sep Change 16 Sep 17 Sep Change S&P 500 1251.7 1192.7 -4.71% 1213.6 1156.39 -4.71% Volatility (VIX) 25.66 31.7 23.54% 30.3 36.22 19.54% TED Spread 1.35 2.01 0.66 2.19 3.02 0.84 13-weeks treasury bill 1.46 0.81 -0.65 0.86 0.02 -0.84 Source: Ayotte and Skeel (2009). Table 1. Comparative Statistics Index between Lehman Brother and AIG What is demonstrated from these numbers is that contrary to the expectations from AIG bailouts, the response from the market was not so positive. On the one hand, this shows how if the confidence is lost, it is difficult to restore it especially in times of crisis. On the other hand, it is evident that bankruptcy has to be kept open as an option in times of need. Deposit Insurance It is suggested that the policy makers in Islamic finance should reconsider the deposit insurance. It is the root cause of moral hazard in conventional finance and has been questioned in conventional finance too. If the investors who fund a financial institution by lending money or buying stock anticipate that the firm will be rescued if it runs into trouble, they may extend funding beyond what they would extend otherwise (Ayotte and Skeel, 2009). It has also been argued that deposit insurance strongly affect market discipline negatively by making depositors indifferent towards the performance of banks. Consequently, banks SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 18 are induced to take risk that is not suitable at all; the depositors will not bother the banks’ risk taking as long as they are sure about their deposits. Even if not abolished, its mechanism should be such that moral hazard is avoided as much as possible. One suggestion given in the wake of recent crisis is that deposit insurance should be dependent on two factors: the size of the financial institution and the risk it takes. These steps will make financial institutions make responsive towards this issue and make lees risky investment, i.e. risk that they can pay for first. CONCLUSIONS AND RECOMMENDATIONS The global financial crisis resulted in great destruction and the world has not yet recovered from it. It is not known how much time it takes the finance world to stand back on its feet again after the collapse. However, there is something positive that the crisis gave the world; it is the regulatory and policy lessons that can be derived from it. Although Islamic finance was quite immune to the global crisis as compared to its conventional peer, concerns still exist. It is time that Islamic finance industry learns from the financial woes of the rest of the world. In fact the world is already changing its financial landscape with heavy reforms and stricter regulations like the new Basel requirements etc. However, the policy makers of Islamic finance should understand that it is one thing to fight against evil; it is absolutely different to curb the very roots of the evil. The West may succeed in its struggle but its success will be temporary only. If crisis management policy for Islamic finance is based on the principles elaborated in this paper like hisbah, maslaha, and sadd al-dhariah, it is hoped that the solutions will be everlasting. In line with the nature of Islamic law in general, such a policy has to be proactive and address the relevant issues before they occur, instead of being reactive and just wait for a crisis in Islamic finance industry to happen first and then think of solutions. It is time for the industry and the policy makers of Islamic finance to revisit their existing strategies and come up with solutions that comply with the spirit of Shariah and can become a role model for the rest of the world. Al-Aamidi. (1983). Al-Ihkam fee Usool al-Ahkam (Vol. 3). Beirut: Dar al- Aafaq al-Jadeedah. REFERENCES Al-Aamidi. (1983). Al-Ihkam fee Usool al-Ahkam (Vol. 3). Beirut: Dar al- Aafaq al-Jadeedah. SHARE | Volume 5| Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management For Islamic Finance Industry_ 19 al-Mawardi. (450 AH). al-Ahkam al-Sultaniya wa al-Wilaya al-Diniyya ("The Laws of Islamic Governance", translated by Asadullah Yate). London: Taha. al-Zuhaily, Wahbah. (1986). Usool al-Fiqh al-Islami. Damascus: Dar al-Fikr. Ayotte, & Skeel., D.A Jr. (2009). Bankruptcy or Bailouts? Journal of Corporation Law. Corporation Law, 35, 469-498. Blinder, A. (2009). Six Blunders En Route to a Crisis. The New York Times, p. 7. Campbell, D, & Griffin, S. (2006). Enron and the End of Corporate Governance. In S.MacLeod (ed.) Global Governanceand the Quest for Justice. Oxford: Hart Publishing. Campbell, J L. (2011). The U.S. Financial Crisis: Lessons for Theories of Institutional Complementarity. Socio-Economic Review, 9, 211-234. Dowd, K. (2009). Moral Hazard and the Financial Crisis. Cato, 29. Laldin, & al, et. (2013). Islamic Legal Maxims and Their Application in Islamic Finance, . Paper presented at the International Shariah Research Academy for Islamic Finance (ISRA) Kuala Lumpur. Leicht, K. T, & Fitzgerald, S. (2007). Postindustrial Peasants. New York. Leijonhufvud, A. (2014). Economics of the crisis and the crisis of economics, Euro. J. History of Economic Thought P. Retrieved from http://dx.doi.org/10.1080/09672567.2014.927519 Marinc, M., & Vlahu, R. (2012). The Economics of Bank Bankruptcy Law. London. Morgan, G. (2008). Market Formation and Governance in International Financial Markets: The Case of OTC Derivatives’, Human Relations. Human Relations, 61, 637-660. Niazi, Liaquat Ali Khan. (1991). Islamic Law of Contract. Lahore, Pakistan: Research Cell, Dyal Sing Trust Library. Rahim, & al, et. (2012). A New Breed of Islamic Accountants And Their Role In Islamic Institutions. Advances in Natural and Applied Sciences. Samwick, A. A. (2009). Moral Hazard in the Policy Response to the 2008 Financial Market Meltdown. Cato, 29. SHARE | Volume 5 | Number 1| January - June 2016 Zada, Lahsasna, & Saleem | Effective Crisis Management for Islamic Finance Industry_ 20 Schluter, M. (2000). Risk, reward and responsibility. Cambridge papers, 9. Schluter, M. (2000). Risk, reward and responsibility. Cambridge papers, 9. Shahatah. (1999). Hurmat al-Maal al-Aam fi Zoai Shariah al-Islamiyya. Cairo: Dar al-Nashar li al-Jamiat. Stiglitz, J. (2003). The Roaring Nineties. New York. Stilt, K. (2008). Price Setting and Hoarding in Mamluk Egypt: The Lessons of Legal Realism for Islamic Legal Studies, in The Law Applied: Contextualizing the Islamic Shari’a. London. Taleb, N N. (2010). Why Did The Crisis of 2008 Happen? Retrieved from http://ssrn.com/abstract=1666042 Usmani. (1414 H). Islam Aor Jaded Maeeshat Aor Tijarat. Karachi: Idaratul Maarif. Wilmarth, A, E Jr. (2011). The Dodd-Frank Act: A Flawed and Inadequate Response to the Too-Big-to-Fail Problem 89 ORE. L. REV., 951-1057. Wolf, M. (2009). Seeds of Its Own Destruction. Financial Times. Yaacob, H, & Donglah, N. K. (2012). Shari’ah audit in Islamic financial institutions: The postgraduates‟ perspective. International Journal of Economic and Finance. SHARE | Volume 5| Number 1| January - June 2016 SHARE | Volume 5| Number 1| January - June 2016
https://openalex.org/W3201760937
https://pure.eur.nl/ws/files/50845742/Remote_magnetic_navigation_shows_superior_long_term_outcomes_in_pediatric_atrioventricular_nodal_tachycardia_ablation_compared_to_manual_radiofrequency_and_cryoablation.pdf
English
null
Remote magnetic navigation shows superior long-term outcomes in pediatric atrioventricular (nodal) tachycardia ablation compared to manual radiofrequency and cryoablation
International journal of cardiology. Heart & vasculature
2,021
cc-by
8,040
IJC Heart & Vasculature Remote magnetic navigation shows superior long-term outcomes in pediatric atrioventricular (nodal) tachycardia ablation compared to manual radiofrequency and cryoablation Remote magnetic navigation shows superior long-term outcomes in pediatric atrioventricular (nodal) tachycardia ablation compared to manual radiofrequency and cryoablation Anna M.E. Noten a, Janneke A.E. Kammeraad b, Nawin L. Ramdat Misier a, Sip Wijchers a, Ingrid M. van Beynum b, Michiel Dalinghaus b, Thomas B. Krasemann b, Sing-Chien Yap a, Natasja M.S. de Groot a, Tamas Szili-Torok a,* a Department of Cardiology, Erasmus MC, University Medical Center, Rotterdam, the Netherlands1 b Department of Pediatric Cardiology, Sophia Children’s Hospital, Erasmus MC, University Medical Center, Rotterdam, the Netherlands1 A B S T R A C T Keywords: Catheter ablation Remote magnetic navigation Cryoablation Radiofrequency ablation Supraventricular tachycardia Atrioventricular reentry tachycardia Atrioventricular nodal reentry tachycardia Pediatrics Background: Catheter ablation (CA) is the first-choice treatment for tachyarrhythmia in children. Currently available CA techniques differ in mechanism of catheter navigation and energy sources. There are no large studies comparing long-term outcomes between available CA techniques in a pediatric population with atrio­ ventricular reentry tachycardia (AVRT) or atrioventricular nodal reentry tachycardia (AVNRT) mechanisms. Objective: This study aimed to compare procedural and long-term outcomes of remote magnetic navigation- guided radiofrequency (RF) ablation (RMN), manual-guided RF ablation (MAN) and manual-guided cryoa­ blation (CRYO). i y y y y Objective: This study aimed to compare procedural and long-term outcomes of remote magnetic navigation- guided radiofrequency (RF) ablation (RMN), manual-guided RF ablation (MAN) and manual-guided cryoa­ blation (CRYO). i Methods: This single-center, retrospective study included all first consecutive CA procedures for AVRT or AVNRT performed in children without structural heart disease from 2008 to 2019. Three study groups were defined by the ablation technique used: RMN, MAN or CRYO. Primary outcome was long-term recurrence of tachyarrhythmia. Results: In total, we included 223 patients, aged 14 (IQR 12–16) years; weighting 56 (IQR 47–65) kilograms. In total, 108 procedures were performed using RMN, 76 using MAN and 39 using CRYO. RMN had significantly lower recurrence rates compared to MAN and CRYO at mean follow-up of 5.5 ± 2.9 years (AVRT: 4.3% versus 15.6% versus 54.5%, P < 0.001; AVNRT: 7.7% versus 8.3% versus 35.7%, P = 0.008; for RMN versus MAN versus CRYO respectively). In AVNRT ablation, RMN had significantly lower fluoroscopy doses compared to CRYO [30 (IQR 20–41) versus 45 (IQR 29–65) mGy, P = 0.040). i Conclusion: In pediatric patients without structural heart disease who underwent their first AV(N)RT ablation, RMN has superior long-term outcomes compared to MAN and CRYO, in addition to favorable fluoroscopy doses. Contents lists available at ScienceDirect Contents lists available at ScienceDirect Available online 1 October 2021 2352-9067/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1016/j.ijcha.2021.100881 Received 1 August 2021; Received in revised form 17 August 2021; Accepted 20 September 2021 IJC Heart & Vasculature 37 (2021) 100881 Remote magnetic navigation shows superior long-term outcomes in pediatric atrioventricular (nodal) tachycardia ablation compared to manual radiofrequency and cryoablation Anna M.E. Noten a, Janneke A.E. Kammeraad b, Nawin L. Ramdat Misier a, Sip Wijchers a, Ingrid M. van Beynum b, Michiel Dalinghaus b, Thomas B. Krasemann b, Sing-Chien Yap a, Natasja M.S. de Groot a, Tamas Szili-Torok a,* a Department of Cardiology, Erasmus MC, University Medical Center, Rotterdam, the Netherlands1 b Department of Pediatric Cardiology, Sophia Children’s Hospital, Erasmus MC, University Medical Center, Rotterdam, the Netherlands1 A R T I C L E I N F O A B S T R A C T Contents lists available at ScienceDirect IJC Heart & Vasculature journal homepage: www.sciencedirect.com/journal/ijc-heart-and-vasculature IJC Heart & Vasculature 37 (2021) 100881 Remote magnetic navigation shows superior long-term outcomes in pediatric atrioventricular (nodal) tachycardia ablation compared to manual radiofrequency and cryoablation Anna M.E. Noten a, Janneke A.E. Kammeraad b, Nawin L. Ramdat Misier a, Sip Wijchers a, Ingrid M. van Beynum b, Michiel Dalinghaus b, Thomas B. Krasemann b, Sing-Chien Yap a, Natasja M.S. de Groot a, Tamas Szili-Torok a,* a Department of Cardiology, Erasmus MC, University Medical Center, Rotterdam, the Netherlands1 b Department of Pediatric Cardiology, Sophia Children’s Hospital, Erasmus MC, University Medical Center, Rotterdam, the Netherlands1 A R T I C L E I N F O A B S T R A C T Contents lists available at ScienceDirect IJC Heart & Vasculature journal homepage: www.sciencedirect.com/journal/ijc-heart-and-vasculature IJC Heart & Vasculature 37 (2021) 100881 1 This author takes responsibility for all aspects of the reliability and freedom from bi 2.3. Definitions radiofrequency (RF) ablation in children [1,4,6]. Comparable to the adult population, other CA techniques also became available in pediatric cardiology over time. Manual guided cryoablation and remote magnetic navigation (RMN) guided RF ablation were first used in children in 2003 [7,8] and 2009 [9–11] respectively and became standard-of-care treat­ ment options. Theoretically, these techniques might have advantages compared to the manual RF technique. Cryothermal energy allows cryomapping (with reversible effect) before establishing a permanent lesion. Furthermore the cryocatheter’s tip adheres to the myocardium during energy application. These could be beneficial, especially when targeting the proximity of vulnerable cardiac structures such as the His bundle [8,12]. With respect to RMN-guided RF ablation, the ablation catheter has a flexible, magnetically enabled distal tip and is navigated toward and held to the ablation surface by magnetic force, facilitating an omnidirectional positioning of the catheter with high stability [13]. Prior reports demonstrated improved lesion formation and superior safety compared to manual RF ablation [14,15], which also could be very beneficial in treatment of younger and smaller patients [9,11]. Total procedure time was defined as the time from first skin puncture until the removal of sheaths. The endpoints of acute procedural success were defined as the elimination of accessory pathway (AP) conduction for AVRT and non-inducibility of the tachycardia (i.e. not even a single echo beat was accepted) for AVNRT. Recurrence was defined as an episode of sustained SVT recorded on 12-lead ECG, 24-hour to 7-day continuous rhythm observation and/or implantable loop recorder or pacemaker or ICD. Minor complications were access site complications, temporary bundle branch block or AV block, pericardial effusion not requiring intervention. Major complications were permanent second- or third-degree atrioventricular block, cardiac tamponade, hemorrhagic shock, stroke and procedure-related death. Access site complications were vessel wall and bleeding complications which required surgical intervention or intervention by radiologist, prolonged hospitalization and/or an Hb drop of > 1.8 mmol/L. 2.1. Study design This single-center, retrospective study investigated all children with a structural normal heart, undergoing their first CA procedure for AVNRT and AVRT. Outcomes were compared between three groups, based on the ablation technique used: RMN-guided RF ablation (RMN), manual guided RF ablation (MAN), and manual guided cryoablation (CRYO), for each of the two SVT mechanisms separately. Primary endpoint was documented recurrence of tachyarrhythmia and/or (if present pre-procedurally) recurrence of pre-excitation on 12-lead ECG during long-term follow-up. We also analyzed the following secondary endpoints: procedural parameters, acute procedural success, the need for repeat procedures and complication rates. The study protocol con­ forms to the ethical guidelines of the 1975 Declaration of Helsinki. Procedural informed consent was obtained from all the patients prior to the EPS. The local medical ethics committee determined this study was not subject to the Dutch Medical Research Involving Human Subjects Act (WMO). The need for individual consent was waived by the local medical ethical committee. 2.4. Data collection i In the pediatric population, several small studies were conducted and reported first experiences with the mentioned techniques [2,3,8,9,11,16,17]. There are no large scale studies comparing ablation techniques in children, nor investigating long-term outcomes. Hence, this study aimed to compare procedural- and long-term outcomes of three ablation techniques - RMN-guided RF ablation, manual-guided RF ablation and manual-guided Cryoablation - in a large group of children diagnosed with AVRT and AVNRT with a structural normal heart. We hypothesized that ablative therapy of AV(N)RT guided by RMN or cry­ othermal energy has improved short- and long-term outcomes compared to manual RF ablation. Baseline demographic and clinical characteristics were collected from the institutional electronic patient dossier [HiX version 6.1 (ChipSoft BV, Amsterdam, NL)]. Procedural data was derived both from the electronic medical files, as well as from the electronic procedural log files recorded with EP-workmate (St. Jude Medical Inc., St. Paul, MN, USA), CryoConsole (Medtronic, Minneapolis, MN, USA) and Odyssey Cinema (Stereotaxis Inc., St. Louis MO, USA) systems. All patient in­ formation was de-identified. 1. Introduction been used as treatment of these SVT since 1990 [1,4]. Over the years, the indication of CA in pediatric patients has evolved from treatment of drug refractory tachyarrhythmia only, to a first-line treatment and patient preference [1,5]. In children without structural heart disease, the majority of supra­ ventricular tachyarrhythmias (SVT) consist of atrioventricular nodal reentrant tachycardia (AVNRT) and atrioventricular reentrant tachy­ cardia (AVRT) [1–3]. Catheter ablation (CA) in pediatric cardiology has Initially, large-scale registries were conducted which reported favorable success and complication rates of manual guided Abbreviations: AP, accessory pathway; AVNRT, atrioventricular nodal reentry tachycardia; AVRT, atrioventricular reentry tachycardia; CA, catheter ablation; CHD, congenital heart defect; CRYO, cryoablation; DAP, dose area product; EAM, electro-anatomic mapping; ECG, electrocardiogram; MAN, manual; RF, radio­ frequency; RMN, remote magnetic navigation; SVT, supraventricular tachycardia. * C di h E-mail address: t.szilitorok@erasmusmc.nl (T. Szili-Torok). 1 Available online 1 October 2021 2352-9067/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). A.M.E. Noten et al. IJC Heart & Vasculature 37 (2021) 100881 2. Methods All antiarrhythmic drugs were discontinued for > 5 half-lives before the procedure. All procedures were performed under general anesthesia. Vascular access was attained with placement of sheaths and diagnostic catheters in left and/or right femoral vein and/or artery, with sheath sizes and catheter types varying based on operator preference. The choice between a transseptal or retrograde aortic approach in case of left-sided AP was also based on operator preference. A standard EPS was performed with diagnostic catheters positioned in the coronary sinus (CS), right ventricular apex, and in the His position. The diagnosis of dual atrioventricular nodal pathways, atrioventricular nodal reentrant tachycardia, or AP-mediated tachycardia was made using standard diagnostic criteria. Regarding AVNRT ablation, Koch’s triangle was mapped by bending and pulling the ablation catheter from the His po­ sition and rotating to the coronary sinus in order to obtain a slow pathway potential with a small preferably fractionated atrial and large ventricular electrogram. Ablation was applied at this site in order to eliminate slow pathway conduction. Appearance of junctional ectopic rhythm was used to guide RF delivery. Ablation was performed using the following RF settings: AVNRT: 10 W gradually increasing to 45 W, non- irrigated, maximum 55 ◦C; AVRT: 55–60 W, non-irrigated, maximum 50–60 ◦C, for 60 s. In AVNRT, the procedural endpoint was non- inducibility of the tachycardia (i.e. not even a single echo beat was accepted). In case of empirical slow pathway ablation, the procedural endpoint was elimination of dual AV nodal conduction. Inducibility was tested after each application. In AVRT, the endpoint of the procedure was persistent absence of AP conduction. Testing with isoprenaline and/ or adenosine was performed based on operator’s preference. A standard of 30 min waiting time was applied to all procedures. Procedures were performed by the same team of operators over time. Table 1 ase e de og ap c a d c ca data. AVNRT ablation RMN N = 39 MAN N = 12 CRYO N = 28 Total N = 79 P-value Female 22 (56.4%) 7 (58.3%) 16 (57.1%) 45 (57.0%) 0.99 Age (year) 15 (12–16) 15 (12–17) 14 (11–16) 15 (12–16) 0.53 Weight (kg) 62 (48–67) 60 (49–67) 53 (43–64) 59 (48–66) 0.25 Length (cm) 169 (161–177) 167 (158–172) 164 (154–175) 166 (158–175) 0.33 BMI (kg/m2) 21 (19–23) 22 (18–23) 19 (17–21) 20 (18–22) 0.19 Palpitations 39 (100.0%) 12 (100.0%) 28 (100.0%) 79 (100.0%) ns Syncope 1 (2.6%) 0 (0.0%) 1 (3.6%) 2 (2.5%) 0.81 Family History Arrhythmias 3 (7.7%) 0 (0.0%) 2 (7.1%) 5 (6.3%) 0.62 Family History Sudden death 1 (2.6%) 1 (8.3%) 0 (0.0%) 2 (2.5%) 0.31 Medication Beta-blocker 1 (2.6%) 0 (0.0%) 2 (7.1%) 3 (3.8%) 0.47 Sotalol 6 (15.4%) 3 (25.0%) 6 (21.4%) 15 (19.0%) 0.70 Digoxine 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) ns Verapamil 2 (5.1%) 0 (0.0%) 3 (10.7%) 5 (6.3%) 0.40 Flecainide 0 (0.0%) 1 (8.3%) 0 (0.0%) 1 (1.3%) 0.06 AVRT ablation RMN N = 69 MAN N = 64 Cryo N = 11 Total N = 144 P-value Female 31 (44.9%) 31 (48.4%) 4 (36.4%) 66 (45.8%) 0.74 Age (year) 14 (12–16) 15 (12–17) 14 (11–16) 14 (12–16) 0.48 Weight (kg) 53 (42–65) 57 (49–65) 53 (43–64) 56 (45–65) 0.65 Length (cm) 165 (155–177) 170 (158–180) 164 (154–175) 167 (156–177) 0.44 BMI (kg/m2) 19 (18–22) 20 (18–21) 19 (17–21) 19 (18–21) 0.98 Palpitations 62 (89.9%) 57 (90.5%) 11 (100.0%) 130 (90.9%) 0.55 Syncope 6 (8.7%) 4 (6.3%) 1 (9.1%) 11 (7.7%) 0.87 Asymptomatic* 6 (8.7%) 8 (12.5%) 0 (0.0%) 14 (9.7%) 0.40 Family History Arrhythmias 11 (16.2%) 5 (7.9%) 2 (18.2%) 18 (12.7%) 0.31 Family History Sudden death 5 (7.5%) 2 (3.2%) 0 (0.0%) 7 (5.0%) 0.39 Medication Beta-blocker 2 (2.9%) 1 (1.6%) 1 (9.1%) 4 (2.8%) 0.37 Sotalol 10 (14.5%) 12 (18.8%) 6 (54.5%) 28 (19.4%) 0.008 Digoxine 1 (1.4%) 0 (0.0%) 0 (0.0%) 1 (0.7%) 0.58 Verapamil 5 (7.2%) 0 (0.0%) 1 (9.1%) 6 (4.2%) 0.08 Flecainide 1 (1.4%) 0 (0.0%) 0 (0.0%) 1 (0.7%) 0.58 BMI: body mass index, CRYO: manual guided cryoablation, MAN: manual-guided radiofrequency ablation, RMN: remote magnetic navigation guided radiofrequency ablation. * Asymptomatic patients in whom accidentally a WPW pattern on 12-lead ECG was discovered at a (e.g. Table 1 sport or profession related) regular medical check-up. BMI: body mass index, CRYO: manual guided cryoablation, MAN: manual-guided radiofrequency ablation, RMN: remote magnetic navigation guided radiofrequency ablation. * whom cross-over between different techniques occurred were also excluded. Accordingly, analysis in 223 patients was performed. Seventy- nine patients were diagnosed with AVNRT and 144 with AVRT. In total, 108 (48.4%) patients were treated with RMN, 76 (34.1%) were treated with MAN and 39 (17.5%) with CRYO ablation technique. The utiliza­ tion of the various techniques over time is presented as supplemental Fig. 1. group were performed using the following ablation catheters: Celcius or ThermoCool (Biosense Webster, Diamond Bar, CA, USA). Cryoablation was performed using Freezor 4 mm (before 2011) and 6 mm (from 2011 until 2019) catheters (Medtronic Inc, Minneapolis, MN, USA). group were performed using the following ablation catheters: Celcius or ThermoCool (Biosense Webster, Diamond Bar, CA, USA). Cryoablation was performed using Freezor 4 mm (before 2011) and 6 mm (from 2011 until 2019) catheters (Medtronic Inc, Minneapolis, MN, USA). 3.1. Baseline demographic and clinical data Baseline demographic and clinical data is presented in Table 1. AVNRT patients had a median age of 15 (IQR 12–16; Range 5–18) years, whereas AVRT patients had a median age of 14 (IQR 12–16; Range 6–17) years. The median weight in the AVNRT subgroup was 59 (IQR 48–66; Range 20–88) kilograms and in AVRT 56 (IQR 45–65; Range 23 –91) kilograms. (Supplemental Fig. 2). The majority of patients experi­ enced palpitations (AVNRT 100.0%, AVRT 90.9%), however a small portion of AVRT patients also had an event of syncope (7.7%). Fourteen children with AVRT (9.7%) were asymptomatic and a pre-excitation on a 12-lead ECG was accidentally found at a medical check-up (e.g. related to sport, school or profession). Baseline demographic and clinical data were comparable between groups, except for sotalol being less frequently used in AVRT ablation in the RMN group (14.5%, compared to 18.8% in MAN and 54.5% in CRYO, P = 0.008). 2.6. Statistical analysis Normality was assessed by the Kolmogorov-Smirnov test, or when appropriate, Shapiro-Wilk test. Mean and standard deviation (SD) were calculated for normally distributed continuous variables. Median, interquartile range (IQR) and range were computed for continuous variables with non-normal distribution. Descriptive statistics for cate­ gorical data were expressed in absolute numbers and percentages. Normally distributed continuous variables were analyzed using the 1- way ANOVA. For continuous variables with non-normal distributions, the Kruskall-Wallis test was used. Post-hoc pairwise comparisons were performed using the Dunn-Bonferroni approach, when a significant main effect was present. For comparing frequencies, the Chi-square test was used, or, when appropriate, Fisher’s exact test. Univariable and Multivariable Cox proportional hazards models were used to examine the relationship between treatment group and long-term outcomes, adjusting for potential confounders. A 2-sided P-value of < 0.05 was considered significant. Data were analyzed using SPSS 26.0 (SPSS Inc., Chicago, IL, USA). 2.2. Study population All consecutive patients, aged < 18 years, undergoing their first CA procedure for AVNRT or AVRT mechanisms between the 1st of January 2008 until 30th of June 2019, were included in this study. Patients were included from a single, high-volume CA center with large experience with all available CA techniques, both in adults as in pediatrics. Redo procedures were excluded. Patients had either documented SVT or symptoms of palpitations or syncope with documented pre-excitation on a 12-lead ECG. We excluded patients in whom cross-over between CA ablation techniques during the procedure occurred and patients with any types of congenital heart defect (CHD). In the RMN group, procedures were performed using the Stereotaxis Niobe system (Stereotaxis, Inc., St Louis, MO, USA). The following ablation catheters were used in RMN: Celsius RMT, ThermoCool RMT (Biosense Webster, Diamond Bar, CA, USA), Trignum Flux gold tip (Biotronik GmbH, Berlin, Germany) or Magnoflush gold tip (MedFact Engineering GmbH, Lorrach, Germany). The procedures in the MAN 2 IJC Heart & Vasculature 37 (2021) 100881 A.M.E. Noten et al. Table 1 Baseline demographic and clinical data. 3.3. Long-term outcome In AVNRT, procedure times were shortest in MAN procedures [88 (IQR 62–99) minutes], then in RMN [101 (IQR 87–121) minutes] and longest in CRYO [120 (IQR 88–143) minutes] (P = 0.018) (Table 2 and Fig. 1). Post-hoc pairwise comparison showed a significant difference between all three groups [MAN versus RMN (P = 0.028); MAN versus CRYO (P = 0.005); RMN versus CRYO (P = 0.029)]. Fluoroscopy dose was significantly different between groups [RMN 30 (IQR 20–41), versus MAN 27 (IQR 17–77), versus CRYO 45 (IQR 29–65) mGy, P = 0.041]. Post-hoc analysis showed that this consisted of a significant difference between RMN compared to CRYO (P = 0.012), whereas RMN versus MAN (P = 0.71) and MAN versus CRYO (P = 0.17) were comparable. Ablation time was significantly shorter in RMN compared to CRYO [RMN 193 (IQR 104–322), MAN 352 (IQR 88–711), CRYO 572 (IQR 480–932) seconds, P = 0.001; post-hoc analysis: RMN versus CRYO P < 0.001, RMN versus MAN P = 0.37 and MAN versus CRYO P = 0.12)]. We found a 100.0% acute success rate in RMN and MAN, compared to 85.7% in CRYO (P = 0.022). Regarding AVNRT ablation, we observed significantly lower recur­ rence rates in RMN and MAN guided ablation, compared to CRYO (7.7% versus 8.3% versus 35.7% respectively, P = 0.008) (Table 3). The mean follow-up time was 5.5 ± 2.9 years. AVNRT procedures performed with RF (i.e. MAN and RMN) were associated with a lower risk of recurrence during the follow-up, when compared to CRYO (Log-Rank P = 0.025; For RMN: HR 0.20, 95%-CI 0.06–0.74; For MAN: HR 0.19, 95%-CI 0.02–1.48; with CRYO as the reference group) (Fig. 2). As a sensitivity analysis, age, gender and weight were consecutively also added to the models, and did not show any significant associations with the primary outcome (data not shown). With respect to AVRT ablation, RMN had superior long-term out­ comes compared to the other techniques (4.3% versus 15.6% versus 54.5%, respectively P < 0.001), at a mean follow-up of 5.5 ± 2.9 years. The need to perform a repeat procedure showed the same pattern. Me­ dian time to recurrence was 6 (IQR 1–12) months in AVNRT and 2 (IQR 1–30) months in AVRT. 3. Results In total, 293 procedures were performed in 247 children within the mentioned time frame. We identified 247 (84.3%) first procedures and 46 (15.7%) redo procedures, of which the latter were excluded. Eleven patients (4.5%) with CHD of varying severity and 13 patients (5.3%) in Standard EPS revealed AVNRT mechanism in 79 (35.4%) patients and AP-mediated tachycardia (AVRT) in 144 (64.6%) patients (Table 2). With respect to left-sided AP’s, a transseptal approach was used in 5.2% of patients whereas a retrograde aortic approach was used in 94.8%. The 3 IJC Heart & Vasculature 37 (2021) 100881 A.M.E. Noten et al. IJC Heart & Vasculature 37 (2021) 100881 Table 2 Procedural parameters. 3. Results Post-hoc pairwise comparison using the Dunn-Bonferroni approach: AP: accessory pathway, AVNRT: atrioventricular nodal reentry tachycardia, AVRT: atrioventricular reentry tachycardia, CRYO: manual guided cryoablation, LA: left atrium, MAN: manual-guided radiofrequency ablation, RA: right atrium, RMN: remote magnetic navigation guided radiofrequency ablation, WPW: Wolff-Parkinson- Whit AP: accessory pathway, AVNRT: atrioventricular nodal reentry tachycardia, AVRT: atrioventricular reentry tachycardia, CRYO: manual guided cryoablation, LA: left atrium, MAN: manual-guided radiofrequency ablation, RA: right atrium, RMN: remote magnetic navigation guided radiofrequency ablation, WPW: Wolff-Parkinson- h respectively, P = 0.05)]. Similar to AVNRT ablation, we observed significantly higher acute success rates in RMN and MAN, compared to CRYO (98.6% versus 95.3% versus 81.8% respectively, P = 0.035). respectively, P = 0.05)]. Similar to AVNRT ablation, we observed significantly higher acute success rates in RMN and MAN, compared to CRYO (98.6% versus 95.3% versus 81.8% respectively, P = 0.035). respectively, P = 0.05)]. Similar to AVNRT ablation, we observed significantly higher acute success rates in RMN and MAN, compared to CRYO (98.6% versus 95.3% versus 81.8% respectively, P = 0.035). majority of AVNRTs were typical with slow-fast AVNRTs (slow-fast: 88.6%, slow-slow: 2.5%, fast-slow: 3.8%) (Table 2). Most AVRTs had a Wolff-Parkinson-White (WPW) pattern on 12-lead ECG with antegrade conduction over the AP (59.7%), whereas in 40.3% a concealed bypass was found (Table 2). 3. Results AVNRT ablation RMN N = 39 MAN N = 12 CRYO N = 28 Total N = 79 P-value Type Typical (slow-fast) 33 (84.5%) 11 (91.7%) 26 (92.9%) 70 (88.6%) 0.54 Atypical (slow–slow) 1 (2.6%) 0 (0.0%) 1 (3.6%) 2 (2.5%) 0.81 Atypical (fast–slow) 2 (5.1%) 0 (0.0%) 1 (3.6%) 3 (3.8%) 0.72 Empirical slow pathway ablation 3 (7.7%) 1 (8.3%) 0 (0.0%) 4 (5.1%) 0.31 Parameters Procedure time (min) 101 (87–121)* 88 (62–99)* 120 (88–143)* 101 (85–128) 0.018 Fluoroscopy dose (mGy) 30 (20–41)** 27 (17–77)** 45 (29–65)** 33 (21–51) 0.041 Ablation time (sec) 193 (104–322)*** 352 (88–711)*** 572 (480–932)*** 279 (118–609) 0.001 Application number 5 (2–10) 3 (2–10) 3 (2–5) 4 (2–7) 0.15 Acute success Acute success 39 (100.0%) 12 (100.0%) 24 (85.7%) 75 (94.9%) 0.022 AVRT ablation RMN N = 69 MAN N = 64 Cryo N = 11 Total N = 144 P-value Type Concealed AP 26 (37.7%) 24 (37.5%) 8 (72.7%) 58 (40.3%) 0.07 WPW 43 (62.3%) 40 (62.5%) 3 (27.3%) 86 (59.7%) 0.07 AP Location LA free wall 46 (66.7%) 36 (56.3%) 0 (0.0%) 82 (56.9%) <0.001 RA free wall 8 (11.6%) 3 (4.7%) 0 (0.0%) 11 (7.6%) 0.20 Septal 13 (18.8%) 20 (31.2%) 3 (27.3%) 36 (25.0%) 0.25 Parahisian 2 (2.9%) 0 (0.0%) 7 (63.6%) 9 (6.3%) <0.001 Multiple AP 0 (0.0%) 5 (7.8%) 1 (9.1%) 6 (4.2%) 0.06 Parameters Procedure time (min) 105 (88–135) 100 (81–160) 150 (107–183) 105 (83–143) 0.15 Fluoroscopy dose (mGy) 42 (26–61) 55 (32–85) 57 (39–98) 48 (30–81) 0.07 Ablation time (sec) 157 (78–365) 195 (74–409) 480 (300–691) 180 (78–415) 0.05 Application number 3 (1–9) 5 (2–12) 2 (2–6) 4 (2–10) 0.36 Acute success Acute success 68 (98.6%) 61 (95.3%) 9 (81.8%) 138 (95.8%) 0.035 Disappearance of pre-excitation on ECG (if present pre-procedural) 41 (97.6%) 39 (95.1%) 3 (100.0%) 83 (96.5%) 0.78 Post-hoc pairwise comparison using the Dunn-Bonferroni approach: * Procedure time: RMN versus MAN P = 0.028; MAN versus CRYO P = 0.005; RMN versus CRYO P = 0.029. A.M.E. Noten et al. p p g pp * Procedure time: RMN versus MAN P = 0.028; MAN versus CRYO P = 0.005; RMN versus CRYO P = 0.029. ** Fluoroscopy dose: RMN versus MAN P = 0.71; MAN versus CRYO P = 0.17; RMN versus CRYO P = 0.012. Post-hoc pairwise comparison using the Dunn-Bonferroni approach: 3.3. Long-term outcome Regarding AVRT, RMN and MAN groups had a significantly lower risk of recurrence, compared to CRYO (Log-rank P = 0.001; For RMN: HR 0.07, 95%-CI 0.02–0.30; For MAN: HR 0.26, 95%– CI 0.09–0.71; with CRYO as the reference group) (Fig. 2). In addition, RMN had significant lower risk of recurrence compared to MAN (Log- rank P = 0.044; HR 0.29 95%-CI 0.08–1.08, with MAN as the reference group). Age, gender and weight did not show any significant Regarding AVRT, procedure times were comparable between groups (Table 2 and Fig. 1). We observed a lower fluoroscopy dose and ablation time in RMN, however this did not reach the level of statistical signifi­ cance [fluoroscopy dose: 42 (IQR 26–61) versus 55 (IQR 32–85), versus 57 (IQR 39–98) mGy, for RMN versus MAN versus CRYO respectively, P = 0.07] [ablation time: 157 (IQR 78–365) versus 195 (IQR 74–904) versus 480 (IQR 300–691) seconds, for RMN versus MAN versus CRYO 4 IJC Heart & Vasculature 37 (2021) 100881 A.M.E. Noten et al. . Procedural parameters. Mean and interquartile range of three procedural parameters: procedure time, fluoroscopy dose and ablation time. Signific Post-hoc pairwise comparison using the Dunn-Bonferroni approach are marked with an asterisk (*): AVNRT procedure time: RMN versus MAN P = 0 CRYO P = 0.005; RMN versus CRYO P = 0.029. AVNRT Fluoroscopy dose: RMN versus CRYO P = 0.012. AVNRT Ablation time: RMN versus CRYO T: atrioventricular nodal reentry tachycardia, AVRT: atrioventricular reentry tachycardia, CRYO: manual guided cryoablation, MAN: manual-gui ncy ablation, RMN: remote magnetic navigation guided radiofrequency ablation. Fig. 1. Procedural parameters. Mean and interquartile range of three procedural parameters: procedure time, fluoroscopy dose and ablation time. Significant results of the Post-hoc pairwise comparison using the Dunn-Bonferroni approach are marked with an asterisk (*): AVNRT procedure time: RMN versus MAN P = 0.028; MAN versus CRYO P = 0.005; RMN versus CRYO P = 0.029. AVNRT Fluoroscopy dose: RMN versus CRYO P = 0.012. AVNRT Ablation time: RMN versus CRYO P < 0.001. AVNRT: atrioventricular nodal reentry tachycardia, AVRT: atrioventricular reentry tachycardia, CRYO: manual guided cryoablation, MAN: manual-guided radio­ frequency ablation, RMN: remote magnetic navigation guided radiofrequency ablation. IJC Heart & Vasculature 37 (2021) 100881 A.M.E. Noten et al. IJC Heart & Vasculature 37 (2021) 100881 Table 3 Long-term outcomes. 4.1. Efficacy of catheter ablation procedures This study observed superior efficacy of the RMN-guided ablation technique in treatment of AP-mediated tachycardias in children. Recurrence of arrhythmia and/or preexcitation on 12-lead ECG was observed only in 4% of cases in this group. Only one other study described long-term outcomes of RMN-guided AP ablation in children [9], where fewer recurrences were observed, though this study had a significantly shorter follow-up duration and only included left free wall AP’s. Ablation of left free wall AP’s has a higher success rate compared to other locations, which is likely to be the explanation for the discrepancy with the current study’s findings [2,18]. Regarding pedi­ atric AVNRT ablation, there are no studies comparing RMN with other ablation techniques. Our study showed significantly higher long-term efficacy in RMN and MAN groups, compared to cryoablation. In the adult population, similar recurrence rates have been reported in RMN and MAN ablation, which were also comparable between these two techniques, corresponding with the current study’s findings [19,20]. Recently published large European and Japanese pediatric CA registries, including primarily manual-guided RF and few cryoablation procedures, reported much higher 12-month recurrence rates compared to our findings, for AVNRT as well as for AVRT ablation [2,3]. Moreover, a higher recurrence rate in cryoablation compared to non-irrigated RF ablation was also observed in one of these registries, whereas the other did no comparison between energy sources [2,3]. Theoretically, the observed differences between cryoablation and RF ablation could be caused by either energy source, catheter stability and/or (mis)identifi­ cation of the right target. In patients with AVNRT undergoing slow pathway ablation, junctional ectopy during the application of RF energy is a sensitive marker of successful ablation [21,22]. During cryoablation, usually accelerated junctional tachycardia is not seen and therefore cannot guide lesion delivery [23]. Lack of this feedback is a possible explanation of the inferior performance of cryoablation in slow-pathway modification. However, the current study observed an inferior perfor­ mance of cryoablation in AVRT ablation as well, where no ectopic feedback is given during ablation. As cryothermal energy allows cry­ omapping (with reversible effect) before establishing a permanent lesion, this technique sometimes is used to avoid permanent damage of vulnerable cardiac structures if the target lesion lies close to these, in order to avoid complications. Despite this advantage, this technique appeared to be less beneficial, even when used in the most easy approachable anatomy (i.e. AVNRT ablation) [24]. 3.5. Adverse events We observed one major complication and two minor complications in total, which were comparable between groups [RMN 0 (0.0%), versus MAN 2 (2.6%), versus CRYO 1 (2.6%), P = 0.24]. One patient had a mild aortic valve regurgitation possibly caused by the retrograde transaortic catheter approach and a transient ischemic attack (TIA) 3 months after the procedure. The minor complications were left bundle branch block (N = 1) and temporary AV block (N = 1) which recovered completely. Another patient died during follow-up due to intestinal ischemia due to a genetic defect (not-procedure related). 3.4. Septal and parahisian accessory pathways APs were predominantly located at the left free wall (56.9%), whereas right free wall, septal and parahisian locations were less frequently observed (Table 2). CRYO was less frequently used for AP’s located at the left free wall (RMN 66.7%, MAN 56.3%, CRYO 0.0%, P < 0.001). In general, recurrence of arrhythmia occurred much more frequently in septal and parahisian AP’s, compared to the other AP lo­ cations (overall recurrence rate per location: left free wall 3.7%, right free wall 18.2%, septal 25.0%, parahisian 55.6%, multiple pathways 0.0%, P < 0.001). Subanalysis of septal and parahisian located AP’s only (N = 45) showed that the three techniques were equally frequently used to treat septal and parahisian AP’s [RMN N = 15 (33.3%), MAN N = 20 (44.4%), CRYO N = 10 (22.2%), P = 0.08] (Table 4). The three tech­ niques had comparable acute success rates (93.3% versus 90.0% versus 80.0% respectively, P = 0.57). We observed superior long-term out­ comes in favor of RMN in this subgroup, with significantly lower recurrence rates (13.3% versus 30.0% versus 60.0% for RMN versus MAN versus CRYO respectively, P = 0.047). Septal and parahisian located pathways had a significantly higher risk of recurrence when compared to AP’s at other locations (Log-rank P = 0.001; univariable HR 3.09, 95%-CI 1.55–6.17; with non-septal/parahisian AP as the reference group). In a multi-variable Cox regression model including both AP location and ablation technique, RMN had significantly the lowest risk of recurrence when compared to the two other techniques (Log-rank P < 0.001; For RMN: multivariable HR 0.14 95%-CI 0.06–0.37, with CRYO as the reference group; For MAN: multivariable HR 0.32, 95%-CI 0.15–0.69, with CRYO as the reference group; For RMN: multivariable HR 0.45, 95%-CI 0.17–1.23, with MAN as the reference group). 3.3. Long-term outcome AVNRT RMN N = 39 MAN N = 12 CRYO N = 28 Total N = 79 P-value Recurrence 3 (7.7%) 1 (8.3%) 10 (35.7%) 14 (17.7%) 0.008 Time to Recurrence (mo) 2 (1–2) 12 (12–12) 6 (1–13) 6 (1–12) 0.49 Recurrence of other arrhythmia 2 (5.1%) 1 (8.3%) 1 (3.6%) 4 (5.1%) 0.82 Redo 2 (5.1%) 1 (8.3%) 9 (32.1%) 12 (15.2%) 0.008 AVRT RMN N = 69 MAN N = 64 CRYO N = 11 Total N = 144 P-value Recurrence 3 (4.3%) 10 (15.6%) 6 (54.5%) 19 (13.2%) <0.001 Time to recurrence (mo) 16 (1–88) 1 (0–29) 6 (1–41) 2 (1–30) 0.33 Recurrence of other arrhythmia 3 (4.3%) 1 (1.6%) 0 (0.0%) 4 (2.8%) 0.52 Redo 5 (7.2%) 9 (14.1%) 5 (45.5%) 19 (13.2%) 0.002 AVNRT: atrioventricular nodal reentry tachycardia, AVRT: atrioventricular reentry tachycardia, CRYO: manual guided cryoablation, MAN: manual-guided radio­ frequency ablation, MO: months, RMN: remote magnetic navigation guided radiofrequency ablation A.M.E. Noten et al. AVNRT: atrioventricular nodal reentry tachycardia, AVRT: atrioventricular reentry tachycardia, CRYO: manual guided frequency ablation, MO: months, RMN: remote magnetic navigation guided radiofrequency ablation associations with the long-term outcome (data not shown). and manual-guided Cryoablation. The main finding of our study is that RMN guided ablation has superior long-term outcomes compared to the other techniques. and manual-guided Cryoablation. The main finding of our study is that RMN guided ablation has superior long-term outcomes compared to the other techniques. 4.1. Efficacy of catheter ablation procedures Therefore, we believe that our results are primarily caused by the combination of en­ ergy source and catheter stability, with RF having a deeper tissue penetration than cryothermal energy. We postulate that the improved 4. Discussion This is the first study to compare long-term outcomes of pediatric AVNRT and AVRT ablation procedures between three recognized abla­ tion techniques: RMN-guided RF ablation, manual-guided RF ablation 6 A.M.E. Noten et al. IJC Heart & Vasculature 37 (2021) 100881 ence rates - cumulative hazard on arrhythmia recurrence. Display of the cumulative hazard on atrioventricular nodal reentry tachyca cular reentry tachycardia (AVRT) recurrence, compared between three techniques: remote magnetic navigation guided radiofreq l guided radiofrequency ablation (MAN) and manual guided cryoablation (CRYO). g 2 R t l ti h d h th i Di l f th l ti h d t i t i l d l t t h di (AVNR Fig. 2. Recurrence rates - cumulative hazard on arrhythmia recurrence. Display of the cumulative hazard on atrioventricular nodal reentry tachycardia (AVNRT) - and atrioventricular reentry tachycardia (AVRT) recurrence, compared between three techniques: remote magnetic navigation guided radiofrequency ablation (RMN), manual guided radiofrequency ablation (MAN) and manual guided cryoablation (CRYO). Fig. 2. Recurrence rates - cumulative hazard on arrhythmia recurrence. Display of the cumulative hazard on atrioventricular nodal reentry tachycardia (AVNRT) - and atrioventricular reentry tachycardia (AVRT) recurrence, compared between three techniques: remote magnetic navigation guided radiofrequency ablation (RMN), manual guided radiofrequency ablation (MAN) and manual guided cryoablation (CRYO). IJC Heart & Vasculature 37 (2021) 100881 A.M.E. Noten et al. AVNRT ablation, Berman et al. evaluated clinical and direct cost per­ spectives of AVNRT ablation using either RMN, conventional manual RF and CRYO techniques [30]. RMN and conventional manual RF appeared to be equally effective and associated with lower AVNRT recurrence rates when compared to CRYO. This study observed significant dispos­ able cost savings of conventional MAN when compared to RMN, despite similar efficacy [30]. Whether the improved outcomes of RMN in pe­ diatrics observed by the current study, justify the additional costs of RMN over conventional manual RF techniques in AVNRT ablation re­ mains a topic of ongoing debate. Table 4 Subanalysis of ablation of septal and parahisian located AP. References [1] J.D. Kugler, D.A. Danford, K. Houston, G. Felix, Radiofrequency catheter ablation for paroxysmal supraventricular tachycardia in children and adolescents without structural heart disease. Pediatric EP Society, Radiofrequency Catheter Ablation Registry, Am. J. Cardiol. 80 (11) (1997) 1438–1443. [2] Krause U, Paul T, Bella PD, Gulletta S, Gebauer RA, Paech C, et al. Pediatric catheter ablation at the beginning of the 21st century: results from the European Multicenter Pediatric Catheter Ablation Registry ’EUROPA’. Europace. 2020. [3] Y. Kato, T. Suzuki, Y. Yoshida, N. Ozaki, S. Kishimoto, H. Aoki, S. Yoshida, S. Watanabe, Y. Nakamura, Catheter ablation in children and patients with congenital heart disease: Review of 1021 procedures at a high-volume single center in Japan, Heart Rhythm. 17 (1) (2020) 49–55. [4] G.F. Van Hare, M.D. Lesh, M. Scheinman, J.J. Langberg, Percutaneous radiofrequency catheter ablation for supraventricular arrhythmias in children, J. Am. Coll. Cardiol. 17 (7) (1991) 1613–1620. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi. org/10.1016/j.ijcha.2021.100881. 4.2. Accessory pathway location AP’s occur at a variety of locations, of which left lateral is the most frequently observed site [18]. Ablation of left lateral located AP’s yields high acute and long-term success rates, whereas right free wall and septal located AP’s are more difficult to ablate [2,3,18]. This is consis­ tent with our findings. The reversible effect of cryomapping, before establishing a permanent lesion during cryoablation, makes this abla­ tion technique very appealing when ablating in the proximity of the His bundle. Nevertheless, we observed high recurrence rates in cryoablation of septal and parahisian pathways. Both RF and cryoablation are sen­ sitive to electrode orientation, contact pressure, and convective thermal effects from blood flow over the electrode-tissue interface, influencing lesion formation [25]. Zones of spherical freezing advance are known as isotherms, with higher temperatures occurring at the periphery of an application, also in cryo-mapping ranges (around −30 ◦C) [26]. Possibly, targets are temporarily paralyzed but not definitively elimi­ nated by this cryo-map effect, resulting in long-term recurrences. On the contrary, in a model with simulated wall motion, RMN-guided RF lesion dimensions were larger compared to manual-guided RF lesions, consis­ tent with greater catheter stability [14]. The improved tissue penetra­ tion of RF energy, combined with superior catheter stability facilitated by RMN are a possible explanation of the superior outcomes of RMN- guided ablation in general, as well as in ablation of septal and para­ hisian located AP’s in specific. In conclusion, in pediatric patients without structural heart disease who underwent their first ablation for AVNRT or AP-mediated tachy­ cardia, RMN has the most favorable long-term outcomes, in addition to favorable fluoroscopy doses and procedure times. Table 4 Table 4 Subanalysis of ablation of septal and parahisian located AP. outcome of RMN over MAN in AVRT ablation, is an affirmation of improved catheter stability facilitated by RMN, in addition to the beneficial effects of RF over cryoablation, resulting in highly effective ablations in the RMN group. outcome of RMN over MAN in AVRT ablation, is an affirmation of improved catheter stability facilitated by RMN, in addition to the beneficial effects of RF over cryoablation, resulting in highly effective ablations in the RMN group. 4.5. Limitations The present study’s retrospective nature and the lack of blinded adjudication might have introduced bias, although this was alleviated by the use of objective measures. The large time window of inclusion might have introduced bias as techniques and approaches changed over time, however this is counterbalanced by the large numbers of inclusion of the present study, which limits its influence on the main outcomes. Principally, the large inclusion period highlights the scarcity of rhythm disorders and ablation procedures in children. In addition, we analyzed the utilization of the various ablation techniques over time and observed that all study groups included patients from the early years as well as from more recent years. However, because of technical reasons the RMN system was not operational from 2012 to 2014 in our center, which may have biased our results. Regarding AVRT, the distribution of cryoa­ blation was skewed towards this technique being more frequently used for ablation of septal and parahisian AP’s than other locations. Never­ theless, in the subanalysis of these specific pathways, other techniques were equally frequently used to treat AP’s at these more challenging ablation sites and the primary outcomes and multivariable Cox regres­ sion analysis results were consistent with our overall findings and showed RMN to be an independent predictor of long-term freedom of recurrence. 4.3. Procedure times and fluoroscopy doses One of the major arguments against RMN in the past was the extra time needed for patient positioning and system set up, which lengthened procedure times [27]. This study observed the shortest procedure times in both RF techniques. In our opinion, RMN’s facilitated catheter navi­ gation, combined with an experienced and dedicated RMN team, resulted in efficient RMN guided ablation procedures. Effects of radia­ tion exposure are major for children, as for instance they are ten times more vulnerable than adults to the induction of cancer by external ra­ diation exposure [28]. Prior studies have demonstrated lower fluoros­ copy times and dosages in favor of RMN [11]. Over time, significant improvement of procedure and radiation dosages in RMN have been observed which can be attributed to the use of EAM and technological advances [27,29]. This study observed the lowest fluoroscopy dosages in RMN and MAN AVNRT ablation, but only a tendency towards lower fluoroscopy dosages in RMN guided AVRT ablation. Further prospective research is needed to clearly define this. 4. Discussion RMN N = 15 MAN N = 20 CRYO N = 10 Total N = 45 P- value Acute success 14 (93.3%) 18 (90.0%) 8 (80.0%) 40 (88.9%) 0.57 Recurrence 2 (13.3%) 6 (30.0%) 6 (60.0%) 14 (31.1%) 0.047 Redo 1 (6.7%) 5 (25.0%) 5 (50.0%) 11 (24.4%) 0.047 AP: accessory pathway, CRYO: manual guided cryoablation, MAN: manual- guided radiofrequency ablation, RMN: remote magnetic navigation guided radiofrequency ablation [5] J.D. Kugler, D.A. Danford, K.A. Houston, G. Felix, Pediatric radiofrequency catheter ablation registry success, fluoroscopy time, and complication rate for supraventricular tachycardia: comparison of early and recent eras, J. Cardiovasc. Electrophysiol. 13 (4) (2002) 336–341. [2] Krause U, Paul T, Bella PD, Gulletta S, Gebauer RA, Paech C, et al. Pediatric catheter ablation at the beginning of the 21st century: results from the European Multicenter Pediatric Catheter Ablation Registry ’EUROPA’. Europace. 2020. [3] Y. Kato, T. Suzuki, Y. Yoshida, N. Ozaki, S. Kishimoto, H. Aoki, S. Yoshida, S. Watanabe, Y. Nakamura, Catheter ablation in children and patients with congenital heart disease: Review of 1021 procedures at a high-volume single center in Japan, Heart Rhythm. 17 (1) (2020) 49–55. 4.4. Cost-effectiveness Cannon, Magnetic versus manual catheter navigation for ablation of free wall accessory pathways in children, Circ. Arrhythm Electrophysiol. 5 (4) (2012) 804–808. [22] D.G. Katritsis, T. Zografos, K.C. Siontis, G. Giannopoulos, R.G. Muthalaly, Q. Liu, R. Latchamsetty, Z. Varga, S. Deftereos, C. Swerdlow, D.J. Callans, J.M. Miller, F. Morady, R.M. John, W.G. Stevenson, Endpoints for Successful Slow Pathway Catheter Ablation in Typical and Atypical Atrioventricular Nodal Re-Entrant Tachycardia: A Contemporary, Multicenter Study. JACC Clin. Electrophysiol. 5 (1) (2019) 113–119. [10] B. Schwagten, L. Jordaens, M. Witsenburg, F. Duplessis, A. Thornton, Y. van Belle, et al., Initial experience with catheter ablation using remote magnetic navigation in adults with complex congenital heart disease and in small children, Pacing Clin. Electrophysiol. 32 (Suppl 1) (2009) S198–S201. p y ( pp ) ( ) [11] B. Schwagten, M. Witsenburg, N.M.S. De Groot, L. Jordaens, T. Szili-Torok, Effect of magnetic navigation system on procedure times and radiation risk in children undergoing catheter ablation, Am. J. Cardiol. 106 (1) (2010) 69–72. [23] M. Sucu, V. Davutoglu, E. Polat, Junctional Beats During Cryo-Ablation Of The Slow Pathway For The Elimination Of Atrioventricular Nodal Reentrant Tachycardia, J. Atr Fibrillation. 9 (1) (2016) 1402. [12] A.C. Skanes, M. Dubuc, G.J. Klein, B. Thibault, A.D. Krahn, R. Yee, D. Roy, P. Guerra, M. Talajic, Cryothermal ablation of the slow pathway for the elimination of atrioventricular nodal reentrant tachycardia, Circulation 102 (23) (2000) 2856–2860. [24] R. Leila, P. Raluca, G. Yves, S. Dirk, S. Bruno, Cryoablation Versus Radiofrequency Ablation in AVNRT: Same Goal, Different Strategy, J Atr Fibrillation. 8 (1) (2015) 1220. [13] M.N. Faddis, W. Blume, J. Finney, A. Hall, J. Rauch, J. Sell, K.T. Bae, M. Talcott, B. Lindsay, Novel, magnetically guided catheter for endocardial mapping and radiofrequency catheter ablation, Circulation 106 (23) (2002) 2980–2985. [ ] h k l i i hik li l [25] B. Parvez, V. Pathak, C.M. Schubert, M. Wood, Comparison of lesion sizes produced by cryoablation and open irrigation radiofrequency ablation catheters, J. Cardiovasc. Electrophysiol. 19 (5) (2008) 528–534. [14] A. Bhaskaran, M.A. Barry, S.I. Al Raisi, W. Chik, D.T. Nguyen, J. Pouliopoulos, C. Nalliah, R. Hendricks, S. Thomas, A. L McEwan, P. Kovoor, A. Thiagalingam, Magnetic guidance versus manual control: comparison of radiofrequency lesion dimensions and evaluation of the effect of heart wall motion in a myocardial phantom, J. Interv. Card Electrophysiol. 44 (1) (2015) 1–8. [26] M. Handler, G. Fischer, M. 4.4. Cost-effectiveness [5] J.D. Kugler, D.A. Danford, K.A. Houston, G. Felix, Pediatric radiofrequency catheter ablation registry success, fluoroscopy time, and complication rate for supraventricular tachycardia: comparison of early and recent eras, J. Cardiovasc. Electrophysiol. 13 (4) (2002) 336–341. Installation of the RMN technology requires an investment both in the essential equipment and specific training of the EP lab staff. In adult 8 A.M.E. Noten et al. IJC Heart & Vasculature 37 (2021) 100881 and the impact of accessory pathway location, Am. J. Cardiol. 74 (8) (1994) 786–789. [6] G.F. Van Hare, H. Javitz, D. Carmelli, J.P. Saul, R.E. Tanel, P.S. Fischbach, R. J. Kanter, M. Schaffer, A. Dunnigan, S. Colan, G. Serwer, Prospective assessment [6] G.F. Van Hare, H. Javitz, D. Carmelli, J.P. Saul, R.E. Tanel, P.S. Fischbach, R. J. Kanter, M. Schaffer, A. Dunnigan, S. Colan, G. Serwer, Prospective assessment after pediatric cardiac ablation: recurrence at 1 year after initially successful ablation of supraventricular tachycardia, Heart Rhythm. 1 (2) (2004) 188–196. after pediatric cardiac ablation: recurrence at 1 year after initially successful ablation of supraventricular tachycardia, Heart Rhythm. 1 (2) (2004) 188–196. [19] M. Shurrab, A. Danon, A. Crystal, B. Arouny, I. Tiong, I. Lashevsky, D. Newman, E. Crystal, Remote magnetic navigation for catheter ablation of atrioventricular nodal reentrant tachycardia: a systematic review and meta-analysis, Expert Rev Cardiovasc Ther. 11 (7) (2013) 829–836. [7] K.K. Collins, M.S. Schaffer, Use of cryoablation for treatment of tachyarrhythmias in 2010: survey of current practices of pediatric electrophysiologists, Pacing Clin. Electrophysiol. 34 (3) (2011) 304–308. [20] L. Parreira, R. Marinheiro, P. Carmo, D. Cavaco, K. Reis-Santos, P. Amador, T. Teixeira, A.S. Soares, F. Costa, P. Adragao, Atrioventricular node reentrant tachycardia: Remote magnetic navigation ablation versus manual ablation - impact on operator fluoroscopy time, Rev. Port. Cardiol. 38 (3) (2019) 187–192. l ll b d [8] F. Drago, M.S. Russo, M.S. Silvetti, A. Des, F. Iodice, M.T. Naso Onofrio, Cryoablation of typical atrioventricular nodal reentrant tachycardia in children: six years’ experience and follow-up in a single center, Pacing Clin. Electrophysiol. 33 (4) (2010) 475–481. [21] J.H. Jentzer, R. Goyal, B.D. Williamson, K.C. Man, M. Niebauer, E. Daoud, S. A. Strickberger, J.D. Hummel, F. Morady, Analysis of junctional ectopy during radiofrequency ablation of the slow pathway in patients with atrioventricular nodal reentrant tachycardia, Circulation 90 (6) (1994) 2820–2826. [9] J.J. Kim, S.L. Macicek, J.A. Decker, N.J. Kertesz, R.A. Friedman, B.C. 4.4. Cost-effectiveness Seger, R. Kienast, F. Hanser, C. Baumgartner, Simulation and evaluation of freeze-thaw cryoablation scenarios for the treatment of cardiac arrhythmias, Biomed. Eng. Online 14 (2015) 12. [27] A.M. Elisabeth Noten, Z. Kis, F. Akca, R. Bhagwandien, S. Wijchers, S.C. Yap, et al., Robotic navigation shows superior improvement in efficiency for atrial fibrillation ablation, J. Atr Fibrillation. 11 (5) (2019) 2108. [15] T. Bauernfeind, F. Akca, B. Schwagten, N. de Groot, Y. Van Belle, S. Valk, B. Ujvari, L. Jordaens, T. Szili-Torok, The magnetic navigation system allows safety and high efficacy for ablation of arrhythmias, Europace. 13 (7) (2011) 1015–1021. d l bl [28] E.J. Hall, Lessons we have learned from our children: cancer risks from diagnostic radiology, Pediatr. Radiol. 32 (10) (2002) 700–706. i [16] J. Papagiannis, K. Papadopoulou, S. Rammos, D. Katritsis, Cryoablation versus radiofrequency ablation for atrioventricular nodal reentrant tachycardia in children: long-term results, Hellenic J. Cardiol. 51 (2) (2010) 122–126. [29] A.M.E. Noten, A.A. Hendriks, S.-C. Yap, D. Mol, R. Bhagwandien, S. Wijchers, I. Kardys, M. Khan, T. Szili-Torok, Contact feedback improves 1-year outcomes of remote magnetic navigation-guided ischemic ventricular tachycardia ablation, Int. J. Cardiol. 315 (2020) 36–44. [17] M. Karacan, N. Çelik, C. Akdeniz, V. Tuzcu, Long-term outcomes following cryoablation of atrioventricular nodal reentrant tachycardia in children, Pacing Clin. Electrophysiol. 41 (3) (2018) 255–260. [30] A.E. Berman, H. Rivner, R. Chalkley, V. Heboyan, A comparative analysis of clinical outcomes and disposable costs of different catheter ablation methods for the treatment of atrioventricular nodal reentrant tachycardia, Clinicoecon Outcomes Res. 9 (2017) 677–683. [18] J.K. Park, B.D. Halperin, J.H. McAnulty, J. Kron, M.J. Silka, Comparison of radiofrequency catheter ablation procedures in children, adolescents, and adults
W4366975137.txt
https://seer.ufrgs.br/index.php/PolisePsique/article/download/117924/88340
pt
Implicações psicossociais da migração rural-urbana para jovens universitários
Revista Polis e Psique
2,023
cc-by
8,468
Implicações psicossociais da migração rural-urbana para jovens universitários Psychosocial implications of young college students’ rural-urban migration Implicaciones psicosociales de la migración rural-urbana para los jóvenes universitários Alexsandra Maria Sousa Silva Faculdade Luciano Feijão (FLF), Sobral, CE, Brasil. Verônica Morais Ximenes Universidade Federal do Ceará (UFC), Fortaleza, CE, Brasil. Resumo Neste trabalho discute-se como os processos migratórios podem produzir efeitos psicossociais, de diferentes ordens, dimensões e com especificidades, quando se trata da relação entre rural e urbano. O objetivo é identificar relações entre os modos de vida dos jovens universitários e a migração rural-urbana na realidade brasileira. Com base qualitativa, foram realizadas 14 entrevistas semiestruturadas com jovens universitários, de comunidades rurais, integrantes da assistência estudantil de uma instituição pública de ensino superior. Os resultados apontaram que o modo de vida dos jovens universitários que vivenciam a migração rural-urbana está atravessado por inúmeros desafios, dentre os quais a pobreza. Apresentou-se dilemas provenientes da migração para os universitários viajantes, que refletem na mudança do modo de socialização, no distanciamento dos vínculos familiares e na mudança cultural. Concluiu-se que é importante pensar políticas de assistência estudantil que considerem a dimensão socioeconômica e subjetiva, favorecendo o acesso e a permanência do estudante na universidade. Palavras-Chave: Juventude; Migração; Ensino Superior. Abstract In this work discusses how the migration processes can produce psychosocial effects of different orders and dimensions, with specificities, regarding the connection between rural and urban. The objective is to identify relations between ways of life of the young college students and the urban-rural migration, in the Brazilian reality. Using a qualitative basis of research, we performed 14 semi structured interviews with the young college students, from rural communities, members of the student aid of a public institution of higher education. The results showed that the way of life of the young college students who experienced rural-urban migration is filled with innumerous challenges, poverty being one of them. The migration of the traveling college students presented dilemmas, which reflect the shift in the means of socialization, distancing from family bonds and cultural change. It has been concluded that it is |68 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ important to think student aid policies that consider the social economic and subjective dimension, enabling the access and permanence of the student in the university. Keywords: Youth; Migration; Higher education. Resumen En este trabajo se analiza cómo los procesos migratorios pueden producir efectos psicosociales de diferentes órdenes y dimensiones, con especificidades, en lo que respecta a la relación entre lo rural y lo urbano. El objetivo es identificar las relaciones entre las formas de vida de los jóvenes universitarios y la migración rural-urbana en la realidad brasileña. Sobre una base cualitativa, se realizaron 14 entrevistas semiestructuradas a estudiantes universitarios de comunidades rurales, integrantes de la asistencia estudiantil en una institución pública de educación superior. Los resultados mostraron que la forma de vida de los jóvenes universitarios que experimentan la migración del campo a la ciudad está atravesada por numerosos desafíos, incluida la pobreza. Se presentaron los dilemas derivados de la migración para los estudiantes universitarios itinerantes, que reflejan el cambio en la forma de socialización, el alejamiento de los lazos familiares y el cambio cultural. Se concluyó que es importante pensar en políticas de atención al estudiante que consideren la dimensión socioeconómica y subjetiva, favoreciendo el acceso y permanencia de los estudiantes en la universidad. Palabras clave: Juventud; Migración; Enseñanza superior. (Caetano & Azevedo, 2017). Os olhares para a realidade da juventude rural tiveram Introdução como marco os anos de 2008 e 2011, com Delimitar o conceito de juventude as respectivas I e II Conferência Nacional permanece um desafio, visto que a de Juventude, a partir das quais denunciou- compreensão de tal categoria é atravessada se a necessidade de espaços para a discussão por históricos, sobre as especificidades desse segmento, culturais, seus modos de vida e desejos de futuro cronológicos e subjetivos. Nesse sentido, (Menezes, Stropasolas & Barcellos, 2014). considera-se importante a incorporação da Com isso, a Secretaria Nacional de multiplicidade de sentidos do “ser jovem” Juventude (SNJ) incluiu demandas da na produção de saberes na modernidade juventude rural, o que culminou na criação (Bittencourt, Matheus & Tavanti, 2018). do Grupo de Trabalho da Juventude Rural marcadores socioeconômicos, políticos, Quando se discute políticas que em 2012. Nesse percurso, quando se busca envolvem a juventude brasileira, constata- estudos e discussões sobre juventude de se que, em sua maioria, referem-se àqueles origem rural e políticas públicas, percebe-se jovens que vivem no espaço urbano que esse arranjo leva à discussão sobre |69 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ agricultura familiar e relação com a terra definitivamente de um lugar e morar em (Barcellos, 2015). outro. Atualmente temos, por exemplo, a A migração está realidade vivida por muitos jovens, cujo da deslocamento para o espaço urbano não tem necessariamente representa uma ruptura impulsionado a mobilidade da população com o rural, mas pode ser característico de rural para os centros urbanos, cujo auge foi um modo de vida cotidiano. Nesse na década de 70, com a concentração das raciocínio, as desigualdades vivenciadas em atividades cidades, realidades rurais têm feito com que os caracterizando uma sociedade de caráter jovens busquem alternativas para satisfazer mais urbano-industrial (Baeninger, 2012). seus anseios, desejos e necessidades – historicamente ligada urbanização. Tal no Brasil à questão fenômeno econômicas nas Leite, Macedo, Dimenstein e Dantas muitas vezes longe de sua comunidade –, (2013) pontuam que o rural passa por reproduzindo uma cultura da migração problemas sociais próprios e agravantes, efetiva ou se propondo a vivenciar a como as dificuldades de acesso a serviços migração transitória. Esse processo se torna de saúde e de educação, e questões como mais complexo na medida em que ocorre insegurança fundiária, exploração de mão em um contexto marcado por desigualdades de obra, etc. Ademais, pode-se dizer que sociais, econômicas e culturais. Marinho novas as (2014) ressalta que essa realidade de especificidades do rural, pois tal espaço migração tem se mostrado mais frequente, vem adquirindo outras roupagens. A principalmente entre os jovens que buscam diferença entre essas realidades, rural e investir nos estudos, nos níveis médio e urbana, também está relacionada ao uso do superior, tempo livre, aos modos de lidar com o qualificação. Nessa perspectiva, este artigo trabalho e às relações sociais. Nesse trata de um recorte de uma pesquisa de tese sentido, é importante problematizar a ideia de doutorado e parte da compreensão de que do rural como um lugar estereotipado e a realidade da migração e da busca pelo submisso necessário acesso ao ensino superior pode sinalizar a questionar discursos dessa natureza e necessidade de colocar em suspensão a evidenciar novos olhares. categoria ‘jovens rurais’, bem como propõe relações ao estão urbano. marcando É Quanto à migração rural-urbana, refletir e querem sobre uma melhor possibilidades para Castro (2009), a relação entre desacomodar permanecer e migrar é muito complexa, sustentam e petrificam a ideia de rural. pois não se trata apenas de certas estruturas de que sair |70 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ A migração de jovens rurais e o ingresso compreender os múltiplos significados no ensino superior históricos e sociais que essas juventudes representam. Nessa direção, parte-se dessa A juventude camponesa é parte mesma autora, para evidenciar a juventude integrante da pluralidade juvenil: são os não urbana, que pode ser chamada de jovens de origem rural, jovens que vêm do juventude rural, juventude da roça ou interior, de pequenas comunidades, regiões juventude do campo. periféricas aos grandes centros urbanos e outras localidades não pois existem jovens em diferentes contextos urbanas. Isso também se aplica a alguns rurais, como em comunidades rurais, no lugarejos que se localizam no interior, que, campo, nos arredores de praias ou serras e mesmo não sendo considerados o rural do em outras regiões com características bem sertanejo, características distintas, dado seus modos de constituição e atravessadas pelas diversas ruralidades. Ao suas histórias. É preciso destacar que os considerar juventude contextos dessas juventudes mudaram, pois camponesa, sobrepomos a dimensão da atualmente envolvem novas e múltiplas classe social, uma vez que, muitas vezes, dimensões, desde os direitos à cultura, ao esse segmento advém do contexto de esporte e ao lazer, até à participação social, pobreza. Por fim, tem-se o recorte dado a política, acesso ao trabalho, renda e partir do nível e do acesso à educação, visto educação (Barcellos, 2015). têm o consideradas A juventude camponesa é ampla, suas recorte da que esta pesquisa priorizou jovens que O modo de vida da juventude na acessaram o ensino superior. Desta feita, realidade rural é marcado pelo trabalho com este estudo situa-se no entrelace desses os pais, seja auxiliando nas atividades principais marcadores, ou seja, busca-se domésticas, seja na agricultura. Nesse caso, evidenciar jovens constata-se uma visão naturalizada que camponeses que vivenciaram a migração perpassa a condição juvenil no que se refere rural-urbana em um contexto de pobreza e à imagem do jovem rural associada ao filho que conseguiram acessar o ensino superior. de agricultores familiares ou camponeses Em relação à categoria juventude, que vive um período de aprendizagens e Castro (2009) chama a atenção para as preparação para suceder seus pais na diversas formas de construção da identidade agricultura. Esse jovem torna-se, então, juvenil, a qual é marcada por uma parte diversidade de experiências, sonhos e reproduzida de uma geração para outra. desejos. A autora ressalta a necessidade de Segundo Castro (2009), essa imagem vem os discursos de de uma relação hierarquizada |71 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ sendo modificada juntamente às (2006), essas transformações estão ligadas estratégias de às demandas por escolarização em nível reprodução das famílias de agricultores, de superior e à busca por novos destinos modo a emergir questões relacionadas à sociais e profissionais. Tal panorama leva juventude e ao êxodo rural, à interação parte da juventude rural a se distanciar do urbano-rural e à crise na relação familiar, campo e da agricultura e apostar na vida marcada pela reprodução dos papéis ao urbana. É preciso considerar que essas longo das gerações. mudanças, situadas a nível global e social, transformações das Desse modo, a migração tem sido são provenientes de uma estrutura um tema bastante debatido na esfera pública capitalista que exige maior nível de estudo e social, e isso tem reverberado também na e especialização sua importância nos centros acadêmicos. impactos em quem vive na zona rural. Tal à Nessa perspectiva, mesmo que ainda possa são haver alguma ligação do jovem com a atividades agrícolas e familiares, ele deixa migração é caracterizada por uma mudança de reproduzir a geração dos pais e passa a se mais permanente, em que o indivíduo fixa constituir como categoria identitária com moradia por um determinado período de novas necessidades de educação, cultura, tempo, enquanto a mobilidade é mais lazer, profissionalização e formação. fenômeno está mobilidade humana, sinônimos. Para relacionado porém Barbieri não (2011), profissional, gerando transitória, envolvendo fatores objetivos, Com isso, observa-se um novo subjetivos e contextuais. Nesse sentido, os cenário, no qual a universidade brasileira jovens que vivem em comunidades rurais passa a contar com uma diversidade maior podem se deslocar para a cidade à procura de de profissionalização e estudo, e isso pode geracionais e de gênero, marcada pela se dar em caráter de migração ou de multiplicidade, mobilidade heterogeneidade. transitória. Assim, esse perfis socioeconômicos, étnicos, diversidade Essa mudança e foi deslocamento aponta que as características favorecida pelas políticas de acesso ao de novas ensino superior entre os anos de 2003 a configurações no que tange às migrações 2011, como as políticas de inclusão e (Baeninger, 2012). democratização do ensino superior nas IES cada O região fato é implicam que atualmente federais brasileiras, as políticas de cotas predominam grandes transformações nas para negros, índios e alunos provenientes da comunidades rurais, principalmente no que escola pública, as políticas de Assistência tange ao segmento juvenil. Segundo Zago Estudantil, o REUNI (Programa de Apoio a |72 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ Planos de Reestruturação e Expansão das relações sociais e fundamentado em uma Universidades Federais), o qual contribuiu compreensão histórica, social, cultural, com o processo de interiorização das econômica e política, Universidades Federais, e o PROUNI importância à relação (Programa Universidade para Todos), que pesquisados para a construção dos dados. conferindo pesquisador- possibilita o acesso do estudante pobre à O processo de inserção ocorreu universidade. No que tange especificamente durante um ano e meio em uma Instituição à realidade da juventude rural, Redin e Pública Silveira (2012) pontuam que o PROUNI localizada no interior do estado do Ceará fortaleceu o sonho do jovem rural de cursar (CE) e considerada um polo de atração o ensino superior. Tal cenário exige novas universitário. Para isso, entrou-se em reflexões, na medida em que a realidade dos contato com o diretor da instituição, de jovens universitários pode ser geradora de modo a ser apresentada a proposta do conflitos estudo. O dirigente desta IES ratificou a para a dinâmica social (Corrochano, 2013). de Ensino Superior (IES), pertinência da temática para a realidade e Não é por acaso que o ensino conferiu a anuência para a execução da superior tem sido tema de discussões pesquisa. Em seguida, buscou-se conhecer fervorosas, principalmente nas duas últimas a Pró-Reitoria de Assuntos Estudantis décadas, (PRAE), apresentá-la a proposta do estudo, nas quai preponderam questionamentos acerca do compromisso além de construir os acordos social das universidades. Acredita-se que o metodológicos. Esse setor contava com a ingresso do jovem rural no ensino superior presença de duas assistentes sociais e dois é um fenômeno complexo, com várias bolsistas, que facilitaram a apropriação e o nuanças que precisam ser evidenciadas. acesso ao Programa Auxílio Moradia, Diante disso, o objetivo deste artigo é destinado a jovens estudantes provenientes identificar relações entre os modos de vida de regiões rurais. Tal programa assegurava dos jovens universitários e a migração rural- um auxílio financeiro, com vistas a garantir urbana na realidade brasileira. a permanência no curso. A partir de então, articulou-se a comunicação com esses Método jovens, inicialmente via e-mail, depois também via Whatsapp, de forma a ser Este estudo tem como base uma pesquisa qualitativa, que concebe o possível convidá-los para participar da pesquisa. conhecimento construído a partir das |73 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ Esses jovens foram selecionados a (Bardin, 1977), com auxílio do Atlas ti 5.2. partir de alguns critérios, como: ser A partir de então, foi possível criar códigos estudante universitário, maior de 18 anos; que sinalizaram as categorias identificadas estar no regularmente matriculado na estudo, tais migratórios; participado do programa de assistência psicossociais. A essas categorias foram estudantil na modalidade Auxílio-Moradia; articuladas ter para provenientes do material produzido nas participação voluntária na pesquisa, com a entrevistas e, assim, seguiu-se às análises. assinatura do Termo de Consentimento Esta pesquisa foi aprovada pelo Comitê de Livre e Esclarecido (TCLE). Ética em Pesquisa (CEP) de uma instituição e disponibilidade Dentre os participantes, foram 9 de processos instituição pesquisada; participar ou já ter interesse modos como: vida; quotations efeitos (citações) pública de ensino superior. homens e 5 mulheres, com idades entre 18 e 29 anos, oriundos de comunidades rurais. Resultados e discussão Esses interlocutores da pesquisa eram estudantes universitários pertencentes a A partir desta pesquisa, conseguiu- cursos de graduação de diferentes áreas, se identificar relações entre os modos de como ciências biológicas e saúde, ciências vida dos jovens universitários e a migração humanas e sociais, ciências exatas e rural-urbana tecnológicas. categorias: processos migratórios; modo de O processo seguiu com a realização de 14 entrevistas através das seguintes vida, com ênfase na pobreza; e, por fim, os semiestruturadas, efeitos psicossociais, com vistas a discutir individuais, em salas de estudo da própria aspectos subjetivos presentes na vivência universidade, a partir de um roteiro migratória. previamente elaborado e semiestruturado, cujo foco foi dividido em três partes: 1) a Os processos migratórios: os jovens que relação do jovem com a comunidade rural; saem do rural para o urbano 2) a vivência da mobilidade e da migração; 3) a relação do jovem com a cidade e a universidade. As entrevistas foram gravadas em Para Barbieri (2011), a migração envolve: fatores objetivos, associados à mudança de residência; perspectiva formato de áudio, com duração entre trinta temporal e espacial; fatores subjetivos, minutos e uma hora, e foram transcritas e quando analisadas, a partir da análise de conteúdo involuntário; e fatores contextuais, quando é de caráter voluntário ou |74 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ o contexto tem variáveis que atraem ou (Participante expulsam as pessoas. Pode-se considerar as Universidade). questões socioeconômicas e a pobreza É possível identificar como a como elementos que influenciam os jovens questão, principalmente das condições de na busca de profissionalização e estudo, vida, é um marcador que ainda caracteriza a principalmente quando não encontram realidade do jovem no rural. É preciso possibilidades em seus lugares de moradia. destacar Por conta dessas condições de vida, o jovem fundamental de seu desenvolvimento, já de comunidades rurais costuma buscar que é considerado uma atividade dominante trabalho mais cedo (Furlani & Bomfim, que orienta o amadurecimento do jovem por 2010). Essa perspectiva se faz presente no ser discurso de um dos jovens, quando profissionalizante, além de influenciar a compartilha sua percepção acerca das posição que tal sujeito vai ocupar nos meios diferenças entre os jovens que vivem na de produção (Bulhões & Abrantes, 2016). cidade e os que vivem no campo: o de trabalho caráter 12, como 17/06/19, mediador produtivo e Para Alves e Dayrell (2015), a Pesquisador: Para você, quais as condição juvenil na realidade rural mostra- diferenças entre os jovens que se fortemente influenciada pela cultura da moram no rural e os que moram na migração. Isso ocorre por conta das cidade? condições de vida presentes na realidade Entrevistado: Ah, porque, às vezes, rural, tais como dificuldades de emprego e a é renda e acesso a diversidades de bens e diferente! Acho que nem sempre, serviços. Tudo isso são fatores que, mas acho que o jovem rural é mais historicamente, estimulam a migração obediente. Assim, é acostumado a campo-cidade trabalhar desde cedo, enfrentar as principalmente mulheres jovens que se dificuldades, porque hoje em dia distanciam cada vez mais do trabalho da tem muitos jovens, mesmo que não roça, demarcando uma diferença em relação seja de classe alta, mas que moram a suas gerações anteriores. Marinho (2014) na cidade. Ele é acostumado com ressalta que essa realidade de migração tem tudo fácil, os pais fazem de tudo pra se mostrado mais frequente, principalmente dar as coisas a ele, eles não ligam entre os jovens que buscam investir nos muito, não se importam com as estudos, nos níveis médio e superior, e conquistas que eles conseguem. querem uma melhor qualificação, pois, de Acho que a diferença é isso um modo geral, no meio rural, quando há educação, os costumes, de muitos jovens, |75 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ escolas, elas atendem ao público até o participantes da pesquisa destacaram “A ensino fundamental. A seguir, ilustra-se a dificuldade acho que é tá longe dos meus percepção do jovem participante acerca das familiares, tipo, quando a gente tá aqui e diferenças entre os meios rural e urbano: adoece, a gente tá sozinho e tem que Pesquisador: Como você vê a resolver as coisas” diferença entre o campo e a cidade? 21/06/19, Universidade). (Participante 5, Entrevistado: Acho que a questão da Assim, apesar da migração ainda localização mesmo. E até o acesso persistir, há uma quebra no imaginário do também, tem as escolas também [na morador da zona rural quando se depara cidade]. As dificuldades no rural são com a complexidade do modo de vida nas maiores, bem maiores do que viver grandes cidades; esse sujeito sente o choque na cidade (Participante 1, 21/06/19, das diferenças, principalmente quando estas Universidade). produzem dificuldades, como a questão da A consequência desse cenário de mobilidade: migração é a redução da juventude rural, Às vezes tinha aula no contraturno e produzindo um envelhecimento do campo eu não ia, porque o transporte era (Marinho, 2014), um desinteresse do jovem caro, o mototáxi, na época, ele já não pela terra (Castro, 2009), uma diminuição era muito barato. E eu, como eu não da nas era daqui, eu nunca soube pegar comunidades rurais e um inchaço nos ônibus aqui na cidade. Aí, então, eu centros urbanos. Com a migração, também dependia se observa que jovens rurais vão para as (Participante cidades e se submetem a empregos Universidade). precários e baixas remunerações, pondo em Há aqueles estudantes que se xeque o ideal e o imaginário construído deslocam diariamente para a cidade em acerca da vida nas grandes cidades. transportes universitários, e há outro perfil, força de trabalho familiar muito 13, do mototáxi 22/05/19, A vida nas cidades impõe novos que são aqueles que moram na cidade para hábitos, a necessidade de lidar com outras estudar e retornam a sua comunidade culturas, diferentes modos de socialização e apenas em alguns finais de semana. Não são inúmeros desafios diários de convivência só as dificuldades que marcam a escolha por para a significação da realidade urbana que morar na cidade, existem fatores que é decisivamente diferente do modo de vida favorecem, como o melhor aproveitamento no do campo. Dentre as facilidades e dificuldades provenientes da migração, os tempo e o próprio esforço do deslocamento contínuo. |76 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ É porque eu tive mais tempo, porque semestre, eu fiquei de AF [avaliação final] eu não perdia esse tempo, e eu não em uma disciplina e no segundo e no ficava muito cansada, porque eu terceiro eu não fiquei em nenhuma. Aí, chegava meia noite, mas, os dias que talvez, tenha sido vir morar aqui também e eu precisava vim pra cidade, eu me esforcei mais” (Participante 1, 21/06/19, tinha que tá acordada 4h30 da Universidade). É importante ressaltar que, manhã já pra tá voltando, aí voltava embora não tenha sido o foco deste estudo, meio dia e tinha que voltar 16h, aí além desse fator, há também as questões chegava meia noite, aí tinha três socioinstitucionais vezes na semana que eu fazia isso podem interferir no rendimento do aluno, na (Participante sua permanência e conclusão do curso. 10, 17/06/19, Universidade). e pedagógicas que A mobilidade cotidiana tem reflexos Há uma diversidade de facetas, entre na constituição da identidade e da facilidades e dificuldades, que geram sociabilidade desses jovens, construindo desafios frente à escolha pela migração ou novas práticas e inúmeras maneiras de fazer pela mobilidade cotidiana. Para migrar e e ser que acontecem até nos transportes morar na cidade, é preciso um mínimo de universitários (Freitas & Braga, 2013). condições socioeconômicas para sustentar Diante disso, essa mobilidade aparece esse modo de vida. Já os que não têm essa possivelmente para tentar romper os condição optam pela mobilidade transitória. círculos de invisibilidade e exclusão social Nesse jovens e cultural cultivada no campo (Oliveira & colocam a questão da mobilidade urbano- Feliciano, 2016). Faz-se urgente refletir rural a sobre o lugar que esse jovem migrante tempo ocupa, especialmente aquele que não saiu “tempo definitivamente de sua localidade, mas perdido”: “Como daqui pra serra são duas vivencia um modo de vida marcado por esse horas de viagem e é uma viagem ‘vai e volta’: por uma realidade em extremamente cansativa, se eu ficasse indo movimento e vindo todo dia, eu perderia muito tempo” incertezas que são impostas por essas (Participante 5, 21/06/19, Universidade). circunstâncias. Nesse aspecto, como relata Sobre a mobilidade como sinonímia de o jovem participante: como mobilidade contexto, fator negativo, produz contraproducente, muitos um considerado pois e pelas insatisfações e perda de tempo, para tais interlocutores, Eu já passei dois semestres, e o essa circunstância interfere diretamente no começo do terceiro indo e vindo, às rendimento acadêmico: “No primeiro vezes duas no dia, porque, antes do |77 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ auxílio moradia, eu tinha a BIA, que Desse modo, constata-se que o rural era a Bolsa de Iniciação Acadêmica, e o urbano são mundos bastante distintos, aqui também da universidade, e aí mas que se entrelaçam intensamente na vida eu vinha e ia duas vezes no dia, desses jovens que decidem migrar para passava mais de 4 horas no ônibus, satisfazer suas necessidades e se deparam muito com grandes desafios, encarados de forma ruim (Participante 10, 17/06/19, Universidade). contraditória e não passiva (Oliveira & Os desejos e a própria relação com Feliciano, 2016). Esses desafios exigem essa mobilidade parecem diferente de transformações psicossociais, readaptações, gerações anteriores e isso se evidencia criações, resistências e enfrentamentos, os quando o jovem diz: “Cara, eu [pausa], quais são produzidos numa relação dialética assim... era uma coisa que eu já queria com as urbanidades e as ruralidades em muito, então, pra mim, foi uma sensação de construção. uma conquista pra mim” (Participante 11, 18/06/19, Universidade). Consideramos, O modo de vida dos jovens migrantes e a portanto, que migrar do rural para a urbano pobreza pode impactar diferentes dimensões, dentre elas pode contribuir com elementos de Alguns dados apontaram elementos fortalecimento e pode representar um que refletem a realidade de pobreza estímulo à liberdade e à autonomia que o vivenciada pelos estudantes universitários, jovem tanto almeja, a partir da relação com dentre estes o fato de serem beneficiários do novas responsabilidades: Programa Bolsa Família. O Programa Bolsa Essa experiência, pra mim, é boa Família é parte da Política de Assistência que a gente vê e a gente vive novas Social (PNAS), a nível de Proteção Social experiências, se Básica (PSB), e se configura como uma locomovendo a gente vai ganhando política de transferência de renda, com foco mais na amenização da extrema pobreza (Brasil, né? A gente responsabilidades, ficando com mais responsabilidade pra nossa própria pessoa. Antes era nossa família que tinha 2010). De modo geral, as discussões sobre essa pobreza têm partido da ideia de ausência de responsabilidade, hoje não. Hoje a renda. É preciso ressaltar como o acesso à gente se torna mais responsável universidade é possibilitado a partir da (Participante garantia de bolsa (renda). Apreende-se, no 6, Universidade). 12/06/19, discurso de um dos jovens entrevistados, os |78 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ impactos da questão econômica para a realizar mais de uma refeição diária. Ao permanência no ensino superior. Filho de analisar outros aspectos, observou-se que a diarista, ele reconhece: entrada falta na universidade implicou a Assim, questões financeiras, no mudança de moradia do rural para urbano e começo, foi um pouco difícil. a necessidade de mudar de residência para Porque recebia cursar o ensino superior. O caminho nenhuma bolsa ainda, minha mãe encontrado por tais jovens migrantes foi a teve que me bancar aqui, né? Como habitação economia só começa no segundo casa/apartamento com os amigos, o que semestre, aí a gente meio que não exige tem acesso no primeiro semestre às cotidianas. bolsas, porque elas geralmente saem Ao como eu não coletiva, rearranjo e dividindo novas adaptações os critérios analisar no começo do ano. Aí minha mãe estabelecidos pela Política Nacional de teve que me bancar aqui e não é tão Assistência barato assim, pra quem trabalha Instituições de Ensino Superior (IES) como diarista. Quanto às questões Federais, financeiras, foi um pouco difícil preponderância da dimensão econômica, (Participante pois o aluno considerado pobre é aquele 5, 21/06/19, Estudantil pode-se (PNAE) constatar das a Universidade). com renda familiar per capita igual ou Além da questão monetária, existe a inferior a um salário mínimo. Segundo de acesso a saúde, educação, Ximenes, Moura Júnior, Cruz, Brito e cultura, lazer, habitação, Sarriera (2016), a perspectiva adotada pelo dentre outros direitos básicos, o que poder público para mensuração da pobreza caracteriza a multidimensionalidade da pauta-se pobreza (Sen, 2000). Nesse sentido, mesmo configurando uma visão unidimensional. A que a família possa auxiliar no aluguel desse consequência disso recai também sobre o estudante que mora na cidade e cursa a estabelecimento de formas de intervenção universidade, há outros elementos que são que, nesse caso, torna-se reducionista por parte da realidade de privação, tais como a naturalizar uma visão de pobreza que se alimentação. Os enfrenta apenas pela via da transferência de realizar principais alimentação, as estudantes relataram refeições no na determinação da renda, renda. Restaurante Universitário (RU), o qual é Ao se pensar na dinâmica social uma política de assistência estudantil que constituída no campo, evidencia-se que, tem contribuído para o fato de conseguirem além das desigualdades globais, existem |79 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ também dificuldades e limitações locais e dedicação e frequência), materiais de específicas, que se mostram nas condições estudo/laboratório/prática, de privação social, cultural e psicológica. socioinstitucionais e pedagógicas, dentre Os modos de vida dos jovens camponeses outros elementos. Para os jovens, o são atravessados pela realidade de pobreza rendimento acadêmico está associado à multidimensional 2000). Tal dedicação de horas de estudo em casa, o que um fator sinaliza a importância do uso do tempo e de preponderante para o modo de vida e o sua aplicação para o estudo. Isso se projeto de futuro da juventude pobre rural, evidencia no seguinte discurso: “Essa coisa o que denota, muitas vezes, uma realidade de atravessada e cidade/urbano] foi bom porque o meu impossibilidades. Miranda, Giagozommi e rendimento acadêmico melhorou e, a partir Fiorott (2021) pontuam que a interseção do momento que eu comecei a morar na entre adolescência e vulnerabilidade social cidade, eu não reprovei mais disciplina” tem produzido diversos estigmas, o que nos (Participante 3, 26/06/19, Universidade). fenômeno (Sen, questões complexo por é limitações instiga a pensar estratégias e espaços de enfrentamento a tais questões. vir morar aqui [referindo-se a Nesse ínterim, pode-se questionar até que ponto as condições objetivas e Ademais, Redin (2017) reconhece concretas da vida influenciam a vida desses que as políticas públicas têm oportunizado jovens, até mesmo diante da escolha do o acesso dessa juventude ao ensino curso. Essa questão também se destaca superior, segmento quando um jovem participante reconhece específico ainda está marginalizado na que escolheu seu curso pelas condições de política educacional, pois não há ainda poder cursá-lo, e não exatamente pela amplas opções de escolha de curso, além de identificação com a área: “a minha escolha precárias condições que viabilizem não só o do curso principalmente se deu porque era o acesso, mas também a permanência e a curso noturno e porque tinha o transporte conclusão do curso. universitário no Embora entanto, se tal reconheça que que vinha à noite” (Participante 3, 26/06/19, Universidade). O atualmente o jovem de classe popular está fato é que alcançando o ensino superior, é necessário financeiras questionar quais os lugares que tal grupo exclusivamente está acessando: quais cursos, estratégias de possibilidades de escolha do curso. De um permanência, relação com as demandas de modo geral, observou-se que os cursos mais cada curso (carga horária, tempo de acessíveis para esses jovens de origem e aqueles com condições tempo para dedicar-se ao estudo têm mais |80 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ popular são os de licenciatura, como pressiona por políticas de permanência e Pedagogia, Geografia, conclusão do ensino superior, bem como História, o que implica diretamente em seus por contribuições dos diversos saberes projetos de futuro na sua comunidade: ser científicos. Matemática, professor. Já os cursos considerados de alta Ao trazer a Psicologia para esse demanda, como o de Medicina, possuem campo em disputa, precisa-se pensar sobre um percentual bem diminuto de jovens as implicações psicológicas da vivência no pobres que conseguem alcançar esse lugar. ensino superior para o modo de vida desses Dessa forma, entendemos que não se trata jovens. Existe um capital que é social, diretamente de incapacidade do jovem cultural e psicológico, assim como uma pobre, mas refere-se à falta de condições mudança de hábito e de cultura. Nesse históricas, estruturais e objetivas para o sentido, surgem novas exigências, modos acesso de tal segmento a esse espaço. Tais de pensar, de fazer e de se relacionar que fatores causam devem ser pensados como impactos na estruturação do indicadores para avaliar as condições de psiquismo desses sujeitos. Mais do que uma acesso ao ensino superior. nova vivência ou a ocupação de um novo As políticas de inclusão têm espaço físico e geográfico, a migração proporcionado o acesso dos jovens pobres à rural-urbana de tais jovens universitários universidade, o que representa um avanço envolve perdas e lutos, novas relações, na realidade social brasileira. Esse avanço outras tem gerados frutos e transformações que mudanças objetivas e subjetivas de vida, demandam novas questões, relacionadas além de dificuldades e desafios ligadas a aos desafios da permanência e da conclusão elas, atravessam a constituição psicossocial do percurso na universidade. Com Zago desses jovens, cuja especificidade é advir de (2006) pode-se reforçar a importância do camadas pobres, comunidades rurais e/ou acesso e deve-se questionar sobre como pequenas cidades do interior. Esse conjunto favorecer a permanência do jovem no de fatores inerentes à experiência da ensino superior. Assim, é preciso repensar migração tendem a gerar implicações uma universidade que esteja comprometida sociais com a realidade desse segmento, cuja desenvolvimento desses jovens. expectativas e e sonhos. psicológicas Tais para o presença é recente nesse nível de ensino. Isso representa novos desafios e caminhos Efeitos psicossociais da migração que precisam considerar um outro perfil social, econômico e subjetivo. Tal realidade |81 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ Pensar sobre migração e seus efeitos psicossociais nos exige considerar um pouco de exclusão, mas só no a início, depois eu já conheço mais ou dimensão da saúde dos sujeitos. Desse menos alguns pontos principais da modo, para além de um paradigma cidade (Participante 7, 12/06/19, biológico ou psicológico, é necessário Universidade). entender e reconhecer esse processo de A partir disso, é perceptível que migrar como um fenômeno que extrapola ocupar um novo território exige reajustes uma mera mudança de lugar físico, pois psicológicos, uma vez que a má adaptação produz efeitos também nas relações sociais pode gerar consequências psíquicas graves. e na constituição do psiquismo dos sujeitos Isso ocorre porque o processo migratório que se mudam. repercute em novos padrões e valores Parte-se da compreensão que as mudanças são intrínsecas culturais no modo de vida dos jovens que ao migram. Evidencia-se que o contato com desenvolvimento e à existência humana. No uma nova cultura impõe um conflito e pode caso da migração, essas mudanças podem gerar uma crise no sujeito, exigindo uma gerar ampliação cultural ou imposição de readaptação (Dantas et al., 2010). Becker e fronteiras e preconceitos (Dantas, Ueno, Martins-Borges Leifert & Suguiura, 2010). Quando se fala dificuldades no processo de adaptação de migração, refere-se a determinados tipos cultural, pois, segundo as autoras, isso está de mudanças, as quais iniciam no lugar de relacionado às perdas do que foi deixado moradia e repercutem em outras dimensões. nas Desse modo, é indubitável que sair de um psicológicas, como a ruptura dos vínculos lugar para morar em outro não significa afetivos, conforme traz o jovem em seu apenas ultrapassar determinada fronteira discurso: (2015) dimensões apontam materiais, sociais as e geográfica, mas também implica enfrentar Pra falar a verdade, eu sempre tive mudanças e acostumado a acordar, ver meu pai, emocionais, que envolvem adaptação e ver minha mãe, ver minhas irmãs em integração, conforme o discurso a seguir: casa e eu acostumado com aquela sociais, psicológicas Assim só foi ruim pra me adaptar, rotina, digamos, de todo dia; porque eu não conhecia a cidade e detalhe: interior é um local que não muitos dos meus colegas são daqui tem movimento, é muito simples, da cidade, aí eu ficava um pouco não tem nenhum trânsito... O perdida no início de não saber onde trânsito dos carros, ‘pelo amor de era tal lugar e tal lugar, aí fica meio Deus’, tipo, não existe isso, é um |82 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ lugar muito sossegado. Aí você, do adaptação psicológica e cultural. De acordo nada, se muda pra um lugar, você tá com Monteiro Neto (2008), a primeira está morando numa cidade, a rotina associada à produção do stress, e a segunda muda abruptamente e aí rapaz... está Você se sente que tá no lugar errado, desenvolvimento de habilidades sociais; essa que é a verdade (Participante 9, para 16/05/19, Universidade). interligadas. No entanto, essas distintas Essa adaptação envolve relação modalidades baseada o na autor, aprendizagem tais dimensões de estão adaptação estão e não com nova cultura e novo clima, perda de necessariamente no âmbito do vínculos e desafios para construção de patológico, principalmente quando se novos. Esse conjunto de fatores em configuram de maneira efêmera, em que se interação coloca a migração como um parte de um processo de inserção e fenômeno desencadeador de mudanças, as reconhecimento social na nova realidade quais podem gerar impactos mais ou menos (Dantas et al., 2010). instáveis na vida dos sujeitos e exigem Mühlen, Dewes e Leite (2010) pensar sobre questões como o processo fazem referência ao estresse de aculturação migratório, a adaptação, a integração e os – termo usado para designar o resultado da novos modos de socialização. Nesse relação do sujeito com duas culturas sentido, segundo um jovem participante da diferentes e seus impactos estressantes, pesquisa, como o conflito de valores e a É, é bastante impactante, porque a discriminação. Segundo Ramos (2009), as gente... um classes sociais mais baixas e com pouca ambiente familiar que a gente, tipo, rede de apoio podem ser indicadoras de se sente protegido, essas coisas, aí maior taxa de estresse entre os que vem morar com gente desconhecido vivenciam a migração. Nesse caso, é de outra cidade e até mesmo de importante outros estados. A gente passa experiências migratórias praticamente, assim, um tempo de psicossociais para adaptação, seis meses, um semestre identidade do sujeito, conforme ilustrado (Participante neste relato: [pausa] tipo, 6, tem 12/06/19, destacar a a relação e entre impactos constituição da Universidade). Foi bem complicado... é diferente, A adaptação está relacionada à por mais que não seja muito dimensão pessoal ou psicológica e ao distante, é diferente as coisas, aqui aspecto interacional, uma vez que existe a são diferentes, você tem que |83 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ resolver tudo sozinha, lá não, lá todo interferir na constituição das redes sociais e mundo se conhece, você pode pedir de pertencimento do sujeito a seu lugar de informação pra qualquer pessoa, origem (Dantas et al., 2010). aqui não, aqui é cada um por si e Nesse aspecto, o suporte social Deus para todos (risos) (Participante exerce um papel fundamental na relação 10, 17/06/19, Universidade). estresse e doença, sendo determinante na Lidar diferenças produção da saúde (Mühlen et al., 2010). provenientes do encontro de duas culturas Percebe-se isso no discurso do jovem sobre aponta para alguns desafios que a migração seu processo de adaptação: “Eu me adaptei impõe à identidade e ao desenvolvimento. rapidamente, fiz amizades, essas coisas, fui Sob essa ótica, as implicações psicológicas me ‘interando’ na faculdade, interagindo envolvidas com outros estudantes. Mas, realmente, no com na as experiência migratória podem também estar relacionadas às início é bem mudanças que refletem no dia-a-dia e nos bastante” hábitos: Universidade). impactante, (Participante 6, influencia 12/06/19, O que eu senti mais falta talvez seja Dentre os aspectos que facilitaram o da rotina mais leve, porque lá eu processo de adaptação a uma nova cultura, assistia televisão, eu estudava na Becker e Martins-Borges (2015) apontaram hora que eu queria, né? Assim, não o apoio social, o envolvimento com práticas tinha uma coisa delimitada; aqui, religiosas e o sentimento de esperança. por exemplo, eu passo o dia todinho Ramos (2009) defende que a constituição de na faculdade, chego em casa à noite, redes de apoio pode atenuar a ocorrência do depende do semestre também, aí, estresse e do sofrimento, uma vez que tais quando chega em casa à noite, ainda relações reduzem as possibilidades de tem que estudar mais, mas você está isolamento do sujeito. Em síntese, Mühlen falando em relação a saudade, né? et al. (2010) acreditam que aspectos como a (Participante relação com a família e o suporte social 7, 12/06/19, Universidade). Essas podem favorecer uma melhor adaptação psicológicas cultural, ou, ao contrário, podem ser podem envolver também as relações dos considerados fatores de risco e gerarem sujeitos migrantes com os seus núcleos stress de aculturação. familiares, implicações por conta do nível de Considera-se que a sobreposição de diferenciação, ligado à maturidade, e nível culturas, de fusão emocional, assim como podem migratório, provenientes produz do processo implicações |84 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ psicossociais, o que exige pensar nessas repercussões a nível da realidade rural- Considerações finais urbana e, especificamente, quando se refere às mudanças socioculturais existentes entre Este estudo é relevante na medida as cidades grandes e cidades do interior ou em que discute a realidade das juventudes rural, refletindo de que modo tais mudanças rurais na perspectiva da Psicologia, com são vivenciadas e significadas pelos jovens. foco nos reflexos psicossociais para a Para Dantas (2017), as instituições precisam experiência realizar fundamental trabalhos preventivos para na universidade, para formulação tema e adaptação e, além disso, promover a desenvolvimento de políticas estudantis. A integração novo partir dos resultados, foi possível identificar ambiente, dados os processos psicológicos que os jovens rurais que optam pela que estão envolvidos na transição e no migração e pelo ensino superior tornam-se encontro de diferentes culturas. uma mão de obra a menos para contribuir dos estudantes Apontamos perspectiva com a produção econômica familiar e, além intercultural como um pensamento que disso, por serem estudantes, passam a pode ser mais discutido no campo da onerar gastos, o que gera um problema educação públicas, diante da realidade da família rural pobre. efeitos Desse modo, a opção pelo estudo pode ser psicossociais vivenciados pelos jovens vista, a priori, com pesar, pois o caminho camponeses que migram. Clemente e esperado para esse jovem rural, exigido Morosini (2020) defendem a importância de pelas suas próprias condições sociais, é o do incentivar ao trabalho, mais comumente a reprodução das competências funções familiares e agrícolas. Diante desse chamando e a ao das políticas atenção para didáticas desenvolvimento que de os visem interculturais no ensino superior. Dessa cenário, maneira, afirmamos que a universidade não naturalização da imagem do jovem rural, é um espaço neutro e imutável e que, associada à figura do filho de agricultores portanto, e familiares ou camponeses que vive um transformada para acolher esses novos período de aprendizagens e preparação para perfis os suceder os pais na agricultura. Esse jovem conhecimentos e vivências desses jovens torna-se, então, parte de uma relação transformadores da própria universidade. hierarquizada, reproduzida de uma geração Por fim, é necessário aprofundar as políticas para outra. A partir dessa desnaturalização, de assistência estudantil no ensino superior. torna-se possível refletir sobre como esses precisa de ser estudantes, questionada tornando deve-se problematizar a |85 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ jovens rurais desacomodam certas vem representando uma conquista por parte estruturas do que é considerado “urbano” e da camada mais pobre da sociedade, no ainda apostar na ideia de juventudes as entanto, o processo de permanência e quais passam a se constituir como categoria conclusão ainda implica enormes desafios. indentitária Consideramos a importância de ampliar a que almeja outras oportunidades de vida. pesquisa, além de contemplar uma maior É necessário enxergar que as diversidade étnica e cultural dos jovens, políticas de estímulo para permanência do visto que foi uma limitação deste artigo. jovem no meio rural podem, por vezes, Vislumbramos ainda que a possibilidade de funcionar como projetos de fixação da aprofundar as peculiaridades da juventude e juventude no campo. Nesse aspecto, a seus entrelaces com a pobreza rural questão não está na fixação do jovem no nordestina pode produzir questões a serem campo, mas na criação de possibilidades de melhor exploradas em futuros estudos. escolha para aqueles que lá querem Referências permanecer, assim como criar condições para aqueles que querem migrar. Isso porque os jovens que vivenciam a realidade de migração para centros urbanos podem se deparar com dificuldades maiores, tais como condições de vulnerabilidade psicossocial, insegurança, alto índice de violência na cidade, maiores desigualdades sociais, diversidades culturais, políticas, econômicas e religiosas, dificuldades de locomoção e de interações com os outros, etc. Isto posto, consideramos que este estudo evidencia o quão necessário e urgente é pensar Alves, M. Z. & Dayrell, J. T. (2015). Transnacionalismo, juventude rural e a busca de reconhecimento. Educação e Pesquisa, 41(spe), 1455-1471. doi: http://dx.doi.org/10.1590/S15179702201508143396 sobre acesso e permanência desses jovens estudantes pobres na universidade, principalmente na última década. Além disso, vale considerar este debate como uma questão ético-política emergente. É fato que o ensino superior Baeninger, R. (2012). Rotatividade migratória: um novo olhar para as migrações internas no Brasil. Revista Interdisciplinar da Mobilidade Humana, 20(39), 77-100. Recuperado de https://www.scielo.br/j/remhu/a/mrV MskqfZGB3w5t7wjfBKHR/?format =pdf&lang=pt Barbieri, A. F. (2011). Mudanças climáticas, mobilidade populacional e cenários de vulnerabilidade para o Brasil. Revista Interdisciplinar da Mobilidade Humana, 19(36), 95-112. Recuperado de http://www.redalyc.org/articulo.oa?i d=407042013005 Barcellos, S. B. (2015). A formulação das políticas públicas para a juventude |86 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ rural no Brasil e os elementos constitutivos desse processo social. Mundo Agrário, 16(32), 1-32. Recuperado de https://www.mundoagrario.unlp.edu. ar/article/view/MAv16n32a10 Bardin, L. (1977). Análise de Conteúdo. Lisboa: Edições 70. Becker, A. P. S. & Martins-Borges, L. M. (2015). Dimensões psicossociais da imigração no contexto familiar. Boletim Academia Paulista de Psicologia, 35(88), 126-144. Recuperado de https://www.redalyc.org/pdf/946/946 40400009.pdf Bittencourt, L. N., Matheus, T. V. & Tavanti, R. M. (2018). Ações culturais e juventudes em m’ boi mirim: um debate sobre políticas públicas e vulnerabilidades. Athenea Digital, 18(3), 1-25. doi: https://doi.org/10.5565/rev/athenea.1 975 Brasil. (2010). Programa Bolsa Família. Brasília: Secretaria Especial de Desenvolvimento. Recuperado de http://mds.gov.br/assuntos/bolsafamilia Bulhões, L. & Abrantes, A. A (2016). Idade adulta e o desenvolvimento psíquico na sociedade de classes – juventude e trabalho. In M. Martins, A. A. Abrantes & M. G. D. Facci (Orgs.), A periodização histórico-cultural do desenvolvimento psíquicodo nascimento a velhice (pp. 241-265). Campinas, SP: Autores Associados. Caetano, E. & Azevedo, E. E. F. N. (2017). Políticas Públicas e Juventudes: Algumas particularidades do caso brasileiro. Polis, 16(48), 37-59. Recuperado de https://www.redalyc.org/pdf/305/305 54026003.pdf Castro, E. G. (2009). Juventude rural no Brasil: processos de exclusão e a construção de um ator político. Revista Latinoamericana de Ciencias Sociales, Niñez y Juventud, Manizales, 7(1), 179-208. Recuperado de https://www.redalyc.org/pdf/773/773 07108.pdf Clemente, F. A. S. & Morosini, M. C. (2020). Competências interculturais: interlocuções conceituais e uma proposta de releitura para a educação superior. Educação e Pesquisa, 46, 120. doi: https://doi.org/10.1590/s16784634202046216262 Corrochano, M. C. (2013). Jovens trabalhadores: expectativas de acesso ao ensino superior. Avaliação, 18(1), 23-44. Recuperado de https://www.redalyc.org/pdf/2191/21 9125744003.pdf Dantas, S., Ueno, L., Leifert, G. & Suguiura, M. (2010). Identidade, migração e suas dimensões psicossociais. Revista Interdisciplinar da Mobilidade Humana, 18(34), 45-60. Recuperado de https://www.redalyc.org/pdf/4070/40 7042011004.pdf Dantas, S. (2017). Saúde mental, interculturalidade e imigração. Revista USP, (114), 55-70. Recuperado de https://www.revistas.usp.br/revusp/ar ticle/view/142368/137500 Freitas, I. C. M. & Braga, J. R. M. (2013). Os universitários viajantes: suas práticas e sociabilidades. O público e o privado, (21), 91-108. Recuperado |87 Rev. Polis e Psique, 2022; 12(3): 68-89. Implicações psicossociais da migração rural-urbana para jovens universitários ___________________________________________________________________________ de http://www.seer.uece.br/?journal=op ublicoeoprivado&page=article&op= view&path%5B%5D=651&path%5 B%5D=755 Furlani, D. D. & Bomfim, Z. A. C. (2010). Juventude e afetividade: tecendo projetos de vida pela construção dos mapas afetivos. Psicologia e Sociedade, 22(1), 50-59. Recuperado de https://www.redalyc.org/pdf/3093/30 9326443006.pdf Leite, J. F., Macedo, J. P. S., Dimenstein, M. & Dantas, C. (2013). A formação em Psicologia para a atuação em contextos rurais. In J. F. Leite & M. D. Dimenstein (Orgs.), Psicologia e contextos rurais (pp. 27-55). Natal: EDUFRN. Marinho, D. L. (2014). Estratégias de Transformação Social no Campo: A Formação de Jovens da Escola Família Agrícola de Marabá-PA. In M. A. Menezes, V. L. Stropasolas & S. B. Barcellos (Orgs.), Juventude Rural e Políticas Públicas no Brasil (p. 164-176). Brasília: Presidência da República. Menezes, M. A., Stropasolas, V. L. & Barcellos, S. B. (2014). Juventude Rural e Políticas Públicas no Brasil. Brasília: Presidência da República. Miranda, P. R. A., Giacomozzi, A. I. & Fiorott, J. G. (2021). Processos Grupais com Adolescentes em Situação de Vulnerabilidade Social. Rev. Polis e Psique, 11(2), 225-244. Recuperado de https://seer.ufrgs.br/PolisePsique/arti cle/view/103393/63739 Monteiro Neto, F. F. (2008). Adaptação psicológica e sociocultural de jovens Oriundos de famílias imigrantes dos PALOP. Psicologia, Educação e Cultura, 12(2), 435-450. Recuperado de https://hdl.handle.net/10216/99793 Mühlen, B. K. V., Dewes, D. & Leite, J. C. C. (2010). Stress e processo de adaptação em pessoas que mudam de país: uma revisão de literatura. Portal Rede Metodista, 12(24). doi: http://dx.doi.org/10.15602/19839480/cmbs.v12n24p59-67 Oliveira, L. B. & Feliciano, C. A. (2016). Entre o campo e a cidade: reflexões sobre o espaço da juventude camponesa. Revista Pegada, 17(2), 108-225. Recuperado de http://revista.fct.unesp.br/index.php/p egada/article/viewFile/4678/3588 Ramos, N. (2009). Saúde, migração e direitos humanos. Mudanças, 17(1), 1-11. doi: https://doi.org/10.15603/21761019/mud.v17n1p1-11 Redin, E. & Silveira, P. R. C. (2012). Juventude rural: experiências e perspectivas. In V. F. Santos, H. A. G. Vela & P. R. C. Silveira (Orgs.), Educação rural no mundo contemporâneo (pp. 175-208). Santa Maria: UFSM. Redin, E. (2017). Políticas educacionais e juventude rural no ensino superior. Educar em Revista, (63), 237-252. doi: https://doi.org/10.1590/01044060.44861 Sen, A. (2000). Desenvolvimento como liberdade. São Paulo: Companhia das Letras. Ximenes, V. M., Moura Júnior, J. F., Cruz, J. M., Brito, L. S. & Sarriera, J. C. (2016). Pobreza Multidimensional e seus Aspectos Subjetivos em Contextos Rurais e Urbanos |88 Rev. Polis e Psique, 2022; 12(3): 68-89. Silva, A. M. S. & Ximenes, V. M. ___________________________________________________________________________ Nordestinos. Estudos de Psicoogia., 21(2), 146-156. doi: 10.5935/16784669.20160015 Zago, N. (2006). Do acesso à permanência no ensino superior: percursos de estudantes universitários de percursos de estudantes universitários de camadas populares camadas populares. Revista Brasileira de Educação, 11(32), 226-370. doi: https://doi.org/10.1590/S141324782006000200003 Alexsandra Maria Sousa Silva. Psicóloga, Doutora Psicologia pela Universidade Federal do Ceará, professora da Faculdade Luciano Feijão (FLF). ORCID: https://orcid.org/0000-00033362-2685 E-mail: alexsandramss88@gmail.com Verônica Morais Ximenes. Psicóloga, Doutora em Psicologia pela Universidad de Barcelona (Espanha), Pós-Doutora em Psicologia pela UFRGS, Docente do Programa de Pós-Graduação em Psicologia da Universidade Federal do Ceará e Pesquisadora do CNPq PQ-2. ORCID: https://orcid.org/0000-00033564-8555 E-mail: vemorais@yahoo.com.br Submetido em: 27/08/2021 1ª rodada: 31/08/2021 2ª rodada: 09/12/2022 Aceito em: 12/01/2023 |89 Rev. Polis e Psique, 2022; 12(3): 68-89.
https://openalex.org/W4362629245
https://figshare.com/ndownloader/files/40006126
English
null
Supplementary Figures from A Transcriptionally Distinct CXCL13&lt;sup&gt;+&lt;/sup&gt;CD103&lt;sup&gt;+&lt;/sup&gt;CD8&lt;sup&gt;+&lt;/sup&gt; T-cell Population Is Associated with B-cell Recruitment and Neoantigen Load in Human Cancer
null
2,023
cc-by
775
Supplementary figure 1 1 A, t-Distributed Stochastic Neighbour Embedding (tSNE) of flow cytometry data of ovarian tumors 2 used for mRNA sequencing. B,C, Relative fluorescent intensity on tSNE per flow cytometry marker 3 shown for ovarian cancer 229 (B) and ovarian cancer 232 (C) used in mRNA sequencing. 4 5 Supplemental figure 2. 6 A,B,C Chemokine arrays for donor 1 using normal exposure (A) and overexposure (B) and for donor 7 2 (C) and donor 3 (D). 8 9 Supplemental figure 3. 10 A, B Survival analyses for tumors with low neo-antigen load, comparing CXCL13highITGAEhigh versus 11 CXCL13lowITGAElow microsatellite stable uterine cancers using TCGA data (A) and CD103high versus 12 CD103low mismatch repair proficient uterine cancers using immunohistochemical staining (B). 13 Supplementary figure 1 Supplementary figure 1 tSNE1 tSNE2 OC229 OC232 CD8+ CD103+ CD8+ CD103- CD56+ CD4+ CD103- CD4+ CD103+ CD3+ TCRαβ+ unclassified OC229 CD3 CD56 TCRαβ CD8α CD8β CD103 CD4 OC232 CD3 CD56 TCRαβ CD4 low high relative fluorescent intensity A B C tSNE1 tSNE2 Supplementary figure 1 A tSNE1 tSNE2 OC229 OC232 CD8+ CD103+ CD8+ CD103- CD56+ CD4+ CD103- CD4+ CD103+ CD3+ TCRαβ+ unclassified A tSNE1 tSNE2 OC229 OC232 CD8+ CD103+ CD8+ CD103- CD56+ CD4+ CD103- CD4+ CD103+ CD3+ TCRαβ+ unclassified A tSNE1 tSNE2 OC229 OC232 CD8+ CD103+ CD8+ CD103- CD56+ CD4+ CD103- CD4+ CD103+ CD3+ TCRαβ+ unclassified tSNE1 tSNE2 OC229 CD3 CD56 TCRαβ CD8α CD8β CD103 CD4 OC232 CD3 CD56 TCRαβ CD8α CD8β CD103 CD4 low high relative fluorescent intensity low high relative fluorescent intensity B C tSNE1 tSNE2 tSNE2 OC229 CD3 CD56 TCRαβ CD8α CD8β CD103 CD4 low high relative fluorescent intensity B tSNE1 tSNE2 B C tSNE1 tSNE2 OC232 CD3 CD56 TCRαβ CD8α CD8β CD103 CD4 low high relative fluorescent intensity tSNE2 Supplementary figure 2 Supplementary figure 2 CXCL13 anti-CD3/CD28 TGF-β anti-CD3/CD28+TGF-β control Array 2 (donor 3; without CXCL13) TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 2) TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 1) TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 1; overexposed) CXCL13 A B c CXCL13 anti-CD3/CD28 TGF-β anti-CD3/CD28+TGF-β control Array 2 (donor 3; without CXCL13) TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 2) TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 1) TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 1; overexposed) CXCL13 A B c A anti-CD3/CD28 control Array 1 (donor 1) anti-CD3/CD28 control Array 1 (donor 1; overexposed) control Array 1 (donor anti-CD3/CD28 verexposed) TGF-β anti-CD3/CD28+TGF-β anti-CD3/CD28+TGF-β TGF-β CXCL13 CXCL13 TGF-β anti-CD3/CD28 anti-CD3/CD28+TGF-β control Array 1 (donor 2) CXCL13 B control Array 1 B B c anti-CD3/CD28 TGF-β anti-CD3/CD28+TGF-β control Array 2 (donor 3; without CXCL13) c Array 2 (donor 3; without CXCL13) anti-CD3/CD28+TGF-β TGF-β CXCL13 CXCL13 Supplementary Figure 3 ++++++ ++++++++++++++++++++++++++++++ +++++++++++++ ++ +++++++++++ ++ ++++ ++ + +++++ + +++++++++++++++++++++++++++++++++++ ++++++++ + + + +++ + ++ + n=86 n=87 0 40 80 100 survival (%) 20 0 5 10 time (years) CXCL13high ITGAEhigh CXCL13low ITGAElow microsatellite stable (MSS) tumors (TCGA) 60 mismatch repair proficient (pMMR) tumors (IHC cohort) 0 40 80 100 disease-specific survival (%) 20 60 0 5 20 time (years) 25 15 10 A B CD103high (>median) CD103low (<median) n=94 n=95 p=0.0005 Supplementary Figure 3 p=0.0499 mismatch repair proficient (pMMR) tumors (IHC cohort) 0 40 80 100 disease-specific survival (%) 20 60 0 5 20 time (years) 25 15 10 B CD103high (>median) CD103low (<median) n=94 n=95 p=0.0005 499 B A ++++++ ++++++++++++++++++++++++++++++ +++++++++++++ ++ +++++++++++ ++ ++++ ++ + +++++ + +++++++++++++++++++++++++++++++++++ ++++++++ + + + +++ + ++ + n=86 n=87 0 40 80 100 survival (%) 20 0 5 10 time (years) CXCL13high ITGAEhigh CXCL13low ITGAElow microsatellite stable (MSS) tumors (TCGA) 60 A mismatch repair proficient (pMMR) tumors (IHC cohort) disease-specific survival (%) survival (%) Supplemental figure 1. Supplemental figure 1. 1 A, t-Distributed Stochastic Neighbour Embedding (tSNE) of flow cytometry data of ovarian tumors 2 used for mRNA sequencing. B,C, Relative fluorescent intensity on tSNE per flow cytometry marker 3 shown for ovarian cancer 229 (B) and ovarian cancer 232 (C) used in mRNA sequencing. 4 Supplemental figure 2. 6 A,B,C Chemokine arrays for donor 1 using normal exposure (A) and overexposure (B) and for donor 7 2 (C) and donor 3 (D). 8 9 Supplemental figure 3. 10 A, B Survival analyses for tumors with low neo-antigen load, comparing CXCL13highITGAEhigh versus 11 CXCL13lowITGAElow microsatellite stable uterine cancers using TCGA data (A) and CD103high versus 12 CD103low mismatch repair proficient uterine cancers using immunohistochemical staining (B). 13 A, B Survival analyses for tumors with low neo-antigen load, comparing CXCL13highITGAEhigh versus 11 CXCL13lowITGAElow microsatellite stable uterine cancers using TCGA data (A) and CD103high versus 12 CD103low mismatch repair proficient uterine cancers using immunohistochemical staining (B). 13 1 1
https://openalex.org/W2135235741
https://hal.science/hal-01523062/document
English
null
Multiprobabilistic Venn Predictors with Logistic Regression
IFIP advances in information and communication technology
2,012
cc-by
4,638
Multiprobabilistic Venn Predictors with Logistic Regression Ilia Nouretdinov, Dmitry Devetyarov, Brian Burford, Stephane Camuzeaux, Aleksandra Gentry-Maharaj, Ali Tiss, Celia Smith, Zhiyuan Luo, Alexey Chervonenkis, Rachel Hallett, et al. To cite this version: Ilia Nouretdinov, Dmitry Devetyarov, Brian Burford, Stephane Camuzeaux, Aleksandra Gentry- Maharaj, et al.. Multiprobabilistic Venn Predictors with Logistic Regression. 8th International Con- ference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2012, Halkidiki, Greece. pp.224-233, ￿10.1007/978-3-642-33412-2_23￿. ￿hal-01523062￿ Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-01523062 https://hal.science/hal-01523062v1 Submitted on 16 May 2017 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Multiprobabilistic Venn Predictors with Logistic Regression Ilia Nouretdinov1,∗, Dmitry Devetyarov1, Brian Burford1, Stephane Camuzeaux2, Aleksandra Gentry-Maharaj2, Ali Tiss3, Celia Smith3, Zhiyuan Luo1, Alexey Chervonenkis1, Rachel Hallett2, Volodya Vovk1, Mike Waterfield2, Rainer Cramer3, John F. Timms2, Ian Jacobs2, Usha Menon2 and Alex Gammerman1 1 Computer Learning Research Centre, Royal Holloway, University of London 2 EGA Institute for Women’s Health, University College London 3 BioCentre and Department of Chemistry, University of Reading ∗Address correspondence to I.Nouretdinov: ilia@cs.rhul.ac.uk. Abstract. This paper describes the methodology of providing multi- probability predictions for proteomic mass spectrometry data. The method- ology is based on a newly developed machine learning framework called Venn machines. They allow us to output a valid probability interval. We apply this methodology to mass spectrometry data sets in order to pre- dict the diagnosis of heart disease and early diagnoses of ovarian cancer. The experiments show that probability intervals are valid and narrow. In addition, probability intervals were compared with the output of a corresponding probability predictor. 1 Introduction Prediction of heart disease (HD) and ovarian cancer (OC) is a critical task. For some of these diseases (e.g., OC) it is especially crucial in their early stages, when the disease has no clinical symptoms. Mass spectrometry techniques are widely deployed in these problems. When predicting diagnosis based on proteomics data, very often, the classical machine learning approach is to predict the diagnosis without any measure of how accurate this prediction is. In this work we describe the methodology of proteomics mass spectrometry data analysis based on hedging predictions that is how strongly we believe in this prediction [3]. The framework of Venn machines was introduced in [5] and represents a new generation of prediction algorithms. These methods have a range of advan- tages over the known techniques. Firstly, the prediction which is made is always tailored to the object; as a result, we output a probability interval to each pa- tient’s diagnosis. Secondly, the only statistical assumption which is used is the exchangeability assumption which can be satisfied when the data sets are in random order. Strict definitions of Venn machines are given in Section 2. The main idea is as follows. We first divide examples into categories, the category assigned to an example may depend not only on the example itself, but also on its relation to the rest of examples. For each hypothesis about the new label, we classify the new object into one of the categories, and then use frequencies of labels in the chosen category as predictable distribution of the new object’s label. Due to different hypotheses, the machine outputs several (two in the binary case) probability distributions (multiprobability distribution) for the new object rather than the single one. Practically any known machine learning algorithm can be used as an underly- ing algorithm in this framework (such as Neural Network in [6] and SVM in [7]). In this work we used logistic regression as an underlying algorithm. It is popular as a method that initially outputs probabilities and provides information about relative weight of features. We will compare Venn machines predictions with probability predictions output by logistic regression. The methodology is designed for the analysis of proteomic mass spectrometry data collected in the UKCTOCS trial (for more information see www.ukctocs.org.uk) 2 Venn Machines Consider a training set consisting of object, xi, label, yi, pairs: (x1, y1) , . . . , (xn, yn). To predict a label yn+1 for a new object xn+1 = xnew, we check different hy- potheses yn+1 = y each time including the pair (xn+1, yn+1) = (xnew, y) into the set. The idea of Venn machines is based on a taxonomy function An, n ∈N, which classifies the relation between an example and the set of the other examples: τi = An+1((xi, yi), {(x1, y1), . . . , (xi−1, yi−1), (xi+1, yi+1), . . . , (xn+1, yn+1)}). τi = An+1((xi, yi), {(x1, y1), . . . , (xi−1, yi−1), (xi+1, yi+1), . . . , (xn+1, yn+1)}). Values τi are called categories and are taken from a finite set T = {τ1, τ2, . . . , τk}. Equivalently, a taxonomy function assigns to each example (xi, yi) its category τi, or in other words grouping all examples to a finite set of categories. This grouping should not depend on the order of examples within a sequence. The conventional way of using Venn ideas was as follows. Categories are formed using only the training set. For each non-empty category τ the following values are calculated: Nτ - the total number of examples from the training set, assigned to category τ, and Nτ(y′) - the number of examples within category τ having label y′. Then empirical probabilities of an object within category τ to have a label y are found as (1) Pτ(y′) = Nτ(y′)/Nτ. (1) Now, given a new object xn+1 with unknown label yn+1, one should assign it somehow to the most likely category of those already found using only the train- ing set; let it be τ ∗. Then the empirical probabilities Pτ ∗(y′) are considered as probabilities of the object xn+1 to have a label y′. The idea of conformal predic- tors [5] allows us to construct several probability distributions (multi probability distribution) of a label y′ for a new object. First we consider a hypothesis that the label yn+1 of a new object xn+1 is equal to y, (yn+1 = y). Then we add the pair (xn+1, y) to the training set and apply to this extended sequence the taxon- omy function A. This groups all the elements of the sequence to categories. 2 Venn Machines Let τ ∗(xn+1, y) be the category containing the pair (xn+1, y). Now for this category we calculate, as previously, the values Nτ ∗, Nτ ∗(y′) and empirical probability distribution (2) Pτ ∗(xn+1,y)(y′) = Nτ ∗(y′)/Nτ ∗. (2) This distribution depends implicitly on the object xn+1 and its hypothetical label y. Trying all possible hypothesises of the label yn+1 being equal to y, we obtain a set of distributions Py(y′) = Pτ ∗(xn+1,y)(y′) for all possible labels y. These distributions in general will be different, as when changing the value of y we change (in general) grouping into categories, the category τ ∗(xn+1, y), containing the pair (xn+1, y), the numbers Nτ ∗and Nτ ∗(y′). So we obtain, as the output of Venn predictors, as many distributions as the number of possible labels. This distribution depends implicitly on the object xn+1 and its hypothetical label y. Trying all possible hypothesises of the label yn+1 being equal to y, we obtain a set of distributions Py(y′) = Pτ ∗(xn+1,y)(y′) for all possible labels y. These distributions in general will be different, as when changing the value of y we change (in general) grouping into categories, the category τ ∗(xn+1, y), containing the pair (xn+1, y), the numbers Nτ ∗and Nτ ∗(y′). So we obtain, as the output of Venn predictors, as many distributions as the number of possible labels. In the two-class problem (Y = {0, 1}), Venn predictors have two probabil- ity distributions, defined by py(1) = P{yn+1 = 1}. Thus, the output can be interpreted as the interval [p−new, p+new] = [min{p0(1), p1(1)}, max{p0(1), p1(1)}] , (3) (3) which is an estimation of probability that yn+1 = 1. We will refer to p−new and p+new as lower Venn prediction and upper Venn prediction, respectively. They can be interpreted as lower and upper bounds for the probability. Thus if one sets a risk threshold θ and takes all the predictions with lower Venn prediction not smaller than θ then the expected percentage of cases between these examples should be between θ and 1 as well. A Venn predictor is entirely defined by its taxonomy. In the next section we describe a taxonomy based on the logistic regression. 3 Logistic Regression Logistic regression outputs the probability distribution of a new label. It pro- duces these distribution as follows. Suppose each object out of the training set x1, . . . , xn is an m-dimensional vector, each with corresponding labels y1, . . . , yn, ∈Y = {0, 1}. The statistical model of logistic regression is based on the assumption that P{yi = 1} = 1/ 1 + e−<xi,b> . The optimization goal for logistic regression is: n X i=1 log  1 + e(−1)yi⟨xi,b⟩ + a ⟨b, b⟩→min b . (4) (4) This formula is based on the maximum likelihood estimation for Logistic regres- sion, with an added regularisation term a ⟨b, b⟩to ensure that a minimum always This formula is based on the maximum likelihood estimation for Logistic regres- sion, with an added regularisation term a ⟨b, b⟩to ensure that a minimum always exists and avoid overfitting. In this work we always set a = 0.1. The above min- imisation problem can be solved by a gradient descent method. Denote by ˆb the solution of the optimisation problem above. For a new object xnew, the probabilistic prediction based on logistic regres- sion will be: 1 pnew = 1 1 + e−⟨xnew,ˆb⟩, (5) (5) which estimates the maximum likelihood probability that ynew = 1 if the data are generated by a distribution from a logistic model. We will call pnew a direct prediction to distinguish it from multi-probabilistic predictions produced by a Venn machine. 3.1 Logistic Regression as an Underlying Algorithm Now we can describe how logistic regression can be plugged in Venn machine as an underlying algorithm. As earlier the aim is to predict labels yi which are equal to 0 for the controls and 1 for the cases, by objects xi — vectors of features, which are intensities of the most frequent peaks in the logarithm scale. The probabilistic method of logistic regression allows us to create a new type of taxonomy. The logistic taxonomy τi, i = 1, . . . , n + 1 is defined as fol- lows. The solution of the optimisation problem ˆb is calculated for the whole set (x1, y1), . . . , (xn, yn), (xn+1, y) as a training set and is used to make direct predictions p1, . . . , pn+1 on the same (training) examples. These predictions are not fair leave-one-out predictions, but it is correct to use them for taxonomy construction. Let p′ i, i = 1, . . . , n + 1 be direct predictions pi sorted in the ascending order. Set a number of taxonomy categories K. Each of them will contain approximately equal number of examples. We use K = 5 in this research (the dependence on the parameter K is discussed in Appendix). Let L0 = 0 and L1, L2, . . . , LK be the integers closest to (n + 1)/K, 2(n + 1)/K, 3(n + 1)/K, . . . , n + 1. The category τi is then defined accordingly to the intervals formed by division points p′ L0, p′ L1, . . . , p′ LK where value pi falls: τi is the number j ∈{1, . . . , K} such that p′ Lj−1 < pi ≤p′ Lj. 4 Mass Spectrometry Data We would like to develop a methodology of providing multiprobability predic- tions for proteomic mass spectrometry data. Hence, we have to take into account the format of the data and peculiarities of the problem. Mass spectra plots can be noisy because of physical, electrical or chemical sources. Pre-processing is applied to mass spectra to get rid of these system- atic artefacts. Auxiliary goals of pre-processing are to normalize the spectra from different samples and reduce the dimensionality of the data. Pre-processing can include the following steps: smoothing by averaging the intensities within a moving window; baseline subtraction; normalization to make sure that the total amounts of ions across different samples are the same. After the true signal is extracted from mass spectra, peaks are identified in each spectrum and then aligned, that is, peaks from different spectra get related to each other and are considered as one peak. Finally, the intensities of identified peaks are calculated. Detailed description of the preprocessing of mass spectrometry samples can be found in [2]. Thus, the data we apply in our methodology is represented as intensities of identified peaks. The peaks are usually sorted by their frequency: the more ex- amples a peak is presented, the higher the rank of the peak. We usually consider a certain number of the most frequent peaks only. Thus, every object xi is a vector of features, which are intensities of the most frequent peaks. Each sample of the OC data set is also assigned a level of OC biomarker CA125 that helps discriminate OC samples from healthy samples. For this reason, for OC we will make our diagnosis based not only on MALDI-TOF data, but also on CA125 levels. The total number of samples is: 561 in HD data set (187 cases and 374 controls); 312 in OC data set (104 cases and 208 controls). Originally, each case was accompanied by two controls matched on patient age, sample collection location and sample collection date/time, among other factors. For this reason, in each data set the number of controls is twice as greater than the number of cases. Each object, xi, only comprises intensities of the most common peaks. It was shown in statistical analysis [8] that the information useful for discrimination between healthy and diseased samples is concentrated in peaks 2 and 3 in OC data. 4 Mass Spectrometry Data For this reason, each OC object comprises the five most common peaks. As for the HD data, we consider the peaks represented in at least 1/3 of samples (41 peaks) in this data set. Each case is assigned a non-negative value T(τ) — time to diagnosis con- firmed by histology/cytology for OC data and time to death for HD data. We will refer to this value as ’time to diagnosis/death’. Each sample in OC data set is also assigned a level C of the biomarker CA125. We will not be trying to predict this time, but we need it to form cross-validation sets for the early diagnostics. 5 Results and Discussion To demonstrate how the proposed methodology works in practice, we applied the designed algorithms to HD and OC proteomic data sets. In all experiments, we use leave-one-out mode: each example (xi, yi) is con- sidered as if it were a new test example and all the remaining examples in the data are treated as the training set. We are applying Venn machines with the taxonomy based on logistic re- gression to MALDI-TOF datasets. Each object xi is a vector comprising the following features: of the most frequent peaks, CA125 value (for OC dataset), additional dimension constantly equal to 1. Since logistic regression also produces probability distributions, we can com- pare the results of the application of the Venn machine based on logistic regres- sion and the probabilistic predictor of logistic regression itself. The experiments were applied to the same type of objects xi in the leave-one-out mode. No. True label Venn prediction Direct prediction 1 0 0.313–0.321 0.508 2 1 0.616–0.616 0.689 3 0 0.321–0.330 0.510 4 0 0.143–0.259 0.371 5 0 0.616–0.634 0.622 Table 1. Leave-one-out Venn predictions for HD data Results of experiments for several controls and cases of HD data are shown in Table 1 for illustrative purposes. For each example, the table contains the true label ynew, and the probability interval [p−new, p+new]. For example, Venn machines output prediction intervals [0.313, 0.321] and [0.616, 0.616] for proba- bilities that examples 1 and 2 are cases (y = 1). As prediction interval indicate, the correct labels for example 1 and 2 are 0 and 1, respectively. The table also includes predictions pnew output by logistic regression for each example. Recall that we call these predictions direct predictions as opposed to Venn predictions output by Venn machines. The table demonstrates that both direct and Venn predictions can be correct or erroneous. First, we would like to demonstrate validity of Venn predictions: true proba- bilities of label distribution are covered or almost covered by the interval between lower and upper Venn prediction. Since we do not know true probabilities of la- bel distribution, we compare empirical probabilities, that is, mean true labels with the mean direct and Venn predictions. Figure 1 is a graphical representation of corresponding cumulative results. The horizontal axis shows the number of observed examples. 5 Results and Discussion The vertical axis shows the cumulative values of: (1) true labels ynew (a solid line); (2) lower and upper Venn predictions p−new, p+new (two dot-dashed lines) and (3) cumulative direct predictions pnew (a dashed line). The examples are sorted according to direct predictions. Firstly, the plot demonstrates validity of Venn machine outputs. Secondly, we can see that probability intervals output by Venn machines are narrow (0.025 on average for HD data); hence, they are almost as precise as single probabilities. Finally, Figure 1 demonstrates that probability intervals can be more accurate than single probabilities produced by logistic regression. It can be seen from the Figure that the true labels are very different from the direct predictions but are only slightly above the upper Venn prediction up to approximately 210 examples and within the upper and lower Venn predictions for the remaining examples after this point. Thus, direct predictions can be misleading (the cancer Fig. 1. Cumulative Venn and direct predictions for the HD data Fig. 1. Cumulative Venn and direct predictions for the HD data probability is oversetimated, while Venn predictions cover true labels: an average true label lies between average low and upper bounds given by the Venn Machine. It can be said that both algorithms relied on the assumption of the mechanism generating the data — the logistic regression statistical model. However, proba- bility predictions used this mechanism directly, and Venn machines deployed the mechanism when defining the taxonomy. As a result, since the statistical model does not hold true (the opposite can be guaranteed only for artificially gener- ated data), probabilities output by logistic regression are different from empirical probabilities. In contrast, Venn machine’s validity was not affected by the fact that the model is not correct. Hence, Venn machine predictions appeared to be more accurate than singleton probability predictions. Even though the Venn machines and logistic regression produce multiproba- bility and probability predictions, respectively, their outputs can be interpreted as usual bare predictions (forced predictions). However, we should bare in mind that when we force Venn machines to output a single prediction, they lose their theoretically proven property of validity. In this section we examine the accuracy of forced predictions when we are not able to use advantages of multiprobability predictions. 5 Results and Discussion We can extract forced predictions out of Venn machines and logistic regres- sion the most intuitive way: we classify a new sample as 1 (case) if and only if pnew > 0.5 for direct prediction or p+new + p−new > 1 for Venn prediction. This will also allow us to compare accuracy of Venn predictions with the direct predictions. Given that the aim is to predict the disease as early as possible. For this reason we consider the dynamics of predictive ability of mass spectrometry peaks across the timeline: the accuracy of the proposed methodology on samples in different time slots of the fixed interval (6 months). Table 2 allows us to compare accuracy of the forced predictions by Venn machine and logistic regression for the OC data. The table demonstrates that Venn machines are comparable with logistic regression in terms of forced pre- diction accuracy: in time slots close to the moment of diagnosis Venn machine is slightly outperformed by logistic regression, then in months 5–7 they have equal accuracy, and in months 8–11 (time slots we are mostly interested in) Time Venn machine Logistic regression slot Accuracy Sensiti Specifi Accuracy Sensiti Specifi vity city vity city 0–6 90.2% 95.6% 87.5% 93.6% 85.3% 97.8% 1–7 88.1% 91.1% 86.6% 92.9% 83.9% 97.3% 2–8 76.6% 59.6% 85.1% 87.9% 78.7% 92.6% 3–9 83.3% 58.3% 95.8% 83.3% 69.4% 90.3% 4–10 75.3% 59.3% 83.3% 82.7% 66.7% 90.7% 5–11 79.7% 52.2% 93.5% 79.7% 56.5% 91.3% 6–12 81.7% 55.0% 95.0% 81.7% 55.0% 95.0% 7–13 70.6% 35.3% 88.2% 70.6% 35.3% 88.2% 8–14 82.4% 52.9% 97.1% 78.4% 47.1% 94.1% 9–15 75.0% 45.0% 90.0% 71.7% 35.0% 90.0% 10–16 73.8% 67.9% 76.8% 67.9% 25.0% 89.3% 11–17 66.7% 50.0% 75.0% 64.3% 17.9% 87.5% 12–18 59.5% 32.1% 73.2% 61.9% 7.1% 89.3% 13–19 66.7% 33.3% 83.3% 65.6% 13.3% 91.7% 14–20 64.0% 24.0% 84.0% 65.3% 12.0% 92.0% 15–21 65.0% 40.0% 77.5% 71.7% 20.0% 97.5% 16–22 33.3% 0.0% 50.0% 63.3% 10.0% 90.0% Table 2. Dynamics of Venn machine and logistic regression performance on the OC dataset Venn machine overperforms logistic regression. Venn machines produce predic- tions with accuracy higher than 73% up to 10 months in advance of the moment of diagnosis. For HD we consider the whole dataset rather than dynamics across the time- line, since it is sufficient to predict this disease at any moment to prevent the consequences. The accuracy of the application of Venn machines to the HD data is 69.9%. 5 Results and Discussion The accuracy is again comparable with the accuracy of underlying algorithms: 67.9%. Personal ID Months in advance Prediction interval 29 13 0.22–0.39 10 0.59–0.71 4 0.88–0.94 39 10 0.53–0.71 4 0.44–0.94 2 0.96–1.00 1 0.97–1.00 Table 3. Dynamics of prediction intervals output by Venn machines for measurements taken from the same OC case Table 3. Dynamics of prediction intervals output by Venn machines for measurements taken from the same OC case Table 3. Dynamics of prediction intervals output by Venn machines for measurements taken from the same OC case Table 3 shows the dynamics of prediction intervals output by Venn machines for samples 29 and 39. Each row corresponds to a single measurement. Column 2 demonstrates how early in advance this measurement was taken. These samples with multiple measurements illustrate two trends in probability interval change. First, the interval is getting narrower when the moment of diagnosis is approach- ing, which means that two probability distributions produced by Venn machines are getting closer to each other, and as a result, the overall prediction is get- ting more precise. This also means that the logistic regression as the underlying model becomes more adequate when the time to diagnostics is closer. Second, the interval is moving towards 1. This implies that we have more trust in our prediction and the prediction is indeed correct. 6 Conclusion This paper introduced the methodology of hedging predictions for proteomic mass spectrometry data. We applied the described methodology to the MALDI- TOF data sets and demonstrated how it works. We empirically confirmed the validity of Venn machines and demonstrated that Venn machines can provide narrow probability intervals that are more accurate than the probabilities pro- vided by its underlying algorithm. Even though Venn machines produce multiprobabilistic predictions, their output can be interpreted as predictions without hedging, similarly to the output of conventional machine learning methods. It was demonstrated that when forced to make single predictions, our methodology provides accuracy similar to the accuracy of the underlying algorithms. As a result, this methodology can provide high accuracy well in advance of the moment of the disease diagnosis. Acknowledgement This work was supported by EraSysBio+ grant funds from the European Union, BBSRC and BMBF ”Living with uninvited guests: comparing plant and animal responses to endocytic invasions” to the Salmonella Host Interactions Project European Consortium; MRC grant G0802594 (Application of conformal predic- tors to FMRI research); MRC grant G0301107 (Proteomic analysis of the human serum proteome); Veterinary Laboratories Agency (VLA) of Department for En- vironment, Food and Rural Affairs (Defra) on Machine learning algorithms for analysis of large veterinary datasets; a grant from The National Natural Sci- ence Foundation of China (No.61128003); and by grant ’Development of New Venn Prediction Methods for Osteoporosis Risk Assessment’ from the Cyprus Research Promotion Foundation. 3. Gammerman, A., Vovk, V. Hedging predictions in machine learning. Computer Journal. Vol. 50, Num. 2, 151–163 (2007). 1. Dawid, A.P., Probability Forecasting. Encyclopedia of Statistical Sciences. Wiley, New York. Vol. 7, 210–218 (1985). References 1. Dawid, A.P., Probability Forecasting. Encyclopedia of Statistical Sciences. Wiley, New York. Vol. 7, 210–218 (1985). 2. Devetyarov, D., Nouretdinov, I., Burford, B., Luo, Z., Chervonenkis, A., Vovk, V., Waterfield, M., Tiss, A., Smith, C., Cramer, R., Gentry-Maharaj, A., Hallett, R., Camuzeaux, S., Ford, J., Timms, J., Menon, U., Jacobs, I., Gammerman, A. Analysis of serial UKCTOCS-OC data: discriminating abilities of proteomics peaks (Technical report), http://www.clrc.rhul.ac.uk/projects/proteomic3.htm (2008). 3. Gammerman, A., Vovk, V. Hedging predictions in machine learning. Computer Journal. Vol. 50, Num. 2, 151–163 (2007). 4. Menon, U., Skates, S.J., Lewis, S. Rosenthal, A.N., Rufford, B., Sibley, K., Macdon- ald, N., Dawnay, A., Jeyarajah, A., Bast-Jr., R.C., Oram, D., Jacobs, I.J. Prospective study using the risk of ovarian cancer algorithm to screen for ovarian cancer. Journal of Clinical Oncology. Vol.23, 7919–7926 (2005). 5. Vovk, V., Gammerman, A., Shafer, G. Algorithmic learning in a random world. Springer, New York (2005). 6. Papadopoulos, H. Reliable Probabilistic Prediction for Medical Decision Support. In Proceedings of the 7th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI 2011), IFIP AICT 364, pp. 265–274. Springer (2011). doi: 10.1007/978-3-642-23960-1 32. 7. Zhou, C., Nouretdinov, I., Luo, Z., Adamskiy, D., Coldham, N., Gammerman, A. A Comparison of Venn Machine with Platt’s Method in Probabilistic Outputs. 12th INNS EANN-SIG International Conference, EANN 2011 and 7th IFIP WG 12.5 International Conference, Artificial Intelligence Applications and Innovations, Corfu, Greece, September 15-18, 2011, Proceedings, Part II. IFIP AICT 364, p.483 (2011). 8. Tiss, A.; Timms, J.F.; Smith, C.; Devetyarov, D.; Gentry-Maharaj, A.; Ca- muzeaux, S.; Burford, B.; Nouretdinov, I.; Ford, J.; Luo, Z.; Jacobs, I.; Menon, U.; Gammerman, A.; Cramer, R. Highly accurate detection of ovarian cancer using CA125 but limited improvement with serum MALDI-TOF MS profiling. Internation Journal of Gynecol. Cancer, 2010, 20, pp.1518–1524. 9. Gammerman, A., Vovk, V., Burford, B., Nouretdinov, I., Luo, Z., Chervonenkis, A., Waterfield, M., Cramer, R., Tempst, P., Villanueva, J., Kabir, M., Camuzeaux, S., Timms, J., Menon, U., Jacobs, I. Serum Proteomic Abnormality Predating Screen Detection of Ovarian Cancer. The Computer Journal (2008). Appendix. Dependence on the taxonomy parameter In this work we used K = 5. On Figure 2 we show what happens if this parameter is changed. It can be seen that with the accuracy becomes satisfactory with at least 4-5 taxa. On the other hand when the number K of taxa increases the interval width becomes wider (less informative) without essential improvement of the accuracy. So the choice of K = 5 was reasonable enough. Fig. 2. Prediction accuracy and interval width for HD data for different K. Fig. 2. Prediction accuracy and interval width for HD data for different K.
https://openalex.org/W2902295519
https://www.scielo.br/j/cta/a/QvLb74nCTgqqR46npRZ8fpk/?lang=en&format=pdf
English
null
Utilization of Uncaria gambir Roxb filtrate in the formation of bioactive edible films based on corn starch
Food Science and Technology
2,019
cc-by
4,316
1 Introduction Bioactive edible film is thin layer made from biopolymer constituents such as starch, protein and lipid that has function as food product packaging and contains natural functional compound. Researches on bioactive edible film had continously increasing, especially in term of natural functional compound used for its processing. Maizura et al. (2007) had described that the use of lemongrass at concentration of 0.1 to 0.4% in edible film formula based on modified cassava starch had produced edible film that capable to inhibit the growth of Escherichia coli O157:H7 bacteria. Pranoto et al. (2005) had stated that incorporation of garlic oil with magnitude of 0.4% (v/v) in formula of alginate-based edible film can inhibit bacteria of Staphylococcus aureus and Bacillus cereus. Edible film with formulation of 3% corn starch concentration and 7% Curcuma aeruginosa Roxb squeeze concentration contains total phenol of 555.16 μg/g and antioxidant activity of 52.10% (Kusumawati & Putri, 2013). The addition of 150 mg catechin compound on Gelidium corneum (GC)-based edible film can inhibit microbia growth of Eschericha coli with magnitude of 1.93 log CFU/g and Listeria monocytogenes with magnitude of 1.44 log CFU/g on sausage (Ku et al., 2008). with inhibition power diameter (IPD) of 8.2 mm. The IPD value from this research was still low and can be elevated by increasing gambier powder quantity in edible film formulation. However, gambier powder addition higher than 3% (w/v) had produced negative impact on edible film. Edible film had dark brown color, became thicker and had low elasticity due to this negative impact which in turn it could not fulfill the stated standard. The use of gambier powder is replaced with gambier powder filtrate in order to overcome the above negative impact. Gambier filtrate is produced by dissolving gambier powder in water followed by stirring and mixture separation into liquid (filtrate) and gambier oilcake. Separation process for gambier filtrate is done through filtration and centrifugation. Filtration is a separation method to separate solid zat from liquid by using porous equipment based on particle size differences. Centrifugation is used as substitute to filtration if solid particles are very fine and mixture volume is less. The use of gambier filtrate with high concentration of catechins compound and low solid content in edible film formulation have positive impact in increasing antibacterial and antioxidant characteristics as well as have no negative impact to physical and chemical characteristics of edible film. 1 Introduction According to Taniguchi et al. (2007), gambier leaves extract contains catechins compound in the range of 28 to 54%. Pambayun et al. (2007) had described that catechins compound in gambier leaves extract had antibacterial characteristic especially for Gram-positive bacteria. Santoso et al. (2016) had reported that ganyong starch-based edible film containing 3% (w/v) gambier powder could inhibit the growth of Staphylococcus aureus bacteria ISSN 0101-2061 (Print) ISSN 1678-457X (Online) ISSN 0101-2061 (Print) ISSN 1678-457X (Online) Food Science and Technology DOI: D https://doi.org/10.1590/fst.06318 Abstract The objective of this research was to determine the effect of Uncaria gambir Roxb filtrate addition on functional edible films based on corn starch. This study used a Factorial Randomized Completely Design with two treatments and three replication for each treatment. The first factor was the concentration of Uncaria gambir Roxb filtrate (20%, 30%, 40%) and the second factor was the mixing separation method (filtering, centrifuge 1000 rpm, centrifuge 2000 rpm). The results showed that treatment of Uncaria gambir Roxb filtrate and the mixing separation method had significant effect on thickness, elongation percentage, water vapour transmission rate, antioxidant and antibacterial activities of edible film. The interaction of Uncaria gambir Roxb filtrate and the mixing separation method had significant effect on water vapour transmission rate. The best treatment based on edible film characteristics was 40% concentration of Uncaria gambir Roxb filtrate and the mixed separation method of 1000 rpm centrifugation. Keywords: corn starch; edible film; filtrate; Uncaria gambir Roxby. ii Practical Application: Edible films applied on food products that play an important role on their conservation, distribution and marketing Practical Application: Edible films applied on food products that play an important role on their conservation, distribution and marketing Practical Application: Edible films applied on food products that play an important role on their conservation, distribution and marketing. Utilization of Uncaria gambir Roxb filtrate in the formation of bioactive edible films based on corn starch di SANTOSO1*, Ranti HAZIRAH1, Gatot PRIYANTO1, dan HERMANTO1, dan SUGITO1 Received 26 Mar., 2018 Accepted 03 Sept., 2018 1Study Program of Agricultural Product Technology, Agricultural Technology Department, Faculty of Agriculture, Sriwijaya University: Ogan Ilir, South Sumatera, Indonesia *Corresponding author: budiunsri@yahoo.com Accepted 03 Sept., 2018 1Study Program of Agricultural Product Technology, Agricultural Technology Department, Faculty of Agriculture, Sriwijaya University: Ogan Ilir, South Sumatera, Indonesia *Corresponding author: budiunsri@yahoo.com 2 Material and methods Equipments used in this study were hot plate, magnetic stirrer, vortex, oven, desicator, Haze meter serie NDH-200 of Nipon Denshoku Kogyo Co Ltd, micrometer (Roch) p p , 1Study Program of Agricultural Product Technology, Agricultural Technology Department, Faculty of Agriculture, Sriwijaya University: Ogan Ilir, South Sumatera, Indonesia *Corresponding author: budiunsri@yahoo com nology, Agricultural Technology Department, Faculty of Agriculture, Sriwijaya University: Ogan Ilir, South Sumatera, Indonesia com duct Technology, Agricultural Technology Department, Faculty of Agriculture, Sriwijaya University: Ogan Ilir, South Sumatera, Indonesia @yahoo.com p p 1Study Program of Agricultural Product Technology, Agricultural Technology Department, Faculty of Agriculture, Sriwijaya University: Ogan Ilir, South Sumatera, Indonesia *Corresponding author: budiunsri@yahoo.com Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 837 Uncaria gambir Roxb filtrate in corn starch edible films (A281500504, Sisaku SHO Ltd, Japan), Testing Machine MPY (Type: PA-104-30, Ltd Tokyo, Japan), and water vapor transmission rate tester Bergerlahr. Materials, 1997), elongation percentage of edible film (American Society for Testing and Materials, 1997), compressive strength (American Society for Testing and Materials, 1997), water vapour transmission rate (American Society for Testing and Materials, 1997). Chemical analysis: water content (American Society for Testing and Materials, 1997) and total phenol content (Chaovanalikit & Wrolstad, 2004) and functional analysis: antioxidant (DPPH method) and antibacterial (Hamilton-Miller & Shah, 2000). Data will be processed by using analysis of variance (Anova). Treatments having significant effect will be further analyzed by using Honestly Significant Different (HSD) test at 5% level. Materials used in this research were as follows: clean water, aquadest, ingredients for analysis, gambier (Uncaria gambir Roxb) obtained from Km 5 Market of Palembang, 5) glycerol, 6) HPMC (Hydroxy propyl methyl cellulose), 7) bacterial culture of Stapylococcus aereus, 8) nutrient agar (NA) and 9) corn starch. 3.1 Thickness Average thickness of the produced edible film was in the range of 0.097 mm to 0.143 mm and had fulfilled the standard of JIS 1975 which stated that maximum thickness for edible film is 0.25 mm. The highest thickness value of edible film was found on A3B2 treatment, whereas the lowest one was found on A1B1 treatment. Average value of edible film thickness was shown in Figure 1. 2.2 Edible film processing (Modified from Harmely et al., 2014) Analysis of variance results showed that treatments of gambier filtrate and mixed separation method had significant effect on edible film thickness, whereas treatment interaction had no significant effect on edible film thickness. Results of HSD test for treatments of gambier powder filtrate and mixed separation method were shown in Table 1 and Table, respectively. Corn stach with magnitude of 6% (w/v) is added with 100 mL aquadest and followed by stirring operation. Starch suspension is heated at temperature of about 60 °C while stirring until complete gelatinization is occurred. HPMC with magnitude of 1.6 g is put into 80 mL hot water and mixed into 100 mL starch suspension. Glycerol with magnitude of 3% (v/v) is added while stirring at temperature of about 60 °C followed by gambier filtrate addition with magnitude of 5% (v/v.final) according to treatments and subsequently is stirred until homogenous. Suspension is poured and leveled into the mould followed by drying in oven at temperature of 45 to 50 °C for 24 hours. After drying operation, edible film is put into dessicator for 6 hours. Finally, edible film is released from the mould and it is ready for analysis. Results of HSD test (Table 1) showed that higher addition of gambier powder filtrate results in thicker edible film. This is due to the fact that gambier powder filtrate contain catechins Figure 1. Average thickness of edible film (mm). 2.1 Processing of gambier powder filtrate Dry gambier powder is crushed until fine using mortar and subsequently is sieved using 80 mesh siever. Weighing of fine gambier powder according to treatments (20%, 30% and 40% (w/v)) and then put it into volumetric glass and added with aquadest until 100 mL boundary mark. The suspension is stirred by using magnetic stirrer for 10 minutes and sieving was done according to the method of treatments (sieve paper, centrifugation at 1000 rpm and centrifugation at 2000 rpm) and followed by filtrate collection. 2.3 Experimental design and statistical analysis This research used Factorial Randomized Completely Design with two treatment factors consisting of filtrate with gambier powder concentration (A): A1 = 20%, A2 = 30% and A3 = 40%(w/v) and mixed separation method (B): B1 = sieve paper, B2 = centrifugation at 1000 rpm and B3 = centrifugation at 2000 rpm. The observed parameters were consisted of physical analysis: edible film thickness (American Society for Testing and Figure 1. Average thickness of edible film (mm). Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 838 838/842 Table 1. HSD test for gambier powder filtrate in term of thickness, elongation percentage, water vapour transmission rate, IC50 and inhibition power diameter (IPD) of the produced edible film. Gambier powder filtrate (%) Thickness (mm) Elongation percentage (%) Water vapour transmission rate (g.m-2.day-1) IC50 (ppm) IPD (mm) A1 0.110+0.015a 14.78+1.262a 20.99+0.705a 413.25+58.324a 7.08+1.171a A2 0.119+0.020b 15.11+1.262ab 21.31+0.684b 368.31+57.475b 7.25+1.333ab A3 0.124+0.018b 15.67+1.453a 21.62+0.691c 341.38+72.096c 7.33+1.526b Remarks: Numbers followed by the same letters at the same columns are not significantly different (α=0.05). Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 838 Santoso et al. starch used in this research contains high amylose component. The highest elongation percentage for edible film was found on A3B2 treatment, whereas the lowest one was found on A1B1 treatment. Average value of elongation percentage for edible film was shown in Figure 2. which is not easily soluble in cold water, but it is only soluble in hot water and develop crystal at dry condition. Crystalic form of catechins compound at dry condition is the dissolved solids in edible film. Higher gambier powder filtrate will cause higher numbers of crystal within film matrix which in turn produce higher concentration of dissolved solids in edible film resulting in increase of edible film thickness. This finding is in accordance to the research that was reported by Santoso (2011) which showed that higher usage of gambier powder results in higher thickness of edible film. Analysis of variance results showed that treatments of gambier filtrate and mixed separation method had significant effect on elongation percentage of edible film, whereas interaction of both treatments had no significant effect on elongation percentage of edible film. Results of HSD test for treatments of gambier powder filtrate and mixed separation method toward elongation percentage of edible film were shown in Table 1 and Table, respectively. 3.2 Elongation percentage HSD test results (Table 2) showed that mixed separation method by using filter paper had lower elongation percentage than that of mixed separation method by using centrifugation. This is related to catechins compound available in gambier filtrate which have semi polar characteristics. Therefore, gambier powder filtrate obtained from mixed separation method by using filtr paper only has dissolved catchins compound, whereas gambier powder filtrate obtained from mixed separation method by using centrifugation method has both dissolved and undissolved catechins compound resulting in higher catechins compound concentration. Average value of elongation percentage for edible film was in the range of 13.33% to 16.67%. Elongation percentage for edible film produced in this research was not fulfill the JIS 1975 standard which requires minimum of 70% elongation percentage for edible film. This is due to the fact that corn Figure 2. Average value of elongation percentage of edible film. 2.3 Experimental design and statistical analysis Table 2 showed that edible film thickness with mixed separation method by using sieve paper was lower than mixed separation method by using centrifugation at 1000 rpm or 2000 rpm. This is due to the fact that separation using filter paper in principle separate liquid and insoluble solids based on the different of particles size of mixture substance. Thus, only dissolved solids which are available in gambier powder filtrate. Centrifugation separation method is only based on the separated fine particles and not separates dissolved or undissolved solids so that filtrate obtained from centrifugation separation contains dissolved or undissolved solids. Total solid content in filtrate affects the increase of edible film thickness. Park et al. (2004) had described that edible film thickness is affected by total solid available within suspension, solution volume and mould area used for edible film processing. Elongation percentage of edible film was increased in accordance to concentration increase of gambier powder filtrate (Table 1) because gambier powder filtrate contains catechins compound which have active group of hydroxyl (OH). The higher the concentration of gambier powder filtrate, the higher the numbers of OH group which are trapped within edible film matrix. The trapped OH group can results in cohesive power decrease of edible film matrix which in turn affect the elasticity of edible film. According to Rodriguez et al. (2006), glycerol which contain high numbers of hydroxyl (OH) gugus in matrix system of edible film can increase water vapour transmission rate and elongation percentage of edible film. 3.3 Water vapour transmission rate HSD test for the effect of gambier powder filtrate concentration and mixed separation method on water vapour transmission rate (g.m-2.day-1) of edible film. Treatment Average HSD 5% = 0.30 A1B1 20.23+0.131 a A2B1 20.57+0.064 B A3B1 20.88+0.063 c A1B3 21.15+0.094 c A2B3 21.4675+0.036 d A3B3 21.6142+0.182 de A1B2 21.7610+0.062 de A2B2 21.9077+0.158 e A3B2 22.2431+0.036 f Remarks: Numbers followed by the same letters at the same columns are not significantly different (α=0.05). HSD test results (Table 2) showed that water vapour transmission rate of edible film using treatments of mixed separation method with 1000 rpm centrifugation and 2000 rpm centrifugation was significantly different than that of mixed separation method using filter paper. It is estimated that differences in mixed separation method cause differences in catechins content available in edible film. The higher numbers of catechins in edible film will results in higher numbers of hydroxyl (OH) group at edible amongst polymers, increase of free space amongst polymers and increase of polymers mobility. These effects in turn cause the decrease of matrix structural integrity of film and have impact on the increase of water vapour transmission rate of edible film. Remarks: Numbers followed by the same letters at the same columns are not significantly different (α=0.05). Figure 4. Average value of IC50 of edible film. Table 3 (HSD) showed that treatments interaction of 20% gambier filtrate cocentration and mixed separation method using filter paper had produced edible film having the lowest average value of water vapour transmission rate. It is due to the fact that A1B1 treatment had the lowest catechins content than other treatments. Low concentration of catechins content within edible film matrix has effect on the decrease of water vapour transmission rate because catechin compound contributes OH group which affect hydrophilic increase of edible film. According to Nemet et al. (2010), glycerol with its hydroxyl gugus affects the increase of water vapour transmission rate. Figure 4. Average value of IC50 of edible film. 3.3 Water vapour transmission rate Average value of water vapour transmission rate for edible film was in the range of 20.2306 to 22.2432 g.m-2.day-1. The highest value of water vapour transmission rate for edible film was found on A3B2 treatment, whereas the lowest one was found on A1B1 treatment. Average value of water vapour transmission rate for edible film was shown in Figure 3. Figure 2. Average value of elongation percentage of edible film. Table 2. HSD test for mixed separation method in term of thickness, elongation percentage, water vapour transmission rate, total phenol, IC50 and inhibition power diameter (IPD) of produced edible film. Mixed separation method Thickness (mm) Elongation percentage (%) Water vapour transmission rate (g.m-2.day-1) IC50 (ppm) IPD (mm) B1 0.101+0.005a 13.67+0.333a 20.85+0.805a 419.16+44.511a 6.83+0.192a B2 0.137+0.001c 16.11+0.509b 21.92+0.315c 304.35+47.426c 7.56+0.509c B3 0.116+0.001b 15.78+0.509b 21.17+0.294b 399.43+19.961b 7.28+0.509 b Remarks: Numbers followed by the same letters at the same columns are not significantly different (α=0.05). ion method in term of thickness, elongation percentage, water vapour transmission rate, total phenol, IC50 ) of produced edible film. Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 2 839 Uncaria gambir Roxb filtrate in corn starch edible films Figure 3. Average value of water vapour transmission rate of edible film. Analysis of variance results showed that treatments of gambier powder filtrate concentration and mixed separation method as well as their interaction had significant effect on water vapour transmission rate of edible film. Results of HSD test for treatments of gambier powder filtrate concentration and mixed separation method toward water vapour transmission rate of edible film were shown in Table 1 and Table 2, respectively. HSD test results (Table 1) showed that higher concentration of gambier powder filtrate results in higher water vapour transmission rate of edible film. This is related to numbers of catechins compound, that is higher concentation of gambier filtrate results in higher numbers of catechins compound within film matrix. Catechins compound containing high numbers of hydroxyl (OH) group causes edible film to have hydrophylic property which in turn affect the increase of water vapour transmission rate of edible film. Sobral et al. (2001) had described that glycerol content in edible film formulation has significant effect on the increase of water vapour transmission rate. Figure 3. Average value of water vapour transmission rate of edible film. Table 3. Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 3.5 Antibacterial activity Test of antibacterial activity was done by using disc method with Staphylococcus aureus as test bacteria. Average value of clear zone or inhibition power diameter (IPD) produced was in the range of 6.67 to 7.67 mm. Average value of inhibition power diameter (IPD) was shown in Figure 5. Kusumawati, D. H., & Putri, W. H. R. (2013). Physical and chemical characterisics of corn starch edible film incorporated with black Curcuma aeruginosa squeeze. Journal of Food and Agroindustry, 1(1), 90-100. Ku K.-J., Hong Y.-H., & Song, K. B. (2008). Mechanical properties of a Gelidium corneum edible film containing catechin and its application in sausages. Journal Food Science, 73(3), 217-222. Analysis of variance results showed that treatments of gambier powder filtrate concentration and mixed separation method had significant effect on inhibition power diameter (IPD) of edible film, whereas interaction of both treatments had no significant effect on inhibition power diameter (IPD) of edible film. Results of HSD test for treatments of gambier powder filtrate concentration and mixed separation method toward inhibition power diameter (IPD) of edible film were shown in Table 1 and 2, respectively. Lucida, H., Bakhtiar, A., & Putri, W. A. (2007). Formulation preparation of mouth antiseptic from cathecin of gambier. Journal of Pharmaceutical Science and Technology, 12(1), 1-7. Maizura, M., Fazilah, A., Norziah, M. H., & Karim, A. A. (2007). Antibacterial activity and mechanical properties of partially hydrolyzed sago starch–alginate edible film containing lemongrass oil. Journal of Food Science, 72(6), 324-330. http://dx.doi.org/10.1111/j.1750- 3841.2007.00427.x. PMid:17995673. HSD test (Table 1) showed that higher concentration of gambier powder filtrate results in higher antibacterial activity of edible film. This is due to the fact that gambier filtrate contains catechins compound which pose antibacterial characteristic. According to Pambayun et al. (2012), catechins compound within commercial gambier product can inhibit Gram-positive bacteria through inhibition property as bactericidal. Molyneux, P. (2004). The use of the stable free radical diphenylpicryl- hydrazyl (DPPH) for estimating antioxidant activity. Science Technology, 26(2), 211-219. Nemet, N. T., Soso, V. M., & Lazic, V. L. (2010). Effect of glycerol content and pH value of film-forming solution on the functional properties of protein-based edible films. Acta Periodica Technologica, 41(41), 57-67. http://dx.doi.org/10.2298/APT1041057N. Table 2 showed that treatment of mixed separation method at 1000 rpm centrifugation had higher antibacterial characteristic than that of other treatments. This is in accordance to antioxidant Pambayun, R., Gardjito, M., Sudarmadji, S., & Kuswanto, K. R. (2012). 3.4 Antioxidanth both treatments had no significant effect on IC50 of edible film. Results of HSD test for treatments of gambier powder filtrate concentration and mixed separation method toward IC50 value of edible film were shown in Table 1 and 2, respectively. The highest IC50 value of edible film was found on A1B1 treatment with magnitude of 258.14 ppm, whereas the lowest one was found on A3B2 treatment with magnitude of 469.32 ppm. The IC50 value is inversely proportional to damping potential of free radical. Low IC50 value found in this research cause higher potential of antioxidant activity, that is edible film concentration required to produce damping activity of free radicals with magnitude of 50% become smaller (Molyneux, 2004). Average value of IC50 was shown in Figure 4. HSD test (Table 1) results showed that higher concentration of gambier powder filtrate in edible film matrix results in lower IC50 value, that is higher concentration of gambier powder filtrate results in higher antioxidant characteristic of edible film. This is due to the fact that catechins compound in gambier powder filtrate has antioxidant characteristic. Gambier extract contains catechins as the main component and catechins is a polyphenol compound that has potential as antioxidant and antibacteria Analysis of variance results showed that treatments of gambier powder filtrate concentration and mixed separation method had significant effect on IC50 of edible film, whereas interaction of Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 840 Santoso et al. Figure 5. Average value of inhibition power diameter of edible film. characteristic of the produced edible film. The higher the antioxidant, the higher the antibacterial characteristic because catechins compound has antibacterial and antioxidant characteristics. 4 Conclusion Corn starch-based edible film added with gambier filtrate has antibacterial and antioxidant characteristics. The best treatment based on physical, chemical and functional characteristics was 40% concentration of Uncaria gambir Roxb filtrate and the mixed separation method of 1000 rpm centrifugation. Figure 5. Average value of inhibition power diameter of edible film. References American Society for Testing and Materials – ASTM. (1997). Annual book of ASTM standards. Philadelphia: ASTM. Chaovanalikit, A., & Wrolstad, R. E. (2004). Total anthocyanins and total phenolics of fresh and processed cheries and their antioxidant properties. JFS. Food Chemistry and Technology, 69(1), 67-72. Hamilton-Miller, J. M. T., & Shah, S. (2000). Activity of the tea component epicatechin gallate and analogue against methicillin-resistant Staphylococcus aureus. The Journal of Antimicrobial Chemotherapy, 46(5), 847-863. http://dx.doi.org/10.1093/jac/46.5.852. Harmely, F., Deviarny, C., & Yenni, W. S. (2014). Formulation and evaluation of edible film prepared from basil leaf (ocimum americanum l.) As Mouth fresh. Journal of Pharmaceutical and Clinical Science, 1(1), 38-47. Acknowledgements This research is part of the results of competitive excellent research which is funded by DIPA of Sriwijaya University Service Public Agency Budget of 2017 No.042.01.2.400953/ 2017 December 5, 2017 with Contract No.988/UN9.3.1/PP/2017 July 21, 2017. (Lucida et al., 2007). Pambayun (2007) had stated that gambeir powder extracted with water solvent at 30 °C produces 45% yield of extracts in phenolic contents and antibacterial properties. Table 2 showed that mixed separation method by using 1000 rpm centrifugation had higher antioxidant activity than that of other treatments. This is related to numbers of catechins compound available in gambier powder filtrate. Separation method by using 1000 rpm centrifugation had higher catechin compound concentration within filtrate because soluble and insoluble components are followed by filtrate, whereas only soluble solid which enter into filtrate in case of separation method by using filter paper and centrifugation at 2000 rpm. These physical differences can be seen from filtrate color, that is filtrate produced from separation method by using 1000 rpm centrifugation was clearer than that of two other treatments. Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 3.5 Antibacterial activity Sensitivity of Gram positive bacteria toward cathecin extracted from gambier (Uncaria gambir Roxb). Agritech, 28(4), 174-179. 841 Food Sci. Technol, Campinas, 39(4): 837-842, Oct.-Dec. 2019 Uncaria gambir Roxb filtrate in corn starch edible films Uncaria gambir Roxb filtrate in corn starch edible films Santoso, B., Marsega, A., Priyanto, G., & Pambayun, R. (2016). Improvement of physical, chemical and antibacterial properties of ganyong starch based edible film. Agritech, 36(4), 379-386. Pambayun, R., Gardjito, M., Sudarmadji, S., & Kuswanto, K. R. (2007). Phenol content and antibacterial properties from several types of gambier extract product (Uncaria gambir Roxb). Indonesia Pharmaceutical Magazine, 18(3), 141-146. Santoso, B. (2011). Integration of Modified Starch, Surfactans, Protein and Gambir Plant Extract in Edible Film Processing (Doctoral dissertation). Program in Agricultural Science, Postgraduate Program, Sriwijaya University, Palembang. Park, D. P., Sung, J. H., Choi, H. J., & Jhon, M. S. (2004). Electroresponsive characteristics of highly substituted phosphate starch. Journal of Materials Science, 39(19), 6083-6086. http://dx.doi.org/10.1023/ B:JMSC.0000041706.76102.b4. Sobral, P. J. A., Menegalli, F. C., Hubinger, M. D., & Roques, M. A. (2001). Mechanical, water vapor barrier and thermal properties of gelatin based edible films. Food Hydrocolloids, 15(4-6), 423-432. http://dx.doi.org/10.1016/S0268-005X(01)00061-3. Pranoto, Y., Salokhe, V. M., & Rakshit, S. K. (2005). Physical and antibacterial properties of alginate-based edible film incorporated with garlic oil. Food Research International, 38(3), 267-272. http:// dx.doi.org/10.1016/j.foodres.2004.04.009. Taniguchi, S., Kuroda, K., Doi, K., Inada, K., Yoshikado, N., Yoneda, Y., Tanabe, M., Shibata, T., Yoshida, T., Hatano, T., & Zasshi, Y. (2007). Evaluation of gambir quality based on quantitative analysis of polyphenolic constituents. Journal of the Pharmaceutical Society of Japan, 127(8), 1291-1300. http://dx.doi.org/10.1248/yakushi.127.1291. PMid:17666883. Rodriguez, M., Oses, J., Ziani, K., & Mate, J. I. (2006). Combined effect of plasticizers and surfactants on the physical properties of starch based edible films. Food Research International, 39(8), 840-846. http://dx.doi.org/10.1016/j.foodres.2006.04.002. 842
https://openalex.org/W4291016126
https://zenodo.org/record/891938/files/138_IJRG14_CC11_127.pdf
Zulu
null
र ंगा े का मना ेवैज्ञानिक प्रभाव एवम र ंग चिकित्सा
Zenodo (CERN European Organization for Nuclear Research)
2,017
cc-by
2,336
ISSN- 2350-0530 ISSN- 2350-0530 Art INTERNATIONAL JOURNAL of RESEARCH –GRANTHAALAYAH Art A knowledge Repository jaxks dk euksoSKkfud izHkko ,oe jax fpfdRlk MkW- izfrHkk lqxa/kh MkW-jktJh uhek ih ,e ch xqtjkrh vkVZl dkyst bUnkSj euq"; bZ'oj dh loZ Js"B d`fr gS vkSj jax l`f"V dk csgrjhu migkj gS A ;g izd`fr dh ,slh HksaV gS ftlls thou vkSj vf/kd vkdZ"kd curk gS A izd`fr ds lkfau/; ls euq"; vius thou dks fu/kkZfjr djrk gS vkSj bl lkektUL; dks LFkkfir djrs gq, LoLFk thou thus dk iz;kl djrk gSA tSls lwjt ds rki o izdk'k dk gekjs thou ij izHkko iM+rk gS oSlk gh jaxks dk izHkko gekjs 'kjhj o eu ij iM+rk gSA jaxks dk egRo dsoy lkSUn;Z o`f) rd lhfer ugha gS vfirq euq"; ds LoHkko mlds Hkfo"; ,oe mlds LokLF; ij Hkh bldk lh/kk izHkko iM+rk gS A izd`fr vius xHkZ esa vla[; jax lek, gq, gS AdHkh cknyks ds cnyrs gq, jax rks dHkh /kjrh ds cnyrs gq, vla[; :iA dHkh ouLifr;ksa ds cnyrs gq, jax rks dHkh Qwyksa ds jaxksa dh fofHkUu NVk,saA dHkh lw;Z dh fdj.kksa ls cnyrs ikuh ds :i rks dgha vkdk'k esa fNrjk bUnz /kuq"k A frrfy;ksa ds ia[kksa ij jaxks dh Nki rks dgh xxu esa mM+rs i[ks:vksa ds fprdcjs ia[k A vkdk'k] i`Foh ty] Fky] uHk izd`fr lEiw.kZ jaxe; gS vkSj bl jaxe; izd`fr dk lh/kk lEcU/k gekjs ru eu vkSj LokLF; ij iM+rk gSA izd`fr esa mifLFkr jax gh gekjs 'kjhj dh izd`fr dk fu/kkjZ.k djrs gSA izd`fr dh lqUnj NVk ds lkFk dHkh eu e;wj ukp mBrk gS rks dHkh cknyksa dh ‘;key NVk ds lkFk d`”.ke; gks tkrk gSA lkxj dh mRrky ygjksa ds lkFk eu epy mBrk gS rks bUnz /kqu”k ds jaxks ds lkFk eu dk fgj.k nksM+ tkrk gS /kwy /kwlfjr fojku izd`fr ds lkFk eu va/ksjh xqQk esa lks tkrk gS rks dHkh vk’kadk ds /kus dkys cknyksa ds ccMaj ls lge tkrk gSA euq”; dh Hkkoukvksa dks O;Dr djus ds fy, jaxksa dk mi;ksx fp=dkj lfn;ksa ls djrs vk jgs gSA blh izdkj lkfgR; esa jaxksa dks euq"; dh Hkkouk dk izrhd ekurs gq, vusd eqgkojsa Hkh x<+s x;s gS tSls rch;r gjh gksuk] psgjk yky gksuk cklarh eu] jaxhu ehtkt+ bR;knh A jaxks dks ,d fuf'pr ifjHkk"kk esa rks ifjHkkf"kr ugha fd;k tk ldrk fdUrq jax dh mRifRr dk foospu fd;k tk, gS rks loZizFke Hkkjrh; /keZ ds vk/kkj jked`".k dk lkaojk jax gh gekjs lkeus vkrk gS A foKku bl rF; dks izekf.kr dj pqdk gS fd lHkh jaxks dh ewy mRifRr dkys jax ls gqbZ gS A izd`fr esa ge vla[; jax ns[krs gS oSls uohu [kkst ds vuqlkj ;g vla[; jax eq[; rhu jaxksa ij gh vk/kkfjr gS A ihyk] uhyk vkSj yky A bu jaxksa dh rjax o yEckbZ vyx vyx gksrh gS A vkSj ;g euq"; ds ekul iVy ij rjg rjg ds fopkj mRiUu djrs gSA fopkjksa vkSj jaxksa esa fof’k"V lEcU/k gSA jaxksa dk ekufld 'kkjhfjd LokLF; ij vlk/kkj.k izHkko iM+rk gSA lw;Z dh fdj.kksa esa lkr jax fuf'pr dze esa gksrs gS izR;sd jax ds vius fo'ks"k LokLF;o/kZd xq.k gksrs gSA lw;Z dh fdj.kksa ds lkr jax cSaxuh] uhyk] vklekuh] gjk] ihyk] ukjaxh vkSj ykyA cSaxuh] uhyk vkSj vklekuh xehZ dks fu;af=r djus esa lgk;d] gjk jax xehZ vkSj lnhZ ds izHkko dks larqfyr j[kus esa lgk;d vkSj ihyk ukjaxh o yky 'kjhj esa xehZ cuk, j[kus esa lgk;d gSA oSKkfudksa us vius iz;ksxksa ls bl fl)kUr dks izfrikfnr fd;k gS fd gekjs 'kjhj esa Hkh ,d lkSje.My gSA ftlesa lkr jaxksa dk leU;o; gSA 'kjhj esa fLFkfr bl lkr jaxksa dks BIGVYOR(B- blue, I =Indigo, G= Green, V= Violate, Y= Yellow, O =Orange, R =Red vk/kqfud HkkSfrdh ds vuqlkj jaxksa dk vk;kstu bysDVªkWu vkSj fofdj.kska dh rjaxksa dh U;wukf/kd yEckbZ ds dkj.k gksrk gSA gj jaxksa dh rjax yEckbZ fHkUu fHkUu gksrh gSA ;g rjaxsa gekjh vka[kksa ds ek/;e ls gekjs efLr"d esa lans'k Hkstrh gS vkSj gekjk 'kjhj mls igpkurk gS A izR;sd ijek.kq ukHkh ds cgkj pDdj yxkrk gSA bysDVªku ijkcSaxuh fdj.kks dks vius esa yhu djrs gq, mRrsftr gksdj bl voLFkk esa ,d fuf'pr rjax yEckbZ ds fdVksu mRlftZr djrk gS A bl mRlftZr fofdj.k dh yEckbZ gh lkr jaxksa esa ls ,d jax dk fu/kkZj.k djrh gS A bu fofHkUu rjax dh yEckbZ;ksa ls mRlftZr fdVksu d.kksa ls gh eq[; lkr jaxksa dh fdj.k dk l`tu gksrk gSA ikSjk.khd 'kL=ksa esa Hkh nsoh nsorkvksa ds Lo:iksa ,oe muds oL=kvyadkjksa ds jaxks ds fo"k; esa cMk 'kL=h; foospu fd;k x;k gS A ,d vkSj ;s fofHkUu jax fpRrkd"kZd] Nch izLrqr djrs gS ogha nwljh vkSj izR;sd nsork esa fufgr xq.kksa ,oe [1-3] Http://www.granthaalayah.com©International Journal of Research -GRANTHAALAYAH ISSN- 2350-0530 Art INTERNATIONAL JOURNAL of RESEARCH –GRANTHAALAYAH Art A knowledge Repository fof'k"V deksZ vFkok mPp mnns';ksa dks Hkh vfHkO;Dr djrs gSA izfl) fo}ku Jh jkepUnzth egsUnz us bl fo"k; esa O;kid 'kks/k fd;k gS tSls & guqekuth 'kfDr cy vkSj rst ds nsork gS vkSj yky o ds'kfj;k jax Hkh mRlkg ,oe LiwfrZ mRiUu djus okyk gS A bl dkj.k guqekuth lnSo ds'kfj;k ;k yky jax esa fn[kkbZ nsrs gSA yky jax euq"; ds 'kjhj dks LoLFk vkSj gf"kZr djus okyk gS vkSj iq:"kkFkZ vkSj vkRe xkSjo dks izdV djrk gS A bl dkj.k izk;% lHkh nsorvksa dh izfrHkk esa yky jksyh dk Vhdk yxk;k tkrk gSA y{ehth dks Hkh eaxydkjh yky oL= iguk, tkrs gS A mUgsa yky dey ij gh izfrf"Br fd;k tkrk gS A y{eh dks fgj.ke; inekgLrk ;k jDro.kkZ dgk x;k gSA D;ksafd yky jax lkSHkkX;] laifRr] ,s'o;Z vkSj le`f) dk izrhd gSA thou ds izR;sd lksaiku esa jaxks dk egRo gS A 'kknh dk tksMk yky jax dk gksrk gS D;ksafd yky jax dkeokluk dk 'keu djrk gSA uo o/kw dh ekax esa flUnwj] yky] pwMh] egkoj] fcUnh ;s lHkh okluk dk 'keu djrs gS iwtk esa Hkh yky Vhdk iq:"kaks ds flj ij yxk;k tkrk gS A gkFkkas esa yky ukMk ck/akk tkrk gS D;ksafd yky Vhdk 'kkSa;Z ,oe fot; dk izrhd gSA g"kZ ds volj ij fookg] tUe] vU; lkaLd`frd mRolksa ij Hkh ge vkuan dh vfHkO;fDr yky & ds'kfj;k jax ls gh djrs gSA yky jax dh rjg Hkxok jax dks Hkh cMk egRo fn;k x;k gSA Hkxok jax lHkh efyurk dks nwj djus okyk vkyL; dyq"krk dks nwj dj ifo= Hkko dk lapkj djus okyk gSA Hkxok jax vfXu dh Tokyk dk jax gS A ;g R;kx leiZ.k riL;k vkSj oSjkX; dk izrhd gSA izd`fr dk nwljk egRoiw.kZ jax ihyk gS A ihyk jax 'kkafr oSjkX; dk lwpd gSA ihyk jax Kku xjhek dk jax gSA ;g ;ksX;rk deZfu"Brk vkSj vkRe fo'okl ,dkxzrk ,oe ckSf)drk dh vkSj vxzsf"kr djrk gS A Hkxoku fo".kq Jhd`".k] x.ks'kth dks ihys jaxksa ds oL=ksa ls gh lqlfTtr fd;k tkrk gSA dgk tkrk gS fd tks O;fDr ihys jax dks viuk ysrk gS Kku ekxZ dks izoZr gksdj vius fo/uksa dks ikj dj eu esa xgjh 'kkarh izkIr djrk gSA gjk jax izd`fr dk jax gS tks O;fDr gjs jax dh oLrqvksa dks /kkj.k djrk gS og izd`fr ds lkFk rknkRE; LFkkfir dj lq[kh fujksxh izlUu vkSj 'kkar jgrk gSA izd`fr esa tgkW gjs jax dh l`f"V O;kid :i ls vofLFkr gS ogha vuar uhys jax dk vkdk'k leqnz vkSj ufn;ksa dk jax uhyk gSA uhyk jax xaHkhjrk ds lkFk lkFk cy] ikS:"k deZ'khrk /kS;Z] lkgl 'kkS;Z vkSj viuh vkLFkkvksa ds fy, fuajrj la?k"kZ djus dh izsj.kk nsus okyk vkSj ladYIk 'kfDr dk ifjpkyd gSA blh dkj.k e;kZnk iq:"kksRre jke vkSj Hkxoku Jhd`".k dk o.kZ uhyk crk;k x;k gS Hkxoku f'ko dks Hkh uhydaB dgk x;k gSA oLrqr% uhyk jax m/kzeh iq:"kksa dk jax gSA bl jax dks /kkj.k djus okyk bfUnz;ks dks o'k esa j[k dj iq:"kkFkZ esa yhu jgrk gSA lQsn jax LoPNrk ,oe ifo=rk Kku ,oa foosd dk izrhd gSA fo|k ds fy, ljLorh dks 'osroL=k ekuk x;k gSA 'osr dey ij fojkteku gSA izR;sd jax dh vius ,d m"ek gksrh gS vkSj ogh m"ek gekjs Kku rUrqvksa ij lh/kk izHkko Mkyrh gSA jaxks dk euksoSKkfud izHkko gh ugha jlk;uhd izHkko Hkh gksrk gSA tks vPNk ;k cqjk nksuks izdkj dk gksrk gS A tc vkneh >wB cksyrk gS rks /;ku ls ns[kus ij irk pykrk gS mlds psgjs dk jax dkyk iM x;k gSA lR;fu"B O;fDr ds psgjs ij vkHkk >ydrh gS A ys';k fl)kUr dk ,d fu;e gS tSlk Hkko gksrk gS oSlk gh jax cu tkrk gS vkSj tSlk jax gksrk gS oSlk gh Hkko cu tkrk gS A gekjs O;fDrRo fopkjksa ds lkFk jaxksa dk xgjk lEcU/k gS ge fdl jax ls iqrs gq, dejksa esa jgrs gS fdl izdkj ds jax dk Hkkstu djrs gS bu lc dk gekjs fopkjksa ij izHkko iMrk gS U;k;ky; esa U;k;k/kh'k dkys jax dk dksV igurs gS odhy Hkh dkys jax dk dksV igurs gS A lnhZ esa dkys uhys jax ds dEcy vks<rs gS bu lc dk dkj.k gS fd dkys jax esa izfrjks/kkRed 'kfDr gksrh gS A og ckgj dh OkLrqvksa dks vkRelkr ugha djrk gS ckgj gh jksd nsrk gSA lQsn jax esa ikjnf'kZrk gksrh gS A jaxks dk euksosKkfud izHkko ;g gS fd jax dzks/k {kerk dk Hkh fu/kkZj.k djrk gS A fofHkUu psruk dsUnzks ij /;ku djus ls dzks/k de gks tkrk gS A gekjs 'kjhj esa ftruh Hkh xzafFk;kW gS mu lcdk Hkh viuk jax gS A 'kjhj ds izR;sd oy; dk viuk jax gS A rFkk 'kjhj dks lkjs jax izHkkfor djrs gSA uhys jax ds iz;ksx ls euq"; ds 'kjhj esa 'kkafr dk vuqHko gksrk gSA gjs jax esa jksx feVkus dh {kerk gksrh gS A gjk jax BaMk gksrk gSA gekjs 'kjhj ds chekfj;ksa ,oe Hkkoksa ds lkFk jaxks dk xgjk lEcU/k gSA jksxks dh fpfdRlk esa dkSu lk jax D;k izHkko djrk gS A bldh Hkh tkudkjh Hkh fpfdRld dks gksuk pkfg, A Hkkjrh; vk;qosZfnd fpfdRlk blh ij vk/kkfjr gSA vk;qosZn us lw{e :i ls fofHkUu jksxksa ds 'keu esa iz;qDr gksus okyh fofHkUu jaxksa okyh tM+h cqfV;ksa ,oa vkS"kf/k;ksa dk fuekZ.k fd;k tkrk gS A _rq fof'k"V] fof'k"V] le; ,oe fof'k"V] nsorkvksa dk vogku djds fof'k"V] vkHkk dk izHkqRo jgrk gS rc vkS"kf/k;ksa dk fuekZ.k djrs gSA fofHkUu jax okys jRuksa ds iz;ksxks ds }kjk jksx fuokj.k vk;qosZfnd foKku dk ,d egRoiw.kZ v/;k; gSA ghjk] eksrh] lksuk pkanh izoky] iUuk ekf.kd iq[kjkt eqaxk vkfn dh HkLe ds }kjk {k; e/kqe; Hkaxnj tyksnj muekn coklhj ân; jksx xfB;k vkfn vlk/; jksxksa dk bykt fd;k tkrk gSA D;ksafd jRuksa dk izHkko mudh fof'k"V] jax iznr vkHkk fdj.kksa ds dkj.k gksrk gSA vk;qosZn dk f=nks"k fl)kUr ¼okn fir] dQ½ mipkj dk vk/kkj gS D;ksafd f=nks"k dh fLFkfr esa 'kjhj dh izd`fr esa ifjorZu vk tkrk gS vkSj chekjh iSnk gks tkrh gS ,slh fLFkfr esa 'kkL=h ;k oSn~; ¼jRuks [1-3] Http://www.granthaalayah.com©International Journal of Research -GRANTHAALAYAH ISSN- 2350-0530 Art INTERNATIONAL JOURNAL of RESEARCH –GRANTHAALAYAH Art A knowledge Repository tM+hcqfV;ksa HkLe vkfn dh vkHkk fdj.kksa ds lkFk rky esy cSBkdj ,slh vkS"k/kh dk lsou djokrs gS tks f=nks"kks esa leUo; LFkkfir dj lds A tM+hcqfV;ksa HkLe vkfn dh vkHkk fdj.kksa ds lkFk rky esy cSBkdj ,slh vkS"k/kh dk lsou djokrs gS tks f=nks"kks esa leUo; LFkkfir dj lds A vk/kqfud oSKkfud vkSj euks'kkL=h bl ckr ls lger gS fd jaxks dh fdj.kksa O;fDr fo'ks"k dh varjeu dh n'kk o mlds O;fdRo dh vfHk:fp ds vuq:i izHkko Mkyrh gS A euksoSKkfud euksfo'ys"k.k fpfdRld MkW esDlewyj us fofHkUu iz;ksxksa ds ek/;e ls ;g izfrikfnr fd;k gS fd O;fDr fof'k"V] jaxks tkequh uhyk gjk yky ihyk dRFkbZ Hkwjk dkyk bu jaxksa ls dksbZ fof'k"V] jaxksa dk p;u djrk gS mlds bl pquko ds vk/kkj ij gh O;fDr fo'ks"k ds fof'k"V] xq.k mldh vfHk:fp o pkfjf=d vfHkO;taukvksa dks Kkr dj muds euksfodkjksa dk lw{e fo'ys"k.k fd;k tk ldrk gSA jax fpfdRlk 'kjhj ds jksx feVkus esa ftruh izHkko'kkyh gS mruh gh ekufld vkSj HkkoukRed jksxksa ds vkjke igwWpkus esa ykHkdkjh gSA euq"; dh izd`fr esa ifjorZu dj nsus dh 'kfDr jax fpfdRlk esa gSA izd`fr dk fo/kku ,slk gS fd tgkW jks'kuh gS ogkW 'kfDr vkSj jax gS jaxksa esa m".krk 'khryrk vkSj Hkkj gSA vesfjdk ds izks- Mxyl thu us fQjsfy;u QksVksxzkQh ij fujUrj chl o"kksZ rd dk;Z djds ukscy iq:Ldkj izkIr fd;k mUgksus ck;ksyDVjksxzkQ uked ,slk ;a= fodflr fd;k ftlds }kjk maxfy;ksa] gFksfy;ksa vkSj 'kjhj ds vU; vaxksa ls mnHkkf"kr fdj.kksa dk fp= fy;k tk lds muds }kjk Lohd`r pkZV ds ek/;e ls dksbZ Hkh tkudkj fpfdRld dqN gh nsj esa jksxksa dh igpku dj ldrk gSA teZu i)fr esa fofHkUu jaxksa okyh cksryks esa ikuh Hkj dj /kwi esa j[k dj ml ikuh ls fofHkUu jksxksa ls xzLr jksxks dh lQy fpfdRlk dh tkrh gSA fpfdRld O;fDr fof'k"V] ls dksbZ jax pquus ds fy, dgrs gS vkSj ml O;fDr dh ml fo'ks"k jax ds izfr laosnu'khyrk irk pyrk gSA og vius fiz; jax dks funsZ'k djrk gS vkSj fQj jaxks ds pkVZ ls ml fof'k"V] jax ds xq.kksa& voxq.kksa dks Kkr fd;k tkrk gS rRi'pkr ml fof'k"V] jax dh cksry esa ikuh Hkj dj /kwi esa j[kk tkrk gS fuajrj lkr fnu /kwi esa j[kus ij ml fof'k"V] cksry esa ml fof'k"V] jax dh dkfLed bysDVks esSfXusfVd rjaxs izokfgr gks tkrh gS A ml ty dk lsou djus ls ml O;fDr fo'ks"k esa mldh vfHk:fp okys jax dk vlarqyu leHko esa vk tkrk gS vkSj lkr fnuksa esa mlds jksx dk 'keu gks tkrk gSA jax i)fr esa mipkj ds nkSjku jksxh dks izfrdwy jaxksa ls cpus dh lykg nh tkrh gSA jksxh dh fpfdRlk esa dkSu lk jax D;k izHkko Mkyrk gS blfy, lQy fpfDRld dks bu jaxksa ds xq.k ,oa izHkko dh iw.kZ tkudkjh vko';d gSA orZeku le; esa O;fDr cgqr vf/kd 'kkjhfjd o ekufld tfVyrkvksa esa f/kjk gqvk gS ,slh fLFkfr esa jax mipkj ds }kjk Hkh og vius thou esa lkektU; LFkkfir dj ldrk gS A viuk ifjos'k viuk QfuZpj vius dejs dk dyj] diM+s dk dyj dk;kZy; dh Vscy] ckFk:e dh VkbZYl] ijns xkM+h LdwVj fcLrj bR;kfn esa Hkh jaxks ds mfpr leoU; ls 'kkafr LFkkfir dj ldrk gSA [1-3] Http://www.granthaalayah.com©International Journal of Research -GRANTHAALAYAH Http://www.granthaalayah.com©International Journal of Research -GRANTHAALAYAH
https://openalex.org/W4293241198
https://www.researchsquare.com/article/rs-1565844/latest.pdf
English
null
Construction of dynamical S-boxes based on piecewise map: Image encryption approach
Research Square (Research Square)
2,022
cc-by
9,309
Research Article Keywords: Digital image Encryption, S-box, Ergodic dynamical systems, Keyspace, Lyapunov exponent Posted Date: April 28th, 2022 DOI: https://doi.org/10.21203/rs.3.rs-1565844/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Noname manuscript No. (will be inserted by the editor) Noname manuscript No. (will be inserted by the editor) Reza Hoseini Department of Physics, Urmia University of Technology, Urmia, Iran. Sohrab Behnia (Corresponding author) Department of Physics, Urmia University of Technology, Urmia, Iran. E-mail: s.behnia@sci.uut.ac.ir Smira Fathizadeh Department of Physics, Urmia University of Technology, Urmia, Iran. Construction of dynamical S-boxes based on: Image encryption approach Reza Hoseini · Sohrab Behnia⇤· Samira Fathizadeh Received: date / Accepted: date Abstract Many cryptographic schemes have been implemented, using the connection between chaos and cryptography. However; present study intro- duces a new algorithm that, as a standard format, can be used in cyberspace. The substitution box (S-box) is one of the most successful examples of these strategies. However, most S-box design schemes contain a fixed point, which may cause a security issue for cryptographic primitives. A dynamic S-box is constructed based on a piecewise map to achieve a low automatic correlation. The introduced piecewise map has an attractive behavior by displaying the ergodic behavior and fractal features in its chaotic domain. The ergodic prop- erty is of great interest in diffusion and confusion processes for encryption. We have examed the performance of to introduced S-box. The applied attacks (Non-linearity, SAC, BIC, LP, and DP) results are close to the optimal value. Then, an image encryption algorithm based on a designated S-box is proposed, and the ability of the encrypted image to resist the attacks has been examed with statistical analysis. The proposed S-box based on the piecewise map has a large and safe Keyspace for encryption. Keywords Digital image Encryption · S-box · Ergodic dynamical systems · Keyspace · Lyapunov exponent 2 Reza Hoseini et al. 1 Introduction Along with developing the communications and multimedia industry, the se- curity of information is a challenging issue. In this regard, requirements for information confidentiality and security of the digital images transmitted over stored or open channels are also expanding [1]. The algorithms proposed for image encryption should retain the intrinsic features of the image, such as the large data aspects of images, high redundancy, and strong correlations between the pixels [2,3]. Chaos-based algorithms have attracted much attention for im- age encryption [4–6]. Chaotic systems are sensitive to their initial conditions. They have a pseudorandom nature in their chaotic domain [7,8]. [ ] Two major categories based on the chaotic encryption algorithms are stream cipher and block cipher. S-box is one of the most significant components used in symmetric block cipher of encryption algorithms of chaotic encryption [9, 10]. To guarantee encrypted information, one should obscure the connection between key and ciphertext (confusion). By spreading the information of plain- text on the ciphertext, it resists the statistical attacks (diffusion). S-box act as a core of the encoder at data encryption standard (DES) and Advanced encryption standard (AES). Network banking and web security are secured based on these standards. S-box is responsible for diffusion and confusion in cryptographic systems and guarantees their performance and their safety [11, 12]. As a heart of the traditional encryption algorithms, it must be resistant to differential and linear cryptanalysis. The security of the cryptographic scheme and its ability to resist differential and linear cryptographic attacks is related to the S-box efficiency [13,14]. An S-box can be designed based on mathe- matical structures, a group of theoretical approaches, the heuristic method, inspired by bee waggle dance, utilizing teaching learning-based optimization, mapping affine, orbit power functions and genetic algorithm [15]. [ ] In 2003, Tang et al. considered an S-box algebraic operation in the chaotic cryptosystem [16]. Some groups constructed S-box based on a chaotic Tent map and Lorenz system [17,18]. Wang designed a bijective S-box with high nonlinearity using chaos and optimization [19]. A dynamical S-box was pro- posed by Liu et al. to improve the image encryption algorithm [20]. Later, Liu group utilized a new dynamic S-box and chaotic system based on DNA sequences [21]. In 2018, a hybrid encryption algorithm based on a chaotic system and S-AES algorithm was developed [22]. 1 Introduction A novel S-box generation al- gorithm including a six-dimensional fractional Lorenz-Duffing chaotic system was proposed by Ye and Zhimao, which used fractional-order chaotic systems with more complex dynamical properties instead of integer-order ones [23]. A novel discrete-space chaotic map based on the multiplication of integer num- bers is proposed for designing an S-box [24]. In the latest researches, a strong S-box is determined thought quantum walks [25]. The ergodic property of the chaotic map could provide the resistivity respect to the brute attacks. In this regard, Wang et al. proposed a dynamical S-Box using two chaotic maps and robust against the attacks [26]. Recently, some S-Boxes are constructed using continuous chaotic systems. Construction of dynamical S-boxes based on: Image encryption approach 3 Wang et al. designed an S-Box based on an autonomous system with a line equilibrium [27]. Also, they proposed an S-Box based on a chaotic system with- out equilibrium [28]. C¸avu¸so˘glu et al. proposed an S-Box algorithm using the scaled chaotic system [29,30]. Also, an S-box based on a hyper-chaotic system has constructed by Solami et al. [31]. y [ ] The present study aims to improve the methods of creating the S-box by ap- plying trigonometric chaotic maps for the block encryption algorithm. As the confusion, diffusion, and keyspace play a vital role in encryption, accurate piecewise chaotic maps with keyspace for enhancing the safety of the cryp- tographic systems have been created. The performance of the constructed S-box is checked by different approaches. We examined the performance of the generated S-box by introducing an image encryption algorithm. The paper is organized as follows: Section 2 briefly introduces the applied chaotic systems. Section 3 contains the details and analysis of the constructed S-box. Section 4 contains a new image encryption system based on the designed S-box. More- over, a security analysis of the proposed image encryption scheme is presented in Section 4. Finally, Section 6 provides conclusions. y [ ] The present study aims to improve the methods of creating the S-box by ap- plying trigonometric chaotic maps for the block encryption algorithm. As the confusion, diffusion, and keyspace play a vital role in encryption, accurate piecewise chaotic maps with keyspace for enhancing the safety of the cryp- tographic systems have been created. The performance of the constructed 2 Applied chaotic maps We have described a hierarchy of one-parameter families of maps with ergodic behavior [32], in the interval [0, 1] as follows [33]: ΦN(x, ↵) = ↵2TN(px) 1 + (↵2 −1)TN(px) (1) (1) we present an example of these types of maps: Φ2(x, ↵) = 4↵2x(1 −x) 1 + 4(↵2 −1)x(1 −x) (2) (2) where the map Φ2(x, ↵) reduces to logistic one for ↵= 1. The nonlinear piecewise chaotic map of the interval [0, 1] with an invariant measure is defined through the following equation [33]: Φ(x, ↵, p) = ↵2F 1 + (↵2 −1)F (3) (3) where where F = 8 < : x p 0 x p x−p 1−p p x 1 (4) (4) where ↵is a control parameter. There is also a probability parameter (p). It should be noted that the construction of S-box is based on the piecewise nonlinear chaotic maps. Also, the conjugate of maps Eq.(1) are as follows [33]: ˜ΦN(x, ↵) = h ◦ΦN(x, ↵) ◦h−1 = 1 ↵2 tan2(N arctan px) (5) (5) 4 Reza Hoseini et al. 4 Reza Hoseini et al. (a) (b) Fig. 1 Lyapunov exponent (a) and bifurcation diagram (b) of Φ(x, α, p) VS. α, where p = 0.35. (b) Fig. 1 Lyapunov exponent (a) and bifurcation diagram (b) of Φ(x, α, p) VS. α, where p = 0.35. where the conjugacy denotes that invertible map h(x) = 1−x x maps I = [0, 1] into [0, 1) (see Appendix A). In the proposed algorithm, we consider the ˜Φ12(x, ↵) for image encryption. 2.0.1 Dynamical Analysis of the chaotic systems 2.0.2 Randomness test 2.0.1 Dynamical Analysis of the chaotic systems - Bifurcation diagram - Bifurcation diagram Bifurcation determines the qualitative changes of the dynamics of a system through the variation of a control parameter [33,34]. If the phase portrait of a dynamical system changes via the change of some control parameters, the bifurcation occurs. In Figs. 1b and 2b, we have demonstrated the bifurcation diagram of Φ(x, ↵, p) and ˜Φ12(x, ↵), respectively. In the non-chaotic region of the parameter, there is only a single period fixed point. -Lyapunov exponent Lyapunov exponents can estimate the sensitivity to initial conditions [33]. For the introduced piecewise mapping can be computed from λ = lim n!1 1 n n−1 X i=0 ln|δn+1| |δn| = lim n!1 1 n n−1 X i=0 ln|(Φi)0(x0)| (6) (6) A positive Lyapunov exponent determines a chaotic motion due to the expo- nentially growth of the distance between nearby points. Figs. 1a and 2a, show the Lyapunov exponent of Φ(x, ↵, p) and ˜Φ12(x, ↵), respectively. In a chaotic region, this map is ergodic, as predicted by Birkohf ergodic theorem [32]. - Time series The simplest method to characterize the properties of attractors is the recon- struction of the phase space with the help of the time series. By considering the same initial conditions (x0 = 0.23) and control parameters (↵= 0.87) after 2000 iterations, we draw the time series of Φ(x, ↵, p) and ˜Φ12(x, ↵) in Figs. 3a and 3b. There is a clear separation between the time series. Construction of dynamical S-boxes based on: Image encryption approach 5 (a) (b) Fig. 2 Lyapunov exponent (a) and bifurcation diagram (b) of ˜Φ12(x, α) VS. α. (a) (b) Fig. 3 Time series of Φ(x, α, p) (a) and ˜Φ12(x, α) (b) where x0 = 0.23, α = 0.87 and p = 0.35 (a) (b) Fig. 2 Lyapunov exponent (a) and bifurcation diagram (b) of ˜Φ12(x, α) VS. α. (a) (b) Fig. 2 Lyapunov exponent (a) and bifurcation diagram (b) of ˜Φ12(x, α) VS. α. (b) (a) Fig. 2 Lyapunov exponent (a) and bifurcation diagram (b) of ˜Φ12(x, α) VS. α. (a) (b) Fig. 3 Time series of Φ(x, α, p) (a) and ˜Φ12(x, α) (b) where x0 = 0.23, α = 0.87 and p = 0.35 (a) Fig. 3 Time series of Φ(x, α, p) (a) and ˜Φ12(x, α) (b) where x0 = 0.23, α = 0.87 and p = 0.35 The security analysis is performed to examine the randomness of the generated sequence. - Bifurcation diagram We use the National Institute of Standards and Technology (NIST) test. The NIST involves nine tests, including frequency, block frequency, runs, universal, discrete Fourier transform, longest run, rank, linear complexity, se- rial and random excursion. The obtained results, according to Table 1, show that the generated sequence of the proposed chaotic system successfully passes these randomness tests. 3 Substitution box and its construction 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 132 201 184 164 42 11 173 39 151 254 200 66 1 53 91 63 2 252 87 172 18 19 148 138 73 236 85 108 114 86 33 116 111 3 97 213 110 7 107 190 218 121 37 220 21 20 71 100 90 179 4 60 202 41 170 209 96 54 83 221 210 122 153 67 75 168 74 5 15 167 92 25 249 223 238 52 146 69 211 248 208 176 196 143 6 72 78 145 134 46 207 194 29 216 231 186 113 14 152 206 129 7 133 203 59 135 195 205 244 192 191 47 119 144 227 80 24 28 8 103 198 124 105 13 101 214 156 84 62 149 154 181 57 246 104 9 55 233 77 165 229 174 112 182 140 166 188 139 128 35 3 212 10 45 255 106 120 197 22 225 115 17 224 130 215 10 6 241 245 11 12 16 161 222 49 178 242 79 199 94 162 228 239 30 117 180 12 187 159 31 219 123 147 243 95 56 82 26 109 48 193 38 169 13 250 61 44 8 0 230 36 204 68 64 2 99 98 81 65 70 14 126 23 150 235 51 50 76 226 88 32 171 253 142 125 136 102 15 27 177 175 4 217 13175 247 118 237 40 141 234 185 155 93 16 58 157 43 9 251 89 160 183 232 240 34 163 189 158 137 127 obtain a next number. Step 6: Steps 4 and 5 are repeated until the S-box is filled. The constructed S-box is presented in Table 2 where we have considered x0 = 0.23, ↵= 0.87 and p = 0.35. These constants build the key of the generated S-box. Table 2 The produced S-box for piecewise chaotic map Φ(x, α, p) where x0 = 0.23, α = 0.87, and p = 0.35. 3 Substitution box and its construction In this study, the chaotic nature of a piecewise chaotic map has been used to generate the S-box. We are using S-box in the confusing process as a nonlinear transformation [1,4]. The generation process contains six steps. Step 1: Input: initial condition of the chaotic map (x0) and the control param- eters (↵). Step 2: The piecewise chaotic map (Eq. (4)) is iterated n = 104 times to deviate the transient condition and produce a random sequence, X = (x1, x2, x3, ..., xN); Step 3: Set i = 1 and j = 1; Step 3: Set i = 1 and j = 1; Step 4: Convert the first sequence of X into an integer sequence using the following formula: following formula: (7) (7) Si = xj ⇥1016 mod 256, (7) This number is the first element of the S-box. Step 5: It is checked that if this number does not exist in the S-box, a new element is added, and i = i + 1 & j = j + 1, otherwise, we repeat step 4 to 6 Reza Hoseini et al. Table 1 Results of NIST SP 800–22 statistical test suite by a piecewise chaotic map (x0 = 0.23, α = 0.87, p = 0.35). We are using 1000 samples of 1 M bits data. Statistical Test P-value Result Frequency 0.5681 Pass Block-frequency 0.4819 Pass Cumulative Sums 0.3851 Pass Runs(M=10000) 0.6915 Pass Discrete fourier transform 0.3419 Pass Longest Runs of Ones 0.3972 Pass Nonperiodic Template 0.2715 Pass Rank 0.4219 Pass Approximate Entropy 0.6782 Pass Overlapping template of all ones 0.3161 Pass Universal Statistical 0.5037 Pass Linear complexcity 0.1277 Pass Serial 0.3908 Pass Random excursion x=-4 0.6057 Pass x=-3 0.5582 Pass x=-2 0.6982 Pass x=-1 0.1278 Pass x=1 0.3900 Pass x=2 0.5082 Pass x=3 0.6618 Pass x=4 0.2764 Pass Table 2 The produced S-box for piecewise chaotic map Φ(x, α, p) where x0 = 0.23, α = 0.87, and p = 0.35. 3 Substitution box and its construction [45] 106 102 104 0.6094 0.3750 0.4980 103.29 0.4971 0.1484 0.0390 Ref. [46] 110 104 106 0.5937 0.4218 0.5039 103.4 0.5058 - - Proposed S-box 106 98 103.75 0.5937 0.3906 0.4965 103.8 0.4628 0.125 0.0468 To test our proposed method, we have applied several attacks to the S-box. To test our proposed method, we have applied several attacks to the S-box. Nonlinearity Nonlinearity is the most significant feature of the S-box. The nonlinearity of a nonlinear alternating box indicates the non-linear order of the Boolean functions of the box, is equal to 112 for the Lorentz replacement box. Nonlinearity has been defined by using Walsh Spectrum [35]. Nonlinearity Nonlinearity is the most significant feature of the S-box. The nonlinearity of a nonlinear alternating box indicates the non-linear order of the Boolean functions of the box, is equal to 112 for the Lorentz replacement box. Nonlinearity has been defined by using Walsh Spectrum [35]. Nf = 2n−1(1 −2−n max ω2GF (2n) | S(f)(!) |) (8) (8) In the above equation, S(f)(!) is the Walsh spectrum for f(x) and is given by: In the above equation, S(f)(!) is the Walsh spectrum for f(x) and is given by: S(f)(!) = X ω2GF (2n) (−1)f(x)L ω.x (9) (9) In this regard, ! belongs to GF(2n), and x and ! are multiplied by a dot product. The ideal value for non-linear S-box is 112 [36]. The maximum accessible value in recent studies is around 106. We compare the results of the non-linearity analysis of our proposed scheme in Table 3. In this regard, ! belongs to GF(2n), and x and ! are multiplied by a dot product. The ideal value for non-linear S-box is 112 [36]. The maximum accessible value in recent studies is around 106. We compare the results of the non-linearity analysis of our proposed scheme in Table 3. – Strict Avalanche Criterion To check whether a given cryptographic transformation satisfies the strict avalanche criterion (SAC), we examined our S-box (Table 2). The next principle test on the generated S-box is SAC. The reported domain for this test is between 0.3906 and 05937, and the ideal value is 0.5 [36,37]. The dependence matrix of the generated S-box is presented in Table 3 and Table 4. It has shown they are close to the optimal value. – Bit Dependent Critica Webster and Tavers introduced the bit dependent critica (BDP) to analyze the S-box [38]. 3 Substitution box and its construction 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 132 201 184 164 42 11 173 39 151 254 200 66 1 53 91 63 2 252 87 172 18 19 148 138 73 236 85 108 114 86 33 116 111 3 97 213 110 7 107 190 218 121 37 220 21 20 71 100 90 179 4 60 202 41 170 209 96 54 83 221 210 122 153 67 75 168 74 5 15 167 92 25 249 223 238 52 146 69 211 248 208 176 196 143 6 72 78 145 134 46 207 194 29 216 231 186 113 14 152 206 129 7 133 203 59 135 195 205 244 192 191 47 119 144 227 80 24 28 8 103 198 124 105 13 101 214 156 84 62 149 154 181 57 246 104 9 55 233 77 165 229 174 112 182 140 166 188 139 128 35 3 212 10 45 255 106 120 197 22 225 115 17 224 130 215 10 6 241 245 11 12 16 161 222 49 178 242 79 199 94 162 228 239 30 117 180 12 187 159 31 219 123 147 243 95 56 82 26 109 48 193 38 169 13 250 61 44 8 0 230 36 204 68 64 2 99 98 81 65 70 14 126 23 150 235 51 50 76 226 88 32 171 253 142 125 136 102 15 27 177 175 4 217 13175 247 118 237 40 141 234 185 155 93 16 58 157 43 9 251 89 160 183 232 240 34 163 189 158 137 127 Construction of dynamical S-boxes based on: Image encryption approach 7 Table 3 Comparison of S-box analysis (Non-linearity, SAC, BIC, LP, and DP ) with pro- posed S-box. Table 3 Comparison of S-box analysis (Non-linearity, SAC, BIC, LP, and DP ) with pro- posed S-box. S-box Nonlinearity SAC BIC LP DP Max Min Ave Max Min Ave Non. SAC Ref. [41] 110 104 106.4 - - 0.503 103.9 0.507 0.113 0.039 Ref. [42] 108 104 106.25 0.5546 0.4453 0.5009 103.6 0.4996 - - Ref. [43] 106 104 104.3 0.5914 0.4111 - - 0.5317 0.1484 0.0625 Ref. [44] - - 103.25 - - 0.5151 103.07 0.5449 0.1562 0.0391 Ref. 3 Substitution box and its construction It is given by DP = max ∆x6=0,∆y | ]{x 2 X|f(x) L f(x L ∆x) = ∆y} 2n −1 2| (11) (11) The smaller the DP value, the stronger the ability to fight against linear transmission attacks. The results of the DP analysis are shown in Table 3. The smaller the DP value, the stronger the ability to fight against linear transmission attacks. The results of the DP analysis are shown in Table 3. We compare the results of our work with the results of others who have cre- ated an S-box using chaos in Table 3. It is shown that the created S-box is best suited for different security goals to construct confusion in cryptography systems. We have used the constructed S-box for image encryption to show the performance of the proposed chaotic map. 3 Substitution box and its construction In this technique, it is tried to understand whether the set of vectors produced with the switch bit of plaintext is autonomous from all avalanche variable set or not. The third criterion is BAC, which includes two tests BIC-nonlinearity and BIC-SAC, with values 112 and 0.5 [36]. A Piecewise map could certify these two tests with 103.8 and 0.4628. The test results are presented in Table 3, Table 5 and Table 6. Reza Hoseini et al. 8 Table 4 The dependence matrix for the proposed S-box. 0 1 2 3 4 5 6 7 0.53125 0.578125 0.484375 0.4375 0.40625 0.515625 0.46875 0.515625 0.53125 0.50000 0.53125 0.46875 0.515625 0.484375 0.525625 0.484375 0.515625 0.59375 0.484375 0.46875 0.5 0.453125 0.53125 0.515625 0.484375 0.546875 0.50000 0.53125 0.46875 0.53125 0.4375 0.46875 0.4375 0.515625 0.546875 0.59375 0.421875 0.5 0.390625 0.421875 0.46875 0.46875 0.453125 0.546875 0.484375 0.546875 0.453125 0.53125 0.453125 0.484375 0.4375 0.4375 0.5 0.5 0.5 0.59375 0.5625 0.484375 0.5 0.5 0.5 0.46875 0.515625 0.5625 Table 4 The dependence matrix for the proposed S-box. – Linear Approximation Probability – Linear Approximation Probability We could compute the masking of all inputs by considering the two masks Γ x and Γ y. The maximum value of masking is known as a maximum linear approximation. If we represent the input value by x and the cor- responding output S(x), the linear approximation probability would be defined as [39]: We could compute the masking of all inputs by considering the two masks Γ x and Γ y. The maximum value of masking is known as a maximum linear approximation. If we represent the input value by x and the cor- responding output S(x), the linear approximation probability would be defined as [39]: LP = max Γ x,Γ y6=0 |]{x|x.Γx = S(x).Γy} 2n −1 2| (10) (10) The smaller the LP value, the stronger the ability to fight against linear transmission attacks, and vise versa. The results of this analysis are shown in Table 3. – Differential Approximation Probability – Differential Approximation Probability The differential approximation probability (DP) is the maximum probabil- ity of y output, when the input is x, and it is used to represent the XOR distribution of input and output of the boolean function [40]. 4 Proposed Image Encryption Algorithm In this algorithm, first, the image is divided into a series of non-overlapping block (Id⇥e). Then, we convert the pixels of the image into the matrix (O(m⇥ n)). At the next step, the key (initial condition and control parameter) for the encryption process is set. The following step consists of the Sub Byte process by using the S-box. The elements of the S-box determine the rows and columns Construction of dynamical S-boxes based on: Image encryption approach 9 Table 5 The BIC-Nonlinearity matrix for proposed S-box. —— 104.000 100.000 98.000 102.000 106.000 108.000 104.000 104.000 —– 106.000 106.000 104.000 104.000 102.000 106.000 100.000 106.000 —— 104.000 108.000 104.000 102.000 102.000 98.000 106.000 104.000 —– 104.000 106.000 106.000 98.000 102.000 104.000 108.000 104.000 —– 104.000 106.000 100.000 106.000 104.000 104.000 106.000 104.000 —– 108.000 106.000 108.000 102.000 102.000 106.000 106.000 108.000 —— 100.000 104.000 106.000 102.000 98.000 100.000 106.000 100.000 —— Table 6 The BIC-SAC matrix for the proposed S-box. —— 0.494141 0.505859 0.527344 0.492188 0.501953 0.500000 0.513672 0.494141 —— 0.523438 0.511719 0.476563 0.505859 0.500000 0.501953 0.505859 0.523438 —— 0.503906 0.482422 0.531250 0.498047 0.515625 0.527344 0.511719 0.503906 —— 0.507813 0.486328 0.509766 0.496094 0.492188 0.476563 0.482422 0.507813 —— 0.527344 0.500000 0.503906 0.501953 0.505859 0.531250 0.486328 0.527344 —— 0.501953 0.492188 0.500000 0.500000 0.498047 0.509766 0.500000 0.501953 —— 0.462891 0.513672 0.501953 0.515625 0.496094 0.503906 0.492188 0.462891 —— Table 5 The BIC-Nonlinearity matrix for proposed S-box. Table 6 The BIC-SAC matrix for the proposed S-box. —— 0.494141 0.505859 0.527344 0.492188 0.501953 0.500000 0.513672 0.494141 —— 0.523438 0.511719 0.476563 0.505859 0.500000 0.501953 0.505859 0.523438 —— 0.503906 0.482422 0.531250 0.498047 0.515625 0.527344 0.511719 0.503906 —— 0.507813 0.486328 0.509766 0.496094 0.492188 0.476563 0.482422 0.507813 —— 0.527344 0.500000 0.503906 0.501953 0.505859 0.531250 0.486328 0.527344 —— 0.501953 0.492188 0.500000 0.500000 0.498047 0.509766 0.500000 0.501953 —— 0.462891 0.513672 0.501953 0.515625 0.496094 0.503906 0.492188 0.462891 —— Table 6 The BIC-SAC matrix for the proposed S-box. on the image matrix for applying the byte transformation. This step allows the proposed encryption algorithm acts as similar to the standard encryption algorithm. on the image matrix for applying the byte transformation. This step allows the proposed encryption algorithm acts as similar to the standard encryption algorithm. Finally, the map ˜Φ12(x, ↵), is used in the diffusion process. The prelimi- nary security analysis is performed to examine the randomness of the gen- erated sequence. Therefore, sequence acts as a chaos-based PRNG (pseudo- random number generator). 4 Proposed Image Encryption Algorithm The generated random decimal sequence X = {x0, x1, ...}, x 2 [0, 255], according to the following relation, diffuses the first block of image. M(i) = XOR(mod(x(i) ⇥109, 256), I(i)) (12) (12) Fig. 4 shows the block diagram of the proposed algorithm. Replacing the keys by considering the generated random sequence for the next blocks increases safety in encryption. The decryption process is the inverse process of encryp- tion. Fig. 4 shows the block diagram of the proposed algorithm. Replacing the keys by considering the generated random sequence for the next blocks increases safety in encryption. The decryption process is the inverse process of encryp- tion. 4.1 Security analysis The pseudo-randomness nature of the chaotic process could provide excellent confusion property for encryption(Fig. 5 and 8). We are using the time series of the introduced piecewise map to produce the S-box. To encrypt the image, we first created an S-box using a chaotic map to allow the message to be diffused and then encrypted it using a second chaotic map. In the encoding step, we divided the image into four blocks. The output map of the first block is employed to encode the next block. Then, the blocks are appended together to be created an encrypted image. 10 Reza Hoseini et al. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Fig. 4 The flowchart of the proposed algorithm for image encryption. Fig. 4 The flowchart of the proposed algorithm for image encryption. We have employed various tests to check the proposed algorithm. We have used the “Al Madinah” gray images (Fig. 5a), with dimensions of 256 ⇥256 for experimental purposes, and the encrypted image is shown in Fig. 5b. The image encryption keys are x0 = 0.65 and ↵= 2.2. We have employed various tests to check the proposed algorithm. We have used the “Al Madinah” gray images (Fig. 5a), with dimensions of 256 ⇥256 for experimental purposes, and the encrypted image is shown in Fig. 5b. The image encryption keys are x0 = 0.65 and ↵= 2.2. 4.1.1 Histogram analysis 4.1.1 Histogram analysis By drawing the number of observations of each brightness intensity, histogram analysis shows how the pixels are distributed in the image [4]. Distribution of relatively uniform image histograms may indicate the quality of encryption. Figs. 5c and 5d represent the histogram of the gray-image before cryptography and after cryptography, respectively. As the image shows, it is uniform. Hence, it does not give any data about the distribution of gray values to the attacker. 4.1.2 Correlation coefficient analysis [50] Proposed Information entropy 7.9971 7.9995 7.9974 7.9785 7.9972 Differential UACI% 0.9959 0.9960 0.9965 0.9960 0.27 attack NPCR% 0.3334 0.3346 0.3341 0.3346 0.99 Correlation Horizontal 0.0015 -0.0006 -0.0027 -0.0082 0.0012 coefficient Vertical 0.001 0.0008 0.0023 -0.0032 0.0033 Diagonal 0.0011 0.0001 0.0088 -0.0405 0.0019 (a) (b) Fig. 6 (a) Correlation analysis of plain image; (b) Correlation analysis of a ciphered image. Reza Hoseini et al. 12 Table 7 Comparison of security analysis on an encrypted image with related work. Security analysis Ref. [47] Ref.[48] Ref. [49] Ref. [50] Proposed Information entropy 7.9971 7.9995 7.9974 7.9785 7.9972 Differential UACI% 0.9959 0.9960 0.9965 0.9960 0.27 attack NPCR% 0.3334 0.3346 0.3341 0.3346 0.99 Correlation Horizontal 0.0015 -0.0006 -0.0027 -0.0082 0.0012 coefficient Vertical 0.001 0.0008 0.0023 -0.0032 0.0033 Diagonal 0.0011 0.0001 0.0088 -0.0405 0.0019 (a) (b) Fig. 6 (a) Correlation analysis of plain image; (b) Correlation analysis of a ciphered image. Table 7 Comparison of security analysis on an encrypted image with related work. Security analysis Ref. [47] Ref.[48] Ref. [49] Ref. [50] Proposed Information entropy 7.9971 7.9995 7.9974 7.9785 7.9972 Differential UACI% 0.9959 0.9960 0.9965 0.9960 0.27 attack NPCR% 0.3334 0.3346 0.3341 0.3346 0.99 Correlation Horizontal 0.0015 -0.0006 -0.0027 -0.0082 0.0012 coefficient Vertical 0.001 0.0008 0.0023 -0.0032 0.0033 Diagonal 0.0011 0.0001 0.0088 -0.0405 0.0019 Fig. 6 (a) Correlation analysis of plain image; (b) Correlation analysis of a ciphered image. 4.1.2 Correlation coefficient analysis 4.1.2 Correlation coefficient analysis In general, each image pixel correlates with neighboring pixels, and when an image is encrypted, the correlation decreases. To calculate the correlation of pixels in the image, we use the following equations for horizontal, vertical, and Construction of dynamical S-boxes based on: Image encryption approach 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 (a) (b) (c) (d) Fig. 5 (a)Original image; (b) Encrypted image; (c) Histogram of original image; (d) His- togram of encrypted image. (b) (a) (b) (b) (d) (a) (b) (c) (d) (a) (c) (c) Fig. 5 (a)Original image; (b) Encrypted image; (c) Histogram of original image; (d) His- togram of encrypted image. diagonal directions. ons. E(x) = 1 N N X i=1 xi (13) D(x) = 1 N N X i=1 (xi −E(x))2 (14) cov(x, y) = 1 N N X i=1 (xi −E(x))(yi −E(y)) (15) rxy = cov(x, y) p D(x) ⇥ p D(y) (16) E(x) = 1 N N X i=1 xi (13) (13) (14) (15) (16) where x and y are the brightness values of the neighboring pixels in the im- age, and N is the number of neighboring pixels selected from the image to consider the correlation. Table 7 shows the correlation of the encrypted im- ages in three horizontal, vertical, and diagonal directions. It is shown that the correlation coefficient is significantly reduced, indicating the efficiency of the proposed layout to resist statistical attacks. Figs. 6a and 6b show the correla- tion graphs of the original and encrypted images, respectively. Also, in Table 7, the correlation coefficients of the proposed method are compared with different references. The small numbers of a correlation coefficient, compared to other references, indicate that the encoded image data is independent and distances from each other. So the cryptographic effect is ideal. 12 Reza Hoseini et al. Table 7 Comparison of security analysis on an encrypted image with related work. Security analysis Ref. [47] Ref.[48] Ref. [49] Ref. 4.1.4 Information entropy The information entropy can be employed as a measure of the degree of per- turbation of gray pixel surfaces. The entropy of an image is used from the following relation [4]: H(m) = 2N−1 X i=1 P(mi)log2 1 P(mi) (20) (20) p(mi) is the probability of grayscale mi and 2N −1 the number of gray levels possible. In a perfectly uniform image with 256 gray levels where the proba- bility of all pixels being the same, the entropy value would be 8. The closeness of the entropy value of the encoded image to 8 means the efficiency of the provided method in the image encryption. We have got entropy 7.9972, which is a good result (Table 7). 4.1.3 Differential attack The number of changing pixel rate (NPCR) and unified averaged changed intensity (UACI) tests are utilized to determine the effect of changing one pixel on the encrypted image. If one-pixel changes from the original image, the NPCR shows the rate of change of pixels. The average intensity difference between an original image and an encoded image is determined using UACI [4]. For the calculation of NPCR and UACI, we consider two ciphered images (C1(i, j) and C2(i, j)) that have a one-pixel difference. The PCR and UACI tests are defined as follows: NPCR = P i,j D(i, j) M ⇥N ⇥100% (17) (17) UACI = 1 M ⇥N [ X i,j | C1(i, j) −C2(i, j) | 255 ] ⇥100% (18) (18) In the above equations, M and N are the image size and D defined as follows: F = ⇢D(i, j) = 1 ifC1(i, j) = C2(i, j) D(i, j) = 0 otherwise (19) (19) The obtained results render NPCR = 0.99% and UACI = 0.27%. Hence, the proposed algorithm can survive a differential attack (Table 7). The obtained results render NPCR = 0.99% and UACI = 0.27%. Hence, the proposed algorithm can survive a differential attack (Table 7). Construction of dynamical S-boxes based on: Image encryption approach 13 Table 8 The key space analysis of piecewise chaotic map. complex map Best range Precision Precision parameter (float) (binary bit) x0(Φ(x, α, p)) [0, 1] 10−14 46 x0( ˜Φ12(x, a)) [0, 1] 10−14 46 control parameter (α) [0.51, 1.25] 10−14 46 control parameter (a) [0.5, 1] 10−14 K Keyspace: 46+46+46+K=140+K bit Table 8 The key space analysis of piecewise chaotic map. 4.1.4 Information entropy 4.1.4 Information entropy 4.3 Analysis of speed 4.3 Analysis of speed When the amount of information is large, the encryption algorithm must be fast. For speed analysis of the proposed image encryption/decryption tech- nique, we used an Intel Core i7 3.40 GHz CPU with 16 GB RAM running on Microsoft Windows 7 and using MATLAB 2014b. We have used 256 grey- scale images of different sizes, to compare the results of encryption/decryption speed test. The total encryption and decryption time is presented in Table 9. The cryptographic image histogram has a uniform distribution for the sam- ple (Fig. 5) and standard test images (Fig. 8). The security analysis of the encrypted images shows that the correlation between the pixels was reduced (Fig. 6). Also, Table 7 shows the acceptable performance of the proposed method against attacks such as information entropy, differential attack, and correlation coefficient. The present comparison shows the performance of the proposed method against attacks [47–50]. Increasing the robustness against general attacks, and increasing the key-space are the main advantages of the introduced piecewise map. When the amount of information is large, the encryption algorithm must be fast. For speed analysis of the proposed image encryption/decryption tech- nique, we used an Intel Core i7 3.40 GHz CPU with 16 GB RAM running on Microsoft Windows 7 and using MATLAB 2014b. We have used 256 grey- scale images of different sizes, to compare the results of encryption/decryption speed test. The total encryption and decryption time is presented in Table 9. p yp yp p The cryptographic image histogram has a uniform distribution for the sam- ple (Fig. 5) and standard test images (Fig. 8). The security analysis of the encrypted images shows that the correlation between the pixels was reduced (Fig. 6). Also, Table 7 shows the acceptable performance of the proposed method against attacks such as information entropy, differential attack, and correlation coefficient. The present comparison shows the performance of the proposed method against attacks [47–50]. Increasing the robustness against general attacks, and increasing the key-space are the main advantages of the introduced piecewise map. 4.2 Keyspace analysis The essential aspect of any cryptographic system is the key. It might be defined in the chaotic domain, which is characterized by the Lyapunov exponent spec- trum. Variation of the Lyapunov exponent concerning the control parameter clarifies the positive domain (Figs. 1(a) and 2(a)). In this study, the initial con- dition x0, control parameters (a and ↵), and the number of mapping iterations (m and n) have arranged the keyspace. It is computed as below. T(x0, a, ↵, m, n) = ✓(x0 ⇥a ⇥↵⇥m ⇥n) (21) (21) T(x0, a, ↵, m, n) = ✓(x0 ⇥a ⇥↵⇥m ⇥n) where x0 2 [0 1] and for Φ(x, ↵, p)(Φ12(x, ↵)) control parameter belongs to [0.51 1.25]([0.5 1]), and m, n indicates the number of mapping iterations. We consider double-precision numbers(1016) for parameters. In standard cryp- tographic systems, the size of the key-space is 2256 [1]. Table 8 lists the best precision for the values of each parameter. The key-space is large enough for many cryptographic systems. Key-sensitivity analysis: To test this sensitivity, we have tried our algorithm with the false key, which is slightly different from the original key (Fig. 7). In the analysis, it was evident that key-sensitivity is of order 10−16. 14 Reza Hoseini et al. (a) (b) Fig. 7 Extract the Encrypted image with (a) Correct key (x0 = 0.65 and α = 2.2 ), (b) Wrong key (x0 = 0.65000000000001 and α = 2.2 ). (a) Fig. 7 Extract the Encrypted image with (a) Correct key (x0 = 0.65 and α = 2.2 ), (b) Wrong key (x0 = 0.65000000000001 and α = 2.2 ). Table 9 Encryption/decryption speed test results. Table 9 Encryption/decryption speed test results. Image size (in pixels) Proposed Akhshani et al. [51] Chen et al. [52] 256 ⇥256 0.19/0.19 0.046/0.046 <0.4 / <0.4 512 ⇥512 0.78/0.75 0.23/0.23 1 / 1 1024 ⇥1024 2.93/2.84 0.953/0.953 3/3 2048 ⇥2048 8.16/8.14 3.89/3.89 14/14 5 Conclusion It is shown that these maps have an attractive property, that is, for even values of N the ΦN(x, α) maps have only a fixed point attractor x = 1 provided that their parameter belongs to interval (N, 1) while at α ≥N they bifurcate to the chaotic regime without having any period doubling or period-n-tupling scenario and remain chaotic for all α 2 (0, N), but for odd values of N, these maps have only fixed point attractor at x = 0 for α 2 ( 1 N , N), again they bifurcate to a chaotic regime at α ≥ 1 N , and remain chaotic for α 2 (0, 1 N ), finally they bifurcate at α = N to have x = 1 as fixed point attractor for all α 2 ( 1 N , 1). Here in this paper we are concerned with their conjugate maps which are defined as: ˜ΦN(x, α) = h ◦ΦN(x, α) ◦h−1 = 1 α2 tan2(N arctan pxn), (23) Conjugacy means that invertible map h(x) = 1−x x maps I = [0, 1] into [0, 1). We have derived analytically their invariant measure for arbitrary values of the parameter α and any integer values of N: µΦN (x,α)(x, β) = 1 π p β p x(1 −x)(β + (1 −β)x) . (24) with β > 0 is the invariant measure of the maps Φ(1,2) N (x, α) provided that, we choose the parameter α in the following form: α = 8 > > > > > < > > > > > : Σ [ (N−1) 2 ] k=0 CN 2k+1β−k Σ [ N 2 ] k=0CN 2kβ−k for odd values of N βΣ [ (N) 2 ] k=0 CN 2kβ−k Σ [ (N−1) 2 ] k=0 CN 2k+1β−k for even values of N (25) where the symbol [ ] means the greatest integer part. Studies based on invariant measure analysis can be useful for confirming the ergodic behaviour of a map. An ergodic system has ‘convergent’ qualities over time, variances are finite and a non-time-dependent process. 5 Conclusion Digital communication rapidly grows owing to the accessibility of digital tech- nology. This requirement restricts encryption methods to the allegiance of standard cryptographic algorithms (such as AES and DES). The critical parts Construction of dynamical S-boxes based on: Image encryption approach 15 (a) (b) (c) (d) (e) (f) (g) (h) Fig. 8 (a),(c),(e),(g) Plain image and (b),(d),(f),(h) corresponding ciphered image. (a) (b) (b) (a) (d) (c) (d) (e) (f) (g) (h) Fig. 8 (a),(c),(e),(g) Plain image and (b),(d),(f),(h) corresponding ciphered image. of a standard cryptography algorithm are the properties of confusion, diffu- sion, and nonlinearity, which guarantee the performance and security of the algorithm. The confusion in encryption meets the purpose of obscuring the relationship between the key and the ciphertext. Also, diffusion stands for spreading the influence of one plaintext bit over other ciphertext bits intended Reza Hoseini et al. 16 to hide the statistical properties of plaintext. In this study, we have used a piecewise ergodic-map and trigonometric map to increase both the number of keys and the complexities involved in the algorithm. The essential point of the proposed maps is their ergodic property. The restriction of proposed methods is in the speed of the required time for encryption. It seems that the proposed algorithm is safe compared to the other known block ciphers. Further stud- ies must be conducted to develop methods for group key management and public-key based on the proposed method. proposed maps is their ergodic property. The restriction of proposed methods is in the speed of the required time for encryption. It seems that the proposed algorithm is safe compared to the other known block ciphers. Further stud- ies must be conducted to develop methods for group key management and public-key based on the proposed method. A One parameter families As an example, some of these maps are given below: Φ2 = 4α2x(1 −x) 1 + 4(α2 −1)x(1 −x) , Φ3 = α2x(4x −3)2 α2x(4x −3)2 + (1 −x)(4x −1)2 . (22) Where α is the control parameter that can potentially be used as the secret keys for secure communication. F substitute with Chebyshev polynomial of type one TN(x). 5 Conclusion 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 A One parameter families As an example, some of these maps are given below: Φ2 = 4α2x(1 −x) 1 + 4(α2 −1)x(1 −x) , Φ3 = α2x(4x −3)2 α2x(4x −3)2 + (1 −x)(4x −1)2 . (22) (22) Where α is the control parameter that can potentially be used as the secret keys for secure communication. F substitute with Chebyshev polynomial of type one TN(x). It is shown that these maps have an attractive property, that is, for even values of N the ΦN(x, α) maps have only a fixed point attractor x = 1 provided that their parameter belongs to interval (N, 1) while at α ≥N they bifurcate to the chaotic regime without having any period doubling or period-n-tupling scenario and remain chaotic for all α 2 (0, N), but for odd values of N, these maps have only fixed point attractor at x = 0 for α 2 ( 1 N , N), again they bifurcate to a chaotic regime at α ≥ 1 N , and remain chaotic for α 2 (0, 1 N ), finally they bifurcate at α = N to have x = 1 as fixed point attractor for all α 2 ( 1 N , 1). Here in this paper we are concerned with their conjugate maps which are defined as: ˜ΦN(x, α) = h ◦ΦN(x, α) ◦h−1 = 1 α2 tan2(N arctan pxn), (23) ˜ΦN(x, α) = h ◦ΦN(x, α) ◦h−1 = 1 α2 tan2(N arctan pxn), (23) (23) Conjugacy means that invertible map h(x) = 1−x x maps I = [0, 1] into [0, 1). We have derived analytically their invariant measure for arbitrary values of the parameter α and any integer values of N: µΦN (x,α)(x, β) = 1 π p β p x(1 −x)(β + (1 −β)x) . (24) µΦN (x,α)(x, β) = 1 π p β p x(1 −x)(β + (1 −β)x) . 5 Conclusion (24) (24) with β > 0 is the invariant measure of the maps Φ(1,2) N (x, α) provided that, we choose the parameter α in the following form: with β > 0 is the invariant measure of the maps Φ(1,2) N (x, α) provided that, we choose the parameter α in the following form: N parameter α in the following form: α = 8 > > > > > < > > > > > : Σ [ (N−1) 2 ] k=0 CN 2k+1β−k Σ [ N 2 ] k=0CN 2kβ−k for odd values of N βΣ [ (N) 2 ] k=0 CN 2kβ−k Σ [ (N−1) 2 ] k=0 CN 2k+1β−k for even values of N (25) where the symbol [ ] means the greatest integer part. Studies based on invariant measure analysis can be useful for confirming the ergodic behaviour of a map. An ergodic system has ‘convergent’ qualities over time, variances are finite and a non-time-dependent process. α = 8 > > > > > < > > > > > : Σ [ (N−1) 2 ] k=0 CN 2k+1β−k Σ [ N 2 ] k=0CN 2kβ−k for odd values of N βΣ [ (N) 2 ] k=0 CN 2kβ−k Σ [ (N−1) 2 ] k=0 CN 2k+1β−k for even values of N (25) (25) where the symbol [ ] means the greatest integer part. Studies based on invariant measure analysis can be useful for confirming the ergodic behaviour of a map. An ergodic system has ‘convergent’ qualities over time, variances are finite and a non-time-dependent process. Construction of dynamical S-boxes based on: Image encryption approach 17 B Compliance with ethical standards Funding: Iranian National Science Foundation (INSF) (grant number 98006790). Funding: Iranian National Science Foundation (INSF) (grant number 98006790). Conflict of Interest: Authors Sohrab Behnia, Reza Hoseini Sani, and Samira Fathizadeh have received research grants from the Iranian National Science Foundation. References 1. C. Paar, J. Pelzl, Understanding cryptography: a textbook for students and practitioners (Springer Science & Business Media, 2009) 2. X. Huang, Y. Dong, H. Zhu, G. Ye, Visually asymmetric image encryption algorithm based on SHA-3 and compressive sensing by embedding encrypted image, Alexandria Engineering Journal 61(10), 1110 (2022). 3. S.D. Isaac, Z.T. Njitacke, N. Tsafack, C.T. Tchapga, J. Kengne, Novel compressive sens- ing image encryption using the dynamics of an adjustable gradient Hopfield neural net- work, The European Physical Journal Special Topics, doi.org/10.1140/epjs/s11734-022- 00472-2 (2022). 4. S. Behnia, A. Akhshani, S. Ahadpour, A. Akhavan, H. Mahmodi, Cryptography based on chaotic random maps with position dependent weighting probabilities, Chaos, Solitons & Fractals 40(1), 362 (2009). 5. S. Behnia, A. Akhshani, H. Mahmodi, A. Akhavan, Chaotic cryptographic scheme based on composition maps, International Journal of Bifurcation and chaos 18(01), 251 (2008). 6. X. Huang, Image encryption algorithm using chaotic Chebyshev generator, Nonlinear Dynamics 67(4), 2411 (2012). 7. S. Behnia, A. Akhavan, A. Akhshani, A. Samsudin, A novel dynamic model of pseudo random number generator, Journal of Computational and Applied Mathematics 235(12), 3455 (2011). ( ) 8. A. Akhshani, S. Behnia, A. Akhavan, S.C. Lim, Z. Hassan, Pseudo random number generator based on synchronized chaotic maps, International Journal of Modern Physics C 21(02), 275 (2010). ( ) ( ) 9. I. Hussain, T. Shah, H. Mahmood, M.A. Gondal, A projective general linear group based algorithm for the construction of substitution box for block ciphers, Neural Computing and Applications 22(6), 1085 (2013). pp ( ), ( ) 10. H. Isa, N. Jamil, M.R. Z’aba, Construction of Cryptographically Strong S-Boxes Inspired by Bee Waggle Dance, New generation computing 34(3), 221 (2016). ( ) ( ) 11. T. Farah, R. Rhouma, S. Belghith, A novel method for designing S-box based on chaotic map and Teaching–Learning-Based Optimization, Nonlinear Dynamics 88(2), 1059 (2017). 12. Y. Wang, K.W. Wong, C. Li, Y. Li, A novel method to design S-box based on chaotic map and genetic algorithm, Physics Letters A, 376(6), 827 (2012). 13. H. Huifang, Z. Hongyan, Application Research of Computers 6(547), 47 (2016) 14. Y. Zhang, A new unified image encryption algorithm based on a lifting transformation and chaos, Information Sciences 547(6), 307 (2016). 15. Y. Wang, K.W. Wong, C. Li, Y. Li, A novel method to design S-box based on chaotic map and genetic algorithm, Physics Letters A 376(6-7), 827 (2012). 16. G. Tang, S. References Wang, H. L¨u, G. Hu, Chaos-based cryptograph incorporated with S-box algebraic operation, Physics Letters A 318(4-5), 388 (2003). lgebraic operation, Physics Letters A 318(4-5), 388 (2003 17. Y. Wang, K.W. Wong, X. Liao, T. Xiang, A block cipher with dynamic S-boxes based on tent map Communications in Nonlinear Science and Numerical Simulation 14(7), 3089 (2009). ( ) 18. F. ¨Ozkaynak, A.B. ¨Ozer, A method for designing strong S-Boxes based on chaotic Lorenz system, Physics Letters A 374(36), 3733 (2010). 19. Y. Wang, P. Lei, K.W. Wong, A Method for Constructing Bijective S-Box with High Nonlinearity Based on Chaos and Optimization, International Journal of Bifurcation and Chaos 25(10), 1550127 (2015). 18 Reza Hoseini et al. 20. Y. Liu, X. Tong, J. Ma, Image encryption algorithm based on hyper-chaotic system and dynamic S-box, Multimedia Tools and Applications 75(13), 7739 (2016). 21. Y. Liu, J. Wang, J. Fan, L. Gong, Image encryption algorithm based on chaotic system and dynamic S-boxes composed of DNA sequences, Multimedia Tools and Applications 75(8), 4363 (2016). ( ), ( ) 22. ¨U. C¸avu¸so˘glu, S. Ka¸car, A. Zengin, I. Pehlivan, A novel hybrid encryption algorithm based on chaos and S-AES algorithm, Nonlinear Dynamics 92(4), 1745 (2018). 23. T. Ye, L. Zhimao, Chaotic S-box: six-dimensional fractional Lorenz–Duffing chaotic system and O-shaped path scrambling, Nonlinear Dynamics 94(3), 2115 (2018). 23. T. Ye, L. Zhimao, Chaotic S-box: six-dimensional fractional Lorenz–Duffing cha system and O-shaped path scrambling, Nonlinear Dynamics 94(3), 2115 (2018). system and O-shaped path scrambling, Nonlinear Dynamics 94(3), 2115 (2018). ( ) ( ) 24. D. Lambi´c, A new discrete-space chaotic map based on the multiplication of integer b d it li ti i S b d i N li D i 100(1) 699 (2020) 24. D. Lambi´c, A new discrete-space chaotic map based on the multiplication of integer numbers and its application in S-box design, Nonlinear Dynamics 100(1), 699 (2020). bi´c, A new discrete-space chaotic map based on the mu . D. Lambic, A new discrete-space chaotic map based on the multiplication of integer numbers and its application in S-box design, Nonlinear Dynamics 100(1), 699 (2020). 25. A.A. Abd el Latif, B. Abd-el Atty, M. Amin, A.M. Iliyasu, Quantum-inspired cascaded discrete-time quantum walks with induced chaotic dynamics and cryptographic applica- tions, Scientific Reports 10(1), 1 (2020). ( ) ( ) 26. X. Wang, Q. References Wang, A novel image encryption algorithm based on dynamic S-boxes constructed by chaos, Nonlinear Dynamics 75(3), 567 (2014). ( ) ( ) 27. X. Wang, A. Akgul, U. Cavusoglu, V.T. Pham, D. Vo Hoang, X.Q. Nguyen, A Chaotic System with Infinite Equilibria and Its S-Box Constructing Application, Applied Sciences 8(11), 1 (2018). ¨ 28. X. Wang, ¨U. C¸avu¸so˘glu, S. Kacar, A. Akgul, V.T. Pham, S. Jafari, F.E. Alsaadi, X.Q. Nguyen, S-Box Based Image Encryption Application Using a Chaotic System without Equilibrium, Applied Sciences 9(4), 2 (2019). ¨ 29. ¨U. C¸avu¸so˘glu, S. Ka¸car, I. Pehlivan, A. Zengin, Secure image encryption algorithm design using a novel chaos based S-Box, Chaos, Solitons & Fractals 95, 92 (2017). ¨ 30. ¨U. C¸avu¸so˘glu, A. Zengin, I. Pehlivan, S. Ka¸car, A novel approach for strong S-Box gen- eration algorithm design based on chaotic scaled Zhongtang system, Nonlinear Dynamics 87(2), 1081 (2017). ( ), ( ) 31. E. Al Solami, M. Ahmad, C. Volos, M.N. Doja, M.M.S. Beg, A New Hyperchaotic System-Based Design for Efficient Bijective Substitution-Boxes, Entropy 20(7), 525 (2018). ( ) 32. J.R. Dorfman, An introduction to chaos in nonequilibrium statistical mechanics (Cam- bridge University Press, 1999) g y , ) 33. M. Jafarizadeh, S. Behnia, S. Khorram, H. Nagshara, Hierarchy of chaotic maps with an invariant measure, Journal of statistical physics 104(5-6), 1013 (2001). ( ) ( ) 34. M. Jafarizadeh, S. Behnia, Hierarchy of one-and many-parameter families of elliptic chaotic maps of cn and sn types, Physics Letters A 310(2), 168 (2003). 35. T.W. Cusick, P. Stanica, Cryptographic Boolean functions and applications (Academic Press, 2017) , ) 36. J. Daemen and V. Rijmen, The design of Rijndael: AES-the advanced encryption stan- dard Springer Science & Business Media, (2013) 37. H.M. Heys, S.E. Tavares, Avalanche characteristics of substitution-permutation encryp- tion networks, IEEE Transactions on Computers 44(9), 1131 (1995). 38. A. Webster, S.E. Tavares, in Conference on the theory and application of cryptographic techniques (Springer, 1985), pp. 523–534 39. M. Matsui, Linear cryptanalysis method for DES cipher (Springer Science & Business Media, 1993) 40. E. Biham, A. Shamir, Differential cryptanalysis of DES-like cryptosystems, Journal of CRYPTOLOGY 4(1), 3 (1991). ( ), ( ) 41. Q. Lu, C. Zhu, X. Deng, An Efficient Image Encryption Scheme Based on the LSS Chaotic Map and Single S-Box, IEEE Access 8, 25664 (2020). 42. M. Farah, A. Farah, T. 12 10 9 47. X. Wang, N. Guan, 2D sine-logistic-tent-coupling map for image encryption, Journal of Ambient Intelligence and Humanized Computing, doi.org/10.1007/s12652-022-03794-0 (2022). 48. M. Yan, J. Xie, A conservative chaotic system with coexisting chaotic-like attractors and its application in image encryption, Journal of Control and Decision 0(0), 1 (2022). 49. J. Zheng, Q. Zeng, An image encryption algorithm using a dynamic S-box and chaotic maps, Applied Intelligence, doi.org/10.1007/s10489-022-03174-3 (2022). 52. G. Chen, Y. Mao, C.K. Chui, A symmetric image encryption scheme based on 3D chaotic cat maps, Chaos, Solitons & Fractals 21(3), 749 (2004). 46. ¨U. C¸avu¸so˘glu, A. Zengin, I. Pehlivan, S. Ka¸car, A novel approach for strong S-Box gen- eration algorithm design based on chaotic scaled Zhongtang system, Nonlinear dynamics 87(2), 1081 (2017). 51. S. Behnia, A. Akhshani, S. Ahadpour, H. Mahmodi, A. Akhavan, A fast chaotic en- cryption scheme based on piecewise nonlinear chaotic maps, Physics Letters A 366(4-5), 391 (2007). / / ( ) 50. N. Rani, S.R. Sharma, V. Mishra, Grayscale and colored image encryption model using a novel fused magic cube, Nonlinear Dynamics, doi.org/10.1007/s11071-022-07276-y (2022). 50. N. Rani, S.R. Sharma, V. Mishra, Grayscale and colored image encryption model using a novel fused magic cube, Nonlinear Dynamics, doi.org/10.1007/s11071-022-07276-y (2022). 51. S. Behnia, A. Akhshani, S. Ahadpour, H. Mahmodi, A. Akhavan, A fast chaotic en- cryption scheme based on piecewise nonlinear chaotic maps, Physics Letters A 366(4-5), 391 (2007). References Farah, An image encryption scheme based on a new hybrid chaotic map and optimized substitution box, Nonlinear Dynamics 99(4), 3041 (2020). 43. U. Hayat, N.A. Azam, A novel image encryption scheme based on an elliptic curve, Signal Processing 155, 391 (2019). 44. M. Khan, Z. Asghar, A novel construction of substitution box for image encryption applications with Gingerbreadman chaotic map and S8 permutation, Neural computing and applications 29(4), 993 (2018). ( ) ( ) 45. G. Chen, A novel heuristic method for obtaining S-boxes, Chaos, Solitons & Fractals 36(4) 1028 (2008) Construction of dynamical S-boxes based on: Image encryption approach 19 / / ( ) 50. N. Rani, S.R. Sharma, V. Mishra, Grayscale and colored image encryption model using a novel fused magic cube, Nonlinear Dynamics, doi.org/10.1007/s11071-022-07276-y (2022).
https://openalex.org/W2317379094
http://koreascience.or.kr:80/article/JAKO201528551642289.pdf
English
null
Value of PAX1 Methylation Analysis by MS-HRM in the Triage of Atypical Squamous Cells of Undetermined Significance
Asian Pacific journal of cancer prevention
2,015
cc-by
3,293
Introduction Optimal treatment option for women with ASCUS is not well established. Most low-grade lesions regress spontaneously. On the other hand, a small but significant proportion of cases with ASCUS may harbour cervical intraepithelial neoplasia (CIN)2-3, or even invasive carcinoma (Arbyn et al., 2004; Yarandi et al., 2009). Persistent infection with high-risk HPV (HR-HPV) is the most important etiologic factor for cervical cancer. In almost every invasive cervical cancer patient, HPV can be recognized (Park et al., 1996; Wang et al., 2013; Siriaunkgul et al., 2014). In consequence, HR-HPV testing has been widely used for the triage of ASCUS patients (Wentzensen et al., 2010). Although HR-HPV testing has a good sensitivity , but the high prevalence of transient human papillomavirus infections limits it’s specificity (Dehn et al., 2007). Thus, there is a need for other markers which have high sensitivity and specificity. Recently, paired boxed gene 1 (PAX1) was reported as a new methylation-silenced gene in cervical cancer (Lai et al., 2008). Recently, methylation- sensitive high-resolution melting (MS-HRM) analysis has been introduced as a sensitive and rapid technique for the quantification of DNA methylation (Wu et al., 2011; Dimitrakopoulos et al., 2012). Abstract Background: Detection of cervical high grade lesions in patients with atypical squamous cells of undetermined significance (ASCUS) is still a challenge. Our study tested the efficacy of the paired boxed gene 1 (PAX1) methylation analysis by methylation-sensitive high-resolution melting (MS-HRM) in the detection of high grade lesions in ASCUS and compared performance with the hybrid capture 2 (HC2) human papillomavirus (HPV) test. Materials and Methods: A total of 463 consecutive ASCUS women from primary screening were selected. Their cervical scrapings were collected and assessed by PAX1 methylation analysis (MS-HRM) and high-risk HPV-DNA test (HC2). All patients with ASCUS were admitted to colposcopy and cervical biopsies. The Chi- square test was used to test the differences of PAX1 methylation or HPV infection between groups. Results: The specificity, sensitivity, and accuracy for detecting CIN2 + lesions were: 95.6%, 82.4%, and 94.6%, respectively, for the PAX1 MS-HRM test; and 59.7%, 64.7%, and 60.0% for the HC2 HPV test. Conclusions: The PAX1 methylation analysis by MS-HRM demonstrated a better performance than the high-risk HPV-DNA test for the detection of high grade lesions (CIN2 +) in ASCUS cases. This approach could screen out the majority of low grade cases of ASCUS, and thus reduce the referral rate to colposcopy. Keywords: ASCUS - PAX1 - methylation - MS-HRM analysis - HPV Asian Pac J Cancer Prev, 16 (14), 5843-5846 Asian Pac J Cancer Prev, 16 (14), 5843-5846 In our study, we tested the efficacy of PAX1 methylation analysis by MS-HRM for the triage of patients with ASCUS, and compared its performance with HR- HPV testing. Value of PAX1 Methylation Analysis by MS-HRM in the Triage of Atypical Squamous Cells of Undetermined Significance Shi-Rong Li1, Zhen-Ming Wang1*, Yu-Hui Wang1, Xi-Bo Wang2, Jian-Qiang Zhao1, Hai-Bin Xue1, Fu-Guo Jiang1 Shi-Rong Li1, Zhen-Ming Wang1*, Yu-Hui Wang1, Xi-Bo Wang2, Jia Zhao1, Hai-Bin Xue1, Fu-Guo Jiang1 1Department of Clinical Laboratory, 2Department of Gynecology, Weifang City People’s Hospital, Weifang, China *For correspondence: wzm196@163.com DOI:http://dx.doi.org/10.7314/APJCP.2015.16.14.5843 Value of PAX1 Methylation Analysis by MS-HRM in the Triage of ASCUS DOI:http://dx.doi.org/10.7314/APJCP.2015.16.14.5843 Value of PAX1 Methylation Analysis by MS-HRM in the Triage of ASCUS RESEARCH ARTICLE Asian Pacific Journal of Cancer Prevention, Vol 16, 2015 5843 Asian Pacific Journal of Cancer Prevention, Vol 16, 2015 5843 1Department of Clinical Laboratory, 2Department of Gynecology, Weifang City People’s Hospital, Weifang, China *For correspondence: wzm196@163.com Statistical analysis 75.0 100.0 25.0 20.3 10.1 6.3 75.0 100.0 25.0 20.3 10.1 6.3 Figure 1. PAX1 MS-HRM Curves Representing the Methylation Level of Two Samples. (From left to right , the curve representing the methylation status:0, 10%, sample1, 30%, 50%, 80%, sample2 , 100%) SPSS 21.0 software was used for statistical analyses. Chi-square test was used to test the differences of PAX1 methylation or HPV infection between groups. p values less than 0.05 was considered statistically significant. Using PAX1 methylation and high-risk HPV test results make receiver operating characteristic curve, calculating the area under the curve. 46.8 56.3 46.8 56.3 Table 1. The Status of PAX1 Methylation and HR-HPV Infection in Patients With ASCU 46 8 56 3 46 8 56 3 able 1. The Status of PAX1 Methylation and HR-HPV 0 25.0 50.0 atment atment urrence emission 31.3 54.2 46.8 56.3 31.3 23.7 38.0 31.3 0 25.0 50.0 atment atment urrence emission 31.3 54.2 46.8 56.3 31.3 23.7 38.0 31.3 Table 1. The Status of PAX1 Methylation and HR-HPV Infection in Patients With ASCU Histopathology n PAX1 MS-HRM HR-HPV inflammation 198 0.0% (0/198) 17.2% (34/198) CIN1 231 8.2% (19/231) 60.2% (139/231) CIN2 19 73.7% (14/19) 57.9% (11/19) CIN3 13 92.3% (12/13) 69.2% (9/13) cervical carcinoma 2 100% (2/2) 100% (2/2) CIN2 + 34 82.4% (28/34)*a 64.7% (22/34) *b CIN2 – 429 4.4% (19/429) 40.3% (173/429) (CIN1 or inflammation) *comparing CIN2 + group with CIN2 – group, aP=0.0000; bP=0.0056 The status of PAX1 methylation and HR-HPV infection in Patients With ASCUS Quantitative assessment of PAX1 methylation by MS- HRM according to disease severity is shown in Table 1. At the cutoff value of 10%, 100% of the cervical carcinoma, 92.3% of the CIN3 and 73.7% of CIN2 specimens were positive for PAX1 methylation compared with 0% for the inflammation, 8.2% for CIN1 specimens (Table 1). At this cut off value, specimens positive for CIN2 + were more significant than CIN2 – specimens (P <0.001). As a comparison, the HC2 HPV test result was positive in 100% of the cervical carcinoma, 69.2% of the CIN3 and 57.9% of CIN2 specimens. Noteworthily, HPV was present in 60.2% and 17.2% of the benign conditions of CIN1 and the inflammation, respectively, but PAX1 methylation was present in only 8.2% and 0%, respectively (Table 1). The detection efficiency of PAX1 methylation and HR-HPV infection for CIN2+ The detection efficiency of PAX1 methylation and HR-HPV infection for CIN2+ We compared the performance of MS-HRM for PAX1 methylation detecting to the HR-HPV test in the detection of CIN2 + by receiver operating characteristic curve (Figure 2). The area under the curve of PAX1 methylation and HR-HPV test were 0.890 and 0.622 (p<0.05). The specificity and sensitivity of PAX1 MS-HRM assay were 75.0 100.0 25.0 20.3 10.1 6.3 75.0 100.0 25.0 20.3 10.1 6.3 Figure 1. PAX1 MS-HRM Curves Representing the Methylation Level of Two Samples. (From left to right , the curve representing the methylation status:0, 10%, sample1, 30%, 50%, 80%, sample2 , 100%) HPV test DNA in a background of unmethylated DNA. At a cutoff value of 10%, In 47 out of 463 ASCUS scrapings, PAX1 methylation was identified. Figure 1 represents the PAX1 MS-HRM results from two representative samples. DNA in a background of unmethylated DNA. At a cutoff value of 10%, In 47 out of 463 ASCUS scrapings, PAX1 methylation was identified. Figure 1 represents the PAX1 MS-HRM results from two representative samples. High-risk HPV-DNA (HR-HPV) infection of the cervical sample was detected by the HC2 test ((Digene Corp., Silver Spring, MD, USA). The specimen DNA was denatured and hybridized with the RNA probes which directed against 13 high-risk human papillomavirus. The hybrids were captured by hybrid-specific antibodies and detected by second antibody and chemiluminescence. MS-HRM analysis of PAX1 DNA of cervical scrapings was extracted using the QIAamp DNA mini kit (Qiagen) and bisulfite converted using the EpiTect Bisulfite Kit (Qiagen) according to the instructions of manufacturer. The bisulfite converted DNA was PCR-amplified with the specific primers (forward, 5’-CGGGAATTAATGAGTTGTTAATT-3’; reverse, 5’-AAACCCAAAATAAACTTCATCC-3’) in a HRM Epigenetics PCR Kit (Qiagen). DNA extracted from normal human peripheral blood leukocytes was used as unmethylated control DNA. DNA of human peripheral blood leukocytes treated with M.SssI methyltransferase served as the fully methylated control. Standards of methylation were constructed by diluting 100% methylated bisulfite-modified control DNA in a pool of normal bisulfite-modified DNA at ratios of 10, 30, 50, 80 and 100% . MS-HRM was conducted in a 25μl volume containing 2μl of modified template DNA, 12.5μl 2×EpiTect HRM PCR Master Mix (HRM Epigenetics PCR Kit, Qiagen), 1.9μl of each primer (10μM), and 6.7μl of PCR grade water. The PCR was subjected to an incubation at 95℃ for 10minutes, followed by 40 cycles of 95℃ for 10 s, annealing for 10 s from 62℃ to 50℃ (decrease 0.5℃per cycle), extension at 72℃ for 10 s; followed by an HRM of 95°C for 1 min, and 40°C for 1 min, 65°C for 15 s, and continuous acquisition to 95°C at one acquisition per 0.02°C, and then detected using the Roche LightCycler (Roche). HRM data were analyzed by use of the dedicated HRM software (LightCycler 480 Gene Scanning Software). All experiments were performed in duplicate. This study was approved by the institutional review board of the Weifang city people’s hospital in Shan dong province, PRC, and informed consents were obtained from all participants. Patient samples A total of 463 consecutive patients (mean age 37 years, range 20-58) who participated in the cervical cytological screening by Thinprep cytologic test (TCT) technique in the Weifang city people’s hospital from January 2013 to December 2014, and with an ASCUS diagnosis based on the cytological screening, were selected. The cytological screening was performed using The 2001 Bethesda System (TBS). The exclusion criteria included current pregnancy, history of cancer at other sites, and history of immune compromise diseases. The cervical scrapings of 463 ASC-US cases were collected and evaluated by using PAX1 methylation analysis (MS-HRM) and high-risk HPV-DNA test (HC2). All patients with ASCUS were admitted for colposcopy and cervical biopsy. Cervical biopsies were cut and stained with hematoxylin and eosin, read by a pathologist, and confirmed by a second independent reading result. 5843 Asian Pacific Journal of Cancer Prevention, Vol 16, 2015 5843 Shi-Rong Li et al Shi-Rong Li et al The sensitivity of the PAX1 MS-HRM assay The sensitivity of the PAX1 MS-HRM was assessed by using the consistency of normalized melting profiles derived from samples with different ratios of methylated template: 10, 30, 50, 80 and 100% methylated. PAX1 MS-HRM was able to reproducibly detect 10% methylated Asian Pacific Journal of Cancer Prevention, Vol 16, 2015 5844 DOI:http://dx.doi.org/10.7314/APJCP.2015.16.14.5843 h l i A l i b MS HRM i h T i f ASCUS DOI:http://dx.doi.org/10.7314/APJCP.2015.16.14.5843 DOI:http://dx.doi.org/10.7314/APJCP.2015.16.14.5843 Value of PAX1 Methylation Analysis by MS-HRM in the Triage of ASCUS detection by HPV test. The NPV, PPV , and accuracy of PAX1 MS-HRM assay are all superior to those of the HC2 HR-HPV test. The MS-HRM for PAX1 methylation testing can screens out the majority high grade lesions cases of ASCUS , and may avoid unnecessary invasive inspection. Because of the influence of prevalence rate, the PPV of MS-HRM for PAX1 methylation testing is not high in this study. i DOI:http://dx.doi.org/10.7314/APJCP.2015.16.14.5843 Value of PAX1 Methylation Analysis by MS-HRM in the Triage of ASCUS detection by HPV test. The NPV, PPV , and accuracy of PAX1 MS-HRM assay are all superior to those of the HC2 HR-HPV test. The MS-HRM for PAX1 methylation testing can screens out the majority high grade lesions cases of ASCUS , and may avoid unnecessary invasive inspection. Because of the influence of prevalence rate, the PPV of MS-HRM for PAX1 methylation testing is not high in this study. i Value o 95.6% and 82.4%, compared with 59.7% and 64.7%, respectively, for detection by HR-HPV test. The negative predictive value (NPV), positive predictive value (PPV), and accuracy of PAX1 MS-HRM assay are all superior to those of the HR-HPV test. The NPV, PPV, and accuracy for detecting CIN2 + lesions were 98.6%, 59.6%, and 94.6%, respectively for PAX1 MS-HRM test; 95 5% 11 3% and Figure 2. The Receiver Operating Characteristic Curve of PAX1 Methylation and HR-HPV Test g y In 2007, MS-HRM analysis was first described by Wojdacz et al for high throughput and sensitive assessment of DNA methylation (Wojdacz et al., 2007). The PCR amplification products generated from bisulfite- modified DNA templates with different contents of methylcytosine show differences in melting temperature, which can be analyzed by a fluorometer thermal cycler. Quantification was implemented by a standard curve generated with serial dilutions of methylated DNA. MS-HRM assay has important advantage over the Methylation-Specific PCR. Discussion Organized cytological screening resulted in a significant decrease in the proportion of women with cervical cancer, and accordingly, the number of ASCUS diagnoses has a marked increase. A significant number of cervical cancers occur after a normal cervical smear following ASCUS (Andrae et al., 1999). In this study, we screens out 34 high grade lesions cases in 463 women with ASCUS. A important problem of cytology screening is the limited sensitivity (Andersson et al., 2005; Naucler et al., 2009). Therefore, if a first cytology screening shows a diagnosis with ASCUS, a second normal cervical smear is not suitable for the exclusion of precancerous cervical lesion. Another strategy is to refer these patients to colposcopy-directed biopsy, but it is costly and invasive. In summary, the MS-HRM for PAX1 methylation testing have a better performance than HR-HPV test in the detection of high grade lesions CIN2 + in ASCUS. This approach could screens out the majority high grade lesions cases of ASCUS, also could reduce the referral rate to colposcopy. Our results may provide a new triage method for women with ASCUS. Because the number of cases particulatrly in the CIN3 and carcinoma groups is small, further studies in larger screening populations are warranted to prove its efficacy in screening ASCUS. Acknowledgements This study was supported by grants from the Weifang Municipal Science and Technology Development Program (No. 201302014). Adding the HR-HPV test is a proposed method to secondary cytology screening. Triage study have shown that HPV test is more sensitive than repeated cytology smear for determining high grade lesions in women with ASCUS (Silverloo et al., 2009). But the low PPV (37%) and specificity (31%)even lead to a worse situation, because they make more patients undergoing unnecessary referrals (Nieh et al., 2005). The sensitivity of the PAX1 MS-HRM assay MS-HRM can scan all of the CpGs in the target sequence. It also resolves heterogeneous methylation (Candiloro et al., 2011), allows gene amplification and methylation analysis in one closed tube. Closed-tube approaches have important advantage, as they minimize the risk of cross contamination. Moreover, MS-HRM experiment doesn’t require a reference assay for normalization. A new Quantitative Methylation-Specific PCR (QMSP) method was used to analyze the gene methylation, but it require a reference assay and expensive fluorescence probe (Huang et al., 2010). All of these advantages make MS-HRM assay rapid, reliable and cost-effective (Wojdacz et al., 2012). Figure 2. The Receiver Operating Characteristic Curve of PAX1 Methylation and HR-HPV Test 95.6% and 82.4%, compared with 59.7% and 64.7%, respectively, for detection by HR-HPV test. The negative predictive value (NPV), positive predictive value (PPV), and accuracy of PAX1 MS-HRM assay are all superior to those of the HR-HPV test. The NPV, PPV, and accuracy for detecting CIN2 + lesions were 98.6%, 59.6%, and 94.6%, respectively, for PAX1 MS-HRM test; 95.5%, 11.3%, and 60.0%, respectively, for HR-HPV test. It also resolves heterogeneous methylation (Candiloro et al., 2011), allows gene amplification and methylation analysis in one closed tube. Closed-tube approaches have important advantage, as they minimize the risk of cross contamination. Moreover, MS-HRM experiment doesn’t require a reference assay for normalization. A new Quantitative Methylation-Specific PCR (QMSP) method was used to analyze the gene methylation, but it require a reference assay and expensive fluorescence probe (Huang et al., 2010). All of these advantages make MS-HRM assay rapid, reliable and cost-effective (Wojdacz et al., 2012). Asian Pacific Journal of Cancer Prevention, Vol 16, 2015 5845 References Andersson S, Dillner L, Elfgren K, et al (2005). A comparison of the human papillomavirus test and Papanicolaou smear as a second screening method for women with minor cytological abnormalities. Acta Obstet Gynecol Scand, 84, 996-1000. This study demonstrates that MS-HRM for PAX1 methylation test is better than HC2 HPV test in the triage of ASCUS. Indeed, it is the most difficult for pathologists to reproduce among all CIN2 diagnosis, because the diagnosis of CIN2 has been a gray area (Carreon et al., 2007). For a long time, CIN2 was considered an intermediate entity that may be under called as CIN1 or over called as CIN3. In the detection of CIN2 + , the specificity and sensitivity of PAX1 MS-HRM assay were 95.6% and 82.4%, compared with 59.7% and 64.7%, for Andrae B, Smith P (1999). Clinical impact of quality assurance in an organized cervical screening program. Acta Obstet Gynecol Scand, 78, 429-35. Arbyn M, Buntinx F, Van Ranst M, et al (2004). Virologic versus cytologic triage of women with equivocal Pap smears: a meta-analysis of the accuracy to detect high-grade intraepithelial neoplasia. J Natl Cancer Inst, 96, 280-93. Candiloro IL, Mikeska T, Dobrovic A (2011). Assessing combined methylation-sensitive high resolution melting Shi-Rong Li et al Shi-Rong Li et al and pyrosequencing for the analysis of heterogeneous DNA methylation. Epigenetics, 6, 500-7. Carreon JD, Sherman ME, Guillen D, et al (2007). CIN2 is a much less reproducible and less valid diagnosis than CIN3: results from a histological review of population-based cervical samples. Int J Gynecol Pathol, 26, 441-6. Dehn D, Torkko KC, Shroyer KR (2007). Human papillomavirus testing and molecular markers of cervical dysplasia and carcinoma. Cancer, 111, 1-14. Dimitrakopoulos L, Vorkas PA, Georgoulias V, Lianidou ES (2012). A closed-tube methylation-sensitive high resolution melting assay (MS-HRMA) for the semi-quantitative determination of CST6 promoter methylation in clinical samples. BMC Cancer, 12, 486. Huang TH, Lai HC, Liu HW, et al (2010). Quantitative analysis of methylation status of the PAX1 gene for detection of cervical cancer. Int J Gynecol Cancer, 20, 513-9.i Lai HC, Lin YW, Huang TH, et al (2008). Identification of novel DNA methylation markers in cervical cancer. Int J Cancer, 123, 161-7.i Naucler P, Ryd W, Tornberg S, et al (2009). Efficacy of HPV DNA testing with cytology triage and/or repeat HPV DNA testing in primary cervical cancer screening. J Natl Cancer Inst, 101, 88-99. References Nieh S, Chen SF, Chu TY, et al (2005). Is p16 (INK4A) expression more useful than human papillomavirus test to determine the outcome of atypical squamous cells of undetermined significance-categorized Pap smear? A comparative analysis using abnormal cervical smears with follow-up biopsies. Gynecol Oncol, 97, 35-40. Park JS (1996). The role of HPV DNA testing in cervical neoplasia. J Obstet Gynaecol Res, 22, 611-20. Silverloo I, Andrae B, Wilander E (2009). Value of high-risk HPV-DNA testing in the triage of ASCUS. Acta Obstet Gynecol Scand, 88, 1006-10. Siriaunkgul S, Settakorn J, Sukpan K, et al (2014). Population- based cervical cancer screening using high-risk HPV DNA test and liquid-based cytology in northern Thailand. Asian Pac J Cancer Prev, 15, 6837-42. Wang JL, Yang YZ, Dong WW, et al (2013). Application of human papillomavirus in screening for cervical cancer and precancerous lesions. Asian Pac J Cancer Prev, 14, 2979-82. Wentzensen N, Wilson LE, Wheeler CM, et al (2010). Hierarchical clustering of human papilloma virus genotype patterns in the ASCUS-LSIL triage study. Cancer Res, 70, 8578-86. Wojdacz TK, Dobrovic A (2007). Methylation-sensitive high resolution melting (MS-HRM): a new approach for sensitive and high-throughput assessment of methylation. Nucleic Acids Res, 35, 41. Wojdacz TK (2012). Methylation-sensitive high-resolution melting in the context of legislative requirements for validation of analytical procedures for diagnostic applications. Expert Rev Mol Diagn, 12, 39-47. Wu W, Zhang J, Yang H, Shao Y, Yu B (2011). Examination of AKAP12 promoter methylation in skin cancer using methylation-sensitive high-resolution melting analysis. Clin Exp Dermatol, 36, 381-5. Yarandi F, Shojaei H, Eftekhar Z, Izadi-Mood N (2009). Comparison of three management strategies for patients with atypical squamous cells of undetermined significance, after six months delay: a three-year experience in an Iranian university hospital. Aust N Z J Obstet Gynaecol, 49, 207-10. Asian Pacific Journal of Cancer Prevention, Vol 16, 2015 5846
W4380877267.txt
https://journals.wlb-stuttgart.de/ojs/index.php/wfr/article/download/6618/6504
de
Rezension von: Mengele, Hans-Peter, Wer zu Späth kommt ...
Württembergisch Franken
2,023
cc-by
509
356 Neue Bücher Adenauer in der Bundesrepublik Deutschland“ (Friedrich P. Kahlenberg, S. 345-358) ge- handelt wird: Es wird deutlich, daß die „Macht der Bilder“ sowohl ein historisches als auch ein aktuelles Problem ist. Als letzter Beitrag, einziger des abschließenden Unterkapitels „Pro memoria“, wurde unter „Versiegende Quellen. Der unbemerkte Untergang kirchlicher Quellen“ von Johann Michael Fritz aufgenommen, ein recht persönlich gehaltene Betrachtung über die sich dem Titel aus der aktuellen, zumindest bedenklich den Erhalt kirchlicher zu nennenden Situation ergebenden Probleme für Kunstgegenstände als Ensembles, die dem aus Wechselspiel-von „Überforderung der staatlichen Denkmalpflege, den finanziellen Nöten von Staat und Kirche, der immer stärker werdenden Entchristlichung bis hin zu dem weitgehenden Desinte- resse der Theologen“ resultieren. Ein Bibliographie des Jubilars, für die jeder an der südwestdeutschen Landesgeschichte Interessierte sicherlich sehr ist, beschließt eine Festschrift, dankbar die gleichzeitig ihren Empfänger ehrt (und sicherlich auch erfreut) und zudem als thematisch überzeugend stringent aufgebautes Werk für sich steht. Hans-Peter Wer Mengele, Rolle in den Umbruch-Jahren, zu Späth kommt... Baden-Württembergs außenpolitische Stuttgart; Tübingen (Silberburg) 1995. 359 S. Hier schreibt ein Autor, der aus zen G. Lubich erster Hand berichtet: Mengele war ab 1982 mit einer kur- Unterbrechung Persönlicher Referent und Berater punkte lagen dabei im internationalen Ton schildert der Autor die Lothar Späth. Seine Schwer- Bereich. In nüchternem außenpolitischen Aktivitäten des früheren Ministerpräsidenten, dessen besonderes Interesse der sozialistischen So erscheint Lothar von und außenwirtschaftlichen Späth nochmals Staatenwelt und den Ländern Asiens in der Rolle, die ihm die liebste war: galt. als innovativer Landespolitiker, für den globales Denken und regionale Wirtschaftspolitik zwei Seiten einer Medaille waren. Mit erstaunlicher Detailgenauigkeit beschreibt der Autor die außenpolitischen Stationen und das außenwirtschaftliche Engagement seines ehemaligen Vorgesetzten, doch bleibt der Hauptakteur auf merkwürdige Weise blaß, denn über den Menschen Lothar Späth erfährt man bei alledem so gut wie nichts. Nichts darüber, ob Späth wirklich die Abvon seinem Bonner Thron zu sicht hatte, Helmut Kohl vorging, als sich die dunklen Wolken der stoßen, nichts darüber, was in Späth „Segeltörn-Affäre“ über ihm zusammenbrauten. Wer Enthüllungen erwartet, wird enttäuscht. Die Diskretion des Verfassers läßt wenig Raum für kritische Wertungen. Dazu werden, wenn überhaupt, Außenstehende bemüht, beispiels- weise Manfred Rommel, der in seiner mild-ironischen Art einmal sagte: „Der Wind, der über Asien weht, er flüstert leise 'Lothar Späth’“. Der das Buch durchziehende selten eine eigene Qualität, zum Berichterstattungston verleiht der Innenansicht der Macht nur etwa in den Jahren 1989/90, in denen der Verfasser plötzlich Zeugen von Siechtum und Untergang des anderen deutschen Staates wird. In diesem so Abschnitt merkt man dem Autor endlich Emotionen langerhoffte Prickeln ein. Nun geschieht, was an, und so der Leser stellt sich beim Leser das eigentlich erwartete: Aus den Marionetten werden Menschen. Das Buch, obzwar nicht einlöst, was glänzend geschrieben, muß sich der Titel den Vorwurf gefallen lassen, daß es reißerisch verspricht. Nicht die Person Lothar Späth steht im Mittelpunkt, sondern das politische Geschehen seiner Amtszeit. Wer die Tagespresse jener Jahre verfolgt hat, wird an vieles zurückerinnert, erfährt aber letztendlich wenig Erhellen- des oder gar Aufregendes. Die Stärke des Buches ist somit gleichzeitig seine Schwäche: Es „referiert“ zu viel. H. Kohl
https://openalex.org/W4386727401
https://www.frontiersin.org/articles/10.3389/frsfm.2023.1260211/pdf?isPublishedV2=False
English
null
Mini review: recent advances in biosurfactant-based association colloids–formation of microemulsions
Frontiers in soft matter
2,023
cc-by
7,369
TYPE Mini Review PUBLISHED 14 September 2023 DOI 10.3389/frsfm.2023.1260211 TYPE Mini Review PUBLISHED 14 September 2023 DOI 10.3389/frsfm.2023.1260211 KEYWORDS biosurfactant, rhamnolipid, saponin, microemulsion, droplets, microstructure, scattering techniques frontiersin.org OPEN ACCESS OPEN ACCESS EDITED BY Francisco Monroy, Complutense University of Madrid, Spain REVIEWED BY Sylvain Prévost, Institut Laue-Langevin, France Ramon Pons, Spanish National Research Council (CSIC), Spain *CORRESPONDENCE Thomas Hellweg, thomas.hellweg@uni-bielefeld.de Thomas Sottmann, thomas.sottmann@ipc.uni-stuttgart.de RECEIVED 17 July 2023 ACCEPTED 01 September 2023 PUBLISHED 14 September 2023 CITATION Hellweg T, Oberdisse J and Sottmann T (2023), Mini review: recent advances in biosurfactant-based association colloids–formation of microemulsions. Front. Soft Matter 3:1260211. doi: 10.3389/frsfm.2023.1260211 EDITED BY Francisco Monroy, Complutense University of Madrid, Spain REVIEWED BY Sylvain Prévost, Institut Laue-Langevin, France Ramon Pons, Spanish National Research Council (CSIC), Spain Thomas Hellweg1*, Julian Oberdisse2 and Thomas Sottmann3* 1Fakultät f. Chemie, Physikalische und Biophysikalische Chemie, Universität Bielefeld, Bielefeld, Germany, 2Laboratoire Charles Coulomb (L2C), Centre National de la Recherche Scientifique, Université de Montpellier, Montpellier, France, 3Fakultät Chemie, Institut f. Physikalische Chemie, Universität Stuttgart, Stuttgart, Germany In the context of a more sustainable economy, bio-surfactants become increasingly important, due to their independence of petrol-based chemistry, their usually mild synthesis conditions, and in certain cases their pharmacological activity. We have recently discussed self-assembly studies in binary systems of bio-surfactants of microbial origin, or saponins extracted from plants (Hellweg et al., Frontiers in Soft Matter, 2023, 2). In the present review, we focus on the formation of microemulsions based on these molecules. We review the formation and structure of microemulsion systems formed by oil, water, and biosurfactants, with a particular focus on Quillaja saponins and rhamnolipids. CITATION Hellweg T, Oberdisse J and Sottmann T (2023), Mini review: recent advances in biosurfactant-based association colloids–formation of microemulsions. Front. Soft Matter 3:1260211. doi: 10.3389/frsfm.2023.1260211 COPYRIGHT © 2023 Hellweg, Oberdisse and Sottmann. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Soft Matter 1 Introduction 2002; Kitamoto et al. 2002). Surface activity is the main characteristic of surfactant molecules, which leads to a decrease of the air-water interfacial tension. For example, the high surface activity of microbial surfactin has been known since the 1960s (Fracchia et al., 2012; Lewinska et al., 2022). On the other hand, the self-assembly properties of biosurfactants lead to the formation of supramolecular aggregates in selective solvents. The self-assembly of such molecules has been reviewed by us (Hellweg et al., 2023), and here we wish to extend this review to the formation of microemulsions with biosurfactants. The size and shape of the structures formed on the nanoscale depend mainly on the surfactants molecular architecture: In particular, the spontaneous curvature of the hydrophilic-hydrophobic interface is the result of opposing forces acting on each moiety: the attractive hydrophobic effect for the tails, and the head group repulsion for the hydrophilic head groups. Ideally, molecules adopt an amphiphilic film curvature corresponding to these interactions, but other constraints like the available surfaces or volumes of each phase may then not be met, and energetic compromises need to be adopted. There is a cost for deviations from the spontaneous curvature, called the bending energy, and the most successful model has been proposed by Helfrich exactly 50 years ago (Helfrich, 1973). Available shapes of the oil or water domains in microemulsions are thus subjected to these free energy contributions. q p Applications of Quillaja saponin-based microemulsions have been proposed in the literature. These mixtures, however, do not always form thermodynamically stable microemulsions. Often, a strong energy input like high pressure homogenization is necessary to form emulsions with relatively unstable droplets in the micro to millimeter range. And sometimes more stable nanoemulsions are formed, where stability before coalescence lasts longer due to the small nanometric droplet size. In spite of the lack of thermodynamic stability, we discuss three articles here which pave the way to “all natural” microemulsions for food and drug incorporation. Schober et al. studied the efficiency of Quillaja saponins to emulsify limonene and alkanes by determining the droplet diameters of mostly bimodal emulsions with a particle sizer (Schober et al., 2017). These studies showed that Quillaja saponins have a significantly higher emulsifying power than various conventional Tween surfactants under the same conditions. 1 Introduction The headgroup area of 1.37 nm2 was found to be slightly larger than that determined by Wojciechowski (Wojciechowski, 2013) and suggests a lay-on configuration of Quillaja saponins in the internal interface of oil-in-water emulsions, where the hydrophobic triterpenoid rings of a saponin molecule reside in the interface and the hydrophilic glucoside tails protrude into the water. Another example is the incorporation of vitamin E, which refers to a fat-soluble group of molecules which have anti-oxydant properties. The transport in a drug carrier is also thought to increase the bioavailability on the intestinal sites. In the form of the more stable vitamin E acetate, which can then be broken down to vitamin E during digestion, the uptake of this lipophilic molecule into nano-emulsion droplets (nonequilibrium emulsion) containing food-grade oils (triglyceride) and surfactants, among which “(marketing) label friendly” saponins, has been studied by Yang and McClements (Yang and McClements, 2013). The formation of nano-emulsion droplets has been shown to be less efficient with saponins in comparison to different Tweens, in the sense that they remain rather macroscopic, with a strong impact on both stability and optical appearance (Mayer et al., 2013). In contrast to this, below the formation of a vitamin E-containing In this short review we address the formation of microemulsions of saponins and microbial surfactants, in particular rhamnolipids. To tune the curvature of biosurfactant-containing amphiphilic films, synthetic surfactants or hydrophobic alcohols were often added. We also briefly discuss emulsions stabilized by biosurfactants, as they could be a useful application of the more readily degradable biosurfactants, given the limited shelf life of most industrial products. Saponin-based microemulsions are exemplarily discussed in Section 2 for Quillaja saponins, while we have focused on rhamnolipids in the third section. In Section 4, some microemulsions stabilized by other biosurfactants are discussed. 1 Introduction The size and shape of the structures formed on the nanoscale depend mainly on the surfactants molecular architecture: In particular, the spontaneous curvature of the hydrophilic-hydrophobic interface is the result of opposing forces acting on each moiety: the attractive hydrophobic effect for the tails, and the head group repulsion for the hydrophilic head groups. Ideally, molecules adopt an amphiphilic film curvature corresponding to these interactions, but other constraints like the available surfaces or volumes of each phase may then not be met, and energetic compromises need to be adopted. There is a cost for deviations from the spontaneous curvature, called the bending energy, and the most successful model has been proposed by Helfrich exactly 50 years ago (Helfrich, 1973). Available shapes of the oil or water domains in microemulsions are thus subjected to these free energy contributions. Kenne, 1999; Wojciechowski, 2013; Wojciechowski et al., 2014; Góral and Wojciechowski, 2020). These saponins have a large structural variety, which impacts their surface activity (Kezwon and Wojciechowski, 2014). In water, they form micelles above a certain concentration, the critical micelle concentration (CMC) (Dargel et al., 2019; Geisler et al., 2019). Nowadays, they are targeted as molecules with potential cosmetic or pharmacological activity (Herzog et al., 2020; Wojciechowski et al., 2021). Regarding the separation and purification of surface-active saponins from multicomponent plant extracts, Obasi et al., (2017) reported an approach reminiscent of the three-phase extraction procedure proposed for the purification of nonionic alkyl polyglycol ethers, in which water- and oil-soluble impurities were extracted into the water and oil excess phases of a Winsor III system (Schubert et al., 1990). Similarly, saponins from the defatted root extract of Securidaca longipedunculata were first dispersed in water to self- assemble into micelles. They report that the addition of ethyl ether initially swells these micelles, and upon further addition causes a phase inversion to a Winsor II type system. The purified saponin was then extracted from the upper ethyl ether-rich phase via three additional purification cycles, while polar impurities were separated in the lower aqueous phase. biosurfactants are their (often) lower toxicity and increased biocompatibility, and sometimes their pharmacological activity, which may be antiviral (Bailly and Vergoten, 2020; Bailly and Vergoten, 2020; Bailly and Vergoten, 2020) or antimicrobial, leading to growing interest in their large-scale production (Kitamoto et al., 2002; Kitamoto et al. 2002; Kitamoto et al. 2002). (Kitamoto et al., 2002; Kitamoto et al. Frontiers in Soft Matter frontiersin.org 1 Introduction Since microemulsions are thermodynamically stable, they are spontaneously formed upon mixing water, oil, surfactant and in some cases a co-surfactant (Hoar and Schulman, 1943; Strey et al., 1994). Microemulsions are clear, isotropic, and homogeneous liquids exhibiting a nanostructure of polar (water) and nonpolar (oil) domains separated by a surfactant layer. The formation of such nanostructures is governed by the combination of constraints of available surface to be covered by surfactant, the volume of each phase, and the interfacial curvature. Resulting nanostructures range from oil-in-water droplets to inverse water-in-oil droplets. The continuous path between both passes through bicontinuous microemulsions and is called the Shinoda cut in the phase diagram (Shinoda and Kunieda, 1973; Olsson et al., 1986). Applications of microemulsions are in particular related to the solubilization of hydrophobic molecules (detergency, drug delivery, . . . ) (Schwuger et al., 1995; Solans and Kunieda, 1997). At present most microemulsions are made by using synthetic surfactants based on petrolchemistry. Biosurfactants, i.e., surfactants from natural resources, represent an attractive alternative, in particular in the context of sustainability. These molecules possess a high degree of chemical diversity, and many different classes exist according the presence of chemical groups, like phospholipids, glycolipids (in particular rhamnolipids discussed here), or lipopeptides (Bezerra et al., 2018). These molecules are usually produced under mild synthesis conditions in bioreactors for microbial biosurfactants (yeast or bacteria) or by extraction for plant biosurfactants (plant saponins). Other advantages of Frontiers in Soft Matter 01 frontiersin.org Hellweg et al. 10.3389/frsfm.2023.1260211 biosurfactants are their (often) lower toxicity and increased biocompatibility, and sometimes their pharmacological activity, which may be antiviral (Bailly and Vergoten, 2020; Bailly and Vergoten, 2020; Bailly and Vergoten, 2020) or antimicrobial, leading to growing interest in their large-scale production (Kitamoto et al., 2002; Kitamoto et al. 2002; Kitamoto et al. 2002). Surface activity is the main characteristic of surfactant molecules, which leads to a decrease of the air-water interfacial tension. For example, the high surface activity of microbial surfactin has been known since the 1960s (Fracchia et al., 2012; Lewinska et al., 2022). On the other hand, the self-assembly properties of biosurfactants lead to the formation of supramolecular aggregates in selective solvents. The self-assembly of such molecules has been reviewed by us (Hellweg et al., 2023), and here we wish to extend this review to the formation of microemulsions with biosurfactants. FIGURE 1 FIGURE 1 The phase boundaries feature the “fish shape” well-known for classical microemulsion systems (rotated by 90° due to the choice of axes), with the phase inversion from Winsor I to Winsor II induced by an increase in the NaCl concentration. Figure adapted from ref. (Nguyen et al., 2008) with permission from Elsevier. FIGURE 1 The phase boundaries feature the “fish shape” well-known for classical microemulsion systems (rotated by 90° due to the choice of axes), with the phase inversion from Winsor I to Winsor II induced by an increase in the NaCl concentration. Figure adapted from ref. (Nguyen et al., 2008) with permission from Elsevier. made of water/NaCl (brine), oil and the same 50/50 mixture of mono- and di-rhamnolipids at 23°C (Nguyen et al., 2008). For toluene as oil, they showed very nicely that at neutral pH the system undergoes a phase inversion from Winsor I to Winsor II when the NaCl concentration is increased and even passes a Winsor III region–where the microemulsion phase coexists with excess water and oil phases–at intermediate salt concentrations. Here, i.e., at the optimum salt concentration, the oil/water interfacial tension passes through a pronounced minimum of 10–5 Nm-1, as is the case for classical microemulsions (Salager et al., 1979; Sottmann and Strey, 1996). Fish-shaped phase boundaries, as known, e.g., from microemulsion systems with non-ionic surfactants (Kahlweit and Strey, 1985), are found when the surfactant concentration is plotted logarithmically against the salt concentration (Figure 1). With the aim of solubilizing hydrophobic oils such as limonene and diesel, they used mixtures of rhamnolipid and sodium bis(2-ethylhexyl) sulfosuccinate (AOT) in a follow-up work and investigated the phase behavior and oil/water interfacial tension of symmetric microemulsions (Nguyen and Sabatini, 2008). When using an equimolar mixture of the two surfactants, the oil/ water interfacial tension was likewise found to pass through a pronounced minimum in the Winsor III region with increasing NaCl concentration. The addition of oleyl alcohol (referred to as a lipophilic linker) to the surfactant mixture resulted in a further decrease in the minimum. As a measure of the tendency to form o/ w- or w/o-microemulsions, the characteristic curvature Cc (Acosta et al., 2008) of the rhamnolipid mixture used was determined to be Cc = −1.41. When compared to the Cc of AOT (Cc(AOT) = -0.92), this means that the rhamnolipid blend is slightly more hydrophilic than AOT. 3.1 Fundamentals of rhamnolipid microemulsions DLS studies performed with a zetasizer gave hydrodynamic radii of 6.2 and 19.3 nm, respectively. Apart from the different production mode, microbial biosurfactants have similar properties as saponins. Their hydrophobic moiety contains one or several branched fatty acid chains, and depending on the hydrophilic head group, they are categorized as phospholipids, lipopeptides, or glycolipids. The latter include threhalolipids, sophorolipids, mannosylerythritol lipids (Kim et al., 2002), or rhamnolipids which are produced by the bacterium Pseudomonas aeruginosa (Rodrigues, 2015; Jahan et al., 2020). Rhamnolipids have a strong surface activity, as they lower the surface tension of the water/air interface by more than a factor of two (Rodrigues, 2015), form micelles above a critical micelle concentration (CMC) and also favor the formation of bacterial biofilms. The potential of rhamnolipids as biodegradable and low-toxicity alternatives to synthetic surfactants for stabilizing microemulsions has been investigated since the beginning of the new millennium. Especially in the deprotonated form, rhamnolipids are quite hydrophilic, resulting in amphiphilic films that are strongly curved around the oil (o/w structures, Winsor I type emulsions). An efficient approach to form microemulsions with hydrophilic surfactants or even drive the system through the phase inversion is to add hydrophobic co-surfactants (Penders and Strey, 1995; Sottmann et al., 2002). In 2005, Ye et al. were among the first to systematically study the influence of alcohol chain length on the phase behavior of a mixture of water/NaOH, n-heptane, rhamnolipid and n-alcohols as co-surfactants (Xie et al., 2005). They used a 50/50 mixture by weight of mono-rhamnolipid and di- rhamnolipids and added sodium hydroxide to adjust a pH of 9 ensuring that the rhamnolipid was completely deprotonated and thus an anionic surfactant. The isothermally recorded Gibbs triangles (25°C) in which the rhamnolipid/n-alcohol mixture is considered as a pseudo-component show that the phase diagrams become more complex as the chain length of the n-alcohol is increased. From n-pentanol, liquid crystalline phases such as lamellar and hexagonal phases appeared, due to the increasing bending rigidity of the mixed amphiphilic film. In a follow-up paper (Xie et al., 2007), the influence of the water concentration on the microstructure was studied in a mixture of 25% n-heptane and 75% of a one-to-one mixture of n-butanol and rhamnolipid. Electrical conductivity showed that the structure reverses from a water-in-oil to an oil-in-water microemulsion upon increasing water concentration. Frontiers in Soft Matter frontiersin.org 3.1 Fundamentals of rhamnolipid microemulsions Apart from the different production mode, microbial biosurfactants have similar properties as saponins. Their hydrophobic moiety contains one or several branched fatty acid chains, and depending on the hydrophilic head group, they are categorized as phospholipids, lipopeptides, or glycolipids. The latter include threhalolipids, sophorolipids, mannosylerythritol lipids (Kim et al., 2002), or rhamnolipids which are produced by the bacterium Pseudomonas aeruginosa (Rodrigues, 2015; Jahan et al., 2020). Rhamnolipids have a strong surface activity, as they lower the surface tension of the water/air interface by more than a factor of two (Rodrigues, 2015), form micelles above a critical micelle concentration (CMC) and also favor the formation of bacterial biofilms. The potential of rhamnolipids as biodegradable and low-toxicity alternatives to synthetic surfactants for stabilizing microemulsions has been investigated since the beginning of the new millennium. Especially in the deprotonated form, rhamnolipids are quite hydrophilic, resulting in amphiphilic films that are strongly curved around the oil (o/w structures, Winsor I type emulsions). An efficient approach to form microemulsions with hydrophilic surfactants or even drive the system through the phase inversion is to add hydrophobic co-surfactants (Penders and Strey, 1995; Sottmann et al., 2002). In 2005, Ye et al. were among the first to systematically study the influence of alcohol chain length on the phase behavior of a mixture of water/NaOH, n-heptane, rhamnolipid and n-alcohols as co-surfactants (Xie et al., 2005). They used a 50/50 mixture by weight of mono-rhamnolipid and di- rhamnolipids and added sodium hydroxide to adjust a pH of 9 ensuring that the rhamnolipid was completely deprotonated and thus an anionic surfactant. The isothermally recorded Gibbs triangles (25°C) in which the rhamnolipid/n-alcohol mixture is considered as a pseudo-component show that the phase diagrams become more complex as the chain length of the n-alcohol is increased. From n-pentanol, liquid crystalline phases such as lamellar and hexagonal phases appeared, due to the increasing bending rigidity of the mixed amphiphilic film. In a follow-up paper (Xie et al., 2007), the influence of the water concentration on the microstructure was studied in a mixture of 25% n-heptane and 75% of a one-to-one mixture of n-butanol and rhamnolipid. Electrical conductivity showed that the structure reverses from a water-in-oil to an oil-in-water microemulsion upon increasing water concentration. Freeze-fracture electron microscopy (FFEM) images might show percolating w/o-droplets at 30 wt% water, while o/ w-droplets might be imaged at 90% water. 2 Ternary oil-water microemulsions based on Quillaja saponins Saponins are biosurfactants which can be extracted from plants, like, e.g., horse chestnuts (used to produce aescin), licorice (glycyrrhizin), soapbark tree Quillaja Saponaria, common foxglove (digitonin), oleander (oleandrin), or Saponaria Oficinalis (gypsogenin). These molecules participate in the defense against invasion by fungi and bacteria. Due to their molecular amphiphilicity based on hydrophobic parts having a triterpenic or steroidic structure with attached sugar moieties, they possess foaming and emulsifying properties. This is for example, the case of Quillaja bark saponins which have been used for washing processes, and their general properties and specifically their interfacial behaviour have been described, see for example, (Nord and 02 frontiersin.org Hellweg et al. 10.3389/frsfm.2023.1260211 FIGURE 1 The phase boundaries feature the “fish shape” well-known for classical microemulsion systems (rotated by 90° due to the choice of axes), with the phase inversion from Winsor I to Winsor II induced by an increase in the NaCl concentration. Figure adapted from ref. (Nguyen et al., 2008) with permission from Elsevier. microemulsion stable over at least a month obtained with microbial biosurfactant will be discussed. 3.1 Fundamentals of rhamnolipid microemulsions Freeze-fracture electron microscopy (FFEM) images might show percolating w/o-droplets at 30 wt% water, while o/ w-droplets might be imaged at 90% water. DLS studies performed with a zetasizer gave hydrodynamic radii of 6.2 and 19.3 nm, respectively. FIGURE 1 A year later, in order to formulate microemulsions without the synthetic surfactants, Sabatini et al. studied a combination of the more hydrophobic biosurfactant sophorolipid (in its nonionic lactone form) and soybean lecithin to compensate Sabatini et al. studied the phase behavior and oil/water- interfacial tension of an alcohol-free symmetric microemulsion 03 frontiersin.org Hellweg et al. 10.3389/frsfm.2023.1260211 for the hydrophilicity of the rhamnolipid (Nguyen et al., 2010). The characteristic phase behavior of microemulsions, reflected in fish- shaped phase boundaries, was obtained when the rhamnolipid to soy lecithin + sophorolipid ratio is used to tune the curvature of the mixed amphiphilic film. In microemulsions with the hydrophobic isopropyl myristate (IPM), the optimal microemulsion (Winsor III and IV), was achieved by a small amount of rhamnolipid, whereas the optimal solubilization of the more hydrophilic limonene required more than twice the amount of rhamnolipid. In 2011, Nguyen and Sabatini provided an overview of microemulsions formulated with rhamnolipids and/or sophorolipids and discussed the use of such microemulsions in various applications (Nguyen and Sabatini, 2011). for the hydrophilicity of the rhamnolipid (Nguyen et al., 2010). The characteristic phase behavior of microemulsions, reflected in fish- shaped phase boundaries, was obtained when the rhamnolipid to soy lecithin + sophorolipid ratio is used to tune the curvature of the mixed amphiphilic film. In microemulsions with the hydrophobic isopropyl myristate (IPM), the optimal microemulsion (Winsor III and IV), was achieved by a small amount of rhamnolipid, whereas the optimal solubilization of the more hydrophilic limonene required more than twice the amount of rhamnolipid. In 2011, Nguyen and Sabatini provided an overview of microemulsions formulated with rhamnolipids and/or sophorolipids and discussed the use of such microemulsions in various applications (Nguyen and Sabatini, 2011). Rhamnolipids exhibit structural similarities to AOT and are therefore rather well suited for the stabilization of oil-continuous microemulsions (see above). Moya-Ramirez and co-workers have produced rhamnolipid-based oil continuous microemulsions (Moya-Ramirez et al., 2017). These microemulsions are used as reaction media for the enzyme catalyzed conversion of used frying oil. The water-in-oil microemulsion system under study contained lipase as catalyst. Compared to the AOT based systems the degree of hydrolysis was 35% higher, which the authors explain by a better compatibility of the lipase and the rhamnolipid in comparison to the synthetic surfactant. Another example for the use of an oil- continuous rhamnolipid-based microemulsion as reaction medium has been given by Palanisamy and Raichur (Palanisamy and Raichur, 2009). 3.2 Applications of rhamnolipid microemulsions Biosurfactant based microemulsions are found in a variety of different applications ranging from Diesel fuel via the use as reaction medium to food products. One example for making Diesel fuel from natural oils like, e.g., castor oil is discussed by Zhu et al. (Zhu et al., 2014). In this work rhamnolipid based microemulsions are used to obtain systems with lower viscosity compared to the original oil leading to less coking in the engine. Moreover, bio-based oil made by liquefaction of biomass is a promising alternative for petrol derived diesel fuel. However, these are usually very complex mixtures containing water and other rather polar compounds. The fuel properties can be improved by microemulsion-formation achieved by simple addition of surfactant. Leng has studied this microemulsification process using a rhamnolipid as micro-emulsifier (Leng, 2018). Unfortunately, the exact composition of the used rhamnolipid is not given. The same group has studied the influence of making biodiesel based microemulsions with small amounts of water and rhamnolipid on the so-called ignition delay (Leng et al., 2022). Biodiesel/rhamnolipid/water microemulsion are found to have a positive effect on the ignition delay and moreover are found to reduce NOx emissions. Hence, such rhamnolipid based biodiesel microemulsions are a more sustainable alternative for conventional Diesel fuel. Besides for Diesel fuel, rhamnolipids can also be used to treat crude oil (Hajimohammadi & Johari-ahar 2017). However, the shown images in this publication indicated that only a crude oil emulsion was made and not a microemulsion, implying that the mixture is not at thermal equilibrium and would sooner or later demix. FIGURE 1 These authors made NiO nanoparticles in a n-heptane/rhamnolipid/water microemulsion. However, as in most studies using oil-continuous microemulsions, no real templating effect of the microemulsion could be proven. The resulting nanoparticles are rather polydisperse and large (micron range). Unfortunately, the authors have not studied the microemulsion droplet sizes. An application of microemulsions in food industry has been described in 2017 by Amiri-Rigi and Abbasi (Amiri-Rigi and Abbasi, 2017). In this work 4 synthetic and 4 bio-surfactants were used to make microemulsions for the extraction of lycopene from tomato processing residuals. The used biosurfactants are lecithin, rhamnolipid, saponin, and sucrose monopalmitate. Lycopene is an interesting compound with strong anti-oxidant activity which can, e.g., be used in skin care products. It is fat soluble and therefore, microemulsions are an interesting extraction medium for it. In the work by Amiri-Rigi and Abbasi only droplet sizes measured by using a zetasizer are presented. No deeper characterization of the structures has been done and therefore the structures obtained by using rhamnolipid and saponin might be vesicles and not microemulsion droplets. However, the extraction process is found to be rather efficient and the lycopene is stabilized by all used surfactants. In this case, e.g., small angle scattering techniques might yield more information about the resulting structures. Guo et al. investigated the effect of styrene added to an aqueous rhamnolipid solution consisting of an equimolar ratio of mono- and di-rhamnolipid on aggregate morphology using SANS adjusting two contrasts (deuterated and hydrogenated styrene) (Guo et al., 2011). Analysis of the data recorded for the aqueous rhamnolipid solution showed that the cylindrical micelles formed at low concentration changed into a mixture of the latter with vesicles as the concentration increased. They rationalized this transition by the fact that the H+-ions released from the carboxyl group of the rhamnolipid lead to a decreasing pH at which they had observed the formation of unilamellar vesicles in a previous SANS study (Dahrazma et al., 2008). Similar transitions have been observed by SANS and electron microscopy in synthetic surfactant systems (Oberdisse et al., 1998). Frontiers in Soft Matter frontiersin.org 4 Some microemulsions formed with other biosurfactants Besides saponins and glycolipid biosurfactants such as rhamnolipids and sophorolipids, also mannosylerythritolipids (MELs), which are abundantly produced by yeast strains of the genus Pseudozyma from renewable resources, were successfully used to stabilize microemulsions. Worakitkanchanakul et al. found that the diacetylated MEL-A on its own is capable of forming water-in- n-decane microemulsions (Worakitkanchanakul et al., 2008). Using DLS, they showed that the diameter of the microemulsion droplets increases from 20 to 60 nm when the ratio of water to surfactant is increased (Figure 2left), as anticipated. This increase in droplet diameter has also been imaged using freeze-fracture electron microscopy (FFEM) (Figure 2, right). Note, however, that the diameters obtained are large for microemulsions, especially given the rather small w0 ratios (Foster et al., 2008). Note that in the DLS analysis the sample viscosity is used instead of n-decane and that FFEM is prone to artifacts. Much more suitable methods to assess the size and shape of the formed structures would be SAXS and 04 frontiersin.org Hellweg et al. 10.3389/frsfm.2023.1260211 FIGURE 2 DLS diameter and FFEM pictures of w/o-microemulsion droplets formed in the system water/n-decane/MEL-A with increasing molar ratio w0 of water-to-MEL-A [from w0 = 1.5 (A), 8.3 (B) and 14.3 (C)] at constant molar n-decane-to-MEL-A ratio of 4.8 at T = 25°C. Figure reprinted from ref (Worakitkanchanakul et al., 2008) with permission from the Japan Oil Chemists’ Society. FIGURE 2 DLS diameter and FFEM pictures of w/o-microemulsion droplets formed in the system water/n-decane/MEL-A with increasing molar ratio w0 of water-to-MEL-A [from w0 = 1.5 (A), 8.3 (B) and 14.3 (C)] at constant molar n-decane-to-MEL-A ratio of 4.8 at T = 25°C. Figure reprinted from ref (Worakitkanchanakul et al., 2008) with permission from the Japan Oil Chemists’ Society. robust results even in complex mixtures. In-house DLS and SAXS are suitable and easily accessible scattering techniques. However, the contrast in electron density, and thus the SAXS intensity, is often limited because typical microemulsion components, as well as the biosurfactants, are composed of H, C, N, and O. Most promising for elucidating complex aggregate morphologies stabilized by mixtures of co-surfactants and biosurfactants is SANS and, in particular, contrast variation SANS using deuteration as demonstrated, for example, for microemulsions stabilized by alcohol/surfactant or amphiphilic diblock copolymer/surfactant mixtures (Endo et al., 2001; Bumajdad et al., 2003). 4 Some microemulsions formed with other biosurfactants Further complementary diffusion NMR and electrical conductivity measurements that probe the continuity of a structure and the study of the dynamic behavior by neutron spin echo will help to elucidate the properties of these promising new microemulsion systems (Hellweg et al., 2001; Reimer et al., 2003). As far as systems are concerned, the “royal path” would be to work with highly purified biosurfactants in order to connect nanostructures to molecular features–probably slowing down progress in an exaggerated manner. We would thus recommend to work with (chemically) polydisperse systems first, and invest energy in purifications only if promising enough. In conclusion, we anticipate biosurfactants to be of growing importance in progressive replacement and improvement of synthetic surfactants by sustainable and biodegradable molecules in many applications, from pharmacology via fuels to detergency. SANS. In a follow-up paper, the same group studied the phase behavior of ternary microemulsions consisting of water, n-decane and MEL-A and the monacetylated MEL-B, respectively (Worakitkanchanakul et al., 2009). The Gibbs phase triangles recorded by polarized optical microscopy and SAXS at 25°C show that at high biosurfactant concentrations the phase diagram of the MEL-A system is dominated by a water-in-n-decane microemulsion, while a lamellar phase is formed in the MEL-B system. These results agree with the phase diagrams of the respective aqueous binary systems summarized by Kitamoto et al. (Kitamoto et al., 2009). While the spontaneous curvature of MEL-A was found to be slightly negative, leading to bilayers forming sponge (L3) and inverted bicontinuous cubic (V2) phases, the spontaneous curvature of MEL-B is almost zero, resulting in the formation of vesicles and/or the lamellar (Lα) phase. Coming back to the incorporation of vitamin E for patients having intake problems, a possible solution is the use of vitamin E microemulsions. In recent work by Kouchi and co-workers such a microemulsion is made by using a crude biosurfactant from Yarrowia lipolytica (Kouchi et al., 2022). The authors present data with droplet sizes below 100 nm, and stability of at least a month. The “crude” biosurfactant mix of microbial origin appears to be better suited than saponins for this purpose, but it is unclear if this is the result of chemical variety, or the precise molecular structure. SANS. In a follow-up paper, the same group studied the phase behavior of ternary microemulsions consisting of water, n-decane and MEL-A and the monacetylated MEL-B, respectively (Worakitkanchanakul et al., 2009). 4 Some microemulsions formed with other biosurfactants The Gibbs phase triangles recorded by polarized optical microscopy and SAXS at 25°C show that at high biosurfactant concentrations the phase diagram of the MEL-A system is dominated by a water-in-n-decane microemulsion, while a lamellar phase is formed in the MEL-B system. These results agree with the phase diagrams of the respective aqueous binary systems summarized by Kitamoto et al. (Kitamoto et al., 2009). While the spontaneous curvature of MEL-A was found to be slightly negative, leading to bilayers forming sponge (L3) and inverted bicontinuous cubic (V2) phases, the spontaneous curvature of MEL-B is almost zero, resulting in the formation of vesicles and/or the lamellar (Lα) phase. Coming back to the incorporation of vitamin E for patients having intake problems, a possible solution is the use of vitamin E microemulsions. In recent work by Kouchi and co-workers such a microemulsion is made by using a crude biosurfactant from Yarrowia lipolytica (Kouchi et al., 2022). The authors present data with droplet sizes below 100 nm, and stability of at least a month. The “crude” biosurfactant mix of microbial origin appears to be better suited than saponins for this purpose, but it is unclear if this is the result of chemical variety, or the precise molecular structure. Author contributions We have reviewed recent progress in the formation of biosurfactant-water-oil microemulsions. One conclusion of our previous mini-review was that the study of biosurfactants from a physical-chemical point of view is still in a rather early stage (Hellweg et al., 2023). This is even more true in the case of biosurfactant-based microemulsions. There is considerable interest in applications, but a lack of systematic and quantitative studies on their properties and especially on the microstructures formed by the amphiphilic biosurfactant film. We therefore recommend a strong focus on structural studies, especially using scattering techniques. The latter are often missing in the literature, although they provide TH: Writing–original draft, Writing–review and editing. JO: Conceptualization, Writing–original draft, Writing–review and editing. TS: Conceptualization, Writing–original draft, Writing–review and editing. Frontiers in Soft Matter References Self-assembly of the bio-surfactant aescin in solution: A small-angle x-ray scattering and fluorescence study. Colloids Interfaces 3 (2), 47. doi:10.3390/colloids3020047 Kouchi, M. M., Amani, H., Naseri, A., and Kariminezhad, H. (2022). Development of an effective and safe system for bioavailability of vitamin e supplements in the stomach. J. Surfactants Deterg. 25 (5), 635–642. doi:10.1002/jsde.12595 Endo, H., Mihailescu, M., Monkenbusch, M., Allgaier, J., Gompper, G., Richter, D., et al. (2001). Effect of amphiphilic block copolymers on the structure and phase behavior of oil-water-surfactant mixtures. J. Chem. Phys. 115 (1), 580–600. doi:10.1063/ 1.1377881 Leng, L. (2018). Biosurfactant rhamnolipid assisted microemulsification of bio-oil components in diesel. Energy Sources, Part A Recovery, Util. Environ. Eff. 41 (7), 829–843. doi:10.1080/15567036.2018.1520362 Foster, T., Sottmann, T., Schweins, R., and Strey, R. (2008). Small-angle neutron scattering from giant water-in-oil microemulsion droplets. i. ternary system. J. Chem. Phys. 128 (5), 054502. doi:10.1063/1.2779322 Leng, L., Wei, L., Li, W., Xu, X., Leng, S., Li, J., et al. (2022). Determination and comparison of the activation energies of biodiesel microemulsion and biodiesel blends. Energy Sources, Part A Recovery, Util. Environ. Eff. 44 (1), 116–125. doi:10.1080/ 15567036.2019.1648594 Fracchia, L., Cavallo, M., Giovanna, M., and Banat, M. (2012). “Biosurfactants and bioemulsifiers biomedical and related applications present status and future potentials,” in Biomedical science, engineering and technology (InTech). doi:10.5772/23821 Lewinska, A., Domza-Kedzia, M., Wójtowicz, K., and Bazylinska, U. (2022). Surfactin-stabilized poly(D,L-lactide) nanoparticles for potential skin application. Colloids Surfaces A Physicochem. Eng. Aspects 648, 129216. doi:10.1016/j.colsurfa. 2022.129216 Geisler, R., Dargel, C., and Hellweg, T. (2019). The biosurfactant β-aescin: A review on the physico-chemical properties and its interaction with lipid model membranes and Langmuir monolayers. Molecules 25 (1), 117. doi:10.3390/molecules25010117 Góral, I., and Wojciechowski, K. (2020). Surface activity and foaming properties of saponin-rich plants extracts. Adv. Colloid Interface Sci. 279, 102145. doi:10.1016/j.cis. 2020.102145 Mayer, S., Weiss, J., and McClements, D. J. (2013). Vitamin e-enriched nanoemulsions formed by emulsion phase inversion: factors influencing droplet size and stability. J. Colloid Interface Sci. 402, 122–130. doi:10.1016/j.jcis.2013.04.016 Guo, Y., Mulligan, C. N., and Nieh, M.-P. (2011). An unusual morphological transformation of rhamnolipid aggregates induced by concentration and addition of styrene: A small angle neutron scattering (SANS) study. Colloids Surfaces A Physicochem. Eng. Aspects 373 (1-3), 42–50. doi:10.1016/j.colsurfa. 2010.10.007 Moya-Ramirez, I., Garcia-Roman, M., and Fernandez-Arteaga, A. (2017). Rhamnolipids: highly compatible surfactants for the enzymatic hydrolysis of waste frying oils in microemulsion systems. ACS Sustain. Chem. &amp Eng. 5 (8), 6768–6775. Funding The author(s) declare that no financial support was received for the research, authorship, and/or publication of this article. 05 frontiersin.org Hellweg et al. Hellweg et al. 10.3389/frsfm.2023.1260211 10.3389/frsfm.2023.1260211 References Jahan, R., Bodratti, A. M., Tsianou, M., and Alexandridis, P. (2020). Biosurfactants, natural alternatives to synthetic surfactants: physicochemical properties and applications. Adv. Colloid Interface Sci. 275, 102061. doi:10.1016/j.cis.2019.102061 Acosta, E. J., Yuan, J. S., and Bhakta, A. S. (2008). The characteristic curvature of ionic surfactants. J. Surfactants Deterg. 11 (2), 145–158. doi:10.1007/s11743-008-1065-7 Acosta, E. J., Yuan, J. S., and Bhakta, A. S. (2008). The characteristic curvature of ionic surfactants. J. Surfactants Deterg. 11 (2), 145–158. doi:10.1007/s11743-008-1065-7 Amiri-Rigi, A., and Abbasi, S. (2017). Stability assessment of lycopene microemulsion prepared using tomato industrial waste against various processing conditions. J. Sci. Food Agric. 97 (14), 4922–4928. doi:10.1002/jsfa.8368 Kahlweit, M., and Strey, R. (1985). Phasenverhalten ternärer systeme des typs h2o-öl- nichtionisches amphiphil (mikroemulsionen). Angew. Chem. 97, 655–669. doi:10.1002/ ange.19850970806 Kahlweit, M., and Strey, R. (1985). Phasenverhalten ternärer systeme des typs h2o-öl- nichtionisches amphiphil (mikroemulsionen). Angew. Chem. 97, 655–669. doi:10.1002/ ange.19850970806 Bailly, C., and Vergoten, G. (2020). Glycyrrhizin: an alternative drug for the treatment of COVID-19 infection and the associated respiratory syndrome? Pharmacol. &amp Ther. 214, 107618. doi:10.1016/j.pharmthera.2020.107618 Kezwon, A., and Wojciechowski, K. (2014). Interaction of quillaja bark saponins with food-relevant proteins. Adv. Colloid Interface Sci. 209, 185–195. doi:10.1016/j.cis.2014. 04.005 Bezerra, K. G. O., Rufino, R. D., Luna, J. M., and Sarubbo, L. A. (2018). Saponins and microbial biosurfactants: potential raw materials for the formulation of cosmetics. Biotechnol. Prog. 34 (6), 1482–1493. doi:10.1002/btpr.2682 Kim, H.-S., Jeon, J.-W., Kim, S.-B., Oh, H.-M., Kwon, T.-J., and Yoon, B.-D. (2002). Surface and physico-chemical properties of a glycolipid biosurfactant, mannosylerythritol lipid, from candida Antarctica. Biotechnol. Lett. 24 (19), 1637–1641. doi:10.1023/a:1020309816545 Bumajdad, A., Eastoe, J., and Heenan, R. K. (2003). Properties of mixed alcohol- zwitterionic surfactant films in quaternary water-in-oil microemulsions. Langmuir 19 (18), 7219–7225. doi:10.1021/la034496k Kitamoto, D., Isoda, H., and Nakahara, T. (2002). Functions and potential applications of glycolipid biosurfactants from energy-saving materials to gene delivery carriers. J. Biosci. Bioeng. 94 (3), 187–201. doi:10.1016/s1389-1723(02) 80149-9 Dahrazma, B., Mulligan, C. N., and Nieh, M.-P. (2008). Effects of additives on the structure of rhamnolipid (biosurfactant): A small-angle neutron scattering (SANS) study. J. Colloid Interface Sci. 319 (2), 590–593. doi:10.1016/j.jcis.2007.11.045 Kitamoto, D., Morita, T., Fukuoka, T., aki Konishi, M., and Imura, T. (2009). Self- assembling properties of glycolipid biosurfactants and their potential applications. Curr. Opin. Colloid &amp Interface Sci. 14 (5), 315–328. doi:10.1016/j.cocis.2009.05.009 Dargel, C., Geisler, R., Hannappel, Y., Kemker, I., Sewald, N., and Hellweg, T. (2019). Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision Frontiers in Soft Matter References Shape similarities of ultra-low interfacial tension curves in ternary microemulsion systems of the water-alkane-ciej type. Ber. Bunsenges. Phys. Chem. 100 (3), 237–241. doi:10.1002/bbpc.19961000309 Olsson, U., Shinoda, K., and Lindman, B. (1986). Change of the structure of microemulsions with the hydrophile lipophile balance of nonionic surfactant as revealed by nmr self-diffusion studies. J. Phys. Chem. 90 (17), 4083–4088. doi:10.1021/j100408a050 Strey, R. (1994). Microemulsion microstructure and interfacial curvature. Colloid and Polym. Sci. 272, 1005–1019. doi:10.1007/bf00658900 Palanisamy, P., and Raichur, A. M. (2009). Synthesis of spherical NiO nanoparticles through a novel biosurfactant mediated emulsion technique. Mater. Sci. Eng. 29 (1), 199–204. doi:10.1016/j.msec.2008.06.008 Xie, Y., Li, Y., and Ye, R. (2005). Effect of alcohols on the phase behavior of microemulsions formed by a biosurfactantrhamnolipid. J. Dispersion Sci. Technol. 26 (4), 455–461. doi:10.1081/dis-200054576 Penders, M., and Strey, R. (1995). Phase behavior of the quaternary system h2io/ n-octane/c8e5/n-octanol: role of the alcohol in microemulsions. J. Phys. Chem. 99, 10313–10318. keine Kopie im Archiv. doi:10.1021/j100025a037 Wojciechowski, K., Jurek, I., Góral, I., Campana, M., Geue, T., and Gutberlet, T. (2021). Surface-active extracts from plants rich in saponins - effect on lipid mono- and bilayers. Surfaces Interfaces 27, 101486. doi:10.1016/j.surfin.2021.101486 Reimer, J., Södermann, O., Sottmann, T., Kluge, K., and Strey, R. (2003). Microstructure of alkyl glucoside microemulsions: control of curvature by interfacial composition. Langmuir 19, 10692–10702. doi:10.1021/la034847v Wojciechowski, K., Orczyk, M., Marcinkowski, K., Kobiela, T., Trapp, M., Gutberlet, T., et al. (2014). Effect of hydration of sugar groups on adsorption of quillaja bark saponin at air/water and Si/water interfaces. Colloids Surfaces B Biointerfaces 117, 60–67. doi:10.1016/j.colsurfb.2014.02.010 Rodrigues, L. R. (2015). Microbial surfactants: fundamentals and applicability in the formulation of nano-sized drug delivery vectors. J. Colloid Interface Sci. 449, 304–316. doi:10.1016/j.jcis.2015.01.022 Wojciechowski, K. (2013). Surface activity of saponin from Quillaja bark at the air/ water and oil/water interfaces. Colloids Surfaces B Biointerfaces 108, 95–102. doi:10. 1016/j.colsurfb.2013.02.008 Salager, J., Morgan, J., Schechter, R., Wade, W., and Vasquez, E. (1979). Optimum formulation of surfactant/water/oil systems for minimum interfacial tension or phase behavior. Soc. Petroleum Eng. J. 19 (02), 107–115. doi:10.2118/7054-pa Worakitkanchanakul, W., Imura, T., Fukuoka, T., Morita, T., Sakai, H., Abe, M., et al. (2009). Phase behavior of ternary mannosylerythritol lipid/water/oil systems. Colloids Surfaces B Biointerfaces 68 (2), 207–212. doi:10.1016/j.colsurfb.2008.10.009 Schober, A., Zhang, J., Subramaniam, A., and Normand, V. (2017). Emulsification efficacy of quillaja saponins at very low concentration: model development and role of alcohols. Colloids Surfaces B Biointerfaces 159, 829–837. doi:10.1016/j.colsurfb.2017. References doi:10.1021/acssuschemeng.7b01008 Nguyen, T., and Sabatini, D. (2011). Characterization and emulsification properties of rhamnolipid and sophorolipid biosurfactants and their applications. Int. J. Mol. Sci. 12 (2), 1232–1244. doi:10.3390/ijms12021232 Hajimohammadi, R., and Johari-ahar, S. (2017). Determination of optimal parameters of rhamnolipid biosurfactant production: an agent for emulsification of heavy crude oil. Petroleum Sci. Technol. 36 (3), 186–192. doi:10.1080/10916466.2017. 1406501 Nguyen, T. T., Edelen, A., Neighbors, B., and Sabatini, D. A. (2010). Biocompatible lecithin-based microemulsions with rhamnolipid and sophorolipid biosurfactants: formulation and potential applications. J. Colloid Interface Sci. 348 (2), 498–504. doi:10.1016/j.jcis.2010.04.053 Helfrich, W. (1973). Elastic properties of lipid bilayers: theory and possible experiments. Z. Naturforsch. 28c, 693–703. doi:10.1515/znc-1973-11-1209 Hellweg, T., Gradzielski, M., Farago, B., and Langevin, D. (2001). Shape fluctuations of microemulsion droplets: A neutron spin–echo study. Colloids Surfaces A 183-185, 159–169. doi:10.1016/s0927-7757(01)00567-2 Nguyen, T. T., and Sabatini, D. A. (2008). Formulating alcohol-free microemulsions using rhamnolipid biosurfactant and rhamnolipid mixtures. J. Surfactants Deterg. 12 (2), 109–115. doi:10.1007/s11743-008-1098-y Hellweg, T., Sottmann, T., and Oberdisse, J. (2023). Recent advances in biosurfactant- based association colloidsself-assembly in water. Front. Soft Matter 2. doi:10.3389/frsfm. 2022.1081877 Nguyen, T. T., Youssef, N. H., McInerney, M. J., and Sabatini, D. A. (2008). Rhamnolipid biosurfactant mixtures for environmental remediation. Water Res. 42 (6-7), 1735–1743. doi:10.1016/j.watres.2007.10.038 Herzog, M., Tiso, T., Blank, L. M., and Winter, R. (2020). Interaction of rhamnolipids with model biomembranes of varying complexity. Biochimica Biophysica Acta (BBA) - Biomembr. 1862 (11), 183431. doi:10.1016/j.bbamem.2020.183431 Nord, L. I., and Kenne, L. (1999). Separation and structural analysis of saponins in a bark extract from quillaja saponaria molina. Carbohydr. Res. 320 (1-2), 70–81. doi:10. 1016/s0008-6215(99)00134-2 Hoar, T. P., and Schulman, J. H. (1943). Transparent water-in-oil dispersions: the oleopathic hydro-micelle. Nature 152, 102–103. doi:10.1038/152102a0 Obasi, T. C., Moldovan, R., Toiu, A., Braicu, C., Bodoki, E., Berindan-Neagoe, I., et al. (2017). Molecular-trapping in emulsion’s monolayer: A new strategy for production Frontiers in Soft Matter 06 frontiersin.org Hellweg et al. 10.3389/frsfm.2023.1260211 10.3389/frsfm.2023.1260211 and purification of bioactive saponins. Sci. Rep. 7 (1), 14511. doi:10.1038/s41598-017- 15067-4 Sottmann, T., Kluge, K., Strey, R., Reimer, J., and Söderman, O. (2002). General patterns of the phase behavior of mixtures of h2o, alkyl glucosides, and cosurfactants. Langmuir 18, 3058–3067. doi:10.1021/la011665x Oberdisse, J., Regev, O., and Porte, G. (1998). Experimental study of the micelle-to- vesicle transition. J. Phys. Chem. B 102 (7), 1102–1108. doi:10.1021/jp972390q Sottmann, T., and Strey, R. (1996). References 08.041 Worakitkanchanakul, W., Imura, T., Morita, T., Fukuoka, T., Sakai, H., Abe, M., et al. (2008). Formation of w/o microemulsion based on natural glycolipid biosurfactant, mannosylerythritol lipid-a. J. Oleo Sci. 57 (1), 55–59. doi:10.5650/jos.57.55 Schubert, K. V., Strey, R., and Kahlweit, M. (1990). A new purification technique for alkyl polyglycol ethers and miscibility gaps for water-CiEj. J. Colloid Interface Sci. 141 (1), 21–29. Das Phasenverhalten von einige binaeren Systemen wird beschrieben. Die Abhaengigkeit von der Tensidreinheit wird dargetstellt. doi:10.1016/0021-9797(91) 90298-m Xie, Y., Ye, R., and Liu, H. (2007). Microstructure studies on biosurfactant- rhamnolipid/n-butanol/water/n-heptane microemulsion system. Colloids Surfaces A Physicochem. Eng. Aspects 292 (2-3), 189–195. doi:10.1016/j.colsurfa.2006. 06.021 Yang, Y., and McClements, D. J. (2013). Encapsulation of vitamin e in edible emulsions fabricated using a natural surfactant. Food Hydrocoll. 30 (2), 712–720. doi:10.1016/j.foodhyd.2012.09.003 Schwuger, M.-J., Stickdorn, K., and Schomäcker, R. (1995). Microemulsions in technical processes. Chem. Rev. 95, 849–864. doi:10.1021/cr00036a003 Shinoda, K., and Kunieda, H. (1973). Conditions to produce so-called microemulsions: factors to increase the mutual solubility of oil and water by solubilizer. J. Colloid Interface Sci. 42 (2), 381–387. doi:10.1016/0021-9797(73)90303-2 Zhu, R., Liang, J., zhong Yuan, X., le Wang, L., jian Leng, L., Li, H., et al. (2014). The formation of rhamnolipid-based water-containing castor oil/diesel microemulsions and their potentiality as green fuels. Energy and Fuels 28 (9), 5864–5871. doi:10.1021/ ef501307e Solans, C., and Kunieda, H. (1997). Industrial applications of microemulsions. New York: Marcel Dekker. 07 07 Frontiers in Soft Matter Frontiers in Soft Matter frontiersin.org
https://openalex.org/W2996678351
https://journal.uinsgd.ac.id/index.php/tamkin/article/download/23925/8593
Malay
null
Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin
Tamkin
2,023
cc-by-sa
7,723
Tamkin: Jurnal Pengembangan Masyarakat Islam Volume 5, Nomor 4, 2020, 475-496 Fakultas Dakwah dan Komunikasi, UIN Sunan Gunung Djati Bandung https://jurnal.fdk.uinsgd.ac.id/index.php/tamkin Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin Aceng Ibrahim1*, Bachrun Rifa’i1, Ratna Dewi1 1Jurusan Pengembangan Masyarakat Islam, Fakultas Dakwah dan Komunikasi Gunung Djati, Bandung *Email : acengibrahim14@gmail.com Aceng Ibrahim1*, Bachrun Rifa’i1, Ratna Dewi1 1Jurusan Pengembangan Masyarakat Islam, Fakultas Dakwah dan Komunikasi Gunung Djati, Bandung *Email : acengibrahim14@gmail.com ABSTRAK Penelitian ini mengungkapkan upaya, program dan dampak kegiatan Pusat Kegiatan Belajar Masyarakat (PKBM) Raharja dalam meningkatkan keterampilan masyarakat miskin. Metode yang digunakan adalah metode deskriptif dengan jenis data kualitatif. Pengumpulan data dilakukan melalui observasi, wawancara, analisis data dilakukan dengan cara mengumpulkan semua informasi, mereduksi data, penyajian data dan penarikan kesimpulan. Hasil penelitian menunjukkan bahwa upaya yang dilakukan untuk meningkatkan keterampilan masyarakat miskin melalui program pendidikan dan ekonomi yaitu dengan persiapan, pendataan dan pelaporan, pemetaan, pelaksanaan, analisis, program dukungan, dan evaluasi. Pusat Kegiatan Belajar Masyarakat (PKBM) Raharja mempunyai program meliputi PAUD, pendidikan kesetaraan, keterampilan seperti pelatihan menjahit, pelatihan memasak, mengemudi, komputer, bahasa asing, dan taman bacaan masyarakat. Kegiatan yang diprioritaskan PKBM Raharja adalah program keterampilan dalam memberdayakan masyarakat miskin. Dampak yang dihasilkan dalam memberdayakan masyarakat miskin ini yaitu perubahan tingkat kesejahteraan masyarakat yang menjadi sejahtera. Kata Kunci : Pemberdayaan; Pusat Kegiatan Belajar Masyarakat; Masyarakat Diterima: Oktober 2020. Disetujui: November 2020. Dipublikasikan: Desember 2020 ABSTRACT This research reviews the efforts, program and impact of community learning center (PKBM) Raharja in improving the skills of the poor. The method used in this research was descriptive qualitative. The study of using data from the database, used to calculate data from databases, data, data, information and data from the data contained in information related to data, as well as information about data that has been analyzed. The results of the study of the expulsion of menstrual program, including training, training, counseling, diagnosis, program planning, and evaluation, as well as training programs for students who have passed the program. In the 475 Diterima: Oktober 2020. Disetujui: November 2020. Dipublikasikan: Desember 2020 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi program Later, it is important to learn how to use the information provided by the Early Childhood Education (PAUD), which is part of the session of equality, which has been involved in the field of training, specializing in learning, development, communicating, teaching and teaching and learning. Community members present in the program, If you have any questions, please contact us anytime. Keywords : development, community learning activity center, society PENDAHULUAN Para ilmuwan sosial dalam memberikan pengertian pemberdayaan mempunyai rumusan yang berbeda-beda dalam berbagai konteks dan bidang kajian. Artinya belum ada definisi yang tegas mengenai konsep tersebut. Namun demikian, bila dilihat secara lebih luas, pemberdayaan sering disamakan dengan perolehan daya, kemampuan dan akses terhadap sumber daya untuk memenuhi kebutuhannya, yang menjadi titik fokus dalam pemberdayaan adalah masyarakat. Menurut Widjaja (2003: 169) pemberdayaan masyarakat adalah upaya meningkatkan kemampuan dan potensi yang dimiliki masyarakat, sehingga masyarakat dapat mewujudkan jati diri, harkat dan martabatnya secara maksimal untuk bertahan dan mengembangkan diri secara mandiri baik di bidang ekonomi, sosial, agama dan budaya. Secara konseptual, pemberdayaan (empowerment) berasal dari kata power (kekuasaan/keberdayaan). Karenanya, ide utama pemberdayaan bersentuhan dengan konsep mengenai kekuasaan. Kekuasaan sering kali dikaitkan dengan kemampuan kita untuk membuat orang lain melakukan apa yang kita inginkan, terlepas dari keinginan dan minat mereka. Ilmu sosial tradisional menekankan bahwa kekuasaan berkaitan dengan pengaruh dan kontrol. Pengertian ini mengasumsikan bahwa kekuasaan sebagai sesuatu yang tidak berubah atau tidak dapat dirubah. Kekuasaan sesungguhnya tidak terbatas pada pengertian diatas. Pemberdayaan menunjuk pada kemampuan orang, khususnya kelompok rentang dan lemah sehingga mereka memiliki kekuatan atau kemampuan dalam memenuhi kebutuhan dasarnya sehingga mereka memiliki kebebasan (freedom), dalam arti bukan saja bebas mengemukakan pendapat, melainkan bebas dari kelaparan, bebas dari kebodohan, bebas dari kesakitan; menjangkau sumber-sumber produktif yang memungkinkan mereka dapat meningkatkan pendapatannya dan memperoleh barang-barang dan jasa-jasa yang mereka perlukan; dan berpartisipasi dalam proses pembangunan dan keputusan-keputusan yang mempengaruhi mereka. Beberapa ahli dibawah ini mengemukakan define pemberdayaan dilihat dari tujuan, proses, dan cara-cara pemberdayaan. (Suharto, 2010: 57-58) Sasaran pemberdayaan masyarakat bukan hanya di bidang ekonomi, bidang politik, bidang lingkungan namun bidang pendidikan pun menjadi sasaran 476 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin pemberdayaan masyarakat terutama masyrakat miskin, sehubungan dengan sasaran diatas Dinas Pendidikan memfalitasi dengan adanya pemberdayaan masyarakat berbasis program ( PKBM ) Pusat Kegiatan Belajar Masyarakat yang di jembatani oleh wadah ( PNFI ) Pendidikan Non Formal Informal yang berpusat pada kegiatan – kegiatan yang bersifat peningkatan kualitas individu maupun kelompok. Dinas Pendidikan menyediakan wadah dalam bidang Pendidikan Nonformal Informal guna untuk membantu dan memudahkan warga belajar untuk bisa mendapatkan pendidikan Nonformal, diruang lingkup PKBM tidak hanya warga belajar usia 9 tahun tetapi di PKBM ini warga belajar mulai dari anak usia sekolah sampai orang dewasa mengikuti kegiatan pembelajaran di PKBM. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 PENDAHULUAN Karena diadakanya lembaga PKBM ini guna untuk memudahkan warga masyarakat yang kurang mampu dalam mendapatkan pendidikan Formal, juga memudahkan masyarakat yang sudah tertinggal usia sekolah di umumnya. Kegiatan-kegiatan yang diadakan oleh PNFI sangat mendukung akan keberdayaan warga belajara yang diadakan oleh lembaga PKBM karena itu pemberdayaan dalam bidang ini meningkatkan keterampilan dan kemahiran warga belajar sampai bisa menjadikan warga belajar itu mandiri dan bisa meningkatkan ekonomi dalam kehidupannya. Pusat Kegiatan Belajar Masyarakat (PKBM) adalah tempat pembelajaran dalam bentuk berbagai macam keterampilan dengan memanfaatkan sarana, prasarana, dan segala potensi yang ada di sekitar lingkungan kehidupan masyarakat, agar masyarakat memiliki keterampilan dan pengetahuan yang dapat dimanfaatkan untuk meningkatkan dan memperbaiki taraf hidupnya. Pusat Kegiatan Belajar Masyarakat ini merupakan salah satu alternative yang dipilih dan dijadikan sebagai ajang proses pemberdayaan masyarakat. Hal ini selaras dengan adanya pemikiran bahwa dengan melembagakan Pusat Kegiatan Belajar Masyarakat, maka akan banyak potensi yang dimiliki oleh masyarakat yang selama ini belum dikembangkan secara maksimal. Pusat Kegiatan Belajar Masyarakat diarahkan untuk dapat mengembangkan potensi-potensi tersebut menjadi bermanfaat bagi kehidupannya.Agar mampumengembangkan potensi- potensi tersebut, maka diupayakan kegiatan pembelajaran yang diselenggarakan di PKBM bervariasi sesuai dengan kebutuhan masyarakat. Kegiatan-kegiatan yang diadakan oleh PNFI sangat mendukung akan keberdayaan warga belajara yang diadakan oleh lembaga PKBM karena itu pemberdayaan dalam bidang ini meningkatkan keterampilan dan kemahiran warga belajar sampai bisa menjadikan warga belajar itu mandiri dan bisa meningkatkan ekonomi dalam kehidupannya. Pusat Kegiatan Belajar Masyarakat (PKBM) adalah tempat pembelajaran dalam bentuk berbagai macam keterampilan dengan memanfaatkan sarana, prasarana, dan segala potensi yang ada di sekitar lingkungan kehidupan masyarakat, agar masyarakat memiliki keterampilan dan pengetahuan yang dapat dimanfaatkan untuk meningkatkan dan memperbaiki taraf hidupnya. Pada tingkat kongkrit, konsep merupakan suatu gambaran mental dari beberapa objek atau kejadian yang sesungguhnya.Pada tingkat abstrak dan komplek, konsep merupakan sentesis sejumlah kesimpulan yang telah ditarik dari pengalaman dengan objek atau kejadian tertentu. Pusat kegiatan belajar masyarakat atau PKBM, merupakan sebuah lembaga pendidikan yang lahir dari pemikiran tentang kesadaran akan pentingnya kedudukan masyarakat dalam proses pembangunan pendidikan nonformal. Oleh sebab itu berdirinya PKBM ditengah masyarakat diharapkan mampu 477 kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi menjadi tulang punggung bagi terjadinya proses pembangunan melalui pemberdayaan potensi-potensi yang ada dimasyarakat. Menurut Sihombing, bahwa PKBM merupakan salah satu alternatif yang dapat dipilih dan dijadikan ajang pemberdayaan masyarakat. PENDAHULUAN Hal ini selaras dengan pemikiran bahwa dengan melambangkan PKBM, akan banyak potensi yang selama ini tidak tergali akan dapat digali, ditumbuhkan, dimanfaatkan, dan didayagunakan melalui pendekatan- pendekatan budaya yang persuasif. PKBM sebagai salah satu mitra kerja pemerintahan dalam mencerdaskan kehidupan mayarakat melalui program-program pendidikan nonformal, diharapkan mampu menumbuhkan masyarakat belajar sehingga pada akhirnya akan meningkatkan kemandirian, keberdayadidikan, an inovatif dalam mencari berbagai informasi baru dalam rangka meningkatkan kehidupannya. Sebagai sebuah pusat pembelajaran, PKBM dibangun atas dasar kebutuhan masyarakat dengan menitik beratkan pada swadaya, gotong royong dan partisipasi masyarakat itu sendiri.Terutama berkaitan dengan pentingnya peningkatan kemampuan, keterampilan dan kecerdasan anggota masyarakat. Potensi masyarakat Kelurahan Antapani Kidul Kecamatan Antapani Kota Bandung sangat tinggi terutama dalam bidang keterampilan seperti, menjahit, tataboga, dan keahlian dalam menggunakan teknologi, namun potensi tersebut dibatasi oleh biaya sehingga PKBM ini menjadi wadah bagi masyarakat Kelurahan Antapani Kidul Kecamatan Antapani Kota Bandung dalam meningkatkan potensi yang dimiliki masyarakat tersebut. Dari latar belakang masalah tersebut dapat dirumuskan beberapa masalah dengan rumusan masalah: Bagaimana upaya PKBM Raharja dalam meningkatkan keterampilan masyarakat miskin? Bagaimana program yang dilakukan PKBM Raharja dalam meningkatkan keterampilan masyarakat miskin ? Bagaimana dampak PKBM Raharja dalam meningkatkan keterampilan masyarakat miskin? Metode yang digunakan dalam penelitian ini adalah metode deskriptif, dengan pendekatan kualitatif, dengan menganalisis data melalui observasi, wawancara dengan Staf di PKBM, dan masyarakat. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 LANDASAN TEORITIS Dalam landasan teoritis ini dijelaskan mengenai pengertian pemberdayaan. Pemberdayaan berasal dari kata “daya” yang mendapat awalan ber- menjadi kata “berdaya” artinya memiliki atau mempunyai daya. Daya artinya kekuatan, berdaya artinya memiliki kekuatan. Kata “berdaya” apabila diberi awalan pe- dengan mendapat sisipan -m- dan akhiran -an menjadi berdaya atau mempunyai daya atau mempunyai kekuatan. Pemberdayaan dalam bahasa Indonesia merupakan terjemahan dari “empowerment” dalam bahasa inggris. Dalam konteks 478 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin pembangunan istilah pemberdayaan pada dasarnya bukanlah istilah baru melainkan sudah sering dilontarkan semenjak adanya kesadaran bahwa faktor manusia memegang peran penting dalam pembangunan. Pemberdayaan disebut sebagai tamkiinu al-dakwah yang memiliki makna kekuatan, kekuasaan, kepedulian dan kemauan yang keras. Arah pemberdayaan diharapkan tepat pada sasaran yang dimulai dari kemiskinan dan simbol-simbol ketidakberdayaan lainnya. Sasaran pemberdayaan dilihat dari segi penyandang masalah kesejahteraan sosial, yaitu: Kemiskinan, yaitu penduduk Indonesia yang termasuk kategori fakir miskin, Ketelantaran, yaitu kondisi tidak terpenuhinya kebutuhan dasar manusia, yang melanda anak-anak, perempuan dan lanjut usia, gelandangan dan pengemis, Kecacatan baik cacat secara fisik ataupun cacat secara mental, Ketuna-sosialan, yaitu kondisi disharmonisasi dengan nilai susila dan sosial budaya yang umum berlaku di masyarakat, dan bencana, baik bencana alam maupun bencana social (Setiawan, 2012: 350-351). Menurut Shardlow istilah pemberdayan membahas bagaimana individu, kelompok ataupun komunitas berusaha mengontrol kehidupan mereka sendiri dan mengusahakan untuk membentuk masa depan sesuai dengan keinginan mereka (Risyanti, 2006: 3). Pemberdayaan ditujukan untuk meningkatkan kekuasaan (power) dari kelompok masyarakat yang kurang beruntung (disadvantaged). Berdasarkan pernyataan ini, pemberdayaan pada dasarnya menyangkut dua kata kunci, yaitu kekuasaan (power) dan kurang beruntung (disadvantaged) (Huda, 2009: 272-273). Pertama, kekuasaan. Realitas yang terjadi di masyarakat, antara satu kelompok dan kelompok masyarakat yang lain sering terjadi kompetisi yang tidak menguntungkan. Kelompok masyarakat yang kaya cenderung mempunyai kekuasaan absolut dan elite politik yang tidak seimbang sehingga pemberdayaan harus mampu membuka dan mendorong akses yang terbuka agar tidak terjadi dominasi. Kedua, keberuntungan, lemahnya kekuatan yang dimiliki salah satu kelompok masyarakat menyebabkan mereka menjadi kurang beruntung. Dengan demikian, pemberdayaan diharapkan mampu menngani masyrakat yang kurang beruntung akibat dari faktor struktural, kultural, dan personal. Melakukan pemberdayaan tentunya ada suatu keterampilan yang masyarakat lakukan. Kata keterampilan berasal dari kata terampil yang berarti cakap dalam menyelesaikan tugas, mampu dan cekatan.Arti secara umum, keterampilan ialah kegiatan yang berhubungan dengan urat-urat syaraf dan otot-otot (neuromuscular) yang lazimnya tampak dalam kegiatan jasmaniah seperti menulis, mengetik, olah raga, dan lain-lain. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 LANDASAN TEORITIS Secara khusus keterampilan dalam belajar adalah suatu cara yang dipakai 479 kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi untuk mendapat, mempertahankan, dan mengungkapkan pengetahuan serta merupakan cara untuk menyelesaikan masalah. Keterampilan menurut para ahli, yang Pertama, menurut Gordon (1994: 55), keterampilan adalah kemampuan untuk mengoperasikan pekerjaan secara mudah dan cermat. Kedua, menurut Nadler (1986: 73), keterampilan adalah kegiatan yang memerlukan praktek atau dapat diartikan sebagai implikasi dari keterampilan. Ketiga, menurut Dunnette (1976: 33), keterampilan adalah kapasitas yang dibutuhkan untuk memelaksanakan beberapa tugas yang merupakan pengembangan dari hasil training dan pengalaman yang didapat. Menurut Robbins (2000: 494-495) pada dasarnya keterampilan dapat dikategorikan menjadi empat, yaitu : Pertama, basic literacy skill (keahlian dasar) merupakan keahlian seseorang yang pasti dan wajib dimiliki oleh kebanyakan orang. Seperti membaca, menulis dan mendengar. Kedua, technical skill (keahlian teknik) merupakan keahlian seseorang dalam pengembangan teknik yang dimiliki. Seperti menghitung secara tepat, mengoperasikan computer. Ketiga,interpersonal skill (keahlian interpersonal) merupakan kemampuan seseorang secara efektif untuk berinteraksi dengan orang lain maupun dengan rekan kerja. Seperti pendengar yang baik, menyampaikan pendapat secara jelas dan bekerja dalam satu tim. Keempat, problem solving (pemecahan masalah) merupakan proses aktivitas untuk menjalankan logika, berargumentasi dan penyelesian masalah serta kemampuan untuk mengetahui penyebab, mengembangkan alternatif dan menganalisa serta memilih penyelesaian yang baik. Jenis keterampilan meliputi keterampilan personal dengan dilakukannya studi dan kebiasaan kerja: misalnya keterampilan menentukan lokasi kerja, mengumpulkan data, menggunakan reference material, membuat kesimpulan dll. Dengan latihan yang benar, siswa diberi peluang untuk memiliki percakapan belajar mandiri dan bekerja mandiri., Keterampilan bekerja dalam kelompok: berkenaan dengan kemampuan seseorang didalam kelompok, seperti menyusun rencana, memimpin diskusi, menilai pekerjaan secara bersama-sama. Keterampilan belajar (continuing learning skills). Keterampilan ini memungkinkan seseorang terampil belajar sepanjang hayat. Untuk tingkat pendidikan dasar sasarannya adalah baru dalam tahapan mengembangkan segenap potensi dirinya dikemudian hari, siswa memiliki semangat, kemampuan dan kepercayaan diri yang sehat. Kemudian keterampilan sosial meliputi kehidupan dan kerja sama, belajar memberi dan menerima tanggung jawab, menghormati hak-hak orang lain, membina kesadaran sosial. Dengan demikian, keterampilan ini maka siswa mampu berkomunikasi dengan sesama manusia, lingkungannya dimasyarakat secara baik. Latihan dan pembinaan yang tampak dalam proses belajar mengajar antara lain mampu melaksanakan dengan baik: diskusi dengan teman, bertanya kepada 480 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin siapapun, menjawab pertanyaan orang lain, menjelaskan kepada orang lain, membuat laporan, dan memerankan sesuatu, dll (Belen dkk, 1990: 348). Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 LANDASAN TEORITIS Keterampilan sosial adalah kemampuan individu untuk berkomunikasi efektif dengan orang lain baik secara verbal maupun nonverbal sesuai dengan situasi dan kondisi yang ada pada saat itu, di mana keterampilan ini merupakan perilaku yang dipelajari. Siswa dengan keterampilan sosial akan mampu mengungkapkan perasaan baik positif maupun negatif dalam hubungan interpersonal, tanpa harus melukai orang lain (Gimpel & Merrell, 1998). Keterampilan sosial membawa siswa untuk lebih berani berbicara, mengungkapkan setiap perasaan atau permasalahan yang dihadapi dan sekaligus menemukan penyelesaian yang adaptif, sehingga mereka tidak mencari pelarian ke hal-hal lain yang justru dapat merugikan diri sendiri adalah memiliki keterampilan sosial (social skill) untuk dapat menyesuaikan diri dengan kehidupan sehari-hari. Keterampilan sosial tersebut meliputi kemampuan berkomunikasi, menjalin hubungan dengan orang lain, menghargai diri sendiri dan orang lain, mendengarkan pendapat atau keluhan dari orang lain, memberi atau menerima feedback, memberi atau menerima kritik, bertindak sesuai norma dan aturan yang berlaku, serta lain sebagainya. Apabila keterampilan sosial dapat dikuasai oleh anak pada fase tersebut maka ia akan mampu menyesuaikan diri dengan lingkungan sosialnya. Hal ini berarti pula bahwa sang anak tersebut mampu mengembangkan aspek psikososial dengan maksimal. Berdasarkan uraian tersebut dapat disimpulkan bahwa keterampilan sosial merupakan kemampuan seseorang untuk berani berbicara, mengungkapkan setiap perasaan atau permasalahan yang dihadapi sekaligus menemukan penyelesaian yang adaptif, memiliki tanggung jawab yang cukup tinggi dalam segala hal, penuh pertimbangan sebelum melakukan sesuatu, mampu menolak dan menyatakan ketidaksetujuannya terhadap pengaruhpengaruh negatif dari lingkungan. Dalam keterampilan yang dihasilkan oleh manusia tentunya ada pelatihan yang menjadi pendorongnya. Pelatihan adalah proses yang didesain untuk meningkatkan pengetahuan dan keterampilan teknis, ataupun meningkatkan kinerja pegawai. Dengan tujuan untuk mengurangi dan menghilangkan kinerja yag buruk, meningkatkan produktivitas, membentuk sikap, loyalitas, da kerja sama yag lebih menguntungkan, memenuhi kebutuhan perencanaan sumberdaya manusia, mengurangi frekuensi dan biaya kecelakaan kerja, dan membantu karyawan dalam peningkatan dan pengembangan pribadi mereka. Tahapan pelatihan dalam SDM meliputi: Pertama, penentuan kebutuhan pelatihan SDM dengan tujuan penentuan kebutuhan pelatihan SDM ini adalah untuk mengumpulkan sebanyak mungkin informasi yang relevan guna mengetahui data atau menentukan apakah perlu tidaknya dilakukan pelatihan SDM dalam organisasi tersebut. Kedua, mendesain 481 kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi program pelatihan SDM melalui metode pelatihan SDM tergantung pada tujuan yang hendak dicapai identifikasi mengenai apa yang diinginkan agar para pekerja harus mengetahui dan harus melakukan. Ketiga, evaluasi efektifitas program pelatihan SDM. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 LANDASAN TEORITIS Dengan pelatihan SDM harus merupakan suatu solusi yang tepat bagi permasahan organisasi, yaitu bahwa pelatihan SDM dimaksudkan untuk memperbaiki kekurangan dari keterampilan setiap pekerja. Tujuan dari tahap ini adalah untuk menguji apakah pelatihan SDM tersebut efektif di dalam mencapai sasaran-sasarannya yang telah ditetapkan. Teknik pelatihan yang dilakukan melalui metode praktis (on job training) metode Pelatihan yang langsung ditangani oleh supervisi/pimpinan perusahaan. Dan metode simulasi dengan metode ini dilakukannya latihan representasi tiruan (artificial). Suatu aspek organisasi dan diminta untuk menanggapinya seperti dalam keadaan sebenarnya. contoh metode simulasi : metode studi kasus, permainan rotasi jabatan, permainan bisnis, dan lain-lain. Masyarakat adalah sekelompok orang yang membentuk sebuah sistem semi tertutup, dimana sebagian besar interaksi adalah individu - individu yang berada dalam kelompok tersebut. Miskin atau kemiskinan diartikan sebagai suatu keadaan dimana seseorang tidak sanggup memelihara dirinya sendiri sesuai dengan taraf kehidupan kelompok dan juga tidak mampu memanfaatkan tenaga, mental, maupun fisiknya dalam kelompok tersebut. Pada dasarnya kemiskinan dapat dibedakan menjadi dua yaitu kemiskinan absolut dan kemiskinan relatif. Kartasasmita (1997:10) mengungkapkan bahwaseseorang dikatakan miskin absolut apabila tidak mampu membiayai kebutuhanyang paling minimal untuk dapat hidup sesuai dengan martabat kemanusiaan.Kemiskinan relatif adalah perbandingan antara kelompok pendapatanmasyarakat, yaitu antara kelompok yang miskin dan kelompok yang tidak miskin,karena mempunyai tingkat pendapatan yang relatif tinggi daripada gariskemiskinan dan merupakan kelompok masyarakat yang relatif lebih maju. Menurut Kartasasmita (1997: 14-15), penyebab kemiskinan ada empat hal sebagai berikut: Pertama, rendahnya taraf pendidikan, taraf pendidikan yang rendah mengakibatkan kemampuan pengembangan diri terbatas dan menyebabkan sempitnya lapangan pekerjaan yang dapat dimasuki. Kedua, rendahnya derajat kesehatan, taraf kesehatan dan gizi yang rendah menyebabkan rendahnya daya tahan fisik, daya pikir dan prakarsa. Terjaminnya kesehatan seseorang akan membuat dirinya rajin bekerjasehingga pendapatannya dapat meningkat dan kesejahteraannya juga meningkat. Ketiga, terbatasnya lapangan pekerjaan, keadaan kemiskinan karena kondisipendidikan dan kesehatan diperparah oleh terbatasnya lapangan pekerjaan.Selama ada lapangan pekerjaan atau kegiatan, selama itu pula ada harapanuntuk memutuskan lingkaran kemiskinan itu. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 482 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin Kondisi keterisoliran, banyak penduduk miskin secara ekonomi tidak berdaya karena terpencil dan terisolasi. Mereka hidup terpencil sehinggasulit atau tidakdapat terjangkau oleh pelayananpendidikan, kesehatan dangerak kemajuan yang dinikmati oleh masyarakat lainnya. Ketidakaktifan keluarga dalam kegiatan pemberdayaan, terutama terjadi setiap tahun disebabkan karena keluarga tidak dapat melaksanakan manajemen keuangan keluarga dengan baik. LANDASAN TEORITIS Lubis (2006) menyatakan bahwa kesalahan mengelola sumber daya merupakan salah satu penyebab keluarga menjadi miskin, dan sebaliknya,pengelolaan sumber daya yang baik dapat mengangkat keluarga menjadi tidak miskin karena keluarga lebih tahan dalam menjalani kehidupannya di masa krisis. Maka, dalam hal ini mencapai keberdayaan keluarga miskin akan mewujudkan kesejahteraan hidup. Kesejahteraan menurut Midgley dalam Adi, 2002: 121-122) adalah kesejahteraan sosial sebagai suatu kondisi sosial dan bukan sekedar kegiatan amal ataupun bantuan sosial yang diberikan pemerintah. Sebagai suatu kondisi (keadaan), kesejahteraan sosial dapat dilihat dari tiga unsur utamanya, yaitu: tingkatan (derajat sampai di mana permasalahan sosial yang ada di masyarakat dapat dikelola, sampai seberapa banyak kebutuhan masyarakat dapat dipenuhidan, sampai seberapa besar kesempatan untuk meningkatkan taraf hidup dapat diperluas pada berbagai lapisan masyarakat. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 HASIL DAN PEMBAHASAN Pusat Kegiatan Belajar Masyarakat Raharja dimulai sejak banyakanya anak-anak yang kurang ilmu pengetahuan karena disibukan oleh bermain yang kurang penting. Pada mulanya yayasan PKBM Raharja ini bekerja sama dengan PKBM Dewi Shinta, hal ini disebabkan karena PKBM Raharja belum memiliki akta notaris, jadi memiliki banyak murid tetapi tempat belajar tidak ada, sedangkan di PKBM Dewi Shinta ada tempat tetapi tidak ada murid. Kemudian pada tahun 2002 pihak yayasan Raharja bekerja sama dengan dinas pemerintah untuk mendapat akta notaris sehingga PKBM ini menjadi yayasan yang legal di pemerintah. PKBM Raharja ini di Jl. Rengasdengklok 9 No.8 Kelurahan Antapani Kidul Kecamatan Antapani Bandung yang merupakan wadah berbagai kegiatan pembelajaran masyarakat yang diarahkan pada pemberdayaan potensi masyarakat untuk menggerak-kan pembangunan di bidang sosial, ekonomi, dan budaya. Pusat Kegiatan Belajar Masyarakat Raharja dibentuk oleh masyarakat, milik masyarakat dan dikelola oleh masyarakat untuk memperluas pelayanan kebutuhan belajar masyarakat. Pembentukan PKBM dilakukan dengan memperhatikan sumber-sumber potensi yang terdapat pada daerah yang bersangkutan terutama jumlah kelompok sasaran dan jenis usaha/keterampilan yang secara ekonomi, sosial dan budaya dapat dikembangkan untuk meningkatkan kesejahteraan warga belajar khususnya dan warga masyarakat sekitarnya. kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 483 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Upaya Pusat Kegiatan Belajar Masyarakat Raharja dalam Memberdayakan Masyarakat Usaha-usaha yang dilakukan Pusat Kegiatan Belajar Masyarakat Raharja dalam memberdayakan masyarakat adalah meningkatkan nya kuantitas Warga Belajar dengan cara menyebar brosur dan memasangkan spanduk lembaga di wilayah yang belum tersentuh. Pelaksanaan program-program pemberdayaan di pusat kegiatan belajar masyarakat dilakukan oleh masyarakat dan difasilitasi oleh yayasan dan perangkat pemerintahan yang dibantu oleh tenaga pengajar. Tahap pelaksanaan kegiatan dilakukan setelah proses perencanaan selesai dan adanya kontribusi dari masyarakat dalam melaksanakan usaha terhadap program pemberdayaan PKBM Raharja adapun langkah-langkah yang dilakukan PKBM dalam melaksanakan usaha program-program pemberdayaan yang meliputi persiapan, pemetaan, pelaksanaan, analisis, program dukungan, dan evaluasi. Hal ini bertujuan untuk mendorong dan meningkatkan kesadaran sosial serta kepedulian partisipasi sosial warga masyarakat desa dalam pelaksanaan pembangunan masyarakat pedesaan (Jamaludin, 2015: 248-249). Adapun implementasi dari setiap program-program pemberdayaan yang dilakukan PKBM Raharja yang Pertama, di bidang pendidikan yaitu dalam program Pendidikan Usia Dini (PAUD), Pendidikan Kesetaraan. Dalam program ini masyarakat dibina dengan pembelajaran agar masyarakat mendapat pendidikan yang layak, sehingga mereka dapat melanjutkan sekolah yang tinggi walaupun dengan pendidikan kesetaraan paket A,B, dan C. Kemudian anak-anak kecil juga dibina dengan menggunakan metode pembelajaran bernyanyi, menghapal doa-doa dan sebagainya. PAUD ini menerapkan pendidikan berbasis karakter dengan menerapkan model yang bukan hanya memberikan rasa aman dan nyaman pada anak, tetapi juga guna merangsang minat belajar anak, sehingga diharapkan siswanya dapat menjadi generasi kreatif dan berdaya pikir tinggi (Higher order Thinking Skills). Sedangkan Taman Bacaan yang terletak bersamaan dengan ruang PAUD, buku-buku tersebut disediakan untuk menambah wahana bacaan bagi anak-anak agar dapat menambah wawasan serta pengetahuannya, adanya bantuan buku-buku tersebut dari beberapa pihak seperti aparat pemerintah desa, dan donator lainnya. Kedua, bidang ekonomi yang dilakukan terhadap program-program pemberdayaan di bidang ekonomi yaitu pertama, dengan adanya keterampilan/kursus seperti mengemudi, kursusu computer, bahasa asing, pelatihan menjahit dan memasak. dalam menjalankan program tersebut PKBM Raharja bekerjasama dengan dinas, dengan adanya usaha peningkatan pendapatan keluarga sejahtera yang membantu dalam kegiatan ekonomi masyarakat di wilayah 484 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin tersebut, ditunjang dengan adanya pelatihan-pelatihan maupun seminar yang diadakan baik dari yayasan Raharja, sehingga warga pun terdorong untuk mencari tambahan penghasilan, oleh karena itu, munculah kegiatan-kegiatan ekonomi di masyarakat di kelurahan Antapani. Upaya Pusat Kegiatan Belajar Masyarakat Raharja dalam Memberdayakan Masyarakat Kegiatan ekonomi ini tetunya dengan mengali potensi masyarakat dengan menjalin komunikasi efektif dengan orang lain baik secara verbal maupun nonverbal sesuai dengan situasi dan kondisi yang ada pada saat itu, di mana keterampilan ini merupakan perilaku yang dipelajari (Gimpel & Merrell, 1998). Dari sekian program-program pemberdayaan yang dilakukan, partisipasi dari masyarakat pun cukup antusias dalam mengikuti kegiatan pemberdayaan yang dilakukan PKBM Raharja, walaupun dalam pelaksanaannya sering adanya pro kontra dari beberapa masyarakat, namun semua itu sedikit demi sedikit dapat diatasi dengan baik. Mengingat pentingnya sebuah partisipasi masyarakat dalam mengikuti program-program pemberdayaan, terdapat tiga alasan utama mengapa partisipasi masyarakat mempunyai sifat penting Menurut Diana Conyers (1991: 154-155). Pertama, partisipasi masyarakat merupakan suatu alat guna memperoleh informasi mengenai kondisi, kebutuhan, dan sikap masyarakat setempat, yang tanpa kehadirannya program pembangunan serta proyek-proyek akan gagal. Alasan kedua, yaitu bahwa masyarakat akan lebih mempercayai proyek atau program pembangunan jika merasa dilibatkan dalam proses persiapan dan perencanannya, Alasan ketiga, partisipasi menjadi urgen karena timbul anggapan bahwa merupakan suatu hak demokrasi jika masyarakat dilibatkan dalam pembangunan masyarakat. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Program Pusat Kegiatan Belajar Masyarakat Raharja dalam Meningkatkan Keterampilan Masyarakat Program yang dilakukan untuk memberdayakan masyarakat yang ada di lembaga PKBM Raharja adalah mengadakannya kegiatan kursus yang tanpa biaya kepada warg belajarnya seperti mengadakan kursus menjahit yang biasanya ada tahap dasar sampai tahap mahir, lembaga memfasilitasi anggaran kursus tingkat dasar sampai selesai, jika warga belajar ada yang berminat untuk melanjutkan kursus ke tingkat mahir, maka warga belajar harus mengeluarkan anggaran sendiri atau mandiri. Usaha ini biasa menjadikan respon positif kepada warga belajar untuk bisa memberdayakan kemampuan dan keahliannya dibidang tersebut. Jadi untuk meningkatkan keterampilan masyarakat PKBM Raharja adalah mengadakannya kegiatan tanpa biaya atau gratis dari yayasan. Program pemberdayaan masyarakat di Pusat Kegiatan Belajar Masyarakat (PKBM) merupakan peningkatan mutu kehidupan mencakup dimensi yang sangat luas dalam dimensi kehidupan. Mulai 485 kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi dari dimensi spiritual, sosial, ekonomi, kesehatan, mentalitas dan kepribadian, seni dan budaya dan sebagainya. Ada komunitas yang hanya menonjolkan satu atau dua dimensi saja. Sementara dimensi lainnya kurang diperhatikan, tetapi ada juga komunitas yang mencoba memandang penting semua dimensi. Dalam mempercepat upaya peningkatan kesejahteraan masyarakat melalui pemberdayaan masyarakat berbasis komunitas, pengembangan ekonomi produktif, dengan melibatkan kelembagaan masyarakat sebagai stakeholder dengan menggerakan partisipasi masyarakat, dalam peningkatan kapasitas kelompok usaha kecil. Dalam melakukan pelatihan ini masyarakat harus mempunyai potensi agar dapat dikembangkan dan dimanfaatkan dengan sebaik mungkin. Hal ini sebagimana yang dikatakan oleh Dina selaku Bendahara beliau mengatakan bahwa: “Banyak masyarakat yang memanfaatkan adanya program ini, terutama masyarakat yang minim pendapatan, disini mereka dibina dengan melalui 3 tahap, yang pertama tahap dasar menjahit, kemudian tahap keterampilan, dan yang terakhir tahap mahir. Tujuan ini dimaksudkan untuk dapat memberdayakan masyarakat terutama dalam peningkatan sumber daya ekonomi”. (Wawancara dengan Dina selaku Bendahara pada tangga 10 November 2017) Berdasarkan hasil wawancara dengan Nenden selaku Sekretaris bahwa adanya pelatihan ini sebagai salah satu untuk memberdayakan masyarakat dalam mensejahterakan perekonomian beliau mengadakan pelatihan ini agar masyarakat dapat mengembangkan potensinya dan memberikan bekal kepada masyarakat agar mereka dapat mengembangkan ilmu mereka dan dapat mengimplementasikan dalam kehidupan sehari-hari. Peran pemberdayaan dalam hal ini untuk mengontrol kehidupan masyarakat dan mengusahakan untuk membentuk masa depan sesuai dengan keinginan mereka (Risyanti, 2006: 3). Dalam melaksanakan pelatihan ini pihak PKBM melakukan kerjasama dengan pihak dinas pemerintahan untuk dapat memudahkan proses berjalannya program di Pusat Kegiatan Belajar Masyarakat (PKBM) Raharja. Program Pusat Kegiatan Belajar Masyarakat Raharja dalam Meningkatkan Keterampilan Masyarakat Kemudian apabila sudah ada yang mahir dalam menjahit mereka akan diberi penghargaan berupa sertifikat, dan sertifikat tersebut dapat digunakan untuk melamar pekerjaan. Melalui kerjasama dengan dinas pemerintahan tentunya Pusat Kegiatan Belajar Masyarakat (PKBM) Raharja ini program yang bersifat legal dari pemerintah karena sudah ada perizinan sebelumnya. Keberadaan Pusat Kegiatan Belajar Masyarakat (PKBM) Raharja menarik simpati banyak orang dari anak-anak sampai orang dewasa. Hal serupa dikatakan oleh Nenden bahwa: “Respon dari pihak masyarakat bahwa dengan adanya pelatihan menjahit Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 486 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin ini mereka sangat senang dan mereka sangat berharap dengan adanya pelatihan ini dapat merubah kehidupan mereka menjadi lebih maju. Serta mereka juga banyak yang termotivasi bahwa pelatihan ini sangat bermanfaat bagi mereka apalagi dapat menggali potensi dari masing-masing masyarakat.” (wawancara dengan Nenden pada tanggal 10 November 2017) Berdasarkan pemaparan tersebut dalam melakukan pelatihan menjahit pihak PKBM Raharja menyelenggarakan pelatihannya ditempat yang strategis dan dapat dijangkau oleh masyarakat sehingga panitia yang lain dapat mendampingi setiap pelatihan menjahit dilaksanakan. Masyarakat yang mengikuti pelatihan ini sangatlah antusias dan dengan adanya pelatihan ini. Dengan dilakukannya pelatihan terhadap masyarakat agar masyarakat di daerahnya bisa lebih maju dan terarah sehingga masyarakat merasakan hasilnya secara tersendiri dan sangat bermanfaat bagi mereka. Ada beberapa masyarakat yang sudah membuka usahanya sendiri yaitu usaha menjahit yang mereka buka di rumah mereka itu semua berkat pelatihan menjahit yang dilakukan oleh PKBM Raharja, mereka banyak belajar dari adanya pelatihan ini dan mereka menjadi tidak merasa terbebani akan biaya dari pelatihan ini.Kebanyakan dari mereka, mereka menjadi tidak perlu mengalami kesulitan dalam melakukan kursus di luar untuk menjahit karena sudah ada pelatihan menjahit. Tidak jauh beda dengan pelatihan kursus di luar sana pelatihan di sini juga mereka diajarkan untuk bagaimana membuat pola dengan benar, mengukur, menggambar baju atau celana yang mereka inginkan dengan benar dan tidak asal- asalan. Semuanya dimulai dari titik nol yang diajarkan terhadap masyarakat yang berkeinginan untuk mengikuti pelatihan ini serta diberikan penghargaan berupa sertifikat bagi yang sudah mahir. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Program Pusat Kegiatan Belajar Masyarakat Raharja dalam Meningkatkan Keterampilan Masyarakat Sama halnya dengan apa yang di paparkan oleh Wiwin beliau mengatakan bahwa: “Pelatihan ini tidak jauh berbeda dengan tempat kursus diluar sana mereka sama-sama di berikan arahan dan gambaran bagaimana menjahit dengan benar dari mulai tahap dasar, yahap terampil, dan tahap mahir.” (wawancara dengan Wawan pada tanggal 10 November 2017) Dengan apa yang dipaparkan oleh Wiwin tersebut membuktikan bahwa pelatihan ini merupakan program PKBM yang harus diikuti oleh setiap masyarakat guna dalam meningkatkan perekonomian mereka. Masyarakat harus serius dalam melaksanaka pelatihan ini karena jika mereka tidak serius dalam melakukan ini maka semuanya akan menjadi hal yang sia-sia untuk mereka dan tidak akan membawa keuntungan bagi mereka. Akan tetapi selama ini masyarakat sangat serius dalam melaksanakan pelatihan ini karena mereka menyadari bahwa pelatihan ini sangatlah penting bagi mereka dengan adanya pelatihan dapat menambah penghasilan jika mereka akan membuka usahanya sendiri. 487 kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Pengawasan dalam mengelola program pelatihan ini diadakan guna untuk mengontrol jalannya program pelatihan agar berjalan sesuai dengan tujuan yang telah ditetapkan, pembinaan dilaksanakan secara langsung dan tidak langsung, pengawasan ini dilaksanaan oleh penyelenggara program kepada anggota-anggota pelaksana program pelatihan menjahit. Hal ini disampaikan oleh Nani selaku ketua yayasan PKBM Raharja: “Adanya pengawasan dalam pengelolaan program ini sangatlah penting bagi kelancaran berjalannya program pelatihan menjahit, anggota memerlukan pengawasan dari agar pelaksanaan tugasnya berhasil, akan tetapi juga tidak mengawasinya dengan ketat namun dengan demikian anggota tetap memberikan laporan guna pertanggung jawaban atas tugasnya.” (wawancara dengan Nani pada tanggal 12 November 2017) Kegiatan pelatihan menjahit ini dilakukan oleh masyarakat dari beberapa RW, yang dilaksanakan pada hari sabtu minggu dengan jadwal pelaksanaan pelatihan dan daftar nama peserta yang sudah tercantum sebagai anggota. Menurut Abdillah Fitri selaku tutor dari penjahitan, dijelaskan bahwa: “Pelatihan terdiri dari 30 orang yang masing-masing tidak mempunyai keterampilan menjahit, di sini belajar mulai dari awal yaitu tingkat dasar sampai mahir, awalnya ya susah tidak mengerti sama sekali tentang menjahit tetapi setelah praktik berlangsung para peserta pelatihan dapat melaksanakan dengan baik” (wawancara dengan Abdillah Fitri selaku tutor dari penjahitan pada tanggal 12 November 2017) Menurut Robbins (2000: 494-495)keterampilan merupakan kemampuan seseorang secara efektif untuk berinteraksi dengan orang lain maupun dengan rekan kerja. Seperti pendengar yang baik, menyampaikan pendapat secara jelas dan bekerja dalam satu tim. Dari pernyataan tersebut diketahui bahwa adanya program pelatihan berlangsung dengan jumlah peserta sebanyak 30 orang. Pertumbuhan jumlah angkatan kerja yang tidak sebanding dengan pertumbuhan lapangan kerja serta didukung dengan adanya krisis global dan gelombang PHK yang mengakibatkan kesenjangan sosial dan ekonomi masyarakat sehingga dalam persaingan di dunia kerja diperlukan kualitas manusia yang berkualitas dan produktif. Dengan adanya pelatihan menjahit ini merupakan salah satu langkah untuk dapat meningkatkan kualitas sumber daya manusia di bidang ketenagakerjaan melalui pelatihan menjahit diharapkan dapat meningkatkan kualitas dan produktifitas sumber daya manusia serta dapat meningkatkan kualitas ekonomi melaui bekerja di suatu perusaan atau digarment dan menciptakan lapangan kerja baru melalui bidang wira usaha mandiri. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Sebagaimana yang dikatakan oleh Abdillah Fitri dalam menjalankan program pemberdayaan masyarakat kelurahan 488 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin ini ada beberapa tahap yang harus dilakukan diantarnya: “Dalam menjalankan pelatihan ini masyarakat dibina terlebih dahulu, atau adanya sosialisasi terhadap masyarakat dari mulai tujuan adanya pelatihan, metodenya gimana, materi pelatihan yang diberikan bentuknya gimana, media yang diberikan seperti apa, penunjangpelatihannya harus jelas, dan yang terakhir harus ada evaluasi” (wawancara dengan Abdillah Fitri pada tanggal 12 November 2017) Berdasarkan wawancara tersebut maka beberapa tahap tersebut harus senantiasan dilakukan dengan tujuan pelatihan , metode pelatihan, materi pelatihan, media pelatihan, komponen penunjang, dan evaluasi. Dari keenam tahap tersebut penulis setuju karena dalam melakukan proses pemberdayaan masyarakat, masyarakat harus dibekali terlebih dahulu dalam hal tujuan sampai hasilnya seperti apa. Hal ini sejalan dengan teori pemberdayaan bahwa mengemukakan bahwa pemberdayaan meliputi tiga hal, yaitu pengembangan (enabling), memperkuat potensi atau daya (empowering), dan terciptanya kemandirian. Bertolak dari pendapat ini, berarti pemberdayaan tidak saja terjadi pada masyarakat yang tidak memiliki kemampuan, akan tetapi pada masyarakat yang memiliki daya yang masih terbatas, dapat dikembangkan hingga mencapai kemandirian Winarni (dalam Ambar Teguh 2004: 79). Hal ini sejalan dengan teori pemberdayaan bahwa tingkat berdaya masyarakat itu dilihat dari partisipasi masyarakat, kemandirian masyarakat, kemauan yang tinggi, jiwa integritas, dan siap untuk diberdayakan. Dalam tahap pemberdayaan juga menurut Suparjan dan Hempri S (2003: 44) dilakukan dengan: Pertama, meningkatkan kesadaran kritis atau posisi masyarakat dalam struktur sosial politik. Kemudian kesadaran kritis yang muncul diharapkan membuat masyarakat mampu membuat argumentasi terhadap berbagai macam eksploitasi serta sekaligus membuat pemutusan terhadap hal tersebut. Peningkatan kapasitas masyarakat harus diperhatiakan bahwa masalah kemiskinan bukan sekear masalah kesejahteraan sosial tetapi berkaitan dengan faktor politik, ekonomi sosial budaya dan keamanan, dan pemberdayaan juga perlu meningkatkan dengan pembangunan budaya masyarakat. Kedua,pelatihan memasakdenganmeningkatkan pendapatan ekonomi memiliki keanekaragaman dapat dilakukan dengan bekerja seperti berdagang, pengrajin, peternak, wiraswasta, dan sebagainya. Tentunya hal ini ditunjang dengan penggalian potensi dari setiap diri masing-masing. Dengan menggali potensi untuk mengasah keterampilan perlu adanya pelatihan yang maksimal. Sekarang ini banyak program-program dari pemerintah untuk meningkatkan perekonomian masyarakat agar tercipta kesejahteraan hidup. kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 489 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi 4 (2020) 475-496 490 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Sebagaimana yang dikatakan oleh Dina bahwa strategi yang dilakukan untuk menjadikan daya tarik untuk masyarakat sebagai berikut: “Strategi yang kami lakukan yaitu dengan mengenalkan makanan yang masyarakat belum pernah melakukannya, kemudian cara pengemasan yang baik dan benar, cara pemasaran, dan cara pengolahan modal dan keuntungan. Kemudian kami juga mendatangkan chef yang terkenal di media sosial, hal ini untuk meningkatkan daya tarik masyarakat” (wawancara dengan Dina pada tanggal 14 November 2017) “Strategi yang kami lakukan yaitu dengan mengenalkan makanan yang masyarakat belum pernah melakukannya, kemudian cara pengemasan yang baik dan benar, cara pemasaran, dan cara pengolahan modal dan keuntungan. Kemudian kami juga mendatangkan chef yang terkenal di media sosial, hal ini untuk meningkatkan daya tarik masyarakat” (wawancara dengan Dina pada tanggal 14 November 2017) Dalam hal ini pemberdayaan harus dilandasi strategi terhadap masyarakat dengan melakukan tahap-tahap pemberdayaan masyarakat menurut Suparjan dan Hempri S (2003: 44): Meningkatkan kesadaran kritis atau posisi masyarakat dalam struktur sosial politik. Hal ini berangkat dari asumsi bahwa sumber kemiskinan berasal dari konstruksi sosial yang ada pada masyarakat itu sendiri. Kesadaran kritis yang muncul diharapkan membuat masyarakat mampu membuat argumentasi terhadap berbagai macam eksploitasi serta sekaligus membuat pemutusan terhadap hal tersebut. Peningkatan kapasitas masyarakat. Dalam konteks ini perlu dipahami, bahwa masalah kemiskinan bukan sekedar masalah kesejahteraan sosial tetapi berkaitan dengan faktor politik, ekonomi sosial budaya dan keamanan. Pemberdayaan juga perlu meningkatkan dengan pembangunan budaya masyarakat. Sebagaimana dikatakan oleh Marni: “Alhamdulillah… setelah saya mengikuti kursus ini saya bisa membuka usaha catering, dan Alhamdulillah saya juga bisa mengkuliahkan anak saya jurusan tata boga” (wawancara dengan Marni salah seorang masyarakat, pada tanggal 15 November 2017) Dari paparan yang telah dijelaskan diatas bahwa banyak masyarakat yang merasakan mafaat dari adanya pelatihan kursus memasak ini, selain mereka belajar dari awal sampai bisa menjadi paham tata caranya mereka juga sudah bisa membuka luang usaha. Dengan adanya kursus memasak merupakan bentuk kuliner yang inovatif yang selalu berkembang dari tahun ke tahun bahkan manusia itu hidup tidak bisa terlepas dari makanan, Apalagi orang-orang Indonesia ini termasuk warga Negara pencinta kuliner. Dengan adanya program dari PKBM Raharja ini dapat memunculkan kreasi inovasi untuk menjadikan bentuk kuliner yang dapat membantu mengupayakan diversifikasi bahan-bahan pangan pokok masyarakat yang telah lama dicanangkan oleh pemerintah Indonesia dalam usaha mengatasi krisis bahan pangan nasional, dan sebagai upaya untuk meningkatkan penghasilan masyarakat Antapani. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Dampak pada Masyarakat setelah Mengikuti Pelatihan Berdasarkan hasil temuan peneliti dilapangan dengan sejumlah data-data yang didapat bahwasanya peran pemberdayaan masyarakat melalui program Pusat Kegiatan Belajar Masyarakat Raharja dalam memberdayakan masyarakat dengan dilakukannya pelatihan menjahit, dan memasak dalam meningkatkan kesejahteraan masyarakat sudah cukup baik walaupun sebaiknya ada yang perlu ditingkatkan, hal ini dapat terlihat dengan upaya-upaya yang dilakukan oleh pihak PKBM dalam mensukseskan program yang sedang dilakukan. Pelatihan ini dilakukan sebagai salah satu upaya untuk memberdayakan masyarakat Antapani, karena tanpa adanya pelatihan masyarakat tidak mempunyai keahlian dalam berbagai bidang apapun. Kemudian pelatihan ini juga dapat membantu terhadap masyarakat yang sedang mangalami penurunan dalam penghasilan mereka. Program pemberdayaan ini untuk meminimalisir angka kemiskinan di masyarakat Antapani, terutama dalam meningkatkan kesejahteraan ekonomi masyarakat. Di samping sebagai upaya mensejahterakan masyarakatnya yaitu dengan adanya program ini ada beberapa masyarakat yang kurang setuju karena kurang motivasi dalam jiwa individu masyarakat sehingga menimbulkan pro dan kontra antara kelurahan dan masyarakat. Namun dengan diadakannya berbagai program ekonomi sedikit dapat mengurangi ketidak setujuannya karena ada beberapa masyarakat yang sudah mengalami perubahan hidup yaitu dengan mendapatkan kesejahteraan hidup. Pelaksanaan proses dan pencapaian tujuan pemberdayaan melalui penerapan pendekatan pemberdayaan yang dapat disingkat menjadi 5 P, yaitu: Pemungkinan, Penguatan, Perlindungan, Penyokong dan Pemeliharaan (Suharto, 2010, hal 67) Hasil penelitian tersebut menunjukkan bahwasannya peran yayasan Raharja dalam memberdayakan masyarakat melalui program Pusat Kegiatan Belajar Masyarakat dalam meningkatkan kesejahteraan ekonomi tentu itu memberikan dampak yang baik terhadap kesejahteraan masyarakat karena segala aspek yang dilakukan oleh PKBM Raharja tersebut tujuannya untuk memberdayakan masyarakat agar taraf hidupnya lebih baik dan sejahtera. Untuk mencapai kesejahteraan hidup perlu adanya dorongan dalam setiap individu, karena berdayanya seseorang dilihat dari usaha itu sendiri dalam menperoleh kesejahteraan. Dengan mengikuti keterampilan, penyuluhan terhadap peningkatan mata pencaharian hal yang vital bagi masyarakat desa sebagai proses untuk memperbaiki dan memenuhi kebutuhan hidup. Karena kebutuhan hidup ditunjang dengan pendapatan perekonomian yang mengalami peningkatan dalam setiap harinya. 491 kin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi Program pemberdayaan masyarakat kelurahan untuk mendorong masyarakat untuk masyarakat berdaya, karena menurut Winarni dalam Ambar Teguh (2004: 79) bahwa pemberdayaan meliputi tiga hal, yaitu: pengembangan (enabling), memperkuat potensi atau daya (empowering), dan terciptanya kemandirian. Bertolak dari pendapat ini, berarti pemberdayaan tidak saja terjadi pada masyarakat yang tidak memiliki kemampuan, akan tetapi pada masyarakat yang memiliki daya yang masih terbatas, dapat dikembangkan hingga mencapai kemandirian. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Dampak pada Masyarakat setelah Mengikuti Pelatihan Maka dari itu dengan adanya program ini dapat memberikan pengaruh besar dalam meningkatkan kekuatan masyarakat, meningkatkan kemampuan terhadap masyarakat khususnya masyarakat Antapani mengalami perubahan sosial dalam peningkatan penghasilan karena keberhasilan program pemberdayaan masyarakat kelurahan dalam meningkatkan kesejahteraan ekonomi masyarakat. Program pemberdayaan masyarakat Antapani, dalam melakukan program-programnya menggunakan pendekatan pembangunan komunitas (community based development) dan partisipatif melalui tiga kegiatan atau tribina yaitu bina sosial, bina ekonomi dan bina fisik lingkungan yang fungsinya untuk melaksanakan paradigma baru pembangunan, yakni "masyarakat membangun" dan bukan lagi "membangun masyarakat" namun yang lebih diprioritaskan yaitu tri bina ekonomi masyarakat. Faktor pendukung dalam program pemberdayaan masyarakat yang dilakukan oleh PKBM, dengan di dorong dengan berbagai upaya maka ada hasil penunjang dari berjalannya suatu program. Hasil sebuah program di dasari dengan faktor-faktor pendukung dan faktor penghambat. Dalam membicarakan hasil program yang dilakukan oleh kelurahan ada beberapa faktor pendukung untuk masyarakat mendapatkan hasil yang bisa mereka pergunakan untuk membuka usaha baik di rumah mereka atau pun dilingkungan yang lainnya. Sebagaimana yang dikatakan oleh Nandang keberadaan program ini dapat berarti masyarakat karena telah terbantu dengan adanya pelatihan menjahit dan pengolahan makanan dengan kata lain masyarakat terbantu. “Alhamdulillah.. dengan adanya program PKBM Raharja ini dapat membantu perekonomian masyarakat Antapani, yaa.. walaupun sedikit tapi dapat membantu masyarakat dalam menambah penghasilan terutama dalam kesejahteraan ekonomi”. (wawancara dengan Nandang pada tanggal 13 November 2017) “Alhamdulillah.. dengan adanya program PKBM Raharja ini dapat membantu perekonomian masyarakat Antapani, yaa.. walaupun sedikit tapi dapat membantu masyarakat dalam menambah penghasilan terutama dalam kesejahteraan ekonomi”. (wawancara dengan Nandang pada tanggal 13 November 2017) Berdasarkan hasil wawancara tersebut bahwa kesejahteraan ekonomi masyarakat mengalami peningkatan dengan banyak manfaat yang dirasakan oleh masyarakat yang mengikuti pelatihan ini. Setidaknya masyarakat menjadi ada peluang untuk membuka usaha baik itu usaha menjahit atau memasak karena 492 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin mereka sudah mendapat pelatihan. Masyarakat mendapatkan penghasilan tambahan untuk mencukupi kebutuhan keluarganya dengan membuka usaha menjahit dan makanan, dapat menambah penghasilan bagi pemilik usaha maupun pegawainya dan, penghasilan keluarga ada peningkatan. Faktor penghambat pemberdayaan masyarakat di berbagai bidang tidak terlepas dari berbagai hambatan yang menyertainya. Hambatan yang sering muncul adalah sulitnya untuk mensinergiskan berbagai pemberdayaan itu dalam suatu program yang terpadu. Dengan memusatkan pada satu dimensi, pengembangan akan mengabaikan kekayaan dan kompleksitas kehidupan manusia dan pengalaman masyarakat. Tidak ada alasan untuk mengatakan bahwa berbagai tindakan untuk memberdayakan masyarakat tidak bisa disinergiskan. Dampak pada Masyarakat setelah Mengikuti Pelatihan Keberhasilan suatu program di dorong dengan faktor-faktor pendukung dan faktor penghambat. Sebagaimana yang dikatakan oleh Nenden beliau mengatakan bahwa: “Faktor penghambat dari berjalannya program ini yaitu minimnya informasi yang diadakan, kemudian anggotapun terbatas kemudian ada juga masyarakat yang tidak mendukung program ini, karena dengan kegiatan yang ada tidak menjamin akan peningkatan perekonomian mereka” (wawancara dengan Nenden pada tanggal 13 November 2017 ) “Faktor penghambat dari berjalannya program ini yaitu minimnya informasi yang diadakan, kemudian anggotapun terbatas kemudian ada juga masyarakat yang tidak mendukung program ini, karena dengan kegiatan yang ada tidak menjamin akan peningkatan perekonomian mereka” (wawancara dengan Nenden pada tanggal 13 November 2017 ) Berdasarkan hasil wawancara tersebut pihak yayasan mengalami kesulitan dalam melakukan perekrutan, karena memang banyak yayasan yang menyelenggarakan program seperti ini. Meskipun telaahan mengenai program pemberdayaan ada faktor penghambat dalam pelaksanaan program dan ketidakberhasilan kelompok sasaran untuk mencapai tujuan namun harus diakui juga bahwa ada banyak program pemberdayaan yang berhasil dan mencapai tujuan yang ditetapkan. Kendala yang terjadi dalam pelaksanaan program pemberdayaan dapat berasal dari kepribadian individu dalam komunitas dan bisa juga berasal dari sistem sosial. Namun dalam program pemberdayaan masyarakat Antapani ini terlihat berhasil karena tingkat keberhasilan ini banyak dirasakan oleh banyak masyarakat yang mengalami kesejahteraan ekonomi dalam kehidupannya. Proses memberdayakan masyarakat berarti melindungi dalam proses pemberdayaan sehingga yang lemah tidak boleh menjadi bertambah lemah karena kekurangberdayaan dalam meghadapi yang kuat (Jamaludin, 2015: 246-247). Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 PENUTUP Berdasarkan penelitian tentang Peran Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin dapat disimpulkan yaitu: Pertama, usaha yang dilakukan Pusat Kegiatan Belajar Masyarakat Raharja dalam meningkatkan keterampilan masyarakat miskin melalui program dengan 493 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi dilakukannya beberapa jalur penerapan di antaranya Persiapan, Pendataan dan Pelaporan, Pemetaan, Pelaksanaan, Analisis, Program dukungan dan Evaluasi. Dengan melakukan usaha dari setiap program-program pemberdayaan yang dilakukan PKBM Raharja yaitu dalam aspek pendidikan dan ekonomi. Kedua, program yang dilakukan oleh pusat kegiatan belajar masyarakat yaitu dengan adanya pelatihan menjahit dan memasak. Pelatihan keterampilan menjahit yang dilaksanakan oleh PKBM Raharja merupakan kegiatan dalam program yang dilakukan agar dapat mendapatkan keahlian khusus untuk yang belum memiliki skill atau keterampilan praktis di bidang menjahit sehingga memiliki bekal keterampilan sebagai penunjang penciptaan lapangan pekerjaan yang akan menjadi mata pencaharian untuk kesiapan berwirausaha agar pendapatan masyarakat dapat mengalami peningkatan yang cukup signifikan. Dalam tahap ini dilakukan dengan beberapa cara seperti dikenalkannya tujuan pelatihan, metode pelatihan, materi pelatihan, media pelatihan, komponen penunjang, dan evaluasi. Kemudian yang kedua pelatihan memasak pelatihan ini bertujuan untuk menggali potensi seperti ibu-rumah tangga dalam menggali potensinya dalam memasak. Ketiga, dampak dari program pemberdayaan masyarakat melalui Pusat Kegiatan Belajar Masyaraakat dalam meningkatkan keterampilan masyarakat miskin telah berhasil dilakukan, banyak masyarakat yang mengembangkan potensinya dalam mengikuti pelatihan yang di lakukan oleh pihak yayasan. Pelatihan menjahit, dan pelatihan memasak sehingga menjadikan masyarakat lebih berdaya dan mengalami kesejahteraan hidup dari sebelumnya. Dengan adanya upaya meminimalisir angka kemiskinan program yang ada di pusat kegiatan belajar masyarakat perlu dimanfaatkan dengan sebaik mungkin, karena program yang ada di PKBM salah satu program untuk meningkatkan keterampilan masyarakat miskin dalam peningkatan kualitas hidup, agar keberdayaan dan kesejahteraan hidup dapat tertutupi dengan ilmu dan pendapatan yang mereka miliki. Apalagi pusat kegiatan belajar masyarakat ini mampu menaungi banyak masyarakat yang berkeinginan untuk belajar dalam menggali potensi dari setiap individu. Pemberdayaan dapat memberikan kekuatan yang dilakukan oleh individu atau masyarakat agar masyarakat dapat berdaya guna dalam memenuhi kebutuhan hidupnya kearah yang lebih sejahtera. Pemberdayaan meliputi tiga hal, yaitu pengembangan (enabling), memperkuat potensi atau daya (empowering), dan terciptanya kemandirian. Artinya pemberdayaan tidak saja terjadi pada masyarakat yang tidak memiliki kemampuan, akan tetapi pada masyarakat yang memiliki daya yang masih terbatas, dapat dikembangkan hingga mencapai kemandirian. Kemandirian masyarakat itu sendiri dapat dilihat dari segi pola fikir, tingkah laku yang bisa mengontrolkan diri atau mengendalikan diri sesuai dengan apa yang 494 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. PENUTUP 5 No. 4 (2020) 475-4 Pemberdayaan Masyarakat Melalui PKBM untuk Meningkatkan Keterampilan Masyarakat Miskin akan mereka lakukan. Manusia dapat dikatakan mandiri jika mereka hidup layak dengan mampu memenuhi Kebutuhan Hidup Minimal (KHM). Kebutuhan hidup minimal dimaksud meliputi sandang, pangan, papan, pendidikan, kesehatan mental, kesehatan fisik, identitas kepribadian, ekspresi diri, kontak sosial, keyakinan, kebebasan memilih, keadilan, pendidikan, ekonomi, mencintai dan dicintai, pengakuan sosial dan pujian, dan pikiran. Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 Aceng Ibrahim, Bachrun Rifa’i, Ratna Dewi DAFTAR PUSTAKA Hempri, S. D. (2003). Pengembangan Masyarakat dari Pembangunan sampai Pemberdayaan. Yogyakarta: Aditya Media. Huda, M. (2009). Pekerjaan Sosial dan Kesejahteraan Sosial. Yogyakarta: Pustaka Pelajar. Jamaludin, A. N. (2015). Sosiologi Pedesaan. Bandung: CV Pustaka Setia. Lubis Z. 2006. Penanggulangan Kemiskinan, diakses 22 November 2017, dari www.waspada.co.id Roesmidi, R. R. (2006). Pemberdayaan Masyarakat. Sumedang: Alqaprint Jatinangor. Setiawan, I. A. (2012) Dakwah Berbasis Pemberdayaan Ekonomi dan Peningkatan Kesejahteraan Mad’u. Ilmu Dakwah: Academic Journal for Homiletic Studies, 6 (2), 347-262 Suharto, E. (2005). Membangun Masyarakat Memberdayakan RakyatKajian Strategis Pembangunan Kesejahteraan Sosial dan Pekerjaan Sosial. Bandung: PT Reflika Aditama. Suharto, E. (2010). Membangun Masyarakat Memberdayakan Rakyat.Bandung: Refika Aditama. Teguh, A. (2004). Kemitraan dan Model-Model Pemberdayaan.Yogyakarta: Gava Media. Teguh, A. (2004). Kemitraan dan Model-Model Pemberdayaan.Yogyakarta: Gava Med 495 Tamkin: Jurnal Pengembangan Masyarakat Islam Vol. 5 No. 4 (2020) 475-496 496 496
W4281728933.txt
https://www.frontiersin.org/articles/10.3389/fimmu.2022.892618/pdf
en
Correlation of HBV DNA and Hepatitis B Surface Antigen Levels With Tumor Response, Liver Function and Immunological Indicators in Liver Cancer Patients With HBV Infection Undergoing PD-1 Inhibition Combinational Therapy
Frontiers in immunology
2,022
cc-by
9,487
ORIGINAL RESEARCH published: 25 May 2022 doi: 10.3389/fimmu.2022.892618 Edited by: Feng Li, Guangzhou Eighth People’s Hospital, China Reviewed by: Jin Hou, Second Military Medical University, China Xiaoyong Zhang, Southern Medical University, China *Correspondence: Fu-Sheng Wang fswang302@163.com Fanping Meng drmengfanping@126.com Ming Shi shiming302@sina.com † These authors have contributed equally to this work and share first authorship ‡ These authors share senior authorship Specialty section: This article was submitted to Viral Immunology, a section of the journal Frontiers in Immunology Received: 09 March 2022 Accepted: 25 April 2022 Published: 25 May 2022 Citation: Pan S, Yu Y, Wang S, Tu B, Shen Y, Qiu Q, Liu X, Su N, Zuo Y, Luan J, Zhang J−Y, Shi M, Meng F and Wang F-S (2022) Correlation of HBV DNA and Hepatitis B Surface Antigen Levels With Tumor Response, Liver Function and Immunological Indicators in Liver Cancer Patients With HBV Infection Undergoing PD-1 Inhibition Combinational Therapy. Front. Immunol. 13:892618. doi: 10.3389/fimmu.2022.892618 Correlation of HBV DNA and Hepatitis B Surface Antigen Levels With Tumor Response, Liver Function and Immunological Indicators in Liver Cancer Patients With HBV Infection Undergoing PD-1 Inhibition Combinational Therapy Shida Pan 1,2†, Yingying Yu 2,3†, Siyu Wang 1,2†, Bo Tu 2, Yingjuan Shen 2, Qin Qiu 2, Xiaomeng Liu 2, Nan Su 2,4, Yanmei Zuo 1,2, Junqing Luan 2, Ji−Yuan Zhang 2, Ming Shi 2,3*‡, Fanping Meng 1,2*‡ and Fu-Sheng Wang 1,2*‡ 1 Chinese People's Liberation Army (PLA) Medical School, Beijing, China, 2 Department of Infectious Diseases, The Fifth Medical Center of Chinese People's Liberation Army (PLA) General Hospital, Beijing, China, 3 Peking University 302 Clinical Medical School, Beijing, China, 4 The Second School of Clinical Medicine, Southern Medical University, Guangzhou, China Background: Thus far, few studies have investigated the safety and efficacy of programmed death-1 (PD-1) immune checkpoint inhibitors (ICIs) and tyrosine kinase inhibitors (TKIs) antibodies in patients with hepatitis B virus (HBV)-related liver cancer. Objective: To investigate the effect of combination therapy with programmed death-1 (PD-1) immune checkpoint inhibitors (ICIs) and tyrosine kinase inhibitors (TKIs) on HBVrelated liver cancer. Methods: Until January 31, 2022, liver cancer patients with hepatitis B surface antigen (HBsAg) or HBV DNA positivity, treated with PD-1 ICIs and TKIs combined with nucleoside analogs (NAs), were retrospectively reviewed. The correlation between the change in HBV DNA and HBsAg levels and tumor response was analyzed using the c2 test. Cox univariate and multivariate survival analyses and Kaplan–Meier curves were used to identify and compare risk factors and overall survival (OS). Results: A total of 48 patients were enrolled in the study, with an objective response rate (ORR) of 31.3%, a disease control rate (DCR) of 66.7%; the incidence of adverse events was mostly mild. A significant decrease in HBV DNA and HBsAg levels was observed at 12 and 24 weeks compared with the baseline (p < 0.05). Compared to patients with progressive disease (PD), patients with disease control showed a more significant decrease in HBV DNA and HBsAg levels at 12 and 24 weeks (p < 0.001). Eleven patients showed elevations in HBV DNA level and one of them showed HBV reactivation; however, the reactivation was not associated hepatitis. Moreover, eight patients showed elevation in HBsAg. Elevation in HBV DNA level was associated with poor tumor response Frontiers in Immunology | www.frontiersin.org 1 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 (P=0.001, OR=18.643 [95% CI: 3.271–106.253]). Cox survival analysis suggested that HBV DNA increase (P=0.011, HR=4.816, 95% CI: 1.439–16.117) and HBsAg increase (P=0.022, HR=4.161, 95% CI: 1.224–16.144) were independent risk factors associated with survival time. Kaplan–Meier curves suggested that patients who exhibited an increase in HBV DNA (6.87 months vs undefined, log-rank test: p= 0.004) and HBsAg (8.07 months vs undefined, log-rank test: p= 0.004) levels had a shorter median survival time (MST). Patients without increased HBsAg showed better baseline liver function and routine blood tests (p<0.05) than patients with increased HBsAg. An increase in C-reactive protein (CRP) and interleukin-6 (IL-6), and a decrease in T lymphocytes, CD4+ T lymphocytes, and B lymphocytes at 1-week post-treatment associated with HBsAg well-controlled. Conclusion: HBV-related liver cancer patients treated with combination therapy showed improved efficacy and safety profiles. Combination therapy has some effect on HBV infection, and a correlation between tumor response and antiviral efficacy was found. Elevation of HBV DNA and HBsAg levels may indicate poorer tumor response and survival time. Better baseline liver function and early immune activation may be associated with decline in HBsAg levels. Keywords: hepatitis B virus, hepatitis B surface antigen, HBV reactivation, median survival time, C-reactive protein, interleukin-6 monotherapy in liver cancer cases due to the promising results in clinical trials (12–14). Although phase I/II trials have shown good efficacy and safety, phase III trials were less favorable. Although some patients have durable response to monotherapy, the benefit remains low, with an objective response rate (ORR) ranging from 15% to 20%. Therefore, combination therapies have been developed to improve ORR. Commonly used combination therapies include ICIs, targeted drugs, radiotherapy, chemotherapy, and interventional therapies (15–19). The efficacy and safety of the combination of atezolizumab and bevacizumab used in phase III IMbrave150 trial were higher compared to those obtained with sorafenib, with the results indicating a higher 12month survival rate (67.2% vs. 54.6%) and median progressionfree survival (PFS) (6.8 months vs 4.3 months); however, the incidence of immune-related adverse events (irAEs) was similar (20). Although ICI treatment has shown better efficacy than sorafenib in liver cancer, there are few reports on its safety and efficacy in patients with HBV-associated liver cancer, especially in those with high viral loads. This is due, firstly, to the HBV viral load that is associated with carcinoma recurrence and prognosis. Patients with HBV DNA > 105 IU/mL before transplantation showed frequent HCC (hepatocellular carcinoma) recurrence after transplantation (21). Patients with a HBV DNA load of < 2000 IU/ mL have longer recurrence-free survival (RFS) rate, and a complete virologic response after operative therapy is associated with a lower risk of cancer recurrence (22). High serum levels of HBV DNA before transarterial chemoembolization (TACE) are associated with poor overall survival (OS) and rapid progression of HCC after TACE (23). Except surgical treatment, the HBV viral load was also associated with poor prognosis in patients receiving systemic INTRODUCTION According to a status report on the cancer burden worldwide, published by the International Agency for Research on Cancer (IARC), liver cancer was predicted to be the sixth most commonly diagnosed cancer and the fourth leading cause of cancer-related deaths worldwide in 2018, with approximately 840,000 new cases and 780,000 deaths per year (1). However, more than half of the new cases and deaths occurred in China (2). Until 2017, liver cancer was the sixth most commonly diagnosed cancer and the fourth leading cause of cancer in China (3).The common causes of liver cancer include hepatitis B virus (HBV) and hepatitis C virus (HCV) infection, aflatoxin contamination, alcohol abuse, obesity, smoking, and type 2 diabetes (4). Common causes of liver cancer vary with region. In developed countries, HCV infection and alcohol abuse contribute the most to liver cancer; in China, the key etiology is chronic HBV infection, with nearly 85% of liver cancer patients having previous or current HBV infection (5). The 5year survival rate of liver cancer patients worldwide is approximately 18%, while that in China is only 12.1% (6, 7). Most liver cancers are always detected in the middle or late stages, resulting in a poor prognosis. Currently, tyrosine kinase inhibitors (TKIs) such as sorafenib and lenvatinib are recommended as first-line treatments for advanced liver cancer (8, 9). However, based on the existing studies, the survival benefits of sorafenib administration are limited (10, 11). Immune checkpoint inhibitors (ICIs), such as PD-1 and PD-L1, provide a new option for the treatment of patients with liver cancer. Several ICIs, such as nivolumab, pembrolizumab, and camrelizumab, have been approved for Frontiers in Immunology | www.frontiersin.org 2 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 function of HBV-specific CD8+ T cells and promoted HBsAg clearance (39–41). Nivolumab, with and without therapeutic vaccination, when used on serum-positive chronic hepatitis B (CHB) patients, resulted in HBsAg reduction and clearance in some patients (42). Although PD-1/PD-L1 ICIs are effective in both tumor control and HBV infection, no study has investigated whether there is an association between tumor response and viral reduction. It is also unknown whether baseline characteristics and dynamics of clinical parameters during PD-1/PD-L1 treatment are associated with decrease HBsAg levels. In our study, liver cancer patients did not need to maintain lower HBV DNA levels when subjected to combined treatment with PD-1 ICIs and TKIs, but concurrent initiation of nucleoside analog (NA) antiviral therapy was required. The aim of the study is to investigate the efficacy, safety (incidence of adverse events and hepatitis related to HBV reactivation) and impact on HBV DNA and HBsAg levels of combination therapy, and to determine the correlation between HBV replication or reactivation with patient response and survival. therapy. Poor OS was recorded for patients with detectable HBV DNA levels after antiangiogenic therapy; the baseline HBV DNA level was an independent predictor of poor OS (24). High baseline HBV load was an adverse prognostic factor for survival in patients treated with sorafenib (25). A high viral load before treatment was an exposure risk factor for survival during systemic chemotherapy and was associated with a higher incidence of severe hepatitis during treatment (26). In most clinical trials for PD-1/PD-L1 ICIs, liver cancer patients with HBV infection were required to have a baseline HBV load <100 IU/mL for eligibility; therefore, the effect of baseline viral load on clinical outcomes could not be assessed. However, several studies have enrolled patients with a high viral load, suggesting an absence of correlation between baseline HBV viral load and patient OS and PFS during PD-1/PD-L1 treatment; however, the presence or absence of antiviral therapy during immunotherapy could affect tumor prognosis (27, 28). Secondly, HBV reactivation could affect the use of PD-1/PDL1 ICIs in HBV-related liver cancer patients. In HBV-related liver cancer patients, common triggers for HBV reactivation include surgery, radiation therapy, chemotherapy, and immunosuppressive drugs. For example, in patients with HBV reactivation and HBV reactivation-associated hepatitis after chemotherapy, use of prophylactic antiviral drugs were independent risk factors for HBV reactivation (29, 30). In patients treated with partial hepatectomy, radiofrequency ablation, and interventional therapy, HBV reactivation and HBV reactivation-related hepatitis may be associated with the immunosuppressive state caused by surgery and anesthesia as well as surgical injury of the liver (31–33). HBV reactivation has also been reported in PD-1PD-L1 antibody therapy. In an Asian cohort of CheckMate 040, patients were subjected to antiviral therapy to achieve an HBV DNA load of ≤ 100 IU/mL; however, five patients exhibited an increase of > 1log10 IU/mL of HBV DNA after PD-1 ICI immunotherapy but did not develop hepatitis (34). Studies that included patients with an HBV DNA load of >100 IU/mL found that the baseline viral load had no effect on HBV reactivation, but antiviral therapy was an independent risk factor for HBV reactivation (35, 36). Although PD-1PD-L1 ICIs exhibit good efficacy and safety in patients with HBV-related liver cancer, there are a few shortcomings that need to be addressed. For example, the load of HBV in patients enrolled in studies is usually low (< 2000 IU/mL), and the efficacy and safety of the drugs against high viral loads (> 2000 IU/mL) remains unclear. For patients with high viral loads, is it necessary to keep the viral load level low before initiating PD-1/PD-L1 antibody immunotherapy? Does the presence of HBV re-replication or reactivation indicate a poor prognosis for patients compared with those with a decrease in HBV DNA? To address these questions, further studies are required. For patients with HBV-related liver cancer, PD-1/PD-L1 antibodies not only showed good tumor response efficacy but also played a role in antiviral treatment of HBV infection (37, 38). In some pre-clinical studies, PD-1 was highly expressed on HBV-specific CD8+ T cells in virally persistently infected mice than in HBsAg-clear mice; PD-1 antibody treatment restored the Frontiers in Immunology | www.frontiersin.org MATERIALS AND METHODS Patients’ Inclusion Between March 2020 and January 2022, patients diagnosed with liver cancer through histological or imaging examination, with HBV DNA or HBsAg seropositivity and treated with at least one cycle of PD-1 ICIs combined with TKIs, were retrospectively reviewed. All patients were treated in the Department of Infectious Diseases, Fifth Medical Center of PLA General Hospital. Sintilimab (Innovent Biologics and Eli Lilly and Company) and camrelizumab (Jiangsu Hengrui Medicine Co. Ltd.), at a fixed dose of 200 mg, and toripalimab at 240 mg were prescribed for a three week cycle. Sorafenib (Bayer Schering Pharma AG) was administered at 400 mg/day, lenvatinib (Eisai Co. Ltd) at 8 to 12 mg/day and regorafenib (Bayer) at 80 mg/day, according to body weight. Treatment options for patients who met the criteria for TACE treatment were decided together by the patient and the physician before or after systemic combination therapies. NAs including entecavir (ETV) (0.5 mg/day), tenofovir (TDF) (300 mg/day), and adefovir dipivoxil (ADV) (10 mg/day) were provided simultaneously with immunotherapy. Radiological response was recorded using computed tomography (CT) or magnetic resonance imaging (MRI) initially as the baseline and then at 3 month intervals. Tumor response was assessed by the modified Response Evaluation Criteria in Solid Tumors (mRECIST) (43, 44). Taking the baseline sum of the diameters of target lesions as a reference, complete response (CR) was defined as the disappearance of any intratumoral arterial enhancement in all target lesions; at least a 30% decrease in the sum of diameters of target lesions was defined as a partial response (PR); an increase in the sum of the diameters of at least 20% was defined as progressive disease (PD); and no change in the sum of the diameters between PR and PD was defined as stable disease (SD). IrAEs were documented according to Version 5 of the Common Terminology Criteria 3 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 a normal distribution. The c2 test or Fisher’s exact test were used to compare categorical data. Comparisons between the two groups were performed using the nonparametric Mann– Whitney U test. Comparisons within groups were analyzed using nonparametric Wilcoxon’s paired test. The cut-off date for survival analysis was January 31, 2022, and OS was estimated using Kaplan–Meier curves and compared using the log-rank test. Hazard ratios (HRs) and 95% confidence intervals (CIs) were analyzed using Cox univariate and multifactorial survival analyses to identify any independent predictive factors that were associated with OS. All figures were generated using GraphPad Prism statistical software (version 9.0; GraphPad Software, San Diego, CA, USA). Differences were considered statistically significant at p < 0.05. for Adverse Events (CTCAE) (45). Patients with spontaneous bacterial peritonitis, a complication of primary liver cancer, were excluded from this study. Depending on the toxicity grade, TKI dose was reduced, suspended, discontinued, or switched to other TKIs in patients who developed irAEs associated with TKI treatment. PD-1 ICIs were suspended or discontinued, and immunosuppressant agents were administered based on the severity of irAEs. HBV reactivation was defined according to the American Association for the Study of Liver Diseases (AASLD) hepatitis B guidance in 2018 for HBsAg-positive and anti-hepatitis B core (HBc)-positive patients, including (1) an increase in HBV DNA ≥ 2 log (100-fold) compared to the baseline level, (2) HBV DNA ≥ 3 log (1000) IU/mL in a patient with previously undetectable levels, or (3) HBV DNA ≥ 4 log (10,000) IU/mL if the baseline level was not available. HBsAg-negative and antiHBc-positive patients who met the following criteria were considered to have HBV reactivation hepatitis: detectable HBV DNA or reverse serological conversion of HBsAg or HBV reactivation accompanied with an increase in alanine aminotransferase (ALT) ≥ 3 times the baseline level and absolute value > 100 U/L (46). The cut-off date for follow-up was January 31, 2022, and all data were obtained from patient medical records. The baseline data included patient demographics, such as Child-Pugh stage, Barcelona Clinic Liver Cancer (BCLC) stage, serum alphafetoprotein (AFP) level, type of combination therapy, and TACE treatment. The data pertaining to the levels of HBV DNA and HBsAg, routine blood tests, liver function, lymphocyte subsets, C-reactive protein (CRP), and interleukin6 (IL-6) at baseline and after 1 week were collected. HBV DNA was quantified using real-time fluorescence PCR (Roche COBAS, USA) with a lower limit of 20 IU/mL. HBsAg was quantified using a Beckman AU5800-03 with a lower limit of 0.05 IU/ml. Serum CRP and liver function parameters were measured using an automatic biochemical analyzer (AU5400; Beckman Coulter, Brea, CA), with the upper limit of the normal value for serum CRP set at 8.2 mg/L. Serum IL-6 was examined using a Roche Cobas 8000 (Roche Diagnostics GmbH, Mannheim, Germany), and the upper limit of the normal value for serum IL-6 was set at 7 pg/mL. Lymphocyte subsets and counts were measured using a FACSCalibur flow cytometer (BD Biosciences, Becton, NJ). Routine blood parameters were measured using an automatic hematology analyzer (HN-2000 series; SYSMEX, Kobe, Japan). RESULTS Baseline Characteristics of Patients A total of 48 patients were enrolled in this study: 41 with HCC and 7 with cholangiocarcinoma. Thirty-four patients were treated with a combination of sintilimab and lenvatinib, while 14 patients received other combination therapies. Twelve patients underwent TACE during the treatment period. Thirty-three patients had received antiviral therapy prior to combination therapy and 15 patients had never received antiviral therapy. Fifteen patients had baseline HBV DNA levels below the lower limit of detection, 16 patients had HBV DNA levels >1000 IU/mL, and other patients had HBV DNA levels between 20 IU/mL and 1000 IU/mL. The baseline HBsAg level of five patients was not available; one patient was seronegative for baseline HBsAg but with HBV DNA > 20 IU/mL, and 45 patients had HBsAg levels > 0.5 IU/mL. Other details are listed in Table 1. Immune-Related Adverse Events and Tumor Response Profiles According to mRECIST, two (4.2%) patients had CR, 13 (27.1%) had PR, 17 (35.4%) had SD, and 16 (33.3%) had PD, with an ORR of 31.3% and Disease control rate (DCR) of 66.7%, as shown in Table 1. There were 22 (45.8%) patients who experienced at least one adverse event of any grade, and 12 patients (25%) developed grade 3 (G3)/grade 4 (G4) irAEs. The most common adverse events in G3/G4 irAEs were fever (n = 9, 18.8%), fatigue (n = 7, 14.6%), lymphopenia (n = 2, 4.2%), bacterial infection (n = 2, 4.2%), diarrhea (n = 2, 4.2%), and hepatitis (n = 2, 4.2%). All patients who experienced severe irAEs improved after receiving glucocorticoids according to clinical guidelines. The overall occurrence of adverse events in the patients is shown in Table 2. During treatment, four patients permanently discontinued sintilimab, three due to irAEs, and one due to tumor progression; five patients had their TKI dose reduced due to adverse AEs; six patients discontinued TKIs, two due to AEs, and four due to complications such as upper gastrointestinal bleeding. Twelve patients switched to regorafenib, three due to AEs, and nine due to tumor progression. Statistical Analysis All statistical analysis were conducted using the IBM SPSS Statistics software application (version 25.0; IBM, Armonk, NY, USA). The baseline data and adverse events were summarized using descriptive statistics. HBV DNA and HBsAg levels below the lower limit were recorded as the lower limit of detection. Continuous variables that conformed to a normal distribution were represented as mean ± SD, non-normal distribution data were presented using median (quartiles), and categorical data by variable numbers (percentages). The student t test was used to compare continuous variables that conformed to Frontiers in Immunology | www.frontiersin.org 4 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 TABLE 1 | Baseline characteristics of patients enrolled in the study. Antiviral Efficacy of Combination Therapy Patients were treated with a combination therapy of PD-1 ICIs and TKIs. All patients received antiviral therapy with NAs and were followed up for more than 24 weeks. One patient was HBsAg negative and HBV DNA positive at the start of treatment. One patient died within 12 weeks. The follow-up time of 12 patients was more than 12 weeks but not more than 24 weeks, during which time, seven patients died, two were lost to followup, and three were still subject to follow-up. Fifteen patients had baseline HBV DNA levels below the lower limit of detection, of which three patients had positive HBV DNA after 24 weeks of treatment, and one of three patients reached HBV reactivation. Twelve patients with baseline HBV DNA positivity had HBV DNA clearance at 12 weeks and five at 24 weeks. HBsAg changes could not be assessed in nine patients at 12 or 24 weeks due to lack of baseline data. Compared with the established baseline levels, overall HBV DNA and HBsAg levels showed a statistically significant difference at 12 and 24 weeks (p < 0.05) (Figure 1). To investigate the correlation between patient tumor response and virus control, we analyzed the changes in HBV DNA and HBsAg levels in patients with different tumor responses. The results suggested that there was no statistically significant difference in HBV DNA and HBsAg levels between patients with tumor control (including CR, PR, SD) and tumor progression (PD), and no significant difference was observed among patients with CR plus PR, SD, and PD (p > 0.05) (Figures 2A, D). Compared to patients with PD, patients with tumor control showed a significant decline in HBV DNA and HBsAg levels at baseline, 12 weeks, and 24 weeks (p<0.05) (Figures 2B, C). In addition, we divided patients with DCR into CR plus PR and SD groups and analyzed the changes in HBV DNA and HBsAg levels in the two groups. The results showed that patients with CR plus PR and SD had similar changes in HBV DNA levels at baseline, 12 weeks, and 24 weeks. However, the change in HBsAg levels was more significantly different in CR plus PR patients than in SD patients (Figures 2E, F). n = 48 Age (year) Sex Male Female Types of tumors hepatocellular carcinoma Cholangiocarcinoma BCLC stage B C (PVTT) C (M) Child-Pugh stage A B AFP <400 (IU/ml) ≥400 (IU/ml) Previous antiviral therapy Yes No HBV DNA ≤20 (IU/ml) >20 and ≤10^3 (IU/ml) >10^3 (IU/ml) HBsAg ≤0.05 (IU/ml) >0.05 (IU/ml) No available HBeAg state Seropositive Seronegative No available Combination treatment Sintilimab+Lenvatinib Sintilimab+Sorafenib Camrelizumab+Lenvatinib Toripalimab+Lenvatinib Combinated TACE therapy Yes No Combination treatment as systemic First line Second line Third line Fourth line Antiviral therapy ETV ADV TDF ETV+TDF Tumor response CR PR SD PD OR (CR+PR) DC (CR+PR+SD) 55.96 ± 9.72 41 (85.4%) 7 (14.6%) 43 (89.6%) 5 (10.4%) 12 (25.0%) 22 (45.8%) 14 (29.2%) 26 (54.2%) 22 (45.8%) 27 (56.2%) 21 (43.8%) 33 (68.7%) 15 (31.3%) 15 (31.2%) 17 (34.5%) 16 (33.3%) 1 (2%) 42 (87.6%) 5 (10.4%) 9 (18.8%) 38 (79.2%) 1 (2%) 34 (70.8%) 7 (14.6%) 3 (6.3%) 4 (8.3%) 12 (25.0%) 36 (75%) 21 (43.8%) 5 (10.4%) 16 (33.3%) 6 (12.5%) Correlation Between Elevated HBV DNA, HBsAg Levels, Tumor Response, and Survival Time After 24 weeks of treatment, 11 patients had elevated HBV DNA and ALT or aspartate aminotransferase (AST) levels, one of whom had HBV reactivation; however, this patient did not develop HBV reactivation-related hepatitis. Two patients developed hepatitis, but it was not associated with HBV reactivation. The characteristics of the 11 patients with elevated HBV DNA levels are shown in Table S1. In addition, 8 patients suffered from elevated HBsAg levels, the characteristics of which are shown in Table S2. Nine of the 11 patients who suffered from HBV DNA elevation had PD, one patient had SD, and one patient had PR. Four of the eight patients with elevated HBsAg levels had SD and four patients had PD. The results of the c2 test suggested that HBV DNA elevation was significantly associated with poor tumor response 38 (79.2%) 1 (2%) 4 (8.3%) 4 (8.3%) 2 (4.2%) 13 (27.1%) 17 (35.4%) 16 (33.3%) 15 (31.3%) 32 (66.7%) ADV, Adefovir dipivoxil; AFP, Alpha‐Fetoprotein; BCLC, Barcelona Clinic Liver Cancer; CR, complete response; DC, disease control; ETV, Entecavir; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B viral; OR, objective response; PD progressive disease; PR, partial response; SD, stable disease; TACE, transcatheter arterial chemoembolization; TDF, Tenofovir. Frontiers in Immunology | www.frontiersin.org 5 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 TABLE 2 | Occurrence of adverse events. Fever Fatigue Hypertension Diarrhea Rash Hypothyroidism Pruritus Proteinuria Renal dysfunction Pneumonia Lymphopenia Thrombocytopenia Hyperthyroidism Cardiotoxicity hoarseness hepatitis Bacterial infection fungal infection Herpes virus infection Intestinal infections G1/G2 G3/4 9 (18.8%) 7 (14.6%) 6 (12.5%) 6 (12.5%) 6 (12.5%) 4 (8.3%) 2 (4.2%) 3 (6.3%) 2 (4.2%) 1 (2%) 1 (2%) 1 (2.1%) 1 (2.1%) 1 (2.1%) 1 (2.1%) 0 0 0 0 0 0 0 0 2 (4.2%) 0 0 0 0 0 1 (2%) 2 (4.2%) 1 (2.1%) 0 0 0 2 (4.2%) 2 (4.2%) 1 (2.1%) 1 (2.1%) 1 (2.1%) (p=0.001, OR=18.643 [95% CI: 3.271–106.253]), but elevated HBsAg was not associated with tumor response (p=0.270, OR=2.440 [95% CI: 0.499–11.965]) (Table 3). The results of Cox univariate and multivariate survival analyses suggested that increase in HBV DNA (P=0.011, HR=4.816, 95 CI%: 1.439–16.117) and HBsAg (P=0.022, HR=4.161, 95 CI%: 1.224– 16.144) levels were independent risk factors associated with patient survival time (Table 4). Patients who had increased HBV DNA (6.87 months vs undefined, log-rank test: p= 0.004) and HBsAg (8.07 months vs undefined, log-rank test: p= 0.004) levels had a shorter median survival time (MST) than patients without an increase in HBV DNA and HBsAg levels (Figure 3). change trends in some clinical parameters. The results suggested that there were higher baseline levels of albumin (ALB), AST/ALT ratio, cholinesterase (CHE), prothrombin activity (PTA), and platelets (PLT), and lower baseline levels of direct bilirubin (DBIL) and absolute value of monocytes (AMC) in patients without increased HBsAg than in patients with increased HBsAg. Compared to patients with increased HBsAg, other patients showed an increase in CRP, IL-6, AST/ ALT ratio, and DBIL and a decrease in T lymphocytes, CD4+ T lymphocytes, B lymphocytes, and CHE one week after the initiation of treatment (Figure 4). Correlation Between Changes in Clinical Parameters and Elevated HBsAg DISCUSSION To identify the biomarkers associated with elevated HBsAg levels, we analyzed the baseline characteristics and early In our study, we investigated the safety and efficacy of a combined treatment with PD-1 ICIs and TKIs in liver cancer patients with FIGURE 1 | The overall changes in HBV DNA and HBsAg levels at different time point. ns: >0.05; ***: ≤0.005; **: ≤0.01; *: ≤0.05. Frontiers in Immunology | www.frontiersin.org 6 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 A D B E C F FIGURE 2 | The change in HBV DNA and HBsAg levels in patients with different tumor responses. (A) Comparison of baseline levels of HBV DNA and HBsAg between patients with DC and PD. (B) The changes in HBV DNA and HBsAg levels at different time points in patients with DC. (C) The changes in HBV DNA and HBsAg levels at different time points in patients with PD. (D) Comparison of baseline levels of HBV DNA and HBsAg among patients with OR, SD and PD. (E) The changes in HBV DNA and HBsAg levels at different time points in patients with OR. (F) The changes in HBV DNA and HBsAg levels at different time points in patients with SD. DC, disease control; OR, objective response; PD, progressive disease; SD, stable disease. ns: >0.05; ***: ≤0.005; **: ≤0.01; *: ≤0.05. HBV DNA and HBsAg levels. Eleven patients showed an increase in HBV DNA level within 24 weeks; one of these patients met the criteria for HBV reactivation, but did not develop HBV-related hepatitis. The baseline level of HBV DNA between patients with and without elevated HBV DNA was not statistically different, and univariate regression analysis suggested that the baseline level of HBV DNA was not a risk factor for HBV DNA elevation. The c2 test showed an association between elevated HBV DNA level and tumor response, and patients with elevated HBV DNA level were more likely to experience tumor progression. Survival analysis HBV infection, the effect of combination therapy on HBV DNA and HBsAg levels, and biomarkers associated with HBsAg elevation. Combination therapy showed good safety and efficacy; the ORR and DCR was 31.5% and 66.7%, respectively. There were 22 (45.8%) patients with irAEs, of which 10 patients suffered from grade 3/4 irAEs, but their symptoms improved upon the administration of glucocorticoids, and no deaths were associated with irAEs. Patients suggested a decrease in overall HBV DNA and HBsAg levels. Compared with PD patients, patients with DCR (including CR, PR and SD) showed a more significant decline in TABLE 3 | c2 test for change trends of HBV DNA and HBsAg levels according to tumor response. The change of HBV DNA Without increase With increase The change of HBsAg Without increase With increase CR+PR+SD PD 29 2 7 9 22 4 9 4 p value OR <0.001 18.643 (3.271-106.253) 0.270 2.44 (0.499-11.965) AFP, Alpha‐Fetoprotein; BCLC, Barcelona Clinic Liver Cancer; HBsAg, hepatitis B surface antigen; HBV, hepatitis B viral; HBsAg, hepatitis B surface antigen; MST, median survival time; TACE, transcatheter arterial chemoembolization. Frontiers in Immunology | www.frontiersin.org 7 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 TABLE 4 | Cox univariate and multifactorial survival analysis to identify any independent predictive factor associated with OS. Cox univariate survival analysis Age Sex Types of tumors BCLC stage Child-Pugh stage Options of combination therapy TACE treatment Line of treatment AFP The occurrence of irAEs Prior antiviral therapy Baseline level of HBV DNA HBV DNA increase HBsAg increase Cox multivariate survival analysis HR 95%CI p value HR 95%CI p value 0.979 0.785 1.062 1.165 0.140 1.154 0.329 0.9103 2.312 0.491 1.176 1.012 4.034 4.869 0.925-1.036 0.420-1.465 0.141-8.030 0.644-2.105 0.160-1.294 0.426-3.125 0.075-1.439 0.579-1.440 0.855-6.257 0.185-1.298 0.406-3.411 0.753-1.358 1.456-11.179 1.476-16.064 0.461 0.446 0.953 0.614 0.140 0.778 0.140 0.696 0.099 0.152 0.765 0.939 0.007 0.009 4.816 4.161 1.439-16.117 1.224-16.144 0.011 0.022 AFP, Alpha‐Fetoprotein; BCLC, Barcelona Clinic Liver Cancer; HBsAg, hepatitis B surface antigen; HBV, hepatitis B viral; TACE, transcatheter arterial chemoembolization. revealed that patients who developed elevated HBV DNA levels had a shorter survival than those who did not, with a median OS of only 6.87 months. To the end, eight patients developed elevated HBsAg levels within 24 weeks, and elevated HBsAg level showed a shorter median OS (8.07 months). Furthermore, patients with increased HBsAg level showed lower levels of ALB, CHE, PTA, monocytes, and PLT and higher levels of DBIL and AST/ALT ratio at baseline. There was a decrease in CHE, PTA, LYM, T lymphocytes, CD4+ T cells, and B cells, and an increase in DBIL, AST/ALT ratio, CRP, and IL-6 in patients without increased HBsAg one week after treatment initiation, but these results were not observed in patients with increased HBsAg. HBV DNA is often associated with poor prognosis in the local or systemic treatment of liver cancer (21–28). However, during PD-1/PD-L1 ICI treatment, patients with HBV-related liver cancer had a similar tumor response to that of all patients. For example, in the CheckMate 040 study, the ORR of nivolumab monotherapy was 19% and 14% in all patients and the HBV subgroup, respectively, and the ORR of nivolumab and ipilimumab was 31% in both the groups (12, 15). In the KEYNOTE-224 study, the DCR was 60% and 52% in all patients and HBV infection group, respectively (13). The ORRs were 36% and 33% in patients treated with atezolizumab and bevacizumab, respectively. However, these patients started immunotherapy with a low viral load (less than 500 IU/mL), which may have contributed to the impact of HBV DNA being overlooked. In our study, 16 patients had HBV DNA levels > 1000 IU/mL, and 11 patients had HBV DNA levels less than 1000 IU/ mL but more than 100 IU/mL. The overall ORR and overall DCR were 31.3% and 66.7%, respectively, most of the irAEs were mild, and no patient died from adverse events. In patients with HBV DNA levels > 100 IU/mL, one patient had CR, seven patients had PR, and ten patients had SD; the ORR and DCR were 29.6% and 66.7%, respectively. From the results of the mate analysis, the overall ORR and DCR were 20% and 60%, respectively, for PD-1/ PD-L1 ICI monotherapy in liver cancer patients, while the ORR was 29% and DCR was 77% for combination therapy with antiVEGF drugs; the overall rate of irAE was 63%, and the occurrence of severe adverse events was 11% (47). Our results suggest that regardless of the HBV DNA levels, there were similar tumor responses and irAEs between patients with HBV-associated liver cancer and other liver cancer patients. Therefore, HBV infection FIGURE 3 | Correlation analysis between tumor response, survival time and the elevation of HBV DNA and HBsAg levels. Kaplan–Meier overall survival estimate curves in patients with increasing HBV DNA and HBsAg and patients exhibiting no increase in HBV DNA and HBsAg. HBsAg, hepatitis B surface antigen; HBV, hepatitis B viral; MST, median survival time. Frontiers in Immunology | www.frontiersin.org 8 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 FIGURE 4 | Baseline characteristics and trends of some clinical parameters in patients with or without increasing HBsAg level. ALB, albumin; AMC, absolute monocyte count; CHE, cholinesterase; CRP, C-reactive protein; DBIL, direct bilirubin; IL-6, interleukin-6; LYM, absolute lymphocytes; PLT, platelets; PTA, prothrombin activity. ns: >0.05; ***: ≤0.005; **: ≤0.01; *: ≤0.05. agent or combined NA therapy may not be satisfactory, and combination therapy with other treatment options should be considered to improve HBsAg clearance. HBV reactivation is a key issue that requires close attention during the treatment of patients with HBV-infected liver cancer. HBV reactivation has been observed in both local and systemic treatments, irrespective of HBV DNA levels, and the use of prophylactic antiviral drugs is an independent risk factor for HBV reactivation (29–36). Although HBV reactivation can also occur during PD-1/PD-L1 ICI treatment, it is less frequent than during surgery, radiation therapy, and other treatments. Treatment such as surgery and radiation therapy can lead to immunosuppression accompanied by dysfunctional HBVspecific T and B cells, resulting in HBV DNA replication and HBV reactivation (53). However, PD-1/PD-L1 antibodies promote the reactivation of the immune system by blocking inhibitory signaling pathways to inhibit viral replication (41). At present, the precise mechanism of HBV re-replication or reactivation in patients treated with PD-1/PD-L1 antibodies is unknown. In a recent study, researchers reported that virological breakthrough(VB) was an indicator of poor outcome in patients with HBsAg positive HCC under ICIs therapy (54). This finding is similar with our study. In our research, all patients underwent PD-1 combined with targeted drug therapies. We found that some patients who had elevated HBV DNA did not meet the requirements for VB or HBV reactivation. We included this group of patients in the study and found similar results, expanding the applicability of our conclusions. Besides HBVDNA, most patients underwent a regular post-treatment HBsAg assay, which enabled us to assess the impact of ICIs on HBsAg. Liver function and immunological characteristics were evaluated based on the change trend of HBsAg in our study. We discussed the relationship between the changes of HBsAg and inflammatory factors, lymphocyte subsets and other may not be a promoter during PD-1/PD-L1 therapy in patients with liver cancer. However, the sample size considered in this study was small; future studies require larger sample sizes to confirm this observation. CD8+ T cells play an important role in virus clearance and are critical for controlling the progression of CHB (48–50). CD8+ T lymphocytes play an important role in tumor immunity and are also crucial in virus clearance and regulation of CHB infection (48–50). Exhaustion of CD8+ T cells could be found in CHB patients, characterized by upregulated expression of PD-1 accompanied by a reduction in secretion and killing functions. Therefore, blocking the PD-1/PD-L1 signaling pathway can restore the function of CD8+ T cells and promote virus clearance. In our study, we investigated the effect of PD-1 antibodies on HBV, and the results suggested that patients showed a decrease in overall HBV DNA and HBsAg levels when using a prophylactic antiviral. Furthermore, patients with DCR showed better efficacy of antiviral therapy and more pronounced decrease in HBV DNA and HBsAg levels compared to those with PD. We hypothesized that this may be related to the reactivation of exhausted CD8+ T cells (51). However, a recent study indicated that tumor-infiltrating HBV-specific CD8 T cells have been proven to be associated with prolonged patient relapse-free survival. More HBV-specific CD8 T cells could be detected in both liver cancer tissue and healthy tissue compared to their levels in the tissues of relapsed patients (52). This indicates that restoring the function of HBVspecific CD8 T cells may be important for improving both the tumor response and virus clearance in patients with HBV-related liver cancer. However, in our study, patients treated with both PD-1 antibodies and NAs did not show HBsAg clearance. Although the non-clearance could be related to the immunosuppressed state, it suggests that when choosing PD-1/ PD-L1 antibodies for antiviral therapy, the efficacy of single- Frontiers in Immunology | www.frontiersin.org 9 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 However, this study had several limitations. First, this was a single-center retrospective study with a relatively small sample size. Second, because fewer patients presented with elevated HBsAg, the baseline and changing characteristics of liver function, blood tests, CRP, IL-6, and lymphocyte subsets were not reflected in this group of patients. Large prospective clinical trials might elaborate the association between CRP, IL-6, lymphocyte subsets, liver function, blood routine, and HBsAg increase, as well as lead to the discovery of other markers. immunological indicators. In our study, 11 patients showed HBV DNA elevation, resulting in HBV reactivation in one of them, but it did not lead to HBV reactivation-related hepatitis. In all patients with HBV DNA elevation, one patient had baseline HBV DNA > 10000 IU/mL, one patient had baseline HBV DNA >1000 IU/mL, two patients had baseline HBV DNA > 100 IU/mL, and other patients had HBV DNA < 100 IU/mL. Univariate regression analysis suggested that the baseline HBV DNA level was not a risk factor for HBV DNA elevation. This result suggests that there is no correlation between baseline HBV DNA levels and HBV DNA increase. Keeping viral load at a lower level when starting PD-1 therapy is therefore not necessary if antiviral therapy is also available, and a high viral load should not be a contraindication to PD-1/PD-L1 ICI therapy in patients with HBV-related liver cancer. In addition, in our study, we found that elevated HBV DNA levels were associated with tumor response and patient survival. These results suggest that elevated or reactivated HBV DNA may occur due to poor tumor control, or that elevated HBV DNA may imply poor prognosis; however, due to the small sample size, the relationship between the two remains unclear. The existence of this correlation may explain why HBV reactivation occurred during treatment with PD-1/PD-L1 ICIs, a class of immune activators. It is possible that patients with PD experience immunosuppression and lack of activated immune cells, such as depleted CD8 T cells, which is consistent with the immunosuppressed state caused by other treatments (53). However, this hypothesis and its exact mechanism require further study. Along with an increase in HBV DNA level, we also observed elevated HBsAg levels during treatment. Up to 24 weeks, eight patients did not show a decrease in HBsAg, and patients who developed HBsAg elevation had shorter OS. Although our results did not suggest a correlation between HBsAg decline and tumor response, this may be due to missing HBsAg data after treatment in some patients, or there may not be an intrinsic correlation. The exact reasons need to be explored considering a larger sample size. Furthermore, we investigated the correlation between changes in clinical parameters and HBsAg levels. These results suggested that better baseline liver function was associated with changes in HBsAg levels. As an immune organ, the liver can mediate adaptive immune tolerance, and better liver function implies better immune regulation, which may play a role in HBsAg clearance (55). Furthermore, an early increase in CRP and IL-6 levels and a decrease in lymphocyte subsets are associated with a decline in HBsAg levels. Although the exact mechanisms responsible for the changes in immune markers are not known, there is an association between changes in CRP and IL-6 levels and changes in lymphocyte subsets. This may be related to the systemic inflammatory response caused by a cytokine storm, which has also been seen in COVID-19 patients. Compared with mild-type COVID-19 patients, severetype patients who experienced cytokine storms had higher levels of CRP and IL-6, which are characterized by high levels of granulocyte colony stimulating factor, interferon-inducible protein-10, monocyte chemotactic Protein-1, macrophage inflammatory protein-1a/b, IL-8, and other cytokines, which can promote chemotaxis or apoptosis of peripheral blood lymphocyte subsets, leading to a decrease in cell numbers (56–59). Frontiers in Immunology | www.frontiersin.org DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors. ETHICS STATEMENT The studies involving human participants were reviewed and approved by Chinese Ethics Committee of Registering Clinical Trials. The patients/participants provided their written informed consent to participate in this study. AUTHOR CONTRIBUTIONS SP, YY, F-SW, FM, and MS conceived the study. SP and YY wrote the manuscript. SP, YY, and SW collected data and performed the data analysis. SW, BT, YS, XL, NS, and YZ participated in the clinical treatment. QQ and JL supervised the clinical treatment. F-SW, FM, MS and J-YZ directed the writing and revision of the manuscript. All authors contributed to the article and approved the submitted version. FUNDING This work was supported by grants from the National Natural Science Foundation of China (82070617); the Innovative Research Group Project of the National Natural Science Foundation of China(81721002); Beijing Municipal Science and Technology Commission (Z201100005520047) and National Key Research and Development Program of China (2019YFC0840704). SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu.2022. 892618/full#supplementary-material 10 May 2022 | Volume 13 | Article 892618 Pan et al. Effects on HBV by PD-1 REFERENCES 17. Xu J, Zhang Y, Jia R, Yue C, Chang L, Liu R, et al. Anti-Pd-1 Antibody Shr1210 Combined With Apatinib for Advanced Hepatocellular Carcinoma, Gastric, or Esophagogastric Junction Cancer: An Open-Label, Dose Escalation and Expansion Study. Clin Cancer Res (2019) 25(2):515–23. doi: 10.1158/ 1078-0432.CCR-18-2484 18. Lee MS, Ryoo B-Y, Hsu C-H, Numata K, Stein S, Verret W, et al. Atezolizumab With or Without Bevacizumab in Unresectable Hepatocellular Carcinoma (Go30140): An Open-Label, Multicentre, Phase 1b Study. Lancet Oncol (2020) 21(6):808–20. doi: 10.1016/S1470-2045(20) 30156-X 19. Kudo M, Motomura K, Wada Y, Inaba Y, Sakamoto Y, Kurosaki M, et al. First-Line Avelumab + Axitinib in Patients With Advanced Hepatocellular Carcinoma: Results From a Phase 1b Trial (Vegf Liver 100). J Clin Oncol (2019) 37:4072. doi: 10.1200/JCO.2019.37.15_suppl.4072 20. Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim T-Y, et al. Atezolizumab Plus Bevacizumab in Unresectable Hepatocellular Carcinoma. N Engl J Med (2020) 382(20):1894–905. doi: 10.1056/NEJMoa1915745 21. Li M-r, Chen G-h, Cai C-j, Wang G-y, Zhao H. High Hepatitis B Virus DNA Level in Serum Before Liver Transplantation Increases the Risk of Hepatocellular Carcinoma Recurrence. Digestion (2011) 84(2):134–41. doi: 10.1159/000324197 22. Sun F, Liu Z, Wang B. Correlation Between Low-Level Viremia and Hepatitis B-Related Hepatocellular Carcinoma and Recurrence: A Retrospective Study. BMC Cancer (2021) 21(1):1103. doi: 10.1186/s12885-021-08483-3 23. Yu SJ, Lee J-H, Jang ES, Cho EJ, Kwak M-S, Yoon J-H, et al. Hepatocellular Carcinoma: High Hepatitis B Viral Load and Mortality in Patients Treated With Transarterial Chemoembolization. Radiology (2013) 267(2):638–47. doi: 10.1148/radiol.13121498 24. Shao Y-Y, Chen P-J, Lin Z-Z, Huang C-C, Ding Y-H, Lee Y-H, et al. Impact of Baseline Hepatitis B Viral DNA Levels on Survival of Patients With Advanced Hepatocellular Carcinoma. Anticancer Res (2011) 31(11):4007. 25. Yang Y, Wen F, Li J, Zhang P, Yan W, Hao P, et al. A High Baseline Hbv Load and Antiviral Therapy Affect the Survival of Patients With Advanced HbvRelated Hcc Treated With Sorafenib. Liver Int (2015) 35(9):2147–54. doi: 10.1111/liv.12805 26. Yeo W, Mo FKF, Chan SL, Leung NWY, Hui P, Lam W-Y, et al. Hepatitis B Viral Load Predicts Survival of Hcc Patients Undergoing Systemic Chemotherapy. Hepatol (Baltimore Md) (2007) 45(6):1382–9. doi: 10.1002/ hep.21572 27. Sun X, Hu D, Yang Z, Liu Z, Wang J, Chen J, et al. Baseline Hbv Loads Do Not Affect the Prognosis of Patients With Hepatocellular Carcinoma Receiving Anti-Programmed Cell Death-1 Immunotherapy. J Hepatocell Carcinoma (2020) 7:337–45. doi: 10.2147/JHC.S278527 28. Yuan G, Li R, Li Q, Hu X, Ruan J, Fan W, et al. Interaction Between Hepatitis B Virus Infection and the Efficacy of Camrelizumab in Combination With Apatinib Therapy in Patients With Hepatocellular Carcinoma: A Multicenter Retrospective Cohort Study. Ann Transl Med (2021) 9(18):1412. doi: 10.21037/atm-21-3020 29. Huang W, Zhang W, Fan M, Lu Y, Zhang J, Li H, et al. Risk Factors for Hepatitis B Virus Reactivation After Conformal Radiotherapy in Patients With Hepatocellular Carcinoma. Cancer Sci (2014) 105(6):697–703. doi: 10.1111/cas.12400 30. Liu S, Lai J, Lyu N, Xie Q, Cao H, Chen D, et al. Effects of Antiviral Therapy on Hbv Reactivation and Survival in Hepatocellular Carcinoma Patients Undergoing Hepatic Artery Infusion Chemotherapy. Front Oncol (2020) 10:582504. doi: 10.3389/fonc.2020.582504 31. Huang G, Lai ECH, Lau WY, Zhou W-p, Shen F, Pan Z-y, et al. Posthepatectomy Hbv Reactivation in Hepatitis B-Related Hepatocellular Carcinoma Influences Postoperative Survival in Patients With Preoperative Low Hbv-DNA Levels. Ann Surg (2013) 257(3):490–505. doi: 10.1097/ SLA.0b013e318262b218 32. Dan JQ, Zhang YJ, Huang JT, Chen MS, Gao HJ, Peng ZW, et al. Hepatitis B Virus Reactivation After Radiofrequency Ablation or Hepatic Resection for Hbv-Related Small Hepatocellular Carcinoma: A Retrospective Study. Eur J Surg Oncol J Eur Soc Surg Oncol Br Assoc Surg Oncol (2013) 39(8):865–72. doi: 10.1016/j.ejso.2013.03.020 33. Zhang S-S, Liu J-X, Zhu J, Xiao M-B, Lu C-H, Ni R-Z, et al. Effects of Tace and Preventive Antiviral Therapy on Hbv Reactivation and Subsequent Hepatitis 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global Cancer Statistics 2018: Globocan Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin (2018) 68 (6):394–424. doi: 10.3322/caac.21492 2. Liu Z, Mao X, Jiang Y, Cai N, Jin L, Zhang T, et al. Changing Trends in the Disease Burden of Primary Liver Cancer Caused by Specific Etiologies in China. Cancer Med (2019) 8(12):5787–99. doi: 10.1002/cam4.2477 3. Feng R-M, Zong Y-N, Cao S-M, Xu R-H. Current Cancer Situation in China: Good or Bad News From The 2018 Global Cancer Statistics? Cancer Commun (Lond) (2019) 39(1):22. doi: 10.1186/s40880-019-0368-6 4. Paik JM, Golabi P, Younossi Y, Mishra A, Younossi ZM. Changes in the Global Burden of Chronic Liver Diseases From 2012 to 2017: The Growing Impact of Nafld. Hepatol (Baltimore Md) (2020) 72(5):1605–16. doi: 10.1002/ hep.31173 5. Wang M, Wang Y, Feng X, Wang R, Wang Y, Zeng H, et al. Contribution of Hepatitis B Virus and Hepatitis C Virus to Liver Cancer in China North Areas: Experience of the Chinese National Cancer Center. Int J Infect Dis IJID Off Publ Int Soc Infect Dis (2017) 65:15–21. doi: 10.1016/j.ijid.2017.09.003 6. Zeng H, Chen W, Zheng R, Zhang S, Ji JS, Zou X, et al. Changing Cancer Survival in China During 2003-15: A Pooled Analysis of 17 Population-Based Cancer Registries. Lancet Glob Health (2018) 6(5):e555–e67. doi: 10.1016/ S2214-109X(18)30127-X 7. Villanueva A. Hepatocellular Carcinoma. N Engl J Med (2019) 380(15):1450– 62. doi: 10.1056/NEJMra1713263 8. Cheng A-L, Kang Y-K, Chen Z, Tsao C-J, Qin S, Kim JS, et al. Efficacy and Safety of Sorafenib in Patients in the Asia-Pacific Region With Advanced Hepatocellular Carcinoma: A Phase Iii Randomised, Double-Blind, PlaceboControlled Trial. Lancet Oncol (2009) 10(1):25–34. doi: 10.1016/S1470-2045 (08)70285-7 9. Ikeda M, Okusaka T, Mitsunaga S, Ueno H, Tamai T, Suzuki T, et al. Safety and Pharmacokinetics of Lenvatinib in Patients With Advanced Hepatocellular Carcinoma. Clin Cancer Res (2016) 22(6):1385–94. doi: 10.1158/1078-0432.CCR-15-1354 10. Chen LT, Martinelli E, Cheng AL, Pentheroudakis G, Qin S, Bhattacharyya GS, et al. Pan-Asian Adapted Esmo Clinical Practice Guidelines for the Management of Patients With Intermediate and Advanced/Relapsed Hepatocellular Carcinoma: A Tos-Esmo Initiative Endorsed by Csco, Ismpo, Jsmo, Ksmo, Mos and Sso. Ann Oncol Off J Eur Soc Med Oncol (2020) 31(3):334–51. doi: 10.1016/j.annonc.2019.12.001 11. Kudo M, Finn RS, Qin S, Han K-H, Ikeda K, Piscaglia F, et al. Lenvatinib Versus Sorafenib in First-Line Treatment of Patients With Unresectable Hepatocellular Carcinoma: A Randomised Phase 3 Non-Inferiority Trial. Lancet (2018) 391(10126):1163–73. doi: 10.1016/S0140-6736(18)30207-1 12. El-Khoueiry AB, Sangro B, Yau T, Crocenzi TS, Kudo M, Hsu C, et al. Nivolumab in Patients With Advanced Hepatocellular Carcinoma (Checkmate 040): An Open-Label, Non-Comparative, Phase 1/2 Dose Escalation and Expansion Trial. Lancet (2017) 389(10088):2492–502. doi: 10.1016/S0140-6736(17)31046-2 13. Zhu AX, Finn RS, Edeline J, Cattan S, Ogasawara S, Palmer D, et al. Pembrolizumab in Patients With Advanced Hepatocellular Carcinoma Previously Treated With Sorafenib (Keynote-224): A Non-Randomised, Open-Label Phase 2 Trial. Lancet Oncol (2018) 19(7):940–52. doi: 10.1016/ S1470-2045(18)30351-6 14. Qin S, Ren Z, Meng Z, Chen Z, Chai X, Xiong J, et al. Camrelizumab in Patients With Previously Treated Advanced Hepatocellular Carcinoma: A Multicentre, Open-Label, Parallel-Group, Randomised, Phase 2 Trial. Lancet Oncol (2020) 21(4):571–80. doi: 10.1016/S1470-2045(20)30011-5 15. Yau T, Kang Y-K, Kim T-Y, El-Khoueiry AB, Santoro A, Sangro B, et al. Efficacy and Safety of Nivolumab Plus Ipilimumab in Patients With Advanced Hepatocellular Carcinoma Previously Treated With Sorafenib: The Checkmate 040 Randomized Clinical Trial. JAMA Oncol (2020) 6(11): e204564. doi: 10.1001/jamaoncol.2020.4564 16. Finn RS, Ikeda M, Zhu AX, Sung MW, Baron AD, Kudo M, et al. Phase Ib Study of Lenvatinib Plus Pembrolizumab in Patients With Unresectable Hepatocellular Carcinoma. J Clin Oncol (2020) 38(26):2960–70. doi: 10.1200/JCO.20.00808 Frontiers in Immunology | www.frontiersin.org 11 May 2022 | Volume 13 | Article 892618 Pan et al. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. Effects on HBV by PD-1 in Hepatocellular Carcinoma: A Meta-Analysis. Jpn J Clin Oncol (2019) 49 (7):646–55. doi: 10.1093/jjco/hyz046 Yau T, Hsu C, Kim T-Y, Choo S-P, Kang Y-K, Hou M-M, et al. Nivolumab in Advanced Hepatocellular Carcinoma: Sorafenib-Experienced Asian Cohort Analysis. J Hepatol (2019) 71(3):543–52. doi: 10.1016/j.jhep.2019.05.014 Zhang X, Zhou Y, Chen C, Fang W, Cai X, Zhang X, et al. Hepatitis B Virus Reactivation in Cancer Patients With Positive Hepatitis B Surface Antigen Undergoing Pd-1 Inhibition. J Immunother Cancer (2019) 7(1):322. doi: 10.1186/s40425-019-0808-5 Lee PC, Chao Y, Chen MH, Lan KH, Lee IC, Hou MC, et al. Risk of Hbv Reactivation in Patients With Immune Checkpoint Inhibitor-Treated Unresectable Hepatocellular Carcinoma. J Immunother Cancer (2020) 8(2): e001072. doi: 10.1136/jitc-2020-001072 Cui T-M, Liu Y, Wang J-B, Liu L-X. Adverse Effects of Immune-Checkpoint Inhibitors in Hepatocellular Carcinoma. Onco Targets Ther (2020) 13:11725– 40. doi: 10.2147/OTT.S279858 Chen J, Hu X, Li Q, Dai W, Cheng X, Huang W, et al. Effectiveness and Safety of Toripalimab, Camrelizumab, and Sintilimab in a Real-World Cohort of Hepatitis B Virus Associated Hepatocellular Carcinoma Patients. Ann Transl Med (2020) 8(18):1187. doi: 10.21037/atm-20-6063 Maier H, Isogawa M, Freeman GJ, Chisari FV. Pd-1:Pd-L1 Interactions Contribute to the Functional Suppression of Virus-Specific Cd8+ T Lymphocytes in the Liver. J Immunol (2007) 178(5):2714–20. doi: 10.4049/ jimmunol.178.5.2714 Tzeng HT, Tsai HF, Liao HJ, Lin YJ, Chen L, Chen PJ, et al. Pd-1 Blockage Reverses Immune Dysfunction and Hepatitis B Viral Persistence in a Mouse Animal Model. PLoS One (2012) 7(6):e39179. doi: 10.1371/journal. pone.0039179 Fisicaro P, Valdatta C, Massari M, Loggi E, Biasini E, Sacchelli L, et al. Antiviral Intrahepatic T-Cell Responses Can Be Restored by Blocking Programmed Death-1 Pathway in Chronic Hepatitis B. Gastroenterology (2010) 138(2):682–93, 93.e1-4. doi: 10.1053/j.gastro.2009.09.052 Gane E, Verdon DJ, Brooks AE, Gaggar A, Nguyen AH, Subramanian GM, et al. Anti-Pd-1 Blockade With Nivolumab With and Without Therapeutic Vaccination for Virally Suppressed Chronic Hepatitis B: A Pilot Study. J Hepatol (2019) 71(5):900–7. doi: 10.1016/j.jhep.2019.06.028 Lencioni R, Llovet JM. Modified Recist (Mrecist) Assessment for Hepatocellular Carcinoma. Semin Liver Dis (2010) 30(1):52–60. doi: 10.1055/s-0030-1247132 Llovet JM, Lencioni R. Mrecist for Hcc: Performance and Novel Refinements. J Hepatol (2020) 72(2):288–306. doi: 10.1016/j.jhep.2019.09.026 Freites-Martinez A, Santana N, Arias-Santiago S, Viera A. Using the Common Terminology Criteria for Adverse Events (Ctcae - Version 5.0) to Evaluate the Severity of Adverse Events of Anticancer Therapies. Actas Dermosifiliogr (Engl Ed) (2021) 112(1):90–2. doi: 10.1016/j.ad.2019.05.009 Terrault NA, Lok ASF, McMahon BJ, Chang K-M, Hwang JP, Jonas MM, et al. Update on Prevention, Diagnosis, and Treatment of Chronic Hepatitis B: Aasld 2018 Hepatitis B Guidance. Hepatol (Baltimore Md) (2018) 67(4):1560– 99. doi: 10.1002/hep.29800 Rao Q, Li M, Xu W, Pang K, Guo X, Wang D, et al. Clinical Benefits of Pd-1/ Pd-L1 Inhibitors in Advanced Hepatocellular Carcinoma: A Systematic Review and Meta-Analysis. Hepatol Int (2020) 14(5):765–75. doi: 10.1007/ s12072-020-10064-8 Tang TJ, Kwekkeboom J, Mancham S, Binda RS, de Man RA, Schalm SW, et al. Intrahepatic Cd8+ T-Lymphocyte Response Is Important for TherapyInduced Viral Clearance in Chronic Hepatitis B Infection. J Hepatol (2005) 43 (1):45–52. doi: 10.1016/j.jhep.2005.01.038 Fioravanti J, Di Lucia P, Magini D, Moalli F, Boni C, Benechet AP, et al. Effector Cd8(+) T Cell-Derived Interleukin-10 Enhances Acute Liver Frontiers in Immunology | www.frontiersin.org 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. Immunopathology. J Hepatol (2017) 67(3):543–8. doi: 10.1016/ j.jhep.2017.04.020 Benechet AP, Iannacone M. Determinants of Hepatic Effector Cd8(+) T Cell Dynamics. J Hepatol (2017) 66(1):228–33. doi: 10.1016/j.jhep.2016.07.011 Wu S-P, Liao R-Q, Tu H-Y, Wang W-J, Dong Z-Y, Huang S-M, et al. Stromal Pd-L1-Positive Regulatory T Cells and Pd-1-Positive Cd8-Positive T Cells Define the Response of Different Subsets of Non-Small Cell Lung Cancer to Pd-1/Pd-L1 Blockade Immunotherapy. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer (2018) 13(4):521–32. doi: 10.1016/j.jtho.2017.11.132 Cheng Y, Gunasegaran B, Singh HD, Dutertre CA, Loh CY, Lim JQ, et al. NonTerminally Exhausted Tumor-Resident Memory Hbv-Specific T Cell Responses Correlate With Relapse-Free Survival in Hepatocellular Carcinoma. Immunity (2021) 54(8):1825–40.e7. doi: 10.1016/j.immuni.2021.06.013 Loomba R, Liang TJ. Hepatitis B Reactivation Associated With Immune Suppressive and Biological Modifier Therapies: Current Concepts, Management Strategies, and Future Directions. Gastroenterology (2017) 152 (6):1297–309. doi: 10.1053/j.gastro.2017.02.009 Wang K, Xia Y, Zhu Y, Yu W, Guo Y, Liu L. Virological Breakthrough After Immune Checkpoint Inhibitor and Nucleos(T)Ide Analog Treatment in Patients With Hepatitis B Surface Antigen Positive Hepatocellular Carcinoma: A Real-World Study. J ImmunoTherapy Cancer (2021) 9(11): e003195. doi: 10.1136/jitc-2021-003195 Zheng M, Tian Z. Liver-Mediated Adaptive Immune Tolerance. Front Immunol (2019) 10:2525. doi: 10.3389/fimmu.2019.02525 Yang P-H, Ding Y-B, Xu Z, Pu R, Li P, Yan J, et al. Increased Circulating Level of Interleukin-6 and Cd8 T Cell Exhaustion Are Associated With Progression of Covid-19. Infect Dis Poverty (2020) 9(1):161. doi: 10.1186/s40249-020-00780-6 Jiang M, Guo Y, Luo Q, Huang Z, Zhao R, Liu S, et al. T-Cell Subset Counts in Peripheral Blood Can Be Used as Discriminatory Biomarkers for Diagnosis and Severity Prediction of Coronavirus Disease 2019. J Infect Dis (2020) 222 (2):198–202. doi: 10.1093/infdis/jiaa252 Zhang W, Li L, Liu J, Chen L, Zhou F, Jin T, et al. The Characteristics and Predictive Role of Lymphocyte Subsets in Covid-19 Patients. Int J Infect Dis IJID Off Publ Int Soc Infect Dis (2020) 99:92–9. doi: 10.1016/j.ijid. 2020.06.079 Tang Y, Liu J, Zhang D, Xu Z, Ji J, Wen C. Cytokine Storm in Covid-19: The Current Evidence and Treatment Strategies. Front Immunol (2020) 11:1708. doi: 10.3389/fimmu.2020.01708 Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer XZ declared a shared parent affiliation with the author NS to the handling editor at the time of review. Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Copyright © 2022 Pan, Yu, Wang, Tu, Shen, Qiu, Liu, Su, Zuo, Luan, Zhang, Shi, Meng and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 12 May 2022 | Volume 13 | Article 892618
https://openalex.org/W2790062957
https://wnus.edu.pl/sip/file/article/view/11419.pdf
Polish
null
Resource management expenditure accounts
Zeszyty Naukowe. Studia i Prace Wydziału Nauk Ekonomicznych i Zarządzania
2,017
cc-by-sa
3,356
DOI: 10.18276/sip.2017.49/1-01 Studia i Prace WNEIZ US nr 49/1 2017 PROBLEMY TEORETYCZNE I METODYCZNE Elżbieta Broniewicz* Politechnika Białostocka *  Adres e-mail: e.broniewicz@pb.edu.pl. Streszczenie W artykule zaprezentowano szereg inicjatyw racjonalnego gospodarowania zasobami, które pojawiły się w ostatnim dziesięcioleciu w Europie i na świecie. Podkreślono trudno- ści w monitorowaniu istniejącego stanu i zmian zachodzących w obszarze gospodarowania zasobami naturalnymi. Rozwiązaniem tego problemu mogłaby być kompilacja europejskich rachunków wydatków na gospodarowanie zasobami (ReMEA), proponowanych do wdro- żenia w krajach Unii Europejskiej po 2020 roku. Celem artykułu jest przedstawienie tych rachunków jako jedynych spośród rachunków dotyczących zasobów naturalnych, prowa- dzonych w jednostkach pieniężnych. Artykuł stanowi przegląd aktów prawnych, koncepcji i planów w dziedzinie gospodarowania zasobami oraz prezentację rachunków ReMEA. Słowa kluczowe: zasoby naturalne, koszty zarządzania zasobami Wprowadzenie Zgodnie z szacunkami Organizacji Narodów Zjednoczonych, przy zachowa- niu obecnego tempa produkcji, konsumpcji i wzrostu liczby ludności, w 2030 roku 10 Gospodarka regionalna i międzynarodowa będzie wymagana około dwukrotna wydajność zasobów Ziemi (Global Footprint Network, 2016). W odpowiedzi na ten niepokojący trend organizacje międzyna- rodowe, rządy poszczególnych państw oraz zainteresowane podmioty gospodarcze zwiększają zaangażowanie w rozwiązywanie problemów związanych z racjonalno- ścią zużycia zasobów (Koh, Morris, Ebrahimi, Obayi, 2015, s. 1576). Racjonalne wykorzystanie zasobów naturalnych polega na przestrzeganiu kilku elementarnych zasad (Czaja, 2013, s. 216): –– bilansowania zasobów i rozchodów, –– bilansowania zasobów i rozchodów, –– nieuszczuplania zasobów kapitału naturalnego, zwłaszcza podstawowego kapitału naturalnego, –– zachowania różnorodności komponentów kapitału naturalnego, –– substytuowania kapitału naturalnego innymi formami kapitału, –– minimalizacji źródła entropii, –– racjonalnego korzystania z zasobów kapitału naturalnego (stosowanie kry- terium maksymalizacji użyteczności osiąganej z kapitału naturalnego), –– efektywności wykorzystania elementów kapitału naturalnego (nadwyżki korzyści nad kosztami). Wprowadzenie tych zasad w życie wymaga odpowiedniej wiedzy, wprowa- dzenia ram prawnych oraz monitorowania kierunku i tempa zachodzących zmian. Celem artykułu jest przedstawienie rachunków wydatków na gospodarowa- nie zasobami jako jedynego spośród rachunków dotyczących zasobów naturalnych, prowadzonego w jednostkach pieniężnych. 2  Ostatnio na temat tej koncepcji pisali między innymi: D. Pieńkowski, E. Kośmicki (2016). Funkcja produkcji gospodarki zamkniętego obiegu. Ekonomia i Środowisko 2 (57), 10–22 oraz M. Burchard-Dziubińska (2017). Cradle to Cradle Approach in Development of Resource-Efficient Economy. Ekonomia i Środowisko 1 (60), 8–17. 1  Wskaźnik ten jest również określany jako wskaźnik produktywności zasobów. 3  Obejmują one na przykład metale ziem rzadkich i inne metale szlachetne oraz fosfor 1. Inicjatywy racjonalnego gospodarowania zasobami Efektywność wykorzystania zasobów naturalnych stała się ważnym przedmio- tem zainteresowania Komisji Europejskiej po 2010 roku. Jedną z inicjatyw „Stra- tegii na rzecz inteligentnego i zrównoważonego rozwoju sprzyjającego włączeniu społecznemu Europa 2020” jest inicjatywa przewodnia dotycząca efektywnego korzystania z zasobów, zawarta w Komunikacie „Europa efektywnie korzystająca z zasobów” (Komisja Europejska, 2011a). Ma ona na celu stworzenie warunków wspierających zmiany prowadzące do przejścia na niskoemisyjną gospodarkę opar- tą na efektywnym korzystaniu z zasobów przy jednoczesnym zapewnieniu wzrostu gospodarczego. Określone w dokumencie narzędzia do monitorowania i mierzenia Elżbieta Broniewicz Rachunki wydatków na gospodarowanie zasobami naturalnymi Elżbieta Broniewicz 11 postępu w zakresie efektywnego korzystania z zasobów dotyczą głównie aspektów technicznych, przykładowo: ograniczenia emisji gazów cieplarnianych, zwiększe- nia udziału odnawialnych źródeł energii w całkowitym zużyciu energii czy zwięk- szenia efektywności wykorzystania energii. W kolejnym Komunikacie z 2011 roku pt. Plan działania na rzecz zasobo- oszczędnej Europy podano dwa poziomy wskaźników (Komisja Europejska, 2011b, s. 4): 1. Tymczasowy podstawowy wskaźnik – „wydajność zasobów”1 – dla oszaco- wania poprawy wyniku ekonomicznego przy jednoczesnym zmniejszeniu presji na zasoby naturalne; obliczany jako iloraz PKB i zużycia materia- łów, jest badany od 2000 roku przez Eurostat. W Unii Europejskiej w 2004 roku przełamano zależność między korzystaniem z zasobów a rozwojem gospodarczym. W latach 2000–2015 wskaźnik wzrósł o 36,5% (Eurostat, 2015, 2017), co oznacza wzrost gospodarczy zachodzący przy zużyciu co- raz mniejszej ilości zasobów naturalnych. 2. Zbiór uzupełniających wskaźników dotyczących kluczowych zasobów na- turalnych, takich jak woda, grunty, surowce i węgiel, w celu uwzględniania globalnego zużycia tych zasobów przez UE, które mają być dookreślone w przyszłości (Komisja Europejska, 2011b, s. 24). Inną znaczącą inicjatywą związaną z racjonalnym gospodarowaniem zaso- bami jest japońskie podejście 3R – reduce, reuse, recycle (Takiguchi, Takemoto, 2008). Jego autorzy, oprócz wskaźnika produktywności zasobów (resources pro- ductivity), zaproponowali dwa dodatkowe wskaźniki: Inną znaczącą inicjatywą związaną z racjonalnym gospodarowaniem zaso- bami jest japońskie podejście 3R – reduce, reuse, recycle (Takiguchi, Takemoto, 2008). Jego autorzy, oprócz wskaźnika produktywności zasobów (resources pro- ductivity), zaproponowali dwa dodatkowe wskaźniki: –– cyclical use rate – reprezentujący odsetek ponownie wykorzystanych mate- riałów w ogólnej ilości materiałów zużytych przez gospodarkę, –– final disposal amount – będący sumą ilości składowanych odpadów komu- nalnych i przemysłowych. 1. Inicjatywy racjonalnego gospodarowania zasobami Nie można też pominąć koncepcji circular economy2, pomysłodawcą której byli Pearce i Turner (1989), a która w 2012 roku stała się przedmiotem manifestu 12 Gospodarka regionalna i międzynarodowa Komisji Europejskiej Manifesto For A Resource-Efficient Europe (Komisja Euro- pejska, 2013) i wezwaniem dla rządzących, przedsiębiorców i całego społeczeństwa do wspierania efektywnego gospodarowania zasobami i transformacji gospodarki i społeczeństwa w kierunku circular economy. Dwa lata później w Komunikacie Komisji Ku gospodarce o obiegu zamkniętym: program „zero odpadów dla Eu- ropy” (2014) zapowiedziano przechodzenie na gospodarkę o bardziej zamkniętym obiegu, a w 2015 roku opracowano kompleksowy dokument: Zamknięcie obiegu – plan działania UE dotyczący gospodarki o obiegu zamkniętym (Komisja Europej- ska, 2015). Odniesiono się w nim do następujących kwestii: produkcji, konsumpcji, gospodarowania odpadami, stymulowania rynku surowców wtórnych i ponownego wykorzystywania wody, a także określono obszary priorytetowe, mające szczególne znaczenie dla gospodarki o obiegu zamkniętym – tworzywa sztuczne, odpady spo- żywcze, surowce krytyczne3, odpady z budowy i rozbiórki, biomasa. Opracowanie ram monitorowania gospodarki o obiegu zamkniętym przewidziano na rok 2017. Odpowiadając na zgłoszone w dokumentach UE problemy z monitorowaniem różnego rodzaju aspektów związanych z gospodarowaniem zasobami, również Biu- ro Statystyczne Unii Europejskiej Eurostat włączyło się w prace w tej sferze. W 2001 roku Eurostat zaproponował rachunki przepływów materiałowych gospodarki Eco- nomy-Wide Material Flow Accounts (EW-MFA) (Eurostat, 2013), które ewidencjo- nują wszystkie przepływy materiałowe (z wyjątkiem wody i powietrza) na granicy dwóch systemów – środowiska przyrodniczego i gospodarki, a także między grani- cami poszczególnych krajów (Telega, 2015, s. 39). Rachunki są prowadzone w jed- nostkach naturalnych (masa materiałów) przez Główny Urząd Statystyczny i dane dla Polski są dostępne od 2000 roku (Baza Danych Eurostat). Innymi europejskimi rachunkami związanymi z zasobami naturalnymi są (Domańska, 2013, s. 6–20): Innymi europejskimi rachunkami związanymi z zasobami naturalnymi są (Domańska, 2013, s. 6–20): –– rachunki emisji do powietrza „Air Emissions Accounts (AEA)” – Polska wypełnia je od 2013 roku), –– rachunki emisji do powietrza „Air Emissions Accounts (AEA)” – Polska wypełnia je od 2013 roku), –– rachunki fizycznych przepływów energii „Physical Energy Flow Accounts (PEFA)” – obecnie prowadzone są prace metodologiczne, a pierwszą trans- misję danych do Komisji Europejskiej przewidziano w 2017 roku, Elżbieta Broniewicz Rachunki wydatków na gospodarowanie zasobami naturalnymi Elżbieta Broniewicz 13 –– rachunki wody, rachunki leśnictwa, rachunki dotyczące odpadów – plano- wane do wdrożenia po 2018 roku. Obecnie w niektórych krajach UE są pro- wadzone badania pilotażowe, jednak Polska w nich nie uczestniczy. 1. Inicjatywy racjonalnego gospodarowania zasobami –– rachunki wody, rachunki leśnictwa, rachunki dotyczące odpadów – plano- wane do wdrożenia po 2018 roku. Obecnie w niektórych krajach UE są pro- wadzone badania pilotażowe, jednak Polska w nich nie uczestniczy. W ramach europejskich rachunków ekonomicznych środowiska Eurostat za- proponował rachunki wydatków na gospodarowanie zasobami „Resource Manage- ment Expenditure Accounts (ReMEA)”. Są to jedyne rachunki dotyczące zasobów, które będą prowadzone w jednostkach pieniężnych, analogicznie do rachunków wy- datków na ochronę środowiska „Environmental Protection Expenditure Accounts (EPEA)”. Rachunki ReMEA są przewidziane do wprowadzenia do podstawy praw- nej Unii Europejskiej po roku 2020. 2. Idea rachunków wydatków na gospodarowanie zasobami Koncepcja rachunków wydatków na gospodarowanie zasobami w pracach Eu- rostatu pojawiła się po raz pierwszy w 1994 roku jako część Europejskiego Sys- temu Gromadzenia Informacji Ekonomicznej o Środowisku (European System for the Collection of Economic Information on the Environment – SERIEE). Rachunki miały obejmować rejestrowanie działań związanych z zarządzaniem zasobami na- turalnymi (SERIEE, 1994, s. 24): –– wodą i innymi zasobami (lasy, gleba, energia itp.), –– działania związane z recyklingiem i odzyskiem, które nie są objęte rachun- kiem wydatków na ochronę środowiska EPEA. –– działania związane z recyklingiem i odzyskiem, które nie są objęte rachun- kiem wydatków na ochronę środowiska EPEA. Określono wówczas bardzo ogólnie ramy rachunków oraz przyjęto, iż prace metodologiczne będą prowadzone w przyszłości. Efektem tego jest wstępna wersja metodyki kompilacji rachunków wydatków na gospodarowanie zasobami opraco- wana w 2014 roku (ReMEA, 2014). Rachunki ReMEA obejmują wszystkie działania związane z zarządzaniem za- sobami naturalnymi i zostały sklasyfikowane jako (CEA, 2011): CReMA 10 – Gospodarka wodna. CReMA 11 – Gospodarka zasobami leśnymi. CReMA 12 – Zarządzanie dziką fauną i florą. CReMA 13 – Gospodarka zasobami energetycznymi. CReMA 13A – Produkcja energii ze źródeł odnawialnych. CReMA 13B – Oszczędzanie ciepła/energii i zarządzanie nimi CReMA 13 – Gospodarka zasobami energetycznymi. CReMA 13A – Produkcja energii ze źródeł odnawialnych. CReMA 13B – Oszczędzanie ciepła/energii i zarządzanie nimi. 14 Gospodarka regionalna i międzynarodowa CReMA 13C – Ograniczanie stosowania paliw kopalnych jako surowców. CReMA 14 – Gospodarka surowcami mineralnymi. CReMA 15 – Działalność badawczo-rozwojowa w zakresie gospodarki za- sobami. CReMA 16 – Pozostała działalność w zakresie gospodarki zasobami. CReMA 16 – Pozostała działalność w zakresie gospodarki zasobami. Działalność uwzględniana w rachunkach ReMEA musi spełniać kryterium głównego celu, a głównym celem jest gospodarowanie zasobami. Wszelkie działa- nia mające korzystny wpływ na zasoby naturalne, ale podejmowane z innych wzglę- dów, nie podlegają zakwalifikowaniu do ReMEA. W rachunkach ReMEA uwzględniane są: W rachunkach ReMEA uwzględniane są: W rachunkach ReMEA uwzględniane są: Charakterystyczne działania gospodarowania zasobami: 1. Charakterystyczne działania gospodarowania zasobami: –– działania mające na celu zmniejszenie zużycia zasobów: odzyskiwanie, ponowne wykorzystanie, recykling, racjonalne użycie, zastąpienie zaso- bów naturalnych, –– działania uzupełniające: zwiększanie odnawialnych zasobów natural- nych (wód, lasów, dzikiej flory i fauny), –– zarządzanie zasobami naturalnymi i działania w zakresie regulacji pro- wadzone przez rząd, –– monitorowanie, kontrola i nadzór, –– działania dydaktyczne, szkoleniowe, informacyjne i komunikacyjne, –– działania w zakresie badań i rozwoju w dziedzinie zarządzania zasobami naturalnymi. 2. Produkty związane z gospodarowaniem zasobami naturalnymi: 2. 2. Idea rachunków wydatków na gospodarowanie zasobami Jako źródło informacji niezbędnej do wypełnienia rachunków, Eurostat propo- uje wykorzystanie istniejących źródeł informacji: Jako źródło informacji niezbędnej do wypełnienia rachunków, Eurostat propo- nuje wykorzystanie istniejących źródeł informacji: –– rachunków produkcji i rachunków tworzenia dochodów; –– tabel podaży i wykorzystania według branż i produktów; –– danych dotyczących inwestycji według branż i sektorów instytucjonalnych; –– danych dotyczących inwestycji według branż i sektorów instytucjonalnych Gospodarka regionalna i międzynarodowa 16 –– danych dotyczących ekwiwalentów czasu pracy według branż i sektorów instytucjonalnych; –– danych dotyczących ekwiwalentów czasu pracy według branż i sektorów instytucjonalnych; –– statystyki sektora instytucji rządowych i samorządowych COFOG. –– statystyki sektora instytucji rządowych i samorządowych COFOG. Jednak już po wstępnym rozeznaniu można stwierdzić, iż w przypadku Pol- ski, istniejące dane statystyczne nie będą wystarczające. Na podstawie doświad- czeń zgromadzonych przy kompilacji rachunków nakładów na ochronę środowiska EPEA, czy też kwestionariusza dotyczącego sektora dóbr i usług środowiskowych EGSS można stwierdzić, iż poziom bezpośrednio dostępnej informacji jest nikły – ze względu na zakres, szczegółowość czy też opóźnienia czasowe. W przypad- ku EPEA konieczne jest prowadzenie celowych badań kosztów bieżących ochrony środowiska, natomiast dane uzyskane przy EGSS opierały się w dużej mierze na k h Jednak już po wstępnym rozeznaniu można stwierdzić, iż w przypadku Pol- ski, istniejące dane statystyczne nie będą wystarczające. Na podstawie doświad- czeń zgromadzonych przy kompilacji rachunków nakładów na ochronę środowiska EPEA, czy też kwestionariusza dotyczącego sektora dóbr i usług środowiskowych EGSS można stwierdzić, iż poziom bezpośrednio dostępnej informacji jest nikły , p p ęp j j j y – ze względu na zakres, szczegółowość czy też opóźnienia czasowe. W przypad- ku EPEA konieczne jest prowadzenie celowych badań kosztów bieżących ochrony środowiska, natomiast dane uzyskane przy EGSS opierały się w dużej mierze na szacunkach. – ze względu na zakres, szczegółowość czy też opóźnienia czasowe. W przypad- ku EPEA konieczne jest prowadzenie celowych badań kosztów bieżących ochrony środowiska, natomiast dane uzyskane przy EGSS opierały się w dużej mierze na szacunkach. 2. Idea rachunków wydatków na gospodarowanie zasobami Produkty związane z gospodarowaniem zasobami naturalnymi: –– produkty charakterystyczne – służące bezpośrednio i wyłącznie ochro- nie zasobów naturalnych, przykładowo: energia odnawialna, doradztwo w zakresie oszczędzania wody i energii, –– produkty powiązane – niecharakterystyczne produkty, służące wyłącz- nie ochronie zasobów naturalnych, których produkcja nie ma wpływu na zarządzanie zasobami, przykładowo: turbiny wiatrowe, pojemniki do magazynowania wody opadowej, specyficzne oprzyrządowanie pomia- rowe, –– produkty zasoboefektywne – niecharakterystyczne produkty, które mogą występować w wersji zasoboefektywnej, przykładowo: pralki energo- i wodooszczędne, świetlówki energooszczędne itp.; uwzględniane są Elżbieta Broniewicz Rachunki wydatków na gospodarowanie zasobami naturalnymi Elżbieta Broniewicz 15 tylko dodatkowe koszty, jeśli zasoboefektywne produkty są droższe niż ich alternatywne odpowiedniki. Zmienne rachunków wydatków na gospodarowanie zasobami są kompatybilne z rachunkami narodowymi według ESA ’95 i zostały przedstawione w tabeli 1. Tabela 1. Zmienne rachunków ReMEA Od strony dochodów Produkcja w sferze gospodarowania zasobami + import – eksport + transfery i podatki = produkcja w cenach zakupu Od strony wydatków Krajowe zużycie produktów związanych z gospodarowaniem zasobami (spożycie lub konsumpcja pośrednia) + akumulacja brutto ± transfery krajowe przekazane i otrzymane ± transfery przekazane i otrzymane z zagranicy = Krajowe wydatki na gospodarowanie zasobami Źródło: opracowanie własne na podstawie ReMEA (2014), s. 15. Tabela 1. Zmienne rachunków ReMEA Rachunki ReMEA składają się z 7 tabel (ReMEA 2014a): Rachunki ReMEA składają się z 7 tabel (ReMEA 2014a): Tabela 1 – Sektor publiczny – instytucje rządowe i samorządowe. Tabela 2 – Sektor gospodarczy – produkcja pomocnicza. Tabela 3 – Przedsiębiorcy jako producenci rynkowi. Tabela 4 – Całkowita podaż gospodarki zasobami. Tabela 5 – Sektor gospodarstw domowych. Tabela 6 – Transfery. Tabela 1 – Sektor publiczny – instytucje rządowe i samorządowe. Tabela 2 – Sektor gospodarczy – produkcja pomocnicza. Tabela 3 – Przedsiębiorcy jako producenci rynkowi. Tabela 4 – Całkowita podaż gospodarki zasobami. Tabela 5 – Sektor gospodarstw domowych. Tabela 6 – Transfery. Tabela 7 – Gospodarka narodowa – w układzie dziedzin gospodarowania zasobami oraz według sektorów gospodarki. Tabela 7 – Gospodarka narodowa – w układzie dziedzin gospodarowania zasobami oraz według sektorów gospodarki. 3. Dotychczasowe doświadczenia polskie związane z ekonomicznymi aspektami gospodarowania zasobami Wspomniany wcześniej rachunek sektora towarów i usług związanych z ochro- ną środowiska EGSS był pilotażowo prowadzony w Polsce w 2015 roku (CSO, 2015). Obejmuje on zarówno dziedziny ochrony środowiska (CEPA, 2000), jak i gospoda- rowania zasobami (CReMA, 2008). W ramach EGSS badane są następujące zmien- ne: produkcja globalna, wartość dodana, eksport i zatrudnienie. Mimo że metoda kompilacji rachunku jest odmienna, wyniki projektu mogą dać pewne wyobrażenie o skali zjawiska. W przypadku produkcji globalnej, ponad połowę (52%) wartości stanowiła produkcja dóbr i usług z zakresu gospodarowania zasobami. Ze względu na brak informacji, nie zostały uwzględnione dziedziny CReMA 11 – Gospodarka zasobami leśnymi, CReMA 12 – Zarządzanie dziką fauną i florą, CReMA 15 – Działalność badawczo-rozwojowa w zakresie gospodarki zasobami oraz CReMA 16 – Pozostała działalność w zakresie gospodarki zasobami. Elżbieta Broniewicz Elżbieta Broniewicz Rachunki wydatków na gospodarowanie zasobami naturalnymi 17 Rachunki wydatków na gospodarowanie zasobami naturalnymi Rysunek 1. Produkcja globalna EGSS w podziale na dziedziny ochrony środowiska i gospodarowania zasobami w Polsce w 2014 roku (mln zł) 8 070 10 183 8 797 6 127 3 249 541 19 240 2 627 12 901 14 270 11 037 1 754 3 760 CEPA 1 CEPA 2 CEPA 3 CEPA 4 CEPA 5 CEPA 6 CEPA 7 CEPA 8 CEPA 9 CReMA 10 CReMA 13 A CReMA 13 B CReMA 13 C CReMA 14 Źródło: Broniewicz, Domańska (2016), s. 23. Rysunek 1. Produkcja globalna EGSS w podziale na dziedziny ochrony środowiska i gospodarowania zasobami w Polsce w 2014 roku (mln zł) Analizując wielkość produkcji globalnej w poszczególnych dziedzinach ochro- ny środowiska i gospodarowania zasobami można stwierdzić, iż najwyższy udział miała produkcja (rynkowa, nierynkowa i pomocnicza) w dziedzinie CReMA 13A – Produkcja energii ze źródeł odnawialnych, CReMA 10 – Gospodarka wodna oraz CReMA 13B – Oszczędzanie ciepła/energii i zarządzanie nimi. Podsumowanie Podjęcie prac nad kompilacją rachunków wydatków na gospodarowanie zaso- bami wydaje się być koniecznością. Prawdopodobnie niezbędne będzie stworzenie polskiej metodyki badania wielkości wydatków na gospodarowanie zasobami. Ana- logicznie do rachunków wydatków na ochronę środowiska, być może wymagane będą badania celowe w tej sferze we wszystkich sektorach gospodarki. Właściwie w badaniach wydatków na ochronę środowiska gospodarstw domowych pewne aspekty wydatków na gospodarowanie zasobami (przykładowo: termoizolacja bu- dynków, wymiana systemu ogrzewania budynku na energooszczędny, zakup okien energooszczędnych) były już uwzględniane i konieczne będzie jedynie wydzielenie tej sfery działalności ze sfery ochrony środowiska. 18 Gospodarka regionalna i międzynarodowa Prowadzenie rachunków wydatków na gospodarowanie zasobami naturalnymi może również przyczynić się do podniesienia świadomości racjonalnej gospodar- ki zasobami w sektorze publicznym, przedsiębiorstwach oraz gospodarstwach do- mowych, co miało miejsce na początku XXI wieku podczas pierwszych lat badań kosztów bieżących ochrony środowiska. To z kolei może skutkować rzeczywistym zwiększeniem stopnia racjonalnego korzystania z zasobów naturalnych. Literatura Broniewicz, E., Domańska, W. (2016). Rachunki sektora towarów i usług związanych z ochroną środowiska (EGSS) jako moduł Europejskich Rachunków Ekonomicznych Środowiska. Wiadomości Statystyczne, 4. CEA (2011). Classification of Environmental Activities (CEA), Classification of Environ- mental Activities (CEA). Pobrano z: https://unstats.un.org/unsd/class/intercop/expert- group/2011/AC234-29.PDF. CSO (2015). Report and the Methodology Description. Module for Environmental Goods and Services Sector (EEEA-EGSS), within the Grant Agreement No. 05121.2013.003- 2013.344. Warszawa. Czaja, S. (2013). Szanse i zagrożenia środowiskowe cywilizacji z punktu widzenia ekonomii zrównoważonego i trwałego rozwoju. Handel Wewnętrzny, 6, 210–221. Global Footprint Network (2016). World Footprint. Do We Fit on the Planet? Pobrano z: http://old.footprintnetwork.org/en/index.php/GFN/page/world_footprint. Domańska, W. (2013). Europejskie Rachunki Ekonomiczne Środowiska (EREŚ). Biało- wieża. Pobrano z: http://bialystok.stat.gov.pl/gfx/bialystok/userfiles/_public/semina- ria_i_konferencje/w.domanska.pdf. Eurostat. Baza danych. Pobrano z: http://ec.europa.eu/eurostat/data/database. Eurostat (2013). Economy-wide Material Flow Accounts (EW-MFA), Compilation Guide 2013. Pobrano z: http://ec.europa.eu/eurostat/web/environment/methodology. Eurostat (2015). Sustainable Development in the European Union. 2015 Monitoring Re- port of the EU Sustainable Development Strategy. Publications Office of the European Union. Pobrano z: p://ec.europa.eu/eurostat/documents/3217494/6975281/KS-GT-15- 001-EN-N.pdf. Eurostat (2017). Sustainable Development Indicators. Theme 2: Sustainable consumption and production. Pobrano z: http://ec.europa.eu/eurostat/web/sdi/indicators/sustaina- ble-consumption-and-production. Elżbieta Broniewicz Rachunki wydatków na gospodarowanie zasobami naturalnymi Elżbieta Broniewicz 19 Koh, S.C.L., Morris, J., Ebrahimi, S.M., Obayi, R. (2015). Integrated Resource Efficiency: Measurement and Management. International Journal of Operations & Production Management, 36 (11), 1576–1600. DOI: 10.1108/IJOPM-05-2015-0266. Komisja Europejska (2011a). Europa efektywnie korzystająca z zasobów – inicjatywa prze- wodnia strategii „Europa 2020”. Komunikat Komisji do Parlamentu Europejskiego, Rady, Europejskiego Komitetu Ekonomiczno-Społecznego i Komitetu Regionów. Bruksela. Komisja Europejska (2011b). Plan działania na rzecz zasobooszczędnej Europy. Komunikat Komisji do Parlamentu Europejskiego, Rady, Europejskiego Komitetu Ekonomiczno- -Społecznego i Komitetu Regionów. Bruksela. Komisja Europejska (2013). Manifesto for a resource-efficient Europe. Brussels, 17 De- cember 2012. Pobrano z: http://europa.eu/rapid/press-release_MEMO-12-989_en.htm. Komisja Europejska (2014). Ku gospodarce o obiegu zamkniętym: program „zero odpadów dla Europy”. Komunikat Komisji do Parlamentu Europejskiego, Rady, Europejskiego Komitetu Ekonomiczno-Społecznego i Komitetu Regionów. Bruksela. Komisja Europejska (2015). Zamknięcie obiegu – plan działania UE dotyczący gospodarki o obiegu zamkniętym. Komunikat Komisji do Parlamentu Europejskiego, Rady, Euro- pejskiego Komitetu Ekonomiczno-Społecznego i Komitetu Regionów. Bruksela. Pearce, D.W., Turner, R.K. (1989). Economics of Natural Resources and the Environment. Johns Hopkins University Press. ReMEA (2014). Resource Management Expenditure Accounts – Draft Guidelines 2014. Ko- misja Europejska. Eurostat. Pobrano z: http://ec.europa.eu/eurostat/web/environment/ methodology. ReMEA (2014a). Resource Management Expenditure Accounts – Tables 2014. Komisja Eu- ropejska. Eurostat. Pobrano z: http://ec.europa.eu/eurostat/web/environment/methodo- logy. SERIEE (1994). Literatura European System for the Collection of Economic Information on the Environment – 1994 Version. European Community. Pobrano z: http://ec.europa.eu/ eurostat/web/environment/methodology. Takiguchi, H., Takemoto, K. (2008). Japanese 3R Policies Based on Material Flow Analy- sis. Journal of Industrial Ecology, 12 (11), 1546–1566. Telega, I. (2015). Kapitał naturalny a proces ekologizacji gospodarki. W: M. Kożuch (red.), Ekologizacja gospodarki (s. 32–47). Kraków: Fundacja Uniwersytetu Ekonomicznego w Krakowie. 20 Gospodarka regionalna i międzynarodowa Abstract The paper presents several of resource management initiatives that have been under- taken in Europe in the last decade. Difficulties were noted in monitoring the existing state and the changes in the management of natural resources. The solution to this problem could be the compilation of European resource management expenditure accounts (ReMEA), pro- posed for implementation in the European Union countries after 2020. The purpose of this paper is to present resource management expenditure accounts as the only ones of the natu- ral resources accounts held in monetary units. This article reviews legislative acts, concepts and plans in the field of resource management and presents ReMEA accounts. Keywords: natural resources, resource management expenditure JEL codes: Q2, Q3, Q56 Keywords: natural resources, resource management expenditure JEL codes: Q2, Q3, Q56
https://openalex.org/W4313033076
http://jurnal.medanresourcecenter.org/index.php/SIN/article/download/115/109
Indonesian
null
Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan
Sintaks
2,021
cc-by
1,788
Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan | 1 Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan | 1 Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan Universitas Islam Sumatera Utara, Indonesia Universitas Islam Sumatera Utara, Indonesia Pengaruh Model g g g g Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan Dwi Oktaviani1, Deliani2, Ali3 Dwi Oktaviani1, Deliani2, Ali3 This study aims to describe the ability to write short stories using the Suggestion-Imagination learning model with the media song of class XI students of YPK Private Vocational School Medan, to describe the ability to write short stories using the Suggestion- Imagination learning model with video media for class XI students at YPK Private Vocational School Medan, Pengaruh, Sugesti-Imajinasi, Menulis Cerpen CITATION (APA 6th Edition) Dwi Oktaviani1, Deliani2, Ali3 2021. Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan. Sintaks: Jurnal Bahasa dan Sastra Indonesia. Volume 1 (1), page 1 - 3 To describe whether there is a significant effect on the ability to write short stories using the Suggestion-Imagination learning model with the media of song for class XI students of YPK Private Vocational School Medan. The sample in this study was 60 students. From the results of data processing, the average value of the experimental class was 71.6 while the control class was 65.5. Thus it can be said that the results of learning to write short stories in the experimental class are higher than the control class. After testing the hypothesis, the value of tcount = 3.144 was then adjusted to ttable at a significant level of = 5% with dk = (n1+n2) – 2 = (30+30 – 2) = 58, then obtained a significant level of 5% = 2.002 ( by interpolation). Then compared between tcount and ttable, it was obtained tcount 3.144 > ttable 2.002 so that Ho (Nil Hypothesis) was rejected and Ha (Alternative Hypothesis) was accepted. CITATION (APA 6th Edition) Dwi Oktaviani1, Deliani2, Ali3 2021. Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan. Sintaks: Jurnal Bahasa dan Sastra Indonesia. Volume 1 (1), page 1 - 3 *CORRESPONDANCE AUTHOR dwioktaviani04@gmail.com deliani@fkip.uisu.ac.id ali@fkip.uisu.ac.id PENDAHULUAN Sebagaimana yang kita ketahui bahwa terdapat empat kemampuan berbahasa yang wajib dikuasai oleh siswa yaitu, menyimak, berbicara, membaca, dan menulis. Menulis merupkan kemampuan berbahasa yang kompleks. Dikatakan demikian, karena menulis adalah hasil mengorganisasikan ide atau gagasan dari proses mendengarkan, berbicara, dan membaca. Yasir Burhan dalam (Asih, 2016:67) mengungkapkan bahwa “Menulis adalah tindakan melakukan pikiran atau perasasaan.” Tetapi pada saat ini minat siswa dalam proses pembelajaran menulis terhitung masih rendah. Muncul paradigma menulis merupakan kegiatan yang sulit. Hal tersebut tidak sepenuhnya benar tetapi tidak dapat juga disalahkan. Jika proses pembelajaran menulis dilakukan dengan model-model pembelajaran yang menarik, maka proses pembelajaran menulis akan menjadi kegiatan yang menyenangkan. Sebagai kegiatan produktif, menulis merupakan kemampuan yang kompleks yang memerlukan keterampilan dan pengetahuan lebih agar dapat menciptakan sebuah produk tulisan yang baik. Dengan tidak mengesampingkan kegiatan pembelajaran yang lain, dalam proses pembelajaran menulis guru sebagai fasilitator dituntut lebih kreatif dalam memberikan stimulus pada siswa agar tujuan pembelajaran dapat tercapai. Ada berbagai macam model pembelajaran yang dapat diterapkan untuk pengajaran menulis, salah satunya adalah model Sugesti-Imajinasi. Model pembelajaran ini dapat mempermudah siswa dalam pengajaran menulis. Landasan dasar dari model Sugesti-Imajinasi ini adalah pemberian sugesti kepada siswa sebagai stimulus atau rangsangan agar mereka mampu mengemukakan ide atau gagasan yang dimiliki dengan cepat dan tepat. Model pembelajaran seperti ini sangat cocok digunakan untuk kegiatan menulis kreatif. Keunggulan model pembelajaran Sugesti-Imajinasi dalam pembelajaran menulis adalah dapat mempermudah siswa dalam mengeksplorasi ide-ide mereka berdasarkan sugesti yang diberikan. Sugesti yang diberikan guru akan membantu siswa untuk lebih mudah menyusun imajinasi atau gagasan-gagasannya 2 | Dwi Oktaviani1, Deliani2, Ali3 2 menjadi sebuah tulisan. Sugesti juga dapat membantu membuka kembali pengalaman yang telah dimiliki siswa sehingga mereka dapat menuliskannya menjadi sebuah tulisan yang menarik. menjadi sebuah tulisan. Sugesti juga dapat membantu membuka kembali pengalaman yang telah dimiliki siswa sehingga mereka dapat menuliskannya menjadi sebuah tulisan yang menarik. Pada penelitian ini media lagu dipilih sebagai alat bantu untuk merangsang ide atau gagasannya dalam menulis cerpen. Media lagu merupakan media audio yang memanfaatkan unsur suara dalam menyaimpaikan pesan informasi dan pengetahuan kepada pendengar. Penelitian ini memilih model pembelajaran Sugesti-Imajinasi dalam pembelajaran menulis cerpen, diharapkan dapat meningkatkan kemampuan menulis cerpen. Berdasarkan uraian di atas, maka peneliti ingin melakukan penelitian dengan judul “Pengaruh Model Sugesti-Imajinasi Dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan.” PEMBAHASAN Setelah melakukan prosedur penelitian seperti uji normalitas, uji homogenitas dan pengujian hipotesis, akhirnya dapat ditemukan hasil penelitian. Pembelajaran menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan menggunakan media lagu bunda ternyata berpengaruh positif dan lebih baik daripada pembelajaran menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan menggunakan video. Hal ini dapat dibuktikan pada hasil penelitian, bahwa nilai rata-rata kemampuan menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media lagu bunda lebih tinggi, yakni sebesar 71,6 dari pada nilai rata-rata kemampuan menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media video yaitu sebesar 65,5. Berdasarkan pengujian normalitas dan pengujian homogenitas, bahwa diketahui data pada kedua kelas yaitu kelas eksperimen dan kelas kontrol berdistribusi normal. Berdasarkan hasil nalisis data dengan menggunakan uji “t” diperoleh thitung = 3,144 selanjutnya disesuaikan dengan ttabel pada taraf signifikan 5% dengan dk = (n1+n2) - 2 = (30+30-2) = 58, maka diperoleh taraf signifikan 5% = 2,002 (dengan interpolasi). Kemudian dibandingkan antara thitung dengan ttabel diperoleh thitung 3,144>ttabel 2,002 sehingga diperoleh Ho (Hipotesis Nihil) ditolak dan Ha (Hipotesis Alternatif) diterima. Dapat disimpulkan bahwa model Sugesti-Imajinsi dengan media lagu bunda berpengaruh positif dalam meningkatkan kemampuan menulis cerpen. Setelah diperoleh hasil dari penelitian ini, selanjutnya akan dibahas mengenai model Sugesti- Imjinasi dengan media lagu lebih berhasil atau lebih baik bila dibandingkan dengan model Sugesti-Imajinasi dengan media video Hal ini dapat dijelaskan bahwa pembelajaran dengan menggunakan model Sugesti- Imajinasi adalah suatu model pembelajaran yang memusatkan pembelajaran pada penciptaan ide kreatif untuk menulis cerpen Kemampuan menulis, khususnya menulis cerpen adalah kemampuan yang dimiliki seseorang untuk menciptakan sebuah karangan dengan suatu imajinasi sehingga menjadi tulisan yang menarik untuk dibaca. Berdasarkan hasil penelitian yang telah dilakukan, terlihat perbedaan kemampuan menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media lagu bunda dengan siswa yang diajarkan dengan model Sugesti-Imajinasi dengan media video. Dari perbedaan tersebut dapat disimpulkan bahwa kelas yang diajarkan dengan model Sugesti-Imajinasi dengan media lagu bunda lebih berpengaruh digunakan dalam proses pembelajaran menulis cerpen dibandingkan dengan model Sugesti-Imajinasi dengan media video pada siswa kelas XI SMK Swasta YPK Medan. SIMPULAN Berdasarkan hasil penelitian dan pembahasan yang yang telah dilakukan, maka dapat disimpulkan sebagai berikut : 1. Hasil pembelajaran menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media lagu bunda lebih efektif dibandingkan dengan hasil pembelajaran menulis cerpen dengan menggunakan model Sugesti- Imajinasi dengan media video. 1. Hasil pembelajaran menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media lagu bunda lebih efektif dibandingkan dengan hasil pembelajaran menulis cerpen dengan menggunakan model Sugesti- Imajinasi dengan media video. 2. Dari hasil pengolahan data, diperoleh nilai rata-rata kelas eksperimen adalah 71,6 sedangkan kelas kontrol 65,5. Dengan demikian dapat dikatakan bahwa hasil pembelajaran menulis cerpen di kelas eksperimen lebih 2. Dari hasil pengolahan data, diperoleh nilai rata-rata kelas eksperimen adalah 71,6 sedangkan kelas kontrol 65,5. Dengan demikian dapat dikatakan bahwa hasil pembelajaran menulis cerpen di kelas eksperimen lebih Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan | 3 Pengaruh Model Sugesti – Imajinasi dengan Media Lagu Bunda Terhadap Kemampuan Menulis Cerpen Siswa Kelas XI SMK Swasta YPK Medan | 3 3 tinggi dari kelas kontrol. Setelah dilakukan pengujian hipotesis diperoleh nilai thitung = 3,144 selanjutnya disesuaikan dengan ttabel pada taraf signifikan α = 5% dengan dk = (n1+n2) – 2 = (30+30 – 2) =58, maka diperoleh taraf signifikan 5% = 2,002 (dengan interpolasi). Kemudian dibandingkan antara thitung dengan ttabel diperoleh thitung 3,144 > ttabel 2,002 sehingga diperoleh Ho (Hipotesis Nihil) ditolak dan Ha (Hipotesis Alternatif) diterima. 3. Terdapat perbedaan yang signifikan antara hasil pembelajaran menulis cerpen dengan model Sugesti- Imajinasi dengan media lagu bunda dengan hasil pembelajaran menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media video. Perbedaan kemampuan menulis cerpen tersebut dapat diketahui dari uji “t” yang dilakukan pada skor post-test kelompok kelas kontrol. 3. Terdapat perbedaan yang signifikan antara hasil pembelajaran menulis cerpen dengan model Sugesti- Imajinasi dengan media lagu bunda dengan hasil pembelajaran menulis cerpen dengan menggunakan model Sugesti-Imajinasi dengan media video. Perbedaan kemampuan menulis cerpen tersebut dapat diketahui dari uji “t” yang dilakukan pada skor post-test kelompok kelas kontrol. Asih.2016.Strategi Pembelajaran Bahasa Indonesia.Bandung: CV Pustaka Setia. Antono, Yayan.2016.”Keefektifan Model Pembelajaran Sugesti – Imajinasi Berbantuan Media Video Klip Dalam P b l j M li C ”Sk i i S1 Y j k P di PBSI Antono, Yayan.2016.”Keefektifan Model Pembelajaran Sugesti – Imajinasi Berbantuan Media Video Klip Dalam Pembelajaran Menulis Cerpen.”Skripsi S1. Yogjakarta.Prodi PBSI. Keefektifan Model Pembelajaran Sugesti – Imajinasi Berbantuan Media Video Klip Dalam kunto, Suharsimi. 2006. Prosedur Penelitian Suatu Pendekatan Praktik. Jakarta: RinekaCipta. FKIP UISU.2019.Panduan Penulisan Skripsi.Medan:FKIP UISU. tahul. 2013. Model-Model Pengajaran Dan Pembelajaran. Yogyakarta : Pustaka pelajar Nurgiantoro, Burhan.1998.Teori pengkajian fiksi.Yogyakarta:Gadjah Mada. University Press. Pribadi, A.Benny.2018.Media Teknologi Dalam Pembelajaran.Jakarta: Kencana. Rianto.2017.”Efektivitas Penggunaan Metode Sugesti-Imajinasi Melalui Media Audio-Visual Dalam Pembelajaran M li C Di SMA K l X ” Sk i i S1 P di PBSI Rianto.2017.”Efektivitas Penggunaan Metode Sugesti-Imajinasi Melalui Media Audio-Visual Dalam Pembelajaran Menulis Cerpen Di SMA Kelas X.” Skripsi S1.Prodi PBSI. Menulis Cerpen Di SMA Kelas X.” Skripsi S1.Prodi PBSI. Sudjono, Anas. 2010. Pengantar Statistik Pendidikan. Jakarta: Rajawali Pers. Menulis Cerpen Di SMA Kelas X.” Skripsi S1.Prodi PBSI. Sudjono, Anas. 2010. Pengantar Statistik Pendidikan. Jakarta: Rajawali Pers. Sugitono. 2011. Statistika Untuk Penelitian. Bandung: Alfabeta, cv Setyosari,P j , g j Sugitono. 2011. Statistika Untuk Penelitian. Bandung: Alfabeta, cv Setyosari,Punaj Sugitono. 2011. Statistika Untuk Penelitian. Bandung: Alfabeta, cv Setyosari,Punaji. 2013. Metode Penelitian Pendidikan dan Pengembangan. Jakarta : Kencana. Pendidikan dan Pengembangan. Jakarta : Kencana. Pendidikan dan Pengembangan. Jakarta : Kencana.
https://openalex.org/W2972055172
https://www.frontiersin.org/articles/10.3389/fgene.2019.00955/pdf
English
null
Molecular Etiology Disclosed by Array CGH in Patients With Silver–Russell Syndrome or Similar Phenotypes
Frontiers in genetics
2,019
cc-by
11,912
Original Research doi: 10.3389/fgene.2019.00955 published: 15 October 2019 Molecular Etiology Disclosed by Array CGH in Patients With Silver–Russell Syndrome or Similar Phenotypes Milena Crippa 1,2, Maria Teresa Bonati 3, Luciano Calzari 1, Chiara Picinelli 1, Cristina Gervasini 4, Alessandra Sironi 1,2, Ilaria Bestetti 1,2, Sara Guzzetti 1, Simonetta Bellone 5, Angelo Selicorni 6, Alessandro Mussa 7, Andrea Riccio 8,9, Giovanni Battista Ferrero 7, Silvia Russo 1, Lidia Larizza 1 and Palma Finelli 1,2* Milena Crippa 1,2, Maria Teresa Bonati 3, Luciano Calzari 1, Chiara Picinelli 1, Cristina Gervasini 4, Alessandra Sironi 1,2, Ilaria Bestetti 1,2, Sara Guzzetti 1, Simonetta Bellone 5, Angelo Selicorni 6, Alessandro Mussa 7, Andrea Riccio 8,9, Giovanni Battista Ferrero 7, Silvia Russo 1, Lidia Larizza 1 and Palma Finelli 1,2* 1 Research Laboratory of Medical Cytogenetics and Molecular Genetics, IRCCS Istituto Auxologico Italiano, Milan, Italy, 2 Department of Medical Biotechnology and Translational Medicine, University of Milan, Milan, Italy, 3 Clinic of Medical Genetics, IRCCS Istituto Auxologico Italiano, Milan, Italy, 4 Medical Genetics, Department of Health Sciences, University of Milan, Milan, Italy, 5 Division of Pediatrics, Department of Health Sciences, University of Piemonte Orientale, Novara, Italy, 6 Department of Pediatrics, ASST-Lariana, Como, Italy, 7 Department of Pediatric and Public Health Sciences, University of Turin, Turin, Italy, 8 Department of Environmental, Biological and Pharmaceutical Sciences and Technologies, University of Campania “Luigi Vanvitelli”,Caserta, Italy, 9 Institute of Genetics and Biophysics “Adriano Buzzati-Traverso,” Consiglio Nazionale delle Ricerche (CNR), Naples, Italy Keywords: Silver–Russell syndrome, Netchine–Harbison clinical scoring system, array CGH, pathogenic CNVs, differential diagnosis Edited by: Specialty section: This article was submitted to Systems Endocrinology, a section of the journal Frontiers in Genetics Received: 11 March 2019 Accepted: 06 September 2019 Published: 15 October 2019 Citation: Crippa M, Bonati MT, Calzari L, Picinelli C, Gervasini C, Sironi A, Bestetti I, Guzzetti S, Bellone S, Selicorni A, Mussa A, Riccio A, Ferrero GB, Russo S, Larizza L and Finelli P (2019) Molecular Etiology Disclosed by Array CGH in Patients With Silver–Russell Syndrome or Similar Phenotypes. Front. Genet. 10:955. doi: 10.3389/fgene.2019.00955 Specialty section: This article was submitted to Systems Endocrinology, a section of the journal Frontiers in Genetics Received: 11 March 2019 Accepted: 06 September 2019 Published: 15 October 2019 Citation: Crippa M, Bonati MT, Calzari L, Picinelli C, Gervasini C, Sironi A, Bestetti I, Guzzetti S, Bellone S, Selicorni A, Mussa A, Riccio A, Ferrero GB, Russo S, Larizza L and Finelli P (2019) Molecular Etiology Disclosed by Array CGH in Patients With Silver–Russell Syndrome or Similar Phenotypes. Front. Genet. 10:955. doi: 10.3389/fgene.2019.00955 Specialty section: This article was submitted to Systems Endocrinology, a section of the journal Frontiers in Genetics Received: 11 March 2019 Accepted: 06 September 2019 Published: 15 October 2019 Edited by: Edited by: Louise Metherell, Queen Mary University of London, United Kingdom Reviewed by: Consolato Sergi, University of Alberta Hospital, Canada Ludmila Matyakhina, Reviewed by: Consolato Sergi, University of Alberta Hospital, Canada Ludmila Matyakhina, GeneDx, United States Thomas Eggermann, RWTH Aachen University, Germany Emily J. H. Cottrell, Queen Mary University of London, United Kingdom Introduction: Silver–Russell syndrome (SRS) is an imprinting disorder primarily caused by genetic and epigenetic aberrations on chromosomes 11 and 7. SRS is a rare growth retardation disorder often misdiagnosed due to its heterogeneous and non-specific clinical features. The Netchine–Harbison clinical scoring system (NH-CSS) is the recommended tool for differentiating patients into clinical SRS or unlikely SRS. However, the clinical diagnosis is molecularly confirmed only in about 60% of patients, leaving the remaining substantial proportion of SRS patients with unknown genetic etiology. University of Alberta Hospital, Canada Ludmila Matyakhina, GeneDx, United States Thomas Eggermann, RWTH Aachen University, Germany Emily J. H. Cottrell, Queen Mary University of London, United Kingdom Materials and Methods: A cohort of 34 Italian patients with SRS or SRS-like features scored according to the NH-CSS and without any SRS-associated (epi)genetic alterations was analyzed by high-resolution array-based comparative genomic hybridization (CGH) in order to identify potentially pathogenic copy number variants (CNVs). *Correspondence: Palma Finelli finelli@auxologico.it; palma.finelli@unimi.it Results and Discussion: In seven patients, making up 21% of the initial cohort, five pathogenic and two potentially pathogenic CNVs were found involving distinct genomic regions either previously associated with growth delay conditions (1q24.3-q25.3, 17p13.3, 17q22, and 22q11.2-q11.22) and with SRS spectrum (7p12.1 and 7p15.3-p14.3) or outlined for the first time (19q13.42), providing a better definition of reported and as yet unreported SRS overlapping syndromes. All the variants involve genes with a defined role in growth pathways, and for two genes mapping at 7p, IGF2BP3 and GRB10, the association with SRS turns out to be reinforced. The deleterious effect of the two potentially pathogenic variants, comprising GRB10 and ZNF331 genes, was explored by targeted approaches, though further studies are needed to validate their pathogenic role in the SRS etiology. In conclusion, we reconfirm the utility of performing a genome-wide scan to achieve a differential diagnosis in patients with SRS or similar features and to highlight novel chromosome alterations associated with SRS and growth retardation disorders. INTRODUCTION Fuke et al., 2013; Inoue et al., 2017; Meyer et al., 2017; Sachwitz et al., 2017; Tumer et al., 2018). To this purpose, and to identify novel rearrangements and genes related to SRS, we performed high-resolution array-based comparative genomic hybridization (aCGH) analysis on a group of 34 patients, fulfilling and not fulfilling the NH-CSS criteria, without a molecular diagnosis. Silver–Russell syndrome (SRS; Online Mendelian Inheritance in Man (OMIM) #180860, also known as Russell–Silver syndrome, RSS) is an imprinting disorder with a broad clinical and molecular heterogeneity. The clinical diagnosis of SRS is difficult, as many of its features are non-specific and overlap with those of several other congenital syndromes; moreover, the severity of clinical presentation can be highly variable (Wakeling et al., 2017). As recommended by the recently published first international consensus statement, the Netchine–Harbison clinical scoring system (NH-CSS) (Azzi et al., 2015) should be applied for SRS diagnosis (Wakeling et al., 2017). The NH-CSS includes six key features: being born small for gestational age (SGA) (birth weight/length ≤ −2 standard deviation (SD)), postnatal growth retardation (weight/length ≤ −2 SD), relative macrocephaly at birth (head circumference ≥ 1.5 SD above birth weight and/or length SD), body asymmetry (leg length discrepancy ≥ 0.5 cm or <0.5 cm with at least two other asymmetrical body parts), feeding difficulties and/or low body mass index (BMI) (BMI ≤ −2 SD at 24 months or current use of a feeding tube or cyproheptadine for appetite stimulation), and a protruding forehead (Azzi et al., 2015). A patient receives a diagnosis of SRS when he/she meets four or more of these six criteria (positive NH-CSS ≥4/6), while patients with a negative NH-CSS (<4/6) are generally classified as unlikely SRS (Azzi et al., 2015). However, despite the high sensitivity of this clinical score, patients scoring 4 or more criteria may have negative molecular testing and vice versa (Tümer et al., 2018). In accordance with the statement, patients meeting NH-CSS criteria including both protruding forehead and relative macrocephaly, but normal at all molecular tests, receive a diagnosis of clinical SRS. The two primary molecular causes of SRS are hypomethylation of the distal imprinting control region H19/IGF2:IG-DMR (ICR1) at 11p15.5, identified in about 30–60% of patients, and maternal uniparental disomy of chromosome 7 (upd(7)mat), detected in about 5–10% of patients (Wakeling et al., 2017). INTRODUCTION In addition chromosome 14q32.2 imprinting defects, underpinning Temple syndrome (TS; OMIM#616222), represent an alternative molecular diagnosis to SRS, as numerous molecularly confirmed TS patients (70–72.7%) were found to fulfill the NH-CSS criteria (Kagami et al., 2017; Geoffron et al., 2018). Further rare genetic alterations, that is, point mutations in imprinted genes at 11p15 or in other disease-causing genes, and non-recurrent copy number variants (CNVs) affecting different chromosome regions have been detected in a small number of SRS patients (Bruce et al., 2010; Lin et al., 2010; Spengler et al., 2012; Fuke et al., 2013; Inoue et al., 2017; Meyer et al., 2017; Sachwitz et al., 2017; Wakeling et al., 2017; Tumer et al., 2018). However, in a significant proportion of SRS patients (up t 30%) th l l ti l i k d Citation: October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 1 aCGH in Unsolved SRS Patients Crippa et al. Frontiers in Genetics  |  www.frontiersin.org Study Cohorth The study population consisted of 34 Italian patients (19 males and 15 females) born SGA and/or with postnatal growth retardation referred as SRS for diagnostic testing and found negative for H19/IGF2:IG-DMR LOM and upd(7)mat.f Clinical information was collected by clinicians at different hospitals who participated in the study, using the same clinical datasheet. SRS clinical scoring was based on the NH-CSS (Azzi et al., 2015) with exceptions when postnatal growth and BMI were assessed at different ages, as performed by Meyer et al. (2017). NH-CSS was assessed in all but two patients due to incomplete clinical data. Neonatal SD was calculated according to the Italian Neonatal Study (INeS) charts (http://www.inescharts. com/), whereas postnatal SD was assessed according to Cacciari et al. (2006). Of the 32 patients, 21(66%) satisfied NH-CSS, 19 out of 21 with a score ≥4/6 and two with a score ≥4/5 as the macrocephaly at birth was missing in the latter. Among the patients fulfilling the NH-CSS criteria, 42% (8/19) showed both protruding forehead and relative macrocephaly at birth, consistent with clinical SRS. The remaining 11 patients turned out to have a negative clinical score but were included in the study since they presented other clinical features suggestive of SRS, for example, triangular face and/or micrognathia and/or fifth finger clinodactyly; in detail, seven patients matched three NH-CSS criteria, and four presented only two clinical items. Furthermore, in all patients with a positive clinical score for SRS, aberrant methylation/uniparental disomy of the imprinted locus MEG3-DMR in 14q32 was ruled out. Table 1 reports the clinical data and the presence/absence of the various NH-CSS criteria in our whole SRS cohort and its two clinical subgroups. Table 1 also provides the frequency of each clinical finding of the scoring system in our clinically confirmed SRS subgroup compared with that in literature, showing a substantial level of consistency between the two cohorts. The study was approved by the ethical committee of IRCSS Istituto Auxologico Italiano, and written informed consent was obtained from the parents for publication of the study and any accompanying images. aCGH and CNV Analysis aCGH analysis was performed using the high-resolution Human Genome CGH Microarray Kits, namely, microarray formats 244K (median resolution of 27 kb) (18 patients) and 2 × 400K (median resolution of 16 kb) (16 patients) (Agilent Technologies, Palo Alto, CA) in accordance with the manufacturer’s instructions. Data were then extracted and analyzed for copy number changes using Agilent CytoGenomics 3.0. However, in a significant proportion of SRS patients (up to 30%), the molecular etiology remains unknown, and SRS represents primarily a clinical diagnosis. Genome-wide approaches such as chromosome microarray analysis and next- generation sequencing in patients referred for SRS testing have been shown to enhance the molecular diagnosis, often unveiling SRS overlapping disorders (Bruce et al., 2010; Spengler et al., 2012; October 2019  |  Volume 10  |  Article 955 2 aCGH in Unsolved SRS Patients Crippa et al. TABLE 1 | Distribution of our SRS cohort, negative to 11p15 defects and upd(7)mat, according to NH-CSS criteria and comparison with literature. SRS study cohort 11p15 LOM and upd(7) mat excluded SRS study cohort NH-CSS <4 group SRS study cohort NH-CSS ≥ 4 group Reference SRS cohort NH-CSS ≥4,11p15 LOM and upd(7)mat excluded (Wakeling et al., 2017) Reference SRS cohort 11p15 LOM and upd(7) mat and/or NH-CSS ≥4 (Wakeling et al., 2017) Patient no. and (%) 34 # 11 (34%) 21 (66%) – – Age at examination (years) 4.0 ± 3.7 5.8 ± 4.2 3.0 ± 3.0 – – NH-CSS factors% and (patients no) Birth weight/height ≤ −2 SD 64% (21/33) 54% (6/11) 71% (15/21) 85.7% (12/14) 91.7% (55/60) Relative macrocephaly at birth § 36% (10/28) 22% (2/9) 42% (8/19) 55.4% (31/56) 85.7% (179/209) Postnatal growth failure ‡ 91% (30/33) 82% (9/11) 95% (20/21) 86.8%* (72/83) 84.2% * (267/317) Protruding forehead 80% (25/31) 50% (5/10) 95% (20/21) 77.6% (59/76) 88.1% (177/201) Body asymmetry 41% (13/32) 9% (1/11) 57% (12/21) 39.5% (62/157) 57.3% (/271473) Feeding difficulties and/ or BMI ≤ −2 SD 79% (26/33) 54% (6/11) 95% (20/21) 54.8%** (40/73) 70.4% ** (216/307) Additional signs% and (patients no) Preterm birth 42% (13/31) 55% (5/9) 38% (8/21) n.k. aCGH and CNV Analysis Triangular face 84% (26/31) 90% (9/10) 80% (16/20) 98.7% (73/74) 93.9% (154/164) Micrognathia 55% (17/31) 60% (6/10) 50% (10/20) 56% (5/9) 61.7% (71/115) Downturned mouth 53% (16/30) 30% (3/10) 65% (13/20) 39.1% (9/23) 47.7% (84/176) Low-set and/or posteriorly rotated ears 30% (9/30) 40% (4/10) 25% (5/20) 35.9% (28/78) 49.3% (131/266) Clinodactyly of 5th finger 55% (17/31) 50% (5/10) 57% (12/21) 77.8% (63/81) 74.6% (238/319) 2nd or 3rd toe syndactyly 16% (5/31) 30% (3/10) 9.5% (2/21) 16% (12/75) 29.9% (79/264) Low muscle mass 28% (9/32) 18% (2/11) 33% (7/21) 34.8% (8/23) 56.3% (58/103) Male genital abnormalities 44% (8/18) 50% (3/6) 41.6% (5/12) 37.5% (3/8) 40% (34/85) Psychomotor delay 33% (9/27) 45% (5/11) 25% (4/16) 41.3%*** (26/63) 36.6%*** (93/254) Speech delay 28.5% (6/21) 33% (3/9) 17% (3/12) 38.5% (20/52) 39.7% (75/189) #Incomplete clinical evaluation in 2 patients; §Head circumference SD ≥1.5 higher than birth weight or length SD; ‡Weight and/or length ≤ −2 SD at the age of evaluation; *Only postnatal height considered **Only feeding difficulties considered, ***Only motor delay considered; n.k., not known; SRS, Silver–Russell syndrome; NH-CSS, Netchine– Harbison clinical scoring system. #Incomplete clinical evaluation in 2 patients; §Head circumference SD ≥1.5 higher than birth weight or length SD; ‡Weight and/or length ≤ −2 SD at the age of evaluation; *Only postnatal height considered **Only feeding difficulties considered, ***Only motor delay considered; n.k., not known; SRS, Silver–Russell syndrome; NH-CSS, Netchine– Harbison clinical scoring system. pubmed/, last access March 2016). The involvement of the identified genes in cell cycle, developmental process, and/or cell growth pathways was investigated using Entrez Gene (https:// www.ncbi.nlm.nih.gov/gene) and potential interaction with already-associated SRS genes was assessed with STRING, a tool for Protein–Protein Interaction Networks (https://string-db. org/). Finally, the evaluation of imprinting status of the genes involved in the regions was carried out using the Geneimprint database (http://www.geneimprint.com/site/genes-by-species), and a search for murine knockout phenotypes with growth impairment was made with the Mouse Genome Informatics (MGI) database (Eppig et al., 2012).h Detected CNVs, identified by at least three consecutive aberrant probes, were compared with the Database of Genomic Variants (DGV) (http://projects.tcag.ca/variation/, release March 2016), to exclude common copy number polymorphisms (frequency >1%). A CNV was classified as rare if unreported or reported at a very low frequency («1%) according to the DGV. aCGH and CNV Analysis Inherited rare CNVs were considered automatically validated, whereas the de novo CNVs (or with an unknown inheritance), detected by a number of oligonucleotide consecutive probes ≤5, were validated using quantitative polymerase chain reaction (qPCR).h The establishment of CNV pathogenicity was made considering inheritance, familial segregation, and the consultation of public databases, to assess (i) whether the CNVs overlap with common microdeletion/microduplication syndromes, (ii) the CNVs’ gene content, and (iii) the clinical information of patients with similar CNVs. Specifically, the following public databases were consulted: University of California Santa Cruz (UCSC) (http:// genome.ucsc.edu/, release February 2009), OMIM (www.ncbi. nlm.nih.gov/OMIM), International Standards for Cytogenomic Arrays (ISCA) (https://www.clinicalgenome.org/, update November 2012), DECIPHER v9.10 (http://www.sanger.ac.uk/ PostGenomics/decipher/, last access December 2016) (Bragin et al., 2014), and PubMed (https://www.ncbi.nlm.nih.gov/ The guidelines suggested by Miller et al. (2010) and successively by the American College of Medical Genetics (Kearney et al., 2011) were followed for CNV classification with minor modifications. Microsatellite Analysis To assess the parental origin of the structural variant at 7p15.3p14.3 identified in patient 20, microsatellite analysis was performed using 8 STR markers, that is, D7S513, D7S493, D7S673, D7S2525, D7S2449, D7S516, D7S2496, and D7S526, within chromosome 7p. All fluorescent PCR amplicons were run October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 3 aCGH in Unsolved SRS Patients Crippa et al. SDS v2.3 software (Applied Biosystems) and processed using RQ Manager 1.2 (Applied Biosystems). Relative quantification of the amount of DNA was obtained using the 2−ΔΔCt method (Livak and Schmittgen, 2001). The primer pairs used are summarized in Supplementary Table 1. on capillary electrophoresis using the 3500 Genetic Analyzer (Applied Biosystems). Data analysis and genotyping were carried out by the Genemapper software (Applied Biosystems), matching parents to proband transmission. Pyrosequencing Pyrosequencing analysis was performed in patients 8 and 18 to assess a perturbation in the methylation level, respectively, of GRB10 and ZNF331 differentially methylated regions (DMRs), possibly mediated by the structural variants identified in the patients. Sodium bisulphite conversion of DNA (500–700 ng) was performed by the EZ DNA Methylation Kit (Zymo Research Corporation, Orange, CA). PCR analysis was performed on bisulphite-treated DNA using forward and reverse primers, one of which was biotinylated. y Two assays specific for GRB10:alt-TSS-DMR, that is, fragment I (chr7:50849713-50850034, hg19) and II (chr7:50850569-50850872, hg19), and one assay specific for ZNF331:alt-TSS-DMR2 (chr19:54058016-54058229, hg19) were designed for patients 8 and 18, respectively. Pyrosequencing experiments were performed using specific sequencing primers to quantify four CpG sites for ZNF331, eight for GRB10 fragment I, and six sites for GRB10 fragment II. Quantitative DNA methylation analysis was performed using a Pyro Mark ID instrument (QIAGEN, Silicon Valley, CA) in the PSQ HS 96 System with the PyroGold SQA reagent kit (Diatech Pharmacogenetics srl, Jesi, Italy) according to the manufacturer’s instructions. Raw data were analyzed using Q-CpG software v1.0.9 (Qiagen srl), which calculates the ratio of converted C’s (T’s) to unconverted C’s at each CpG, giving the percentage of methylation. Sequencing Analysis g In patient 28, carrier of a genomic deletion involving the TRIM37 gene, causative of the recessive disease MULIBREY nanism (OMIM#253250), the entire coding sequence, intron– exon junctions, and untranslated exons of the TRIM37 gene (RefSeq Accession: NM_001005207) were amplified for mutation screening by PCR using the AmpliTaq Gold® kit (Thermo Fisher Scientific). Sequencing was performed using the Big Dye® Terminator v.3.1 Cycle Sequencing kit (Thermo Fisher Scientific). Sequences were then aligned to the human reference genome sequence (human genome assembly GRCh37/hg19) and analyzed with the ChromasPro 1.5 software (Technelysium Pty Ltd., Tewantin QLD, Australia). The primer pairs and amplification conditions are summarized in Supplementary Table 2. ( ) Quantitative gene expression analysis was performed in patients 18 and 20 to explore a potential dysregulation in gene expression mediated by the CNVs identified in the patients. Total RNA of patients, their parents, and 10 healthy controls were collected using Tempus Blood RNA tubes (Thermo Fisher Scientific, Waltham, MA), isolated using the Tempus Spin RNA Isolation kit (Thermo Fisher Scientific), and reverse transcribed using the High-Capacity cDNA Reverse Transcription kit (Thermo Fisher Scientific). Quantitative real-time RT-PCR, based on the TaqMan methodology, was performed using an ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Foster City, CA). In patient 18, the amounts of GRB10 mRNA were calculated using the 2−ΔΔCt method (Livak and Schmittgen, 2001), with GAPDH, HMBS, RPRPL0, and TBP as the endogenous-normalizing genes, while in patient 20 the amounts of IGF2BP3 mRNA were calculated with the same method, using the GAPDH, GUSB, and TBP as the endogenous-normalizing genes. All assays were provided by Thermo Fisher Scientific (TaqMan Gene Expression Assays: ID# Hs00959293_m1, GRB10; Hs00946580_m1, IGF2BP3; Hs99999905_m1, GAPDH; Hs00939627_m1, GUSB; Hs00609297_m1, HMBS; Hs99999902_m1, RPLP0; and Hs99999910_m1, TBP). Real-time data were analyzed using the RQ Manager 1.2 software (Applied Biosystems). For each gene of interest, we established the proper range of gene expression in 10 healthy controls by calculating the mean value ± 2 SD. If the expression level of the CNV carrier was out of the control range, a likely dysregulation of the index gene mediated by the CNV was inferred. Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients Overall, chromosomal imbalances with obvious pathogenicity could be identified in five patients (7, 17, 20, 23, and 28) (Table 2). Figure 1 shows the genomic regions containing the pathogenic CNVs of the carrier patients. DNA samples from 34 patients with a clinical suspicion of SRS were screened by aCGH analysis. In total, in 22 of the 34 patients (65%), 40 CNVs classified as rare in the Database of Genomic Variants were detected. Of the 40 variants, 18 were deletions (ranging from 11 kb to 11 Mb), and 22 were duplications (ranging from 22.8 kb to 3 Mb). The aCGH of parental DNA revealed that in four patients, the variants were de novo, whereas in 15 patients, they had been inherited from an unaffected parent. Parental DNA samples were not available in three patients (Supplementary Table 4).h p We identified in patient 7 a de novo deletion of 11 Mb at 1q, arr[GRCh37] 1q24.3q25.3(172652343_183538289)x1 dn, comprising 98 RefSeq genes (Table 2, Figure 1A). Deletions within the critical region 1q24q25 (Figure 1A) have been associated with a distinctive syndromic phenotype characterized by facial dysmorphism, small hands and feet with fifth finger clinobrachydactyly, severe proportionate short stature with microcephaly, and severe cognitive disability (Burkardt et al., 2011; Ashraf et al., 2015). Patient 7 was born preterm with cesarean section due to severe intrauterine growth restriction (IUGR) in the context of altered umbilical arterial flow and insufficient placental supply. In the first months of life, he had feeding difficulties, requiring gastric tube feeding (Table 3). At clinical evaluation at 28 months, he showed growth retardation, slight dolichocephaly, triangular face, slight micro-retrognathia, and teeth and genital abnormalities (Table 3, Figure 2A). He also displayed bilateral fifth finger clino/brachydactyly, flat feet, and delayed bone maturation, as revealed by a skeletal survey performed at 11 months (Figure 2A). Further medical problems included slight global developmental delay, attention deficit, and The role of CNVs in SRS-like clinical presentation was assessed by integrating the analysis of gene content, with a focus on genes involved in growth and development, with literature and database information. We categorized as pathogenic five CNVs, including four de novo CNVs and one inherited unmasking a recessive allele, and as potentially pathogenic (PP) two inherited CNVs (Table 2). All the remaining CNVs (n = 33) were considered either VOUS, that is, variants of unknown significance, or likely benign (LB) CNVs (Supplementary Table 4). Quantitative PCR qPCR analysis was performed on genomic DNA of patients 8 and 20 using SYBR Green methodology (Thermo Fisher Scientific). qPCR was used in patient 8 to finely map the proximal CNV boundary at 19q13.42, and in patient 20 to validate the presence of a small de novo deletion at 7p21.1 detected by three aCGH probes. All amplicons were chosen within non-repeated portions of the chromosomes using Primer3 software (http://bioinfo.ut.ee/primer3-0.4.0/). A control amplicon was selected with the same parameters in the PCNT gene at 11q14.1. Size (approximately 60–100 bp) and Tm (60°C) were the same for all amplicons. Amplification and detection were performed on the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems); thermal cycling conditions were 50°C for 2 min and 95°C for 10 min, followed by 40 cycles at 95°C for 15 s and 60°C for 1  min. Each experiment was performed in triplicate on patients, parents, and two controls. The results were acquired with the For each sample, the methylation value represents the mean between at least two independent PCR and pyrosequencing experiments. For each assay, we established the proper range of methylation in healthy controls by calculating the mean value ± 2 SD; 11 controls for ZNF331, 24 controls for GRB10 fragment I, and eight controls for GRB10 fragment II were used. October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 4 Crippa et al. aCGH in Unsolved SRS Patients Table 3 outlines the NH-CSS on the basis of the main clinical features at birth and at last clinical evaluation of the seven patients found to carry pathogenic (n = 5) and PP (n = 2) CNVs. The primer pairs and amplification conditions are summarized in Supplementary Table 3. Frontiers in Genetics  |  www.frontiersin.org Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients All pathogenic and PP-CNVs identified in patients were submitted to the ClinVar database (http://www.ncbi.nlm.nih.gov/clinvar/, 19 December 2018, date last accessed; accession numbers SCV000863548 to SCV000863554). TABLE 2 | Genomic imbalances detected using array CGH and categorized as pathogenic (P) or potentially pathogenic (PP) variants. Patient# Gain/ loss CNV description according to the ISCN nomenclaturea Size Inheritance No. of affected genes DECIPHERb Genes with functions in cell cycle, developmental process, and cell growth pathways P-CNV #7 Loss arr[GRCh37] 1q24.3q25.3(172652343_183538289)x1 dn 11 Mb De novo 98 13 CENPL, GAS5, TNN, RFWD2, PAPPA2, BRINP2, RASAL2, ABL2, QSOX1, LHX4, CACNA1E, RGS16, LAMC1 #17 Gain arr[GRCh37] 17p13.3(1130776_1361490)x3 dn 231 kb De novo 5 TUSC5, YWHAE, CRK #20 Loss arr[GRCh37] 7p15.3p14.3(23236782_30690453)x1 dn 7.5 Mb De novo 80 5 GPNMB, IGF2BP3, NPY, SKAP2, HOXA1, HOXA2, HOXA3, HOXA4, HOXA5, HOXA6, HOXA7, HOXA9, HOXA10, HOXA11, HOXA13 #23 Loss arr[GRCh37] 22q11.2q11.22(21808950_22963000)x1 dn 1.15 Mb De novo 21 10 UBE2L3, MAPK1 #28 Loss arr[GRCh37] 17q22(57075470_57235248)x1 mat 159.8 kb MAT 3 1 TRIM37 PP-CNV #8 Gain arr[GRCh37] 19q13.42(54039784_54484439)x 3 mat 445 kb MAT 58 ZNF331 #18 Gain arr[GRCh37] 7p12.1(50981149_51956510)x 3 mat 975 kb MAT 1 (GRB10) a International System for Human Cytogenomic Nomenclature (ISCN 2016). bNumber of reported patients with overlapping CNVs and prenatal and/or postnatal growth retardation. () genes possibly deregulated by a position effect mechanism. CGH, comparative genomic hybridization; CNV, copy number variant. October 2019  |  Volume 10  |  Article 955 5 aCGH in Unsolved SRS Patients Crippa et al. TABLE 3 | Clinical data of patients from the study cohort carrying pathogenic (P) or potentially pathogenic (PP) variants. Patient P-CNV PP-CNV 7 17 20 23 28 8 18 NH-CS 3/6 4/6 4/6 2/6 5/6 n.d. 4/5 Sex Male Male Female Female Male Female Female Gestational age 34 38 38+2 35 36 39 32 Birth length(SD) 38 cm (−2.36) 46 cm (−1.76) 43.4 cm (−2.45) 41 cm (−1.99) n.k. (−2) 48 cm (−0.61) 38 cm (−1.51) Birth weight (SD) 1087 g (−2.51) 2500 g (−1.69) 1925 g (−2.6) 1570 g (−2.01) 2060 g (−1.42) 2940 g (−0.54) 1230 g (−1.36) OFC at birth (SD) 27 cm (−2.49) 33.5 cm (−0.57) 31.1 cm (−1.89) 29 cm (−2.02) n.k. (−2) n.k. n.k. Relative macrocephalyat birth No Yes No No No n.k. n.k. Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients Age at examination 28 months 11 months 6 years 1 month 3 years 3 years 2 months 12 years 7 months 11 years Length (SD) 76 cm (−3.77) 75 cm (0.19) 99.8 cm (−3.16) 85.3 cm (−2.45) 79.6 cm (−3.76) 141.5 cm (−1.95) 130 cm (−2.23) Weight (SD) 8.7 kg (−4.55) 7.1 kg (−2.56) 12.1 kg (−4.43) 9.8 kg (−3.52) 8.2 kg (−6.3) 36.5 kg (−1.39) 22.5 kg (−2.9) BMI (SD) 15.3 (−0.88) 12.6 (−3.78) 12.1 (−3.25) 13.47 (−1.84) 12.94 (−3.39) 18.2 (−0.66) 13.3 (−2.7) Relative macrocephaly No Yes Yes No No n.k. No Feeding difficulties Yes Yes Yes No Yes n.k. Yes Protruding forehead No Yes Yes No Yes No Yes Body asymmetry No No No No Yes n.k. Yes Craniofacial dysmorphisms Slight dolichocephaly, triangular face, downturned corners of the mouth, diastema of maxillary central incisors, fusion of upper right central and lateral incisor, slight micro-retrognathia Triangular face, downturned mouth, thin lips Triangular face Triangular face, downturned mouth, thin lips, abnormal ears (Mozzillo et al., 2016) Triangular face, thin lips, micrognathia Deep-set eyes, prognathism, bushy eyebrows Psychomotor developmental delay Yes, only developmental No Yes No (Mozzillo et al., 2016) n.k. Yes Other findings 5th finger clino/ brachydactyly,flat feet, delayed bone maturation, balanic hypospadias with chordae of penis and schisis of prepuce,attention deficit, poor social interaction 5th finger clino/ brachydactyly, right undescended testicle, muscle hypotonia Myopia and astigmatism, small hands with 5th finger clino/ brachydactyly, small feet with short and broad halluces and symmetric skeletal anomalies, duplex collecting system of right kidney 5th finger clino/ brachydactyly, 2nd and 3rd toe syndactyly (Mozzillo et al., 2016) n.k. Gastroesophageal reflux, gastrointestinal abnormalities n.k., not known; CNV, copy number variant; OFC, occipitofrontal head circumference; BMI, body mass index. TABLE 3 | Clinical data of patients from the study cohort carrying pathogenic (P) or potentially pathogenic (PP) variants. No (Mozzillo et al., 2016) n.k. Yes 5th finger clino/ brachydactyly, 2nd and 3rd toe syndactyly (Mozzillo et al., 2016) n.k. Gastroesophageal reflux, gastrointestinal abnormalities and downturned mouth with thin lips. Further medical problems included hypotonia and muscular hypotrophy, typical of 17p13.3 duplication syndrome. reduced social interaction. The small number of clinical features reminiscent of SRS is consistent with a clinical score of 3/6 (Table 3). Frontiers in Genetics  |  www.frontiersin.org Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients Based on the partial overlap of the patient’s CNV with the 1q24q25 microdeletion syndrome critical region, the gene content of this region was taken into account in the re-evaluation of the overall patient’s phenotype. p y In patient 20, a de novo deletion of 7.5 Mb at 7p was detected, arr[GRCh37] 7p15.3p14.3(23236782_30690453)x1 dn, including 80 RefSeq genes (Table 2, Figure 1C). The patient was born at term and showed prenatal as well as postnatal growth restriction (Table 3). She had feeding difficulties and motor delay as she sat up alone at 9 months and walked autonomously at 17 months. Clinical evaluation at 4 years revealed relative macrocephaly, triangular face, and prominent forehead (Figure 2B). Her features satisfied four NH-CSS criteria; however, clinical SRS diagnosis cannot be confirmed, as she did not show relative macrocephaly at birth. Neuropsychological evaluation at age 10 using Wechsler Intelligence Scale for Children (WISC)-III scales (Wechsler, 2002) showed borderline intellectual functioning (QIT: 79; QIV: 78; and QIP: 85), as the girl needed a support teacher at high Patient 17 was found to carry a de novo 230-kb duplication in 17p13.3 affecting five genes, arr[GRCh37] 17p13.3(1130776_1361490)x3 dn, mapping inside the region causative of the 17p13.3 duplication syndrome (OMIM#613215) and the Miller–Dieker syndrome (MDS, OMIM#247200) (Table 2, Figure 1B). The patient was born at term with slight growth restriction, which worsened in the next months when he also presented feeding difficulties (Table 3). His features are consistent with clinical SRS diagnosis: he fulfilled four out of six criteria and showed typical SRS facial appearance characterized by relative macrocephaly, triangular face, prominent forehead, October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 6 aCGH in Unsolved SRS Patients Crippa et al. h l H h i ht i d f 3 16 SD ( 6 ) t 1 84 and feet with symmetric skeletal anomalies (Figure 2B) a IGURE 1 | Physical map of the genomic regions containing the pathogenic CNVs identified by array CGH. (A) 1q24.3-q25-5 region showing the deletion detecte n patient 7. (B) 17p13.3 region showing the duplication detected in patient 17. (C) 7p21.1-p14.3 region showing the deletion detected in patient 20. D) 22q11.21-q11.22 region showing the deletion detected in patient 23 and the low copy repeats (LCR-E and LCR-D) surrounding the deletion bkps. Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients (E) 17q22 egion showing the deletion and the mutation c.1949-12 A > G involving TRIM37 gene detected in patient 28. The RefSeq genes are depicted in dark blue, and he OMIM disease genes in green; the disease critical regions are indicated by black bars, the deletions by red bars, the duplications by blue bars, the deletions of reviously reported SRS patients by red-white bars, and the candidate genes by red boxes. The images are a modification of a version obtained from the UCSC Genome Browser (human genome assembly GRCh37/hg19). CNVs, copy number variants; CGH, comparative genomic hybridization; OMIM, Online Mendelian nheritance in Man; SRS, Silver–Russell syndrome; UCSC, University of California Santa Cruz. IGURE 1 | Physical map of the genomic regions containing the pathogenic CNVs identified by array CGH. (A) 1q24.3-q25-5 region showing the deletion detect n patient 7. (B) 17p13.3 region showing the duplication detected in patient 17. (C) 7p21.1-p14.3 region showing the deletion detected in patient 20. D) 22q11.21-q11.22 region showing the deletion detected in patient 23 and the low copy repeats (LCR-E and LCR-D) surrounding the deletion bkps. (E) 17q22 FIGURE 1 | Physical map of the genomic regions containing the pathogenic CNVs identified by array CGH. (A) 1q24.3-q25-5 region showing the deletion detected in patient 7. (B) 17p13.3 region showing the duplication detected in patient 17. (C) 7p21.1-p14.3 region showing the deletion detected in patient 20. (D) 22q11.21-q11.22 region showing the deletion detected in patient 23 and the low copy repeats (LCR-E and LCR-D) surrounding the deletion bkps. (E) 17q22 region showing the deletion and the mutation c.1949-12 A > G involving TRIM37 gene detected in patient 28. The RefSeq genes are depicted in dark blue, and the OMIM disease genes in green; the disease critical regions are indicated by black bars, the deletions by red bars, the duplications by blue bars, the deletions of i l t d SRS ti t b d hit b d th did t b d b Th i difi ti f i bt i d f th UCSC FIGURE 1 | Physical map of the genomic regions containing the pathogenic CNVs identified by array CGH. (A) 1q24.3-q25-5 region showing the deletion detected in patient 7. (B) 17p13.3 region showing the duplication detected in patient 17. (C) 7p21.1-p14.3 region showing the deletion detected in patient 20. Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients (D) 22q11.21-q11.22 region showing the deletion detected in patient 23 and the low copy repeats (LCR-E and LCR-D) surrounding the deletion bkps. (E) 17q22 region showing the deletion and the mutation c.1949-12 A > G involving TRIM37 gene detected in patient 28. The RefSeq genes are depicted in dark blue, and the OMIM disease genes in green; the disease critical regions are indicated by black bars, the deletions by red bars, the duplications by blue bars, the deletions of previously reported SRS patients by red-white bars, and the candidate genes by red boxes. The images are a modification of a version obtained from the UCSC Genome Browser (human genome assembly GRCh37/hg19). CNVs, copy number variants; CGH, comparative genomic hybridization; OMIM, Online Mendelian Inheritance in Man; SRS, Silver–Russell syndrome; UCSC, University of California Santa Cruz. and feet with symmetric skeletal anomalies (Figure 2B), and duplex collecting system of the right kidney. school. Her height improved from −3.16 SD (age 6 years) to −1.84 SD (age 10 years) in response to growth hormone (GH) therapy, and bone age assessment at 11 years showed a correspondence between bone age and chronological age. She also presented features not belonging to the SRS spectrum, such as small hands Segregation analysis from parents to proband of D7S493 and D7S2525 microsatellites, mapping inside the 7.5-Mb deletion, showed that the deletion occurred on the paternal chromosome October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 7 Crippa et al. aCGH in Unsolved SRS Patients FIGURE 2 | Photographs of patients 7 (A) and 20 (B). (A-1) Face and whole body of patient 7 at 28 months showing proportionate short stature, mild craniofacial dysmorphisms, that is, triangular face and slight micrognathia, and flat feet. (A-2 and 3) Left hand and feet X-rays of the patient at 11 months showing (A-2) absence of the carpal bone nuclei, suggestive of a delayed bone age, fifth finger clinodactyly with hypoplastic and triangular shaped intermediate phalanx and distal phalanx hypoplasia, and (A-3) short first toes. The absence of bone nuclei at the carpus (A-2) and their presence at the tarsus (A-3) suggest a non-uniform pattern of skeletal maturation. (B-1, B-2, and B-3) Facial appearance of patient 20 at 4 years (B-1) and 10 years (B-2, B-3) showing prominent forehead and triangular face more evident in childhood. DISCUSSION Lastly, in patient 28, who presents five NH-CSS criteria except for relative macrocephaly at birth (Table 3), we identified a deletion on chromosome 17q, arr[GRCh37] 17q22(57075470_57235248) x1 mat, of 159.8 kb, involving the TRIM37, SKA2, and PRR11 genes (Table 2, Figure 1E). Mutations in the tripartite motif containing 37 (TRIM37) gene (OMIM*605073) are causative of the MULIBREY nanism (OMIM#253250), a recessive syndrome with SRS overlapping clinical features (Hämäläinen et al., 2006). Indeed, TRIM37 gene sequencing revealed a pathogenic single base change c.1949-12 A > G in intron 18 (NM_001005207) inherited from the patient’s healthy father (Figure 1E), as reported by Mozzillo et al. (2016). This study aimed to evaluate, through high-resolution aCGH, a cohort of 34 patients with clinical signs resembling SRS but negative to (epi)genetic tests. Five individuals, accounting for 15% of the starting cohort, were found to harbor pathogenic CNVs (P-CNVs), attesting a frequency higher than that shown by similar studies, ranging from 6% (Inoue et al., 2017) or 7% (Azzi et al., 2015) to 12% (Sachwitz et al., 2017). Furthermore, our detection rate increases to 21% by including two cases carrying PP-CNVs, whose exact pathogenic role awaits confirmation by further studies. Regarding the NH-CSS, three out of five patients with P-CNVs (patients 17, 20, and 28) and one (patient 18) out of two with PP-CNVs fulfill the criteria, and one patient (patient 7) with P-CNV scored positive for three clinical items, reinforcing the recommendation to process by aCGH patients with a clinical suspicion of SRS. Only one patient (patient 17) can be diagnosed clinical SRS, while the other six patients, despite matching four or more criteria, are unlikely SRS.i Potentially Pathogenic Variants (PP-CNV) Two CNVs were classified as PP variants on the basis of gene content and genomic position criteria. Figure 3 shows the genomic regions containing the PP-CNVs of patients 8 and 18, as well as the integrated molecular analyses performed to assess their pathogenic effect. Our results confirm the genetic heterogeneity found in similar studies (Tumer et al., 2018), increasing the number of structural variants identified to date in patients with SRS or similar features. The detected CNVs involve distinct genomic regions either previously associated with growth delay conditions (1q24.3-q25.3, 17p13.3, 17q22, and 22q11.2-q11.22) and with SRS spectrum (7p12.1 and 7p15.3-p14.3) or outlined for the first time (19q13.42), hence enhancing the definition of reported and novel SRS overlapping syndromes. Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients It is worth noting that patient 20 has two further deletions on proximal 7p, the first at 7p21.3 sized 81 kb and the second at 7p21.1 of 11 kb, both categorized as VOUS. p g A de novo deletion of 1.15 Mb at 22q, arr[GRCh37] 22q11.2q11.22(21808950_22963000)x1 dn, encompassing 21 RefSeq genes, was identified in patient 23 (Table 2, Figure 1D). The deletion overlaps the critical region of distal 22q11.2 microdeletion syndrome (OMIM#611867), which is associated with a variable phenotype with prenatal growth delay. The girl was born preterm, at 35 weeks of gestation, and showed SGA, a triangular face, and fifth finger clinodactyly (Table 3). Growth retardation persisted at the age of 3 years. No cardiac defects or other features frequently associated with distal 22q11.2 microdeletion were present. The molecular findings did not confirm the initial clinical suspicion of SRS in keeping with the clinical presentation and the negative clinical score (2/6 items).i Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients It is worth noting that patient 20 has two further deletions on proximal 7p, the first at 7p21.3 sized 81 kb and the second at 7p21.1 of 11 kb, both categorized as VOUS. A de novo deletion of 1.15 Mb at 22q, arr[GRCh37] 22q11.2q11.22(21808950_22963000)x1 dn, encompassing 21 RefSeq genes, was identified in patient 23 (Table 2, Figure 1D). The deletion overlaps the critical region of distal 22q11.2 microdeletion syndrome (OMIM#611867), which is associated with a variable phenotype with prenatal growth delay. The girl was born preterm, at 35 weeks of gestation, and showed SGA, a triangular face, and fifth finger clinodactyly (Table 3). Growth retardation persisted at the age of 3 years. No cardiac defects or other features frequently associated with distal 22q11.2 microdeletion were present. The molecular findings did not confirm the initial clinical suspicion of SRS in keeping with the clinical presentation and the negative clinical score (2/6 items). In patient 18, a duplication of 975 kb at 7p, arr[GRCh37] 7p12.1(50981149_51956510)x3 mat (Table 2, Figure 3B-1), was detected. The patient fulfills the NH-CSS criteria with a score of 4/5. No information about her head circumference at birth was available to establish a proper clinical SRS diagnosis. She did not show severe birth retardation, but psychomotor developmental delay and her facial dysmorphisms were not associated with SRS (Table 3). Her mother did not show any SRS features. The duplication fully involves the cordon-bleu (COBL) gene (OMIM*610317) and interestingly, its distal bkp maps 130 kb from the promoter of the growth factor receptor-bound protein 10 (GRB10) gene (OMIM*601523), which is a candidate for SRS (Wakeling et al., 2017; Miyoshi et al., 1998). Further molecular analyses revealed that the duplication is transmitted by the phenotypically healthy grandfather (Figure 3B-2). Neither GRB10 expression nor GRB10:alt-TSS-DMR methylation are impaired by the structural variant in the carrier family members (Figures 3B-3–5), with the exception of a slight hypomethylation detected in the mother compared with her daughter and controls (Figures 3B-4 and 5). (Supplementary Figure 1). In order to assess whether the insulin-like growth factor 2 mRNA binding protein 3 (IGF2BP3) gene (OMIM*608259)—involved in the deletion and potential candidate for the patient’s phenotype—might be downregulated, gene expression analysis was performed. Real-time RT-PCR analysis confirmed the hypothesis, identifying a halved amount of IGF2BP3 transcript in the patient compared with controls (Supplementary Figure 1). Genomic Imbalances by aCGH and Clinical Records of the Carrier Patients (B-4) Small hands with proximal placement and ulnar side bowing of first fingers, and clinodactyly of fifth fingers. (B-5) Left hand X-rays of the patient at 10 years showing hypoplasia of distal phalanges. (B-6) Small feet with short and broad halluces, right over-riding toes, and small/deep-set nails. (B-7) Feet X-rays of the patient at 4 years showing proximal placement of fifth metatarsals, absence of distal phalanges in second and fifth toes, absence of middle phalanges in fifth toes, and phalanges hypoplasia. FIGURE 2 | Photographs of patients 7 (A) and 20 (B). (A-1) Face and whole body of patient 7 at 28 months showing proportionate short stature, mild craniofacial dysmorphisms, that is, triangular face and slight micrognathia, and flat feet. (A-2 and 3) Left hand and feet X-rays of the patient at 11 months showing (A-2) absence of the carpal bone nuclei, suggestive of a delayed bone age, fifth finger clinodactyly with hypoplastic and triangular shaped intermediate phalanx and distal phalanx hypoplasia, and (A-3) short first toes. The absence of bone nuclei at the carpus (A-2) and their presence at the tarsus (A-3) suggest a non-uniform pattern of skeletal maturation. (B-1, B-2, and B-3) Facial appearance of patient 20 at 4 years (B-1) and 10 years (B-2, B-3) showing prominent forehead and triangular face more evident in childhood. (B-4) Small hands with proximal placement and ulnar side bowing of first fingers, and clinodactyly of fifth fingers. (B-5) Left hand X-rays of the patient at 10 years showing hypoplasia of distal phalanges. (B-6) Small feet with short and broad halluces, right over-riding toes, and small/deep-set nails. (B-7) Feet X-rays of the patient at 4 years showing proximal placement of fifth metatarsals, absence of distal phalanges in second and fifth toes, absence of middle phalanges in fifth toes, and phalanges hypoplasia. October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 8 aCGH in Unsolved SRS Patients Crippa et al. (Supplementary Figure 1). In order to assess whether the insulin-like growth factor 2 mRNA binding protein 3 (IGF2BP3) gene (OMIM*608259)—involved in the deletion and potential candidate for the patient’s phenotype—might be downregulated, gene expression analysis was performed. Real-time RT-PCR analysis confirmed the hypothesis, identifying a halved amount of IGF2BP3 transcript in the patient compared with controls (Supplementary Figure 1). DISCUSSION gf Specifically, in patient 8, the aCGH analysis showed a duplication of 445 kb in 19q, arr[GRCh37] 19q13.42(54039784_54484439) x3 mat (Table 2, Figure 3A-1). The patient’s clinical data were insufficient for NH-CSS evaluation. She showed facial dymorphisms reminiscent of SRS and growth retardation during early childhood (Table 3). The proximal duplication breakpoint (bkp) lies within the zinc finger protein 331 (ZNF331) gene (OMIM*606043) (isoform NM_018555), an imprinted gene yet unrelated to SRS. qPCR experiments clarified that the ZNF331 gene and its two maternally methylated DMRs (ZNF331:alt-TSS- DMR1 and ZNF331:alt-TSS-DMR2) are fully duplicated (Figure 3A-2), and consistently, pyrosequencing analysis of DMR2 showed an increase of methylation in the patient and her mother as compared with controls (Figure 3A-3). Patient 7 carries a P-CNV associated with 1q24q25 microdeletion syndrome (Figure 1A) and his phenotype, matching that of the described patients (Burkardt et al., 2011; Ashraf et al., 2015), recapitulates the SRS phenotype in terms of a few minor dysmorphisms and growth retardation. Three “key” genes for growth delay map within the deletion: CENPL, October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 9 aCGH in Unsolved SRS Patients Crippa et al. FIGURE 3 | Physical map of the genomic regions and molecular characterization of the potentially pathogenic CNVs identified by array CGH. The 19q13.42 (A-1) an 7p12.1 (B-1) regions showing in blue the duplication identified in patients 8 (A-1) and 18 (B-1). The RefSeq genes are depicted in dark blue, and the OMIM disease genes in green; the candidate genes are illustrated by red boxes, and the differentially methylated regions (DMRs) in black. The positions of pyrosequencing assays and qPCR probes used in the CNVs’ molecular characterization are reported in black and blue, respectively. The images are a modification of a version obtained from the UCSC Genome Browser (human genome assembly GRCh37/hg19). (A-2) Copy number analysis by using qPCR probes specific for ZNF331 gene promoter confirmed the entire duplication of ZNF331 gene in both patient 8 and her mother (M). (A-3) Pyrosequencing analysis performed on blood DNA of the ZNF331:alt- TSS-DMR2 showed an hypermethylation level in both patient 8 and her mother (M) compared with healthy controls. (B-2) Patient 18’s family tree showing duplication transmission. (B-3) Relative expression of GRB10 blood mRNA in patient 18 and her mother (M), compared with healthy controls. DISCUSSION Data were normalized against GAPDH as housekeeping gene; similar results were obtained using TBP, HMBS, and RPLP0 as normalizer (data not shown). (B-4, B-5) Pyrosequencing analysis of the GRB10:alt-TSS-DMR performed on blood DNA showed a methylation level in patient 18 and her maternal grandfather (MGF) compared with healthy controls with both assays, while in patient 18’s mother (M), a hypomethylation level was detected with both assays. CNVs, copy number variants; CGH, comparative genomic hybridization; OMIM, Online Mendelian Inheritance in Man; qPCR, quantitative polymerase chain reaction; UCSC, University of California Santa Cruz. FIGURE 3 | Physical map of the genomic regions and molecular characterization of the potentially pathogenic CNVs identified by array CGH. The 19q13.42 (A-1) an 7p12.1 (B-1) regions showing in blue the duplication identified in patients 8 (A-1) and 18 (B-1). The RefSeq genes are depicted in dark blue, and the OMIM disease genes in green; the candidate genes are illustrated by red boxes, and the differentially methylated regions (DMRs) in black. The positions of pyrosequencing assays and qPCR probes used in the CNVs’ molecular characterization are reported in black and blue, respectively. The images are a modification of a version obtained from he UCSC Genome Browser (human genome assembly GRCh37/hg19). (A-2) Copy number analysis by using qPCR probes specific for ZNF331 gene promoter confirmed the entire duplication of ZNF331 gene in both patient 8 and her mother (M). (A-3) Pyrosequencing analysis performed on blood DNA of the ZNF331:alt- TSS-DMR2 showed an hypermethylation level in both patient 8 and her mother (M) compared with healthy controls. (B-2) Patient 18’s family tree showing duplication ransmission. (B-3) Relative expression of GRB10 blood mRNA in patient 18 and her mother (M), compared with healthy controls. Data were normalized against GAPDH h k i i il l b i d i TBP HMBS d RPLP0 li (d h ) (B 4 B 5) P i l i FIGURE 3 | Physical map of the genomic regions and molecular characterization of the potentially pathogenic CNVs identified by array CGH. The 19q13.42 (A-1) and 7p12.1 (B-1) regions showing in blue the duplication identified in patients 8 (A-1) and 18 (B-1). The RefSeq genes are depicted in dark blue, and the OMIM disease genes in green; the candidate genes are illustrated by red boxes, and the differentially methylated regions (DMRs) in black. DISCUSSION The positions of pyrosequencing assays and qPCR probes used in the CNVs’ molecular characterization are reported in black and blue, respectively. The images are a modification of a version obtained from the UCSC Genome Browser (human genome assembly GRCh37/hg19). (A-2) Copy number analysis by using qPCR probes specific for ZNF331 gene promoter confirmed the entire duplication of ZNF331 gene in both patient 8 and her mother (M). (A-3) Pyrosequencing analysis performed on blood DNA of the ZNF331:alt- TSS-DMR2 showed an hypermethylation level in both patient 8 and her mother (M) compared with healthy controls. (B-2) Patient 18’s family tree showing duplication transmission. (B-3) Relative expression of GRB10 blood mRNA in patient 18 and her mother (M), compared with healthy controls. Data were normalized against GAPDH as housekeeping gene; similar results were obtained using TBP, HMBS, and RPLP0 as normalizer (data not shown). (B-4, B-5) Pyrosequencing analysis of the GRB10:alt-TSS-DMR performed on blood DNA showed a methylation level in patient 18 and her maternal grandfather (MGF) compared with healthy controls with both assays, while in patient 18’s mother (M), a hypomethylation level was detected with both assays. CNVs, copy number variants; CGH, comparative genomic hybridization; OMIM, Online Mendelian Inheritance in Man; qPCR, quantitative polymerase chain reaction; UCSC, University of California Santa Cruz. The duplication at 17p13.3 of the clinically SRS patient 17 includes only five genes and to the best of our knowledge is the smallest reported to date. Two non-overlapping deletions of the same region have been identified in clinically suspected SRS patients (Spengler et al., 2012). Microduplications at 17p13.3, involving the YWHAE and CRK genes, similarly to our patient, are associated with macrosomia in the majority of cases (Henry et al., 2016). Conversely, patient 17 displays growth retardation and does not show the centromere protein L (OMIM*611503), involved in kinetochore function and mitotic progression (Burkardt et al., 2011); PAPPA2, pappalysin 2, encoding a protease for IGFBP3 and IGFBP5 involved in IGF-1 availability (Dauber et al., 2016), whose biallelic mutations account for short stature (Dauber et al., 2016); and LHX4, LIM homeobox 4 (OMIM*602146) causative of the dominant pituitary hormone deficiency-4 syndrome (CPHD4; OMIM*262700). October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 10 aCGH in Unsolved SRS Patients Crippa et al. 2006), while the adjacent cluster of 46 miRNAs is paternally expressed in the placenta (Noguer-Dance et al., 2010). DISCUSSION Recently, a paternally inherited duplication of 1.06 Mb, fully involving ZNF331 and C19MC, was stated to be pathogenic in a fetus with IUGR (Petre et al., 2018) that in our opinion might be ascribed to the duplication of the paternally expressed C19MC cluster. The fetus’s father, in turn, inherited the duplication from his mother mirroring, our patient 8 and her mother, but his phenotype is not described. ZNF331 is worth considering as a candidate gene for growth disorders, though further studies are needed to elucidate the pathomechanism.i microduplication-associated neurologic dysfunctions, confirming the incomplete penetrance and highly variable expressivity of this duplication syndrome. Among the duplicated genes playing a role in cellular growth pathways, CRK (OMIM*164762) has been proposed as a candidate for body size determination (Østergaard et al., 2012). Patient 23 was found to carry the distal 22q11.2 microdeletion syndrome (OMIM#611867), a recurrent rearrangement associated with prenatal and postnatal growth retardation (Ben- Shachar et al., 2008). She is unlikely SRS, though the same 1.1 Mb deletion has been reported in an SRS patient matching five NH-CSS criteria (Bruce et al., 2010), and an atypical distal 22q11.2 microdeletion has been described in a patient with clinical features reminiscent of SRS (Garavelli et al., 2011). The main candidate genes of this syndrome are UBE2L3, ubiquitin conjugating enzyme E2 L3 (OMIM*603721), whose homozygous mutations in mice lead to placental defects and growth retardation (Harbers et al., 1996), and MAPK1, mitogen-activated protein kinase 1 (OMIM*176948), encoding a protein kinase with an essential role in embryonic development (Saba-El-Leil et al., 2003). p A second maternal duplication at 7p12 was identified and categorized as PP-CNV in case 18 presenting a positive clinical score of 4 out of 5 (relative macrocephaly at birth is unknown). The CNV maps about 130 kb from the promoter of GRB10 (Figure 3B-1), an imprinted gene with biallelic expression in many tissues, but with maternal expression in muscle and placenta (Blagitko et al., 2000; Monk et al., 2009). Genomic imbalances affecting GRB10, in particular, maternal duplications involving GRB10, lead to SRS- like features and growth retardation (Tumer et al., 2018; Yuan et al., 2016). Since the duplication is inherited from the healthy maternal grandfather via the healthy mother, an imprinting transmission pattern might be envisaged. DISCUSSION Even if concordant results from both expression and methylation analyses on family member blood cells did not support this hypothesis, the pathogenic role of the duplication remains challenging as a perturbed gene regulation cannot be ruled out during the fetal development, in particular in tissues where the gene is only maternally expressed. y p ( ) Patient 28 was diagnosed to be affected by MULIBREY nanism (OMIM#253250), as he carries a 17q22 chromosome deletion and a concomitant TRIM37 splicing mutation affecting the non-deleted chromosome, as already described (Mozzillo et al., 2016), reconfirming the clinical overlap between SRS and MULIBREY nanism (Hämäläinen et al., 2006; Jobic et al., 2017). Finally, patient 20 showed a de novo deletion of 7.5 Mb on the paternal chromosome 7 at p15.3-14.3, proximal to the candidate 7p12 SRS region. She fulfills the NH-CSS except for body asymmetry, less common in upd(7)mat than in 11p15 LOM molecularly confirmed SRS patients (Wakeling et al., 2017), and relative macrocephaly at birth, although displayed in early childhood. Out of the numerous deleted genes that may account for SRS similar features, IGF2BP3 is the most relevant, as it encodes for an RNA- binding factor specific to the 5′UTR of IGF2 mRNA, regulating the IGF2 transcript levels during fetal growth and development (Monk et al., 2002). A similar IGF2BP3 deletion was found in an SRS patient with classical SRS phenotype, supporting the relevance of this locus in SRS etiology (Lin et al., 2010). Another interesting disease gene is HOXA13, homeobox A13 (OMIM*142959), causative of the hand-foot-genital syndrome (HFGS, OMIM#140000), which is characterized by small feet with unusually short and big toes and abnormal thumbs, and urogenital malformations. This gene is also involved in heterogeneous chromosomal aberrations of the 7p15- p14 region, and like the reported cases, our patient shows feet and kidney anomalies (Devriendt et al., 1999; Dunø et al., 2004; Kosaki et al., 2005; Fryssira et al., 2011; Jun et al., 2011; Hosoki et al., 2012; Pezzani et al., 2015). In conclusion, we found five clinically relevant CNVs and two VOUS—PP-CNVs in distinct chromosomal regions in 15% and 6% of clinically suspected SRS patients, respectively. Out of these variants, two, harbored by patients matching the NH-CSS criteria, map to 7p distinct regions, reinforcing the contribution of the short arm of chromosome 7 to SRS-like etiology. DISCUSSION Further investigations on a larger cohort of SRS/SRS-like patients are needed, firstly, to confirm the observed relevant CNVs detection rate, to exclude coincidental findings unrelated to the phenotype of the reported patients, and to support the role of the appointed genes in SRS and similar conditions. Genome-wide aCGH scan integrated by in-depth studies is re-confirmed as an appropriate and powerful tool for achieving a differential diagnosis in both clinically and unlikely SRS diagnosed patients and to highlight novel chromosome alterations and genomic regions associated with SRS and similar growth retardation disorders. DATA AVAILABILITY STATEMENT The datasets generated for this study can be found in ClinVar database (http://www.ncbi.nlm.nih.gov/clinvar/), accession numbers SCV000863548 to SCV000863554. Regarding PP-CNVs, the maternal duplication at 19q13.42, identified in patient 8, comprises the chromosome 19 microRNA cluster (C19MC) and the ZNF331 gene, whose ZNF331:alt-TSS- DMR2 was found to be hypermethylated in both patient and mother. ZNF331, encoding a zinc finger protein that acts as a tumor suppressor and negative regulator of cellular growth (Yu et al., 2013; Wang et al., 2013), was found to be differentially expressed between normal and IUGR placentas (Diplas et al., 2009). Up to now, data on maternal or paternal ZNF331 expression are scarce and contradictory (Lambertini et al., 2012; Daelemans et al., 2010; Pollard et al., 2008; Pant et al, Frontiers in Genetics  |  www.frontiersin.org REFERENCES Devriendt, K., Jaeken, J., Matthijs, G., Van Esch, H., Debeer, P., Gewillig, M., et al. (1999). Haploinsufficiency of the HOXA gene cluster, in a patient with hand- foot-genital syndrome, velopharyngeal insufficiency, and persistent patent ductus botalli. Am. J. Hum. Genet. 65, 249–251. doi: 10.1086/302452 Ashraf, T., Collinson, M. N., Fairhurst, J., Wang, R., Wilson, L. C., and Foulds, N. (2015). Two further patients with the 1q24 deletion syndrome expand the phenotype: a possible role for the miR199-214 cluster in the skeletal features of the condition. Am. J. Med. Genet. A 167A, 3153–3160. doi: 10.1002/ajmg.a.37336h Diplas, A. I., Lambertini, L., Lee, M. J., Sperling, R., Lee, Y. L., Wetmur, J., et al.. (2009). Differential expression of imprinted genes in normal and IUGR human placentas. Epigenetics 4, 235–240. doi: 10.4161/epi.9019f Dunø, M., Hove, H., Kirchhoff, M., Devriendt, K., and Schwartz, M. (2004). Mapping genomic deletions down to the base: a quantitative copy number scanning approach used to characterise and clone the breakpoints of a recurrent 7p14.2p15.3 deletion. Hum. Genet. 115, 459–467. doi: 10.1007/ s00439-004-1174-y Azzi, S., Salem, J., Thibaud, N., Chantot-Bastaraud, S., Lieber, E., Netchine,  I., et al. (2015). A prospective study validating a clinical scoring system and demonstrating phenotypical–genotypical correlations in Silver–Russell syndrome. J. Med. Genet. 52, 446–453. doi: 10.1136/jmedgenet-2014-102979 Ben-Shachar, S., Ou, Z., Shaw, C. A., Belmont, J. W., Patel, M. S., Hummel, M., et al.. (2008). 22q11.2 distal deletion: a recurrent genomic disorder distinct from DiGeorge syndrome and velocardiofacial syndrome. Am. J. Hum. Genet. 82, 214–221. doi: 10.1016/j.ajhg.2007.09.014 Eppig, J. T., Blake, J. A., Bult, C. J., Kadin, J. A., and Richardson, J. E.Mouse Genome Database Group (2012) The Mouse Genome Database (MGD): comprehensive resource for genetics and genomics of the laboratory mouse. Nucleic Acids Res. 40, D881–D886. doi: 10.1093/nar/gkr974 Blagitko, N., Mergenthaler, S., Schulz, U., Wollmann, H. A., Craigen, W., Eggermann, T., et al. (2000). Human GRB10 is imprinted and expressed from the paternal and maternal allele in a highly tissue- and isoform-specific fashion. Hum. Mol. Genet. 9, 1587–1595. doi: 10.1093/hmg/9.11.1587 Fryssira, H., Makrythanasis, P., Kattamis, A., Stokidis, K., Menten, B., Kosaki, K., et al. (2011). Severe developmental delay in a patient with 7p21.1-p14.3 microdeletion spanning the TWIST gene and the HOXA gene cluster. Mol .Syndromol. 2, 45–49. doi: 10.1159/000334313 Hum. Mol. Genet. 9, 1587–1595. doi: 10.1093/hmg/9.11.1587 Bragin, E., Chatzimichali, E. A., Wright, C. F., Hurles, M. E., Firth, H. V., Bevan, A. P., et al. (2014). ETHICS STATEMENT The study was approved by the ethical committee of Istituto Auxologico Italiano and written informed consent was obtained from the parents for publication of this study and any accompanying images. October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 11 aCGH in Unsolved SRS Patients Crippa et al. REFERENCES DECIPHER: database for the interpretation of phenotype- linked plausibly pathogenic sequence and copy-number variation. Nucleic Acids Res. 42, D993–D1000. doi: 10.1093/nar/gkt937 Fuke, T., Mizuno, S., Nagai, T., Hasegawa, T., Horikawa, R., Miyoshi, Y., et al. (2013). Molecular and clinical studies in 138 Japanese patients with Silver– Russell syndrome. PLoS One 8, e60105. doi: 10.1371/journal.pone.0060105 Garavelli, L., Rosato, S., Wischmeijer, A., Gelmini, C., Esposito, A., Mazzanti, L., et al. (2011). 22q11.2 distal deletion syndrome: description of a new case with truncus arteriosus type 2 and review. Mol. Syndromol. 2, 35–44. doi: 10.1159/000334262 Bruce, S., Hannula-Jouppi, K., Puoskari, M., Fransson, I., Simola, K. O., Lipsanen- Nyman, M., et al. (2010). Submicroscopic genomic alterations in Silver–Russell syndrome and Silver–Russell-like patients. J. Med. Genet. 47, 816–822. doi: 10.1136/jmg.2009.069427 Geoffron, S., Abi Habib, W., Chantot-Bastaraud, S., Dubern, B., Steunou, V., Azzi, S., et  al. (2018). Chromosome 14q32.2 imprinted region disruption as an alternative molecular diagnosis of Silver–Russell syndrome. J. Clin. Endocrinol. Metab. 103, 2436–2446 .doi: 10.1210/jc.2017-02152 Burkardt, D. D., Rosenfeld, J. A., Helgeson, M. L., Angle, B., Banks, V., Smith, W. E., et al. (2011). Distinctive phenotype in 9 patients with deletion of chromosome 1q24- q25. Am. J. Med. Genet. A 155A, 1336–1351. doi: 10.1002/ajmg.a.34049 Cacciari, E., Milani, S., Balsamo, A., Spada, E., Bona, G., Cavallo, L., et al. (2006). Italian cross-sectional growth charts for height, weight and BMI (2 to 20 yr). J. Endocrinol. Invest. 29, 581–593. doi: 10.1007/BF03344156 Hämäläinen, R. H., Mowat, D., Gabbett, M. T., O'brien, T. A., Kallijärvi, J., and Lehesjoki, A. E. (2006). Wilms' tumor and novel TRIM37 mutations in an Australian patient with MULIBREY nanism. Clin. Genet. 70, 473–479. doi: 10.1111/j.1399-0004.2006.00700.x Daelemans, C., Ritchie, M. E., Smits, G., Abu-Amero, S., Sudbery, I. M., Forrest, M. S., et al. (2010). High-throughput analysis of candidate imprinted genes and allele-specific gene expression in the human term placenta. BMC Genet. 11, 25. doi: 10.1186/1471-2156-11-25 Harbers, K., Müller, U., Grams, A., Li, E., Jaenisch, R., and Franz, T. (1996). Provirus integration into a gene encoding a ubiquitin-conjugating enzyme results in a placental defect and embryonic lethality. Proc. Natl. Acad. Sci. U. S. A. 93, 12412–12417. doi: 10.1073/pnas.93.22.12412 Dauber, A., Muñoz-Calvo, M. T., Barrios, V., Domené, H. M., Kloverpris, S., Serra- Juhé, C., et al. (2016). Mutations in pregnancy-associated plasma protein A2 cause short stature due to low IGF-I availability. EMBO Mol. Med. 8, 363–374. doi: 10.15252/emmm.201506106 Henry, R. K., Astbury, C., Stratakis, C. ACKNOWLEDGMENTS The authors thank patients and families for their cooperation in this project. The authors thank Patrizia Colapietro for her technical contribution to this work. SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fgene.2019.00955/ full#supplementary-material All authors read and approved the manuscript for submission. All authors read and approved the manuscript for submission. All authors read and approved the manuscript for submission. FUNDING MC: study design, analysis, and interpretation of clinical and experimental data, and manuscript preparation; MB: clinical data collection and interpretation; LC: performed preliminary diagnostic methylation-specific multiplex ligation-dependent probe amplification (MS-MLPA) analysis and collection of clinical data; CP: analysis and interpretation of data from aCGH; CG: performed pyrosequencing analysis; ASi: performed qPCR analysis; IB: analysis and interpretation of data from aCGH and CNVs submission to ClinVar; SG: performed preliminary diagnostic MS-MLPA analysis; ASe, AM, and SB: clinical data collection; AR: critical revision of the manuscript; GF: clinical data collection; SR: clinical data collection and critical revision of the manuscript; LL: manuscript preparation and critical revision of the manuscript; PF: study design, interpretation of experimental data, manuscript preparation, and critical revision of the manuscript. This study was partially supported by the Italian Ministry of the University and Research PRIN grants 2009 and 2015 (Prot. 2009MBHZPR_004 and Prot. 2015JHLY35) to PF and by Italian Ministry of Health grant “Ricerca Corrente” to IRCCS Istituto Auxologico Italiano (PF 08C301-2013). REFERENCES Temple syndrome: comprehensive molecular and clinical findings in 32 Japanese patients. Genet. Med. 19, 1356–1366. doi: 10.1038/gim.2017.53h Saba-El-Leil, M. K., Vella, F. D., Vernay, B., Voisin, L., Chen, L., Labrecque, N., et al. (2003). An essential function of the mitogen-activated protein kinase Erk2 in mouse trophoblast development. EMBO Rep. 4, 964–968. doi: 10.1038/ sj.embor.embor939 Kearney, H. M., Thorland, E. C., Brown, K. K., Quintero-Rivera, F., and South, S. T.Working Group of the American College of Medical Genetics Laboratory Quality Assurance Committee. (2011). American College of Medical Genetics standards and guidelines for interpretation and reporting of postnatal constitutional copy number variants. Genet. Med. 13, 680–685. doi: 10.1097/ GIM.0b013e3182217a3a Sachwitz, J., Meyer, R., Fekete, G., Spranger, S., Matulevičienė, A., Kučinskas, V., et al. (2017). NSD1 duplication in Silver–Russell syndrome (SRS): molecular karyotyping in patients with SRS features. Clin. Genet. 91, 73–78. doi: 10.1111/ cge.12803 Kosaki, R., Higuchi, M., Mitsui, N., Matsushima, K., Ohashi, H., and Kosaki, K. (2005). Deletion involving the TWIST locus and the HOXA cluster: a contiguous gene syndrome on 7p? Congenit. Anom. (Kyoto) 45, 35–38. doi: 10.1111/j.1741-4520.2005.00059.x Sheng, H. Z., Zhadanov, A. B., Mosinger, B. Jr, Fujii, T., Bertuzzi, S., Grinberg, A., et al. (1996). Specification of pituitary cell lineages by the LIM homeobox gene Lhx3. Science 272, 1004–1007. doi: 10.1126/science.272.5264.1004 Spengler, S., Begemann, M., Ortiz Brüchle, N., Baudis, M., Denecke, B., Kroisel, P. M., et al. (2012). Molecular karyotyping as a relevant diagnostic tool in children with growth retardation with Silver–Russell features. J. Pediatr. 161, 933–942. doi: 10.1016/j.jpeds.2012.04.045 Lambertini, L., Marsit, C. J., Sharma, P., Maccani, M., Ma, Y., Hu, J., et al. (2012). Imprinted gene expression in fetal growth and development. Placenta 33, 480– 486. doi: 10.1016/j.placenta.2012.03.001 Lin, S. Y., Lee, C. N., Hung, C. C., Tsai, W. Y., Lin, S. P., Li, N. C., et al. (2010). Epigenetic profiling of the H19 differentially methylated region and comprehensive whole genome array-based analysis in Silver–Russell syndrome. Am. J. Med. Genet. A 152A, 2521–2528. doi: 10.1002/ajmg.a.33629 Tümer, Z., López-Hernández, J. A., Netchine, I., Elbracht, M., Grønskov, K., Gede, L. B., et al.. (2018). Structural and sequence variants in patients with Silver–Russell syndrome or similar features-Curation of a disease database. Hum. Mutat. 39, 345–364. doi: 10.1002/humu.23382 Wakeling, E. L., Brioude, F., Lokulo-Sodipe, O., O'Connell, S. M., Salem, J., Bliek, J., et al. (2017). Diagnosis and management of Silver–Russell syndrome: first international consensus statement. Nat. Rev. Endocrinol. 13, 105–124. REFERENCES doi: 10.1038/nrendo.2016.138 Livak, K. J., and Schmittgen, T. D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(–Delta Delta C(T)) Method. Methods 25, 402–408. doi: 10.1006/meth.2001.1262 Meyer, R., Soellner, L., Begemann, M., Dicks, S., Fekete, G., Rahner, N., et al. (2017). Targeted next generation sequencing approach in patients referred for Silver–Russell syndrome testing increases the mutation detection rate and provides decisive information for clinical management. J. Pediatr. 187, 206– 212. doi: 10.1016/j.jpeds.2017.04.018 Wang, H., Chen, G., Wang, H., and Liu, C. (2013). RITA inhibits growth of human hepatocellular carcinoma through induction of apoptosis. Oncol. Res. 20, 437– 445. doi: 10.1038/onc.2012.54 Wechsler, D. (2002). Wechsler preschool and primary scale of intelligence – third edition. San Antonio, TX: The Psychological Corporation.i Miller, D. T., Adam, M. P., Aradhya, S., Biesecker, L. G., Brothman, A. R., Carter, N. P., et al. (2010). Consensus statement: chromosomal microarray is a first-tier clinical diagnostic test for individuals with developmental disabilities or congenital anomalies. Am. J. Hum. Genet. 86, 749–764. doi: 10.1016/j.ajhg.2010.04.006 Yu, J., Liang, Q. Y., Wang, J., Cheng, Y., Wang, S., Poon, T. C., et al. (2013). Zinc-finger protein 331, a novel putative tumor suppressor, suppresses growth and invasiveness of gastric cancer. Oncogene 32, 307–317. doi: 10.1038/onc.2012.54 Yuan, H., Huang, L., Hu, X., Li, Q., Sun, X., Xie, Y., et al. (2016). FGFR3 gene mutation plus GRB10 gene duplication in a patient with achondroplasia plus growth delay with prenatal onset. Orphanet J. Rare Dis. 11, 89. doi: 10.1186/ s13023-016-0465-4 Miyoshi, N., Kuroiwa, Y., Kohda, T., Shitara, H., Yonekawa, H., Kawabe, T., et al. (1998). Identification of the Meg1/Grb10 imprinted gene on mouse proximal chromosome 11, a candidate for the Silver-Russell syndrome. Gene. Proc. Natl. Acad. Sci. USA 3, 1102–1107. doi: 10.1073/pnas.95.3.1102 Monk, D., Arnaud, P., Frost, J., Hills, F. A., Stanier, P., Feil, R., et al. (2009). Reciprocal imprinting of human GRB10 in placental trophoblast and brain: evolutionary conservation of reversed allelic expression. Hum. Mol. Genet. 18, 3066–3074. doi: 10.1093/hmg/ddp248 Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer TE declared a past co-authorship with several of the authors AR, LC, SG, AS, AM, SR to the handling editor. Monk, D., Bentley, L., Beechey, C., Hitchins, M., Peters, J., Preece, M. A., et al. (2002). REFERENCES A., and Hickey, S. E. (2016). 17p13.3 microduplication including CRK leads to overgrowth and elevated growth October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 12 Crippa et al. aCGH in Unsolved SRS Patients factors: a case report. Eur. J. Med. Genet. 59, 512–516. doi: 10.1016/j. ejmg.2016.09.006 factors: a case report. Eur. J. Med. Genet. 59, 512–516. doi: 10.1016/j. ejmg.2016.09.006 Østergaard, J. R., Graakjær, J., Brandt, C., and Birkebæk, N. H. (2012). Further delineation of 17p13.3 microdeletion involving CRK. The effect of growth hormone treatment. Eur. J. Med. Genet. 55, 22–26. doi: 10.1016/j. ejmg.2011.09.004 Hosoki, K., Ohta, T., Fujita, K., Nishigaki, S., Shiomi, M., Niikawa, N., et al.. (2012). Hand-foot-genital syndrome with a 7p15 deletion: clinically recognizable syndrome. Pediatr. Int. 54, e22–e25. doi: 10.1111/j.1442-200X.2011.03550.x Pant, P. V., Tao, H., Beilharz, E. J., Ballinger, D. G., Cox, D. R., and Frazer, K. A. (2006). Analysis of allelic differential expression in human white blood cells. Genome Res. 16, 331–339. doi: 10.1101/gr.4559106f Inoue, T., Nakamura, A., Fuke, T., Yamazawa, K., Sano, S., Matsubara, K., et al. (2017). Genetic heterogeneity of patients with suspected Silver–Russell syndrome: genome-wide copy number analysis in 82 patients without imprinting defects. Clin. Epigenet. 9, 52–62. doi: 10.1186/s13148-017-0350-6 Petre, G., Lorès, P., Sartelet, H., Truffot, A., Poreau, B., Brandeis, S., et al. (2018). Genomic duplication in the 19q13.42 imprinted region identified as a new genetic cause of intrauterine growth restriction. Clin. Genet. 94, 575–580. doi: 10.1111/cge.13449 Jobic, F., Morin, G., Vincent-Delorme, C., Cadet, E., Cabry, R., Mathieu-Dramard, M., et al.. (2017). New intragenic rearrangements in non-Finnish MULIBREY nanism. Am. J. Med. Genet. A 173, 2782–2788. doi: 10.1002/ajmg.a.38381 Pezzani, L., Milani, D., Manzoni, F., Baccarin, M., Silipigni, R., Guerneri, S., et al. (2015). HOXA genes cluster: clinical implications of the smallest deletion. Ital. J. Pediatr. 41, 31–37. doi: 10.1186/s13052-015-0137-3 Jun, K. R., Seo, E. J., Lee, J. O., Yoo, H. W., Park, I. S., and Yoon, H. K. (2011). Molecular cytogenetic and clinical characterization of a patient with a 5.6-Mb deletion in 7p15 including HOXA cluster. Am. J. Med. Genet. A 155A, 642–647. doi: 10.1002/ajmg.a.33860 Pollard, K. S., Serre, D., Wang, X., Tao, H., Grundberg, E., Hudson, T. J., et al. (2008). A genome-wide approach to identifying novel-imprinted genes. Hum. Genet. 122, 625–634. doi: 10.1007/s00439-007-0440-1 Kagami, M., Nagasaki, K., Kosaki, R., Horikawa, R., Naiki, Y., Saitoh, S., et al. (2017). REFERENCES Characterisation of the growth regulating gene IMP3, a candidate for Silver–Russell syndrome. J. Med. Genet. 39, 575–581. doi: 10.1136/jmg.39.8.575 Copyright © 2019 Crippa, Bonati, Calzari, Picinelli, Gervasini, Sironi, Bestetti, Guzzetti, Bellone, Selicorni, Mussa, Riccio, Ferrero, Russo, Larizza and Finelli. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Mozzillo, E., Cozzolino, C., Genesio, R., Melis, D., Frisso, G., Orrico, A., et al. (2016). MULIBREY nanism: two novel mutations in a child identified by array CGH and DNA sequencing. Am. J. Med. Genet. A 170, 2196–2199. doi: 10.1002/ajmg.a.37770 Noguer-Dance, M., Abu-Amero, S., Al-Khtib, M., Lefèvre, A., Coullin, P., Moore, G. E., et al. (2010). The primate-specific microRNA gene cluster (C19MC) is imprinted in the placenta. J. Hum. Mol. Genet. 19, 3566–3582. doi: 10.1093/hmg/ddq272 October 2019  |  Volume 10  |  Article 955 Frontiers in Genetics  |  www.frontiersin.org 13
https://openalex.org/W4376626901
https://petsymposium.org/popets/2023/popets-2023-0089.pdf
English
null
Differential Privacy for Black-Box Statistical Analyses
Proceedings on Privacy Enhancing Technologies
2,023
cc-by
14,008
1 INTRODUCTION Consider a scenario involving a holder of personal data and a re- searcher. The researcher has written an analysis script and would like to apply it to the data in order to gain insight. The data holder is concerned with the privacy of individuals in the data, and may not necessarily trust the researcher. Furthermore, the data holder may have contractual obligations that prevent them from sharing information about individuals in the data. The researcher gives the data holder their analysis script, but the data holder does not have the resources or expertise to analyze the code for privacy threats. Instead, the data holder desires a simple way to add privacy guarantees to the researcher’s script, without looking inside it, so that the output can be safely returned. This motivates the central question of this study: A malicious researcher has a variety of techniques to extract private information from a dataset. Suppose a researcher is looking for information about a target individual, 𝑡. The researcher could write a script that hides𝑡’s income in the trailing digits of the output. They could code the script so the sum of the digits in the output is even if 𝑡is in the data, and odd otherwise. They could make the script deliberately fail if a certain medical diagnosis appears in 𝑡’s records. Providing further challenges, a malicious researcher may obfuscate their code, making it difficult to determine its true purpose. Obfuscation may be performed by hand, or a variety of automated obfuscation tools may be used by researchers without special expertise [2, 20]. What privacy guarantees can be added to a statistical analysis script from an untrusted source, treating it in a black box fashion? What privacy guarantees can be added to a statistical analysis script from an untrusted source, treating it in a black box fashion? Although many organizations are interested in increasing re- searcher access to data, the costs given current technology can be considerable. For example, the US Census Bureau has a program to allow researchers to interact with raw data. To maintain privacy, however, the Bureau maintains a set of secure locations around the country and puts all applicants through a background check.1 In another example, Facebook and Social Science One teamed up to Following common practice, the data holder might hope to ap- ply differential privacy by adding noise related to the researcher script’s sensitivity [11]. 2See Data & Security of Social Science One, available at https://socialscience.one/data- security-privacy. 1.1 Considerations for Untrusted Code Differential Privacy, Untrusted Code, Black Box, Statistics Running code from untrusted sources introduces special challenges to the data holder. To borrow terms from the security literature, some researchers that submit scripts may be honest, meaning they abide by the intended use of the system. Others may be considered malicious, meaning that their true goal is to extract private infor- mation from the data. A lack of trust suggests that all researchers must be considered potentially malicious. ABSTRACT make data about shared URLs available to researchers using differ- ential privacy [18]. However, to work with the most sensitive data, researchers must be pre-approved by a university review board, pass an application process, and undergo monitoring - even at the level of individual keystrokes.2 make data about shared URLs available to researchers using differ- ential privacy [18]. However, to work with the most sensitive data, researchers must be pre-approved by a university review board, pass an application process, and undergo monitoring - even at the level of individual keystrokes.2 We formalize a notion of a privacy wrapper, defined as an algorithm that can take an arbitrary and untrusted script and produce an out- put with differential privacy guarantees. Our novel privacy wrapper, named TAHOE, incorporates two design ideas: a type of stability under subsetting, and randomization over subset size. We show that TAHOE imposes differential privacy for every possible script. When the data alphabet is finite and small enough, TAHOE can be practically run on a single computer. Performance simulations show that TAHOE has greater accuracy than a benchmark algo- rithm based on a subsample-and-aggregate approach for certain scenarios and parameter values. In contrast with existing approaches, the goal of this study is to enable a researcher to work with private data in an automatic fashion, without the need for screening procedures to establish trust. We will refer to a system that achieves this as a privacy wrapper. Akin to a function wrapper in programming, the idea is to write an algorithm that mediates all interaction with the researcher’s script, producing output that is based on the behavior of the script, while also yielding strong privacy guarantees. Differential Privacy for Black-Box Statistical Analyses Nitin Kohli UC Berkeley Center for Effective Global Action Berkeley, CA, USA nitin.kohli@berkeley.edu Paul Laskowski UC Berkeley School of Information Berkeley, CA, USA paul@ischool.berkeley.edu Nitin Kohli UC Berkeley Center for Effective Global Action Berkeley, CA, USA nitin.kohli@berkeley.edu 1For information on US Census’ Secure Research Environment, see https://www.census. gov/about/adrm/fsrdc/about/secure_rdc.html. 1For information on US Census’ Secure Research Environment, see https://www.census. gov/about/adrm/fsrdc/about/secure_rdc.html. This work is licensed under the Creative Commons Attribu- tion 4.0 International License. To view a copy of this license visit https://creativecommons.org/licenses/by/4.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA. Proceedings on Privacy Enhancing Technologies 2023(3), 418–431 © 2023 Copyright held by the owner/author(s). https://doi.org/10.56553/popets-2023-0089 1.2 Motivating a Novel Privacy Wrapper Algorithm 1: Exploiting fixed 𝑛through non-response Input: To overcome the challenges of untrusted code, this study intro- duces a novel privacy wrapper, called TAHOE. TAHOE treats the researcher script as a black box and uses subsets of data to ensure a type of sensitivity bound. It has the following properties: • Dataset 𝑆⊆𝐷 • Dataset 𝑆⊆𝐷 1 (Return 1) If |𝑆| < 𝑛or 𝑡∈𝑆, return 1 (1) Privacy: Given any researcher script, the mechanism gener- ated by the wrapper meets approximate differential privacy. If 𝑡∈𝐷, the script will output 1 for any subset of size 𝑛that contains 𝑡. Moreover, any such subset is stable, because if any rows are removed, the size will be less than 𝑛, so the script will again return 1. On the other hand, if 𝑡∉𝐷, the script will return ⊥for any subset of size 𝑛, so no stable subsets of size 𝑛exist. Thus, a stable subset of size 𝑛exists if and only if 𝑡∈𝐷. Therefore the adversary will know whether the target is in the data, based on whether or not the wrapper returns an output. f (2) Accuracy: The output of the wrapper is related to the re- searcher script in the sense that it is found by adding (prede- termined) noise to the output of the script on some subset of data. (3) Flexibility: TAHOE places no limitation on the code writ- ten by a researcher. Any script that returns a real vector, including those that may terminate without returning an output, can be used. To defend against attacks like this, our second idea is to random- ize over the size of subsets we consider, 𝑛. Intuitively, by making the number of individuals we remove random, we may conceal how many individuals are removed because they are influential.i A disadvantage of our algorithm is that the runtime can be prohibitively large, depending on the privacy parameters chosen. As we will show in Section 6, TAHOE can be optimized to run efficiently for the special case of a finite data alphabet. In Section 4, we define a distribution 𝐺, used to select a subset size 𝑛. The size ranges from 𝑁−𝑀to 𝑁, where 𝑁is the size of the dataset and 𝑀is a parameter representing the maximum number of individuals removed. The specific form of 𝐺is chosen to optimize privacy parameters. 1.2 Motivating a Novel Privacy Wrapper It is necessary that the internal parameter 𝑛is kept secret, in order our privacy guarantees to hold. fi y pi p We will describe TAHOE formally in Section 4. Its design is motivated by two central ideas, which we present at an intuitive level. 1.2.1 Remove influential individuals. An individual may be called influential, roughly speaking, if removing them causes a large change in the output of a script. For example, suppose that a re- searcher submits a script to compute the mean age of participants in a medical trial. If a target individual happens to be very young, an adversary may be able to distinguish whether they are in the data or not by observing the mean. On the other hand, if a privacy wrapper removes the influential individual before running the script, the presence of the individual can no longer be determined in such a direct manner. 1 INTRODUCTION Informally, sensitivity refers to how much the output of a script can change, if one were to change a single row of the data. Unfortunately, determining the sensitivity of a researcher script may be difficult, or even impossible. For an instructive example, again suppose that that an adversary is interested in a specific target, 𝑡, which may or may not be in the data. The adversary writes a script that outputs 0 unless 𝑡is in the data, in which case the script outputs some large 𝑉≫1. If 𝑡is not in the data, there may be no way to detect that the script is capable of outputting any value other than 0 - especially if the script is obfuscated. The adversary may thus hope to trick the data 418 Proceedings on Privacy Enhancing Technologies 2023(3) Differential Privacy for Black-Box Statistical Analyses holder into assuming the script has low sensitivity, leading them to add a small amount of noise. 1.2.2 Randomize number of individuals removed. A further prob- lem arises when basing a privacy wrapper on stable subsets: the existence of a stable subset of a given size can itself reveal infor- mation about individuals in the data. Suppose that a wrapper was programmed to find a stable subset of some fixed size 𝑛, and release some output if it exists. If a stable subset of size 𝑛does not exist, the system terminates without returning a value, denoted by the non-response symbol ⊥. A malicious researcher could exploit this wrapper design to learn whether a target individual, 𝑡, is inside the data. An example of a script that does this is given in Algorithm 1. There is also a theoretical reason to believe that determining the sensitivity of a script is difficult. In the language of the theory of computation, the property of meeting a sensitivity bound is semantic and non-trivial. Rice’s theorem tells us that the problem of deciding if arbitrary code meets such properties is undecidable [27]. This suggests that a traditional strategy of adding noise based on sensitivity cannot work in all cases. 1.3 Summary of Results Theorem 2 shows that when 𝑅returns a statistically consistent estimate for a population parameter, and TAHOE is configured to never return ⊥, TAHOE’s output will be consistent as well. To better understand the accuracy of TAHOE’s output, we per- form a set of simulation exercises in Section 7. Each exercise is based around an honest researcher that wishes to perform a styl- ized analysis task involving a finite data alphabet. As a benchmark for comparison, we also simulate the same tasks under the GUPT system, which is another algorithm that can fill the role of a privacy wrapper [23]. While these systems are very different from each other, we establish guidelines to make our comparison as fair as possible.i The first task we consider is generating a histogram showing the proportion of each possible data entry. In our experiments, we find that GUPT outperforms TAHOE for small datasets. However, above a certain data size, TAHOE’s output becomes more accurate, reflecting favorable scaling properties. TAHOE also performs rela- tively well as the size of the alphabet increases, but less well as 𝜖 approaches 0. In the Lipschitz extension approach, a researcher script is re- placed by an approximating function that has low global sensitivity but matches the output of the script on at least one reference dataset. The resulting function is known as a Lipschitz extension, and may be useful if its output matches the original script on typical datasets. Furthermore, the low global sensitivity means that a small amount of noise is sufficient to impose differential privacy. This approach was pioneered by Jha and Raskhodnikova, who developed algo- rithms for constructing Lipschitz extensions when each data entry takes on a finite set of values [14]. The original script is run on possible databases in a predetermined order, and the output is used to calibrate the extension. The second task we consider is computing the sample entropy of the data. We once again find that GUPT outperforms TAHOE for small datasets. TAHOE does gain an advantage as the dataset grows large, but it takes a much larger dataset (above 106 rows) before TAHOE becomes more accurate than GUPT. We argue that this is related to the fact that entropy is a one-dimensional quantity, and to the shape of the entropy function. This approach was extended by Cummings and Durfee, who pre- sented algorithms for infinite database domains [5]. 1.3 Summary of Results Within the differential privacy literature, three high-level strate- gies can be applied to black boxes. We will refer to these as sub- sample and aggregate, Lipschitz extension, and restricted language. In the subsample and aggregate approach, a dataset is divided into (small) subsets and the researcher script is run on each piece. The results are then aggregated together in some way to yield a final answer. Privacy protection emerges from the fact that a single individual can only affects the output resulting from a single subset (or a small fraction). Nissim, Raskhodnikova, and Smith were the first to outline this strategy, providing an example using the median as the aggregation function with noise calibrated to smooth sensitivity [24]. GUPT is a similar system, in which the mean is used as an aggregation function [23]. A feature of both of these systems is that a bound is required on the script output. As an alternative, a script can be run on subsamples, then the median can be released using the the Propose-Test-Release framework of Dwork and Lei [10]. Such a system does not require any bounds on the script output, but there is always some probability that the script returns a non-response character ⊥instead of returning an output. The privacy wrapper we propose similarly requires no bounds on script output, and returns ⊥in some cases. In cases when the output of a script is categorical instead of metric, aggregation can be performed through noisy voting [16, 26]. Compared to this lineage, our privacy wrapper also examines subsets of data, but does not require an aggregation step. Instead, our work is focused on finding subsets of data with favorable privacy characteristics prior to computing statistics. By contrast, we consider scenarios in which the data alphabet has finite size, |D| = 𝑓. In this case, we describe optimizations that allow TAHOE to run efficiently. Proposition 6 shows that the runtime now becomes O𝜏𝜖−𝑓ln(𝛿−1)𝑓 as 𝜖,𝛿→0, where 𝜏is a bound on the runtime of 𝑅. In Section 7 we describe the results of experiments involving finite data alphabets, in which TAHOE is able to run on a single computer.fi In general, measuring the accuracy of TAHOE’s output is difficult, because its behavior depends crucially on what is in the script 𝑅. Nevertheless, some guarantees can be derived for special cases. 1.3 Summary of Results Our algorithm TAHOE (Trim And withHold Or Execute), which is formally defined in Section 4, combines the two previous ideas. At a high level, TAHOE begins by selecting a random size 𝑛from distribution 𝐺. TAHOE then checks if any stable subsets of size 𝑛exist for a dataset 𝐷, which can be viewed as trimming influen- tial individuals from the data. If no stable subset of size 𝑛exists, TAHOE withholds from responding, represented by non-response ⊥. Otherwise, TAHOE selects a stable subset 𝑆of size 𝑛uniformly at random and executes the researcher’s script 𝑅on it, with the addition of Laplace noise.f Based on this observation, we consider looking for a subset of data, such that removing further individuals does not change the output very much. This intuition will be formalized in our notion of stable subsets. We prove in Theorem 1 that TAHOE imposes (𝜖,𝛿)-differential privacy. More precisely, fixing any researcher script 𝑅, the resulting behavior of TAHOE can be viewed as a function from the set of possible datasets to distributions over possible outputs. The dif- ferential privacy bound applies to this function, and it holds no matter what script 𝑅is chosen. Even if every individual is highly influential with respect to script 𝑅, differential privacy continues to hold, though the lack of stable subsets implies that TAHOE will always return ⊥in this case. Definition 1. (Informal Version of Definition 7) Given dataset 𝐷, researcher script 𝑅, and threshold 𝛼𝜆∈R, a subset 𝑆⊆𝐷is stable if for any 𝐴, 𝐵⊆𝑆of some minimum size, 𝑅returns a real vector for both 𝐴and 𝐵, and ||𝑅(𝐴) −𝑅(𝐵)||1 ≤𝛼𝜆. Definition 1. (Informal Version of Definition 7) Given dataset 𝐷, researcher script 𝑅, and threshold 𝛼𝜆∈R, a subset 𝑆⊆𝐷is stable if for any 𝐴, 𝐵⊆𝑆of some minimum size, 𝑅returns a real vector for both 𝐴and 𝐵, and ||𝑅(𝐴) −𝑅(𝐵)||1 ≤𝛼𝜆. A formal definition of this property is given in Section 4. Based on our intuition, we search for algorithms that run the script on a stable subset and add noise to protect privacy. 419 Proceedings on Privacy Enhancing Technologies 2023(3) Kohli and Laskowski Without restrictions on the set of possible data entries D, TAHOE is too computationally expensive to be practical. We provide a big-O bound in Proposition 5, and argue that (when maintaining𝛿< 1/𝑁), the runtime is superpolynomial in the size of the dataset. 1.3 Summary of Results In their system, a Lipschitz extension is constructed by first running the source script on the empty database. The output for any other database is found recursively, based on the output of its subsets. A theme that emerges from this work is that a Lipschitz extension is accurate when the output of the original script on the starting reference dataset is close to the output on the actual data. Another feature is that Lipschitz extensions take exponential time to compute, unless the researcher script is a white box with additional structure that can be exploited. 3Technically speaking, Δ(Ω) is the set of all probability distributions over some 𝜎- algebra of Ω. 4.1 Stable Subsets Before defining stability, a preliminary concept is that of responsive datasets. Essentially, this requires that a researcher script does not return ⊥when applied to the dataset or any subset of a minimum size. Given a nonempty set Ω, let Δ(Ω) be the set of all probabil- ity distributions over Ω.3 We define an algorithm as a function 𝐴: D★→Δ(R𝑘∪{⊥}). Here, ⊥is used to represent non-response, meaning an algorithm fails to return a real value. For ease of expo- sition, we define a researcher’s script as a deterministic function 𝑅: D★→R𝑘∪{⊥}. All results in this paper can be extended to researcher scripts that are randomized by adding a sampling step to the wrapper. Let R be the set of all deterministic algorithms.f Definition 6. (Responsive Subset) A subset 𝑆⊆D★is called (𝑅,𝑙)- responsive if for every 𝑋⊆𝑆of size at least 𝑙, we have 𝑅(𝑋) ∈R𝑘. Note that any multiset 𝑆of size less than 𝑙is trivially responsive by definition. Our definition of stable subsets refines responsiveness to convey the intuition that removing further individuals does not greatly affect script output. The notion of differential privacy is rooted in the indistinguisha- bility of probability distributions [9]. Definition 7. (Stability: Formal Version of Informal Definition 1) A multiset 𝑆∈D★is called (𝛼, 𝜆,𝑙, 𝑅)−stable if 𝑆is (𝑅,𝑙)-responsive, and for any subsets 𝑋,𝑌⊆𝑆of size at least 𝑙, 𝐿(𝑅(𝑋), 𝜆) ∼𝛼,0 𝐿(𝑅(𝑌), 𝜆).4 Definition 2. For any nonempty set Ω, two distributions 𝑑1,𝑑2 ∈ Δ(Ω) are (𝜖,𝛿)-indistinguishable if for every measurable set 𝐸⊆Ω, for all 𝑖, 𝑗∈{1, 2}, 𝑑𝑖(𝐸) ≤exp(𝜖)𝑑𝑗(𝐸) + 𝛿. Because 𝛼, 𝜆, 𝑙, and 𝑅will not change in this study, we will omit them and simply refer to the set 𝑆as stable. Additionally, any multiset 𝑆of size less than 𝑙is trivially stable by definition. We will denote the (𝜖,𝛿)-indistinguishability of 𝑑1 and 𝑑2 as 𝑑1 ∼𝜖,𝛿𝑑2. If 𝑑1, 𝑑2, and 𝑑3 are distributions, then the following transitive property holds. It will be useful to rewrite stability in terms of the following definitions. Let 𝑈be the set of all vectors of the form (𝑢1,𝑢2, ...,𝑢𝑘) where 𝑢𝑗∈{−1, +1} for all 𝑗. For 𝑢∈𝑈and 𝑆∈D★of size at least 𝑙, define 𝑚𝑖𝑛𝑢(𝑆) and 𝑚𝑎𝑥𝑢(𝑆) as follows: Lemma 1. [12] If 𝑑1 ∼𝜖,0 𝑑2, and 𝑑2 ∼𝜖′,0 𝑑3, then 𝑑1 ∼𝜖+𝜖′,0 𝑑3. Lemma 1. [12] If 𝑑1 ∼𝜖,0 𝑑2, and 𝑑2 ∼𝜖′,0 𝑑3, then 𝑑1 ∼𝜖+𝜖′,0 𝑑3. 4 ALGORITHM DESCRIPTION: TAHOE As noted in Section 1.2, our privacy wrapper TAHOE leverages stable subsets of random sizes to impose (𝜖,𝛿)-differential privacy. In this section, we formalize each of these notions and specify our algorithm. 2 RELATED WORKSf Differential privacy was introduced by Dwork et al. as a rigorous standard for mechanisms that compute real-valued statistics from personal data [8, 9, 11]. The authors pioneered privacy analysis based on global sensitivity, which is defined as the maximal change in a statistic resulting from a one row change to any dataset. Sub- sequent papers developed variants of the definition of differential privacy [3, 4, 6, 13, 22] whose analyses leverage global sensitivity. A number of studies have since developed differentially private mech- anisms leveraging local sensitivity, which is often much smaller than global sensitivity [15, 24, 31]. Such approaches cannot imme- diately be applied to black box researcher scripts, as neither local nor global sensitivity may be estimated. Similarly to the above work, our privacy wrapper leverages sub- setting to avoid searching through the entire space of possible datasets. However, we only consider subsets above a minimum size. As long as the researcher script has low sensitivity over this local domain, our privacy wrapper will output accurate results, even if the sensitivity bound is broken on other typical datasets. Addi- tionally, the number of relevant subsets may be far fewer than the 420 Differential Privacy for Black-Box Statistical Analyses Proceedings on Privacy Enhancing Technologies 2023(3) Our privacy wrapper will leverage the 𝑘-dimensional Laplace distribution. number considered under a Lipschitz extension approach, though it is still too large for practical computation when the data alphabet is infinite. Definition 5. The 𝑘-dimensional Laplace distribution with mean 𝜇∈R𝑘and scale 𝜆> 0, denoted as 𝐿𝑘(𝜇, 𝜆), is the distribution in Δ(R𝑘) with density function 𝑓(𝑧) = (2𝜆)−𝑘exp(−||𝑧−𝜇||1/𝜆). In the restricted language approach, researchers are permitted to design their own script, but must use a limited language that restricts their access to data. An example of such a system is PINQ, which presents programmers with a SQL-like interface with privacy guarantees [21]. Other differentially private SQL systems have been subsequently developed [15, 19, 32]. In a similar fashion, Kifer et al. propose an architecture in which access to data is mediated by a privacy layer that implements differentially private mechanisms [17]. A necessary and sufficient condition to determine whether two Laplace distributions with the same scale 𝜆are (𝛼, 0)-indistinguishable is given below. Lemma 2. For 𝜆> 0, ||𝜇−𝜈||1 ≤𝛼𝜆⇐⇒𝐿𝑘(𝜇, 𝜆) ∼𝛼,0 𝐿𝑘(𝜈, 𝜆). Lemma 2. For 𝜆> 0, ||𝜇−𝜈||1 ≤𝛼𝜆⇐⇒𝐿𝑘(𝜇, 𝜆) ∼𝛼,0 𝐿𝑘(𝜈, 𝜆). 3 PRELIMINARIES Let D be the set of possible entries that represent one individual in a dataset. A dataset is represented as a multiset of finite size with entries in D. Let D★be the set of all possible datasets. We say two datasets 𝐷1, 𝐷2 ∈D★are neighbors if one can be obtained from the other by switching exactly one element. That is, 𝐷1 and 𝐷2 are neighbors if there exists 𝑥∈𝐷1 and 𝑦∈𝐷2 such that 𝐷2 = (𝐷1 ∪{𝑦}) \ {𝑥}. 4.1 Stable Subsets Differential privacy applies the notion of indistinguishable proba- bility distributions to randomized algorithms on neighboring datasets. 𝑚𝑖𝑛𝑢(𝑆) = ( min𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) if 𝑆is (𝑅,𝑙)-responsive −∞ otherwise Definition 3. (Differential Privacy [8]) An algorithm 𝐴: D★→ Δ(R𝑘∪{⊥}) is (𝜖,𝛿)-differentially private if for any neighboring datasets 𝐷1 and 𝐷2, 𝐴(𝐷1) ∼𝜖,𝛿𝐴(𝐷2). Definition 3. (Differential Privacy [8]) An algorithm 𝐴: D★→ Δ(R𝑘∪{⊥}) is (𝜖,𝛿)-differentially private if for any neighboring datasets 𝐷1 and 𝐷2, 𝐴(𝐷1) ∼𝜖,𝛿𝐴(𝐷2). and A wrapper is a function𝑊: D★×R →Δ(R𝑘∪{⊥}) that takes a dataset 𝐷and an algorithm 𝑅as input and outputs a noisy response based on 𝑅and 𝐷. We extend the notion of differential privacy to wrappers as follows. 𝑚𝑎𝑥𝑢(𝑆) = ( max𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) if 𝑆is (𝑅,𝑙)-responsive ∞ otherwise Now stability can be expressed as follows: Proposition 1. A multiset 𝑆∈D★is (𝛼, 𝜆,𝑙, 𝑅)−stable if and only if either |𝑆| < 𝑙, or |𝑆| ≥𝑙and Proposition 1. A multiset 𝑆∈D★is (𝛼, 𝜆,𝑙, 𝑅)−stable if and only if either |𝑆| < 𝑙, or |𝑆| ≥𝑙and Definition 4. (Imposition of Differential Privacy) A wrapper 𝑊 imposes (𝜖,𝛿)−differential privacy if for any researcher algorithm 𝑅, the function 𝑊(·, 𝑅) is (𝜖,𝛿)−differentially private. max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  ≤𝛼𝜆 max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  ≤𝛼𝜆 By Lemma 2 𝐿(𝑅(𝑋) 𝜆) ∼𝛼0 (𝑅(𝑌) 𝜆) ⇐⇒||𝑅(𝑋) −𝑅(𝑌) ||1 ≤𝛼𝜆 4By Lemma 2, 𝐿(𝑅(𝑋), 𝜆) ∼𝛼,0 (𝑅(𝑌), 𝜆) ⇐⇒||𝑅(𝑋) −𝑅(𝑌) ||1 ≤𝛼𝜆. 421 Proceedings on Privacy Enhancing Technologies 2023(3) Kohli and Laskowski The proof of this proposition can be found in the appendix. From a computational perspective, 𝑚𝑖𝑛𝑢and 𝑚𝑎𝑥𝑢satisfy a recurrence relationship which will enable us to check the stability of subsets more efficiently. 4.2 Randomized Subset Sizes Our algorithm randomly chooses how many individuals to exclude when searching for a stable subset, making it harder for an adver- sary to leverage the size of stable subsets to extract information from the data. Throughout this paper, we will use 𝑁to represent the size of a dataset, and 𝑀to represent the maximum number of entries that a privacy wrapper may exclude. In our algorithm, 𝑀 will be set to the output of the following helper function Proposition 2. For 𝑆∈𝐷★of size at least 𝑙, the following hold: (a) If |𝑆| = 𝑙and 𝑅(𝑆) ≠⊥, then 𝑚𝑖𝑛𝑢(𝑆) = 𝑚𝑎𝑥𝑢(𝑆) = 𝑢· 𝑅(𝑆) for all 𝑢∈𝑈. Alternatively, if |𝑆| = 𝑙and 𝑅(𝑆) =⊥then 𝑚𝑖𝑛𝑢(𝑆) = −∞and 𝑚𝑎𝑥𝑢(𝑆) = ∞. (a) If |𝑆| = 𝑙and 𝑅(𝑆) ≠⊥, then 𝑚𝑖𝑛𝑢(𝑆) = 𝑚𝑎𝑥𝑢(𝑆) = 𝑢· 𝑅(𝑆) for all 𝑢∈𝑈. Alternatively, if |𝑆| = 𝑙and 𝑅(𝑆) =⊥then 𝑚𝑖𝑛𝑢(𝑆) = −∞and 𝑚𝑎𝑥𝑢(𝑆) = ∞. (a) If |𝑆| = 𝑙and 𝑅(𝑆) ≠⊥, then 𝑚𝑖𝑛𝑢(𝑆) = 𝑚𝑎𝑥𝑢(𝑆) = 𝑢· 𝑅(𝑆) for all 𝑢∈𝑈. Alternatively, if |𝑆| = 𝑙and 𝑅(𝑆) =⊥then 𝑚𝑖𝑛𝑢(𝑆) = −∞and 𝑚𝑎𝑥𝑢(𝑆) = ∞. ℎ(𝜖,𝛿, 𝛼) = ⌈𝑄−1 ln(𝛿−1 exp(𝜖)𝑄+ 1)⌉ ( ) ( ) (b) If |𝑆| > 𝑙and 𝑅(𝑆) ≠⊥, letting 𝐶(𝑆) be the set of subsets of 𝑆 of size |𝑆| −1, then for all 𝑢∈𝑈, where 𝑄= 𝜖(𝜖−4𝛼)(2𝜖−4𝛼)−1. Note that 𝑀decreases in both 𝜖 and 𝛿.i 𝑚𝑖𝑛𝑢(𝑆) = min  𝑢· 𝑅(𝑆), min 𝑐∈𝐶(𝑆) 𝑚𝑖𝑛𝑢(𝑐) 𝑚𝑖𝑛𝑢(𝑆) = min  𝑢· 𝑅(𝑆), min 𝑐∈𝐶(𝑆) 𝑚𝑖𝑛𝑢(𝑐) d 𝑚𝑎𝑥𝑢(𝑆) = max  𝑢· 𝑅(𝑆), max 𝑐∈𝐶(𝑆) 𝑚𝑎𝑥𝑢(𝑐) To select a size between 𝑁−𝑀and 𝑁, we define the distribution 𝐺(𝜖, 𝛼, 𝑁, 𝑀) as follows: 𝐺(𝜖, 𝛼, 𝑁, 𝑀)(𝑛) equals and 𝛿′ exp  min n (𝜖−4𝛼)(𝑛−𝑁+ 𝑀) −2𝛼,𝜖(𝑁−𝑛) o 𝑚𝑎𝑥𝑢(𝑆) = max  𝑢· 𝑅(𝑆), max 𝑐∈𝐶(𝑆) 𝑚𝑎𝑥𝑢(𝑐) for 𝑛∈{𝑁−𝑀, ..., 𝑁} and 0 otherwise, where 𝛿′ is a constant selected so that the total probability sums to 1. As we will see in the following section, the normalization constant 𝛿′ of 𝐺is closely related to the privacy parameter 𝛿in the (𝜖,𝛿)-differential privacy guarantee of TAHOE. We will normally omit the arguments to 𝐺 for readability purposes. A specific example of the shape of 𝐺when 𝑁= 100, 𝑀= 42,𝜖= 0.1, and 𝛼= 0.01 is provided in Figure 1. For these values, 𝛿′ ≈0.0098. for 𝑛∈{𝑁−𝑀, ..., 𝑁} and 0 otherwise, where 𝛿′ is a constant selected so that the total probability sums to 1. 4.2 Randomized Subset Sizes As we will see in the following section, the normalization constant 𝛿′ of 𝐺is closely related to the privacy parameter 𝛿in the (𝜖,𝛿)-differential privacy guarantee of TAHOE. We will normally omit the arguments to 𝐺 for readability purposes. A specific example of the shape of 𝐺when 𝑁= 100, 𝑀= 42,𝜖= 0.1, and 𝛼= 0.01 is provided in Figure 1. For these values, 𝛿′ ≈0.0098. Alternatively, if |𝑆| > 𝑙and 𝑅(𝑆) =⊥, then 𝑚𝑖𝑛𝑢(𝑆) = −∞ and 𝑚𝑎𝑥𝑢(𝑆) = ∞. Stable subsets obey certain algebraic properties that facilitate algorithmic analysis. As the following lemma shows, subsetting a stable subsets creates another stable subset. Lemma 3. If 𝑆is (𝛼, 𝜆,𝑙, 𝑅)−stable, and 𝑆′ ⊆𝑆, then 𝑆′ is also (𝛼, 𝜆,𝑙, 𝑅)−stable. Figure 1: The distribution 𝐺(𝑛) when 𝑁= 100, 𝑀= 42,𝜖= 0.1, and 𝛼= 0.01. For these values, 𝛿′ ≈0.0098. Proof. Consider 𝑆′ ⊆𝑆. If |𝑆′| < 𝑙, then 𝑆′ is trivially stable. Hence, consider the case when |𝑆′| ≥𝑙. For any 𝑋⊆𝑆′ of size at least 𝑙, 𝑋⊆𝑆, so 𝑅(𝑋) ≠⊥since 𝑆is (𝑅,𝑙)-responsive. Hence, 𝑆′ is (𝑅,𝑙)-responsive. Take any 𝑋,𝑌⊆𝑆′ of size at least 𝑙. Since 𝑆′ ⊆𝑆, the (𝛼, 𝜆,𝑙, 𝑅)−stability of 𝑆implies 𝐿(𝑅(𝑋), 𝜆) ∼𝛼,0 𝐿(𝑅(𝑌), 𝜆). Hence, 𝑆′ is also (𝛼, 𝜆,𝑙, 𝑅)−stable. □ Corollary 1. If 𝑆and ˆ𝑆are (𝛼, 𝜆,𝑙, 𝑅)−stable, then 𝑆∩ˆ𝑆is also (𝛼, 𝜆,𝑙, 𝑅)−stable. Corollary 1. If 𝑆and ˆ𝑆are (𝛼, 𝜆,𝑙, 𝑅)−stable, then 𝑆∩ˆ𝑆is also (𝛼, 𝜆,𝑙, 𝑅)−stable. Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩ˆ𝑆. Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩ˆ𝑆. 5 PRIVACY ANALYSIS OF TAHOE In this section, we show that our privacy wrapper TAHOE imposes (𝜖,𝛿)-differential privacy. 6 Return 𝜔 6 Return 𝜔 Throughout this section, we will use T𝐷∈Δ(R𝑘∪{⊥}) to denote the probability distribution induced by TAHOE on dataset 𝐷, researcher script 𝑅, and parameters 𝜖,𝛿, 𝛼, and 𝜆. Since 𝐷is the only input that will change in the upcoming proofs, we omit writing the other inputs for readability purposes. For any 𝐷∈D★, write T𝐷(·|𝑛) to represents the conditional probability distribution of the wrapper when 𝑛has been chosen in Step 1 of Algorithm 2. Algorithm 2: TAHOE If 𝑆is stable, add it to set S. 4 (Non-response condition) If S = ∅, set 𝜔=⊥. 4 (Non-response condition) If S = ∅, set 𝜔=⊥. 5 (Execute condition) Otherwise, select a stable subset 𝑆from S uniformly at random, and sample 𝜔∼𝐿(𝑅(𝑆), 𝜆). 6 Return 𝜔 5 (Execute condition) Otherwise, select a stable subset 𝑆from S uniformly at random, and sample 𝜔∼𝐿(𝑅(𝑆), 𝜆). 5 (Execute condition) Otherwise, select a stable subset 𝑆from S uniformly at random, and sample 𝜔∼𝐿(𝑅(𝑆), 𝜆). Algorithm 2: TAHOE (From Data Holder) (From Data Holder) • Dataset 𝐷with size 𝑁 • Privacy parameters 𝜖> 0 and 0 < 𝛿≤1 • Distinguishability parameter 𝛼< 𝜖/4 such that ℎ(𝜖,𝛿, 𝛼) < (𝑁−1)/2. (From Researcher) 𝝀: This parameter is chosen by the researcher and represents the scale of added Laplace noise. Decreasing 𝜆results in more accurate output; however, this also makes the definition of stable subsets stricter, which increases the probability that TAHOE halts without returning an output. We propose two heuristics for setting 𝜆. One is to decide how much noise can be added to the result while preserv- ing its utility. In our experiments, we found that the greatest source of error was the Laplace noise added in Step 5 of the algorithm. Further, the standard deviation of this noise is given by √ 2𝜆. The other heuristic we propose is to compute a value for 𝜆such that TAHOE is guaranteed to never (or rarely) to output ⊥. We give examples of such computations in Section 7. • Researcher script 𝑅 • Scale parameter 𝜆 Output: 𝜔∈R𝑘∪{⊥} 1 (Choose subset size) Set 𝑀= ℎ(𝜖,𝛿, 𝛼) and sample 𝑛∼𝐺(𝜖, 𝛼, 𝑁, 𝑀). 2 (Compute 𝑚𝑖𝑛𝑢and 𝑚𝑎𝑥𝑢) (a) For all subsets 𝑆of size 𝑁−2𝑀−1, execute the researcher script 𝑅on 𝑆. Compute 𝑚𝑖𝑛𝑢(𝑆) = 𝑚𝑎𝑥𝑢(𝑆) for all 𝑢∈𝑈according to the base case in Proposition 2(a). (b) Set 𝑖= 𝑁−2𝑀. 𝜶: This parameter is chosen by the data holder, and controls how indistinguishable 𝐿(𝑅(𝐴), 𝜆) and 𝐿(𝑅(𝐵), 𝜆) must be when 𝐴and 𝐵are subsets of a stable set. If 𝛼is too small, TAHOE will tend to return ⊥with high probability. If𝛼is too close to𝜖/4, 𝑀will become large, increasing the probability of returning ⊥. Additionally, setting 𝛼too high will violate the requirement that 𝑀= ℎ(𝜖,𝛿, 𝛼) < (𝑁− 1)/2. In our simulation studies, we have found that 𝛼= 𝜖/5 is a reasonable rule of thumb for all scenarios we considered. (c) While 𝑖≤𝑛 (I) For each subset 𝑆of size 𝑖, execute the researcher script 𝑅on 𝑆. Compute 𝑚𝑖𝑛𝑢(𝑆) and 𝑚𝑎𝑥𝑢(𝑆) for all 𝑢∈𝑈according to Proposition 2(b). (II) Set 𝑖= 𝑖+ 1. 3 (Check stability) Initialize S = ∅. For each subset 𝑆of size 𝑛, check whether 𝑆is stable according to Proposition 1. If 𝑆is stable, add it to set S. 3 (Check stability) Initialize S = ∅. For each subset 𝑆of size 𝑛, check whether 𝑆is stable according to Proposition 1. Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩ˆ𝑆. (c) While 𝑖≤𝑛 (I) For each subset 𝑆of size 𝑖, execute the researcher script 𝑅on 𝑆. Compute 𝑚𝑖𝑛𝑢(𝑆) and 𝑚𝑎𝑥𝑢(𝑆) for all 𝑢∈𝑈according to Proposition 2(b). (II) Set 𝑖= 𝑖+ 1. that increasing 𝑀will make it harder to find stable subsets, resulting in a higher probability of returning ⊥. Moreover, the requirement 𝑀= ℎ(𝜖,𝛿, 𝛼) < (𝑁−1)/2 places a lower bound on 𝜖and 𝛿. A second disadvantage is that a small 𝜖also requires 𝛼to be smaller, which may cause TAHOE to return ⊥with high probability, or require more noise. 4.3 Our General Algorithm Our specific privacy wrapper TAHOE (Trim And withHold Or Execute) described in Algorithm 2 takes as input a researcher script 𝑅: D★→Δ(R𝑘∪{⊥}) and a dataset 𝐷∈D★of size 𝑁. Additionally, TAHOE uses the following auxiliary parameters: 𝜖> 0, 𝛿> 0, 𝛼< 𝜖/4, and 𝜆> 0.i Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩ˆ𝑆. Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩ˆ𝑆. □ Figure 1: The distribution 𝐺(𝑛) when 𝑁= 100, 𝑀= 42,𝜖= 0.1, and 𝛼= 0.01. For these values, 𝛿′ ≈0.0098. Corollary 2. If 𝑆⊆𝐷is (𝛼, 𝜆,𝑙, 𝑅)−stable, and 𝐷and 𝐷′ are neigh- bors, then 𝑆∩𝐷′ is also (𝛼, 𝜆,𝑙, 𝑅)−stable. 𝐺is designed to rise and fall in an exponential manner. In the privacy analysis of Section 5, it will be helpful to relate the proba- bility at 𝑛to the total probability below 𝑛and the total probability above 𝑛. This is done by the following two propositions. Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩𝐷′. □ Proof. Apply Lemma 3 with 𝑆as is and 𝑆′ = 𝑆∩𝐷′. □ □ Given data 𝐷∈D★and researcher script 𝑅, let 𝑚(𝐷, 𝑅) be the size of the largest stable subset of 𝐷. The following lemma relates 𝑚to the idea of neighboring datasets. Proposition 3. For any 𝑟in the support of 𝐺,︁ 𝐺(𝑟) ≤(exp(𝜖−4𝛼) −1) ∑︁ 𝑛<𝑟 𝐺(𝑛) + exp(−2𝛼)𝛿′ Lemma 4. When 𝐷1, 𝐷2 ∈D★are neighbors,𝑚(𝐷1, 𝑅) and𝑚(𝐷2, 𝑅) differ by at most 1. Lemma 4. When 𝐷1, 𝐷2 ∈D★are neighbors,𝑚(𝐷1, 𝑅) and𝑚(𝐷2, 𝑅) differ by at most 1. ︁ Proposition 4. For any 𝑟in the support of 𝐺,︁ ︁ Proposition 4. For any 𝑟in the support of 𝐺,︁ 𝐺(𝑟) ≤(exp(𝜖) −1) ∑︁ 𝑛>𝑟 𝐺(𝑛) + 𝛿′ Proof. Suppose𝑆1 is a maximal stable subset of 𝐷1. Then𝑆1∩𝐷2 is a stable subset of 𝐷2 by Corollary 2. Further, |𝑆1∩𝐷2| ≥|𝑆1|−1 = 𝑚(𝐷1, 𝑅) −1. Therefore 𝑚(𝐷2, 𝑅) ≥𝑚(𝐷1, 𝑅) −1. By symmetric argument, 𝑚(𝐷1, 𝑅) ≥𝑚(𝐷2, 𝑅) −1. □ ︁ A proof of Proposition 3 is provided in the Appendix; the proof of Proposition 4 is similar to the previous one, so we omit it. □ 422 Proceedings on Privacy Enhancing Technologies 2023(3) Differential Privacy for Black-Box Statistical Analyses Algorithm 2: TAHOE Input: (From Data Holder) • Dataset 𝐷with size 𝑁 • Privacy parameters 𝜖> 0 and 0 < 𝛿≤1 • Distinguishability parameter 𝛼< 𝜖/4 such that ℎ(𝜖,𝛿, 𝛼) < (𝑁−1)/2. (From Researcher) • Researcher script 𝑅 • Scale parameter 𝜆 Output: 𝜔∈R𝑘∪{⊥} 1 (Choose subset size) Set 𝑀= ℎ(𝜖,𝛿, 𝛼) and sample 𝑛∼𝐺(𝜖, 𝛼, 𝑁, 𝑀). 2 (Compute 𝑚𝑖𝑛𝑢and 𝑚𝑎𝑥𝑢) (a) For all subsets 𝑆of size 𝑁−2𝑀−1, execute the researcher script 𝑅on 𝑆. Compute 𝑚𝑖𝑛𝑢(𝑆) = 𝑚𝑎𝑥𝑢(𝑆) for all 𝑢∈𝑈according to the base case in Proposition 2(a). (b) Set 𝑖= 𝑁−2𝑀. Theorem 1. TAHOE imposes (𝜖,𝛿)-differential privacy. Theorem 1. TAHOE imposes (𝜖,𝛿)-differential privacy. Theorem 1. TAHOE imposes (𝜖,𝛿)-differential privacy. Configuring TAHOE is complicated by the large number of pa- rameters, and by the different ways one might want to assess its behavior. In general, a multi-way trade-off exists among how much noise TAHOE adds, how likely TAHOE is to return ⊥, the privacy parameters, and how much sampling variation is introduced by sub- setting. The relationship among these quantities depends on what is in the researcher script, and therefore cannot be known precisely. In the following table, we provide some guidance by discussing the effects of each parameter in general terms. Before providing the proof, we sketch an outline of our argu- ment. In order to show that TAHOE imposes (𝜖,𝛿) differential pri- vacy, we first show in Lemma 5 that the value of 𝑀set in TAHOE produces 𝛿′ < 𝛿. Since (𝜖,𝛿′) differential privacy implies (𝜖,𝛿)- differential privacy, it is sufficient to show that TAHOE imposes (𝜖,𝛿′)-differential privacy. Next, we proceed by considering the behavior of TAHOE on two neighboring datasets 𝐷1 and 𝐷2. If there are no stable subsets of size 𝑁−𝑀for both of these datasets, then TAHOE returns ⊥with probability 1 on both datasets. Hence, the behavior of TAHOE is constant for all neighboring databases, so TAHOE trivially imposes (𝜖,𝛿′)-differential privacy. Lemma 5. Given the value of 𝑀chosen by TAHOE, 𝛿′ < 𝛿. T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) Let 𝐸⊆R𝑘∪{⊥} be measurable, and let 𝐷1 and 𝐷2 be neigh- boring datasets. We will prove the bound, T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) ≤ Í 𝑛<𝑟exp(2𝛼)𝐻(𝐸)𝐺(𝑛) + exp(2𝛼)𝐻(𝐸)𝐺(𝑟) −𝛿′ Í 𝑛<𝑟exp(−2𝛼)𝐻(𝐸)𝐺(𝑛) = exp(4𝛼) Í 𝑛<𝑟𝐺(𝑛) + 𝐺(𝑟) −exp(−2𝛼)𝐻(𝐸)−1𝛿′ Í 𝑛<𝑟𝐺(𝑛) ≤exp(4𝛼) Í 𝑛<𝑟𝐺(𝑛) + 𝐺(𝑟) −exp(−2𝛼)𝛿′ Í 𝑛<𝑟𝐺(𝑛) T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) ≤exp(𝜖) ≤ Í 𝑛<𝑟exp(2𝛼)𝐻(𝐸)𝐺(𝑛) + exp(2𝛼)𝐻(𝐸)𝐺(𝑟) −𝛿′ ≤ Í 𝑛<𝑟exp(−2𝛼)𝐻(𝐸)𝐺(𝑛) = exp(4𝛼) Í 𝑛<𝑟𝐺(𝑛) + 𝐺(𝑟) −exp(−2𝛼)𝐻(𝐸)−1𝛿′ Í 𝑛<𝑟𝐺(𝑛) ≤exp(4𝛼) Í 𝑛<𝑟𝐺(𝑛) + 𝐺(𝑟) −exp(−2𝛼)𝛿′ Í 𝑛<𝑟𝐺(𝑛) If there are no stable subsets of size 𝑁−𝑀of both 𝐷1 and 𝐷2, then T𝐷1 and T𝐷2 are the same distribution (giving probability 1 to ⊥), so the bound follows immediately. If there are any stable subsets of size 𝑁−𝑀of either 𝐷1 or 𝐷2, choose one and call it 𝐾. Define reference distribution 𝐻= 𝐿(𝑅(𝐾), 𝜆). Lemma 6. 𝐻is (2𝛼, 0)−indistinguishable from T𝐷𝑖(·|𝑛) for any 𝑖∈{1, 2} and for any 𝑛for which this algorithm does not return ⊥. where the last inequality follows since 𝐻(𝐸) ≤1. Plugging in 𝐺(𝑟) from Proposition 3 and simplifying bounds the ratio by exp(𝜖). Case ⊥∈𝐸: By Lemma 6, T𝐷𝑖(·|𝑛) ∼2𝛼,0 𝐻(·) when T𝐷𝑖doesn’t return ⊥. Proof. Let S𝑛 𝑖be the set of stable subsets of size 𝑛of 𝐷𝑖. For 𝑖∈{1, 2}, if T𝐷𝑖(·|𝑛) does not return ⊥, then for all measurable sets 𝐸⊆R𝑘∪{⊥},︁ When 𝑛< 𝑟, T𝐷𝑖doesn’t return ⊥, so T𝐷𝑖(𝐸|𝑛) = T𝐷𝑖(𝐸\{⊥}|𝑛), hence exp(−2𝛼)𝐻(𝐸\ {⊥}) ≤T𝐷𝑖(𝐸|𝑛) ≤exp(2𝛼)𝐻(𝐸\ {⊥}). Additionally, T𝐷𝑖({⊥}|𝑛) = 1 for all 𝑛> 𝑟, so T𝐷𝑖(𝐸|𝑛) = 1 for all 𝑛> 𝑟. Hence,︁︁ T𝐷𝑖(·|𝑛)(𝐸) = 1 |S𝑛 𝑖| ∑︁ 𝑆∈S𝑛 𝑖 𝐿(𝑅(𝑆), 𝜆)(𝐸)fi ︁ So it is sufficient to show that every distribution of the form 𝐿(𝑅(𝑆), 𝜆) where 𝑆⊆𝐷𝑖is a stable subset of size 𝑛≥𝑁−𝑀is 2𝛼-indistinguishable from 𝐻.i T𝐷1 (𝐸|𝑛) = ∑︁ 𝑛<𝑟 T𝐷1 (𝐸|𝑛)𝐺(𝑛) + T𝐷1 (𝐸|𝑟)𝐺(𝑟) + ∑︁ 𝑛>𝑟 T𝐷1 (𝐸|𝑛)𝐺(𝑛) ≤ ∑︁ 𝑛<𝑟 exp(2𝛼)𝐻(𝐸\ {⊥})𝐺(𝑛) + 𝐺(𝑟) + ∑︁ 𝑛>𝑟 𝐺(𝑛) For stable 𝑆⊆𝐷𝑖of size 𝑛≥𝑁−𝑀, we first establish that |𝑆∩𝐾| ≥𝑁−2𝑀−1. By inclusion-exclusion, |𝑆∪𝐾| = |𝑆|+|𝐾|− |𝑆∩𝐾|. Since |𝑆| ≥𝑁−𝑀, |𝐾| = 𝑁−𝑀, and 𝑆∪𝐾⊆𝐷1 ∪𝐷2 implies |𝑆∪𝐾| ≤|𝐷1∪𝐷2| = 𝑁+1, we have |𝑆∩𝐾| ≥𝑁−2𝑀−1. Because 𝑆and 𝐾are stable, and 𝑆∩𝐾is a subset of size at least 𝑁−2𝑀−1, we have 𝐿(𝑅(𝑆), 𝜆) ∼𝛼,0 𝐿(𝑅(𝑆∩𝐾), 𝜆) by Corollary 1. Additionally, since 𝐾is stable, and 𝑆∩𝐾is a subset, 𝐿(𝑅(𝐾), 𝜆) ∼𝛼,0 𝐿(𝑅(𝑆∩𝐾), 𝜆). Table 1: Description of Parametersf Table 1: Description of Parameters 𝝐and 𝜹: These are the differential privacy parameters, which are chosen by the data holder. Decreasing 𝜖and 𝛿corresponds to a stronger privacy guarantee. However, setting these parameters too small presents two disadvantages. One disadvantage is that a smaller 𝜖and 𝛿results in a higher 𝑀, requiring the algorithm to consider smaller subsets of data. At least in typical cases, we expect On the other hand, if there is a stable subset of size 𝑁−𝑀of at least one of the datasets, we can pick any stable subset 𝐾and define a reference distribution 𝐻= 𝐿(𝑅(𝐾), 𝜆). For any value of 𝑛 for which TAHOE does not return ⊥, we then show in Lemma 6 423 Proceedings on Privacy Enhancing Technologies 2023(3) Kohli and Laskowski where the inequality follows from Lemma 6. Additionally,︁ that both conditional distributions𝑇𝐷1 (·|𝑛) and𝑇𝐷2 (·|𝑛) are (2𝛼, 0)- indistinguishable from 𝐻. Using this result, we finish the proof by showing that, for any measurable set 𝐸⊆R𝑘∪{⊥}, 𝑇𝐷1 (𝐸) ≤ exp(𝜖)𝑇𝐷2 (𝐸) + 𝛿′ based on two cases: when ⊥∉𝐸and when ⊥∈𝐸. T𝐷2 (𝐸) = ∑︁ 𝑛<𝑟 T𝐷2 (𝐸|𝑛)𝐺(𝑛) + T𝐷2 (𝐸|𝑟)𝐺(𝑟) ≥ ∑︁ 𝑛<𝑟 T𝐷2 (𝐸|𝑛)𝐺(𝑛) ≥ ∑︁ 𝑛<𝑟 exp(−2𝛼)𝐻(𝐸)𝐺(𝑛) Proof. We will show that TAHOE imposes (𝜖,𝛿′)-differential privacy, where 𝛿′ is the normalization constant in the definition of 𝐺. This is sufficient because of the following lemma, which is proven in the Appendix. ︁ where the last inequality results from Lemma 6. Combining these two observations, Lemma 5. Given the value of 𝑀chosen by TAHOE, 𝛿′ < 𝛿. 5In practice, to prevent side-channel attacks, we believe the data holder must impose a timeout on the researcher script. In case 𝑅breaches the timeout on subset 𝑆, TAHOE should assign 𝑅(𝑆) to be ⊥. The length of the timeout could, in principle, depend on 𝑁and other parameters. Lemma 5. Given the value of 𝑀chosen by TAHOE, 𝛿′ < 𝛿. When |D| = 𝑓, the number of subsets of a dataset 𝐷∈D𝑁 (a) of size 𝑁−𝑖for 𝑖≥0 is upper bounded by 𝑖+𝑓−1 𝑓−1  (b) of size at least 𝑁−2𝑀−1 is upper bounded by 2𝑀+1+𝑓 𝑓 . (a) of size 𝑁−𝑖for 𝑖≥0 is upper bounded by 𝑖+𝑓−1 𝑓−1  (b) of size at least 𝑁−2𝑀−1 is upper bounded by 2𝑀+1+𝑓 𝑓 . ︁ ≤exp(𝜖) To optimize TAHOE for this setting, we index subsets of 𝐷ac- cording to their histograms, in order to avoid duplicating computa- tions on equivalent subsets. Thus, in Step 2(a), TAHOE must run 𝑅 on every histogram of size 𝑁−2𝑀−1. By Lemma 7(a), there are at most 𝑁−2𝑀−2+𝑓 𝑓−1  such histograms. By similar analysis, in Step 2(c)(I), TAHOE runs 𝑅on at most 𝑖+𝑓−1 𝑓−1  such histograms of size 𝑁−𝑖. □ 6.1 Runtime Analysis for Finite Alphabets While the runtime bound of Proposition 5 is impractical for all but small datasets with weak privacy parameters, there is one situation in which TAHOE can be modified to run much faster. In particular, when the set of possible data entries D is finite of size 𝑓, the number of unique subsets of 𝐷∈D★that TAHOE must consider is greatly reduced. Intuitively, when 𝑁≥𝑓, by the pigeonhole principle some of the rows in the dataset will necessarily hold the same value; removing any one of them results in the same subset as removing any other. Proposition 6. For fixed 𝑅,𝑘, and D with |D| = 𝑓, and 𝜖= 𝑔𝛼for constant𝑔> 4, the worst-case runtime of TAHOE is O𝜏𝜖−𝑓ln(𝛿−1)𝑓 as 𝜖,𝛿→0. Proposition 5. For fixed 𝑅and 𝑘, and 𝜖= 𝑔𝛼for constant 𝑔> 4, the worst-case runtime of TAHOE is O  (𝜏+ 𝑁)(𝑁+ 1) ˆ𝑔𝜖−1(1−ln(𝛿))+3  as 𝑁→∞and 𝜖,𝛿→0, where ˆ𝑔= (4𝑔−8)(𝑔−4)−1. Notice that the dependency on 𝑁has disappeared, and a depen- dency on 𝑓has been introduced. At an intuitive level, the histogram representation provides a lower dimensional representation of sub- sets, reducing the number of subsets that need to be examined. In upcoming Section 7, we run experiments involving data alphabet sizes from 2 to 4 and find that the algorithm can readily run on a single machine.i A proof of this proposition can be found in the Appendix. For fixed privacy parameters, 𝜖and 𝛿, and fixed 𝜏, the runtime of TAHOE is polynomial in 𝑁. However, a best practice first sug- gested by Dwork, et al. is to maintain 𝛿< 1/𝑁[11]. This causes the runtime to be superpolynomial in 𝑁. For finite data alphabets that are too large for TAHOE to run on a single machine, parallel computation can expand the applicability of the algorithm. In particular, the most computationally expensive steps are 2(a) and 2(c)(I). Both of these can be distributed over a parallel cluster. 6 RUNTIME ANALYSIS OF TAHOE To analyze the execution time of TAHOE, we let 𝜏represent the worst case runtime of 𝑅.5 Without any restrictions, TAHOE takes prohibitively long to execute. The reason is that, in the worst case, Step 2 of the algorithm requires executing the researcher script on every possible subset of data with size ranging from 𝑁−2𝑀−1 to 𝑁. The following proposition places a bound on the execution time without any restrictions on |D|. To attain this result, we must specify how the input parameter 𝛼changes as the other parameters change. We suppose that the data holder sets 𝛼proportional to 𝜖, which is consistent with the rule of thumb (𝜖= 5𝛼) we use in our simulations in Section 7. Overall, when TAHOE samples 𝑛in Step 1, in Step 2 TAHOE runs 𝑅on every histogram with size in {𝑁−2𝑀−1, ...,𝑛}. In the worst case, 𝑛= 𝑁in Step 1. By Lemma 7(b), the total number of histograms to check is upper bounded by 2𝑀+1+𝑓 𝑓 . When |D| = 𝑓, the runtime of TAHOE is given in the following proposition, with the proof deferred to the appendix. Proposition 6. For fixed 𝑅,𝑘, and D with |D| = 𝑓, and 𝜖= 𝑔𝛼for constant𝑔> 4, the worst-case runtime of TAHOE is O𝜏𝜖−𝑓ln(𝛿−1)𝑓 as 𝜖,𝛿→0. Lemma 5. Given the value of 𝑀chosen by TAHOE, 𝛿′ < 𝛿. Thus 𝐿(𝑅(𝑆), 𝜆) ∼2𝛼,0 𝐿(𝑅(𝐾), 𝜆) by Lemma 1. □ ︁ Additionally,︁ T𝐷2 (𝐸|𝑛) = ∑︁ 𝑛<𝑟 T𝐷1 (𝐸|𝑛)𝐺(𝑛) + T𝐷2 (𝐸|𝑟)𝐺(𝑟) + ∑︁ 𝑛>𝑟 T𝐷2 (𝐸|𝑛)𝐺(𝑛) ≥ ∑︁ 𝑛<𝑟 exp(−2𝛼)𝐻(𝐸\ {⊥})𝐺(𝑛) + T𝐷2 (𝐸|𝑟)𝐺(𝑟) + ∑︁ 𝑛>𝑟 𝐺(𝑛) ≥ ∑︁ 𝑛<𝑟 exp(−2𝛼)𝐻(𝐸\ {⊥})𝐺(𝑛) + ∑︁ 𝑛>𝑟 𝐺(𝑛) Recall that 𝑚(𝐷𝑖, 𝑅) is the size of the largest stable subset in 𝐷𝑖. Let 𝑟= max(𝑚(𝐷1, 𝑅),𝑚(𝐷2, 𝑅)). By Lemma 4, we know that 𝑚(𝐷1, 𝑅),𝑚(𝐷2, 𝑅) ∈{𝑟−1,𝑟}. For 𝑖∈{1, 2}, TAHOE returns ⊥ on data 𝐷𝑖if and only if 𝑛> 𝑚(𝐷𝑖, 𝑅). This means that TAHOE will return ⊥for any 𝑛> 𝑟and will not return ⊥for any 𝑛< 𝑟. ︁︁ where the last inequality follows as 𝐺(𝑟) ≥0. Combining these observations, T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) ≤ Í 𝑛<𝑟exp(2𝛼)𝐻(𝐸\ {⊥})𝐺(𝑛) + 𝐺(𝑟) + Í 𝑛>𝑟𝐺(𝑛) −𝛿′ Í 𝑛<𝑟exp(−2𝛼)𝐻(𝐸\ {⊥})𝐺(𝑛) + Í 𝑛>𝑟𝐺(𝑛) For 𝑖∈{1, 2} the law of total probability implies︁︁ T𝐷𝑖(𝐸) = ∑︁ 𝑛<𝑟 T𝐷𝑖(𝐸|𝑛)𝐺(𝑛) + T𝐷𝑖(𝐸|𝑟)𝐺(𝑟) + ∑︁ 𝑛>𝑟 T𝐷𝑖(𝐸|𝑛)𝐺(𝑛) ︁ We consider two cases: ︁ We consider two cases: Substituting for 𝐺(𝑟) in the numerator from Proposition 4 and rearranging, the right-hand side is upper-bounded by Case ⊥∉𝐸: For 𝑛> 𝑟, TAHOE returns ⊥, so T𝐷𝑖(𝐸|𝑛) = 0. Hence,︁ Substituting for 𝐺(𝑟) in the numerator from Proposition 4 and rearranging, the right-hand side is upper-bounded by T𝐷1 (𝐸) = ∑︁ 𝑛<𝑟 T𝐷1 (𝐸|𝑛)𝐺(𝑛) + T𝐷1 (𝐸|𝑟)𝐺(𝑟) ≤ ∑︁ 𝑛<𝑟 exp(2𝛼)𝐻(𝐸)𝐺(𝑛) + exp(2𝛼)𝐻(𝐸)𝐺(𝑟) exp(4𝛼) Í 𝑛<𝑟𝐻(𝐸\ {⊥})𝐺(𝑛) + exp(𝜖−2𝛼) Í 𝑛>𝑟𝐺(𝑛) Í 𝑛<𝑟𝐻(𝐸\ {⊥})𝐺(𝑛) + exp(2𝛼) Í 𝑛>𝑟𝐺(𝑛) 424 Proceedings on Privacy Enhancing Technologies 2023(3) Differential Privacy for Black-Box Statistical Analyses number of excluded elements is Í𝑓 𝑗=1 𝑤𝑗= 𝑖. The following lemma places bounds on the number of such histograms, with the proof deferred to the Appendix. Since 𝜖> 4𝛼, the fraction is greater than 1, so we can subtract the first terms from the numerator and denominator and maintain the inequality:︁ Since 𝜖> 4𝛼, the fraction is greater than 1, so we can subtract the first terms from the numerator and denominator and maintain the inequality:︁ T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) ≤exp(4𝛼) exp(𝜖−4𝛼) ∑︁ 𝑛>𝑟 𝐺(𝑛) = exp(𝜖) ∑︁ 𝑛>𝑟 𝐺(𝑛) ≤exp(𝜖) T𝐷1 (𝐸) −𝛿′ T𝐷2 (𝐸) ≤exp(4𝛼) exp(𝜖−4𝛼) ∑︁ 𝑛>𝑟 𝐺(𝑛) = exp(𝜖) ∑︁ 𝑛>𝑟 𝐺(𝑛) ≤exp(𝜖) Lemma 7. When |D| = 𝑓, the number of subsets of a dataset 𝐷∈D𝑁 Lemma 7. 7.2 Normalized Histogrami Theorem 2. If 𝑅is consistent for parameter 𝜃and TAHOE is con- figured to never return ⊥, lim𝑁→∞𝜆= 0, and lim𝑁→∞𝑁−𝑀= ∞ then the random variable formed by composing TAHOE with 𝑅is consistent for parameter 𝜃. The first script we consider is one that returns a normalized his- togram representing the data. Given alphabet D = (𝜎1, ..., 𝜎𝑓) and dataset 𝐷∈D𝑁, define the (un-normalized) histogram H (𝐷) = (𝑣1, ..., 𝑣𝑓), where 𝑣𝑗denote the number of times 𝜎𝑗appears in 𝐷. A script to return a normalized histogram is then specified by P(𝐷) = H (𝐷)/|𝐷|. A proof is given in the appendix. A similar result holds for re- searcher scripts that implement unbiased estimators of 𝜃. At a high level, the normalized histogram is well-matched with a subsample-and-aggregate strategy, because the proportion of each 𝜎𝑗on a small subset of data tends to represent the overall proportion of 𝜎𝑗well [24]. In fact, as long as the subsets have equal size, GUPT introduces no sampling variation, so the only error comes from added Laplace noise. To configure GUPT, we set the bound on script output in each dimension to be [0, 1]. 6We define a topology over R𝑘∪{⊥} using the Alexandroff Extension [7] of the standard topology of R𝑘with the singleton {⊥}. 7By Lemma 2, a necessary and sufficient condition to ensure TAHOE does not return ⊥is to set 𝛼and 𝜆such that ||𝑅(𝑋) −𝑅(𝑌) ||1 ≤𝛼𝜆for all sets 𝑆,𝑋,𝑌such that 𝑋,𝑌⊆𝑆⊆𝐷, where |𝑆| ∈{𝑁−𝑀, ..., 𝑁} and |𝑋|, |𝑌| ≥𝑁−𝑀. 7 ACCURACY ANALYSIS OF TAHOE This is done so that both systems always output real vectors, and accuracy can be measured using a common metric (standard error). 𝐵) →1 as 𝑁→∞. When 𝑅has bounded global sensitivity, it is possible to configure TAHOE so that it preserves consistency. For this discussion, we allow that parameters 𝜖, 𝛼, 𝑀, 𝜆may be chosen as functions of 𝑁. Three conditions are required. First, 𝛼and 𝜆must be chosen so that TAHOE never returns ⊥.7 Second, the noise scale 𝜆must approach zero as 𝑁→∞. Third, 𝑀must be fixed or grow slowly enough that lim𝑁→∞𝑁−𝑀= ∞. An advantage of GUPT that we do not capture in our results is that it uses pure differential privacy. We nevertheless use a common 𝜖for all systems. We allow TAHOE to have the extra advantage of approximate differential privacy, but require that 𝛿< 1/𝑁[1, 25]. p y Three conditions are required. First, 𝛼and 𝜆must be chosen so that TAHOE never returns ⊥.7 Second, the noise scale 𝜆must approach zero as 𝑁→∞. Third, 𝑀must be fixed or grow slowly enough that lim𝑁→∞𝑁−𝑀= ∞. 6We define a topology over R𝑘∪{⊥} using the Alexandroff Extension [7] of the standard topology of R𝑘with the singleton {⊥}. 7fi 7.1 Accuracy Simulations To gain insight into additional accuracy characteristics of our pri- vacy wrapper, we consider a set of example scripts that an (honest) researcher may want to run. To make the runtime practical, we chose stylized analysis tasks involving finite data alphabets. For each script, our goal is to understand how well our privacy wrapper performs, and compare it to a benchmark. Unlike GUPT, TAHOE introduces both sampling variation and added noise. Several parameters must be configured to run TAHOE. We assume that the end user follows our recommendation of setting 𝛼= 𝜖/5. We follow the practice of setting 𝛿= 1/(𝑁+ 1). For this exercise, the benchmark that we select is GUPT, which follows the sample-and-aggregate strategy [23]. To summarize the algorithm, GUPT partitions the dataset into 𝑁0.4 pieces, runs the script on each piece, then releases the average output with Laplace noise. Unlike TAHOE, GUPT has the advantage of using pure differ- ential privacy. As an additional benchmark, we consider the Laplace mechanism with global sensitivity as a white-box privacy preserv- ing algorithm [11]. This provides a measure of the cost arising from treating a script in a black-box fashion. To guarantee that TAHOE never returns ⊥, it is necessary and sufficient to make sure it doesn’t return ⊥when 𝑛= 𝑁is cho- sen in Step 1. This is true whenever the entire dataset 𝐷is sta- ble. Equivalently, for any subsets 𝐴, 𝐵⊆𝐷, with size at least 𝑁−2𝑀−1, we must check that ||P(𝐴) −P(𝐵)||1 ≤𝜆𝛼. Let vector 𝑎= (𝑎1, ..,𝑎𝑓) = H (𝐷) −H (𝐴) represent how many in- stances of each 𝜎𝑗were removed from 𝐷to form 𝐴. Likewise, let 𝑏= (𝑏1, ..,𝑏𝑓) = H (𝐷) −H (𝐵) . Then ||𝑎||1 = Í𝑓 𝑗=1 𝑎𝑗≤2𝑀+ 1 and ||𝑏||1 = Í𝑓 𝑗=1 𝑏𝑗≤2𝑀+ 1. We can then write, A comparison of TAHOE and GUPT is complicated by funda- mental differences between the algorithms. First, the parameters of each algorithm are different, and setting them requires judge- ment on the part of the researcher. Second, GUPT guarantees pure differential privacy while TAHOE uses approximate differential privacy. Finally, GUPT always provide an output, while TAHOE sometimes returns ⊥, making it difficult to compare accuracy on the same scale. 7By Lemma 2, a necessary and sufficient condition to ensure TAHOE does not return ⊥is to set 𝛼and 𝜆such that ||𝑅(𝑋) −𝑅(𝑌) ||1 ≤𝛼𝜆for all sets 𝑆,𝑋,𝑌such that 𝑋,𝑌⊆𝑆⊆𝐷, where |𝑆| ∈{𝑁−𝑀, ..., 𝑁} and |𝑋|, |𝑌| ≥𝑁−𝑀. 7 ACCURACY ANALYSIS OF TAHOE The performance of privacy-protecting systems is often assessed via a statement that relates the privacy level to accuracy, which may be measured in terms of mean squared error (MSE), standard error, or some other metric. For researcher scripts in general, there is no way to know what accuracy TAHOE provides for given parameter values, because that depends crucially on what is inside the script. However, we are able to provide some results related to accuracy for specific scripts, or classes of scripts.i More formally, when D = {𝜎1, ..., 𝜎𝑓} any dataset 𝐷∈D𝑁can be written in histogram form as 𝐷= (𝑣1, ..., 𝑣𝑓) where 𝑣𝑗denotes the number times 𝜎𝑗appears in the dataset and Í𝑓 𝑗=1 𝑣𝑗= 𝑁[12]. Any subset 𝑆⊆𝐷of size at least 𝑁−𝑖is represented by a histogram (𝑣1−𝑤1, 𝑣2−𝑤2, ..., 𝑣𝑓−𝑤𝑓), where each 𝑤𝑗∈{0, ..., 𝑣𝑗} represents the number of excluded data elements with value 𝜎𝑗, and the total Our first result motivates the use of TAHOE for inferential sta- tistics. When performing inference tasks, it is common to assume that the data entries are drawn independently from a probability distribution [29, 30]. Let F be a distribution over the set of data val- ues D, described by a parameter 𝜃∈R𝑘. Further, let 𝑫𝑁represent a random dataset of 𝑁rows drawn independently and identically from F. A useful set of scripts are those that estimate 𝜃consistently. 425 Proceedings on Privacy Enhancing Technologies 2023(3) Kohli and Laskowski Kohli and Laskowski Definition 8. A script 𝑅: D★→R𝑘∪{⊥} is consistent for parameter 𝜃∈R𝑘if In the case of GUPT, the main parameter we must set is the bounding rectangle for the script output. For the sake of simplicity, we use the most extreme values that are mathematically possible, meaning that the output is never censored. In the case of TAHOE, we must set the noise scale 𝜆. For each script, we compute a value of 𝜆large enough to guarantee that the algorithm never returns ⊥. This is done so that both systems always output real vectors, and accuracy can be measured using a common metric (standard error). plim𝑁→∞𝑅(𝑫𝑁) = 𝜃 y p meaning that the output is never censored. In the case of TAHOE, we must set the noise scale 𝜆. For each script, we compute a value of 𝜆large enough to guarantee that the algorithm never returns ⊥. 7.1 Accuracy Simulations These differences between the systems prevent us from comparing them “straight out of the box.”i ||P(𝐴) −P(𝐵)||1 = ||(P(𝐷) −P(𝐵)) −(P(𝐷) −P(𝐴))||1 ||P(𝐴) −P(𝐵)||1 = ||(P(𝐷) −P(𝐵)) −(P(𝐷) −P(𝐴))||1 = 𝑏 |𝐵| −𝑎 |𝐴| 1 ≤||𝑏||1 |𝐵| + ||𝑎||1 |𝐴| ≤2𝑀+ 1 |𝐵| + 2𝑀+ 1 |𝐴| ≤2(2𝑀+ 1) 𝑁−2𝑀−1 To configure TAHOE and GUPT and make the comparison as fair as possible, we adopt a set of guidelines. First, we follow simple heuristics to set parameter values when possible. This is aligned with our view that a privacy wrapper should not require specialized knowledge on the part of a researcher. Second, we set parameters in a way that makes the privacy wrappers as comparable as possible. where the last inequality follows as |𝐴|, |𝐵| ≥𝑁−2𝑀−1. Thus setting 𝜆= 2(2𝑀+ 1)/((𝑁−2𝑀−1)𝛼) where 𝑀= ℎ(𝜖,𝛿, 𝛼) is sufficient to guarantee that TAHOE never returns ⊥. where the last inequality follows as |𝐴|, |𝐵| ≥𝑁−2𝑀−1. Thus setting 𝜆= 2(2𝑀+ 1)/((𝑁−2𝑀−1)𝛼) where 𝑀= ℎ(𝜖,𝛿, 𝛼) is sufficient to guarantee that TAHOE never returns ⊥. 426 Proceedings on Privacy Enhancing Technologies 2023(3) Differential Privacy for Black-Box Statistical Analyses (a) Total error when 𝜖= 1. (b) View of Figure 2(a)’s area of detail. (c) Total error when 𝑓= 2 and 𝑁= 105. Figure 2: The total error, including sampling error and added noise, for the normalized histogram script under different systems in log-scale. (c) Total error when 𝑓= 2 and 𝑁= 105. (a) Total error when 𝜖= 1. (b) View of Figure 2(a)’s area of detail. (b) View of Figure 2(a)’s area of detail. (c) Total error when 𝑓= 2 and 𝑁= 105. (a) Total error when 𝜖= 1. Figure 2: The total error, including sampling error and added noise, for the normalized histogram script under different systems in log-scale. Given these configurations, we perform experiments in which GUPT and TAHOE are each executed on a randomly generated dataset. Each row of data is an independent draw from a discrete uniform distribution. The parameter 𝜖is set to values ranging from 0.25 to 2 and 𝑁is set to values in {102, 103, 104, 105}. The alphabet size is set to values in {2, 3, 4}. To estimate error, each experiment was replicated 50 times. All experiments were completed on a single computer with TAHOE and GUPT implemented in Python. Finally, as seen in Figure 2(c), decreasing 𝜖results in greater error for every system. 7.3 Sample Entropy We now turn our attention to statistics other than the normalized histogram. Before we begin, we note that a histogram script could be used as a building block to compute other statistics. A researcher could first submit the normalized histogram script from the previous section to TAHOE, then compute a statistic of interest directly from the privatized histogram. While this approach is general, there are situations in which a researcher can achieve more accurate results by supplying a script that directly computes the statistic of interest. First, as can be seen in Figure 2 (a) and (b), the error of every system decreases as the data size grows along the horizontal axis. For small datasets, GUPT outperforms TAHOE, featuring an order of magnitude less error in some cases. In this range, our internal parameter 𝑀is large relative to 𝑁, meaning that TAHOE must consider relatively small subsets. These subsets can have signifi- cantly different proportions, so a lot noise is required to make them indistinguishable. By contrast, the number of subsets in GUPT is large as a fraction of 𝑁, reducing the sensitivity of the output to any one subset. In this section, we consider a researcher who supplies a script to compute sample entropy. As a general rule, we have found that for roughly balanced datasets, a researcher can achieve greater accu- racy by first computing a histogram using TAHOE, then computing the entropy of that histogram. For unbalanced datasets with low entropy, a researcher can achieve greater accuracy by submitting a script that computes entropy directly. For larger datasets, TAHOE’s advantages grow. The internal parameter 𝑀grows slowly compared to 𝑁, meaning that the sub- sets under consideration have more similar proportions. Beyond a certain dataset size, TAHOE features less total error than GUPT. Again letting 𝑓represent the size of the data alphabet D, the sample entropy of database 𝐷with normalized histogram P(𝐷) = (𝑝1, ..., 𝑝𝑓) is defined as, Next, as also seen in Figure 2 (a) and (b), increasing the alphabet size 𝑓from 2 to 3, and to 4, leads to more error for both privacy wrappers. As the dimensionality increases, TAHOE shows a relative advantage, in the sense that it takes fewer rows of data before TAHOE’s error drops below that of GUPT. In the case of TAHOE, the noise scale 𝜆does not depend on 𝑓. 7.1 Accuracy Simulations In the case of GUPT, the curve is approximately hyperbolic, since the added noise scale increases with 1/𝜖. On the other hand, TAHOE’s total error increases faster as 𝜖approaches zero, crossing the curve for GUPT. Examining the expression for TAHOE’s noise scale 𝜆, there is an 𝛼in the denominator, and 𝛼is set proportionally to 𝜖, which would suggest a hyperbolic relationship. However, there is an extra dependency on 𝜖, since 𝑀increases as 𝜖 decreases, driving 𝜆higher for small values of 𝜖. Each panel of Figure 2 displays the total standard error (root- mean-square error) for different parameter values. The error for each histogram is measured using the L1 norm, and includes both sampling error and added noise as a function of dataset size. In practice, we found that sampling variation accounted for less than 1% of the total error in typical simulations. REFERENCES [1] John M Abowd, Robert Ashmead, Ryan Cumings-Menon, Simson Garfinkel, Micah Heineck, Christine Heiss, Robert Johns, Daniel Kifer, Philip Leclerc, Ash- win Machanavajjhala, et al. 2022. The 2020 Census Disclosure Avoidance System TopDown Algorithm. arXiv preprint arXiv:2204.08986 (2022). [1] John M Abowd, Robert Ashmead, Ryan Cumings-Menon, Simson Garfinkel, Micah Heineck, Christine Heiss, Robert Johns, Daniel Kifer, Philip Leclerc, Ash- win Machanavajjhala, et al. 2022. The 2020 Census Disclosure Avoidance System TopDown Algorithm. arXiv preprint arXiv:2204.08986 (2022). g p p [2] Arini Balakrishnan and Chloe Schulze. 2005. Code obfuscation literature survey. CS701 Construction of compilers 19 (2005). f p [3] Raef Bassily, Adam Groce, Jonathan Katz, and Adam Smith. 2013. Coupled-worlds privacy: Exploiting adversarial uncertainty in statistical data privacy. In 2013 IEEE 54th Annual Symposium on Foundations of Computer Science. IEEE, 439–448.f [3] Raef Bassily, Adam Groce, Jonathan Katz, and Adam Smith. 2013. Coupled-worlds privacy: Exploiting adversarial uncertainty in statistical data privacy. In 2013 IEEE 54th Annual Symposium on Foundations of Computer Science. IEEE, 439–448.f Figure 3: The total error, including sampling error and added noise, for the entropy script under different systems in log- scale. y p f p [4] Mark Bun and Thomas Steinke. 2016. Concentrated differential privacy: Simpli- fications, extensions, and lower bounds. In Theory of Cryptography Conference. Springer, 635–658. [4] Mark Bun and Thomas Steinke. 2016. Concentrated differential privacy: Simpli- fications, extensions, and lower bounds. In Theory of Cryptography Conference. Springer, 635–658. [5] Rachel Cummings and David Durfee. 2020. Individual sensitivity preprocessing for data privacy. In Proceedings of the Fourteenth Annual ACM-SIAM Symposium on Discrete Algorithms. SIAM, 528–547.f g [6] Damien Desfontaines and Balázs Pejó. 2020. Sok: differential privacies. Proceed- ings on privacy enhancing technologies 2020, 2 (2020), 288–313. g p y g g [7] Richard M Dudley. 2018. Real analysis and probability. CRC Press. y y p y [8] Cynthia Dwork. 2006. Differential Privacy. In Automata, Languages and Pro- gramming, Michele Bugliesi, Bart Preneel, Vladimiro Sassone, and Ingo Wegener (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 1–12. p g g g [9] Cynthia Dwork, Krishnaram Kenthapadi, Frank McSherry, Ilya Mironov, and Moni Naor. 2006. Our data, ourselves: Privacy via distributed noise generation. In Annual international conference on the theory and applications of cryptographic techniques. Springer, 486–503.f q p g [10] Cynthia Dwork and Jing Lei. 2009. Differential privacy and robust statistics. In Proceedings of the forty-first annual ACM symposium on Theory of computing. 371–380. 7.3 Sample Entropy Since the same amount of Laplace noise is added to each dimension of the histogram, the total added noise increases linearly with 𝑓. By contrast, the noise scale used in GUPT is tied to the L1 diameter of the output space, which is proportional to 𝑓. Summing over the 𝑓histogram dimensions, the total added noise is therefore proportional to 𝑓2. E(𝐷) = − 𝑓∑︁ 𝑗=1 𝑝𝑗log2(𝑝𝑗)i ︁ To configure TAHOE, we set 𝛼= 𝜖/5 and 𝛿= 1/(𝑁+1) as before. To make the privacy wrappers as comparable as possible, we again set 𝜆so that TAHOE never returns ⊥. Applying the same argument we used for the proportion script, we must ensure that for any subsets 𝐴, 𝐵⊆𝐷, with size at least 𝑁−2𝑀−1, ||E(𝐴) −E(𝐵)||1 ≤ 𝜆𝛼. While the left hand side can be bounded mathematically, we 427 Proceedings on Privacy Enhancing Technologies 2023(3) Kohli and Laskowski performed a brute force computational search to find the smallest value of 𝜆.i We present our own design for a privacy wrapper, which we call TAHOE. Our algorithm operationalizes two core ideas: find subsets fulfilling a certain stability primitive, and randomize over the subset size. Due to the large number of subsets involved, TAHOE is impractically slow in most scenarios. On the other hand, we consider the special case of a finite data alphabet and describe a set of optimizations that allow TAHOE to run efficiently. With our systems configured this way, we use simulation to measure the error of each system for randomly generated datasets. Each datapoint is drawn from a discrete uniform distribution over D. We found experimentally that varying the size of the alphabet does not significantly alter the results, so we fix the alphabet size at 3. Parameter 𝜖is set to values from {1, 2} and 𝑁is set to values in {104, 105, 106, 107}. Each experiment was replicated 50 times and all experiments were run on a single computer. Performance simulations show that TAHOE’s performance is comparable to GUPT, a benchmark algorithm from the subsample- and-aggregate lineage, with TAHOE displaying better accuracy for some scenarios and parameter values. TAHOE performs relatively well when the dataset is large, when the dimensionality of the output is high, and when 𝜖is not too small. In Figure 3, we plot the total standard error (root-mean-square error) as a function of dataset size, for different parameter values. ACKNOWLEDGMENTS We are grateful for funding from the UC Berkeley Center for Long- Term Cybersecurity. This work was greatly improved by comments from our anonymous reviewers, Adam Smith, and participants in the workshop, Data Privacy: Foundations and Applications Reunion at the Simons Institute for the Theory of Computing. 7.3 Sample Entropy As in the histogram example, we found that GUPT outperforms TAHOE for small datasets. TAHOE does overtake GUPT’s perfor- mance for large enough datasets, but the number of datapoints required is considerably greater than we observed for the normal- ized histogram. Several factors can help explain this result. First, the slope of the entropy function is greatest for unbalanced datasets with entropy near zero. To set the noise scale 𝜆, we must consider subsets containing all of one letter, for which changing 2𝑀+ 1 entries can result in a large change in entropy. TAHOE therefore requires a relatively large amount of noise. Second, entropy is a one-dimensional quantity, so GUPT doesn’t suffer a penalty for each dimension, as we saw in the previous experiment. Finally, because the entropy function has slope close to zero for relatively balanced datasets, and subsets are likely to have proportions close to 𝐷, GUPT introduces very little sampling variation. REFERENCES Figure 3: The total error, including sampling error and added noise, for the entropy script under different systems in log- scale. [11] Cynthia Dwork, Frank McSherry, Kobbi Nissim, and Adam Smith. 2006. Cali- brating noise to sensitivity in private data analysis. In Theory of cryptography conference. Springer, 265–284.f [12] Cynthia Dwork and Aaron Roth. 2014. The algorithmic foundations of differential privacy. Foundations and Trends in Theoretical Computer Science 9, 3-4 (2014), 211–407.f Differential Privacy for Black-Box Statistical Analyses Differential Privacy for Black-Box Statistical Analyses systems. arXiv preprint arXiv:2002.04049 (2020). systems. arXiv preprint arXiv:2002.04049 (2020). y p p [18] Gary King and Nathaniel Persily. 2020. Unprecedented facebook urls dataset now available for academic research through social science one. Social Science One 13 (2020). where the second equality follows because maximizing a sum of functions, each with a different independent variable 𝑢𝑗, is equivalent to maximizing each function in the sum individu- ally. It further follows that where the second equality follows because maximizing a sum of functions, each with a different independent variable 𝑢𝑗, is equivalent to maximizing each function in the sum individu- ally. It further follows that ( ) [19] Ios Kotsogiannis, Yuchao Tao, Xi He, Maryam Fanaeepour, Ashwin Machanava- jjhala, Michael Hay, and Gerome Miklau. 2019. Privatesql: a differentially private sql query engine. Proceedings of the VLDB Endowment 12, 11 (2019), 1371–1384. q q y g g f [20] Joshua Alexander Kroll. 2015. Accountable algorithms. Ph. D. Dissertation. Prince- ton University. [21] Frank D McSherry. 2009. Privacy integrated queries: an extensible platform for privacy-preserving data analysis. In Proceedings of the 2009 ACM SIGMOD International Conference on Management of data. 19–30.f max 𝑋,𝑌∈2𝑆 𝑙 ||𝑅(𝑋) −𝑅(𝑌)||1 = max 𝑋,𝑌∈2𝑆 𝑙 max 𝑢∈𝑈  𝑢· 𝑅(𝑋) −𝑢· 𝑅(𝑌)  = max 𝑢∈𝑈  max 𝑋,𝑌∈2𝑆 𝑙  𝑢· 𝑅(𝑋) −𝑢· 𝑅(𝑌)  = max 𝑢∈𝑈  max 𝐴∈2𝑆 𝑙 𝑢· 𝑅(𝐴) −min 𝐴∈2𝑆 𝑙 𝑢· 𝑅(𝐴)  = max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  [22] Ilya Mironov. 2017. Rényi differential privacy. In 2017 IEEE 30th computer security foundations symposium (CSF). IEEE, 263–275. [23] Prashanth Mohan, Abhradeep Thakurta, Elaine Shi, Dawn Song, and David Culler. 2012. GUPT: privacy preserving data analysis made easy. In Proceedings of the 2012 ACM SIGMOD International Conference on Management of Data. 349–360. f g f [24] Kobbi Nissim, Sofya Raskhodnikova, and Adam Smith. 2007. Smooth sensitivity and sampling in private data analysis. In Proceedings of the thirty-ninth annual ACM symposium on Theory of computing. 75–84.f [25] Hector Page, Charlie Cabot, and Kobbi Nissim. 2018. Differential privacy an introduction for statistical agencies. NSQR. Government Statistical Service (2018), 1–53. [26] Nicolas Papernot, Martín Abadi, Ulfar Erlingsson, Ian Goodfellow, and Kunal Talwar. 2016. Semi-supervised knowledge transfer for deep learning from private training data. arXiv preprint arXiv:1610.05755 (2016). Differential Privacy for Black-Box Statistical Analyses where the second to last equality follows because maximiz- ing a difference of two functions with different independent variables is equivalent to maximizing the first function and minimizing the second one, and the last equality follows by applying the definition of 𝑚𝑎𝑥𝑢(𝑆) and 𝑚𝑖𝑛𝑢(𝑆). □ [27] Henry Gordon Rice. 1953. Classes of recursively enumerable sets and their decision problems. Transactions of the American Mathematical society 74, 2 (1953), 358–366. [28] Kenneth H Rosen. 1999. Discrete mathematics & applications. McGraw-Hill.fif [29] Adam Smith. 2008. Efficient, differentially private point estimators. arXiv preprint arXiv:0809.4794 (2008). [30] Adam Smith. 2011. Privacy-preserving statistical estimation with optimal conver- gence rates. In Proceedings of the forty-third annual ACM symposium on Theory of computing. 813–822.f Now, we prove both directions of Proposition 1. ( =⇒) Suppose 𝑆is stable. If |𝑆| < 𝑙the result follows immedi- ately. Otherwise, |𝑆| ≥𝑙. Then by the definition of stability, 𝑆is responsive and for any 𝑋,𝑌∈2𝑆 𝑙, 𝐿𝑘(𝑅(𝑋), 𝜆) ∼𝛼,0 𝐿𝑘(𝑅(𝑌), 𝜆). By Lemma 2, the last relation can be written as, f p g [31] Salil Vadhan. 2017. The complexity of differential privacy. In Tutorials on the Foundations of Cryptography. Springer, 347–450. [32] Royce J Wilson, Celia Yuxin Zhang, William Lam, Damien Desfontaines, Daniel Simmons-Marengo, and Bryant Gipson. 2019. Differentially private sql with bounded user contribution. arXiv preprint arXiv:1909.01917 (2019). [32] Royce J Wilson, Celia Yuxin Zhang, William Lam, Damien Desfontaines, Daniel Simmons-Marengo, and Bryant Gipson. 2019. Differentially private sql with bounded user contribution. arXiv preprint arXiv:1909.01917 (2019). max 𝑋,𝑌∈2𝑆 𝑙 ||𝑅(𝑋) −𝑅(𝑌)||1 ≤𝛼𝜆 APPENDIX Since 𝑆is responsive of minimum size 𝑙, by Lemma 8, Since 𝑆is responsive of minimum size 𝑙, by Lemma 8, 8 DISCUSSION [13] Cynthia Dwork and Guy N Rothblum. 2016. Concentrated differential privacy. arXiv preprint arXiv:1603.01887 (2016). [14] Madhav Jha and Sofya Raskhodnikova. 2013. Testing and reconstruction of Lipschitz functions with applications to data privacy. SIAM J. Comput. 42, 2 (2013), 700–731.f This study formalizes the notion of a privacy wrapper: an algorithm that can pass data to a researcher script and observe the return values, before returning an output to the user. We believe that this is a useful framework for reasoning about untrusted code. Differential privacy extends naturally to this setting, with the standard proba- bility bound required to hold for every possible script. Moreover, the script is treated as a black box, avoiding the limitations inherent in analyzing code. [15] Noah Johnson, Joseph P Near, and Dawn Song. 2018. Towards practical differential privacy for SQL queries. Proceedings of the VLDB Endowment 11, 5 (2018), 526– 539. [16] James Jordon, Jinsung Yoon, and Mihaela Van Der Schaar. 2018. PATE-GAN: Generating synthetic data with differential privacy guarantees. In International Conference on Learning Representations. [17] Daniel Kifer, Solomon Messing, Aaron Roth, Abhradeep Thakurta, and Danfeng Zhang. 2020. Guidelines for implementing and auditing differentially private 428 Proceedings on Privacy Enhancing Technologies 2023(3) Proof of Proposition 5 Proof. To start, we bound the growth rate of 𝑀as 𝜖,𝛿→0. When 𝜖= 𝑔𝛼for constant 𝑔> 4, 𝑄= 𝜖(𝑔−4)(2𝑔−4)−1. Denote ˆ𝑔= (4𝑔−8)(𝑔−4)−1, so 𝑄= 2𝜖/ ˆ𝑔. Then, 𝑀= & ˆ𝑔 2𝜖ln  𝛿−1 exp(𝜖) ˆ2𝜖 ˆ𝑔+ 1 ' ≤ ˆ𝑔 2𝜖ln  𝛿−1 exp(𝜖) 2𝜖 ˆ𝑔+ 1  + 1 ︁ Subtracting Í𝑟−1 𝑁−𝑀𝐺(𝑛) from both sides, we have,︁ (exp(𝜖−4𝛼) −1) 𝑟−1 ∑︁ 𝑛=𝑁−𝑀 𝐺(𝑛) = 𝐺(𝑟) −𝐺(𝑁−𝑀) ≥𝐺(𝑟) −exp(−2𝛼)𝛿′ Because we are interested in the limit as 𝜖→0, we restrict attention to 𝜖< 𝑒−1. Then 2 exp(𝜖)𝜖/ ˆ𝑔< 2 exp(𝑒−1)𝑒−1/ ˆ𝑔= 2 exp(𝑒−1 −1)/ ˆ𝑔< 2/ ˆ𝑔< 1. Since ln(·) is an increasing function, it follows that Rearranging gives the desired result. □ Rearranging gives the desired result. □ □ Rearranging gives the desired result. □ 𝑀≤ ˆ𝑔 2𝜖ln  𝛿−1 + 1  + 1 Proof of Proposition 1 Proof of Proposition 1 Proof. To ease notation, let 2𝑆 𝑙= {𝐴⊆𝑆: |𝐴| ≥𝑙}. We will use the following lemma in the proof: max 𝑋,𝑌∈2𝑆 𝑙 ||𝑅(𝑋) −𝑅(𝑌)||1 = max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  Lemma 8. If multiset 𝑆∈D★is (𝑅,𝑙)−responsive with |𝑆| ≥𝑙, then Combining these results yields the desired inequality of Combining these results yields the desired inequality of max 𝑋,𝑌∈2𝑆 𝑙 ||𝑅(𝑋) −𝑅(𝑌)||1 = max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  ≤𝛼𝜆i ( ⇐= ) Suppose |𝑆| < 𝑙. Then 𝑆is stable by definition. Suppose |𝑆| ≥𝑙and ( ⇐= ) Su |𝑆| ≥𝑙and Proof. Take any𝑋,𝑌∈2𝑆 𝑙, and consider ||𝑅(𝑋)−𝑅(𝑌)||1 = Proof. Take any𝑋,𝑌∈2𝑆 𝑙, and consider ||𝑅(𝑋)−𝑅(𝑌)||1 = Í𝑘 𝑗=1 |𝑅(𝑋)𝑗−𝑅(𝑌)𝑗|. For all 𝑗∈{1, ...,𝑘}, define 𝑢(𝑋,𝑌) 𝑗 as 1 if 𝑅(𝑋)𝑗≥𝑅(𝑌)𝑗, and −1 otherwise. Then, |𝑅(𝑋)𝑗−𝑅(𝑌)𝑗| = 𝑢(𝑋,𝑌) 𝑗 (𝑅(𝑋)𝑗−𝑅(𝑌)𝑗) for all 𝑗. Even more so, for all 𝑗we have 𝑢(𝑋,𝑌) 𝑗 (𝑅(𝑋)𝑗−𝑅(𝑌)𝑗) = max𝑢𝑗∈{−1,+1} 𝑢𝑗(𝑅(𝑋)𝑗−𝑅(𝑌)𝑗) by construction. Hence,︁ max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  ≤𝛼𝜆 max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  We know 𝑆is responsive, because otherwise, there must be some 𝑋∈2𝑆 𝑙with 𝑅(𝑋) =⊥. But then 𝑚𝑖𝑛𝑢(𝑋) = −∞and 𝑚𝑎𝑥𝑢(𝑋) = ∞ for all 𝑢∈𝑈. This would imply that max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  = ∞> 𝛼𝜆 ||𝑅(𝑋) −𝑅(𝑌)||1 = 𝑘 ∑︁ 𝑗=1 max 𝑢𝑗∈{−1,+1}𝑢𝑗(𝑅(𝑋)𝑗−𝑅(𝑌)𝑗) = max 𝑢∈𝑈 𝑘 ∑︁ 𝑗=1 𝑢𝑗(𝑅(𝑋)𝑗−𝑅(𝑌)𝑗) = max 𝑢∈𝑈  𝑢· 𝑅(𝑋) −𝑢· 𝑅(𝑌)  giving a contradiction. Finally, by Lemma 8 we have max 𝑋,𝑌∈2𝑆 𝑙 ||𝑅(𝑋) −𝑅(𝑌)||1 = max 𝑢∈𝑈  𝑚𝑎𝑥𝑢(𝑆) −𝑚𝑖𝑛𝑢(𝑆)  ≤𝛼𝜆 completing the proof. □ 429 Proceedings on Privacy Enhancing Technologies 2023(3) Kohli and Laskowski Kohli and Laskowski To simplify computations, we replace 2𝛼with 𝜖, which makes the integral smaller. Define ˆ𝛿−1 as Proof of Proposition 3 ≤exp (𝑄−1 ln(𝛿−1 exp(𝜖)𝑄+ 1)𝑄) −1 𝛿 Proof. Note the recurrence: for any 𝑁−𝑀≤𝑛≤𝑁, 𝐺(𝑛+1) ≤ exp(𝜖−4𝛼)𝐺(𝑛). Using this, we see that︁︁ We have shown 𝛿′ < ˆ𝛿≤𝛿, as required. □ We have shown 𝛿′ < ˆ𝛿≤𝛿, as required. □ We have shown 𝛿′ < ˆ𝛿≤𝛿, as required. □ exp(𝜖−4𝛼) 𝑟−1 ∑︁ 𝑛=𝑁−𝑀 𝐺(𝑛) = 𝑟−1 ∑︁ 𝑛=𝑁−𝑀 exp(𝜖−4𝛼)𝐺(𝑛) ≥ 𝑟−1 ∑︁ 𝑛=𝑁−𝑀 𝐺(𝑛+ 1) = 𝑟∑︁ 𝑛=𝑁−𝑀+1 𝐺(𝑛) Proof of Proposition 2 Proof. To prove (a), observe that when |𝑆| = 𝑙, the set 𝑆it- self is the only subset of size at least 𝑙. Hence, when 𝑅(𝑆) ≠⊥, 𝑚𝑖𝑛𝑢(𝑆) = min𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) = 𝑢· 𝑅(𝑆) = max𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) = 𝑚𝑎𝑥𝑢(𝑆). In the case where 𝑅(𝑆) =⊥, then 𝑆is not (𝑅,𝑙)- responsive, hence 𝑚𝑖𝑛𝑢(𝑆) = −∞and 𝑚𝑎𝑥𝑢(𝑆) = ∞by definition.i 𝑁 ∫ 𝑁−𝑀−1 exp min n (𝜖−4𝛼)(𝑛−𝑁+ 𝑀) −𝜖,𝜖(𝑁−𝑛−1) o! 𝑑𝑛 Then ˆ𝛿−1 < (𝛿′)−1. We set the two arguments to the min function equal to one another to find the intersection point 𝑛★= 𝑁−𝑀(𝜖− 4𝛼)(2𝜖−4𝛼)−1. Then splitting the integral into two integrals over (𝑁−𝑀−1,𝑛★) and (𝑛★, 𝑁) and evaluating them yields To prove (b), we first note that for any 𝐴⊆𝑆with |𝐴| ≥𝑙, either 𝐴= 𝑆or 𝐴⊆𝑐for some 𝑐∈𝐶(𝑆). Hence, we can write {𝐴⊆𝑆: |𝐴| ≥𝑙} = {𝑆} ∪ Ø 𝑐∈𝐶(𝑆) {𝐴⊆𝑐: |𝐴| ≥𝑙} {𝐴⊆𝑆: |𝐴| ≥𝑙} = {𝑆} ∪ Ø 𝑐∈𝐶(𝑆) {𝐴⊆𝑐: |𝐴| ≥𝑙} Then, when 𝑅(𝑆) ≠⊥, 𝑚𝑖𝑛𝑢(𝑆) = min 𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) = min  𝑢· 𝑅(𝑆), min 𝑐∈𝐶(𝑆) min 𝐴⊆𝑐,|𝐴|≥𝑙𝑢· 𝑅(𝐴)  = min  𝑢· 𝑅(𝑆), min 𝑐∈𝐶(𝑆) 𝑚𝑖𝑛𝑢(𝑐) exp(𝜖) ˆ𝛿−1 = exp (𝑀𝑄) 𝑄 −  exp(−(𝜖−4𝛼)) 𝜖−4𝛼 −1 𝜖  > exp (𝑀𝑄) 𝑄 −  1 𝜖−4𝛼−1 𝜖  = exp (𝑀𝑄) −1 𝑄i Then, when 𝑅(𝑆) ≠⊥, 𝑚𝑖𝑛𝑢(𝑆) = min 𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) 𝑚𝑖𝑛𝑢(𝑆) = min 𝐴⊆𝑆,|𝐴|≥𝑙𝑢· 𝑅(𝐴) = min  𝑢· 𝑅(𝑆), min 𝑐∈𝐶(𝑆) min 𝐴⊆𝑐,|𝐴|≥𝑙𝑢· 𝑅(𝐴)  = min  𝑢· 𝑅(𝑆), min 𝑐∈𝐶(𝑆) 𝑚𝑖𝑛𝑢(𝑐) where the first inequality follows from 𝜖> 4𝛼. By construction in TAHOE, 𝑀≥𝑄−1 ln(𝛿−1 exp(𝜖)𝑄+ 1). Hence, In the case where 𝑅(𝑆) =⊥, then 𝑆is not (𝑅,𝑙)-responsive, hence 𝑚𝑖𝑛𝑢(𝑆) = −∞by definition. The proof for 𝑚𝑎𝑥𝑢(𝑆) follows by a similar argument. □ ˆ𝛿= exp(𝜖)𝑄 exp (𝑀𝑄) −1 ≤ exp(𝜖)𝑄 exp (𝑄−1 ln(𝛿−1 exp(𝜖)𝑄+ 1)𝑄) −1 = 𝛿 ˆ𝛿= exp(𝜖)𝑄 exp (𝑀𝑄) −1 Proof of Lemma 5 Using standard analysis, ln(𝐴+ 1) ≤ln(𝐴) + 1 whenever 𝐴≥1. Since 𝛿≤1,𝛿−1 ≥1, so Proof. The privacy parameter 𝛿′ in the distribution 𝐺can be computed precisely, as (𝛿′)−1 equals︁ Since 𝛿≤1,𝛿−1 ≥1, so 𝑀≤ ˆ𝑔 2𝜖ln  𝛿−1  + ˆ𝑔 2𝜖+ 1 = ˆ𝑔 2𝜖 1 −ln(𝛿) + 1 𝑁 ∑︁ 𝑛=𝑁−𝑀 exp min  (𝜖−4𝛼)(𝑛−𝑁+ 𝑀) −2𝛼,𝜖(𝑁−𝑛) ! In the worst-case, TAHOE samples 𝑛= 𝑁during Step 1. TAHOE is then required to check stability for all subsets of 𝐷of sizes be- tween 𝑁−2𝑀−1 and 𝑁. The researcher’s script 𝑅must run on Í2𝑀+1 𝑗=0 𝑁 𝑗  subsets in Step 2. For each subset𝑆in Step 2, we compute 𝑚𝑖𝑛𝑢(𝑆) and 𝑚𝑎𝑥𝑢(𝑆) for each 𝑢∈𝑈where |𝑈| = 2𝑘. Computing each of these quantities requires taking a minimum and a maximum ︁ By standard analysis arguments, we can lower bound the sum with the following integral. 𝑁 ∫ 𝑁−𝑀−1 exp min n (𝜖−4𝛼)(𝑛−𝑁+ 𝑀) −2𝛼,𝜖(𝑁−𝑛−1) o! 𝑑𝑛 430 Proceedings on Privacy Enhancing Technologies 2023(3) Differential Privacy for Black-Box Statistical Analyses 2(b), which is bounded by 𝑓. Hence, a big-O bound on Step 2 is over 𝐶(𝑆) from Proposition 2(b), which is bounded by 𝑁. Hence, the big-O runtime of Step 2 is Í2𝑀+1 𝑗=0 𝑁 𝑗 (𝜏+ 𝑁2𝑘). Since over 𝐶(𝑆) from Proposition 2(b), which is bounded by 𝑁. Hence, the big-O runtime of Step 2 is Í2𝑀+1 𝑗=0 𝑁 𝑗 (𝜏+ 𝑁2𝑘). Since 2𝑀+ 1 + 𝑓 𝑓  (𝜏+ 𝑓2𝑘) = (2𝑀+ 1 + 𝑓)...(2𝑀+ 2) 𝑓! (𝜏+ 𝑓2𝑘) ≤(2𝑀+ 1 + 𝑓)...(2𝑀+ 2) (𝑓−1)! (𝜏+ 2𝑘) ≤(2𝑀+ 1 + 𝑓)𝑓(𝜏+ 2𝑘) 2𝑀+1 ∑︁ 𝑗=0 𝑁 𝑗  ≤(𝑁+ 1)2𝑀+1 Since 𝑘and 𝑓are fixed by the proposition, the bound is O(𝜏𝑀𝑓). 𝑀 𝑓 Since 𝑘and 𝑓are fixed by the proposition, the bound is O(𝜏𝑀𝑓). 2𝑀1 𝑓 and 𝑘is fixed by the proposition, the runtime is and 𝑘is fixed by the proposition, the runtime is In Step 3, TAHOE computes on at most 𝑓 𝑓  subsets and on 2𝑘vectors 𝑢∈𝑈, which is less than O(𝜏𝑀𝑓). It can be checked that Steps 1, 4, and 5 require fewer operations than Step 2, and so they do not alter the big-O. Plugging in the upper bound for 𝑀, the runtime is bounded by O(𝜏+ 𝑁)(𝑁+ 1)2𝑀+1 In Step 3, TAHOE computes on at most 𝑁 𝑁−2𝑀−1  subsets and on 2𝑘vectors 𝑢∈𝑈, which is less than O(𝜏+ 𝑁)(𝑁+ 1)2𝑀+1. Proof of Lemma 5 It can be checked that Steps 1, 4, and 5 require fewer operations than Step 2, and so they do not alter the big-O. Plugging in the upper bound for 𝑀, the runtime is bounded by O  𝜏  ˆ𝑔 2𝜖ln(𝛿−1) + ˆ𝑔 2𝜖+ 1 𝑓 ⊆O𝜏𝜖−𝑓ln(𝛿−1)𝑓 □ O  (𝜏+ 𝑁)(𝑁+ 1) ˆ𝑔𝜖−1(1−ln(𝛿))+3  Proof of Lemma 7 Let 𝑓𝑛represent the distribution of 𝑅(𝑫𝑛). Since the data is drawn independently from D with distribution F, conditional on 𝑛, 𝑆has the distribution F𝑛, and so 𝑅(𝑺) has the distribution 𝑓𝑛. Proof. We start by proving (a). The number of histograms of size 𝑁−𝑖is equal to the number of integer solutions to Í𝑓 𝑗=1 𝑤𝑗= 𝑖 where each 0 ≤𝑤𝑗≤𝑣𝑗. This is never exceeds the number of integer solutions to Í𝑓 𝑗=1 𝑤𝑗= 𝑖where each 0 ≤𝑤𝑗, dropping the upper bound on each 𝑤𝑗. By “stars and bars” [28] there are 𝑖+𝑓−1 𝑓−1  solutions. 𝑓 Therefore, the distribution of 𝑅(𝑺) is Í𝑁 𝑛=𝑁−𝑀𝐺(𝑛)𝑓𝑛. Given an open set 𝐵⊆R𝑘∪{⊥} containing 𝜃, and probability 𝑝< 1, by consistency of 𝑅, there exists 𝑚such that for any 𝑛> 𝑚, 𝑓𝑛(𝐵) > 𝑝. Since lim𝑁→∞𝑁−𝑀= ∞, we can choose 𝑁such that 𝑁−𝑀> 𝑚, so︁︁︁ Next, we prove (b). The number of histograms of size at least 𝑁−2𝑀−1 is equal to the number of integer solutions to Í𝑓 𝑗=1 𝑤𝑗≤ 2𝑀+ 1 where each 0 ≤𝑤𝑗≤𝑣𝑗. This is never exceeds the number of integer solutions to Í𝑓 𝑗=1 𝑤𝑗≤2𝑀+ 1 where each 0 ≤𝑤𝑗, dropping the upper bound on each 𝑤𝑗. Introduce a new variable 𝑤𝑓+1 ≥0. Then the number of non-negative integer solutions to Í𝑓 𝑗=1 𝑤𝑗≤2𝑀+ 1 is the same as the number of non-negative integer solutions to Í𝑓+1 𝑗=1 𝑤𝑗= 2𝑀+ 1. By “stars and bars” there are 2𝑀+1+𝑓 𝑓  solutions. □ 𝑁 ∑︁ 𝑛=𝑁−𝑀 𝐺(𝑛)𝑓𝑛(𝐵) > 𝑁 ∑︁ 𝑛=𝑁−𝑀 𝐺(𝑛)𝑝= 𝑝 𝑁 ∑︁ 𝑛=𝑁−𝑀 𝐺(𝑛) = 𝑝 ︁︁︁ Therefore 𝑅(𝑺) converges in probability to 𝜃. Since 𝜆→0 as 𝑁→ ∞, 𝐿converges in probability to 0. And since addition is continuous, by the continuous mapping theorem, ︁︁︁ Therefore 𝑅(𝑺) converges in probability to 𝜃. Since 𝜆→0 as 𝑁→ ∞, 𝐿converges in probability to 0. And since addition is continuous, by the continuous mapping theorem, plim𝑁→∞  𝑅(𝑺) + 𝐿  = plim𝑁→∞𝑅(𝑺) + plim𝑁→∞𝐿= 𝜃 □ □ Proof of Theorem 2 Proof. Let random variable 𝑺be the subset chosen by TAHOE, and let random variable 𝐿∼𝐿𝑘(0, 𝜆) be the Laplace noise added by TAHOE. Since TAHOE is configured to never halt, the output of TAHOE is then the random variable 𝑅(𝑺) + 𝐿. □ Proof of Proposition 6 Proof. By the proof of Proposition 5 𝑀≤ ˆ𝑔 2𝜖ln(𝛿−1) + ˆ𝑔 2𝜖+ 1 where ˆ𝑔= (4𝑔−8)(𝑔−4)−1. Next, we examine the runtime of TAHOE as a function of 𝑀. In the worst-case when |D| = 𝑓, TAHOE is required to run 𝑅on at most 2𝑀+1+𝑓 𝑓  subsets during Step 2. For each subset 𝑆in Step 2, we compute 𝑚𝑖𝑛𝑢(𝑆) and 𝑚𝑎𝑥𝑢(𝑆) for each 𝑢∈𝑈where |𝑈| = 2𝑘. Computing each of these quantities requires taking a minimum and a maximum over 𝐶(𝑆) from Proposition where ˆ𝑔= (4𝑔−8)(𝑔−4)−1. Next, we examine the runtime of TAHOE as a function of 𝑀. In the worst-case when |D| = 𝑓, TAHOE is required to run 𝑅on at most 2𝑀+1+𝑓 𝑓  subsets during Step 2. For each subset 𝑆in Step 2, we compute 𝑚𝑖𝑛𝑢(𝑆) and 𝑚𝑎𝑥𝑢(𝑆) for each 𝑢∈𝑈where |𝑈| = 2𝑘. Computing each of these quantities requires taking a minimum and a maximum over 𝐶(𝑆) from Proposition 431
https://openalex.org/W2775079125
https://bmcresnotes.biomedcentral.com/track/pdf/10.1186/s13104-017-3029-3
English
null
Development of a Saw Bones Model for training pedicle screw placement in scoliosis
BMC research notes
2,017
cc-by
3,663
© The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Objective:  The purpose of this study was to determine if a Sawbones Scoliosis Model could be used as a simulator to train residents in placing pedicle screws—a complex procedure with a steep learning curve. Surgical simulation, a common tool teaching residents complex procedures in a safe environment, was staged using a Sawbones Scoliosis Model. Ten junior and ten senior residents out of 25 total possible residents (80%) were instructed how to place pedi- cle screws using the free-hand technique. They were then asked to place them unilaterally from T4 to L4 and were assessed on completion time, accuracy placement accuracy, and overall competency using an objective rating scale. Results:  Senior residents had an average time to completion of 38.9 ± 4.7 min vs. junior’s 50.1 ± 11.7 min, and a pedicle screw accuracy of 43.6 ± 6.4% vs. junior’s 44.4 ± 17.4%. Overall competency scores were similar for both groups; however, senior residents scored higher in the time and motion subdomain. Senior residents had a faster completion time and were more efficient, suggesting greater experience in spine surgery. The low rate of screw accu- racy in both groups validates that simulation is a safe way for trainees to learn complex tasks. Keywords:  Saw bones, Surgical simulation, Scoliosis, Pedicle screw Simulation as an educational tool for complex proce- dures has been used in other industries, such as airlines, with much success. Only recently has surgical simula- tion been developed to train surgical residents. In ortho- paedics, simulators have been described in arthroscopy, fracture surgery and joint arthroplasty [11–14]. Pedicle screw simulation has been described before, but not with a Scoliosis Model [15, 16]. Idiopathic scoliosis is associ- ated with dysplastic pedicle morphology and a complex three-dimensional deformity that makes pedicle screw placement more challenging [17, 18].h Introduction Complex surgical procedures are often described as having a learning curve where technique improves with experience. One such procedure is placement of pedicle screws in the treatment of scoliosis [1–5]. With knowl- edge of spinal anatomy and surface landmarks, pedicle screws can be placed safely using the free hand technique described by Lenke [6, 7]. The rate of screw misplace- ment has been characterized as 14–25% [8, 9]. Screw malposition has the potential for catastrophic neuro- logic and vascular complications [10]. While placement of pedicle screws remains a complex procedure best left to experienced surgeons, orthopaedic surgical residents must learn this technique during their training. The purpose of this study is to validate a Scoliosis Training Model as a simulation technique in training res- idents how to place pedicle screws. Our hypothesis is that senior residents will have a faster completion time, pedi- cle screw accuracy rate, and higher score on an objective rating scale compared to junior residents. Historically, surgical education included trainees per- forming procedures on real patients while attending sur- geons supervised. With the ever-increasing complexity of procedures, this method of training may not prove safe or provide maximal educational benefit. Development of a Saw Bones Model for training pedicle screw placement in scoliosis Gregory Tanner, Saman Vojdani*  , David E. Komatsu and James M. Barsi © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. *Correspondence: saman.vojdani@stonybrookmedicine.edu Stony Brook University Hospital, Stony Brook, USA Tanner et al. BMC Res Notes (2017) 10:678 https://doi.org/10.1186/s13104-017-3029-3 Tanner et al. BMC Res Notes (2017) 10:678 https://doi.org/10.1186/s13104-017-3029-3 BMC Research Notes Open Access Methodsh This project was reviewed by the institutional review board and determined to be exempt. *Correspondence: saman.vojdani@stonybrookmedicine.edu Stony Brook University Hospital, Stony Brook, USA *Correspondence: saman.vojdani@stonybrookmedicine.edu Stony Brook University Hospital, Stony Brook, USA Tanner et al. BMC Res Notes (2017) 10:678 Page 2 of 5 The model screws (Medtronic, Memphis, TN) were laid out next to the model. Subjects were timed from start to task comple- tion. Videos of the simulation were delivered to a blinded, experienced pediatric spinal deformity surgeon for review. A checklist of steps and global rating scale (Table 1) was developed for this exercise. To prevent bias, the videos were cropped so as only to only show the model and the hands of the subject. The global scale used is a modified version previously validated in the literature [19–21]. A Scoliosis Spine Model (Model 1323021; Sawbones, Vashon, Washington) comprising of T1 to the sacrum was used for the simulation. This model has a 55° right thoracic curve with an associated 25° rotation. A scoliosis spine holder (Model 1703-100; Sawbones, Vashon, Wash- ington) was used to secure the model during instrument insertion. To simulate surrounding musculature, the model was supported in the holder by polystyrene foam peanuts and the model secured in place up to the level of the transverse processes. Peanuts were also inserted between the spinous processes and vertebral foramen in order to prevent direct visual assistance of insertion of the pedicle or spinal cord. After insertion of screw from T4 to L4, each model was photographed before removing the implants. Each screw was evaluated for accuracy, and misplaced screws were recorded as to their location. Average screw length and diameter were obtained. The final results of screw placement were recorded and shared with the blinded reviewer. Subjects and simulation Ten senior (fourth and fifth year) and ten junior (first, second, or third year) residents from a US orthopaedic residency program, out of a possible 25 total residents (80%), were recruited for this study. Participation in the study was voluntary. Statistical analyses F h i i l For the statistical comparisons, data from post-graduate year (PGY) 4 and 5 residents were grouped as seniors, and data from PGY1, 2, and 3 residents were grouped as juniors. The data were then presented as group mean  ±  standard deviation. An f test was performed using Excel (Microsoft, Redmond, WA) to assess data normality with multiple outcomes found to be non- normally distributed. As such, significant differences between groups were assessed using Mann–Whitney tests in SPSS (Version 19, SAS Institute, Cary, NC). For all tests, p values less than 0.05 were considered significant. Prior to beginning the simulation, the residents were given verbal instructions describing the simulation by a fellowship trained orthopaedic surgeon. The free hand technique for placement of pedicle screws was explained in detail. The subjects were randomized either the right or left side of the spine. Pedicle screws were placed from T4 to L4, and each participant was instructed to place one screw at a time. The hand dominance of the sub- jects was recorded. All required instruments and pedicle Table 1  Objective rating scale Grading rubric for steps and global rating scale for resident performance 1 2 3 4 5 Preparation Did not organize, has to stop procedure frequently to prepare equipment Equipment generally organized All equipment neatly prepared and ready for use 1 2 3 4 5 Time and motion Many unnecessary movements Efficient time and motion Clear economy of movement and maximum efficiency 1 2 3 4 5 Instrument handling Repeatedly makes tentative or awkward moves with instru- ments Competent use of instruments, occasionally appeared stiff and awkward Fluid moves with instruments and no awkwardness 1 2 3 4 5 Flow of procedure Frequently stopped procedure and seemed unsure of next moves Demonstrated some forward plan- ning with reasonable progression of procedure Obviously planned course of proce- dure with effortless flow from one move to the next 1 2 3 4 5 Knowledge of procedure Deficient knowledge Knew all important steps Demonstrated familiarity with all aspects of the procedure 1 2 3 4 5 Overall performance Very poor Competent Clearly superior Tanner et al. BMC Res Notes (2017) 10:678 Page 3 of 5 Table 2  Scores on objective global exam Compiled scores for residents based on rubric. This table presents the mean ± standard deviation for each of the five evaluated surgical skills demonstrated by the junior and senior residents. Results Completion time was 50.1  ±  11.7 min for junior resi- dents and 38.9 ± 4.7 min for senior residents, revealing a significant increase in speed for the senior residents (p  =  0.018) (Fig. 1a). Screw accuracy defined as the pedicle screw being completely within the pedicle was 44.4  ±  17.4% for junior residents and 43.6  ±  6.4% for senior residents, demonstrating no significant differences in accuracy (p = 0.847) (Fig. 1b). The objective global rat- ing scale showed no overall difference between junior and senior residents; however, there was a statistically significant difference in the time and motion subdomain (Table 2). Statistical analyses F h i i l p values obtained from Mann– Whitney tests comparing the groups are presented in the last column with significant values (i.e., p > 0.05) highlighted in italics Junior residents Senior residents p value Preparation 3.29 ± 0.49 3.43 ± 0.53 0.591 Time and motion 2.71 ± 0.49 3.71 ± 0.95 0.030 Instrument handling 2.86 ± 0.90 3.14 ± 0.69 0.496 Flow 3.29 ± 0.49 3.57 ± 0.53 0.298 Knowledge 3.29 ± 0.49 3.71 ± 0.76 0.244 Overall 3.29 ± 0.49 3.57 ± 0.53 0.298 Table 2  Scores on objective global exam Discussion procedure even in experienced hands. In order to teach this skill in a safe setting that had no associated risk to real patients, a Scoliosis Model was tested. A common dilemma among those who educate residents is how to impart procedural knowledge to the trainee. Historically, resident education used an apprenticeship model. With the increasing complexity of procedures and breadth of knowledge required of residents during their surgical education, educational tools have been devel- oped to enhance the learning experience. As we expected, senior residents, who have had more surgical experience and training, had a faster completion time for the task. On the objective rating scale, senior residents had more economy of motion compared to less experienced trainees.hi Simulation of the patient encounter has been the standard in medical student education. Only recently has simulation of complex procedures been developed. Simu- lation as an educational tool has been given paramount importance as the American Board of Orthopaedic Sur- gery in concert with the Accreditation Council for Grad- uate Medical Education mandated the development of a surgical skills curriculum for orthopaedic interns. The finding of low screw accuracy rate in both groups suggests that placement of pedicle screws in scoliosis remains an extremely challenging task. Most ortho- paedic surgeons who place pedicle screws in practice have completed advanced fellowship training. Given the high malposition rate and the associated risk of injury of misplacement, the effectiveness of simulation of this technique becomes evident. Lieberman et  al. showed that robotic guidance was able to improve the accuracy of pedicle screw placement in a cadaveric study of nor- mal spines [16], but our simulation may be able to offer similar improvements without the costs associated with guidance systems. Similarly, Luciano et  al., studied the With increasing complexity of a procedure, so increases the corresponding learning curve. Pedicle screw place- ment in idiopathic scoliosis is one such complex proce- dure. The combination of a three dimensional deformity along with dysplastic anatomy make this a difficult Fig. 1  a Completion time in minutes for junior and senior residents. b Screw accuracy for junior and senior residents. Graphic illustrating compara- tive group findings Tanner et al. BMC Res Notes (2017) 10:678 Page 4 of 5 Availability of data and materials ability of a haptic feedback system to improve pedicle screw placement [15], but did not find a significant effect for their training. Limitationsh This study was funded by the OMeGA Medical Grants Association. The funds were used to purchase the models used in the study. OMeGA had no role in study design, data collection, data analysis, data interpretation, or drafting of the manuscript. The limitations of this study include selection bias since it was limited to trainees at a single institution. Residency programs are unique in their curriculum development. Including institutions with more emphasize on orthopae- dic spine surgery might have altered the results. Another bias is the very nature of simulation. Even though attempts were made to make this simulation feel as real as possible, nothing can substitute for the operative expe- rience on live patients. Finally, we did not have prelimi- nary data to run a power analysis to establish a proper sample size. Due to these limitations, other institutions should weight the merits of our simulation against others before adopting any specific training program. However, it is our intent to further develop this training program as validate protocol for improving the accuracy of pedicle screw placement are needed. Further studies will confirm if the skills learned in this exercise are transferable to real clinical scenarios. Abbreviations US: United States; T4: 4th thoracic vertebrae; L4: 4th lumbar vertebrae; PGY: post-graduate year; SPSS: Statistical Package for the Social Sciences; Min: minutes. 9. Liljenqvist UR, Halm HF, Link TM. Pedicle screw instrumentation of the thoracic spine in idiopathic scoliosis. Spine. 1997;22(19):2239–45. 9. Liljenqvist UR, Halm HF, Link TM. Pedicle screw instrumentation of the thoracic spine in idiopathic scoliosis. Spine. 1997;22(19):2239–45. 10. Brown CA, Lenke LG, Bridwell KH, et al. Complications of pediatric thora- columbar and lumbar pedicle screws. Spine. 1998;23(14):1566–71. 10. Brown CA, Lenke LG, Bridwell KH, et al. Complications of pediatric thora- columbar and lumbar pedicle screws. Spine. 1998;23(14):1566–71. Consent for publication Not applicable. Ethics approval and consent to participate This project was reviewed by the Stony Brook University Hospital Institutional Review Board and determined to be exempt as per 41315. All volunteer par- ticipants provided verbal consent to participation. The committee approved this verbal consent procedure. Authors’ contributions 11. Van Heest A, Putnam M, Agel J, et al. Assessment of technical skills of orthopaedic surgery residents performing open carpal tunnel release surgery. J Bone Joint Surg Am. 2009;91(12):281–7. 11. Van Heest A, Putnam M, Agel J, et al. Assessment of technical skills of orthopaedic surgery residents performing open carpal tunnel release surgery. J Bone Joint Surg Am. 2009;91(12):281–7. GT and SV facilitated testing. JB reviewed and ranked resident performances. DK analyzed data. All authors read and approved the final manuscript. GT and SV facilitated testing. JB reviewed and ranked resident performances. DK analyzed data. All authors read and approved the final manuscript. 12. Insel A, Carofino B, Leger R, et al. The development of an objective model to assess arthroscopic performance. J Bone Joint Surg Am. 2009;91(9):2287–95. 12. Insel A, Carofino B, Leger R, et al. The development of an objective model to assess arthroscopic performance. J Bone Joint Surg Am. 2009;91(9):2287–95. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Received: 4 August 2017 Accepted: 29 November 2017 Received: 4 August 2017 Accepted: 29 November 2017 References 1. Lonner BS, Auerbach JD, Estreicher MB, Kean KE. Thoracic pedicle screw instrumentation: the learning curve and evolution in the technique in the treatment of adolescent idiopathic scoliosis. Spine. 2009;34(20):2158–64. 1. Lonner BS, Auerbach JD, Estreicher MB, Kean KE. Thoracic pedicle screw instrumentation: the learning curve and evolution in the technique in the treatment of adolescent idiopathic scoliosis. Spine. 2009;34(20):2158–64. p p 2. Bergeson RK, Schwend RM, DeLucia T, et al. How accurately do novice surgeons place thoracic pedicle screws with the free hand technique? Spine. 2008;33(15):E501–7. 3. Samdani AF, Ranade A, Sciubba DM, et al. Accuracy of free-hand place- ment of thoracic pedicle screws in adolescent idiopathic scoliosis: how much of a difference does surgeon experience make? Eur Spine J. 2010;19(1):91–5. In conclusion, a Sawbones Scoliosis Model can be used as an educational tool in teaching orthopaedic residents how to perform the complex task of placing pedicle screws. Senior residents, who had more surgical experi- ence, scored higher on the objective measures of comple- tion time and economy of motion subdomain on a global rating scale. However, given the limitations of this study, other residency programs will have to decide if this simu- lation is appropriate for their training goals. 4. Gang C, Haibo L, Fancai L, et al. Learning curve of thoracic pedicle screw placement using the free-hand technique in scoliosis: how many screws needed for an apprentice? Eur Spine J. 2012;21(6):1151–6. 5. Gonzalvo A, Fitt G, Liew S, et al. The learning curve of pedicle screw place- ment: how many screws are enough? Spine. 2009;34(21):E761–5. 6. Kim YJ, Lenke LG, Bridwell KH, et al. Free hand pedicle screw placement in the thoracic spine: is it safe? Spine. 2004;29(3):333–42. 7. Suk SI, Lee CK, Kim WJ, et al. Segmental pedicle screw fixation in the treat- ment of thoracic idiopathic scoliosis. Spine. 1995;20(12):1399–405. 8. Abul-Kasim K, Ohlin A. The rate of screw misplacement in segmental pedicle screw fixation in adolescent idiopathic scoliosis. Acta Orthop. 2011;82(1):50–5. Discussion Combined with our study, these stud- ies show there are many options for instructing residents in pedicle screw placement in a safe environment with- out any associated patient risk. Moreover, using the tools outlined in our study, the experienced surgeon can use objective measures to assess the trainee in any of these modalities. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Consent for publication Not applicable. Acknowledgements 13. Yehyawi TM, Thomas TP, Ohrt GT, et al. A simulation trainer for complex articular fracture surgery. J Bone J Surg Am. 2013;95(e92):1–8. 13. Yehyawi TM, Thomas TP, Ohrt GT, et al. A simulation trainer for complex articular fracture surgery. J Bone J Surg Am. 2013;95(e92):1–8. The authors would like to thank Mike Mazella of MedTronic for supplying the utensils for the study. 14. Podolsky DJ, Martin AR, Whyne CM, et al. Exploring the role of 3-dimen- sional simulation in surgical training: feedback from a pilot study. J Spinal Disord Tech. 2010;23(8):e70–4. 14. Podolsky DJ, Martin AR, Whyne CM, et al. Exploring the role of 3-dimen- sional simulation in surgical training: feedback from a pilot study. J Spinal Disord Tech. 2010;23(8):e70–4. 10. Brown CA, Lenke LG, Bridwell KH, et al. Complications of pediatric thora- columbar and lumbar pedicle screws. Spine. 1998;23(14):1566–71. 15. Luciano CJ, Banerjee PP, Bellotte B. Learning retention of thoracic pedicle screw placement using a high-resolution augmented reality simulator with haptic feedback. Neurosurgery. 2011;69(1 Suppl Operative):ons14-9. scoliosis: is pedicle fixation an anatomically viable technique? Spine. 2000;25(18):2285–93. 19. Reznick RK, MacRae H. Teaching surgical skills: changes in the wind. N Engl J Med. 2006;355(25):2664–9. 20. Winckle C, Reznick RK, Cohen R, Taylor B. Reliability and construct validity of a structured technical skills assessment form. Am J Surg. 1994;167:423–7. 18. O’Brien MF, Lenke LG, Mardjetko S, Lowe TG, Kong Y, Eck K, Smith D. Pedicle morphology in thoracic adolescent idiopathic 16. Lieberman IH, Hardenbrook MA, Wang JC, Guyer RD. Assessment of pedicle screw placement accuracy, procedure time, and radiation expo- sure using a miniature robotic guidance system. J Spinal Disord Tech. 2012;25(5):241–8. 21. Martin JA, Regehr G, Reznick R, Macrae H, Murnaghan J, Hutchison C, Brown M. Objective structured assessment of technical skill (OSATS) for surgical residents. Br J Surg. 1997;84:273–8. 17. Berryman F, Pynsent P, Fairbank J, Disney S. A new system for measuring three-dimensional back shape in scoliosis. Eur Spine J. 2008;17(5):663–72. Tanner et al. BMC Res Notes (2017) 10:678 Competing interests Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. Tanner et al. BMC Res Notes (2017) 10:678 Page 5 of 5 • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: 15. Luciano CJ, Banerjee PP, Bellotte B. Learning retention of thoracic pedicle screw placement using a high-resolution augmented reality simulator with haptic feedback. Neurosurgery. 2011;69(1 Suppl Operative):ons14-9. 16. Lieberman IH, Hardenbrook MA, Wang JC, Guyer RD. Assessment of pedicle screw placement accuracy, procedure time, and radiation expo- sure using a miniature robotic guidance system. J Spinal Disord Tech. 2012;25(5):241–8. 17. Berryman F, Pynsent P, Fairbank J, Disney S. A new system for measuring three-dimensional back shape in scoliosis. Eur Spine J. 2008;17(5):663–72. 18. O’Brien MF, Lenke LG, Mardjetko S, Lowe TG, Kong Y, Eck K, Smith D. Pedicle morphology in thoracic adolescent idiopathic scoliosis: is pedicle fixation an anatomically viable technique? Spine. 2000;25(18):2285–93. 19. Reznick RK, MacRae H. Teaching surgical skills: changes in the wind. N Engl J Med. 2006;355(25):2664–9. 20. Winckle C, Reznick RK, Cohen R, Taylor B. Reliability and construct validity of a structured technical skills assessment form. Am J Surg. 1994;167:423–7. 21. Martin JA, Regehr G, Reznick R, Macrae H, Murnaghan J, Hutchison C, Brown M. Objective structured assessment of technical skill (OSATS) for surgical residents. Br J Surg. 1997;84:273–8.
https://openalex.org/W2971296819
https://ttu-ir.tdl.org/bitstream/2346/92257/1/Main%20article%20with%20TTU%20Libraries%20cover%20page.pdf
English
null
Suitability of Scoria as Fine Aggregate and Its Effect on the Properties of Concrete
Sustainability
2,019
cc-by
7,232
Received: 29 July 2019; Accepted: 22 August 2019; Published: 26 August 2019 Received: 29 July 2019; Accepted: 22 August 2019; Published: 26 August 2019 Abstract: The increase in the demand for concrete production for the development of infrastructures in developing countries like Ethiopia leads to the depletion of virgin aggregates and high cement demand, which imposes negative environmental impacts. In sustainable development, there is a need for construction materials to focus on the economy, efficient energy utilization, and environmental protections. One of the strategies in green concrete production is the use of locally available construction materials. Scoria is widely available around the central towns of Ethiopia, especially around the rift valley regions where huge construction activities are taking place. The aim of this paper is therefore to analyze the suitability of scoria as a fine aggregate for concrete production and its effect on the properties of concrete. A differing ratio of scoria was considered as a partial replacement of fine aggregate with river sand after analyzing its engineering properties, and its effect on the mechanical properties of concrete were examined. The test results on the engineering properties of scoria revealed that the material is suitable to be used as a fine aggregate in concrete production. The replacement of scoria with river sand also enhanced the mechanical strength of the concrete. Generally, the findings of the experimental study showed that scoria could replace river sand by up to 50% for conventional concrete production. Keywords: green concrete; scoria; river sand; virgin aggregate; sustainable development Sustainability 2019, 11, 4647; doi:10.3390/su11174647 Suitability of scoria as fine aggregate and its effect on the properties of concrete The Texas Tech community has made this publication openly available. Please share how this access benefits you. Your story matter to us. Citation Warati, G.K., Darwish, M.M., Feyessa, F.F., & Ghebrab, T.. 2019. Suitability of scoria as fine aggregate and its effect on the properties of concrete. Sustainability (17). https://doi.org/10.3390/su11174647 Citable Link https://hdl.handle.net/2346/92257 Terms of Use © 2019 by the authors. cc-by Title page template design credit to Harvard DASH Title page template design credit to Harvard DASH. sustainability sustainability sustainability 1 Civil Engineering department, Jimma University Institute of Technology, Jimma-Ethiopia, P.O. Box 378, Ethiopia 1 Civil Engineering department, Jimma University Institute of Technology, Jimma-Ethiopia, P.O. Box 378, Ethiopia 2 Civil, Environmental, and Construction Engineering, Texas Tech University, Lubbock, TX 79409-1023, USA 3 Environmental Engineering Department, Jimma University Institute of Technology, Jimma-Ethiopia, P.O. Box 378, Ethiopia g g y 3 Environmental Engineering Department, Jimma University Institute of Technology, Jimma-Ethiopia, P.O. Box 378, Ethiopia * Correspondence: getchitaly@yahoo.com 1. Introduction Impurities, such as organic matter, silt, and clay content, in the sand contribute to a reduction in the compressive strength of the concrete. Olanitori [10] revealed that the compressive strength of hardened concrete decreases as the percentage of the silt and clay contents increases. Thus, assessing the quality of construction materials is very important to build strong, durable, and cost-effective structures for sustainable infrastructure development [11]. In order to minimize environmental degradation in the depletion of natural sand, and the associated cost of transportation, the use of locally available alternative materials is a critical solution. The use of locally available construction materials that are suitable for concrete production is recommended for the improvement of concrete performance at a comparable environmental incidence and lifetime [12]. Quarry dust wastes are used as alternative construction materials for natural sand replacement in order to minimize the shortage of river sand in concrete production in some countries. For instance, quarry dust was used to replace the river sand in India to conserve the sustainable development of scarcely available natural sand [13,14]. Quarry wastes are also used as a partial replacement for natural sand in the production of concrete in Malaysia [15]. Some alternative materials, such as fly ash, copper slag, limestone, and siliceous stone powder, are also used as a partial or full replacement of river sand in the production of mortar and concrete [16]. The strength and durability properties of concrete containing copper slag as a fine aggregate have been evaluated and an improvement has been observed up to a replacement level of 40% as investigated by [17]. Another research was done in the use of ferronickel slag as fine aggregate in which the workability and compressive strength properties of mortar were enhanced up to a 50% replacement for river sand [18]. Scoria, which is a product of explosive volcanic eruptions, has been used for centuries in the world as a construction material. Different researchers have examined the use of scoria as a construction material in concrete production. According to [19], scoria used as a coarse aggregate was found to be very useful in the production of lightweight concrete, with sufficient strength giving it the advantage of reducing the dead load in building structures. 1. Introduction Concrete is one of the most widely used construction materials in the construction industry of Ethiopia, like other countries of the world. The production of concrete-making materials or ingredients usually needs high energy, costs, and causes environmental pollution. Green concrete is characterized by the application of industrial by-products or alternative materials for wise utilization of natural resources and to save energy besides minimizing environmental pollution [1]. The increase in the demand for concrete production leads to the depletion of virgin concrete ingredients, which is a critical issue for all stakeholders of the construction industry for sustainable economic growth, particularly in developing countries. Therefore, there is a need to provide further attention to construction materials with regards to the economy, wise energy utilization, and environmental protection for sustainable development [2]. Fine aggregate is one of the ingredients of concrete that needs to get considerable attention in the production of green concrete, and it usually comprises up to 30% to 35% of the total volume of concrete. Natural or river sand is the most commonly used fine aggregate in the production of concrete. www.mdpi.com/journal/sustainability 2 of 15 Sustainability 2019, 11, 4647 The construction industry exploits sand mainly from stream beds, sandy marine sediments, and alluvial deposits [3]. Currently, a shortage of natural or river sand has become a problem for the construction industry in Ethiopia as a result of growing construction activities in many parts of the country, leading to significant depletion of natural sand [4,5]. The primary source of sand that is provided to construction worksites of the nearby towns of Addis Ababa city is Awash basin, located at a distance of about 70 to 120 km from Addis Ababa. The supply of river sand from such a long distance makes the production and supply of sand energy-intensive and costly. Moreover, the quality of the sand supplied from this source is poor with regards to its silt content, which is sometimes as high as 20%, going far beyond the standard limit [6]. The quality of constituent materials used in the production of concrete plays a principal role in determining the physical as well as the strength properties of the resultant concrete [7]. The properties of sand, specifically the gradation and fineness modulus of sand, are among the principal factors known to affect the performance of both fresh and hardened concrete [8,9]. 1. Introduction Scoria is also used as a lightweight aggregate with silica fume and fly ash mineral admixture in the production of lightweight structural concrete in which an outstanding performance was observed with regards to the strength to unit weight ratio [20]. Improvements in the mechanical strength of mortar were also observed when using volcanic scoria as sand in the production of Portland cement mortar [21]. Volcanic materials are comprehensively utilized in the production of blended cement due to their availability and properties. Volcanic materials are used as supplementary cementitious materials to produce environmentally friendly concrete [22]. The use of scoria as a replacement of cement was investigated in the production of self-consolidating concrete, which was found to be economically and environmentally pleasant with the required strength and durability properties of concrete [23]. The use of scoria as a replacement of cement in concrete production also improved the durability of the concrete by increasing its resistance to chloride penetration and giving longer periods of corrosion Sustainability 2019, 11, 4647 3 of 15 initiation [24]. Scoria, as a cement additive in Portland cement, was found to be effective in improving the property of the cement paste, particularly in controlling the alkali–silica reaction [25]. Scoria is widely available around Addis Ababa and in the rift valley regions of Ethiopia and thus can be used as an alternative construction material in concrete production [26]. The quarry sites of scoria are limited to the use of sub-base road construction and are found a short distance of about 0.5 to 4.2 km from the nearby towns, such as Adama, Bishoftu, and Mojo, unlike natural or river sand. Hence, the use of this material as fine aggregate in concrete production helps to save energy and cost [27,28]. The physical and chemical properties of an aggregate are attributed to the parent material. However, the shape, size, and surface texture(s) are essential for the concrete’s workability, and bond characteristics between the aggregate and cement paste are attributed to the mode of production. It is essential that the properties of the aggregate and its modes of production are understood to produce the required quality of concrete [6,29]. The aim of this study was to investigate the suitability of scoria as fine aggregate and its effect on the physical and mechanical properties of concrete. 2.1. Scoria The scoria sample was collected from the quarry site indicated in Figure 1 around Adama town, is found in the eastern part of Ethiopia at a geographical location of 8.54◦N and 39.27◦E at a distance of 100 km from Addis Ababa. The grain size distributions of the sample taken directly from the quarry site have the range of fine aggregate. Thus, the physical properties of this sample were determined to check its suitability as a fine aggregate in concrete production. Figure 1. Scoria quarry site near Adama town. Figure 1. Scoria quarry site near Adama town. 2.1.1. Gradation Sieve analysis was conducted to determine the proportion of the different sizes of aggregates and whether it satisfies the specification requirement in accordance with ASTM C33 and C136 [30,31] for use as a fine aggregate in concrete production. The series of standard sieves of square openings were placed in consecutive sizes in which the largest sieve size was placed at the top and the smallest was at the bottom, followed by a pan. The standard sample size was added to the sieve, allowing for shaking by the machine for 15 minutes. From the particle size distribution of the aggregate, the fineness modulus of the material was also determined to identify the coarseness or fineness of the material. As per the ASTM C33 specification [30], the fineness of fine aggregate needs to be in the range of 2.3 to 3.1. Based on the fineness modulus of fine aggregate, sand can be categorized as fine sand (2.2 to 2.6), medium sand (2.6 to 2.9), or coarse sand (2.9 to 3.2) as per the Ethiopian standard specification of building and road materials [32]. Sustainability 2019, 11, 4647 4 of 15 A blending of aggregates is undertaken when there is a deficiency in the grading of an aggregate. If two aggregates, designated A and B, are mixed together, with a fineness modulus of FMA and FMB, the resultant blend fineness is calculated using a formula [33]. A blending of aggregates is undertaken when there is a deficiency in the grading of an aggregate. If two aggregates, designated A and B, are mixed together, with a fineness modulus of FMA and FMB, the resultant blend fineness is calculated using a formula [33]. The results of the sieve analysis of the scoria sample taken from the quarry site of the study area are shown in Table 1. As it is shown in Table 1 and Figure 2, the cumulative percentage passes required for sieve sizes of 2.36, 0.6, and 0.3 mm are out of the ASTM C33 standard limit [30]. Thus, the particle size distribution of scoria does not fulfill the required specification of the standard and needs blending for use in concrete production. Table 1. Percentage passes of scoria fine aggregates. Sieve Sizes Cum. 2.1.1. Gradation % Passed Scoria Percentage Passing (ASTM C33 Standard) 9.75 mm 100 100 4.75 mm 95.30 95–100 2.36 mm 75.40 80–100 1.18 mm 55.30 50–85 0.6 mm 19.15 25–60 0.3 mm 5.20 10–30 0.15 mm 2.05 2–10 Figure 2. Scoria aggregate sieve analysis graphical representation. Table 1. Percentage passes of scoria fine aggregates. Table 1. Percentage passes of scoria fine aggregates Figure 2. Scoria aggregate sieve analysis graphical representation. Figure 2. Scoria aggregate sieve analysis graphical representation. 2.1.2. Fineness Modulus The percentages retained on the sieve sizes are indicated in Table 2, from which the fineness modulus of the scoria sample was determined. The fineness modulus (FM) of fine aggregate generally needs to lie between 2.3 and 3.1 as per the ASTM C136 specification as already mentioned. However, the FM of the scoria sample, which is 3.47, is out of the range of the limit set by the standard and thus should not be used without blending. Since the scoria sample is a coarser fine aggregate, it has to be blended with fine aggregate or sand with smaller fineness modulus to meet the standard specification after being blended. 5 of 15 Sustainability 2019, 11, 4647 Table 2. Percentage retained of scoria fine aggregates. Table 2. Percentage retained of scoria fine aggregates. Table 2. Percentage retained of scoria fine aggregates. Sieve Sizes % Retained Cumulative % Retained 9.75 mm 0 0 4.75 mm 4.70 4.70 2.36 mm 19.90 24.60 1.18 mm 20.00 44.60 0.6 mm 36.20 80.80 0.3 mm 14.00 94.80 0.15 mm 3.15 97.95 2.1.3. Silt Content, Specific Gravity, and Water Absorption Capacity The material in the fine aggregates that has a smaller size than 75 µm is generally regarded as silt. Silt in the fine aggregate can have a severe effect on the quality of the concrete. It mainly affects the workability of the concrete, and also results in a reduction in strength. The silt content of the sample was determined in the laboratory using the ASTM C117 standard specification [34]. According to the standard, the maximum silt content is limited to 3%. Specific gravity is the ratio between the weight of the substance and that of the same volume of water. In this definition, the substance or material is assumed to be solid. The apparent specific gravity of an aggregate refers to the solid materials excluding the pores, and the bulk specific gravity refers to the total volume, including the pores of the aggregate. Absorption capacity refers to the amount of water a solid aggregate material can absorb. The maximum water absorption capacity of fine aggregate is limited to only 3% as per the ASTM C128 specification [35]. Both the specific capacity and water absorption capacity are determined using the Pycnometer method. 2.1.2. Fineness Modulus The unit weight effectively measures the volume that the graded aggregates occupy in concrete and includes both the solid aggregate particles and the voids between the aggregates. The unit weight of the sample was measured using ASTM C29/29M [36] by filling the sample in a container of a known volume (0.005 m3) and weighing it. A summary of the physical properties of scoria is shown in Table 3. The result obtained from the laboratory test showed the bulk specific gravity and the apparent specific gravity of the scoria sample was found to be 2.397 and 2.457, respectively. In addition, the water absorption capacity of scoria was found to be very low, with a numerical value of 0.6%. From the experiment conducted, the scoria seems to have a smaller bulk density of 1165.70 kg/m3 as compared to common river sand. The silt content of the scoria sample was found to be 0.76%, which is less than the ASTM standard [34]. Table 3. Summary of the physical properties of the scoria sample. Physical Properties of Scoria Fine Aggregate Silt content (%) 0.76 Fineness modulus (FM) 3.47 Bulk Unit weight (kg/m3) 1165.70 Bulk Specific gravity 2.397 Apparent Specific gravity 2.457 Water absorption capacity (%) 0.60 2.1.4. Chemical Composition of Scoria Table 3. Summary of the physical properties of the scoria sample. Table 3. Summary of the physical properties of the scoria sample. Physical Properties of Scoria Fine Aggregate Silt content (%) 0.76 Fineness modulus (FM) 3.47 Bulk Unit weight (kg/m3) 1165.70 Bulk Specific gravity 2.397 Apparent Specific gravity 2.457 Water absorption capacity (%) 0.60 2.1.4. Chemical Composition of Scoria Physical Properties of Scoria Fine Aggregate Physical Properties of Scoria Fine Aggregate 2.1.4. Chemical Composition of Scoria 2.1.4. Chemical Composition of Scoria 4. Chemical Composition of Scoria The scoria sample was ground or milled, and the chemical composition of the scoria powder shown in Figure 3 was determined at the Geochemical Laboratory of Geological Survey of Ethiopia. The analytical method of elemental analysis was conducted to determine the mineral composition of the scoria powder, and the obtained result is shown in Table 4. From the result obtained, the scoria powder contains a greater amount of silicon dioxide (52.53%), which is followed by aluminate and 6 of 15 Sustainability 2019, 11, 4647 ferrite, with a percentage of 15.49% and 11.00%, respectively. The summation of these three oxide compositions was found to be 79.02%, and the material is classified as class F pozzolanic material according to ASTM C-618 [37]. Figure 3. Picture of scoria sample: (a) Scoria aggregate; (b) Scoria powder. Figure 3. Picture of scoria sample: (a) Scoria aggregate; (b) Scoria powder. Table 4. Chemical composition of scoria. Composition SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O MnO P2O5 TiO2 H2O LOI Content (%) 52.53 15.49 11.0 10.1 4.4 2.56 <0.1 0.18 0.17 0.88 0.64 0.85 2.2. Sand Natural river sand, which is supplied from the Gambella region in Ethiopia, was used in this study. All the physical properties that were determined for scoria were also determined following the same procedure for use of the sand fully or partially as a fine aggregate in concrete production. 2.2.1. Gradation and Fineness Modulus 2.2.2. Specific Gravity and Water Absorption Capacity 2.2.2. Specific Gravity and Water Absorption Capacity 2.2.2. Specific Gravity and Water Absorption Capacity A summary of the physical properties of the river sand is presented in Table 6. The bulk specific gravity and apparent specific gravity of river sand were determined to be 2.58 and 2.67, respectively. The silt content test result of the river sand was found to be 4.26%, which is above the maximum allowable limit of the ASTM C117 specification [34]. Therefore, the river sand was washed, and its silt content was reduced to 1.56% to keep the standard specification for concrete production. The bulk density of the river sand determined from the laboratory result was 1482 kg/m3. The water absorption capacity of the sand was found to be 0.67%. Table 6. Summary of the physical properties of the river sample. Table 6. Summary of the physical properties of the river sample. Physical Properties of River Sand Fine Aggregate Silt content (%) 1.56 Fineness modulus (FM) 2.36 Bulk Unit weight (kg/m3) 1482 Bulk Specific gravity 2.58 Apparent Specific gravity 2.67 Water absorption capacity (%) 0.67 Physical Properties of River Sand Fine Aggregate 2.2.1. Gradation and Fineness Modulus The particle size distribution of river sand was analyzed, and the obtained result shown in Table 5. The obtained result fulfills the ASTM C33 [30] standard specification for concrete production. The grain size distribution of the river sand used, indicated in Figure 4 shows that it is fine sand since its fineness modulus, which is 2.36, lies between 2.2 to 2.6. Table 5. Gradation of river sand. Sieve Sizes in mm Cumulative % Retained Cumulative % Passed Cumulative % Pass (ASTM C33) 9.50 0 100 100 4.75 0.74 99.26 95–100 2.36 3.45 96.55 80–100 1.18 18.23 81.77 50–85 0.60 48.28 51.72 25–60 0.30 72.91 27.09 5–30 0.15 92.61 7.39 0–10 0.075 99.62 0.68 Table 5. Gradation of river sand. 7 of 15 Sustainability 2019, 11, 4647 Figure 4. River sand grain size distribution curve. 2 2 2 Specific Gravity and Water Absorption Capacity Figure 4. River sand grain size distribution curve. 2.3. Coarse Aggregate The coarse aggregate used in this study was crushed stone gravel of basaltic origin. The maximum nominal size of the coarse aggregate used was 20 mm. Gradation of the coarse aggregate conducted to check whether it fulfilled the ASTM C-33 specification [30] for concrete production. The test result of the grain size distribution of the coarse aggregate is indicated in Figure 5. The unit weight, specific gravity, moisture content, and water absorption of the aggregate were also determined using ASTM C-29, ASTM C-127, and ASTM C-136 [31,36,38]. The test results of these properties are presented in Table 7. The physical properties of the aggregate satisfied the ASTM standard specification for concrete production. 8 of 15 8 of 15 Sustainability 2019, 11, 4647 Figure 5. Gradation of coarse aggregate. Table 7. Physical properties of coarse aggregate. Coarse Aggregate Property Bulk Specific Gravity 2.65 Apparent Specific Gravity 2.73 Water Absorption (%) 1.08 Moisture Content (%) 0.23 2.4. Cement Figure 5. Gradation of coarse aggregate. Table 7. Physical properties of coarse aggregate. Coarse Aggregate Property Bulk Specific Gravity 2.65 Apparent Specific Gravity 2.73 Water Absorption (%) 1.08 Moisture Content (%) 0.23 C t Figure 5. Gradation of coarse aggregate. Table 7. Physical properties of coarse aggregate. 2.5. Water Tap water, which is used for drinking purposes, was used in this experimental work. 2.4. Cement Dangote ordinary Portland cement (OPC) with a grade of 42.5 R was used in this study. The physical properties of Dangote OPC cement were determined as per ASTM standard specification(s). The physical properties of Dangote OPC, which include the standard consistency, setting time, fineness, and specific gravity, were determined using the standard specifications of ASTM C 187, ASTM C 191, ASTM C-184, and ASTM C 188, respectively [39–42]. The obtained result of the laboratory experiment on the cement fineness indicated that 100% passed the 150 µm, 97.6% passed the 100 µm, and 92% and 78.2% passed the 90 µm and 63 µm, respectively. The standard consistency, setting time, and specific gravity test results are indicated in Table 8. Table 8. Physical properties of cement. Properties Dangote OPC Specific Gravity 3.13 Standard Consistency (%) 32.4 Initial Setting Time (min) 80 Final Setting Time (min) 290 Table 8. Physical properties of cement. Properties Dangote OPC Specific Gravity 3.13 Standard Consistency (%) 32.4 Initial Setting Time (min) 80 Final Setting Time (min) 290 Table 8. Physical properties of cement. 9 of 15 Sustainability 2019, 11, 4647 2.6. Concrete Mix Design The American Concrete Institute (ACI) method of the mix design method was used in designing the concrete mixes. The trial mix was prepared for a concrete grade of C-25 compressive strength. A maximum aggregate size of 20 mm was used in the experiment. A water to cement ratio of 0.50 was used based on the maximum aggregate size and the concrete grade. By considering the workability of the concrete in line with the aggregate size and strength of the concrete, the amount of water required was determined following the ACI mix design table. Accordingly, the amount of water required for the mix is 185 kg/m3. Based on the fineness modulus of the fine aggregate, the amount of the coarse aggregate was determined for the specified maximum aggregate size. The proportion of the concrete ingredients used in a unit cubic meter is indicated in Table 9, considering the associated physical properties of the materials. Table 9. Proportion of the concrete constituents. Constituents Water (kg/m3) Cement (kg/m3) River Sand (kg/m3) Scoria (kg/m3) Coarse Aggregate (kg/m3) 100% RS + 0% Scoria 185 370 673.40 0.00 1087.80 90% RS + 10% Scoria 185 370 606.06 67.34 1087.80 75% RS + 25% Scoria 185 370 505.05 168.35 1087.80 65% RS + 35% Scoria 185 370 437.71 235.69 1087.80 50% RS + 50% Scoria 185 370 336.70 336.70 1087.80 Table 9. Proportion of the concrete constituents. Table 9. Proportion of the concrete constituents. In this study, the fine aggregate used was a combination of scoria and river sand. Replacement of scoria was made with a percentage of 0% (control), 10%, 25%, 35%, and 50% on a weight basis. The effects of the different levels of percentages were investigated on the workability and compressive strength of the concrete by measuring the slump height and compression machine, respectively. 2.7. Concrete Tests Concrete tests were carried out to determine the properties of both fresh and hardened concrete. The slump cone method was used for the determination of the concrete consistency or workability of the fresh concrete using ASTM C143 [43]. A slump cone with 10 cm diameter at the top, 20 cm diameter at the bottom, and vertical height of 30 cm was used. The internal surface of the cone was cleaned and placed on a smooth horizontal surface. The cone was filled in three layers of equal height with concrete, and each layer was tamped 25 times with a 16 mm tamping rod. The slump value for each concrete mix was determined. The precision of the test result was checked as per the ASTM C143 to accept the test results. The observed slump values for each level of replacement are presented in Section 3.1. The compressive strengths of the test specimens were determined using an automatic compression testing machine (UTC-4321) using ASTM C-39/39M [44]. Cubic steel molds of a 15 cm dimension were used to cast the concrete test specimens. Once the molds were firmly fixed, and their interior faces greased, the molds were filled in three layers each, with 5 cm being tapped 25 times manually with a steel rod. Three specimens were prepared for each test, and the specimens were stored in curing water tank for 7, 28, and 56 day periods at a room temperature of about 25 ◦C [45]. The reliability in the precision of the measurement in the data set was checked by determining the standard deviation, and the error bars are indicated in the graphs presented in Figures 7–9. The standard deviations of the data sets varied from 0.14 to 0.478, which is less than the acceptable precision of 1.4 MPa applicable for a laboratory with in-test variation [46]. 10 of 15 Sustainability 2019, 11, 4647 3.1. Workability of Fresh Concrete The test result of the slump height observed is indicated in Table 10. The test result of the experimental work shows that the workability of the fresh concrete slightly increases as the amount of scoria replaced as fine aggregate increases. As the percentage of scoria increases, the fineness of the blended fine aggregate increases. The increase in the workability of the concrete is supposed to be due to the relative coarseness of the fine aggregate as a result of the replacement of scoria with the higher fineness of modulus as compared to the river sand (RS). Table 10. Slump height of scoria replaced concrete. Mix Designation Slump Value (mm) 100% RS + 0% Scoria 32 90% RS + 10% Scoria 33 75% RS + 25% Scoria 35 65% RS + 35% Scoria 36 50% RS + 50% Scoria 38 3.2. Density of Hardened Concrete Table 10. Slump height of scoria replaced concrete. 3.2. Density of Hardened Concrete 3.2. Density of Hardened Concrete The density of the test sample at the age of the 56th day for the different levels or proportions of scoria replacement was determined, and the result indicated in Table 11. It can be seen from the table that the density of the concrete decreases as the percentage of scoria increases due to the lower unit weight of scoria. This finding confirms a research study made by Shannag et al. in 2014 [20]. The reduction in the density of concrete helps to decrease the self-weight of the concrete in the determination of the design dead of the concrete structural member to have reduced concrete cross-sectional dimension supporting the structural member and thus saves the amount of concrete volume relatively. Table 11. Density of hardened concrete. Mix Designation Density, kg/m3 100% RS + 0% Scoria 2501.93 90% RS + 10% Scoria 2482.68 75% RS + 25% Scoria 2465.19 65% RS + 35% Scoria 2458.62 50% RS + 50% Scoria 2452.05 Table 11. Density of hardened concrete. The experimental test results show that the density of the hardened concrete decreases with increasing amounts of scoria percentage, which is expressed in a polynomial function with an equation of Y = 0.018X2 −1.894X + 2501.1 as shown in Figure 6, where X represents the percentage of scoria replacement. The density of the hardened concrete at any given percentage of scoria can be estimated using this equation. 11 of 15 11 of 15 Sustainability 2019, 11, 4647 Figure 6. Density of hardened concrete as a function of scoria replacement. Figure 6. Density of hardened concrete as a function of scoria replacement. Figure 6. Density of hardened concrete as a function of scoria replacement. 3.3. Compressive Strength 3.3. Compressive Strength The average compressive strength of scoria-replaced concrete of different curing ages is presented in Figures 7–9. The compressive strength test result of the experiment shows that there is an increase in the compressive strength at the curing ages of 7, 28, and 56 days with an increase in the amount of the scoria used as fine aggregate as compared to that of the control sample. The average compressive strength of concrete for a scoria replacement of 10%, 25%, 35%, and 50% increased by 1.879%, 4.212%, 5.616%, and 0.829%, respectively, for the seven days age of curing compared to the control mix. For the age of the 28-day strength of the concrete, the levels of increments observed were 2.49%, 4.62%, 5.94%, and 0.451% for the respective scoria replacement percentages. Almost the same amounts of increments were also observed for the 56-day concrete compared to the control. Figure 7. Compressive strength of seven-day age with an error bar. Figure 7. Compressive strength of seven-day age with an error bar. 12 of 15 12 of 15 Sustainability 2019, 11, 4647 Figure 8. Compressive strength of 28-day age with error bar. Figure 9. Compressive strength of 56-day age with error bar. The enhancement in the properties of the concrete observed was due to the inherent physica properties of the scoria sample. The silt and clay content of scoria is smaller than the river sand whil the water absorption capacity is almost similar. The decrease in the silt content of the blended concret mixes helps the proper cement hydration processes to take place, favoring the compressive strength o the concrete to occur due to proper bondage between the aggregates. The blending of the finer rive sand with the coarser scoria resulted in a well-graded particle size distribution, increasing the volum Figure 8. Compressive strength of 28-day age with error bar. Figure 9. Compressive strength of 56-day age with error bar. Figure 9. Compressive strength of 56-day age with error bar. The enhancement in the properties of the concrete observed was due to the inherent physical properties of the scoria sample. The silt and clay content of scoria is smaller than the river sand while the water absorption capacity is almost similar. 4. Conclusions Funding: This research was funded by the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), and the APC was funded by Jimma University Institute of Technology and Texas Tech University. Acknowledgments: The authors are thankful to Texas Tech University, GIZ, and Jimma University Institute of Technology for their cooperation in providing the necessary support for carrying out this research study. Acknowledgments: The authors are thankful to Texas Tech University, GIZ, and Jimma University Institute of Technology for their cooperation in providing the necessary support for carrying out this research study. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. 3.3. Compressive Strength The decrease in the silt content of the blended concrete mixes helps the proper cement hydration processes to take place, favoring the compressive strength of the concrete to occur due to proper bondage between the aggregates. The blending of the finer river sand with the coarser scoria resulted in a well-graded particle size distribution, increasing the volume solid of the concrete and thereby increasing the strength of the concrete. Sustainability 2019, 11, 4647 13 of 15 The maximum compressive strength attained at a replacement of 35% of scoria. As the amount of scoria gets more and more, the amount of finer sand aggregates decreases, and thereby, the volume of solid decreases. Besides, the formation of the interaction of paste and aggregate also decreased due to the lowering of the filler or packing effect. As a result, the compressive strength of concrete declined beyond 35%. However, the compressive strength with 50% replacement is still higher than that of the control sample due to the well-graded particle distribution of the blended fine aggregate as compared to that of the river sand. Generally, this experimental study shows that the scoria sample under consideration can replace the river sand up to 50% in order to produce concrete for the target mix design considered. 4. Conclusions The findings of the study can be concluded as such: The evaluation of the properties of scoria showed that scoria is suitable to be used as a partial replacement of fine aggregate in the production of concrete. Scoria was found to have outstanding quality, especially with regards to its silt content compared to the river sand. Thus, the use of scoria as fine aggregate helps to save energy and cost of washing the river sand. It also produces improved concrete quality. The use of scoria in concrete production was found to increase the workability of the fresh concrete as compared to the control, favoring a better quality concrete. The compressive strength of the hardened concrete increased up to a 50% replacement of scoria as a fine aggregate compared to the control concrete in which the river the sand was used. Thus, scoria can be used up to this level of replacement to produce the target design strength of concrete of C-25 grade. The river sand used in this study had a well-graded gradation of particle size distribution by using a 35% replacement of the scoria sample, thus causing the hardened concrete to have maximum compressive strength. The density of the hardened concrete decreases as the ratio of scoria replacement increases due to its lesser specific gravity and induces the concrete to have a relatively decreased self-weight, and thus increases the strength to unit weight ratio or structural efficiency as compared to the control sample. In general, scoria is suitable for use to produce concrete as a fine aggregate and partial replacement of river sand. Its local availability aids its choice as an alternative construction material. Its availability at a shorter distance compared to river sand around the central towns of Ethiopia helps to save energy and cost, and thus meets the strategy of green concrete production. Author Contributions: Conceptualization, methodology, data collection and data analysis, validation, original draft writing and editing—G.K.W.; Supervision and editing—M.M.D., F.F.F., T.G. Author Contributions: Conceptualization, methodology, data collection and data analysis, validation, original draft writing and editing—G.K.W.; Supervision and editing—M.M.D., F.F.F., T.G. Funding: This research was funded by the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), and the APC was funded by Jimma University Institute of Technology and Texas Tech University. References 1. Hameed, M.S.; Sekar, A. Properties of green concrete containing quarry rock dust and marble sludge powder as fine aggregate. ARPN J. Eng. Appl. Sci. 2009, 4, 83–89. 1. Hameed, M.S.; Sekar, A. Properties of green concrete containing quarry rock dust and marble sludge powder as fine aggregate. ARPN J. Eng. Appl. Sci. 2009, 4, 83–89. 2. Yang, L. Sustainability and innovative construction: Green building with concrete. J. Civ. Eng. Environ. 2012, 2, 1–2. 3. Mengistu, T.; Haile, M. Industrial Minerals and Rocks Resource Potential of Ethiopia; Federal Democratic Republic of Ethiopia, Ministry of Mines, Geological Survey of Ethiopia: Addis Ababa, Ethiopia, 2003. Available online: http://etd.aau.edu.et/bitstream/handle/123456789/2268/Shewaferaw%20Dinku%20.pdf?sequence= 1&isAllowed=y (accessed on 8 May 2017). y y 4. Shiferaw, B.; Abebe, D. The Use of Manufactured Sand in Concrete Production: Test Result and Cost Comparison. MSC Thesis, Faculty of Technology, Addis Ababa University, Addis Ababa, Ethiopia, 2006. 4. Shiferaw, B.; Abebe, D. The Use of Manufactured Sand in Concrete Production: Test Result and Cost Comparison. MSC Thesis, Faculty of Technology, Addis Ababa University, Addis Ababa, Ethiopia, 2006. 4. Shiferaw, B.; Abebe, D. The Use of Manufactured Sand in Concrete Production: Test Result and Cost Comparison. MSC Thesis, Faculty of Technology, Addis Ababa University, Addis Ababa, Ethiopia, 2006. Sustainability 2019, 11, 4647 14 of 15 14 of 15 5. Ararsa, W.; Emer, T.Q.; Abraham, A. Suitability of Ambo Sandstone Fine Aggregate as an Alternative River Sand Replacement in Normal Concrete Production. Am. J. Civ. Eng. Archit. 2018, 6, 140–146. [CrossRef] 6. Abebe, D. The Need for Standardization of Aggregates for Concrete Production in Ethiopian Construction Industry. Available online: https://scholarworks.wmich.edu/cgi/viewcontent.cgi?article=1089&context= africancenter_icad_archive (accessed on 12 July 2018). 7. Ngugi, H.N.; Mutuku, R.N.; Gariy, Z.A. Effects of sand quality on compressive strength of concrete: A case study of Nairobi County and its Environs. J. Civ. Eng. 2014, 4, 255. [CrossRef] 8. Luo, F.J.; He, L.; Pan, Z.; Duan, W.H.; Zhao, X.L.; Collins, F. Effect of very fine particles on workability and strength of concrete made with dune sand. Construct. Build. Mater. 2013, 47, 131–137. [CrossRef] 9. Sabih, G.; Tarefder, R.A.; Jamil, S.M. Optimization of gradation and fineness modulus of naturally fine sands for improved performance as fine aggregate in concrete. Int. Conf. Sustain. Des. Eng. Constr. 2016, 145, 66–73. [CrossRef] 10. Olanitori, L. Mitigating the effect of clay content of sand on concrete strength. References In Proceedings of the 31st Conference on Our World in Concrete and Structures, Singapore, 15–17 August 2006. 11. Savitha, R. Importance of Quality Assurance of Materials for Construction Work. Building Materials Research and Testing Division, National Building Research Organization, 2012. Available online: www.nbro.gov.lk/ images/content_image/publications/symposia/2010/importance_of_quality_assurance.pdf. (accessed on 10 August 2019). 12. Muller, H.S.; Haist, M.; Vogel, M. Assessment of the sustainability potential of concrete and concrete structures considering their environmental impact, performance and lifetime. Constr. Build. Mater. 2014, 67, 321–337. [CrossRef] 13. Azhagarsamy, S.; Sumaiya Fathima, A.M.; Thilagavathi, K. Effect of Quarry Dust on High Performance Concrete. Int. Res. J. Eng. Tech. 2017, 4, 1223–1226. 14. Mamta, K.; Rabbani, A. Use of Quarry Dust as fine aggregates by Partial Replacement of Sand in Concrete. Int. Res. J. Eng. Technol. 2017, 4, 851–853. 15. Safiuddin, M.; Raman, S.N.; Zain, M.F.M. Utilization of Quarry Waste Fine Aggregate in Concrete Mixtures. J. Appl. Sci. Res. 2007, 3, 202–208. 6. Sukesh, C.; Krishna, K.B.; Teja, P.S.L.S.; Rao, S.K. Partial replacement of sand with quarry dust in conc Int. J. Innov. Technol. Explor. Eng. 2013, 2, 254–258. 7. Patnaik, B.; Seshadri, S.; Rao, S.C. Strength and durability properties of copper slag admixed concrete. IJ Int. J. Res. Eng. Technol. 2015. [CrossRef] 18. Saha, A.K.; Sarker, P.K. Compressive strength of mortar containing ferronickel slag as replacement of natural sand. Procedia Eng. 2017, 171, 689–694. [CrossRef] 19. Aho, I.; Uungwa, S. Engineering properties of scoria concrete as a construction material. Glob. J. Eng. Res. 2015, 14, 59–63. [CrossRef] 20. Shannag, M.; Charif, A.; Naser, S.; Faisal, F.; Karim, A. Structural Behavior of Lightweight Concrete Made with Scoria Aggregates and Mineral Admixtures. Int. J. Civ. Archit. Sci. Eng. 2014, 8, 105–109. 21. Tchamdjou, W.H.J.; Grigoletto, S.; Michel, F.; Courard, L.; Abidi, M.L.; Cherradi, T. An investigation on the use of coarse volcanic scoria as sand in Portland cement mortar. Case Stud. Constr. Mater. 2017, 7, 191–206. [CrossRef] 2. Cai, G.; Noguchi, T.; Degée, H.; Zhao, J.; Kitagaki, R. Volcano-related materials in concretes: A comprehen review. Environ. Sci. Pollut. Res. 2016, 23, 7220–7243. [CrossRef] [PubMed] 23. Hossain, K.; Lachemi, M. Fresh, mechanical, and durability characteristics of self-consolidating concrete incorporating volcanic ash. J. Mater. Civ. Eng. 2009, 22, 651–657. [CrossRef] 24. Al-Swaidani, A.M.; Aliyan, S.D. Effect of adding scoria as cement replacement on durability-related properties. Int. J. Concr. Struct. Mater. 2015, 9, 241–254. [CrossRef] 25. Hossain, K.M.A. References Blended cement and lightweight concrete using scoria: Mix design, strength, durability and heat insulation characteristics. Int. J. Phys. Sci. 2006, 1, 5–16. 26. Niguse, T.; Shiferaw, T. Scoria sand replacement in structural concrete. Zede J. 1984, 6, 27–32. 27. Bontempi, E. A new approach for evaluating the sustainability of raw materials substitution based on embodied energy and the CO2 footprint. J. Clean. Prod. 2017, 162, 162–169. [CrossRef] 15 of 15 Sustainability 2019, 11, 4647 28. Kupwade, P.K.; DeWolf, C.; Chin, S.; Ochsendof, J.; Hajiah, A.E.; Al-Mimin, A.; Büyüköztürk, O. Impact of Embodied Energy on materials/buildings with partial replacement of ordinary Portland Cement (OPC) by natural Pozzolanic Volcanic Ash. J. Clean. Prod. 2018, 177, 547–554. [CrossRef] Neville, A.; Brooks, J. Concrete Technology (Edition-2001 Standards); Addison Wesley Longman: Harlow, UK 2001. 30. ASTMC33. Standard Specification for Concrete Aggregates; ASTM International: West Conshohocken, PA, USA, 2013. 31. ASTMC136. Standard Test Method for Sieve Analysis of Fine and Coarse Aggregates; ASTM International: West Conshohocken, PA, USA, 2014. 32. EBCS. Ethiopian Standard Specification of Building and Road Materials; EFDR, Ethiopian Federal Democratic Republic Addis Ababa: Addis Ababa, Ethiopia, 2002. 33. Zongjin, L. Advanced Concrete Technology; John Wiley & Sons: Hoboken, NJ, USA, 2011. 34. ASTMC117. Standard Test Method for Materials Finer than 75 µm Sieve in Mineral Aggregates by Washing; ASTM International: West Conshohocken, PA, USA, 2004. ASTMC128-15. Standard Test Method for Relative Density (Specific Gravity) and Absorption of Fine Aggregate ASTM International: West Conshohocken, PA, USA, 2015. ASTM International: West Conshohocken, PA, USA, 2015. 36. ASTMC29/C29M. Standard Test Method for Bulk Density (“Unit Weight”) and Voids in Aggregate; ASTM International: West Conshohocken PA USA 2001 6. ASTMC29/C29M. Standard Test Method for Bulk Density (“Unit Weight”) and Voids in Aggregate; AS International: West Conshohocken, PA, USA, 2001. 37. ASTMC618. Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use in Concrete; ASTM International: West Conshohocken, PA, USA, 2012. 38. ASTMC127-15. Standard Test Method for Relative Density (Specific Gravity) and Absorption of Coarse Aggregate; ASTM International: West Conshohocken, PA, USA, 2015. 39. ASTMC187-11. Standard Test Method for Normal Consistency of Hydraulic Cement; ASTM International: West Conshohocken, PA, USA, 2011. 40. ASTMC191-13. Standard Test Methods for Time of Setting of Hydraulic Cement by Vicat Needle; ASTM International: West Conshohocken, PA, USA, 2013. 41. ASTMC184. Standard Test Method for Fineness of Hydraulic Cement by the 150-Micrometer (No. 100) and 75-Micrometer (No. References 200) Sieves; ASTM International: West Conshohocken, PA, USA, 1994. 42. ASTMC188-17. Standard Test Method for Density of Hydraulic Cement; ASTM International: West Conshohocken, PA, USA, 2017. 43. ASTMC143/C143M-15a. Standard Test Method for Slump of Hydraulic-Cement Concrete; ASTM International: West Conshohocken, PA, USA, 2015. 44. ASTMC39/39M. Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens; ASTM International: West Conshohocken, PA, USA, 2016. 45. ASTMC192/C192M. Standard Practice for Making and Curing Concrete Test Specimens in the Laboratory; ASTM International: West Conshohocken, PA, USA, 2007. 46. ACI-214R-11. Guide to Evaluation of Strength Test Results of Concrete; ACI Committee: Framington Hills, MI, USA, 2011. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4368376986
https://link.springer.com/content/pdf/10.1007/s10531-023-02612-z.pdf
English
null
Impact of herbivory by Zerenopsis lepida (Lepidoptera: Geometridae) on the endangered Encephalartos eugene-maraisii under field conditions
Biodiversity and conservation
2,023
cc-by
10,381
Biodiversity and Conservation (2023) 32:2451–2468 https://doi.org/10.1007/s10531-023-02612-z Biodiversity and Conservation (2023) 32:2451–2468 https://doi.org/10.1007/s10531-023-02612-z ORIGINAL RESEARCH P. D. Janse van Rensburg pjvrensburg18@gmail.com Communicated by Daniel Sanchez Mata. Keywords  Cycad · Encephalartos · Zamiaceae · Zerenopsis · Herbivory · Conservation biology Impact of herbivory by Zerenopsis lepida (Lepidoptera: Geometridae) on the endangered Encephalartos eugene- maraisii under field conditions P. D. Janse van Rensburg1  · H. Bezuidenhout2,3  · J. Van den Berg1 Received: 13 June 2022 / Revised: 13 March 2023 / Accepted: 16 April 2023 / Published online: 3 May 2023 © The Author(s) 2023 1 Unit of Environmental Sciences and Management, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa 2 Scientific Services, Conservation Services, P.O. Box 110040, Hadison Park, Kimberley 8306, SANParks, South Africa 3 Applied Behavioural Ecology and Ecosystem Research Unit, University of South Africa, Private Bag X6, Florida 1717, South Africa Introduction The consumption of plants by herbivores is among the most important species interactions in nature. Insects are one of the most abundant groups of herbivores and have significant impacts on plant population dynamics (Ancheta and Heard 2011; Agrawal et al. 2012; Myers and Sarfraz 2017). Most herbivores remain at low densities and impose minor but chronic damage on their host plant, a phenomenon referred to as background insect her­ bivory (Kozlov et al. 2015; Barrio et al. 2017; Shao et al. 2020). However, numerous studies have demonstrated that even minor herbivory can reduce plant health and reduce the rates of population growth (Baron and Bros 2005; Kettenring et al. 2009; Zvereva et al. 2012; Stephens and Westoby 2014; Zvereva and Kozlov 2014). It is important to consider how insect herbivores impact populations of rare plant spe­ cies as part of conservation programs (Myers and Sarfraz 2017). Specialist herbivores on rare plants raise interesting conservation questions because they may be rare and at risk of extinction themselves (Ancheta and Heard 2011; Moir et al. 2011, 2016; Ramírez-Restrepo et al. 2016; Koi 2017). Since rare plant species characteristically have small and isolated populations, with a specific habitat, specialist herbivores will likely have lower rates of colonization and higher probabilities of extinction during the decline of their host popula­ tion (Moir et al. 2011; Taylor and Moir 2014). Indeed, reduced herbivore incidence and levels of herbivory, and in some cases even the complete absence of herbivores, have been reported in smaller host plant populations (Colling and Matthies 2004; Baron and Bros 2005). Increased herbivory has only rarely been reported in small host populations (Elzinga et al. 2005). Globally, several groups of Lepidoptera feed on cycads although many cycadivorous Lepidoptera remains understudied (Whitaker and Salzman 2020). In Africa, an entire tribe of Geometridae moths, the Diptychini, consume leaves of the cycad genera Encephalartos and Stangeria (Sihvonen et al. 2015). The majority of Diptychini moths have only been recorded from a few cycad hosts and some are of conservation concern, e.g., Calliora­ tis millari (Lepidoptera: Geometridae) in South Africa (Staude and Sihvonen 2014; Louw and Armstrong 2018; Terblanche 2018). Sufficient host plant populations that can sustain herbivore populations are becoming increasingly scarce and fragmented (Staude and Sih­ vonen 2014). Zerenopsis lepida (Lepidoptera: Geometridae), however, is widely distributed and occurs on wild populations of several species of Encephalartos and Stangeria eriopus. Abstract Encephalartos eugene-maraisii is an endangered, endemic plant in the Waterberg Moun­ tain range of South Africa. This is the first report of herbivory by Zerenopsis lepida (Lepidoptera: Geometridae) on natural populations of E. eugene-maraisii. The impact of herbivory by folivorous arthropods on this cycad species has not been studied before. This study aimed to determine the incidence and intensity of Z. lepida herbivory on E. eugene- maraisii under field conditions. Local plant dynamics were studied by recording the plant location, sex, size and phenological stage for each plant. Damage surveys were carried out over two consecutive seasons (2020–2021 and 2021–2022). Leaf damage was assessed on 227 plants comprising 692 stems. Approximately 40% of the stems had signs of leaf her­ bivory by Z. lepida and 10% of the total leaf area was removed by Z. lepida larvae. Less herbivory was recorded in areas where plants synchronously produced new leaf flushes compared to areas with asynchronous leaf flushes. Greater apparency of plants led to a greater incidence of herbivory and level of damage. Repeated damage to plants may delay future leaf and cone flushing events and contribute to the threat against this cycad species. 1 Unit of Environmental Sciences and Management, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa 2 Scientific Services, Conservation Services, P.O. Box 110040, Hadison Park, Kimberley 8306, SANParks, South Africa 3 Applied Behavioural Ecology and Ecosystem Research Unit, University of South Africa, Private Bag X6, Florida 1717, South Africa 3 Applied Behavioural Ecology and Ecosystem Research Unit, University of South Africa, Private Bag X6, Florida 1717, South Africa 1 3 3 Biodiversity and Conservation (2023) 32:2451–2468 2452 1 3 Introduction Zerenopsis lepida also made the transition to cultivated cycads outside their native range, including cycad genera from other world regions such as Cycas and Dioon (Donaldson and Bösenberg 1995; Staude 2008; Donaldson 2015). However, the unique genetic diversity of the various wild Z. lepida populations is still under threat from the large-scale removal of cycads from the wild and some wild populations of Z. lepida have been reported to have disappeared (Staude and Sihvonen 2014). Zerenopsis lepida and other Diptychini have several characteristics that set them apart from other Geometridae i.e., gregarious larvae, aposematic colouration, diurnal adults, and lek mating behaviour whereby males choose specific lek sites from where they attract females (Staude et al. 2011; Bayliss et al. 2009). Early-instar larvae of Z. lepida always feed on new cycad leaves that are still soft while larger instars (≥ third instar) can switch to a variety of secondary angiosperm host plants growing in the vicinity of their primary cycad host (Donaldson and Bösenberg 1995). Carissa bispinosa, C. macrocarpa, Diospyros 1 3 1 Biodiversity and Conservation (2023) 32:2451–2468 2453 lycioides, D. whyteana, Apodytes dimidiata, Maesa alnifolia and M. lanceolata have been recorded as secondary host plants for Z. lepida (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). Cycad-host preference amongst ovipositing females and early-instar larvae may primarily be influenced by the presence of methylazoxymethanol (MAM) glycosides in cycad tissue and secondary host plants are likely included in the larval diet when new and soft cycad leaves are depleted (Donaldson and Bösenberg 1995).i lycioides, D. whyteana, Apodytes dimidiata, Maesa alnifolia and M. lanceolata have been recorded as secondary host plants for Z. lepida (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). Cycad-host preference amongst ovipositing females and early-instar larvae may primarily be influenced by the presence of methylazoxymethanol (MAM) glycosides in cycad tissue and secondary host plants are likely included in the larval diet when new and soft cycad leaves are depleted (Donaldson and Bösenberg 1995).i This study reports for the first-time evidence of Z. lepida feeding on a wild population of Encephalartos eugene-maraisii, an endangered endemic plant in the Waterberg Mountain range of South Africa (IUCN Red List Criteria A2ad + 4ad; B1ab(v)) with between 400 and 620 plants remaining in the wild (Bösenberg 2022). Encephalartos eugene-maraisii is a large cycad with aerial stems (up to 4 m long) that often becomes procumbent with age. Introduction Individual plants are often multi-stemmed through the production of basal suckers (Fig. 1). Vegetative production of suckers and the longevity of stems imply that individual plants can persist over long periods. Like all cycads, E. eugene-maraisii is dioecious although cones are produced infrequently. Leaves are produced from the stem apex with new leaf flushes first appearing in October and peaking in January-February. Leaves are initially soft and harden with maturity and can remain alive for many years. Encephalartos eugene- maraisii has a very slow growth rate, producing only one set of new leaves annually with intervals between leaf flushes spanning several years. No seedlings have been recorded in recent history. Seeds produced naturally do not germinate, whereas cones pollinated by hand produce viable seeds, indicating that pollinators may be absent (Bezuidenhout 2019). The slow-growing nature of E. eugene-maraisii and its lack of natural recruitment has made it especially sensitive to poaching. Zerenopsis lepida is abundant and widespread in habitats that are generally frost-free and moist in the eastern parts of southern Africa while previous reports indicate that it is absent Fig. 1  Characteristic structure of an Encephalartos eugene-maraisii plant (P.D. Janse van Rensburg) Fig. 1  Characteristic structure of an Encephalartos eugene-maraisii plant (P.D. Janse van Rensburg) 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2454 from drier and colder habitats above 1500 m a.s.l., including the Waterberg (Staude and Sih­ vonen 2014). The increasing availability of cultivated cycads (native and non-native) as host plants in transformed habitats may have potentially facilitated the expansion of its range (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). Host range expansion by Z. lepida is potentially problematic because herbivory is often severe. In the absence of sec­ ondary host plants, Z. lepida larvae complete their entire life cycle on cycads, often causing extensive damage. However, even though high levels of damage from Z. lepida have been observed on cultivated cycads, no published references addressed Z. lepida damage under field conditions. In this paper, we present quantitative estimates of Z. lepida damage on E. eugene-maraisii under field conditions. Field observations indicate that not all individuals in the population are equally damaged. We aim to investigate whether differences in her­ bivory are either random, dependent on plant features or related to environmental variables. 1 3 Herbivory scoring Due to the low number of plants remaining in Marakele and difficulty in reaching plants, damage surveys were confined to Entabeni. Marakele was, however, visited and although damaged leaves were observed no new leaves and no eggs or larvae could be located to confirm the presence of Z. lepida. A total of 227 plants were selected at Entabeni after which the geographic position of each plant was recorded and a photo was taken for future identification of each of these plants. Herbivory scoring was conducted over two consecu­ tive seasons (2020–2021 and 2021–2022). The scoring was done between March and April since, during this period, most of the new leaves were fully expanded and the majority of larvae have moved off of plants. Scoring of damage severity could be done up to 3 months after herbivory and was postponed this long to avoid underestimating damage on new leaves because feeding lesions may expand during leaf expansion. Damage scoring during the 2020–2021 season was done for new and old leaves while only new leaves were assessed during the 2021–2022 season. New leaves refer to the leaves produced in the current year and are distinctively coloured for several months. Old leaves were those that developed during previous years. The feeding damage symptoms of Z. lep­ ida herbivory are easy to observe and remain visible for several years, due to the toughness of leaves. The level of herbivory was expressed as incidence (presence or absence of herbivory) and level of damage (percentage of leaf area removed). To determine the level of damage the percentage of leaf area removed for each leaf was estimated visually using different damage classes: 0%, 1–25%, 26–50%, 51–75%, and > 75%. These broad herbivory bins are effective for higher levels of herbivory and although potentially unsuited for estimating low levels of herbivory, we rarely observed leaves with low levels of herbivory (1–5%) due to the close association of Z. lepida with the new and expanding leaves of E. eugene-maraisii and the gregarious feeding behaviour of larvae. Visual damage estimation has been used in similar herbivory studies and was shown to be reliable (Johnson et al. 2016; Barrio et al. 2017; Brezzi et al. 2017; Castagneyrol et al. 2018). To avoid bias a single person performed all the herbivory assessments. Herbivory scoring We calculated the percentage of leaf area consumed by larvae by multiplying the number of leaves from each damage class with the midpoint of each damage class category e.g., 13% for the 1–25% class (Kozlov et al. 2015). The values of all classes were then summed and divided by the total number of leaves per stem. Study sites Historically, fire has been a regular occurrence in the study areas due to a very high frequency of lightning strikes in this mountainous region. Fire likely plays an important role in the patterns and processes of cycad habitats, for example, stimulating the production of new leaves and cones (Cousins and Witkowski 2017) and fire scars were visible on cycads in both populations. than 400 mm per year. Mean frost days per annum is 16 days with the earliest occurrence on 14 May and the latest occurrence on 23 July (Van Staden et al. 2021). Historically, fire has been a regular occurrence in the study areas due to a very high frequency of lightning strikes in this mountainous region. Fire likely plays an important role in the patterns and processes of cycad habitats, for example, stimulating the production of new leaves and cones (Cousins and Witkowski 2017) and fire scars were visible on cycads in both populations. Study sites Encephalartos eugene-maraisii is distributed in isolated sub-populations in the Waterberg Mountain range in the Limpopo province, South Africa (Bezuidenhout et al. 2020). Two main conservation areas (Marakele National Park and Entabeni Safari Conservancy) remain in the wild situated at either extreme of the species’ geographical range. Detailed locality information is not revealed to protect sensitive information on the distribution of remaining E. eugene-maraisii plants. Both areas form part of the Waterberg Biosphere Reserve. Entabeni contains the weight of E. eugene-maraisii distribution as a species and is situ­ ated at the north-eastern extreme of its species distribution. Most of the remaining plants occur at high altitudes (1500–1750 m a.s.l.) where it grows on rocky outcrops in mainly the Waterberg-Magaliesberg Summit Sourveld (Gm 29) (Mucina et al. 2006). The vegetation is characterised by patches of open woodland of Protea caffra and open shrubland of Englero­ phytum magalismontanum and Ancylobotrys capensis (Steyn and Bezuidenhout 2020). The plants are growing in three main groups on individual rocky outcrops that are separated by large flood plains and were treated as three sub-populations (A1-A3). Marakele is situated at the southwestern extreme of the distribution of E. eugene-maraisii in formally protected areas managed by SANParks. The site consists of small isolated sub­ populations separated by mountains with rugged topography and steep cliffs in some areas. In a phytosociological classification, mapping and description of the Waterberg Mountain vegetation 12 plant communities were identified. One of these plant communities is the Indigofera mollicoma-Encephalartos eugene-maraisii shrubland, which is characterised by the presence of E. eugene-maraisii along with the trees Heteropyxis natalensis and Protea caffra and the grasses Themeda triandra, Loudetia simplex and Trachypogon spicatus (Van Staden et al. 2021). The climate of both sites is warm in summer and cold and prone to frost in winter. Winter temperatures (coldest months June–July) are as low as − 3.7 °C, whilst the summer tem­ peratures (warmest months December to February) can reach a maximum temperature of 40.0 °C recorded at the Marakele weather stations. The rainfall, which is mainly during the spring and summer (September–March) seasons, can range from as high as 1196.9 to lower 1 3 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2455 than 400 mm per year. Mean frost days per annum is 16 days with the earliest occurrence on 14 May and the latest occurrence on 23 July (Van Staden et al. 2021). Statistical analysis Data were analyzed using non-parametric methods since the assumption of normality was not met for either data set. The level of damage was compared between plant sex, altitude, aspect, host-plant density and density of surrounding vegetation and tested for significant differences with the Kruskal-Wallis test. We analysed correlations between the incidence of herbivory and the level of damage with the different plant size parameters, using Spear­ man rank correlation analysis. Finally, we used Chi-square (χ2) analysis to compare the frequency of new leaf flushes and the incidence of herbivory on new leaf flushes between subpopulations (site A1-A3). All analyses were done using SPSS version 27 (IBM Corp 2020). Descriptive data A total of 227 plants were counted comprising 692 stems. No seedlings could be found. The total number of stems/suckers on an individual plant ranged from one to eight (3.01 ± 1.57 [mean ± SD]) and only 18.94% of the plants were single-stemmed. Stem length averaged 63.37 ± 75.13 cm and the longest stem recorded was 370 cm. By searching plants for the presence of live cones and old decaying cone material, the sex of 86 plants (37.89%) could be identified. Of these, 49 (56.98%) were male and 37 (43.02%) were female. The smallest stem with evidence of a previous coning event was 40 cm in height. This was taken as the minimum size of mature stems. In the 2020–2021 season, 335 stems flushed new leaves and 522 stems had old leaves while only 179 flushed leaves in the 2021–2022 season and 573 stems had old leaves (Table 1). Most new leaf flushes occurred in the summer with the first flushes appearing in October and the last flushes in March. Apparency of plants We also tested whether the apparency of plants affected the level of herbivory. The appar­ ency of plants is affected by several factors including the plant size, host-plant density and density of surrounding vegetation. Five plant size parameters were taken into account: (1) 3 Biodiversity and Conservation (2023) 32:2451–2468 2456 the total number of leaves per plant, (2) the number of stems/suckers per plant, (3) basal circumference of the largest stem, (4), the length of the longest stem (distance between stem base and the base of the most distal petiole) and since stems were often procumbent the (5) vertical plant height (distance between the ground and the tip of the most distal leaflet of the tallest stem) were also measured. Based on heat maps of existing plants, we rated dense and sparse host-plant density areas, and areas in between as intermediate. The density of sur­ rounding vegetation was scored as dense (host plant was completely covered with foliage from other plants), moderate (intermediate covering), and sparse (host was not covered by any other vegetation). Field observations of leaves per stem 40 62 74 35 52 58 Date/location of egg clusters Number of egg clusters Mean number of eggs per cluster (± SD) Total num­ ber of eggs October-December 40 (76.92%) 119.93 ± 48.57 4917 January-March 12 (23.08%) 110.91 ± 17.35 1220 New leaves 34 (65.38%) 123.32 ± 42.52 4193 Old leaves 18 (34.62%) 108.00 ± 45.79 1944 Upper surface 10 (19.23%) 123.50 ± 55.75 1235 Lower surface 42 (80.77%) 116.00 ± 41.24 4902 Total 52 118.02 ± 43.85 6137 Table 2  Number of eggs and egg clusters counted on Encepha­ lartos eugene-maraisii plants during the 2021–2022 season Table 1  Numbers of new and old leaves on stems of Encephalartos eugene-maraisii counted during 2020– 2021 and 2021–2022 Date/location of egg clusters Number of egg clusters Mean number of eggs per cluster (± SD) Total num­ ber of eggs October-December 40 (76.92%) 119.93 ± 48.57 4917 January-March 12 (23.08%) 110.91 ± 17.35 1220 New leaves 34 (65.38%) 123.32 ± 42.52 4193 Old leaves 18 (34.62%) 108.00 ± 45.79 1944 Upper surface 10 (19.23%) 123.50 ± 55.75 1235 Lower surface 42 (80.77%) 116.00 ± 41.24 4902 Total 52 118.02 ± 43.85 6137 Table 2  Number of eggs and egg clusters counted on Encepha­ lartos eugene-maraisii plants during the 2021–2022 season October to December. The majority of egg clusters (80.77%) were located on the underside of leaves and several clusters were often found on a single leaf flush. The majority (65.38%) of egg clusters were recorded from new leaves (Table 2). Although egg clusters were often found on old and even dead leaves, they were always laid on stems with robust developed cataphylls that were about to either push leaves or cones (data not shown). Subsequent visits to these plants revealed no new damage to old leaves. Four cases of early-instar larvae attacking the crown of the plant were observed. After a seven-day period, no larvae could be observed on any of these plants. Subsequent visits to these plants revealed the plants were still capable of flushing new leaves in the same season despite damage to the crown although leaf flushing was delayed. Four cases were recorded where egg clusters were laid on plants in the process of pushing male cones. Only a single female plant produced cones during the study period and no eggs or larvae could be located on this plant. Field observations Field observations were made of all development stages of Z. lepida. No adults could be observed during the winter months probably due to very low temperatures and frost that occur during some nights. Egg clusters were first observed in October when new leaf flushes first appear. Eggs were laid in clusters containing from 17 to 231 eggs (118.02 ± 43.85) (Table 2). The majority (76.92%) of egg clusters were observed early in the season from 3 1 Biodiversity and Conservation (2023) 32:2451–2468 2457 Table 1  Numbers of new and old leaves on stems of Encephalartos eugene-maraisii counted during 2020– 2021 and 2021–2022 Leaf numbers 2020–2021 2021–2022 New Old Total New Old Total No. of stems with leaves 335 (48.41%) 522 (75.43%) 665 (96.10%) 179 (25.87%) 573 (82.80%) 653 (94.36%) Total no. of leaves counted 4854 7772 12,626 2084 9150 11,234 Mean no. of leaves per stem (± SD) 14.49 ± 9.63 14.89 ± 12.71 18.46 ± 15.12 11.64 ± 8.79 15.75 ± 11.18 16.18 ± 12.75 Mean no. of leaves per mature stem (± SD) 21.80 ± 5.58 22.83 ± 12.51 29.88 ± 12.51 21 ± 4.91 23.27 ± 8.52 25.76 ± 9.80 Maximum no. Field observations Early instar larvae were attacking emerging male cones by scraping the sur­ face of sporophylls but damage to the cones was only superficial. Larvae could also not be located on these plants after a seven-day period. Early instar larvae fed gregariously on new leaves by scraping the surface beneath leaf­ lets producing irregular patches (Fig. 2). The larvae spin a fine mat of silk on which they 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2458 rest on the underside of leaves. Late instar larvae feed on the edge of leaflets often causing severe damage. As larvae became larger the number of larvae per group decreased. When larvae hatched on very young leaves (leaflets still folded tightly against each other) damage tended to be severe and the entire leaf flush was destroyed. Once the leaves have extended beyond ∼50 cm in length and started to harden, larvae started dispersing either to neighbour­ ing plants with new emerging leaves or secondary host plants. Of the known secondary host plants, Apodytes dimidiata and Diospyros lycioides were found in the study area. Apodytes dimidiata is abundant in the study area and Z. lepida larvae (≥ fourth instar) were often observed feeding on A. dimidiata. We observed no larvae feeding on D. lycioides. No other host plant species were recorded. Table 3  Incidence of herbivory and estimated percentage leaf area of Encephalartos eugene- maraisii consumed by Zerenop­ sis lepida Levels of herbivory The severity of damage to old leaves measured in the 2020–2021 season was lower than on new leaves for that season (Table 3). Estimates of old leaves likely under-estimated some aspects of damage because we recorded several instances where new leaf flushes were extensively damaged or destroyed that would otherwise not be visible on old leaf flushes. Incidence of herbivory on new leaf flushes ranged from 40.60% on new leaves from 2020 to 2021 to 40.78% on new leaves from 2021 to 2022 (Table 3). While the leaves on some Fig. 2  Different development stages of Zerenopsis lepida observed in the Waterberg: a adult; b egg clus­ ter; c newly hatched larvae; d mature larvae on cycad host; e mature larvae on secondary host plant (P.D. Janse van Rensburg) Fig. 2  Different development stages of Zerenopsis lepida observed in the Waterberg: a adult; b egg clus­ ter; c newly hatched larvae; d mature larvae on cycad host; e mature larvae on secondary host plant (P.D. Janse van Rensburg) 1 Biodiversity and Conservation (2023) 32:2451–2468 2459 cycad stems were heavily attacked, neighbouring stems or plants with newly flushed leaves often remained free of damage. The leaves from the majority of the stems did not have heav­ ily damaged leaves (> 25% leaf area removed). The mean amount of photosynthetic tissue removed ranged from 9.01% on new leaves from the 2020–2021 season to 10.61% on new leaves from the 2021–2022 season (Table 3).i cycad stems were heavily attacked, neighbouring stems or plants with newly flushed leaves often remained free of damage. The leaves from the majority of the stems did not have heav­ ily damaged leaves (> 25% leaf area removed). The mean amount of photosynthetic tissue removed ranged from 9.01% on new leaves from the 2020–2021 season to 10.61% on new leaves from the 2021–2022 season (Table 3).i There was no significant association between the level of damage and aspect, altitude and the sex of plants (data not shown). A Kruskal-Wallis test did, however, show significant dif­ ferences in the level damage between different densities of surrounding vegetation for new leaves 2020–2021 (H = 7.132, P = 0.028), new leaves 2021–2022 (H = 11.529, P = 0.003), total leaves 2020–2021 (H = 9.059, P = 0.011) but not for old leaves 2020–2021 (H = 4.415, Level of her­ bivory 2020–2021 2021–2022 New Old Total New No. Levels of herbivory of stems with signs of leaf her­ bivory (Inci­ dence) 136 (40.60%) 200 (38.31%) 301 (44.01%) 73 (40.78%) No. of stems heavily damaged (> 25% leaf area removed) 38 (11.34%) 43 (8.24%) 63 (9.21%) 37 (20.67%) No. of leaves with signs of her­ bivory 900 (18.54%) 1375 (17.67%) 2275 (18.01%) 543 (26.05%) No. of heavily damaged leaves (> 25%) 528 (10.87%) 765 (9.84%) 1293 (10.24%) 414 (19.86%) Mean % leaf area lost per stem (stems with signs of leaf her­ bivory) (± SD) 22.27 ± 24.17 17.62 ± 18.75 16.81 ± 17.81 26.02 ± 15.91 Mean % leaf area lost per stem (all stems) (± SD) 9.01 ± 18.85 6.81 ± 14.46 7.62 ± 14.62 10.61 ± 16.34 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2460 P = 110). Pairwise comparisons indicated significantly less leaf area consumed on cycads that are surrounded by dense vegetation compared to sparse vegetation (Table 4). There were also significant differences in the level damage between different host-plant densi­ ties for old leaves 2020–2021 (H = 13.694, P = 0.001), new leaves 2020–2021 (H = 11.278, P = 0.004), total leaves 2020–2021 (H = 10.029, P = 0.007) but not for new leaves 2021– 2022 (H = 1.008, P = 0.604). Pairwise comparisons indicated that significantly less herbivory occurred in low host-density areas compared to high host-density areas (Table 4).i P = 110). Pairwise comparisons indicated significantly less leaf area consumed on cycads that are surrounded by dense vegetation compared to sparse vegetation (Table 4). There were also significant differences in the level damage between different host-plant densi­ ties for old leaves 2020–2021 (H = 13.694, P = 0.001), new leaves 2020–2021 (H = 11.278, P = 0.004), total leaves 2020–2021 (H = 10.029, P = 0.007) but not for new leaves 2021– 2022 (H = 1.008, P = 0.604). Pairwise comparisons indicated that significantly less herbivory occurred in low host-density areas compared to high host-density areas (Table 4).i Significant correlations were found between the incidence of Z. lepida herbivory and the level of damage with the number of leaves, the number of stems, stem circumference and stem height (Table 5). Levels of herbivory The strongest correlations were recorded between the number of leaves and the incidence of herbivory for new leaves 2020–2021 (r = 0.476, P = < 0.001) and total leaves 2020–2021 (r = 0.504, P = < 0.001) and between the number of stems with the incidence of herbivory for new leaves 2020–2021 (r = 0.451, P = < 0.001) and total leaves 2020–2021 (r = 0.452, P = < 0.001). The vertical plant height only had significant correla­ tions with both the levels of herbivory (r = 0.255, P = 0.019) and incidence of herbivory (r = 0.216, P = 0.049) with new leaves from the 2021–2022 season.l The frequency of new leaf flushes varied dramatically between sub-populations dur­ ing both the 2020–2021 (χ2 = 77.092, df = 2, P < 0.001) and 2021–2022 (χ2 = 54.578, df = 2, P < 0.001) seasons (Table 6). The proportion of stems with new leaf flushes that were dam­ aged by Z. lepida decreased with an increasing number of stems flushing new leaves in 2020–2021 (χ2 = 16.431, df = 2, P < 0.001) but not significantly in 2021–2022 (χ2 = 4.117, df = 2, P = 0.128). (Table 6). Field observations The biology of Z. lepida has been studied and described in previous studies (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). In this study, we report on a few idiosyncra­ sies regarding the behaviour of Z. lepida in the Waterberg region.f The biology of Z. lepida has been studied and described in previous studies (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). In this study, we report on a few idiosyncra­ sies regarding the behaviour of Z. lepida in the Waterberg region.f Zerenopsis lepida is widespread and has colonized a broad range of cycads from different clades and lineages in East and Southern Africa. This is the first record of Z. Field observations lepida in the Waterberg region of the Limpopo Province, which supports earlier indications that its host Table 4  Pairwise comparisons of the effect of host-plant density and the density of surrounding vegetation on the percentage leaf area of Encephalartos eugene-maraisii consumed by Zerenopsis lepida Density Sparse Intermediate Dense P Host-plant density Old 2020–2021 2.50 ± 6.15a 7.29 ± 15.64b 7.91 ± 14.93b 0.001* New 2020–2021 5.18 ± 13.27a 8.40 ± 16.54ab 12.39 ± 24.47b 0.004* Total 2020–2021 4.09 ± 6.45a 6.84 ± 9.13ab 9.72 ± 13.13b 0.007* New 2021–2022 9.64 ± 14.65 11.98 ± 17.62 9.77 ± 15.83 0.604 Density of surrounding vegetation Old 2020–2021 7.56 ± 11.71 7.77 ± 12.55 6.34 ± 16.38 0.110 New 2020–2021 10.84 ± 18.98a 6.90 ± 10.89ab 0.59 ± 1.68b 0.028* Total 2020–2021 8.29 ± 10.49a 6.80 ± 7.08a 5.84 ± 16.51b 0.011* New 2021–2022 13.99 ± 15.21a 5.69 ± 10.60b 3.10 ± 8.59b 0.003* Means with different superscript symbols indicate significant differences le 4  Pairwise comparisons of the effect of host-plant density and the density of surrounding vegetation on percentage leaf area of Encephalartos eugene-maraisii consumed by Zerenopsis lepida 1 Biodiversity and Conservation (2023) 32:2451–2468 2461 Table 5  Effect of plant apparency on the incidence and the level of damage from Zerenopsis lepida herbivory Plant size Incidence Level of damage 2020–2021 2021– 2022 2020–2021 2021– 2022 New Old Total New New Old Total New Number of leaves r 0.476 0.334 0.504 0.319 0.389 0.215 0.320 0.295 P < 0.001* < 0.001* < 0.001* 0.003* < 0.001* 0.002* < 0.001* 0.007* Number of stems r 0.451 0.229 0.452 0.201 0.379 0.203 0.339 0.170 P < 0.001* 0.001* < 0.001* 0.067 < 0.001* 0.004* < 0.001* 0.123 Stem circumference r 0.207 0.239 0.231 0.292 0.163 0.176 0.184 0.303 P 0.020* 0.001* < 0.001* 0.010* 0.068 0.014* 0.007* 0.007* Stem height r 0.194 0.196 0.257 0.302 0.098 0.090 0.132 0.315 P 0.025* 0.005* < 0.001* 0.005* 0.261 0.203 0.047* 0.004* Vertical plant height r 0.105 0.116 0.126 0.216 0.064 0.087 0.075 0.255 P 0.234 0.100 0.057 0.049* 0.467 0.216 0.259 0.019* Asterisks represent significant relationships Table 6  Percentage of Encephalartos eugene-maraisii stems with new leaf flushes and percentage of new leaf flushes with signs of Zerenopsis lepida herbivory Leaf and herbivory phenology 2020–2021 2021–2022 Site A1 Site A2 Site A3 Site A1 Site A2 Site A3 Stems with new leaf flushes Count 68 108 159 100 49 30 % 28.9 48.9 69.7 42 21.90 12.90 New leaf flushes with signs of herbivory Count 41 43 49 42 15 17 % 60.3 39.8 30.8 42 30.6 53.3 Leaf area removed % 15.86 5.92 4.92 14.10 7.82 15.78 le 6  Percentage of Encephalartos eugene-maraisii stems with new leaf flushes and percentage of new leaf hes with signs of Zerenopsis lepida herbivory and distribution range is expanding (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). Field observations This species probably arrived in the study area only during the last few years since it seems unlikely that its presence went unnoticed during previous surveys of E. eugene- maraisii in this region. Its ability to colonize new regions is probably aided by the increased use of its cycad hosts as ornamental plants in home gardens and the movement of plants between regions. and distribution range is expanding (Donaldson and Bösenberg 1995; Staude and Sihvonen 2014). This species probably arrived in the study area only during the last few years since it seems unlikely that its presence went unnoticed during previous surveys of E. eugene- maraisii in this region. Its ability to colonize new regions is probably aided by the increased use of its cycad hosts as ornamental plants in home gardens and the movement of plants between regions. Wu et al. (2010) discussed a similar case of breakaway populations of Chilades pandava on cycads in Taiwan that were caused by range expansions of native populations of C. pan­ dava. Host range expansion of native herbivores may have certain consequences for both the host and herbivore populations. Breakaway populations colonize host population previ­ ously unaffected by Z. lepida, including cultivated cycads in cities and will likely exhibit lower genetic diversity than wild populations (Olivier 2006; Wu et al. 2010). However, these breakaway population may also function as a bridge between natural populations of Z. lepida that have historically been isolated and have unique genetic features. This might increase interpopulation movement threatening the unique isolated wild populations of Z. lepida with genetic contamination (Olivier 2006; Staude and Sihvonen 2014). In addition to the threat of host range and distribution expansion, introduced cycad hosts in home gar­ dens may increase the biomass of Z. lepida increasing its herbivory pressure on wild cycad populations. 1 3 2462 Biodiversity and Conservation (2023) 32:2451–2468 Records generally indicate that Z. lepida is active throughout the year with two peak flight periods in mid-summer and autumn (Staude and Sihvonen 2014). However, the cold and frost-prone habitat of E. eugene-maraisii and the unavailability of food in the winter months means the area is likely unsuitable for Z. lepida to be active year-round and both immature and mature stages are only present during the spring and summer months of South Africa. Moths first appeared in the spring between October and November. Field observations The majority of egg clusters were found early in the season between October and December and new egg clusters were rarely observed later than January-February. Donaldson and Bösenberg (1995) indicated that they could not locate any egg clusters on the leaves of Encephalartos altensteinii once its leaves had extended beyond 30 cm. How­ ever, we frequently recorded egg clusters on old leaves during the early months of summer. During this period new leaf flushes are still scarce and the ovipositing females likely have no choice in laying eggs on old leaves. Similarly, Eumaeus atala (Lepidoptera: Lycaenidae) were also observed laying eggs on older leaves and even stems and cones of Zamia integri­ folia when the availability of new leaf flushes was low (Koi and Daniels 2015). Although eggs were laid on old leaves, the stems of plants that they oviposited on had robust cataphyll complexes that were in the early stages of either flushing leaves or push­ ing cones. However, at such an early stage of an organ flush, the moths did not seem to be able to distinguish between a leaf flush or cone flush as several egg clusters were deposited on plants that were in the process of pushing cones. While several egg clusters were often recorded on single stems that were in the process of a new organ flush, no clusters were observed on adjacent stems or neighbouring plants that were not about to undergo an organ flush. Various Lepidoptera species select young leaves for oviposition. Young leaves are thought to be selected because they provide a better food source for young larvae and they also lack mechanical protection. Cycads produce rich amounts of unique secondary compounds that may impact herbivore populations (Snyder and Marler 2011; Whitaker et al. 2022). It’s been suggested that the host preference of Z. lepida and other cycadivorous Lepidoptera may be based on a factor ubiquitous in most cycad lineages such as MAM glycosides (Donaldson and Bösenberg 1995; Koi 2017). Before a new leaf or cone is flushed, the cataphyll complex becomes mitotic and robust, mobilizing MAM-glycosides which may inadvertently func­ tion as a chemical cue to stimulate Z. lepida oviposition and feeding behaviour. The MAM glycoside concentrations tend to decrease as leaves expand and become older (Prado et al. 2014). Levels of herbivory The proportion of leaf area removed by larval feeding is a good measure of the loss of pho­ tosynthetic tissues. Herbivory by Z. lepida is easy to identify by searching leaves for typical feeding damage which persists for several years. Over two years we reported that a mean leaf area ranging from 9.01 to 10.61% was eaten away which seems fairly high compared to losses of foliage due to background herbivory that is generally reported in terrestrial ecosystems (Turcotte et al. 2014; Kozlov et al. 2015; Meyer et al. 2017; Zang et al. 2017; Gálman et al. 2018). Although very few studies have investigated the level of herbivory by folivorous cycad consumers, it falls into the same range as estimates in other reported cases for cycads (Clark and Clark 1991; Beltrán-Valdez and Torres-Hernández 1995; Prado et al. 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2463 2014; Santos Murgas and Abrego 2016). However, even minor, but chronic herbivory may have significant effects on the growth and reproduction of plants (Baron and Bros 2005; Kettenring et al. 2009; Zvereva et al. 2012; Myers and Sarfraz 2017). This may have far- reaching effects on rare plant populations such as the already endangered E. eugene-maraisii. 2014; Santos Murgas and Abrego 2016). However, even minor, but chronic herbivory may have significant effects on the growth and reproduction of plants (Baron and Bros 2005; Kettenring et al. 2009; Zvereva et al. 2012; Myers and Sarfraz 2017). This may have far- reaching effects on rare plant populations such as the already endangered E. eugene-maraisii. Cycads are slow-growing with long-lived leaves and make large investments in defence. Plant species with slow growth rates and low turnover rates of tissue in resource limited environments are expected to make large investments in anti-herbivore defences because the potential impact of herbivory could be very high (Endara and Coley 2011). Species such as this generally exhibit lower levels of herbivory (Cebrián and Duarte 1994; Fine et al. 2006). It is expected that fast-growing plant species are capable of tolerating higher levels of herbivory because they can more quickly replace lost leaf area while the health costs of a given level of herbivory will be higher for slow-growing species (Massad 2013; Gianoli and Salgado-Luarte 2017). Due to the slow turnover of plant tissue, slow-growing species might also take longer to recover should herbivory occur. Levels of herbivory Furthermore, even minor herbivory may reduce leaf life span (Zvereva and Kozlov 2014). Levels of damage may have been under­ estimated because measures of leaf area loss do not record leaves that were prematurely senesced due to herbivory. Damage may further be underestimated if damage to the crown and potential damage to cones is considered. Flushing new leaves expose the vulnerable growth point for potential stem damage and death (Clark and Clark 1991). We reported several instances of larvae attacking the crowns of plants in the early stages of leaf flushes. These plants were moni­ tored and although damage to the crown delayed the upcoming leaf flushes, plants were still able to flush new leaves. Zerenopsis lepida has previously been reported to opportunistically feed on female cones and burrow in the crown of Encephalartos species although damage is generally low (Donaldson and Bösenberg 1995; Zunckel 1995; Staude and Sihvonen 2014). Only one female plant coned during the study period but no Z. lepida larvae were present. However, we observed early instar Z. lepida larvae attacking emerging male cones, but that damage was only superficial.l Since cycads do not flush leaves continuously, plants are not able to regenerate damaged leaves until the following flush. During the 2020–2021 season, new leaf flushes from 38 stems were heavily damaged (> 25% leaf area removed). Six of those re-flushed leaves dur­ ing the 2021–2022 season. For most, however, it may take several years to recover and flush new leaves depending on the severity of the damage. Repeated severe damage will likely reduce growth and plant health. This may also delay coning and reduce the reproductive output of E. eugene-maraisii. Due to the absence of pollinators, there is no natural recruit­ ment. Recovery programs are in place (Bezuidenhout 2019) but are dependent on wild plants coning in order to perform hand pollination. Cone production is already a rare event which may be exacerbated by chronic herbivory. However, the chances of experiencing severe herbivory in successive leaf flushes also appear to be low, e.g., of the 43 stems that had heavily damaged old leaves in 2020–2021, 27 flushed new leaves during the 2020–2021 and 2021–2022 seasons and only four of those were heavily damaged. Effect of plant apparency The level of damage was significantly higher on more apparent plants. An individual plant may be more apparent if it is taller, multi-stemmed, occurs in high host-density areas and 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2464 surrounded by sparse vegetation. This supports earlier observations that female Z. lepida moths avoid ovipositing on cycads under a closed canopy (Staude and Sihvonen 2014). This is also similar to findings by Bayliss et al. (2009) who found that the incidence of larval damage of Callioratis grandis (Lepidoptera: Geometridae) was associated with more mature individuals of its cycad host Encephalartos gratus. While large plants may have certain advantages such as high competitiveness or increased fecundity, they may also have higher costs such as increased apparency to herbi­ vores (Blanckenhorn 2000). Larger plants may be easier for herbivores to locate and colo­ nize and larger plants may also be more attractive since they provide greater amounts of resources (Schlinkert et al. 2015). Therefore, a trade-off might exist between the benefits of plant size and the cost of herbivory. The effects of herbivory are likely counterbalanced by increased abilities for tolerance of herbivory damage due to their larger size and capacity for storage of energy (Boege and Marquis 2005) and higher ability to compete with surrounding vegetation for resources (Schlinkert et al. 2015). 1 3 Potential regulatory role of synchronized leaf flushes In addition to the production of secondary metabolites, cycad leaves are also mechanically protected through leaf toughness and spines (Prado et al. 2014, 2016). Cycad leaves are initially soft and harden with maturity. Young leaves are vulnerable to herbivore attack dur­ ing leaf expansion when the majority of cycad folivorous arthropods consume new leaves (Staude and Sihvonen 2014; Prado et al. 2014; Koi and Daniels 2015). Two phenological defence mechanisms are suggested to reduce damage to expanding leaves i.e., the produc­ tion of leaves when herbivore abundance is low and satiating herbivores with a superabun­ dance of edible tissue by synchronising leaf flushes (Clark and Clark 1991; Gong and Zhang 2014; Lamarre et al. 2014).l Our results support the idea that synchronised leaf flushes may satiate herbivore popula­ tions. The results were especially visible for site A3 during 2020–2021 when the majority of stems flushed new leaves and correspondingly the proportion of plants with damage and the severity of damage was the lowest in site A3. During mass leaf flushing events the sheer number of new leaves would satiate the Z. lepida population meaning the majority of plants escape damage. Unsynchronised leaf flushes would lead to a yearly steady supply of new leaves which should lead to a consistently high population of Z. lepida. Synchronised leaf flushes at long unpredictable intervals would ensure difficulty for adult moths to locate suit­ able sites for larval development (Clark and Clark 1991). A possible explanation for the higher synchrony at site A3 is a severe hail storm that occurred at the start of the 2020–2021 season that severely damaged and defoliated sev­ eral plants. Environmental disturbances (e.g., fire and strong winds) causing leaf damage and defoliation have been reported to result in a synchronised leaf flush afterwards (Hirsh and Marler 2002). The habitat of E. eugene-maraisii is fire-prone due to frequent lightning strikes, and fire likely plays an important role in the ecological processes in its habitat. Fire scars occur on the majority of plants in both populations and the plants are seemingly well adapted to fire. However, the majority of the Entabeni sites haven’t burned in recent years. Fire could potentially synchronise leaf flushes (Swart et al. 2019). Manipulating the timing of fire may also be used to ensure the majority of new leaf flushes emerge outside peak herbivore densities (Zunckel 1995). We observed the majority of Z. Potential regulatory role of synchronized leaf flushes lepida egg clus­ 1 3 1 3 1 Biodiversity and Conservation (2023) 32:2451–2468 2465 ters between October and December. Stems that flush leaves later in the season may avoid severe damage. ters between October and December. Stems that flush leaves later in the season may avoid severe damage. Conclusion The colonization by Z. lepida of this region where one of the last populations of E. eugene- maraisii occur adds to the threat against this species. The levels of damage recorded in this study are fairly high compared to what is often reported for terrestrial plants, and may be especially important in the context of E. eugene-maraisii. Slow growing species such as E. eugene-maraisii that make large investments in defence and have low turnover rates of tissue will take long to recover should herbivory occur. Damage to the crowns of plants delays upcoming leaf flushes, and plants are not able to regenerate damaged leaves until the following flush. Since some plants may take several years to recover and flush new leaves, repeated severe damage will likely reduce growth and plant health which may delay coning and reduce the reproductive output of E. eugene-maraisii. This study emphasises that young leaves are an ephemeral resource and that aspects of leaf phenology may serve as a defence mechanism. Plants may escape high levels of damage by producing leaves in synchrony to satiate herbivores or leaves may be produced during periods when herbivore abundances are low. Lastly, we recorded that Z. lepida feeding damage increases with plant size and canopy openness, as the apparency of large and open plants is likely more attractive to Z. lepida and larger plants likely provide greater amounts of resources. Acknowledgements  Comments on the manuscript from Bianca Greyvenstein are appreciated. We are thank­ ful to Simphiwe Dubazane for assisting us during field surveys. This project was supported by the Wild Cycad Conservancy. Author Contribution  All authors contributed to the study’s conception and design. Data collection and analy­ sis were performed by P.D. Janse van Rensburg with assistance from J. Van den Berg and H. Bezuidenhout. The first draft of the manuscript was written by P.D. Janse van Rensburg and all authors commented on previ­ ous versions of the manuscript. All authors read and approved the final manuscript. Funding  Open access funding provided by North-West University. Data Availability  The datasets generated during the study are available from the corresponding author on reasonable request. Declarations Ethics approval and research permits  Ethics approval (no: NWU-01552-20-A9) for this study was granted by the North-West University, Faculty of Natural and Agricultural Sciences Ethics Committee (FNASREC). A permit (no: ZA/LP/111179) to do scientific research on plants in the Limpopo Province of South Africa was granted by the Department of Economic Development, Environment and Tourism (LEDET). Competing interests  The authors have no relevant financial or non-financial interest to disclose. Open Access  This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is 3 3 Biodiversity and Conservation (2023) 32:2451–2468 2466 not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References Agrawal AA, Hastings AP, Johnson MTJ, Maron JL, Salminen JP (2012) Insect herbivores drive real-time ecological and evolutionary change in plant populations. Science 338:113–116. https://doi.org/10.1126/ science.1225977 Agrawal AA, Hastings AP, Johnson MTJ, Maron JL, Salminen JP (2012) Insect herbivores drive real-time ecological and evolutionary change in plant populations. Science 338:113–116. https://doi.org/10.1126/ science.1225977 Ancheta J, Heard SB (2011) Impacts of insect herbivores on rare plant populations. Biol Conserv 144:2395– 2402. https://doi.org/10.1016/j.biocon.2011.06.019l Ancheta J, Heard SB (2011) Impacts of insect herbivores on rare plant populations. Biol Conserv 144:2395– 2402. https://doi.org/10.1016/j.biocon.2011.06.019l Baron S, Bros SM (2005) Herbivory and the endangered robust spineflower (Chorizanthe robusta var. robusta). Madroño 52:46–52. https://doi.org/10.3120/0024-9637(2005)52[46:HATER]2.0.CO;2 Barrio IC, Lindén E, Te Beest M et al (2017) Background invertebrate herbivory on dwarf birch (Betula glandulosa-nana complex) increases with temperature and precipitation across the tundra biome. Polar Biol 40:2265–2278. https://doi.org/10.1007/s00300-017-2139-7 p g Bayliss J, Burrow C, Martell S, Staude H (2009) An ecological study of the relationship between two living fossils in Malawi: the Mulanje tiger moth (Callioratis grandis) and the Mulanje cycad (Encephalartos gratus). Afr J Ecol 48:472–480. https://doi.org/10.1111/j.1365-2028.2009.01135.x gratus). Afr J Ecol 48:472–480. https://doi.org/10.1111/j.1365-2028.2009.01135.x Beltrán-Valdez JM, Torres-Hernández L (1995) Aspectos cualitativos de la herbivoría de Eumaeus debora Hubner. (Lepidoptera: Lycaenidae) sobre Ceratozamia mexicana Bröng. (Cycadales: Zamiaceae). La ciencia y el Hombre 19:91–104 y Bezuidenhout H (2019) Supplementation of the Encephalartos eugene-maraisii population at Marakele National Park. Internal report 02 / 2019, Scientific Services, Kimberley, South African National Parks pi y Bezuidenhout H, Steyn T, Mamatsharaga T, Froneman W, Rodgers S, Lessing J, Van den Berg J, Van Eeden W (2020) Management Plan for the Encephalartos eugene-maraisii (Waterberg cycad) June 2020. Inter­ nal report 08 / 2020, Scientific Services, Kimberley, South African National Parks Blanckenhorn WU (2000) The evolution of body size: what keeps organisms small? Q Rev Biol 75:385–4 https://doi.org/10.1086/393620 p g Boege K, Marquis RJ (2005) Facing herbivory as you grow up: the ontogeny of resistance in plants. Trends Ecol Evol 20:441–448. https://doi.org/10.1016/j.tree.2005.05.001 Bösenberg JD (2022) Encephalartos eugene-maraisii. The IUCN Red List of Threatened Species 2022 eT41904A50906421. https://doi.org/10.2305/IUCN.UK.2022-1.RLTS.T41904A50906421.en. Accessed 10 March 2023 Brezzi M, Schmid B, Niklaus PA, Schuldt A (2017) Tree diversity increases levels of herbivore damage in a subtropical forest canopy: evidence for dietary mixing by arthropods? J Plant Ecol 10:13–27. https:// doi.org/10.1093/jpe/rtw038f g jp tagneyrol B, Jactel H, Moreira X (2018) Anti-herbivore defences and insect herbivory: interactive effects of drought and tree neighbours. J Ecol 106:2043–2057. References IBM Corp, New York p g j IBM Corp (2020) IMB SPSS Statistics for Windows, Version 27.0. IBM Corp, New York Johnson MTJ Bertrand JA Turcotte MM (2016) Precision and accuracy in quantifying herbivory E Johnson MTJ, Bertrand JA, Turcotte MM (2016) Precision and accuracy in quantifying herb Entomol 41:112–121. https://doi.org/10.1111/een.12280l Kettenring KM, Weekley CW, Menges ES (2009) Herbivory delays flowering and reduces fecundity of Liatris ohlingerae (Asteraceae), an endangered, endemic plant of the Florida scrub. J Torrey Bot Soc 136:350–362. https://doi.org/10.3159/08-RA-113.1l p g Koi S (2017) A butterfly picks its poison: cycads (Cycadaceae), integrated pest management and Eumaeus atala Poey (Lepidoptera: Lycaenidae). Entomol Ornithol Herpetol 6:191. https://doi. org/10.4172/2161-0983.1000191 Koi S, Daniels J (2015) New and revised life history of the Florida hairstreak Eumaeus atala (Lepidop­ tera: Lycaenidae) with notes on its current conservation status. Fla Entomol 98:1134–1147. https://doi. org/10.1653/024.098.0418 Kozlov MV, Lanta V, Zverev V, Zvereva EL (2015) Global patterns in background losses of woody plant f age to insects. Global Ecol Biogeogr 24:1126–1135. https://doi.org/10.1111/geb.12347 Lamarre GPA, Mendoza I, Fine PVA, Baraloto C (2014) Leaf synchrony and insect herbivory among tree habitat specialists. Plant Ecol 215:209–220. https://doi.org/10.1007/s11258-013-0290-y Lamarre GPA, Mendoza I, Fine PVA, Baraloto C (2014) Leaf synchrony and insect herbivory among tropic tree habitat specialists. Plant Ecol 215:209–220. https://doi.org/10.1007/s11258-013-0290-y Louw SL, Armstrong AJ (2018) Field work undertaken during 2017–2018 on the critically endangered moth callioratis millari (Lepidoptera: Geometridae, Diptychininae). Metamorphosis 29:40–44f Massad TJ (2013) Ontogenetic differences of herbivory on woody and herbaceous plants: a meta-analysis demonstrating unique effects of herbivory on the young and the old, the slow and the fast. Oecologia 172:1–10. https://doi.org/10.1007/s00442-012-2470-1 Meyer ST, Scheithe L, Hertzog L, EbelingbA, Wagg C, Roscher C, Weisser WW (2017) Consistent increase in herbivory along two experimental plant diversity gradients over multiple years. Ecosphere 8:e01876. https://doi.org/10.1002/ecs2.1876 Moir ML, Vesk PA, Brennan KEC, Keith DA, McCarthy MA, Hughes L (2011) Identifying and managing threatened invertebrates through assessment of coextinction risk. Conserv Biol 25:787–796. https://doi. org/10.1111/j.1523-1739.2011.01663.x Moir ML, Coates DJ, Kensington WJ, Barrett S, Taylor GS (2016) Concordance in evolutionary history of threatened plant and insect populations warrant unified conservation management approaches. Biol Conserv 198:135–144. https://doi.org/10.1016/j.biocon.2016.04.012 p g j Mucina L, Hoare DB, Lötter MC et al (2006) Grassland Biome. In: Mucina L, Rutherford MC (eds) The vegetation of South Africa, Lesotho and Swaziland. References https://doi.org/10.1111/1365-2745.12956 Cebrián J, Duarte CM (1994) The dependence of herbivory on growth rate in natural plant communit Func Ecol 8:518–525. https://doi.org/10.2307/2390077 Clark DB, Clark DA (1991) Herbivores, herbivory and plant phenology: patterns and consequences in a tropi­ cal rain forest cycad. In: Price PW, Lewinsohn TM, Fernandes WG, Benson WW (eds) Plant-animal interactions: evolutionary ecology in tropical and temperate regions. John Wiley and Sons, New York, pp 209–225 pp Colling G, Matthies D (2004) The effects of plant population size on the interactions between the endan­ gered plant Scorzonera humulis, a specialised herbivore, and a phytophagous fungus. Oikos 105:71–78. https://doi.org/10.1111/j.0030-1299.2004.12783.x p g j Cousins SR, Witkowski ETF (2017) African cycad ecology, ethnobotany and conservation: a synthesis. Bot Rev 83:152–194. https://doi.org/10.1007/s12229-017-9183-4 Donaldson JS (2015) Cycads. In: Prinsloo GL, Uys VM (eds) Insects of cultivated plants and natural pastures in Southern Africa. Entomological Society of Southern Africa, Pretoria, pp 668–679 Donaldson JS, Bösenberg JD (1995) Life history and host range of the leopard magpie moth, Zerenopsis leopardina Felder (Lepidoptera: Geometridae). Afr Entomol 3:103–110. https://hdl.handle.net/10520/ AJA10213589_82f Elzinga JA, Turin H, Van Damme JMM, Biere A (2005) Plant population size and isolation affect herbivory of Silene latifolia by the specialist herbivore Hadena bicruris and parasitism of the herbivore by parasit­ oids. Oecologia 144:416–426. https://doi.org/10.1007/s00442-005-0096-2 1 3 Biodiversity and Conservation (2023) 32:2451–2468 2467 Endara MJ, Coley PD (2011) The resource availability hypothesis revisited: a meta-analysis. Funct Ecol 25:389–398. https://doi.org/10.1111/j.1365-2435.2010.01803.x Fine PVA, Miller ZJ, Mesones I, Irazuzta S, Appel HM, Stevens MHH, Sääksjärvi I, Schultz JC, Coley PD (2006) The growth-defence trade-off and habitat specialization by plants in amazonian forests. Ecology 87:S150–S162. https://doi.org/10.1890/0012-9658(2006)87[150:TGTAHS]2.0.CO;2 Gálman A, Abdala-Roberts L, Zhang S, Berny-Mier y Teran JC, Rasmann S, Moreira X (2018) A global analysis of elevational gradients in leaf herbivory and its underlying drivers: effects of plant growth form, leaf habit, and climatic correlates. J Ecol 106:413–421. https://doi.org/10.1111/1365-2745.12866 Gianoli E, Salgado-Luarte C (2017) Tolerance to herbivory and the resource availability hypothe 13:20170120. https://doi.org/10.1098/rsbl.2017.0120 Gianoli E, Salgado-Luarte C (2017) Tolerance to herbivory and the 13:20170120. https://doi.org/10.1098/rsbl.2017.0120 g ( ) y 3:20170120. https://doi.org/10.1098/rsbl.2017.0120 p g Gong B, Zhang G (2014) Interactions between plants and herbivores: a review of plant defences. Act Sin 34:325–336. https://doi.org/10.1016/j.chnaes.2013.07.010 p g j Hirsh H, Marler T (2002) Damage and recovery of Cycas micronesica after Typhoon Paka. Biotropica 34:598–602. https://doi.org/10.1111/j.1744-7429.2002.tb00579.x p g j IBM Corp (2020) IMB SPSS Statistics for Windows, Version 27.0. y ( ) doi.org/10.4161/cib.14084 Staude HS (2008) An annotated report on further host-plant associations for african loopers (Lepidopte Geometridae). Metamorphosis 19:193–209 Staude HS, Sihvonen P (2014) Revision of the African geometrid genus Zerenopsis C. & R. Felder – moths with peculiar life histories and mating behaviour (Geometridae: Ennominae: Diptychini). Metamorpho­ sis 25:11–55 Staude HS, Bayliss J, Sihvonen P (2011) The Mulanje tiger moth Callioratis grandis Prout, 1922, a new status, a critically endangered species from Malawi (Lepidoptera: Geometridae: Diptychinae). Meta­ morphosis 22:49–64f p Stephens AEA, Westoby M (2014) Effects of insect attack to stems on plant survival, growth, reproduction and photosynthesis. Oikos 124:266–273. https://doi.org/10.1111/oik.01809 Steyn T, Bezuidenhout H (2020) Entabeni biodiversity and conservation value. Internal report for E Safari Conservancy Scientific Services Kimberley South African National Parks yn T, Bezuidenhout H (2020) Entabeni biodiversity and conservation value. Internal report for Entabeni Steyn T, Bezuidenhout H (2020) Entabeni biodiversity and conservation value. Internal report Safari Conservancy, Scientific Services, Kimberley, South African National Parks y ( ) y Safari Conservancy, Scientific Services, Kimberley, South African National Parks Swart C, Rowswell R, Donaldson J, Barker N (2019) Population structure and survival of the critically endan­ gered cycad Encephalartos latifrons in South Africa. S Afr J Bot 127:80–90. https://doi.org/10.1016/j. sajb.2019.08.034 j Taylor GS, Moir ML (2014) Further evidence of the coextinction threat for jumping plant-lice: three new Acizzia (Phyllidae) and Trioza (Triozidae) from western Australia. Insect Syst Evol 45:283–302. https:// doi.org/10.1163/1876312X-00002107 g Terblanche RF (2018) Nectar plants, pollination and implications for dispersal corridors of the critically endangered Callioratis millari (Lepidoptera: Geometridae, Diptychinae). Metamorphosis 29:47–50 Turcotte MM, Davies TJ, Thomsen CJM, Johnson MTJ (2014) Macroecological and macroevolutionary patterns of leaf herbivory across vascular plants. Proc R Soc B: Biol Sci 281:20140555. https://doi. org/10.1098/rspb.2014.0555i Van Staden PJ, Bredenkamp GJ, Bezuidenhout H, Brown LR (2021) A reclassification and description of the Waterberg Mountain vegetation of the Marakele National Park, Limpopo province, South Africa. Koedoe 63:a1689. https://doi.org/10.4102/koedoe.v63i1.1689 p g Whitaker MRL, Salzman S (2020) Ecology and evolution of cycad-feeding Lepidoptera. Ecol Lett 23:1862– 1877. https://doi.org/10.1111/ele.13581 Whitaker MRL, Gilliéron F, Skirgaila C, Mescher MC, De Moraes CM (2022) Experimental evidence chal­ lenges the presumed defensive function of a “slow toxin” in cycads. Sci Rep 12:6013. https://doi. References South African Biodiversity Institute, Pretoria, pp 349–437 Myers JH, Sarfraz RM (2017) Impacts of insect herbivores on plant populations. Annu Rev Entomol 62:207– 230. https://doi.org/10.1146/annurev-ento-010715-023826f Olivier JC (2006) Population genetic effects of human-mediated plant range expansions on native phytopha­ gous insects. Oikos 112: 456–463. https://doi.org/10.1111/j.0030-1299.2006.14185.x Prado A, Sierra A, Windsor D, Bede J (2014) Leaf traits and herbivory levels in a tropical gymnospe Zamia stevensonii (Zamiaceae). Am J Bot 101:437–447. https://doi.org/10.3732/ajb.1300337 Prado A, Runio-Mendez G, Yañez-Espinosa L, Bede JC (2016) Ontogenetic changes in azoxyglyco­ side levels in the leaves of Dioon edule Lindl. J Chem Ecol 42:1142–1150. https://doi.org/10.1007/ s10886-016-0774-z Ramírez-Restrepo L, Koi S, MacGregor-Fors I (2016) Tales of urban conservation: Eumaeus butter­ flies and their threatened cycad host plants. Urban Ecosyst 20:375–378. https://doi.org/10.1007/ s11252-016-0599-0 Santos Murgas A, Abrego JC (2016) Historia natural de Eumaeus godarti (Lycaenidae, Lepidoptera) y her­ bivoria en Zamia manicata Revista Colón Ciencias, Tecnología y Negocios 3:36–48 1 3 3 Biodiversity and Conservation (2023) 32:2451–2468 2468 Schlinkert H, Westphal C, Clough Y, Ludwig M, Kabouw P, Tscharntke T (2015) Feeding damage to plants increases with plant size across 21 Brassicaceae species. Oecologia 179:455–466. https://doi. org/10.1007/s00442-015-3353-zf g o X, Zhang Q, Liu Y, Yang X (2020) Effects of wind speed on background herbivory of an insect herbi­ vore. Écoscience 27:71–76. https://doi.org/10.1080/11956860.2019.1666549 Sihvonen P, Staude HS, Mutanen M (2015) Systematic position of the enigmatic african cycad moths: an integrative approach to a nearly century old problem (Lepidoptera: Geometridae, Diptychini). Syst Entomol 40:606–627. https://doi.org/10.1111/syen.12125 p g y der LR, Marler TE (2011) Rethinking cycad metabolite research. Commun Integr Biol 4:86–88. https:// doi.org/10.4161/cib.14084 y ( ) doi.org/10.4161/cib.14084 org/10.1038/s41598-022-09298-3 g Wu LW, Yen SH, Lees DC, Hsu YF (2010) Elucidating genetic signatures of native and introduced popula­ tions of the Cycad Blue, Chilades pandava to Taiwan: a threat both to sago palm and to native Cycas populations worldwide. Biol Invasions 12:2649–2669. https://doi.org/10.1007/s10530-009-9672-4 p p p g Zang S, Zhang Y, Ma K (2017) The association of leaf lifespan and backgro specific level. Ecology 98:425–432. https://doi.org/10.1002/ecy.1649 Zang S, Zhang Y, Ma K (2017) The association of leaf lifespan and background specific level. Ecology 98:425–432. https://doi.org/10.1002/ecy.1649 Zunckel K (1995) The ecology and management of the Kaapsehoop & Burtt Davy). Dissertation, University of Cape Townf Zunckel K (1995) The ecology and management of the Kaapsehoop cycad (Encephalartos laevifolius St & Burtt Davy). Dissertation, University of Cape Townf y) , y p Zvereva EL, Kozlov MV (2014) Effect of herbivory on leaf life span in woody plants: a meta-analysis. J Ecol 102:873–881. https://doi.org/10.1111/1365-2745.12252 Zvereva EL, Kozlov MV (2014) Effect of herbivory on leaf life s 102:873–881. https://doi.org/10.1111/1365-2745.12252 Zvereva EL, Kozlov MV (2014) Effect of herbivory on leaf life span in woody plants: a meta-analysis. J Ecol 102:873–881. https://doi.org/10.1111/1365-2745.12252 102:873–881. https://doi.org/10.1111/1365-2745.12 Zvereva EL, Zverev V, Kozlov MV (2012) Little strokes fell great oaks: minor but chronic herbivory substan­ tially reduces birch growth. Oikos 121:2036–2043. https://doi.org/10.1111/j.1600-0706.2012.20688.x Zvereva EL, Zverev V, Kozlov MV (2012) Little strokes fell great oaks: minor but chronic herbivory substan­ tially reduces birch growth. Oikos 121:2036–2043. https://doi.org/10.1111/j.1600-0706.2012.20688.x Publisher’s Note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s Note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manu­ script version of this article is solely governed by the terms of such publishing agreement and applicable law. 1 3